MCU 2015 ERS Practical Handbook of Noninvasive Ventilation

practical
Handbook
Noninvasive
Ventilation
Editor
Anita K. Simonds
PUBLISHED BY
THE EUROPEAN RESPIRATORY SOCIETY
CHIEF EDITOR
Anita K. Simonds (London, UK)
ERS STAFF
Eddy Baker, Alice Bartlett, Matt Broadhead, Lee Dodd, Jonathan Hansen,
Catherine Pumphrey, Elin Reeves, David Sadler, Claire Turner, Ben Watson
© 2015 European Respiratory Society
Cover image: Juanmonino, iStockphoto
Design by Claire Turner, Lee Dodd, ERS

Typeset by Techset Composition Ltd
Printed by Latimer Trend
Indexed by Merrall-Ross International Limited
All material is copyright to the European Respiratory Society.

It may not be reproduced in any way including electronically without the
express permission of the society.
CONTACT, PERMISSIONS AND SALES REQUESTS:

European Respiratory Society, 442 Glossop Road, Sheffield, S10 2PX, UK
Tel: +44 114 2672860 Fax: +44 114 2665064 e-mail: info@ersj.org.uk
ISBN 978-1-84984-075-0
Table of contents
Contributors vii
Preface xi
Get more from this Practical Handbook xii
List of abbreviations xiii
Chapter 1 – Introduction
NIV: past, present and future 1

Anita K. Simonds
Chapter 2 – Getting the basics right: equipment
Basic principles of ventilators 10

Alanna Hare and Michelle Chatwin
Matching mode and settings to the patient: an introduction 18

Jan H. Storre and Jens C. Callegari
Choosing the interface 26

Anne-Kathrin Brill
Supplemental oxygen and humidication 35

Sundeep Kaul and Anita K. Simonds
Chapter 3 – Getting the basics right: patient selection
The patient with an acute hypercapnic exacerbation of COPD 41

Mark W. Elliott
Patients with acute hypercapnic respiratory failure and 49

neuromuscular or chest wall disease
Anita K. Simonds
Patients with acute hypercapnic respiratory failure and OHS 56
Nicholas Hart and Patrick B. Murphy
Patients with acute-on-chronic hypercapnic respiratory failure 60

due to non-COPD obstructive lung disease and interstitial disorders
Marieke L. Duiverman and Peter J. Wijkstra
The patient with acute hypoxaemic respiratory failure excluding 67

pulmonary oedema
Pongdhep Theerawit, Yuda Sutherasan and Paolo Pelosi
The patient with acute hypoxaemic failure and cardiogenic 72

pulmonary oedema
João C. Winck and Luís F. Azevedo
Chapter 4 – Paediatric indications for NIV
Acute NIV in children, including ventilator and interface choice 79

Alessandro Amaddeo and Brigitte Fauroux
Chapter 5 – Airway clearance and physiotherapy
Airway clearance methods and nebulised therapy in acute NIV 86

Michelle Chatwin
Chapter 6 – Monitoring progress in acute NIV
Monitoring choices in acute NIV 93

Raffaele Scala
Starting and stopping acute NIV: when and why? 102

Bernd Schönhofer
Problem-solving: case studies of NIV problems and their 111

management
Alanna Hare
Chapter 7 – NIV and the intensive care unit
NIV and weaning 118

Miquel Ferrer
NIV to avoid re-intubation 127

Paolo Navalesi and Federico Longhini
NIV in the perioperative period 135

Yuda Sutherasan, Maria Vargas and Paolo Pelosi
NIV for endoscopic procedures 142

Leo M.A. Heunks, Lisanne Roesthuis and Erik H.F.M. van der Heijden
NIV in respiratory pandemics 148

Anita K. Simonds
Stepping up and down from NIV to tracheostomy ventilation 155

Mark W. Elliott
Chapter 8 – Long-term NIV
Chronic NIV in hereditary neuromuscular disorders 163

Anita K. Simonds
Chronic NIV in motor neurone disease/ALS 176

Joan Escarrabill
Chronic NIV in chest wall disorders 182

Chronic NIV in COPD 190

Wolfram Windisch and Jan H. Storre
Chronic NIV in OHS 197

Patrick B. Murphy and Nicholas Hart
Chronic NIV in bronchiectasis, CF and interstitial lung 204

disease
Amanda J. Piper
Chronic NIV in heart failure patients: ASV, NIV and CPAP 211
João C. Winck, Marta Drummond, Miguel Gonçalves and Tiago Pinto
Chronic NIV in children: indications, outcomes and transition 217

Hui-Leng Tan and Anita K. Simonds
Practicalities of and guide to cough augmentation and 226

daytime mouthpiece ventilation
Michelle Chatwin
Long-term NIV failure: causes and problem solving 234

Jean-Paul Janssens, Dan Adler and Jesus Gonzalez-Bermejo
Long-term NIV case histories 246

Alanna Hare
Chapter 9 – NIV for symptom palliation
NIV in palliative care and at the end of life 253

Anna Maria Cuomo
Chapter 10 – Discharge planning and community care
Discharging the ventilator-dependent adult and child 260

Joan Escarrabill
Home monitoring and follow-up of long-term NIV 265

Dan Adler, Claudio Rabec and Jean-Paul Janssens
Assessing quality of life and outcome of long-term NIV 276

Sophie E. Huttmann and Wolfram Windisch
Patient and carer education, and risk management 282

Joan Escarrabill
Chapter 11 – Setting up an NIV service
Setting up and staffing your NIV unit 289

Joan Escarrabill
Index 295
Contributors
Chief Editor Jens C. Callegari

Cologne Merheim Hospital, Dept
Anita K. Simonds of Pneumology, Kliniken der Stadt
NIHR Respiratory Biomedical Köln gGmbH, Witten/Herdecke
Research Unit, Royal Brompton and University, Cologne, Germany
Hareeld NHS Foundation Trust, CallegariJ@kliniken-koeln.de
London, UK
A.Simonds@rbht.nhs.uk Michelle Chatwin
Clinical and Academic Department
of Sleep and Breathing,
Authors Royal Brompton and Hareeld NHS
Foundation Trust,
Dan Adler London, UK
Division of Pulmonary Diseases, m.chatwin@rbht.nhs.uk
Geneva University Hospital, Geneva,
Switzerland Anna Maria Cuomo
Dan.Adler@hcuge.ch Freelance oncologist,
Bologna, Italy
Alessandro Amaddeo anniecuomo@gmail.com
Pediatric Noninvasive Ventilation
and Sleep Unit, Necker University Marta Drummond
Hospital, Paris, France Pulmonology Dept, Faculty of
alessandro.amaddeo@gmail.com Medicine, Porto University, Porto,
and Sleep and Non-invasive
Luís F. Azevedo Ventilation Unit, Pulmonology Dept,
Dept of Health Information and Centro Hospitalar São João, Porto,
Decision Sciences, Center for Health Portugal
Technology and Services Research marta.drummond@gmail.com
– CINTESIS (Centro de Investigação
em Tecnologias e Serviços de Marieke L. Duiverman
Saúde), Faculdade de Medicina da Department of Pulmonology/Home
Universidade do Porto, Portugal Mechanical Ventilation, University
lazevedo@med.up.pt Medical Center Groningen,
University of Groningen, Groningen,
Anne-Kathrin Brill The Netherlands
University Hospital Berne, m.l.duiverman@umcg.nl
Inselspital, Dept of Pneumology,
Bern, Switzerland Mark W. Elliott
anne-kathrin.brill@insel.ch Dept of Respiratory Medicine,
Sleep and Non-invasive Ventilation
Services, St. James’s University
Hospital, Leeds, UK
mark.elliott2@nhs.net
Joan Escarrabill Nicholas Hart
Hospital Clínic, Chronic Care Lane Fox Clinical Respiratory
Program, Barcelona, Spain Physiology Research Centre, Guy’s
ESCARRABILL@clinic.ub.es and St Thomas’ NHS Foundation
Trust, Division of Asthma, Allergy
Brigitte Fauroux and Lung Biology, King’s College
Pediatric Noninvasive Ventilation London, and Lane Fox Respiratory
and Sleep Unit, Necker University Unit, Guy’s and St Thomas’ NHS
Hospital, Paris, France Foundation Trust, London, UK
brigitte.fauroux@nck.aphp.fr nicholas.hart@gstt.nhs.uk
Miquel Ferrer Leo M.A. Heunks

Respiratory Intensive and Dept of Intensive Care Medicine,
Intermediate Care Unit, Dept of Radboud University Medical
Pneumology, Thorax Institute, Centre Nijmegen, Nijmegen, The
Hospital Clínic, IDIBAPS, Barcelona, Netherlands
Spain Leo.Heunks@radboudumc.nl
miferrer@clinic.ub.es
Sophie E. Huttmann
Miguel Gonçalves Cologne Merheim Hospital,
Pulmonology Dept, Faculty of Department of Pneumology,
Medicine, Porto University, Kliniken der Stadt Köln gGmbH
and Sleep and Non-invasive Witten/Herdecke University, Faculty
Ventilation Unit, Pulmonology Dept, of Health/School of Medicine,
Centro Hospitalar São João, Porto, Cologne, Germany
Portugal HuttmannS@kliniken-koeln.de
goncalvesmr@gmail.com
Jean-Paul Janssens
Jesus Gonzalez-Bermejo Division of Pulmonary Diseases,
Service de Pneumologie et Geneva University Hospital, Geneva,
Réanimation, GH Pitié Salpêtrière, Switzerland
Paris, France jean-paul.janssens@hcuge.ch
jesus.gonzalez@psl.aphp.fr
Sundeep Kaul
Alanna Hare Royal Brompton and Hareeld NHS
Royal Brompton and Hareeld NHS Foundation Trust, Hareeld Hospital,
Foundation Trust, Hareeld, UK
London, UK sunnykaul@aol.com
A.Hare@rbht.nhs.uk
Federico Longhini
Anesthesia and Intensive Care,
Sant’Andrea Hospital, ASL VC,
Vercelli, Italy
longhini.federico@gmail.com
Patrick B. Murphy Claudio Rabec
Lane Fox Clinical Respiratory Service de Pneumologie et
Physiology Research Centre, Guy’s Réanimation, Centre Hospitalier et
and St Thomas’ NHS Foundation Universitaire de Dijon, Dijon, France
Trust, Division of Asthma, Allergy Claudio.Rabec@chu-dijon.fr
and Lung Biology, King’s College
London, and Lane Fox Respiratory Lisanne Roesthuis
Unit, Guy’s and St Thomas’ NHS Dept of Intensive Care Medicine,
Foundation Trust, London, UK Radboud University Medical
patrick.murphy@gstt.nhs.uk Centre Nijmegen, Nijmegen, The
Netherlands
Paolo Navalesi Lisanne.Roesthuis@radboudumc.nl
Anesthesia and Intensive Care,
Sant’Andrea Hospital, ASL VC, Raffaele Scala
Vercelli, Dept of Translational Pulmonary Unit and Respiratory
Medicine, Eastern Piedmont Intensive Care Unit, S. Donato
University “A. Avogadro”, Novara, Hospital, Arezzo, Italy
and CRRF Mons. L. Novarese, raffaele_scala@hotmail.com
Moncrivello, Italy
paolo.navalesi@med.unipmn.it Bernd Schönhofer
Pneumologie, Internistische
Paolo Pelosi Intensivmedizin und Schlafmedizin,
IRCCS AOU San Martino-IST, Dept KRH Klinikum Siloah- Oststadt-
of Surgical Sciences and Integrated Heidehaus, Hannover, Germany
Diagnostics, University of Genoa, Bernd.Schoenhofer@t-online.de
Genoa, Italy
ppelosi@hotmail.com Jan H. Storre
Tiago Pinto of Pneumology, Kliniken der Stadt
Sleep and Non-invasive Ventilation Köln gGmbH, Witten/Herdecke
Unit, Pulmonology Dept, Centro University, Cologne, and
Hospitalar São João, Porto, Portugal Dept of Pneumology, University
tiagoffpinto@gmail.com Hospital Freiburg, Germany
storrej@kliniken-koeln.de
Amanda J. Piper
Dept of Respiratory and Sleep Yuda Sutherasan
Medicine, Royal Prince Alfred Ramathibodi Hospital, Mahidol
Hospital, Camperdown, and University, Bangkok, Thailand
Woolcock Institute of Medical sutherasan_yuda@yahoo.com
Research, University of Sydney,
Sydney, Australia Hui-Leng Tan
amanda.piper@sydney.edu.au Dept of Paediatric Respiratory
Medicine, Royal Brompton Hospital,
Royal Brompton and Hareeld NHS
Foundation Trust, London, UK
H.Tan@rbht.nhs.uk
Pongdhep Theerawit João C. Winck
Ramathibodi Hospital, Mahidol Pulmonology Dept, Center for
University, Bangkok, Thailand Health Technology and Services
pongdhep@yahoo.com Research – CINTESIS (Centro de
Investigação em Tecnologias e
Erik H.F.M. van der Heijden Serviços de Saúde), Faculdade de
Dept of Pulmonary Diseases, Medicina da Universidade do Porto,
Radboud University Medical Centre Porto, Portugal
Nijmegen, Nijmegen, jcwinck@mail.telepac.pt
The Netherlands
Erik.vanderheijden@radboudumc.nl Wolfram Windisch
Maria Vargas of Pneumology, Kliniken der Stadt
Dept of Neuroscience Köln gGmbH Witten/Herdecke
and Reproductive and University, Faculty of Health/School
Odontostomatological Sciences, of Medicine, Cologne, Germany
University of Naples “Federico II”, windischw@kliniken-koeln.de
Naples, Italy
vargas.maria82@gmail.com
Peter J. Wijkstra
Department of Pulmonology/Home
Mechanical Ventilation, University
Medical Center Groningen,
University of Groningen, Groningen,
The Netherlands
p.j.wijkstra@umcg.nl
Preface
Bearing in mind the growing applications for noninvasive ventilation (NIV)

and the success of the ERS NIV courses, we felt the time had come for an ERS
Practical Handbook of Noninvasive Ventilation. Gaining knowledge in NIV is
not difficult as there are many resources available – what makes the difference
in clinical practice is expertise in applying the technique and acquiring skills
in problem-solving. It is these practical issues this book attempts to address,
together with providing an up-to-date guide to which patients to select for
therapy, and the evidence on which this is based.
NIV is a rapidly changing eld. In recent months we have had trial results on
the use of adaptive servo ventilation, suggesting that this does not improve
outcome in heart failure patients with central sleep apnoea; and indications
for home NIV in chronic COPD have been informed by recent trial results.
One of the greatest values of ERS NIV courses is the chance to discuss
challenges in patient care. Some of these are perennial – persuading a
confused, hypercapnic patient or small child to cope with a mask, avoiding
interface leaks, and balancing ventilatory needs with patient tolerance. In this
book we have invited those in the front line of NIV to share their views and
experiences. This means there will be a variety of approaches suggested in
some cases, whereas in others there may be a greater degree of consensus, or
even complete agreement. We very much hope this will help the reader not
only to make the right decision for their patients but also to gain condence
from the very practical advice offered.
I am extremely grateful to all the contributors and reviewers, and to the ERS
Publications Office team for their great support and enthusiasm.
Anita K. Simonds
Chief Editor
Get more from this
Handbook
By buying the ERS Practical Handbook of Noninvasive Ventilation, you also
gain access to the electronic version of the book, as well as an accredited
online CME test.
Simply visit http://erspublications.com/titles and add the ERS Practical

Handbook of Noninvasive Ventilation to your cart. At the checkout, enter
the unique voucher code printed on the inside front cover of the book. You
will then be able to download the entire book in PDF format to read on your
computer or mobile device.
You’ll also be able to take the online CME test. This Practical Handbook has
been accredited by the European Board for Accreditation in Pneumology
(EBAP) for 12 CME credits.
Also available from the ERS
ERS Handbook: Self-Assessment in Respiratory Medicine

Edited by Konrad E. Bloch, Anita K. Simonds and Thomas Brack
The new and updated second edition of Self-Assessment in Respiratory

Medicine is an invaluable tool for any practitioner of adult respiratory
medicine. The 261 multiple-choice questions cover the full breadth
of the specialty, using clinical vignettes that test not only readers’
knowledge but their ability to apply it in daily practice.
List of abbreviations
AHI Apnoea–hypopnoea index

AIDS Acquired immunodeciency syndrome
ALS Amyotrophic lateral sclerosis
ARDS Acute respiratory distress syndrome
ARF Acute respiratory failure
ASB Assisted spontaneous breathing
ASV Adaptive servo ventilation
ASSPCV Assisted pressure-controlled ventilation
AVAPS Average volume-assured pressure support
BMI Body mass index
CF Cystic brosis
COPD Chronic obstructive pulmonary disease
CPAP Continuous positive airway pressure
ECG Electrocardiogram
EPAP Expiratory positive airway pressure
FEV1 Forced expiratory volume in 1 s
FIO2 Inspiratory oxygen fraction
FVC Forced vital capacity
HIV Human immunodeciency virus
ICU Intensive care unit
IPAP Inspiratory positive airway pressure
IPPV Intermittent positive pressure ventilation
IVAPS Intelligent volume-assured pressure support
NIV Noninvasive ventilation
NPV Negative pressure ventilation
OHS Obesity hypoventilation syndrome
OSA(S) Obstructive sleep apnoea (syndrome)
PaCO2 Arterial carbon dioxide tension
PaO2 Arterial oxygen tension
PAV Proportional assist ventilation
PCV Pressure-controlled ventilation
PEEP Positive end-expiratory pressure
PSV Pressure support ventilation
PtcCO2 Transcutaneous carbon dioxide tension
SaO2 Arterial oxygen saturation
RCT Randomised controlled trial
SpO2 Arterial oxygen saturation measured by pulse oximetry
TB Tuberculosis
TLC Total lung capacity
VCV Volume-controlled ventilation
VT Tidal volume
Introduction
NIV: past, present and future
Anita K. Simonds
The history of NIV is an intertwining chronicle of the development of negative

and positive pressure modes, as at different times in history, each mode has
dominated. For example, in the mid-1980s, all patients using respiratory support
received NPV, whereas the vast majority of our patients now use positive-pressure
NIV, and there are hundreds of thousands of individuals with sleep apnoea world-
wide receiving CPAP and many thousands using NIV for respiratory failure.
Indeed, the history of NIV dates right back to the beginning:
“And the LORD God formed man of the dust of the ground, and breathed
into his nostrils the breath of life; and man became a living soul”.
Genesis 2: 7
So NIV existed before invasive ventilation! More scientifically, from an invasive
ventilation perspective, there are descriptions going back to antiquity of trache-
ostomy, and these are found in the medieval period and 1500s too. Vesalius, in
the 16th century, was aware that positive pressure applied to the trachea would
inflate the lungs and Hooke demonstrated that it was possible to keep a dog alive
by applying bellows to the upper airway in 1667. For verifiable human accounts,
we need to advance to the 18th century – and first deal with NPV.
Woollam (1976) credits John Dalziel, a Scot, as the first to describe a tank ventila-
tor/iron lung type device and a fellow Scot, Alexander Graham Bell, developed the
notion further. Bell is best known for inventing the forerunner of the telephone in
1876 but, in 1881, his first child Edward died of respiratory distress a few hours
after birth. The story goes that shortly afterwards Bell was walking along a shingle
beach in Ontario when the idea of a negative-pressure jacket to assist the breath-
ing of infants came to him. He went on to patent this device. Figure 1 is an extract
from his notebook, although it is quite difficult to decipher.
Key points
• The development of negative- and positive-pressure NIV is
inextricably linked.
• NIV is one of the most evidence-based areas of respiratory
medicine and indications for NIV continue to increase in
number.
ERS Practical Handbook Noninvasive Ventilation 1

Introduction
Figure 1. Drawing of a negative pressure device for children by Alexander Graham Bell.
Image: US Library of Congress, Washington, DC, USA.
In fairness, similar ideas were flourishing elsewhere, in Europe, but it was not until
the 1920s that an iron lung with a motorised pump was developed by Drinker in
1928 in the USA and demonstrated in London in 1931. This concept then entered
the medical mainstream, disseminated by a brisk correspondence in the Lancet.
That was fortuitous, as the coming scourge was epidemics of poliomyelitis, which
had begun in the First World War and swept across Europe and the USA in the
1930s–1950s. Polio paralyses the respiratory muscles as well as limbs muscles,
resulting in respiratory failure. Iron lungs were pressed into action, including the
intimidating multitier versions in figure 2. There is no doubt that iron lungs saved
thousands of lives but they were big, cumbersome and expensive (the original
Drinker ventilator cost $1500 – equivalent to the cost of a US new-build house at
the time), and so were not going to be a practical way forward in respiratory care.
In their observations on the use of negative-pressure respirators in polio, Plum
and Wolff (1951) found that the tank ventilator was safest for managing respira-
tory insufficiency and that in the acute phase of polio, the cuirass was too ineffi-
cient. Upper airway obstruction provoked by the negative pressure was a common
problem.
Practical limitations were compounded by the huge outbreak of polio in Denmark
in 1952, which was associated with a very high prevalence of cases with bulbar
weakness. Not only was an insufficient number of iron lungs available but these
were also inadequate in caring for patients with bulbar problems – mortality rose
to 90% and the only solution open to Ibsen (1954) and the Danish anaesthetic
and medical teams was invasive positive-pressure ventilation via a tracheostomy
2 ERS Practical Handbook Noninvasive Ventilation
Introduction
Figure 2. Multitier iron lung used in poliomyelitis epidemics. Reproduced from Kacmarek
(2011) Respir Care; 56: 1170–1180 with permission from the publisher.
or endotracheal tube. This switch to positive pressure continued and iron lungs
began to disappear, heralding the arrival of the modern ICU. There was a brief
resurgence of NPV in the 1970s and 1980s, but mainly to care for those with
chronic ventilatory failure.
Turning to the development of noninvasive positive pressure, this started at a
slightly earlier time. Possibly the first well documented use of mask ventilation,
in the 1760s, was in “resuscitation boxes”, which contained bellows to insufflate
the lungs, tubing and glass nasal masks, and were placed by the Royal Humane
Society (London, UK) to be used in the rescue of “drowned persons”. The first was
located by the Serpentine Lake in Hyde Park, London. Ice skating was much in
vogue at that time and the Serpentine froze in the winter. It seems children fre-
quently fell through the ice and had to be rescued; there is even a protocol for the
resuscitation of children rescued from a frozen lake. In retrospect, it is difficult to
think of a better prognostic group to resuscitate – young, fit and cooled, providing
they were retrieved quickly enough.
But this was all manually applied positive-pressure ventilation and the first true
motorised ventilator did not appear until the turn of the 20th century. In May
1908, under the headline “Smother small dog to see it revived”, The New York Times
described a demonstration to the King County Medical Society, in Brooklyn, NY,
USA, of a mechanised ventilator developed by Prof. George Poe (spookily related
to mystery writer Edgar Allan Poe). A young boy was given a quarter to find a stray
dog on the streets and this “cur” was smothered till apparently lifeless and then
successfully resuscitated with the ventilator, to the acclaim of the audience! So, a
good day for ventilators, but a bad day for stray dogs in Brooklyn.
The best description of use in real clinical practice comes from Germany with
the Dräger Pulmotor (Drägerwerk AG, Lubeck, Germany). This was patented by
Heinrich Dräger in 1907 and was an innovative time-cycled device that deliv-
ered positive pressure during inspiration and negative pressure during expira-
tion. However, it had a flaw in that the mask was connected to the ventilator by a
Introduction
single limb of tubing, meaning that the carbon dioxide in the exhaled breath was
rebreathed, which could eventually result in asphyxiation. Fortunately, Heinrich’s
son Bernard redesigned the circuitry with two sets of tubing, one for inhalation,
the other for exhalation, which solved that problem; this modification went into
production and 30 years later, the 12 000th Pulmotor rolled off the production
line in Lubeck. The Pulmotors were supplied to mines for poisoning accidents, to
deal with victims of fires and for other acute uses. This is crucial, as the stimulus
for ventilator use had been entirely for ARF up to this point.
It was not until the 1970s and 1980s that long-term chronic use began to be
the spur to ventilatory progress. This was partly related to better understanding
of the physiology of breathing during sleep, the rediscovery of sleep apnoea and
CPAP therapy, and underlying global trends in the switch from acute to chronic
healthcare. OSA is associated with recurrent episodes of upper airway obstruc-
tion, which can lead to a number of vascular complications if not addressed, but
which Sullivan showed in 1981 could be effectively treated with CPAP, as the air-
flow splints the airway open. In addition, developments in masks and technology
extended NIV to respiratory failure in patients with neuromuscular disease.
The original CPAP machines were very large – about the size of a vacuum cleaner –
but have improved, and become smaller and portable over time. Importantly,
mask design and comfort have improved too. To complete the timeline, the devel-
opments from the end of the 1980s to the present are shown in figure 3.
A great deal of progress has occurred such that NIV is now one of the most
evidence-based areas of respiratory medicine, as this handbook will describe.
Really significant interventions are the discovery and confirmation by RCT that
NIV halves mortality and morbidity in acute exacerbations of COPD, and this
Chronic
Improved
outcome in
NIV in Duchenne MD
restrictive NIV in motor
CPAP,
disorders neurone NIV in
neuromuscular
disease chronic COPD
disease Paediatric NIV
Late
1980s 1990s 2000 2005 2014
1990s
NIV in acute hypoxaemic respiratory failure,

pulmonary oedema and weaning
NIV in acute COPD Acute
Figure 3. Timeline of developments in NIV from the 1980s to the present day. MD: muscular
dystrophy.
Introduction
provides the rationale for NIV to be available in every acute unit that admits res-
piratory patients, and for NIV to be used post-operatively in high-risk patients and
for weaning. An additional major change in the past 30–40 years has been the
increasing indications for long-term, chronic NIV and, of course, long-term appli-
cation of CPAP in OSA.
For patients with a range of causes of ventilatory failure, the natural history pro-
gresses from normal breathing, to a gradual loss in lung volumes and then, ini-
tially, changes in blood gases are seen at night due to hypoventilation, and if that
is not addressed, ultimately, progression to daytime respiratory failure, cardiac
decompensation and premature death. The interval between the onset of respira-
tory failure and death may be as short as a few years. In Duchenne muscular dys-
trophy, once a patient has developed a raised carbon dioxide level during the day,
there is a 90% chance that they will be dead within a year.
Figure 4 shows the long-term outcome of different groups of patients treated
with NIV having developed severe ventilatory failure or progressed to cor pulmo-
nale pretreatment. In post-polio patients, 5-year survival with NIV is 100% and
it appears that these individuals will live to their normal life expectancy. 5-year
survival is ∼80% in the other restrictive conditions. Results are less good in COPD
and bronchiectasis for two reasons: these are intrinsic lung disease conditions
rather than being restrictive disorders with normal lungs, and the patients were
severely end-stage when treated – with some being on the transplant waiting list.
In COPD, recent trials have shown NIV may be of benefit in stable hypercapnic
patients, as discussed further in the section entitled “Chronic NIV in COPD”. In
Duchenne patients, median survival is now nearly 30 years, and around a third of
our Duchenne patients are living into their late thirties and early forties.
NIV has been extended to the paediatric age range, with the feasibility of using
NIV to control nocturnal hypoventilation in children initially being demonstrated
100 Polio
TB
Neuromuscular
80
Kyphoscoliosis
Continuing NIV %
60
COPD
40
20
Bronchiectasis
0
0 1 2 3 4 5 6
Years
Figure 4. Probability of continuing NIV long term, which is equivalent to survival in most
cases. Reproduced from Simonds et al. (1995) with permission from the publisher.
Introduction
predominantly in neuromuscular conditions. Many of these children now survive to

adolescence or adulthood, as shown in the section entitled “Chronic NIV in heredi-
tary neuromuscular disorders”. Furthermore, characterisation of the genotypes and
phenotypes of some congenital disorders (e.g. congenital myasthenia and congenital
muscular dystrophies) has clarified the natural history of these conditions, thereby
facilitating anticipatory care plans and enabling personalised ventilatory care.
Present trends
There is also growing interest in NIV in cardiology. There is no doubt that patients
with heart disease and OSA benefit from treatment of the OSA. By contrast,
Cheyne-Stokes respiration is a form of central sleep apnoea that has been rec-
ognised for centuries in chronic heart failure. It was previously thought to be
simply a marker of severe disease and an epiphenomenon, but recently, the link
to the progression of disease has been explored and it has been found to be more
prevalent in milder cases. This is important, as heart failure is common and the
majority of those affected have mild cardiac impairment. Recent work shows that
around half of patients with mild heart failure have sleep disordered breathing
too. CPAP can be used to treat OSA but does not work in central sleep apnoea or
Cheyne–Stokes respiration. The new ventilatory concept of ASV aims to smooth
out the Cheyne–Stokes pattern and, in doing so, may reduce associated sym-
pathetic stimulation and arousals from sleep (see the section entitled “Chronic
NIV in heart failure patients: ASV, NIV and CPAP”). Several large international
multicentre trials of ASV in heart failure patients with predominant central sleep
apnoea are now in progress, with cardiac and all-cause mortality or unplanned
admissions as major outcome measures. However early results suggest ASV may
not improve outcome contrary to expectations, and may cause harm in severe
heart failure patients with central sleep apnoea.
Palliative care
NIV is now being used in some situations to palliate symptoms without the aim
of prolonging survival or substantially modifying arterial blood gas tensions.
Here, goals such as reduction in dyspnoea and control of symptoms of nocturnal
hypoventilation should be set pre-emptively so that if these are not met, NIV
can be discontinued and palliative efforts directed elsewhere. Nava et al. (2013)
have shown that in oncology patients with solid tumours complicated by ARF and
an expected life expectancy of <6 months, NIV reduced dyspnoea more rapidly
than oxygen therapy alone and patients required less morphine. The benefit was
most marked in hypercapnic patients and within the first hour of therapy, sug-
gesting that responses can be rapidly gauged. NIV combined with cough-assist
devices can also be used to manage severely ill type 1 spinal muscular atrophy
infants with the aim of discharging the patient to their home and managing
breathlessness.
However, these approaches have been used in units familiar with NIV, and wide
translation to oncology units and other palliative care centres needs to be carefully
considered and managed. The use of NIV to palliate symptoms is discussed further
in the section entitled “NIV in palliative care and at the end of life”.

Introduction
Implementation of NIV in clinical practice

The newer applications discussed above underscore a problem of implementa-
tion of “good practice” in NIV, which affects all units to a greater or lesser extent.
There is evidence that patients who would benefit from NIV are not receiving it
even for gold standard indications, such as acute hypercapnic exacerbation of
COPD, for a number of reasons, but mainly because the medical team does not
feel skilled enough to deliver it. Of course, the answer to this problem is to pro-
vide knowledge and skills to remedy these deficiencies, which is the purpose of
this handbook, many NIV courses and the ERS Skills-based Simulator Training in
Non-Invasive Ventilation.
Intelligent ventilators
The question arises, if medical teams are inexperienced, can you make the ventila-
tor intelligent? A variety of approaches have been adopted in the past few years
to try to combine bilevel pressure support with the delivery of an assured minute
ventilation or VT. The underlying aim of these modes is to better adapt to the
patient’s own ventilatory requirements, which will vary during different stages of
sleep and with different activities during the day. Some devices also have an “intel-
ligent” backup rate and a “learn” mode in which the ventilator adapts to patient’s
respiratory effort and pattern.
AVAPS was one of the first of these new modes. An initial randomised cross
over trial of AVAPS versus standard pressure support in obesity hypoventilation
patients showed a small improvement on nocturnal carbon dioxide tension but no
long-term quality of life improvement. Murphy et al. (2012) confirmed there was
no long-term advantage of using AVAPS over optimally titrated bilevel pressure
support in very obese patients, and results in COPD patients are equivocal. The
IVAPS ventilator targets VT rather than minute ventilation and has been shown
to produce equivalent control of nocturnal hypoventilation to a group of patients
expertly set-up on NIV. In addition, in a group with predominantly restrictive dis-
orders starting NIV for the first time, IVAPS resulted in improved adherence over-
night and a reduction in stage 1 sleep, suggesting sleep initiation when starting
NIV was improved. These results suggest that intelligent modes of ventilation may
have a role in certain subgroups but they have not been demonstrated to be supe-
rior to conventional pressure support NIV in all patient groups.
Other developments
Interface development has also advanced very significantly, with better choice
and design particularly the use of softer contoured material, rather than rigid plas-
tic or vinyl. However, problems with pressure sores and midfacial hypoplasia have
not yet been solved and tactics to overcome these issues are discussed in the sec-
tion entitled “Choosing the interface”.
Organisation and delivery of ventilatory care is likely to evolve too. In acute NIV,
there has been a trend to manage sicker patients successfully in high-dependency
units. For homecare, greater information from ventilator software enables prob-
lems to be solved remotely, and telemonitoring approaches are increasing but
need to be validated. There is every hope that the future of NIV will be as exciting
as its past!

Introduction
Further reading
• Bradley TD, et al. (2003). Sleep apnea and heart failure. Part I: obstructive sleep
apnea. Circulation; 107: 1671–1678.
• Bradley TD, et al. (2005). Continuous positive airway pressure for central sleep
apnea and heart failure. N Engl J Med; 353: 2025–2033.
• Carlucci A, et al. (2003). Changes in the practice of non-invasive ventilation in
treating COPD patients over 8 years. Intensive Care Med; 29: 419–425.
• Chatwin M, et al. (2011). Outcome of goal-directed non-invasive ventilation and
mechanical insufflation/exsufflation in spinal muscular atrophy type I. Arch Dis
Child; 96: 426–432.
• Fauroux B, et al. (1995). Home treatment for chronic respiratory failure in children:
a prospective study. Eur Respir J; 8: 2062–2066.
• Ibsen B (1954). The anaesthetist’s viewpoint of the treatment of respiratory com-
plications in poliomyelitis during the epidemic in Copenhagen, 1952. Proc R Soc
Med; 47: 72–74.
• Jaye J, et al. (2009). Autotitrating versus standard noninvasive ventilation: a ran-
domised crossover trial. Eur Respir J; 33: 566–571.
• Jenkinson C, et al. (1999). Comparison of therapeutic and subtherapeutic nasal
continuous positive pressure airway pressure for obstructive sleep apnoea: a ran-
domised prospective parallel trial. Lancet; 353: 2100–2105.
• Köhnlein T, et al. (2014). Non-invasive positive pressure ventilation for the treat-
ment of severe stable chronic obstructive pulmonary disease: a prospective multi-
centre, randomised, controlled clinical trial. Lancet Respir Med; 2: 698–705.
• Lloyd-Owen SJ, et al. (2005). Patterns of home mechanical use in Europe: results
from the Eurovent survey. Eur Respir J; 25: 1025–1031.
• Maheshwari V, et al. (2006). Utilization of noninvasive ventilation in acute care
hospitals: a regional survey. Chest; 129: 1226–1233.
• Murphy PB, et al. (2012). Volume targeted versus pressure support non-invasive
ventilation in patients with super obesity and chronic respiratory failure: a ran-
domised controlled trial. Thorax; 67: 727–734.
• Nava S, et al. (1998). Noninvasive mechanical ventilation in the weaning of
patients with respiratory failure due to chronic obstructive pulmonary disease.
A randomized controlled trial. Ann Intern Med; 128: 721–728.
• Nava S, et al. (2013). Palliative use of non-invasive ventilation in end-of-life patients
with solid tumours: a randomised feasibility trial. Lancet Oncol; 14: 219–227.
• Plant PK, et al. (2000). Early use of noninvasive ventilation for acute exacerbations
of chronic obstructive pulmonary disease on general respiratory wards: a multi-
centre randomised controlled trial. Lancet; 355: 1931–1935.
• Plum F, et al. (1951). Observations on acute poliomyelitis with respiratory insuf-
ficiency. JAMA; 146: 442–446.

Introduction
• Shaw LA, et al. (1929). An apparatus for the prolonged administration of artificial
ventilation. J Clin Invest; 8: 33–46.
• Simonds AK, et al. (1995). Outcome of domiciliary nasal intermittent positive pres-
sure ventilation in restrictive and obstructive disorders. Thorax; 50: 604–609.
• Simonds AK, et al. (2000). Outcome of paediatric domiciliary mask ventilation in
neuromuscular and skeletal disease. Eur Respir J; 16: 476–481.
• Storre JH, et al. (2006). Average volume-assured pressure support ventilation in
obesity hypoventilation. A randomised crossover trial. Chest; 130: 815–821.
• Sullivan CE, et al. (1981). Reversal of obstructive sleep apnoea by continuous posi-
tive pressure applied through the nares. Lancet; 1: 862–865.
• Vazir A, et al. (2007). A high prevalence of sleep disordered breathing in men with
mild symptomatic chronic heart failure due to left ventricular systolic dysfunction.
Eur J Heart Fail; 9: 243–250.
• Vianello A, et al. (1994). Long-term nasal intermittent positive pressure ventilation
in advanced Duchenne’s muscular dystrophy. Chest; 105: 445–448.
• Woollam CHM (1976). The development of apparatus for intermittent negative
pressure respiration (1) 1832–1918. Anaesthesia; 31: 537–547.
• Woollam CHM (1976). The development of apparatus for intermittent negative
pressure respiration (2) 1919–1976. Anaesthesia; 31: 666–685.
Online resources
• Hare A, et al. Skills-based Simulator Training in Non-Invasive Ventilation. www.
ers-education.org/Media/Media.aspx?idMedia=234264

The basics: equipment
Basic principles of ventilators
Alanna Hare and Michelle Chatwin
Ventilatory assistance may be provided by the delivery of intermittent positive

pressure to the airway or intermittent negative pressure to the chest wall. Invasive
ventilation involves airway intubation (via an endotracheal tube or tracheostomy)
but may also be provided by the use of electrodes, as in diaphragmatic pacing. NIV,
by contrast, avoids intubation and its inherent risks (including ventilator-acquired
pneumonia, tracheal stenosis and the complications associated with sedation)
while preserving functions such as speaking and swallowing. As a result, how-
ever, noninvasive modes of ventilation may be less appropriate in patients with
impaired airway reflexes or excessive airway secretions.
Terminology
It is important to understand the terminology used in NIV. Figure 1 illustrates
some of the setting parameters that can be adjusted when setting up NIV and
their relationship to the normal respiratory cycle. Not all settings are required in
all modes: the different modes will be covered in more detail later in this section.
Inspiratory positive airway pressure IPAP is the pressure delivered by the ventilator
while the patient is inhaling. This establishes the pressure support (IPAP minus
EPAP), which provides assistance to inspiration and decreases the work of breath-
ing by unloading the respiratory muscles, leading to increased VT and minute ven-
tilation, and improved gas exchange.
Expiratory positive airway pressure EPAP is the pressure delivered by the ventila-
tor while the patient is exhaling. EPAP assists with the maintenance of upper
airway patency in sleep, which may be important in patients with an unstable
Key points
• NIV is the provision of ventilatory assistance without the use of
an invasive airway.
• NIV may be provided as either positive pressure via a facial
mask, mouthpiece or helmet, or as negative pressure, for
example, via a tank ventilator, cuirass or pneumojacket.
• Positive pressure ventilation may be pressure preset or volume
preset. Newer modes of ventilation are able to combine
aspects of both volume- and pressure-preset ventilation.

One breath cycle

tI
tImax
tImin
IPAP
Pressure
support
EPAP
PEEP
Rise time Fall time
Time
Figure 1. Common parameters than can be adjusted within the different modes when setting
up NIV and their relationship to the normal respiratory cycle. tI: inspiratory time; tImax: maximum
inspiratory time; tImin: minimum inspiratory time. Reproduced and modified from Skills-based
Simulator Training in Non-Invasive Ventilation. © ResMed Limited. All rights reserved.
upper airway (e.g. in OSA) and helps to recruit/maintain lung volume, improving
oxygenation. In obstructive lung disease, EPAP helps to overcome the inspiratory
threshold load when intrinsic PEEP is present, reducing the work of breathing and
maximising effective triggering. EPAP is necessary to ensure sufficient expiratory
flow to flush carbon dioxide from the ventilatory dead space, enabling removal
of carbon dioxide from the ventilator circuit and preventing rebreathing. Some
ventilators enable the clinician to set an EPAP range that adjusts the level of PEEP
applied in response to patient-related changes (auto-EPAP).
Volume Some ventilators require the practitioner to set a VT or minute volume to
be delivered by the ventilator while the patient is inhaling, rather than an inspira-
tory pressure. This ventilatory mode will be discussed further later in this section.
Backup rate In some ventilator modes, a backup rate can be set by the clinician to
ensure a minimum ventilation level for the patient. If the patient’s spontaneous
breathing rate falls below this level, the ventilator will deliver timed breaths to
ensure the minimum breath rate is met. This is important in patients whose cen-
tral respiratory drive is impaired in sleep (e.g. in central hypoventilation syndrome
and some patients with neuromuscular disease).
Inspiratory time limits Some ventilators in certain ventilatory modes (described
later in this section) enable the clinician to set an inspiratory time (tI). This ensures
the ventilator delivers the inspiratory pressure for a set time only, regardless of the
patient’s spontaneous breathing pattern. Other modes of ventilation will require
parameters for inspiratory time to be set, for example, minimum and maximum tI.
Inspiratory time limits aim to avoid the problems of:
• premature cycling, in which the inspiratory time may be too short for an adequately
supported breath
• late cycling, in which the inspiratory time may be too long (often as a result of
circuit leaks) causing insufficient time for expiration and leading to the devel-
opment of auto-PEEP
Triggering As the patient begins to inhale, the ventilator detects a change in flow
or pressure, which triggers the ventilator to change from expiratory to inspira-
tory pressure. Leaks can affect ventilator triggering, as they may cause a pressure
or flow drop resulting in auto-triggering. Leaks may also contribute to ineffective
inspiratory effort by preventing the detection of the patient’s inspiratory effort.
Some ventilators enable the practitioner to set the trigger sensitivity in order to
overcome these issues.
Cycling As the patient exhales, the ventilator detects a change in flow, which
causes the ventilator to cycle into its lower expiratory pressure setting. Leaks can
also affect cycling, as if the leak flow is greater than the ventilator’s flow cycle
criteria, the inspiratory phase will continue. In some modes, cycling can also occur
after a preset time through the setting of inspiratory time limits, as described earlier.
Some ventilators enable the practitioner to set the cycle sensitivity.
Rise time This is the time taken to reach IPAP after the onset of the inspiratory
phase. Patients with a high ventilatory demand may require a short rise time,
especially those needing a shorter inspiratory and a longer expiratory time (e.g.
patients with obstructive lung disease). In patients with a slower respiratory rate or
lungs that empty rapidly, such as patients with chest wall deformities, increasing
rise time can improve comfort. An overly long rise time may result in a reduction
in the effective VT delivered to the patient, and the actual rise time achieved is
influenced by several factors including lung compliance, leaks and patient breath-
ing patterns.
Fall time This is the time taken for the inspiratory pressure to fall after the ven-
tilator cycles into expiration. Some ventilators enable the practitioner to set the
fall time.
Ventilator modes
Continuous positive airway pressure CPAP provides a single continuous pressure
throughout the respiratory cycle. Strictly speaking, CPAP is not NIV, as it does not
actively support the respiratory muscles or assist with the delivery of VT. If the
patient fails to make a spontaneous respiratory effort, the ventilator will continue
to deliver a constant pressure, which provides no ventilatory assistance to the
patient. This mode, therefore, is not effective in the presence of apnoea.
Pressure support ventilation PSV provides additional support during inspiration,
unloading the respiratory muscles and increasing VT. In this mode, the practi-
tioner selects both an IPAP and an EPAP to be delivered by the ventilator. PSV
may be provided in spontaneous or spontaneous/timed mode. In spontaneous
mode, the ventilator is triggered to deliver the preset IPAP by an increase in the
patient’s inspiratory flow. The ventilator cycles into expiration when it detects a
fall in the patient’s respiratory flow. No backup respiratory rate is provided by the
ventilator and no limits are set on the inspiratory time. By contrast, in spontane-
ous/timed mode, a backup rate is also set by the practitioner, so that, should
the patient’s r espiratory rate fall below a certain preset level, the ventilator will
deliver a backup timed breath. This backup rate is useful in the setting of apnoea
or periodic breathing.
Pressure-controlled ventilation In PCV, as in the spontaneous/timed mode of PSV,
the ventilator is triggered by inspiration by the patient, or after a set period of
time if the patient’s spontaneous breathing rate falls below the backup rate. In
this mode, however, the inspiratory time (tI) is also preset by the practitioner, so
that the ventilator cycles into expiration after a set period of time, rather than in
response to a drop in the patient’s respiratory flow. This function is useful when
the inspiratory phase is prolonged due to leak or lung mechanics (e.g. in COPD) as
it ensures adequate time is given over to the expiratory phase.
Timed ventilation Some ventilators have a timed mode. In this mode, the ventilator
is not triggered or cycled in response to patient effort. Instead, the practitioner
determines the respiratory rate and the ventilator is triggered into inspiration at
this preset time. The clinician also determines the tI and the ventilator cycles into
expiration after this time has elapsed. This mode of ventilation provides little inter-
action between the patient and the ventilator, and therefore, is often uncomfort-
able and poorly tolerated by non-sedated patients.
Proportional assist ventilation In PAV, rather than providing a fixed pressure
throughout inspiration regardless of the patient’s spontaneous effort, instead,
the ventilator generates pressure in proportion to the patient’s effort. The aim
of this ventilatory strategy is to improve patient–ventilator synchrony. In this
mode, positive pressure is delivered during inspiration in response to patient
flow in the form of flow assist and volume assist, such that an increase in patient
demand and effort leads to increased support from the ventilator, while the
patient controls the timing and size of each breath. Thus, with PAV, the ven-
tilator amplifies the patient’s inspiratory effort without any preselected target
volume or pressure. Studies have not shown PAV to have significant advantages
over more standard pressure-preset ventilation, but it may be useful in assist-
ing with weaning or during exercise. Importantly, as this mode relies upon the
patient’s respiratory effort, PAV should not be used in patients with depressed
ventilatory drive, particularly in those patients whose intrinsic respiratory drive
is depressed during sleep.
Volume-preset ventilation Sometimes referred to as volume-targeted ventilators,
volume-preset ventilators require selection of the VT or minute volume, rather than
the pressure, to be delivered. In this mode, the ventilator delivers a fixed VT and
inspiratory flow with each breath; however, there is poor leak compensation, such
that the delivered VT will fall in the presence of a leak. Figure 2 and table 1 highlight
some of the main differences between pressure- and volume-preset ventilation.
Note that in a volume-preset mode of ventilation, a spontaneous mode is not pos-
sible, and instead, a fixed tI must always be set.
In clinical practice, some of the differences between these volume- and pressure-
preset ventilation modes may be less important, and short-term comparisons of
the different modes have shown little difference in terms of their effects on VT,
minute volume, respiratory rate, arterial blood gas tensions, nocturnal oxygen
saturations and PtcCO2, and inspiratory effort. However, some studies have sug-
gested that some patients do seem to show clinical improvement using one mode
over the other and, if sufficient clinical improvement is not achieved in one mode,
switching to the alternate mode may be appropriate.
Volume-assured pressure support and adaptive servoventilation The term volume-
assured pressure support (VAPS) refers to various hybrid modes of ventilation in
which, in the simplest terms, the aim is to provide a minimum level of ventilation
a) b)
Pressure Pressure
Flow Flow
Volume Volume
Figure 2. Positive pressure ventilators. a) Pressure-preset ventilators: pressure is

constant; flow and volume differ with each breath. These include bilevel ventilators, which
have been designed specifically for NIV. b) Volume-preset ventilators: volume and flow
are constant; pressure is variable with each breath. Reproduced and modified from Skills-
based Simulator Training in Non-Invasive Ventilation. © ResMed Limited. All rights
reserved.
by varying the level of pressure support provided by the ventilator. Various VAPS
modes exist and may require the practitioner to set a target VT, minute ventilation
or target alveolar ventilation (i.e. total ventilation minus the ventilation estimated
to be “wasted” within the patient’s anatomical dead space). Different approaches
exist to enable the practitioner to establish this ventilatory target. In all cases, the
ventilator adjusts the pressure support, within practitioner-determined limits,
to meet the preset ventilatory target. Some VAPS modes also have the ability
to automatically adjust the EPAP (auto-EPAP) in order to eliminate upper airway
obstruction, which may be particularly helpful in patients with unstable upper
airways.
Table 1. Comparison of pressure- and volume-preset ventilators
Pressure-preset ventilators Volume-preset

(including bilevel) ventilators
Main clinician- Level of inspiratory and/or V T or minute volume to be
determined expiratory pressure to be delivered
setting delivered
VT delivery Delivered V T will fall if airway Delivers constant V T despite
resistance increases or lung changes in airway resistance
compliance decreases or lung compliance
Airway Pressure generated Generates increased pressures
pressures predetermined by clinician to achieve target V T: may
generated generate high peak airway
pressures
Leak Good leak compensation: Poor leak compensation:
compensation flow increases to compensate no increase in flow to
for leak and delivered V T is compensate for leak;
maintained therefore, delivered V T falls
in the presence of significant
leaks
Reproduced from Skills-based Simulator Training in Non-Invasive Ventilation. © ResMed Limited.
All rights reserved.

The aim of the VAPS mode is to adapt the delivered inspiratory pressure to
changes in respiratory impedance, such that more stable ventilation is achieved
while improving patient comfort and reducing side-effects such as gastric dis-
tension. It is not yet clear which patient groups may benefit from this ventilatory
strategy and studies in the acute setting are lacking. Nevertheless, the approach
may be useful in long-term ventilated patients who tolerate higher inspiratory
pressures less well, as the average pressure delivered may be lower using this
strategy.
ASV uses an algorithm involving the patient’s average respiratory rate, direction,
magnitude and rate of change of the patient’s respiratory flow, and a backup
respiratory rate in order to synchronise the level of pressure support delivered
during periods of apnoea/hypopnoea. During periods of apnoea/hypopnoea,
pressure support is delivered and matched to the patient’s usual breathing
pattern. If the apnoea/hypopnoea persists, backup breaths are supplied and,
when breathing resumes, the delivered pressure support is reduced to a minimum
level.
Ventilator circuits
Figure 3 shows some examples of different ventilator circuits. Most noninvasive
ventilators use a single-limb circuit through which inspiratory and expiratory
pressures are alternately delivered. Dual-limb circuits, with separate inspiratory
and expiratory lines, are more commonly used in critical care ventilators, as they
provide the ability to closely monitor a patient’s exhaled volumes and minimise
rebreathing through separation of the inspiratory and expiratory gases. In a dual-
limb circuit, active inspiratory and expiratory valves are incorporated into the ven-
tilator and circuitry.
In a single-limb circuit, exhaled carbon dioxide is exhausted either:
• through a passive exhalation port, an intentional, calibrated leak, either in the
circuit or in the interface itself (known as a vented mask); or
• via an active exhalation valve.
There is a theoretical risk of rebreathing with a passive exhalation port if the
patient’s expiratory flow exceeds the flow capacity of the port. This can be reduced
by siting the leak port in the mask rather than the tubing, by using a higher EPAP
and by titrating oxygen into the mask rather than the tubing. The minimum EPAP
setting on most bilevel ventilators is 4 cmH2O to reduce the risk of rebreathing of
carbon dioxide.
An active exhalation valve actively closes during the inspiratory phase to prevent
loss of VT. An active exhalation valve effectively reduces rebreathing, but may
increase expiratory resistance and is noisier than a traditional passive exhalation
port. With an active exhalation valve, it is not essential to set an EPAP.
Oxygen delivery
Critical care ventilators usually use an oxygen blender, which enables the accurate
titration and monitoring of the FIO2 applied, as well as the provision of higher con-
centrations of oxygen. Some ventilators outside the critical care setting have this
ability; however, it is usually required that oxygen be entrained into the circuit or

a)
Noninvasive interface
(vented mask or mouthpiece)
b)
(non-vented mask)
Single limb
c)
(non-vented mask)
Double limb
Figure 3. Circuits and ventilators. a) Leak circuit: single-limb circuit with a leak port in
the mask. b) Single- and c) double-limb valve circuits with an active exhalation valve. Leak
compensation is designed to compensate for unintentional leaks in the circuit or vented mask
that can result in patient–ventilator asynchrony.

interface. This means that the delivered oxygen concentration is variable, as it is

affected by:
• where the oxygen is entrained (higher concentrations are usually achieved
when oxygen is entrained into the mask)
• the ventilator settings
• type of exhalation port
• breathing pattern
• leaks
Alarms
Noninvasive ventilators have a range of alarms that can be set, including low pres-
sure alarms that can aid in the detection of disconnection or excessive leaks, and
high pressure alarms that are useful in volume-preset modes to warn of exces-
sively high pressures that may occur if a patient’s physiology changes such that it
is difficult to deliver the preset VT. Flow alarms may also be preset on some ven-
tilators and enable the ventilator to warn the clinician of worsening leak, airflow
obstruction or partially blocked ventilator tubing. As there are a wide variety of
indications for noninvasive respiratory support, in both acute and chronic settings,
it is important to set ventilator alarms that are appropriate for the patient’s condi-
tion. For example, a patient who is dependent on ventilation will have a greater
need for closer monitoring and alarms.
Further reading
• Hess DR (2010). Positive pressure ventilators. In: Elliott M, et al., eds. Non-invasive
Ventilation and Weaning: Principles and Practice. London, Hodder Arnold; pp. 13–23.
• Kacmarek RM (2011). Proportional assist ventilation and neurally adjusted venti-
latory assist. Respir Care; 56: 140–152.
• Mehta S, et al. (2011). Non-invasive ventilation. Am J Respir Crit Care Med; 163:
540–577.
• Nava S, et al. (2014). (Almost) everything you wanted to know about a ventilator.
In: Nava S, et al., eds. Non-invasive artificial ventilation. Milan, Springer; pp. 9–14.
Online resources
• Hare A, et al. Journey around the ventilator module. Skills-based Simulator Training
in Non-Invasive Ventilation. www.ers-education.org/e-learning/simulators.aspx
• Janssens JP, et al. (2014). Ventilators used for NIV. ERS Course Noninvasive ven-
tilation: basic concepts, Hanover 2014. www.ers-education.org/events/courses/
noninvasive-ventilation-basic-concepts,-hanover-2014.aspx
• Köhnlein T, (2014). Basic principles of ventilators – all you need to know. ERS Course
Noninvasive ventilation: basic concepts, Hanover 2014. www.ers-education.org/
events/courses/noninvasive-ventilation-basic-concepts,-hanover-2014.aspx

Matching mode and settings to

the patient: an introduction
Jan H. Storre and Jens C. Callegari
From a historical point of view, NIV was first established in clinical routine in the
1950s during the poliomyelitis epidemic. During that period, NPV was the first
mode of ventilation, which was introduced using the iron lung. A negative pres-
sure was applied to the chest during inspiration to augment alveolar ventilation,
followed by passive exhalation. Following this, further tools like tank respirators or
cuirass devices became available for the purpose of NPV. In the following decades,
NPV was used for NIV, but its influence was decreased by the establishment of
NIV using positive pressure ventilation via a nasal or oronasal mask, which was
labelled as noninvasive positive pressure ventilation. The terms used in the litera-
ture for different kinds of ventilation modes are often quite confusing and a clear
definition is lacking. In addition, this matter is promoted by different manufactur-
ers of ventilators, who label their own modes of ventilation for marketing reasons.
The most frequently used terms for NIV modes can be found in the list of abbre-
viations at the start of this book.
Key points
• Today, NIV is predominantly used in the pressure-
preset mode, which is better tolerated by the patients in
comparison with the volume-preset mode; the combined
hybrid mode of pressure-preset NIV with target volume does
not provide further benefits.
• Settings of pressures and backup rates have to be titrated
individually with regard to the patient’s tolerance and the
underlying pathology, as well as sleep disordered breathing.
• In chronic NIV, nasal masks are the interface most often used,
but oronasal masks should be applied for mouth breathers or
in case of significant mouth leaks.
• When patients with NIV are dependent on oxygen, circuits with
active exhalation valves are recommended since intentional
leaks have a negative impact on FIO2 at the interface.

Modes of ventilation
In general, there are different options for setting the ventilator when applying NIV.
The first option is to provide a fixed inspiratory volume with varying pressures
during inspiration, which is known as volume-preset NIV. The second option is
to provide a defined inspiratory pressure with varying volumes, which is labelled
pressure-preset NIV. The characteristic ventilation patterns for both of these are
given in figure 1. These two modes have been widely used with great success.
Nevertheless, both options have advantages and disadvantages, which are illus-
trated in table 1. For more details on ventilator modes, see the section entitled
“Basic principles of ventilators”.
Today, most patients are ventilated using pressure-preset ventilators (fig. 2).
However, in the early 1990s the majority were ventilated successfully using
volume-preset ventilators. The reasons for this development were as follows:
1) setting a pressure-preset mode was better tolerated by the patient (table 1)
2) the devices were smaller as well as cheaper
However, the latter argument cannot be maintained today, since many of the
blower-driven devices offer both options of ventilatory support due to technical
refinements. Even today, clear data showing benefits for pressure-preset NIV in
high-impact end-points such as improvements in gas exchange or sleep quality,
in health-related quality of life or mortality are not available. Nevertheless, due to
the higher comfort, pressure-preset NIV has become the mode most frequently
used in surveys and scientific studies.
Technical improvements have enabled manufacturers to combine the advantages
of pressure- and volume-preset NIV into one mode. These so-called hybrid modes,
containing pressure-preset NIV with target volume (such as AVAPS or IVAPS), were
introduced to clinical practice and investigated in several studies, but there is no
evidence that these hybrid modes show significant clinical benefits compared
with the classical modes of pressure- or volume-preset NIV. One benefit of these
new modes might be a titration benefit during NIV initiation. Conversely, it could
be argued that automatic modes result in less attention being given by the staff
to the important pathophysiology of individual patients, possibly resulting in a
negative impact on outcome. However, these advantages and disadvantages are
speculative.
In general, an individualised approach is necessary to apply the best mode and
setting for the patient. For the best setting of the ventilator, multiple pieces of
information are needed. First, the pathology of ventilatory failure and the pres-
ence of sleep-related breathing disorders have to be investigated. The settings
should be targeted with respect to this information. Standardised recommen-
dations cannot be given for the varying patient subgroups. Patient subgroups
such as neuromuscular patients might benefit from different modes during the
day and night. In these patients, daytime ventilation via a mouthpiece using the
volume-preset mode is needed in combination with a pressure-preset mode
via nasal or oronasal mask during the night. Consequently, devices with both
modes of ventilation (dual-mode ventilators) are clearly needed. Nevertheless,
it was reported that patients with thoracic cage or neuromuscular disorders
needed less IPAP (e.g. levels between 10 and 20 mbar) for positive outcomes,

20
a) 1600 Volume-preset mode b) 1600 Pressure-preset mode
1400 Inspiratory volume 1400

Expiratory volume
1200 1200
1000 1000
800 800
Volume mL
Volume mL
600 600
400 400
200 200
0 0
00:00 00:28 00:57 01:26 01:55 02:24 02:52 03:21 03:50 04:19 04:48 00:00 01:12 02:24 03:36 04:48 06:00 07:12
Time h Time h
c) 35 d) 35
30 30
25 25
20 20
15 15
Pressure mbar
Pressure mbar
10 10
5 5
Peak inspiratory pressure
0 0
00:00 00:28 00:57 01:26 01:55 02:24 02:52 03:21 03:50 04:19 04:48 00:00 01:12 02:24 03:36 04:48 06:00 07:12
Time h Time h
Figure 1. a, c) Volume-preset and b, d) pressure-preset NIV in a subject with thoracic restrictive disorder due to kyphoscoliosis. Ventilation patterns were
recorded using a pneumotachograph, which was placed between the mask and the expiratory valve.
ERS Practical Handbook Noninvasive Ventilation

Table 1. Advantages and disadvantages of pressure- and volume-preset ventilators
Aspect Pressure-preset Volume-preset

Constant inspiratory volume - ++
Constant inspiratory pressure ++ -
Improvement in sleep quality + +
Improvement in gas exchange + +
Improvement of quality of life + +
Leak compensation ++ −
Gastrointestinal side-effects + −
Comfort + −
Advantages range from + to ++ and disadvantages are shown as -. Reproduced from Storre et al.
(2008), with permission from the publisher.
compared with patients with lung/airway diseases such as COPD. As pointed out
in more detail in the section entitled “Chronic NIV in COPD”, a high-intensity
mode of NIV is needed in COPD subjects. This mode is characterised by higher
IPAP levels (mostly >20 mbar) and a controlled form of NIV with higher settings
of breathing frequency, aiming for normocapnia, and is needed to improve long-
term survival, lung function, exercise tolerance and health-related quality of
life in COPD patients. In line with this, the need for higher IPAP levels is often
reported in patients suffering from OHS when targeting to approach normocapnia.
As mentioned, this has to be titrated on the individual patient and their tolerance.
Pressure-preset Volume-preset
100
80
Ventilation %
60
40
20
0
All
Austria
Belgium
Denmark
Finland
France
Germany
Greece
Ireland
Italy
Netherlands
Norway
Poland
Portugal
Spain
Sweden
UK
Figure 2. Percentages of pressure- and volume-preset positive pressure ventilators used for
home mechanical ventilation in different countries. n=21 526. Reproduced from Lloyd-Owen
et al. (2005), with permission from the publisher.
The occurrence of OSA as a comorbidity suggests a need for higher EPAP to sta-

bilise the upper airways. However, with a higher EPAP, an even higher IPAP is
needed, in order to apply the same level of inspiratory pressure support to aug-
ment the respiratory system. This has to be remembered during treatment to
ensure a sufficient alveolar ventilation and reduction of hypercapnic carbon diox-
ide tension levels, which is one of the major goals of NIV.
Furthermore, NIV can be applied using an assist or controlled mode or by com-
bining both settings in the assist-control mode. All these options have been
used successfully. As mentioned, especially in COPD patients, high-intensity NIV
including a high backup rate and the target of achieving controlled ventilation
has been reported to give positive outcomes. These positive outcomes could not
be achieved with low-intensity NIV in the assist mode. A controlled mode of NIV
(assisted pressure-controlled ventilation (APCV or ASSPCV)) reduces the workload
of respiratory muscles more effectively than is possible using the assist mode.
However, the assist mode (PSV) is primarily used in the acute setting. Irrespective
of the chosen mode, the interaction between the patient and the ventilator is of
high impact and the patient’s tolerance to either setting has to be monitored and
reflected. A harmony between the patient and the ventilator is needed and “auto-
triggering” (fig. 3) or “fighting against the ventilator” have to be monitored and
minimised.
Interfaces and circuits for NIV
The interface is crucial for applying NIV successfully. Interfaces are discussed further
in the section entitled “Choosing the interface”. The most common interfaces are
nasal or oronasal masks, which both show advantages and disadvantages (table 2).
In acute care, where patients suffer from severe dyspnoea, mouth breathing is
common. Thus, an oronasal mask covering mouth and nose is the better interface
Patient–ventilator synchrony Patient–ventilator asynchrony: “auto-triggering”

Pressure by
polygraphy
ventilator movement
Pressure by Thoracic
Figure 3. Patient–ventilator asynchrony during mechanical ventilation after five regular

breaths. Periodic breathing patterns by the subject (thoracic movements) are not detected by
the ventilator and breathing frequency (pressure curves) is increased without patient effort
(so-called “auto-triggering”). Reproduced and modified from Storre et al. (2014a), with
permission from the publisher.
Table 2. Comparison of oronasal# and nasal masks
Aspect Oronasal mask Nasal mask

Mouth leak No Yes
Mouth breathing Possible Decreases NIV quality
Dead space Higher Low
First choice interface Acute care Chronic care
Communication Reduced Possible
Eating and drinking No Possible
Expectoration No Possible
Risk of aspiration Elevated Reduced
Risk of aerophagia Elevated Reduced
Claustrophobia Elevated Reduced
Comfort Lower Higher
#:sometimes also known as facial or face masks. Reproduced and modified from Storre et al.
(2008), with permission from the publisher.
for these patients, providing sufficient NIV within a short period of time, which is
essential for a good outcome. In addition to oronasal masks, total face masks can
be used, which cover the complete face. Total face masks can be used to prevent
pressure marks on the bridge of the nose. However, this is mostly at the cost of
an increase in the amount of leaks, since sealing the interface is more challeng-
ing with a larger interface. If an oronasal mask is not tolerated, treatment can also
be performed in acute ventilatory failure using a nasal mask. In chronic care and
home mechanical NIV, a nasal mask is the preferred interface, and occasionally
nasal pillows are used (table 2 and fig. 4). Surprisingly, despite the great importance
of correct interface choice, there is little scientific evidence and few studies have
focused on the choice of interface or systematically studied differences regarding
the quality of NIV. Thus, the published recommendations are expert opinions rather
than evidence-based medicine. Today, a broad variety of different commercial inter-
faces exists, which enhances the chance of making an appropriate choice for most
patients. Nevertheless, due to varying facial anatomies, there are still some patients
who need individually modelled masks. In patients with neuromuscular disorders
and permanent daily application of NIV, mouthpiece ventilation is reported to be a
successful tool.
Besides the different options when choosing the right interface, there are varying
options for the selection of the ventilatory circuits between the patient and the
ventilator. In chronic care, single-limb circuits are most often used, which can be
divided into circuits using an integrated active exhalation valve and those with
a passive exhalation valve (or intentional leaks). These intentional leaks can be
integrated either into the circuit itself or into the interface, and nowadays these
are the circuits most often used in chronic care. This trend has been influenced
more by the manufacturers than scientific evidence. Most devices manufactured
in the past decade have become smaller and cheaper and have used the passive
a) Nasal mask Facial mask Tracheostomy

100
80
Lung HMV users %
60
40
20
b) 100
80
Thoracic HMV users %
60
40
20
c) 100
Neuromuscular HMV users %
80
60
40
20
0
All
Austria
Belgium
Denmark
Finland
France
Germany
Greece
Ireland
Italy
Netherlands
Norway
Poland
Portugal
Spain
Sweden
UK
Figure 4. Proportions of interfaces used for home mechanical ventilation (HMV) in different
countries for users with a) lung, b) thoracic and c) neuromuscular diseases. No data were
available for Polish lung disease users. n=21 526. Reproduced from Lloyd-Owen et al. (2005),
with permission from the publisher.
exhalation valve circuits only. In these passive exhalation valve circuits, leakage
is always present during expiration as well as inspiration, which is necessary for
the elimination of carbon dioxide. However, this amount of leakage can be dis-
advantageous when patients are dependent on supplemental oxygen in addition
to mechanical ventilation. In these cases, the low-pressure oxygen flow used in
home care is added to the circuit and is partly lost due the large amount of inten-
tional leakage. As a consequence, the FIO2 is reported to be substantially lower
when compared with a circuit with an active exhalation valve, where the leak-
age in the circuit is only present during expiration. Consequently, using a circuit
with an active exhalation valve is suggested when supplemental oxygen has to be
added to NIV. For more details, see the section entitled “Supplemental oxygen and
humidification”.
Further reading
• Bayarassou AH, et al. (2013). Common mistakes leading to NIV failure. Minerva
Pneumol; 52: 39–53.
• Dreher M, et al. (2010). High-intensity versus low-intensity non-invasive ven-
tilation in patients with stable hypercapnic COPD: a randomised crossover trial.
Thorax; 65: 303–308.
• Elliott MW (2004). The interface: crucial for successful noninvasive ventilation. Eur
Respir J; 23: 7–8.
ment of severe stable chronic obstructive pulmonary disease: a prospective, multi-
• Lloyd-Owen SJ, et al. (2005). Patterns of home mechanical ventilation use in
Europe: results from the Eurovent survey. Eur Respir J; 25: 1025–1031.
• Mehta S, et al. (2001). Noninvasive ventilation. Am J Respir Crit Care Med; 163:
540–577.
• Storre JH, et al. (2008). Noninvasive mechanical ventilation in chronic respiratory
failure: ventilators and interfaces. In: Muir JF, et al., eds. Noninvasive Ventilation.
2nd Edn. ERS Monogr; 41: 319–337.
• Storre JH, et al. (2014a). Monitoring des Beatmungspatienten [Monitoring of
patients receiving mechanical ventilation]. Pneumologie; 68: 532–541.
• Storre JH, et al. (2014b). Oxygen supplementation in noninvasive home mechan
ical ventilation: the crucial roles of CO2 exhalation systems and leakages. Respir
Care; 59: 113–120.
• Windisch W, et al. (2012). Target volume settings for home mechanical ventila-
tion: great progress or just a gadget? Thorax; 67: 663–665.

Choosing the interface
Anne-Kathrin Brill
Interfaces connect the ventilator via circuit tubing to the patient and thereby allow
the delivery of pressurised air into the upper airways and subsequently into the
lungs. Choosing an appropriate interface is essential for successful NIV. The inter-
face has to provide a good seal and needs to be tolerated by the patient at the
same time.
Interfaces can broadly be classified into six categories: nasal masks, nasal pillows,
oronasal masks, total face or full face masks, oral masks or mouthpieces, and the
helmet (fig. 1). With the exception of mouthpieces and the helmet, most inter-
faces are made of a soft cushion and a mask frame. They are normally secured
with a head frame, headgear, or straps with velcro, clips or hooks.
Key points
• Interface choice depends on clinical setting, equipment
availability, the patient’s individual characteristics and safety
considerations. NIV services should have a range of different
masks and accessories.
• Fit a correctly sized mask and headgear to minimise leak and
increase comfort. Ensure that the interface complies with the
ventilator and circuit tubing.
• All interfaces can be used to provide NIV successfully, but in
the acute setting interfaces covering the mouth and nose are
advantageous.
• If the patient cannot adapt to NIV, changing the interface
can be helpful but should not delay intubation in critically ill
patients if NIV is clearly failing.
• Patients using NIV for >12 h per day should have a spare mask
and can benefit from having two alternative interface types to
alternate with.

Oronasal mask# Full face mask#
Covers mouth and nose Also called total face mask, cephalic or
Special subtype: hybrid masks (a combination integral mask
of nasal pillows and an oral mask) Covers mouth, nose and eyes and seals
With or without forehead spacer around the perimeter of the face
Nasal mask# Nasal pillows

Covers the whole nose but not the mouth Subtype of nasal mask, also nasal plugs or

With or without forehead spacer nasal slings
Applied externally to the nares
Oral masks and mouthpieces Helmet

Placed between the patient’s lips Transparent hood with collar
Mouthpieces have various degrees of Covers the whole head and all or part of the
flexion and are held in place by a lip seal neck, no contact with the head
or the teeth Has at least two ports
Oral masks can also have headgear as a Most helmets have an anti-asphyxia valve
securing system
Figure 1. The six main interface types for NIV. #: available as a vented or non-vented version. Image of the human head by Patrick J. Lynch reproduced
from Wikimedia Commons under CC BY 2.5 licence.
27
Today, we can choose from a broad range of commercially available interfaces

and the different types, shapes, materials, sizes and accessories allow a suitable
interface to be found for nearly every patient. Custom-made masks are another
option and can be helpful for patients with a prominent facial anatomy, but in
ARF there is usually not enough time for their realisation. Many aspects have to
be considered when the interface is initially chosen (table 1). These include the
overall setting of NIV, availability of interfaces, and technical and patient-related
factors, as well as safety aspects, experience of staff and costs. All interfaces can
be used to treat respiratory failure and there is not enough evidence in the litera-
ture to provide strong general recommendations towards a single type of interface
in terms of tolerance or efficacy. Each interface type has advantages and disadvan-
tages (table 2), and the most suitable interface will depend on the situation and
the patient’s individual characteristics.
The first aspects to consider when choosing the interface are technical compati
bility, i.e. which ventilator and circuit is used, safety aspects, size of the interface
and the overall setting in which NIV is started (table 3). NIV masks are provided
as vented or non-vented masks. Vented masks have integrated holes in the mask
frame or a swivel elbow to remove carbon dioxide and to prevent rebreathing.
Non-vented masks do not have an opening and need a separate option for carbon
dioxide removal from the circuit. It is important to be aware of which interface and
circuit is used. Therefore, prior to the start of NIV the method of carbon dioxide
removal should be identified, its patency assured, and compatibility of circuit and
interface checked.
Table 1. Factors influencing interface selection in NIV
Setting Technical Patient-related Others
Respiratory failure Availability Age Safety

Acute Vomiting
Circuit used Facial anatomy
Chronic Quick-release
Vented Recent facial
strap
Location of NIV Non-vented surgery
Anti-asphyxia
ICU/HDU Nasal patency
CO2 rebreathing valve#
Emergency Dentures
Location of
room Motor function Experience of
exhalation
Post-operative staff
Interface/
recovery room Ventilator used
headgear size Airborne disease
Respiratory ward Leak
TB
General ward compensation Skin conditions
Influenza
Ventilator
Estimated length Claustrophobia
settings Accessories
of use Mouth breathing Dressings
Short-term Type of headgear
Chin strap
Long-term Eating/speaking
Tube adaptors
Expectoration Lining
The main factors that may influence interface selection in NIV are listed, but the list is not
exhaustive. HDU: high-dependency unit. #: present in vented masks and most helmets.

Table 2. Characteristics of the different interfaces
Oronasal Total Helmet Nasal Nasal Mouth

mask face mask prongs pieces
mask
Acute setting • • • ⚬ ⚬ ⚬
Use outside HDU/ICU • • • • •
Chronic setting • • • • •
Less claustrophobic ⚬ ⚬ • • •
More likely to have • • • •
leaks in the acute
setting
Nasal patency • •
required
Coughing and • •
expectoration
is easier
Useful for prominent • • • •
facial anatomy
High level of noise •
No pressure on the • • • •
nasal bridge
High gas flow required •
Chance of eye • ⚬ ⚬ •
irritation
Speaking is easier • • •
HDU: high-dependency unit; closed circles: applicable to the interface; open circles: an
alternative, but less common or less frequent option. Reproduced and modified from Brill (2014)
Other safety considerations include:

• the risk of asphyxiation in case of vomiting
• the risk of asphyxiation in case of ventilator malfunction
• the patient’s level of consciousness and ability to remove the mask
autonomously
• potential disease transmission
With nasal masks and the helmet the risk in case of vomiting or ventilator mal-
function is usually lower, but patients using masks covering their mouth and nose
should either be able to remove the mask themselves, be able to call for immedi-
ate help, or receive closer monitoring. Vented masks covering the mouth and nose
and most helmets are usually equipped with an anti-asphyxiation valve that opens
in case of unexpected ventilator failure. These valves should never be obstructed
intentionally.
Table 3. Vented and non-vented interfaces and suitable respiratory circuits
Interface Suitable respiratory circuits

Vented
The interface has one or Single-limb “vented” or “intentional leak”
multiple little holes in the circuit: CO2 removal occurs through the
frame or on the swivel elbow orifices sited within the interface or the
allowing passive CO2 removal swivel elbow
Non-vented
The interface is completely Double-limb “non-vented” circuit: CO2
closed removal occurs through the ventilator
Requires an option for CO2 Single-limb “non-vented” circuit: CO2 removal
removal within the circuit or occurs through an active non-rebreathing
ventilator expiratory valve at the distal end of the
inspiratory circuit or at the end of a short
expiratory limb
Single-limb “vented” or “intentional leak”
circuit: CO2 removal occurs through a
whisper swivel or a hole orifice at the end
of the circuit, or a plateau exhalation valve
placed between circuit and mask
NIV is a droplet-generating procedure that may expose healthcare workers who

stand close to patients on NIV to potentially infectious droplets. If NIV has to be
applied to patients with highly transmittable airborne disease, the use of a non-
vented mask covering the nose and mouth within a non-vented tubing circuit and
filtering of the exhaled gases can reduce environmental spread.
Interfaces are available as disposable and reusable versions. Disposable inter-
faces are mainly used for shorter episodes of NIV in the acute setting (e.g. in
emergency rooms, ICUs or post-operative recovery rooms). A change to a mask
for long-term use should be considered if NIV has to be prolonged, as masks for
long-term NIV:
• are often easier for patients to handle themselves
• can be used for longer periods of time
• often allow more equal distribution of pressure over the face
Interface strategies differ in acute and chronic settings (fig. 2). In ARF the priority
is to quickly achieve efficient ventilation with the patient being able to tolerate the
interface. In this situation most patients are in respiratory distress and breathe at
a high respiratory rate. Patients often start mouth breathing to bypass the higher
resistance of the nose. This can result in mouth leaks and compromise the efficacy
of ventilation if nasal masks are used. Only a few studies have compared different
interfaces in ARF, but overall masks covering the nose and mouth were associated
with a quicker improvement in blood gases and less NIV failure, while nasal masks
were often better tolerated in the longer term (Kwok et al., 2003; Girault et al.,
2009; Navalesi et al., 2000). Therefore, the recommendation is to use interfaces
Check compatibility of interface, ventilator and circuit. Identify where exhalation occurs. Choose correct headgear.
ARF CRF
Start with nasal mask

Start with oronasal mask (or interface of patient’s choice; consider patient’s
abilities, needs and comorbidities)
Check for non-intentional leaks, asynchrony, Check for non-intentional leaks, asynchrony,
ventilator settings, circuit; adjust if necessary ventilator settings, circuit; adjust if necessary

Adapted to NIV? Yes Yes Adapted to NIV?
No Check for side-effects: No
Pressure marks
Patient still suitable for NIV? Leak into the eyes Try other types of interface and accessories
Comfort and sleep
Yes No Adjust if necessary
Try alternative Consider NIV failure Patient improvement Consider custom-made mask
interface
NIV success
Consider switching to a smaller interface, nightly use only Consider additional alternative interface
Account for patient’s needs and preferences e.g. additional mouthpiece in NMD
Account for patient’s needs and preferences
Figure 2. Interface strategies for NIV in adult patients with ARF and chronic respiratory failure (CRF). NMD: neuromuscular disease.
31
covering the nose and mouth as the first choice in ARF and to try switching to a
smaller interface once the patient is more stable. Mainly, oronasal masks are used
in acute NIV. Full face masks or helmets can also be used in selected patients, but
they are often a second choice because of higher costs, and the necessity of having
experienced staff and a higher level of monitoring to use the helmet. In acute NIV
there is often not very much time to try different masks and more than one change
of interface type is often not possible. Slightly more stable patients might benefit
from a second change of interface if it can help to increase tolerance and efficacy of
NIV and thereby prevent intubation. However, trying different interfaces should not
delay intubation in critically ill patients if NIV is clearly failing despite best efforts.
In the chronic setting there is more time to choose the most appropriate interface
for the patient, leading to more individual choices. Patients can usually try dif-
ferent interfaces and with the exception of the helmet all other interface types
are used in long-term NIV for chronic respiratory failure. Smaller masks are often
used in this setting and it is appropriate to start NIV with a nasal mask unless:
• the nose is not patent
• the patient cannot keep the mouth closed (e.g. due to muscular weakness)
• the patient wishes to use a mask covering mouth and nose
In addition to a good mask fit, it is also important that the interface meets the
patient’s needs and does not interfere too much with the patient’s daily routine
and sleeping habits. Masks should be easy to apply and remove by the patient or
caregivers. Quick-release straps or headgear that can be placed with one hand
can be an option if the patient has impaired motor function. In addition, some
masks allow for wearing glasses more easily than others. Patients who need
NIV for >12 h per day or who are at risk of pressure sores may benefit from hav-
ing different interface types to alternate between. This allows for intermittent
pressure relief of the skin and can give patients more flexibility with eating and
speaking.
Independent of the clinical situation and the selected interface type, interfaces
should always be fitted in the correct size and with a correctly sized and suitable
securing system, to improve fit and comfort and to reduce unpleasant side-effects
such as air-leaks or pressure marks. The use of sizing gauges (usually provided by
the manufacturer) is encouraged. Masks should not overlay the eyes or lips and,
at least in the chronic setting, masks should be fitted in the situation in which the
interface will be used (i.e. with the patient seated for daytime use, or lying down
supine in his/her favourite sleeping position, where a cushion might be able to
support or interfere with mask fit).
Large air-leaks affect the efficacy of NIV, promote patient–ventilator asynchronies

and should be corrected immediately by refitting or changing the interface. Most
ventilators designed for NIV can compensate for small air-leaks and small leaks
can mostly be tolerated if they do not irritate the patient (e.g. eyes or noise) or
disturb sleep. Over-tightening of the mask straps should be avoided as it will cause
discomfort and might lead to skin damage. Leaks can be reduced with the help of
accessories, for example chin straps in the case of mouth leaks or use of lining
and dermal dressings to improve the seal. The latter are also useful to protect the
skin by reducing pressure and friction during NIV. For patients with prominent
anatomy or at risk for the development of pressure ulcers, masks with softer cush-
ions can be used. An adjustable forehead spacer or multipoint headgears can also
help to distribute the pressure more evenly over the face and reduce pressure on
the bridge of the nose. Regular reassessments of the mask fit and side-effects are
essential to provide comfortable and successful NIV.
Breathing patterns change with age and recommendations for adults cannot be
translated directly into paediatric NIV. Interface choice and special considerations
for children are described in the section entitled “Acute NIV in children, including
ventilator and interface choice”.
Further reading
• Brill AK (2014). How to avoid interface problems in acute noninvasive ventilation.
Breathe; 10: 230–242.
• Carron M, et al. (2013). Complications of non-invasive ventilation techniques:
a comprehensive qualitative review of randomized trials. Br J Anaesth; 110:
896–914.
• Crimi C, et al. (2010). A European survey of noninvasive ventilation practices. Eur
Respir J; 36: 362–369.
• Fraticelli AT, et al. (2009) Physiological effects of different interfaces during non
invasive ventilation for acute respiratory failure. Crit Care Med; 37: 939–945.
• Garuti G, et al. (2014). Open circuit mouthpiece ventilation: concise clinical review.
Rev Port Pneumol; 20: 211–218.
• Girault C, et al. (2009). Interface strategy during noninvasive positive pressure ven-
tilation for hypercapnic acute respiratory failure. Crit Care Med; 37: 124–131.
• Kwok H, et al. (2003). Controlled trial of oronasal versus nasal mask ventilation in
the treatment of acute respiratory failure. Crit Care Med; 31: 468–473.
• Nava S (2013). Behind a mask: tricks, pitfalls, and prejudices for noninvasive ven-
tilation. Respir Care; 58: 1367–1376.
• Nava S, et al. (2009). Interfaces and humidification for noninvasive mechanical
ventilation. Respir Care; 54: 71–84.
• Navalesi P, et al. (2000). Physiologic evaluation of noninvasive mechanical ven-
tilation delivered with three types of masks in patients with chronic hypercapnic
respiratory failure. Crit Care Med; 28: 1785–1790.
• Schettino GP, et al. (2003). Position of exhalation port and mask design affect
CO2 rebreathing during noninvasive positive pressure ventilation. Crit Care Med;
31: 2178–2182.
• Sferrazza Papa GF, et al. (2012). Recent advances in interfaces for non-invasive
ventilation: from bench studies to practical issues. Minerva Anestesiol; 78:
1146–1153.

Online resources
• Escarrabill J. How do I choose the interface? ERS Course Noninvasive ventila-
tion: basic concepts, Hanover 2014. www.ers-education.org/events/courses/
noninvasive-ventilation-basic-concepts,-hanover-2014.aspx
• Escarrabill J, et al. Interfaces used for NIV. ERS Course Noninvasive ventilation: basic
concepts, Hanover 2014. www.ers-education.org/events/courses/noninvasive-
ventilation-basic-concepts,-hanover-2014.aspx
• Wijkstra P. Air leakage during NIV: How important, how to avoid, how to handle?
ERS Course Noninvasive ventilation: basic concepts, Hanover 2014. www.
ers-education.org/events/courses/noninvasive-ventilation-basic-concepts,-
hanover-2014.aspx

Supplemental oxygen and humidification
Sundeep Kaul and Anita K. Simonds
Supplemental oxygen
Patients receiving NIV for either ARF or chronic respiratory failure, irrespective of
aetiology, often require additional oxygen therapy to increase PaO2 and, therefore,
improve tissue oxygen delivery. The introduction of ventilator support via NIV
may itself correct the hypoxia; however, in many cases, supplemental oxygen is
required. Some ventilators that deliver NIV are equipped with an oxygen blender,
thereby allowing the operator to dial in the FIO2 and titrate according to oxygen
saturations (SpO2) or PaO2. However, portable bilevel pressure ventilators com-
monly used to deliver NIV do not typically have an oxygen control. These NIV
ventilators entrain room air and, on many machines, oxygen enrichment requires
supplemental oxygen to be fed proximally into the circuit or directly into the
mask.
Indications Supplemental oxygen is indicated in type 1 or 2 ARF and chronic res-

piratory failure requiring NIV.
Key points
• Supplemental oxygen is often required by patients receiving NIV.
• Adding oxygen via the mask yields greater oxygen delivery to
the patient rather than altering inspiratory pressures.
• Achieving higher oxygen tensions without reducing patient
comfort is optimal.
• Humidification during NIV is a standard of care.
• Reduced humidification can lead to increased airway
resistance and mucus plugging.
• Appropriate humidification can reduce patient discomfort and
improve NIV tolerance.
• Different humidification systems are available and each should
be considered according to the clinical scenario.

Set-up and practical tips

• Ensure the oxygen saturation monitor has good contact with patient’s fingertip
and that the waveform is reliable (if the analogue signal is available).
• Connect one end of the oxygen tube to the oxygen source (e.g. wall outlet or
cylinder).
• Connect the other end of oxygen tube to either the oxygen connector in the
ventilator circuit or directly to the mask via an inlet or nipple.
• Increase the oxygen flow rate (at the source) until the required SpO2 is achieved
(typically above 92%).
• Perform arterial blood gas analysis to confirm the readings.
• Ensure that the ventilator is switched on and the oxygen supply is connected,
as there have been fatal incidents due to unintended disconnections or failure
to switch on the ventilator.
The concentration of oxygen delivered to the patient depends on a number of
factors, which include ventilator pressures and flow rates, mask leakage, the
site of the exhalation port, oxygen flow rate, the site of oxygen entrainment, and
patient’s inspiratory flow rate.
If the clinical situation requires the amount of inspired oxygen to be increased
swiftly, then manoeuvres available to the operator, in addition to increasing the
oxygen flow rate, include optimising the site of oxygen entrainment and altering
the inspiratory/expiratory pressures. Different entrainment sites are associated
with the delivery of different inspired oxygen concentrations (fig. 1). Entraining
supplemental oxygen directly through the mask yields higher SaO2 than adding
it to other points in the ventilator circuit. Lower inspiratory pressures tend to
be associated with higher oxygen concentrations but this may reduce the VT
delivered. High expiratory pressures may reduce tolerability.
Humidification
The normal, unsupported human airways are capable of delivering a VT of air at
100% relative humidity, at core temperature, when it reaches the alveolar surface
30 Site 1
Site 2
25 ▲ Site 3
PO2 mmHg
20 Site 4
15
10 ▲
5 ▲
0 ▲
0 2 4 6 8 10 12 14 16
Oxygen flow rate L·min–1
Figure 1. Effect of increasing flow rates on oxygen tension (PO2 ) at different entrainment
sites at IPAP 10 cmH2O and EPAP 4 cmH2O. Supplemental oxygen was added at four sites
along the ventilatory circuit. Site 1: between mask and exhalation port. Site 2: just distal to
exhalation port. Site 3: at ventilator outlet, between ventilator circuit and ventilator. Site 4:
directly into the mask via an inlet. Data from Kaul et al. (2006).
for optimal gas exchange. Humidification of inspired gas is typically provided by

evaporation of water from tracheobronchial secretions. Evaporation results in air-
way cooling, which is partially compensated for by heat and water provided by the
bronchial blood flow. During expiration, some of the moisture is recovered as the
air flows back over the mucosa on its way out of the airways.
Patients receiving NIV require sufficient humidification of inspired air for the
same reasons as those who breathe spontaneously. Humidifying inspired gas
to 100% relative humidity precludes heat and moisture exchange with the
mucosa, and thereby excludes the need for recovery of heat and moisture from
expired gas. NIV delivers air at higher inspiratory flow rates than the patient’s
own respiratory flow rates, which can overwhelm the capacity of the patient’s
humidification system. This, in turn, can lead to airway dryness, altered airway
mechanics, inspissated secretions and increased work of breathing, and predis-
pose to NIV failure.
Humidification during NIV is an accepted standard of care. The aim of humidifica-
tion during NIV is to reduce the risk of:
• increased airway resistance
• mucus plugging of the airways
• ciliary dysmotility
• epithelial desquamation
Maintenance of normal mucosal function also sustains airway patency and lung
compliance, thus preserving lung mechanics. Correct application of a humidifi-
cation system may enhance patient comfort and reduce the work of breathing,
thereby improving NIV compliance and success.
When optimising the humidification process during NIV, it is important to take the
following factors into account.
Types of humidification systems There are two methods of providing humidifica-
tion. Firstly, via heated humidifiers that utilise an external power source and water
supply, and secondly, using a heat and moisture exchanger (HME) system, which
recycles the patient’s own moisture and heat.
Use the appropriate humidification system according to the patient’s needs. The
pros and cons of each system are highlighted in table 1.
Leaks Air leaks (via the mask or mouth) result in unidirectional nasal air flow and,
therefore, reduced mucosal recovery of moisture during the expiratory phase,
leading to dry nasal mucosa and increased nasal airway resistance. This, in turn,
reduces NIV compliance and efficiency. Performance of HME is also reduced, as
expired air is less moist.
One should institute measures to reduce leakage and consider heated humidifica-
tion systems.
Interfaces Nasal masks are associated with mouth leaks and, therefore, reduced
humidification and its consequences, as described above. Within a helmet inter-
face, there is a pool of exhaled gas that will contribute to humidification.
Consider substituting a nasal for an oronasal mask, especially in mouth-breathers,
those with secretion management issues and in the acute setting. If using a
Table 1. Advantages and disadvantages of HME filter and heated humidifier (HH) s ystems
during NIV
System Advantages Disadvantages

HME Cost effective Increased dead space
Extended use in the ICU Reduced efficacy in case of leaks
Eliminates circuit condensation Efficacy depends on body and
(hygroscopic models are environmental temperatures
recommended) May lead to an increase in airway
A booster system applied to a resistance in patients with
hydrophobic HME may preserve heavy secretions and respiratory
AH capacity when incoming gases tract bleeding
are delivered at a temperature
<26°C and at high flow
Does not need electricity
HH Less work of breathing than with Less efficacy with high
HME environmental temperature
Limited or no effect on dead space Needs electricity
ventilation so CO2 retention is Performance of different devices
minimal varies
Achieves RH and AH values
sufficient for gas conditioning
Clinically effective, especially in
patients with mild-to-severe
hypercapnic ARF
AH: absolute humidity; RH: relative humidity. Reproduced from Esquinas Rodriguez et al. (2012)
helmet interface, then consider altering the humidification (i.e. it may require
lower external humidification than a nasal or full face interface).
Ventilator Double-circuit ICU ventilators are associated with lower humidification
levels than single-circuit home NIV machines. Home ventilators entrain room
air, unlike ICU ventilators, which use dry gas from an outlet. In the ICU setting, if
higher oxygen concentrations are required, then there is an increased susceptibil-
ity to the consequences of low humidification, as medical oxygen is anhydrous.
Higher levels of humidification are required in hospital settings and with ICU ven-
tilators, especially if higher oxygen levels are required.
High-flow oxygen therapy systems
A variety of high-flow humidified oxygen devices have been developed and,
recently, compared with NIV in a range of clinical situations. An example of a
high-flow humidified oxygen system is shown in figure 2. These devices combine
a source of oxygen together with compressed air, with the gas mixture heated to
∼37°C and delivered at almost 100% humidification. The high flow tends to gen-
erate a low level of CPAP. FIO2 can be adjusted and can be set to target a specific
SpO2 value, e.g. ≥92%.
There are a variety of anecdotal reports on the success of high-flow oxygen ther-
apy, e.g. in paediatric sleep disordered breathing and exacerbations of COPD. It has
a) b)
Figure 2. a) Example of a high-flow humidified oxygen therapy system (Optiflow delivery
system with AIRVO humidifier; Fisher and Paykel Healthcare, Maidenhead, UK). b) A high-
flow humidified oxygen therapy system in use.
been advocated for use in patients with hypercapnic excacerbations of COPD who
cannot tolerate NIV because the nasal cannulae system, through which high-flow
humidified oxygen therapy is delivered, may be easier for some patients to use
than nasal masks. To date, there have been no RCTs of high-flow oxygen therapy
versus NIV in acute COPD exacerbations.
However, Frat et al. (2015) performed an RCT comparing high-flow oxygen
delivered via nasal cannulae with standard oxygen therapy and NIV in acute
hypoxaemic, non-hypercapnic respiratory failure patients in an ICU setting. The
participants had no history of chronic respiratory failure. The primary outcome was
intubation rate at day 28, and secondary end-points included all-cause mortality
in the ICU and at 90 days, and the number of ventilator-free days at 1 month. The
trial showed no significant difference in intubation rates between the groups, but
28-day mortality was lower in the high-flow oxygen therapy group.
Stéphan et al. (2015) carried out a non-inferiority trial of high-flow oxygen ther-
apy versus NIV in post-cardiothoracic surgery patients who had developed acute
hypoxaemic respiratory failure or were deemed at risk of doing so. Recruits
included those who had undergone cardiothoracic procedures, of which coronary
artery bypass, valvular repair and pulmonary thromboendarterectomy were the
most common interventions. No significant differences were found in terms of
ICU mortality (23 (5.5%) patients using NIV and 28 (6.8%) using high-flow nasal
oxygen therapy; p = 0.66, absolute difference 1.2%, 95% CI −2.3–4.8%), indicat-
ing that high-flow oxygen therapy is not inferior to bilevel pressure support NIV.
In both these studies, high-flow oxygen therapy tended to be better tolerated than
NIV and, in the latter trial, resulted in fewer nasal pressure sores.
Take home message Both of the recent trials described here were carried out in
patients with acute hypoxaemic respiratory failure (discussed in the sections enti-
tled “The patient with acute hypoxaemic respiratory failure excluding pulmonary
oedema” and “The patient with acute hypoxaemic respiratory failure and cardio-
genic pulmonary oedema”) and, notably, hypercapnic patients were excluded.
High-flow oxygen therapy may well have a role in this acute hypoxaemic group,
but it is clear that a spectrum of noninvasive respiratory support modes should be
available.
NIV remains the treatment of choice in hypercapnic excerbations of COPD. However,
further trials would be welcome in those with acute hypoxaemic exacerbations of
COPD, or with interstitial lung disease and acute hypoxaemic decompensation,
especially where barotrauma is a concern. A further role for consideration is the
application of high-flow oxygen therapy in acute COPD patients who fail NIV but,
here, success might be more likely in those with primary NIV failure due to mask
or high pressure intolerance, rather than secondary failure due to progression of
hypercapnic respiratory failure. In all situations, close monitoring is vital.
Further reading
• Branson R, et al. (2010). Is humidification always necessary during noninvasive
ventilation in the hospital? Respir Care; 55: 209–216.
• Esquinas Rodriguez AM, et al. (2012). Humidifiers during non-invasive ventilation –
key topics and practical implications. Crit Care; 16: 203.
• Frat J-P, et al. (2015) High-flow oxygen therapy through nasal cannula in acute
hypoxaemic respiratory failure. N Engl J Med 372: 2185–2196.
• Kaul S, et al. (2006). The effect of entrainment site and inspiratory pressure
on the delivery of oxygen therapy during non-invasive mechanical ventilation
(NIMV) in acute COPD patients. Eur Respir Rev; 15: 190–191.
• NHS England. Risk of severe harm and death from unintentional interruption of
non-invasive ventilation. www.england.nhs.uk/wp-content/uploads/2015/02/
psa-niv.pdf Date last updated: February 13, 2015.
• Samolski D, et al. (2006). Inspired oxygen fraction achieved with a portable ventila-
tor: determinant factors. Respir Med; 100: 1608–1613.
• Schwartz AR, et al. (2004). Factors affecting oxygen delivery with bi-level positive
airway pressure. Respir Care; 49: 270–275.
• Stéphan F, et al. (2015). High-flow nasal oxygen therapy vs noninvasive positive
airway pressure in hypoxemic patients after cardiothoracic surgery: a randomised
clinical trial. JAMA; 313: 2331–2339.

The basics: patient selection
The patient with an acute hypercapnic

exacerbation of COPD
Mark W. Elliott
NIV is well established in the management of acute-on-chronic hypercapnic

respiratory failure secondary to an acute exacerbation of COPD (AECOPD). RCTs
conducted in different countries and in different clinical settings have consist-
ently shown that NIV reduces the need for endotracheal intubation and, in
some studies, improves survival. In no study has NIV been shown to be harmful,
although in one, there was a trend towards worse survival, attributable to a delay in
endotracheal intubation. A reduction in nosocomial infection is a consistent find-
ing. NIV has been compared head-to-head with invasive ventilation and shown to
be comparable in terms of survival. In those who could be successfully managed
using NIV, there were advantages in the short-term: shorter ICU stay and, in the
subsequent year, a lower readmission rate and less need for de novo long-term
oxygen prescription. However, NIV is best seen as providing an option between no
ventilatory support and full support delivered invasively via an endotracheal tube.
It is complementary, not an alternative; NIV will prevent a significant proportion of
patients deteriorating to the point when endotracheal intubation, and its attendant
complications, becomes necessary.
Patient selection and initial management
pH is the single most important predictor of outcome in an AECOPD. Un
controlled oxygen therapy can lead to worsening hypercapnia and acidosis in
at-risk individuals, and has been shown to be associated with an increased need
Key points
• Oxygen should be given to a target saturation of 88–92%
during spontaneous breathing and NIV.
• NIV should be started if pH <7.35 after standard medical
therapy.
• There is no lower pH below which a trial of NIV is inappropriate.
• If pH <7.30, the risk of endotracheal intubation and death is
substantially increased, and NIV is strongly advised.
• When a patient is deteriorating despite NIV, distinguish
between “technical”, “noninvasive” and “ventilation” failure.

Box 1. When should NIV be started in AECOPD?

After 1 h of standard medical therapy, including oxygen targeted to a
saturation of 88–92%, if PaCO2 >6 kPa (45 mmHg), then:
• pH >7.35 means that NIV not indicated
• pH 7.30–7.35 means that NIV is advised
One patient will avoid intubation for every 10 treated but 80% will get better
without NIV.
If a patient is reluctant to use NIV, it is reasonable to repeat arterial blood
gas analysis after 1–2 h to establish whether the patient is one of the 80%
who will get better just with conservative treatment, then:
• pH <7.30 means that NIV is strongly advised
50% of patients will deteriorate to the point at which endotracheal
intubation should be considered; even if the patient is reluctant to use NIV,
they should be strongly encouraged to do so.
for endotracheal intubation and ICU admission. In a RCT performed in patients

during ambulance transfer to hospital, mortality was higher in patients receiving
high-flow oxygen than in those in whom it was titrated to achieve an oxygen
saturation of 88–92%. Approximately 20% of patients presenting with acidosis
to the emergency department will correct their pH completely into the normal
range with standard medical therapy, most importantly including properly con-
trolled oxygen and this, therefore, should be the initial focus of therapy. If after
1 h, the patient remains acidotic (pH <7.35), NIV should be started. There is no
lower pH value below which NIV should not be used; however, the more acidotic
the patient, the greater the risk of failure of NIV (box 1).
There are few absolute contraindications to NIV. These are intuitively reason-
able and have been exclusion criteria for inclusion in the RCTs rather than hav-
ing been shown to be harmful. For coma, which was previously considered as a
contraindication, it has subsequently been shown that the outcome with NIV
is as good as in other patient groups. Therefore, it is better to consider some
situations as warranting special monitoring and care (box 2). For instance, if a
pneumothorax is identified on a chest radiograph, an intercostal drain should be
inserted. However, this is not always technically straightforward and in this case,
a careful trial of NIV using low inflation pressures and with very close monitoring
is possible.
Monitoring
Patients must be monitored, as a proportion will deteriorate despite NIV (box 3).
Oxygen saturation should be monitored continuously and although there is no
direct clinical trial evidence, there seems no reason not to target an oxygen satur
ation of 88–92%, as in the spontaneously breathing patient. Indeed, in a large
study from China in which NIV was started 24–48 h after admission and a reduc-
tion in endotracheal intubation rate seen even in nonacidotic patients, injudicious
oxygen therapy may have been a factor. The high intubation rate in nonacidotic
patients was surprising and the mean PaO2 in both the NIV-treated patients and
Box 2. Situations that warrant special care and monitoring, and may seriously
limit the use/effectiveness of NIV
• Facial burns, recent facial or upper airway surgery, or facial trauma
• Vomiting
• Fixed upper-airway obstruction
• Undrained pneumothorax
• Upper gastrointestinal surgery
• Inability to protect the airway
• Copious respiratory secretions
• Life threatening hypoxaemia
• Haemodynamically unstable requiring inotropes/pressors
• Confusion/agitation
• Bowel obstruction
controls was 9.6–11.3 kPa (72–85 mmHg) in the first 24 h after randomisation,

suggesting that these patients may have deteriorated because they were given too
much oxygen in hospital.
Patients require intermittent arterial blood gas tensions, the frequency depend-
ing upon severity, with the more acidotic patient requiring more regular arterial
sampling. Respiratory rate should be assessed regularly. An improvement in pH
and a fall in respiratory rate are markers of a likely successful outcome from NIV.
Similarly, an improvement in Glasgow Coma Scale and APACHE (Acute Physiology
and Chronic Health Evaluation) scores is predictive of a favourable outcome. To
ensure optimal delivery of NIV, leaks from around the interface should be sought
and minimised, and the patient carefully observed for any evidence of asynchrony
between their breathing pattern and ventilator-delivered breaths. With more
sophisticated, “high-end” ventilators specifically designed for acute care, analysis
of displayed waveforms may yield useful additional information. There may also
be a role for transcutaneous carbon dioxide monitoring and although this does
not give a measurement of pH, this can be estimated; further research is required
before this can be recommended for routine clinical practice.
Choice of interface
Because patients with an AECOPD generally mouth-breathe, an oronasal mask
is the interface of choice. If this cannot be tolerated, nasal masks or pillows are
Box 3. Monitoring
Continuous
• Oxygen saturation
• ECG if the patient is tachycardic or there is evidence of arrhythmia
Intermittent
• Arterial blood gas sampling
• Recording of respiratory rate
• Observation for leaks around the interface
• Observation of patient–ventilator synchrony

an alternative. A mask that covers the whole face – the cephalic or total face
mask – may be more acceptable to some patients and has been shown to be of
comparable efficacy to an oronasal mask. The helmet mask has also been used
in patients with COPD.
How to set the ventilator
There is a wide variety of ventilators available from a number of manufacturers.
The majority, if not all now, will be bilevel devices. Staff should be trained with the
device that they are using; it is not possible to give an exhaustive description of
how to set the ventilator, as this will vary from machine to machine. There are,
however, some parameters that will need to be set on all machines.
Many patients find the sensation of positive pressure applied to the upper air-
way uncomfortable, and tolerance is likely to be improved by starting low and
increasing the pressure over time. There will often have to be a compromise
between comfort and response. In other words, the pressure required to achieve
the desired response may be at the cost of patient discomfort, which may reach
a point at which the patient will not use the ventilator. It is better to achieve 80%
of the response with the patient prepared to use NIV than to strive for 100% and
the patient refusing. Where the line is drawn between response and comfort will
depend upon severity. At the milder end of the spectrum, response can be com-
promised to ensure comfort, whereas this is not true at the severe end of the spec-
trum. Usually, as the patient gets used to NIV, particularly if it relieves dyspnoea,
the level of support can be increased.
EPAP This is important in overcoming the inspiratory threshold load imposed by
intrinsic PEEP, reducing ineffective efforts and preventing carbon dioxide rebreath-
ing by lavaging exhaled gas from the mask. It should usually be set at 5 cmH2O
and rarely increased above 8 cmH2O.
IPAP This should start at 10 cmH2O and be increased, according to comfort and
response, to 15 cmH2O and then, if possible, to 20 cmH2O. The speed with which
this is done will depend upon illness severity. The patient should then be re-
evaluated with an assessment of respiratory rate and arterial blood gas tensions.
There is some controversy about the use of higher inflation pressures. They can
cause an increase in patient–ventilator asynchrony because of worsening hyper-
inflation and may cause more leakage, but there is still a worthwhile increase in
ventilation and offloading of the respiratory muscles. If necessary, IPAP can be
increased to the maximum generated by the ventilator.
Backup rate Finally, a backup rate should be set. Periods of apnoea are rare in
patients with an AECOPD, who usually have a high respiratory rate. The backup
rate should be set at 15 breaths⋅min−1.
How to recognise the patient who is deteriorating despite NIV
A falling pH, rising carbon dioxide, increasing tachypnoea or inability to ad
equately oxygenate indicate that NIV is failing. Failure can be classified under
three headings.
1) Technical failure This refers to a failure to achieve adequate ventilation non
invasively. Causes include an inappropriately set ventilator or, rarely, equipment
malfunction.
2) Failure of “noninvasive” This relates to problems with the interface, either of

patient tolerance, usually due to claustrophobia, or of inability to achieve a
satisfactory fit without excessive leaks.
3) Failure of “ventilation” This refers to the situation in which, despite patient
acceptance, minimal leaks and good patient–ventilator synchrony, it is not pos-
sible to ventilate the patient adequately.
In cases of “technical” failure, the abnormality should be corrected and the
patient re-evaluated soon thereafter. “Noninvasive” failure will usually occur
early, while there is still time for medical therapy, such as steroids, to have its full
effect, and replacement of an ineffective interface with a more effective one (e.g.
an endotracheal tube) is likely to result in successful ventilation. Before recourse
is made to intubation, other interfaces, differing in size and type, should be tried.
In contrast, failure of “ventilation” usually occurs later, at which time medical
therapy has usually had its maximum effect and replacement of an interface
that is functioning effectively with a different interface (an endotracheal tube)
is unlikely to confer much additional benefit (fig. 1). Late failure of NIV usu-
ally occurs in patients with advanced disease and carries a poor prognosis. It
may be more appropriate to consider switching to a more palliative approach.
However, determining prognosis for patients with COPD is notoriously difficult
and clinicians underestimate the likelihood of success with invasive ventilation.
Although symptom burden is high, many would choose intubation again where
it to be required.
Troubleshooting
Patient intolerance of the mask, leaks, and asynchrony between the patient and
ventilator are the commonest causes of NIV failure (box 4). Before accepting that
the patient cannot tolerate a particular interface, is important to establish that
it is the correct size, has the correct head gear and is being used appropriately.
Sometimes, the patient is right not to tolerate the interface because it is not being
used correctly. A switch to a smaller, less intrusive interface, such as a nasal mask
or nasal pillows, sometimes proves acceptable, but this is at the cost of mouth
leaks and because there is less contact with the face, these masks tend to be
less stable and more easily dislodged. A strategy of rotating interface types has
been described. Leaks result in inadequate pressure support, and can interfere
with both triggering into inspiration and cycling into expiration. It is usually a con-
sequence of an incorrectly sized interface or, most commonly, insufficient tension
Failure of “noninvasive” Failure of “ventilation”

tends to be early tends to be later
Time for medical Medical therapy has

therapy to work already had its chance
IMV success
Figure 1. Failure of NIV. IMV: invasive mechanical ventilation.

Box 4. Checklist for patients uncomfortable with NIV or failing to improve

1) Check that the interface and head gear are the correct size and
appropriately applied
2) Minimise leak
3) Carefully observe the excursion of the chest wall while listening to the
way that the ventilator cycles. Each inspiratory movement of the chest
wall should be accompanied by a machine-delivered breath. The onset
of the machine-delivered breath should coincide with inspiratory
movement of the chest wall and the machine should cycle into
expiration when the chest wall is maximally expanded.
If patient breaths are not accompanied by a machine breath:
• Minimise leak
• Increase trigger sensitivity
• Increase EPAP in 1-cmH2O increments to 8 cmH2O
• Consider reducing IPAP
• Consider increasing backup rate
in the restraining straps, which should be tightened. Masks for NIV are designed so
that as the pressure in the interface rises, the mask surface is pressed flat against
the face; paradoxically, if the mask is applied too tightly, this effect is lost; leaks can
be reduced by loosening the mask straps.
Patient–ventilator asynchrony is a consequence of a trigger that has been set
incorrectly (on some ventilators, it is not possible to change the trigger sensitivity)
or, because of intrinsic PEEP, inspiratory airflow at the mouth lags behind inspir
ation, as defined by activation of the inspiratory muscles. Careful observation will
show a lag between the onset of ventilator inspiration and outward chest wall
movement. This can be reduced by increasing EPAP. If the patient is receiving high
levels of pressure support, consideration should be given to reducing the level of
IPAP. Alternatively, the backup rate can be increased but care must be taken not
to worsen hyperinflation.
Future care planning
Many patients and their families regard COPD as a way of life rather than a
disease, and some are surprised when patients with clearly very severe and
advanced disease die. An admission with an AECOPD, particularly if it requires NIV,
is an important event in the natural history of COPD. Patients are at high risk of
readmission and death in the year following an AECOPD. The opportunity should
be taken to discuss future care including the patient’s wishes about how future
exacerbations should be managed, in particular their desire or otherwise for
invasive ventilation.
Audit
The application of NIV is a practical skill and the good results obtained in clin
ical trials in AECOPD are not necessarily guaranteed in everyday clinical prac-
tice. Auditing is important for identifying practice that does not reach acceptable
standards and driving change, primarily in education and service delivery.

Further reading
• Antonelli M, et al. (2004). Noninvasive positive pressure ventilation using a helmet
in patients with acute exacerbation of chronic obstructive pulmonary disease: a
feasibility study. Anesthesiology; 100: 16–24.
• Appendini L, et al. (1999). Physiologic response of ventilator-dependent
patients with chronic obstructive pulmonary disease to proportional assist ven-
tilation and continuous positive airway pressure. Am J Respir Crit Care Med; 159:
1510–1517.
• Austin MA, et al. (2010). Effect of high flow oxygen on mortality in chronic obstruc-
tive pulmonary disease patients in prehospital setting: randomised controlled trial.
BMJ; 341: c5462.
• Chu CM, et al. (2004). Readmission rates and life threatening events in COPD sur-
vivors treated with non-invasive ventilation for acute hypercapnic respiratory fail-
ure. Thorax; 59: 1020–1025.
• Collaborative Research Group of Noninvasive Mechanical Ventilation for Chronic
Obstructive Pulmonary Disease (2005). [Early use of noninvasive positive pressure
ventilation for patients with acute exacerbations of chronic obstructive pulmonary
disease: a multicentre randomized controlled trial]. Zhonghua Jie He He Hu Xi Za
Zhi; 28: 680–684.
• Colombo D, et al. (2008). Physiologic response to varying levels of pressure sup-
port and neurally adjusted ventilatory assist in patients with acute respiratory fail-
ure. Intensive Care Med; 34: 2010–2018.
• Confalonieri M, et al. (2005). A chart of failure risk for noninvasive ventilation in
patients with COPD exacerbation. Eur Respir J; 25: 348–355.
• Conti G, et al. (2002). Noninvasive vs. conventional mechanical ventilation in
patients with chronic obstructive pulmonary disease after failure of medical treat-
ment in the ward: a randomized trial. Intensive Care Med; 28: 1701–1707.
• Cuvelier A, et al. (2009). Cephalic versus oronasal mask for noninvasive ventilation
in acute hypercapnic respiratory failure. Intensive Care Med; 35: 519–526.
• Diaz GG, et al. (2005). Noninvasive positive-pressure ventilation to treat hypercap-
nic coma secondary to respiratory failure. Chest; 127: 952–960.
• Elliott M, et al., eds (2010). Non-invasive Ventilation and Weaning: Principles and
Practice. Boca Raton, CRC Press.
• Elliott MW, et al. (2012). Noninvasive ventilation for acute exacerbations of chronic
obstructive pulmonary disease: “Don’t think twice, it’s alright!” Am J Respir Crit
Care Med; 185: 121–123.
• Ferguson GT, et al. (1995). CO2 rebreathing during BiPAP ventilatory assistance.
Am J Respir Crit Care Med; 151: 1126–1135.
• Girault C, et al. (2009). Interface strategy during noninvasive positive pressure
ventilation for hypercapnic acute respiratory failure. Crit Care Med; 37: 124–131.

• Jeffrey AA, et al. (1992). Acute hypercapnic respiratory failure in patients with
chronic obstructive lung disease: risk factors and use of guidelines for manage-
ment. Thorax; 47: 34–40.
• Lightowler JV, et al. (2003). Non-invasive positive pressure ventilation to treat
r espiratory failure resulting from exacerbations of chronic obstructive pulmonary
disease: Cochrane systematic review and meta-analysis. BMJ; 326: 185.
• Moretti M, et al. (2000). Incidence and causes of non-invasive mechanical ventila-
tion failure after initial success. Thorax; 55: 819–825.
• Muir J-F, et al., eds (2008). Noninvasive Ventilation. ERS Monogr; 41.
• Plant PK, et al. (2000). One year period prevalence study of respiratory acidosis in
acute exacerbations of COPD: implications for the provision of non-invasive venti-
lation and oxygen administration. Thorax; 55: 550–554.
• Plant PK, et al. (2001). Non-invasive ventilation in acute exacerbations of chronic
obstructive pulmonary disease: long term survival and predictors of in-hospital
outcome. Thorax; 56: 708–712.
• Roberts CM, et al. (2011). Acidosis, non-invasive ventilation and mortality in hos-
pitalised COPD exacerbations. Thorax; 66: 43–48.
• Tuggey JM, et al. (2006). Titration of non-invasive positive pressure ventilation in
chronic respiratory failure. Respir Med; 100: 1262–1269.
• Wildman MJ, et al. (2007). Implications of prognostic pessimism in patients with
chronic obstructive pulmonary disease (COPD) or asthma admitted to intensive
care in the UK within the COPD and asthma outcome study (CAOS): multicentre
observational cohort study. BMJ; 335: 1132.
• Wildman MJ, et al. (2009). Survival and quality of life for patients with COPD or
asthma admitted to intensive care in a UK multicentre cohort: the COPD and
Asthma Outcome Study (CAOS). Thorax; 64: 128–132.
• Wood KA, et al. (1998). The use of noninvasive positive pressure ventilation in
the emergency department: results of a randomized clinical trial. Chest; 113:
1339–1346.
Online resources
• Hare A, et al. Acute COPD module. Skills-based Simulator Training in Non-Invasive
Ventilation. www.ers-education.org/e-learning/simulators.aspx

Patients with acute hypercapnic

respiratory failure and neuromuscular
or chest wall disease
Anita K. Simonds
Patients with neuromuscular disease (NMD) and chest wall disorders comprise a
heterogeneous group; even those with the same diagnosis may be at a different
stage of the natural history or the condition itself may affect individuals variably.
For this reason, there are few RCTs on the use of NIV to treat ARF in NMD and
chest wall disorders. Increased knowledge of genotype–phenotype links in NMD
has aided estimates of prognosis but, in many situations, long-term outcomes
may be hard to gauge, especially when faced with a patient presenting with ARF
for the first time. It is crucial that:
1) Patients are assessed on an individual basis without preconceptions
2) Where the natural history and disease progression in that individual is known,
advanced plans are formulated (anticipatory care plan)
Anticipatory care plans ensure that care decisions can be made on a considered
basis and involve all parties, including the patient and family. This will involve
choice between hospital admission or care at home, and decisions on escalation
from noninvasive to invasive ventilation, ICU admission and resuscitation prefer-
ences. The care plan should be reviewed, modified if required, and updated over
time depending on disease progression, comorbidities, response to therapy and
patient preference (see later). This management plan can be applied to conditions
as divergent as congenital muscular dystrophy, motor neurone disease and chest
wall disorders, providing they are individualised appropriately.
Particular issues which impact on acute respiratory care and NIV in NMD are:
• Difficulties with cough and ineffective secretion clearance
• Bulbar weakness – swallowing function may worsen at the time of an acute
infection, and/or aspiration may be the cause of, or have contributed to, the
current infective episode
Key points
• Patients with inspiratory and expiratory muscle weakness and
an acute chest infection benefit from a combination of NIV
and cough assist approaches, which may reduce the need for
• Anticipatory care plans are a key aspect of management.

• Upper limb weakness may make it difficult for patient to place and remove the
interface without assistance, or communicate
• Cardiomyopathy complicates a number of neuromuscular diseases including
Duchenne muscular dystrophy, Emery Dreifuss muscular dystrophy and sarco-
glycanopathies; left ventricular function may worsen during a chest infection
or conversely an acute decompensation may be caused by acute cardiogenic
pulmonary oedema
• Patients may need high-dependency unit care to manage nursing/self-care
needs even if there is no acute medical indication for the admission
• Practicalities involve the need for special pressure-control beds, hoists, accom-
modation for carers, and space for assistive equipment (e.g. wheelchair,
seating, plus adjacent disability toilet and shower facilities)
Combination of NIV and physiotherapy/cough assist techniques

Assessment of cough and cough assist techniques are described in the section
entitled “Practicalities of and guide to cough augmentation and daytime mouth-
piece ventilation”. Patients with early onset scoliosis or other forms of chest wall
disease may have preserved expiratory muscle and bulbar strength, and so are less
likely to require cough augmentation interventions or aspirate; NIV alone usually
suffices.
The combined use of NIV and cough assist strategies in case series of acute
ventilatory failure have produced encouraging results in Duchenne muscular dys-
trophy and spinal muscular atrophy, in which the joint application seems to reduce
the need for invasive ventilation and produces better outcomes. It is, therefore,
logical to apply this strategy to most acute exacerbations in NMD, providing obvi-
ous contraindications, e.g. profound bulbar weakness, aspiration, rapidly progres-
sive hypoxaemic respiratory failure, multisystem failure and coma, are absent.
NIV should be introduced if tachypnoea, high work of breathing and hypercapnia
are evident. Depending on the size of the patient, it is prudent to start at a fairly
low IPAP, e.g. 10 cmH2O with an EPAP of 4 cmH2O, increasing these settings as
required to control PaCO2 and maintain comfort.
Tips for management of patients on home NIV with an acute chest infection
• Have a low threshold to start antibiotics if pyrexia, purulent/increased
sputum and malaise.
• Ensure adequate hydration.
• Use NIV intensively.
• Carry out physiotherapy while the patient is using NIV – the physiotherapy
session is likely to be more productive and the patient will tire less.
• Add supplemental oxygen to NIV to maintain SpO2 >93%.
• Do not use uncontrolled oxygen therapy without close monitoring of
carbon dioxide tension (PCO2) or the addition of ventilatory support, as
hypercapnia may worsen.
• Always monitor PCO2 as well as SpO2. PtcCO2 measurements may be useful
to assess trends.
• Add humidification if the patient is using daytime and nocturnal NIV and/
or if bronchial secretions are tenacious, or mucous plugging is present.

• Consider increasing IPAP setting by 2–5 cmH2O according to PCO2 level

and comfort, and increase EPAP by 1–2 cmH2O to a maximum of
7 cmH2O to recruit lung volume and prevent or manage atelectasis.
Higher levels of EPAP reduce pressure support and may not be well
tolerated. Backup rate can be adjusted to just below spontaneous
breathing rate to improve PCO2 control.
• If the above alterations in IPAP and EPAP do not optimise arterial blood
gas tensions and reduce respiratory distress, use of a volume-assured
bilevel device (e.g. IVAPS or AVAPS) may be of value, although this has
not been confirmed in randomised trials.
• Consider using a full face mask to reduce leaks; or a Total Face Mask
(Philips Respironics, Best, The Netherlands) or helmet if difficulties
matching ventilatory needs are experienced using a nasal interface.
• Assess swallowing – if impaired consider nasogastric or percutaneous
gastric feeding to ensure nutrition is maintained. Remember swallowing
function may worsen during an acute exacerbation.
• Add nebulised bronchodilator if the patient has asthma, wheeze or
evidence of airflow obstruction.
General medical care

Serum creatinine levels are below the normal range in many NMD patients when they
are well, as a result of reduced muscle mass. Thus, a creatinine level within the normal
range when the patient is unwell can suggest pre-renal or renal insufficiency. Muscle
glycogen stores are also reduced due to low muscle bulk and so NMD patients may
become rapidly and unexpectedly hypoglycaemic during acute episodes. Therefore,
blood glucose levels should be checked regularly. NMD patients are also prone to
metabolic acidosis during acute events. This has been observed relatively often in
Duchenne muscular dystrophy and spinal muscular atrophy patients.
Duchenne muscular dystrophy patients and other myopathy patients may have a
raised creatinine phosphokinase or troponin level as part of their muscle condi-
tion. It is therefore trends in these results which matter, not absolute values.
Individuals with central core disease and multiminicore disease, especially involv-
ing mutations in the RYR1 gene, are at risk of malignant hyperpyrexia. This rarely
occurs spontaneously but is precipitated by muscle relaxants and anaesthetic
agents, such as suxamethonium chloride and halothane, respectively.
Endotracheal intubation and tracheostomy in acute chest infections
As discussed earlier the aim is to use a combination of NIV and cough insufflation–
exsufflation to manage acute exacerbations. In some patients with severe infections,
pneumonia or excessive secretions whom it is impossible to manage noninvasively,
a short period of endotracheal intubation and invasive ventilation may enable
secretions to be cleared optimally and recovery to occur. Fibre-optic bronchoscopy
may be required in some cases, and the cough insufflator–exsufflator machine can
be applied effectively via an endotracheal tube (fig. 1). The aim is then to extu-
bate the individual back onto NIV. Some authorities and even patients believe that
intubation should be avoided in all circumstances. However, unless NIV has been
agreed as a ceiling of care, a brief period of intubation and invasive ventilation with
52
a) b) c)
Figure 1. A patient with motor neurone disease and left lower lobe and lingular atelectasis a) on admission following intubation, b) after fibre-optic bron-
choscopy, and c) following addition of cough insufflation–exsufflation via an endotracheal tube, resulting in re-inflation of the atelectatic area.

extubation back onto NIV may shorten the duration of illness and so be in the
patient’s best interests, where NIV has failed and the precipitating cause is revers-
ible. Of course, difficulties arise if the patient cannot be extubated back onto NIV
because of extreme ventilator dependence or swallowing impairment/aspiration.
This may be temporary, in which case it is logical to perform a tracheostomy with
a view to decannulating and stepping down to NIV on recovery (see the section
entitled “Stepping up and down from NIV to tracheostomy ventilation”).
The justifiable concern is that after performing a tracheostomy, it may not be pos-
sible to step down to NIV. In this situation it may be the patient’s wish to continue
with tracheostomy ventilation long term, in which case the long-term care plan
will need to be changed accordingly. Decisions about the role of tracheostomy
should be made on an individual basis taking into account the reversible nature
of an acute event, progression of the condition over the period leading to the
current episode and its likely course, and the patient/family and team views on
the outcome with a tracheostomy versus the outcome without performing this.
Vianello et al. (2011) advocate the temporary placement of a minitrach combined
with NIV to aid secretion clearance. This may be of value in some cases – we have
found it of more use for stepping down from invasive ventilation, via minitrach, to
NIV rather than for avoiding the need for a formal tracheostomy. It is not possible
to use a minitrach to deliver ventilatory support.
Preventing chest infections All patients should receive the influenza and pneumo-
coccal vaccinations, if not contraindicated. Patients and their family/carers should
receive physiotherapy advice, and those with a reduced peak cough flow should be
provided with a home cough-assist device where indicated. Patients should have a
low threshold to start an antibiotic if they develop symptoms of a chest infection
and the provision of a reserve course is helpful. In a small subgroup of patients who
develop bronchiectasis or pseudomonal colonisation due to repeated infections/
aspiration, nebulised colistin or another anti-pseudomonal antibiotic may help. In
patients with asthma or those who develop wheeze or bronchial hyperreactivity at
the time of a chest infection, a bronchodilator should be added. As inspiratory flow
is often poor, triggering of an inhaler may be unreliable, and therefore, a nebulised
bronchodilator is preferable. It is important to consider that wheeze may not be
audible in NMD and chest wall disorder patients with very low VT.
NIV in patients with an acute pneumothorax

For patients with NMD, or any other condition in which the risk of acute pneumo
thorax is high (e.g. cystic fibrosis or bullous lung disease), who require NIV the
following strategies can be used:
• Insert a chest drain into the pneumothorax before applying positive pressure
ventilation
• Use a pressure-preset, pressure-limited ventilator to reduce peak airway pressure
• Use the lowest IPAP and EPAP settings that are compatible with control of
arterial blood gas tensions and symptoms. Be prepared to accept a higher
level of PaCO2 than normal (permissive hypercapnia), provided the patient is not
becoming acidotic
• Explore the possibility of temporarily withdrawing NIV
• Have a low threshold to repeat chest radiography to assess progress
• Consider a computed tomography scan of the thorax to differentiate between

pneumothorax and bulla
• High-flow oxygen therapy (e.g. AIRVO 2; Fisher & Paykel Healthcare Limited
Auckland, New Zealand) might be suitable in selected cases with mild hyper-
capnia or normocapnia, but very careful monitoring of PaCO2 is mandatory
Decision making and anticipatory care plans

The best decisions are based on the best information. Over the past decade better
care, including the application of NIV, has altered the natural history of many con-
ditions, and molecular genetics breakthroughs have facilitated the understand-
ing that some conditions (e.g. limb girdle and congenital muscular dystrophies)
consist of a variety of genetic and phenotypic disorders with variable outcome.
These findings are discussed the sections entitled “Chronic NIV in hereditary
neuromuscular disorders” and “Chronic NIV in children: indications, outcomes
and transition”, and will inevitably will influence decisions on the management
of acute events. It should also be understood that in some conditions, e.g. spinal
muscular atrophy type 1, there is a spectrum within the classification, with some
infants being profoundly weak shortly after birth and others having better pre-
served function. Many chest wall disorders such as early onset scoliosis are not
progressive, but lung volume will be lost with normal ageing. Previously, there
has been evidence of widely varying medical advice given to patients and families
on prognosis and available care options; some of which was based on outdated
expectations (Gibson, 2001). This unhelpful situation has probably improved
(Kinali et al., 2006), but it is always sensible to seek advice from specialist centres
who frequently manage NMD and chest wall disorder patients.
A number of organisations issue alert cards for patients that cover medical advice
for chest infections and an anaesthetic guide for procedures (e.g. Motor Neurone
Disease Association, www.mndassociation.org; and Muscular Dystrophy Campaign,
www.muscular-dystrophy.org). These can be customised and are helpful.
Anticipatory care plans should include preferences and decisions on the use of:
• Hospital admission versus care at home or, for example, in a hospice
• Intravenous antibiotics
• NIV
• Cough-assist devices
• NIV as a ceiling of care
• Nasogastric and/or percutaneous endoscopic gastrostomy insertion for hydra-
tion and/or feeding
• ICU admission
• Endotracheal intubation
• Insertion of a tracheostomy (with short- or long-term aim)
• Resuscitation
Palliative care
In some patients, palliative care strategies to improve symptom control may sit logi-
cally alongside interventional plans including temporary ICU admission for an acute
episode; in other situations NIV as a ceiling of care (or indeed a preference to avoid or
withdraw NIV) combined with palliative care input will be the management of choice.
Further reading
• Bach JR, et al. (2000). Spinal muscular atrophy type I: a noninvasive respiratory
management approach. Chest; 117: 1100–1105.
• Bach JR, et al. (2010). Extubation of patients with neuromuscular weakness: a new
management paradigm. Chest; 137: 1033–1039.
Child; 96: 426–432.
• Finder JD, et al. (2004). Respiratory care of the patient with Duchenne muscular
dystrophy: ATS consensus statement. Am J Respir Crit Care Med; 170: 456–465.
• Gibson B (2001). Long-term ventilation for patients with Duchenne muscular dys-
trophy: physicians’ beliefs and practices. Chest; 119: 940–946.
• Kinali M, et al. (2006). UK Physicians’ attitudes and practices of long term non-
invasive ventilation of children with Duchenne muscular dystrophy. Pediatr
Rehabil; 9: 351–364.
• Racca F, et al. (2010). Respiratory management of acute respiratory failure in
neuromuscular diseases. Minerva Anestesiol; 76: 51–62.
• Simonds AK (2004). Pneumothorax: an important complication of non-invasive
ventilation in neuromuscular disease. Neuromusc Disord; 14: 351–352.
• Simonds AK (2007). Acute non-invasive ventilation in neuromuscular disease,
chest wall disorders and cystic fibrosis amd bronchiectasis. In: Simonds AK, ed.
Non-Invasive Respiratory Support: a Practical handbook. 3rd edn. London, Hodder
Arnold; pp. 73–80.
• Tzeng AC, et al. (2000). Prevention of pulmonary morbidity for patients with
neuromuscular disease. Chest; 118: 1390–1396.
• Vianello A, et al. (2000). Non-invasive ventilatory approach to treatment of acute
respiratory failure in neuromuscular disorders. A comparison with endotracheal
intubation. Intensive Care Med; 26: 384–390.
• Vianello A, et al. (2011). Prevention of extubation failure in high risk patients with
neuromuscular disease. J Crit Care; 26: 517–524.
Online resources
• Hare A, et al. Acute neuromuscular disease module. Skills-based Simulator Training
in Non-Invasive Ventilation. www.ers-education.org/e-learning/simulators.aspx
• Schönhofer B (2014). NIV in acute hypercapnic respiratory failure. ERS Course

Patients with acute hypercapnic

respiratory failure and OHS
Nicholas Hart and Patrick B. Murphy
OHS is increasingly common, with both increasing obesity and increasing severity of
OSA. Although the true population prevalence is unknown, it can be demonstrated
that significant numbers of acute medical patients meet the criteria for the
diagnosis of OHS, although this often goes unrecognised and untreated, even if it
is recognised. The original description of treatment of OHS with positive pressure
ventilation was reported in the ICU in patients with significant comorbidity, and
this has become the increasingly common cohort within critical care, recently
termed malignant OHS. Despite the recognised morbidity and mortality of patients
presenting with obesity-related acute decompensated respiratory failure, it
remains underrepresented in primary research and guidelines on NIV. ∼50% of
patients presenting with decompensated OHS will have another contribu tory
acute pathology, such as pneumonia, heart failure, veno-thromboembolism or
sepsis. The other half of these patients present with decompensated chronic
respiratory failure.
Key points
• Differences in symptom burden influence the mode of clinical
presentation.
• High morbidity and mortality of obese patients within critical
care is attributable to comorbidities rather than obesity per se,
and the presence of obesity should not preclude escalation of
care during acute decompensated respiratory failure.
• NIV is associated with reduction in morbidity and mortality
compared with endotracheal intubation, and is the preferred
mode of ventilatory support in those with primarily acute and
acute-on-chronic respiratory failure.
• Patients with poor adherence to domiciliary CPAP and NIV
or with super obesity (BMI >60 kg·m−2) are most likely to fail
initial acute management with NIV and such patients should
be appropriately managed within critical care to facilitate
escalation to invasive ventilation if required.

There is evidence that patients presenting acutely have a lower perceived symp-
tom burden. Such patients do not present themselves at an early stage of disease,
and therefore are more likely to present with acute decompensated chronic re
spiratory failure. It is important to highlight that morbid obesity is not a poor prog-
nostic feature in patients admitted to critical care, as supported by a number of
meta-analyses and epidemiological studies, despite a perception to the contrary.
It is the comorbidities of obesity, such as diabetes and ischaemic heart disease,
that are prognostically more important than obesity per se in determining patient
outcomes from critical care.
Patients presenting with acute decompensated respiratory failure secondary to
OHS can be managed successfully with NIV and this should be the preferred mode
of ventilatory support, with the exception of the most severely unwell or those
with other associated organ dysfunction in whom endotracheal intubation and
mechanical ventilation on the ICU is mandated. It is important to acknowledge that
the outcome in patients with multi-organ failure complicating decompensated
ARF remains poor. Furthermore, the management of obesity-related respiratory
failure with oxygen therapy alone, either at high or intermediate concentrations, is
strongly discouraged due to the adverse effect on minute ventilation with subse-
quent worsening of hypercapnic respiratory failure. The strongest predictor of NIV
failure in patients already established on domiciliary therapy for obesity-related
respiratory failure is their previous compliance. Indeed, patients with poor adher-
ence to home CPAP or NIV are more likely to fail a trial of acute NIV during a
decompensated episode. NIV is also more likely to fail in the super-obese patients
with a BMI >60 kg⋅m−2 and in those patients who have a poor initial response to
NIV. It is essential that such patients undergo a careful risk assessment and that
NIV is administered in the most appropriate clinical setting, which is dependent
on the balance between the risk of NIV success or failure.
There are few data to support specific ventilator strategies in the management of
acute decompensated OHS, but as is the case with all administration of NIV, an
experienced multidisciplinary team is vital for its successful application. However,
a requirement of NIV application is to control upper airways obstruction and over-
come the respiratory muscle load and, as a consequence, a high EPAP and high
inspiratory driving pressure are required. Specifically, the EPAP should be set to
optimise respiratory mechanics by modifying lung volume and the operating pos
ition on the pressure–volume curve as well as overcoming upper airways obstruc-
tion in order to reduce the work of breathing. IPAP should be titrated to ensure
satisfactory chest wall excursion with supplementary oxygen entrained at the
lowest level to achieve target oxyhaemoglobin saturations of between 88% and
92%. The inspiratory time should be set to between 1.2 s and 1.4 s in line with
the patient’s restrictive lung defect and the backup rate should be used to ensure
mandatory ventilation during sleep. A target backup rate of between 12 and 14
breaths⋅min−1 is recommended. The details of NIV titration are shown in figure 1.
To deliver such noninvasive ventilatory support, an oronasal mask is the preferred
interface in the acute setting to reduce mask leak and maximise ventilation. As
with all NIV prescriptions, the interface should be individually sized and fitted by
experienced staff to reduce the risk of nasal bridge ulceration during therapy and
to optimise patient comfort. Treatment duration should be ongoing, with short
breaks for oral medication, oral hygiene, hydration and nutrition, during the first
58
Assess for underlying cause(s) of acute
Assess suitability for NIV
decompensation# and treat appropriately
NIV contraindicated
Suitable for trial of NIV
Liaise with critical care to assess for invasive

Establish interface and initial settings ventilatory support
EPAP IPAP Other settings

Start at 6 cmH2O and titrate by 2 cmH2O to: Start at 16 cmH2O and titrate by 2 cmH2O to: Supplemental oxygen:
Abolish upper airways obstruction Obtain adequate chest wall excursion If persistent hypoxia (SpO2 <88%, PaO2 <8 kPa
Stop snoring/chest wall paradox Obtain adequate VT (if measured accurately (<60 mmHg)) despite adequate IPAP and
Stop repetitive desaturations aim for 8–10 mL·kg-1 ideal body weight) EPAP
Aim: SpO2 >88%, PaO2 >8 kPa (>60 mmHg) Reduce respiratory distress (respiratory Backup rate:
Usual setting: 8–10 cmH2O rate <30 breaths·min-1) 12–16 breaths·min-1
Aim: pH >7.35 Inspiratory time:
Usual settings: 18–24 cmH2O 1.2 s – 1.4 s
Continuous physiological monitoring

First 24-48 h
Oxygen saturations and PtcCO2
Intermittent arterial blood gas analysis
Figure 1. Flow chart summarising management of and NIV settings for acute decompensated episodes of obesity-related respiratory failure. #: e.g. pneumonia,
right heart failure, veno-thromboembolism and sepsis.

24–48 h of therapy. When clinical stability has been achieved, weaning of NIV
support from 24 h per day to night-time only can be undertaken with continu-
ous and then intermittent cardiorespiratory physiological monitoring. In patients
without a pre-existing diagnosis of obesity-related respiratory failure, this needs
to be confirmed and other causes of chronic respiratory failure excluded based on
overnight limited respiratory polygraphy, lung function testing and imaging, as per
standard practice. In particular, a coexistent diagnosis of COPD should be sought.
It has been established that the hospital outcome of patients receiving NIV for
treatment of acute decompensated respiratory failure is better in patients with
obesity-related respiratory failure than in those with COPD. However, the long-
term outcome is similar between these two patient groups if the patients with
obesity-related respiratory failure are not established on domiciliary NIV treat-
ment on discharge from hospital. This highlights the potential benefit of home
mechanical ventilation for obese patients with chronic respiratory failure and,
although these patients can be safely discharged in the short term following an
acute episode of respiratory decompensation, an assessment for long-term CPAP
or NIV treatment is required in the early post-discharge period.
Clinical tips
• The initial NIV interface should be an oronasal mask or a total face mask.
• During the initial acute period, when day and night ventilation is required,
settings can be manipulated to improve daytime tolerance. For example, lower
levels of EPAP can be applied during the day to improve tolerance, as OSA does
not occur outside of sleep, and the backup rate can be reduced or the spontan
eous mode can be applied.
• Optimisation of cardiorespiratory morbidity and treatment of cor pulmonale
with aggressive diuresis is required concurrently with application of NIV.
• Allow patients regular breaks from NIV as necessitated by the need for medica-
tion, meals and personal hygiene in order to prevent skin breakdown on the
nasal bridge and promote adherence.
• It is essential to correct upper airways obstruction with adequate EPAP and an
optimal sleeping position to prevent incremental IPAP increases due to fail-
ure to deliver pressure to the lower airways rather than inadequate pressure
support.
Further reading
• Carrillo A, et al. (2012). Noninvasive ventilation in acute hypercapnic respiratory
failure caused by obesity hypoventilation and chronic obstructive pulmonary dis-
ease. Am J Respir Crit Care Med; 186: 1279–1285.
• Duarte AG, et al. (2007). Outcomes of morbidly obese patients requiring mechan
ical ventilation for acute respiratory failure. Crit Care Med; 35: 732–737.
• Masa JF, et al. (2010). Non-invasive ventilation in acute and chronic respiratory
failure secondary to obesity. In: Elliott M, et al., eds. Non-Invasive Ventilation and
Weaning: Principles and Practice. 1st Edn. London, Hodder Arnold; pp. 408–417.

Patients with acute-on-chronic

hypercapnic respiratory failure
due to non-COPD obstructive lung
disease and interstitial disorders
Respiratory failure occurs when the respiratory system cannot maintain gas
exchange, causing dysfunction of other organs and threatening life. Respiratory
failure can primarily affect oxygenation, manifested by hypoxaemia, or ventilation,
manifested by additional hypercapnia. Hypoxaemic respiratory failure can be clas-
sified into four pathophysiological mechanisms:
• impaired diffusion
• shunt
• ventilation–perfusion mismatch
• hypoventilation
While the first three mechanisms lead primarily to hypoxaemia and not hyper-
capnia, hypoventilation leads to both. Hypercapnic respiratory failure caused by
hypoventilation develops secondary to a decreased ventilatory drive or a load
placed on the system that cannot be overcome by the respiratory muscles.
Key points
• In acute-on-chronic respiratory failure, the worsened imbalance
between the load placed on the system and the capacity of the
respiratory muscles can no longer be compensated.
• In obstructive lung diseases, this imbalance will soon lead to
hypercapnia, while in interstitial lung diseases, as these are
lung parenchymal diseases, hypoxaemia occurs primarily, with
hypercapnia only at later stages.
• The evidence for NIV in acute exacerbations of nonobstructive
lung diseases is limited but in practice, it is often worth trying
to prevent intubation.
• The evidence for NIV in acute exacerbations of interstitial
lung diseases is even more limited; it often fails, especially
in idiopathic pulmonary fibrosis, but might be tried to avoid
endotracheal intubation.

In patients with pre-existing lung diseases, pathophysiological mechanisms may

coexist, leading to variable degrees of hypoxaemic and hypercapnic respiratory
failure.
Respiratory failure can be chronic, when the respiratory failure develops over
weeks to months, or acute, when there is a sudden deterioration over hours to
days. ARF can occur without pre-existing respiratory failure or on top of chronic
respiratory failure (CRF). In chronic hypercapnic respiratory failure, excess carbon
dioxide is buffered through bicarbonate retention in the kidney. As a consequence,
pH remains normal. However, when there is a sudden deterioration in patients’
baseline status, the kidney does not have time to retain enough bicarbonate to
buffer the increasing amount of carbon dioxide and respiratory acidosis develops,
being a typical sign of acute-on-chronic hypercapnic respiratory failure (fig. 1).
In this section, we will focus on the situation where ARF develops on top of chronic
hypercapnic respiratory failure in obstructive pulmonary disease other than COPD
and in interstitial lung diseases (ILDs). In COPD, the use of NIV to treat acute-on-
chronic hypercapnic respiratory failure is proven and incorporated into guidelines.
However, for non-COPD obstructive lung disease and interstitial disorders, the use
of NIV is outside the current guidelines but seems to be used rather frequently.
Acute-on-chronic respiratory failure in non-COPD obstructive lung diseases
Why patients with CRF can deteriorate so rapidly seems to be a consequence of
an imbalance between an increased load (caused by increased airway resistance
and a hyperinflation-driven decreased lung compliance) and a decreased capacity
of the respiratory system. Although one might expect that during this process,
respiratory muscles become fatigued, the occurrence of respiratory failure seems
to be a consequence of decreasing VT in order to prevent fatigue (fig. 2).
When optimal settings are applied, NIV counterbalances several of these factors.
First, the application of external PEEP to a level close to, but not above, intrinsic
Respiratory failure
Lung failure Pump failure
Hypoventilation
Impaired Hypoventilation (Central) Imbalanced load and

diffusion depression of capacity of respiratory
Ventilation–perfusion respiration muscles
Shunt
mismatch
Exacerbation
Respiratory muscle
weakness
Hypoxaemia Hypercapnia
Figure 1. Causes of respiratory failure.

1.0
0.8 Fatigue zone
0.6
tI/ttot
0.4
0.2
0
0 0.2 0.4 0.6 0.8
Pdi/Pdi,max
Figure 2. Relationship between inspiration time (tI) as a fraction of total duty cycle (ttot) and
force output of the diaphragm (Pdi/Pdi,max). In the COPD patient, in stable condition, this
relationship lies close to the fatigue zone (triangle). This is in contrast to a healthy subject
(circle). When an exacerbation occurs, more force has to be delivered by the diaphragm
(arrow), and patients would enter the fatigue zone (rectangle). However, to prevent entering
the fatigue zone, patients decrease their tI leading to a decrease in VT, leading to hyper-
capnia (oblique rectangle). Reproduced from Demedts M, et al., eds (1999), Longziekten
[Lung Diseases], with permission from the publisher (Koninklijke Van Gorcum, Assen, The
Netherlands).
PEEP helps to reduce the pressure gradient between the distal and central air-
ways that is established during dynamic hyperinflation. By doing so, patients need
less reduction in intrathoracic pressure and, thus, less inspiratory muscle work
to initiate inspiration. The addition of (enough) pressure support helps to further
unload the inspiratory muscles while it also augments ventilation. NIV has also
been shown to have bronchodilatory effects in asthma. Finally, NIV might improve
ventilation–perfusion matching. None of these is provided with the application of
oxygen therapy as the sole intervention.
Asthma In asthma, overt respiratory failure is very rare, as the treatment of asthma
has improved. Unfortunately, some patients do not adhere to their therapy, are
not treated optimally or have severe exacerbations despite optimal therapy, so
ventilatory failure in asthma does still occur.
Acute ventilatory failure in asthma usually occurs only with severe respiratory
distress. Patients hyperventilate until they become so fatigued that they can no
longer sustain the hyperventilation and start to retain carbon dioxide. Combined
respiratory and metabolic acidosis is often present at this stage. Therefore, patients
should be monitored closely and ventilator assistance should not be delayed until
patients reach this ‘crossover’ point.
Despite aggressive medical management with supplemental oxygen and
pharmacotherapy, some patients fail to improve, and require ICU admission and
mechanical ventilation. In this case, if the condition allows, NIV is preferable to
endotracheal intubation (ETI) and invasive mechanical ventilation (IMV), as this

carries the risk of complications such as ventilator-associated pneumonia and
barotrauma. In asthma, there have been studies investigating NIV as a treatment
of ARF. However, as this occurs so infrequently, studies have suffered from small
numbers and a lack of meaningful clinical outcomes. Some studies have shown
shorter length of hospital stay, reduced airflow limitation and improvement in
respiratory rate. However, as results from these studies are controversial, NIV for
the management of acute asthma is not recommended in guidelines. Nevertheless,
it might be beneficial in a subgroup of patients in the ICU, under careful observation,
and with careful consideration of ventilator type, mask choice, settings (inspiratory
pressures that are too high may lead to barotraumas and too much PEEP may lead
to increased lung hyperinflation) and nebuliser position.
Bronchiolitis Obliterative bronchiolitis is a disease process characterised by subepi-

thelial inflammatory and fibrotic narrowing of the bronchioles, and is caused by a
range of medical conditions and exposures. Clinically, it leads to a disease presen-
tation with progressive dyspnoea and nonproductive cough over a period of weeks
to months, and abnormal lung function frequently characterised by an obstructive
pattern. The natural history and prognosis of obliterative bronchiolitis is highly
variable; it may lead to respiratory failure. In adults, there is no evidence to sup-
port the use of NIV, both as a treatment of CRF due to obliterative bronchiolitis
and ARF. In children, acute bronchiolitis, usually due to (viral) infections, occurs
more frequently and can sometimes develop into CRF due to the development of
obliterative bronchiolitis. NIV can be effective in both these situations.
Cystic fibrosis Although CF is a multisystem disease, the primary cause of death

is respiratory failure. Inflammatory processes in the airway wall, mucus plugging
and lung parenchyma destruction due to bronchiectasis lead to progressive airway
obstruction, hyperinflation, ventilation–perfusion mismatch and, together with a
decrease in respiratory muscle strength, eventually may cause (chronic) respira-
tory failure.
During an exacerbation of the disease, inflammation worsens, and both airway

obstruction and the amount of sputum increase, leading to detrimental effects
on gas exchange and patient condition. Although initially it was stated that NIV
should not be used routinely as a treatment for an exacerbation of CF or bronchi-
ectasis because the excessive amount of sputum would limit its effectiveness,
studies (although not RCTs) and several reports have shown that NIV can be of
benefit in this situation. The excessive amount of sputum does not seem to be
a major problem; indeed, two studies have shown that during an exacerbation,
NIV might even aid in sputum mobilisation. It has been hypothesised that NIV is
of benefit during sputum expectoration manoeuvres as it might prevent respira
tory muscle fatigue and airway closure during prolonged expirations, leading to
improvement of alveolar ventilation and better sputum mobilisation. In practice,
NIV has become a therapy worth trying during a CF exacerbation with acute-
on-chronic respiratory failure in order to prevent intubation, with its associated
complications.
Non-CF bronchiectasis Bronchiectasis might develop in association with several

conditions other than CF (e.g. in pulmonary TB or infections with nontuberculous
mycobacteria, Kartagener’s syndrome, or due to prior pneumonia) or might occur

without a known cause (idiopathic). These patients often die from bronchiectasis-
related disease with or without ARF. Similar to CF, it is often thought that the
excessive amounts of sputum characteristic of non-CF bronchiectasis would limit
the use of NIV in these patients, especially during an acute exacerbation of the dis-
ease, which is often elicited by infection. However, NIV might be of use in selected
cases with less severe ARF due to bronchiectasis exacerbations, in which it might
prevent ETI and shorten the length of ICU stay. One has to bear in mind that NIV
failure rates are high in more severe ARF. In a study by Phua et al. (2010), failure was
caused by an inability of the NIV to correct gas exchange sufficiently and not due to
problems with sputum handling. Recently, a case was presented with a patient with
exacerbated non-CF bronchiectasis being successfully treated using a pump-assisted
venovenous system for extracorporeal carbon dioxide removal as an alternative to ETI
following NIV failure. Although this is only a case, alternatives to IMV are more than
welcome, as mortality rates of up to 40% once on IMV have been reported.
Interstitial lung diseases

ILDs are a heterogeneous group of disorders of the lung parenchyma with a
known cause (such as those associated with drugs or collagen vascular disease)
or unknown cause (idiopathic pulmonary fibrosis (IPF) and interstitial idiopathic
pneumonia other than IPF, such as nonspecific interstitial pneumonia). As the
pulmonary interstitium, blood vessels and air spaces may be affected, diffusion
impairment, shunting and ventilation–perfusion mismatching may lead to re
spiratory failure. As a consequence, these diseases are characterised by breath-
lessness and increasing hypoxaemia.
Patients may experience acute deterioration of their condition without any
identifiable cause, histopathologically characterised by diffuse alveolar dam-
age. Deterioration may also be caused by an infection, heart failure or thrombo
embolism. In all cases of deterioration, severe hypoxaemia is the primary problem.
At late stages, the lungs become so stiff that the respiratory muscles can no longer
sustain the imposed load and relative hypoventilation may lead to hypercapnia on
top of the usually worsened hypoxaemia.
When ARF develops in ILDs, outcomes are very poor, especially in IPF. Use of
conventional IMV by ETI does not seem to improve outcome, especially in IPF
and especially when no reversible cause is present or no alternative, such as lung
transplantation, is provided. Little is known about NIV during an exacerbation of
ILDs. Theoretically, prevention of complications caused by IMV, such as ventilator-
associated pneumonia, may be of benefit. Some small observational studies have
reported that NIV can be an effective treatment to prevent ETI in patients with less
severe disease. Results seem to be better with non-IPF ILD. However, in more than
half of cases, NIV is unsuccessful and, especially in these cases, mortality rates are
equally high.
In general, a more “lung-protective” ventilatory strategy with low inspiratory and
expiratory pressures is advised, as the lungs are usually already very damaged. As
the primary problem is usually hypoxaemia, high FIO2 must be provided to achieve
adequate oxygenation. However, the optimal ventilatory strategies are unproven
and numbers of study subjects are too small to make conclusions; consideration
of ventilatory strategies is mainly on theoretical grounds. Overall, it seems that

with progressive disease, adjustments made to the ventilator are often unsuc-
cessful in correcting gas exchange sufficiently. Insufficient ventilation on NIV may
warrant the institution of other treatment options, such as extracorporeal mem-
brane oxygenation, especially once a reversible cause is expected or when lung
transplantation is an option. However, NIV might be used as a palliative measure
in patients with do-not-intubate orders to reduce dyspnoea sensation. However,
little is known about quality of life during these last hours in patients on NIV com-
pared with other attractive alternatives available to relieve dyspnoea once death
is approaching.
Further reading
• Alzeer AH, et al. (2007). Survival of bronchiectatic patients with respiratory failure
in ICU. BMC Pulm Med; 7: 17.
• Arcaro G, et al. (2014). The successful management of a patient with exacerbation
of non-cystic fibrosis bronchiectasis and bilateral fibrothorax using a venovenous
extracorporeal carbon dioxide removal system. Respir Care; 59: e197–e200.
• Barker AF, et al. (2014). Obliterative bronchiolitis. N Engl J Med; 370: 1820–1828.
• Bellemare F, et al. (1982). Effect of pressure and timing of contraction on human
diaphragm. J Appl Physiol; 53: 1190–1195.
• Bellemare F, et al. (1983). Force reserve of the diaphragm in patients with chronic
obstructive pulmonary disease. J Appl Physiol; 55: 8–15.
• British Thoracic Society Standards of Care Committee (2002). Non-invasive venti-
lation in acute respiratory failure. Thorax; 57: 192–211.
• Carson KV, et al. (2014). Noninvasive ventilation in acute severe asthma: current
evidence and future perspectives. Curr Opin Pulm Med; 20: 118–123.
• De Backer L, et al. (2011). The effects of long-term noninvasive ventilation in
hypercapnic COPD patients: a randomized controlled pilot study. Int J Chron
Obstruct Pulmon Dis; 6: 615–624.
• Ganesh A, et al. (2015). Use of noninvasive ventilation in adult patients with acute
asthma exacerbation. Am J Ther [In press DOI: 10.1097/MJT.0000000000000184].
• Gifford AH (2014). Noninvasive ventilation as a palliative measure. Curr Opin
Support Palliat Care; 8: 218–224.
• Holland AE, et al. (2003). Non-invasive ventilation assists chest physiotherapy in
adults with acute exacerbations of cystic fibrosis. Thorax; 58: 880–884.
• Lazner MR, et al. (2012). Non-invasive ventilation for severe bronchiolitis: analysis
and evidence. Pediatr Pulmonol; 47: 909–916.
• Mallick S (2008). Outcome of patients with idiopathic pulmonary fibrosis (IPF) ven-
tilated in intensive care unit. Respir Med; 102: 1355–1359.
• Mason RJ, et al. (2010). Murray and Nadel’s Textbook of Respiratory Medicine. 5th
Edn. Philadelphia, Saunders.

• Mollica C , et al. (2010). Mechanical ventilation in patients with end-stage idiopathic

pulmonary fibrosis. Respiration; 79: 209–215.
• Moran F, et al. (2013). Non-invasive ventilation for cystic fibrosis. Cochrane
Database Syst Rev; 4: CD002769.
• Pallin M, et al. (2014). Noninvasive ventilation in acute asthma. J Crit Care; 29:
586–593.
• Papiris S, et al. (2002). Clinical review: severe asthma. Crit Care; 6: 30–44.
• Phua J, et al. (2010). Noninvasive and invasive ventilation in acute respiratory fail-
ure associated with bronchiectasis. Intensive Care Med; 36: 638–647.
• Placidi G, et al. (2001). Short-term effects of positive airway pressure on sputum
clearance by directed coughing: a crossover randomized study. Pediatr Pulmonol;
32: Suppl. 22, 313.
• Ram FS, et al. (2004). Non-invasive positive pressure ventilation for treatment of
respiratory failure due to exacerbations of chronic obstructive pulmonary disease.
Cochrane Database Syst Rev; 3: CD004104.
• Soma T, et al. (2008). A prospective and randomized study for improvement of
acute asthma by non-invasive positive pressure ventilation (NPPV). Intern Med;
47: 493–501.
• Tomii K, et al. (2010). Role of non-invasive ventilation in managing life-threat-
ening acute exacerbation of interstitial pneumonia. Intern Med; 49: 1341–1347.
• Vestbo J, et al. (2013). Global strategy for the diagnosis, management, and preven-
tion of chronic obstructive pulmonary disease: GOLD executive summary. Am J
Respir Crit Care Med; 187: 347–365.
• Vianello A, et al. (2014). Noninvasive ventilation in the event of acute respiratory
failure in patients with idiopathic pulmonary fibrosis. J Crit Care; 29: 562–567.
• Westhoff M (2015). Akut-auf-chronisches respiratorisches Versagen bei intersti-
tiellen Pneumonien. [Acute on chronic respiratory failure in interstitial pneumo-
nias]. Med Klin Intensiv Med Notfmed; 110: 188–196.

The patient with acute hypoxaemic

respiratory failure excluding
pulmonary oedema
Pongdhep Theerawit, Yuda Sutherasan and Paolo Pelosi
NIV has been used for patients with respiratory failure since the 1940s. Initially,
NIV aimed to support ventilatory respiratory failure, e.g. during exacerbations of
COPD. Later, the benefit of NIV was confirmed for cardiogenic pulmonary oedema.
Nevertheless, the application of NIV in hypoxaemic respiratory failure is still con-
troversial. In acute hypoxaemic respiratory failure (mostly patients with ARDS), the
pathogenesis and physiological alterations are substantially different from those of
cardiogenic pulmonary oedema. The alveolar damage, including injury of pulmonary
capillary vessels, occurs in almost the entire lung, leading to fluid leakage into pul-
monary tissues and marked inflammation. Although invasive mechanical ventilation
is a standard treatment as a protective strategy in these patients, NIV can also play a
role in acute hypoxaemic respiratory failure. The benefits of NIV, in terms of improve-
ments in gas exchange abnormalities, protective effects for lung injury and improved
mortality outcomes, need be addressed when selecting patients. Selecting suitable
patients should be performed on an individual basis. The aims of this section are to
describe the physiological effects of NIV in acute hypoxaemic respiratory failure, for
alleviation of symptoms as well as for improvement of mortality outcomes.
Physiological rationale for NIV in acute hypoxaemic respiratory failure
Currently, ventilator strategies for protection of injured lungs are recommended
by the standard guidelines for invasive mechanical ventilation. PEEP should be
applied at a high enough level to reduce cyclic opening and closing of injured
alveoli, whereas the V T should be low to prevent injury by overstretching.
Although the use of neuromuscular blocking agents in early ARDS can improve
the adjusted 90-day mortality and increase ventilator-free days, an animal model
Key points
• NIV can improve oxygenation in acute hypoxaemic respiratory
failure.
• NIV relieves dyspnoea by reducing the work of breathing.
• As there is a high rate of NIV failure for patients with acute
hypoxaemic respiratory failure, it is very important that there is
appropriate patient selection, intensive monitoring and prompt
intubation if signs of NIV failure develop.

of acute hypoxaemic respiratory failure has demonstrated improvement of lung

aeration during spontaneous breathing in the ventilator mode of airway pressure
release ventilation compared with controlled ventilation (Wrigge et al., 2003).
This is explained by the active diaphragmatic function in the synchronised ven-
tilator setting, promoting alveolar recruitment in the dependent lung area. As a
result of this, the synchronised spontaneous breathing provided by mechanical
ventilation, including NIV, may be of benefit to acute hypoxaemic respiratory failure
patients. NIV has also been demonstrated to reduce the work of breathing in non-
severe ARDS. After applying NIV, the patients feel more comfortable, as shown by a
reduction in respiratory rate and parameters related to work of breathing.
Improvement of oxygenation during NIV has been observed in ARF patients. This
is caused by several mechanisms:
• Modern NIV machines can provide a high concentration of oxygen, which is
adjustable, as with the invasive mechanical ventilator.
• The applied CPAP can work as PEEP in order to recruit and keep open the col-
lapsed alveoli. This increases functional residual capacity and makes the lungs
more uniform.
• The CPAP decreases the gradient pressure across pulmonary capillaries, prob-
ably resulting in decreased extravascular lung water.
In terms of inflammation, application of bilevel positive airway pressure ventilation
can reduce the concentrations of inflammatory cytokines such as interleukin-8
measured in serum and bronchoalveolar lavage fluid. In addition, NIV minimises
alveolar injury, as shown by the lung wet-to-dry weight ratio and the pathological
score of lung injury, in an animal model with ARDS.
NIV and outcomes in acute hypoxaemic respiratory failure
NIV to avoid intubation in ARDS patients NIV has been used as the first-line
ventilatory mode for ARDS in many centres. A multicentre survey by Antonelli
et al. (2007) revealed that in ∼30% of 147 cases of ARDS, NIV was given as
the first-line treatment. An important issue to note when using NIV as a first-
line treatment for ARDS is the high failure rate. This failure rate may be as
high as 50%, especially in patients with severe community-acquired pneumo-
nia (CAP). Moreover, these patients in whom NIV fails have a very high mortal-
ity rate. However, providing NIV in ARDS patients can improve oxygenation,
decrease work of breathing and decrease rate of ventilator-associated pneu-
monia. Thille et al. (2013) presented data from 113 acute hypoxaemic respira-
tory failure patients, with 82 cases of diagnosed ARDS. The rate of NIV failure
in ARDS was 61% and increased according to the severity of ARDS. The mor-
tality rate in the ICU was 25%. Among acute hypoxaemic respiratory failure
patients, ARDS patients had a higher mortality rate than non-ARDS patients.
Interestingly, the mortality rate of moderate-to-severe ARDS patients intubated
after NIV failure was similar to those intubated without prior NIV (47% versus
44%). Additionally, no difference in time to intubation was observed between
survivors and nonsurvivors in patients receiving NIV and requiring intubation
within the first 96 h.
NIV in pneumonia Applying NIV in severe pneumonia shows a high rate of fail-
ure. A prospective observational study conducted by Jolliet et al. (2001) enrolled
24 severe CAP patients who required ICU admission. The definition of severe
pneumonia included bilateral or multilobar involvement, a respiratory rate of
>30 breaths⋅min−1, PaO2/FIO2 ratio <250 mmHg (<33.3 kPa), presence of hypo-
tension and use of a vasopressor. With clearly defined criteria for intubation, 16
(66%) patients had NIV failure and eight (33%) patients died, all of whom were
intubated. As expected, the NIV group had a shorter ICU stay. However, early pro-
vision of NIV to severe CAP patients may be more effective, as shown by Nicolini
et al. (2014), who demonstrated a 25% failure rate in 127 cases of severe CAP
(PaO2/FIO2 ratio <250 mmHg (<33.3 kPa)).
NIV in immunocompromised patients The outcomes of applying NIV in immunocom-
promised patients have been shown in RCTs. Antonelli et al. (2000) administered
NIV to solid organ transplantation patients, and compared it with standard oxygen
therapy in a total of 40 randomised patients. ∼40% of patients had ARDS and 10%
had pneumonia. The NIV group showed higher improvement of PaO2/FIO2 than the
standard oxygen therapy group. Four (20%) patients on NIV required intubation.
Among these, three patients had ARDS and one patient had pneumonia. The failure
rate was significantly higher in the standard oxygen therapy group. The numbers of
ARDS and pneumonia patients requiring intubation were similar for both groups.
The length of ICU stay in survivors receiving NIV was shorter than for standard oxy-
gen therapy. Four patients died in the ICU (three patients developed ARF from ARDS
and one from pneumonia). The hospital mortality in the NIV group was 35% (seven
cases) and there was no statistically significant difference in mortality between the
NIV and standard oxygen therapy groups.
Another RCT by Hilbert et al. (2001) included diverse immunocompromised
patients: some with haematological malignancy, and some with other diseases
requiring immunosuppressive agents and/or organ transplantation. A total of
52 patients was randomly assigned to receive NIV or standard oxygen therapy.
Failure occurred in 12 (46%) patients in the NIV group, significantly fewer than
in the standard oxygen therapy group. Most of the failure patients had haema-
tological malignancies. The mortality rate was found to be significantly lower in
the NIV group than in the standard oxygen therapy group (10 versus 18 patients,
respectively). A comparison between NIV and invasive mechanical ventilation as
the type of ventilation initially provided in a cohort study in haematological malig-
nancy showed no difference in mortality rate but a high failure rate for NIV (69%)
(Depuydt et al., 2004). Patients who failed with NIV had a very high mortality rate
(91.7%). Patients successful with NIV had a mortality rate of 37.5%.
Evidence from meta-analyses and systematic reviews
In a meta-analysis by Agarwal et al. (2010), 13 studies with diverse causes of ARDS
were included. The pooled intubation rate was 48% and the mortality rate was
35%. Another meta-analysis by Luo et al. (2014), including six RCTs comparing
NIV and standard oxygen therapy, found a favourable outcome for NIV regard-
ing a decrease in endotracheal intubation rate, but a mortality benefit was not
demonstrated.
A systematic review in chest trauma patients also showed some benefits of
NIV. Chiumello et al. (2013) included four RCTs and six observational studies
in their systematic reviews. The results showed benefits in terms of improve-
ment in the PaO2/FIO2 ratio (data from five studies: one RCT and four prospective
observational studies) and reduction of mortality (data from four RCTs and one
prospective observational study).
Which acute hypoxaemic respiratory failure patients are suited to NIV?

Because of the high NIV failure rate and perhaps high mortality among patients
with ARDS in whom NIV fails, appropriate patient selection is needed for the prac-
tice of NIV.
• ARDS patients who develop shock should avoid NIV as the first-line mechanical
ventilation.
• A very low PaO2/FIO2 ratio at baseline or its deterioration after 1 h of applying
NIV frequently results in NIV failure.
• Severity of the disease causing respiratory failure, with a high severity score
such as Simplified Acute Physiology Score II (SAPS II), is another risk factor for
failure.
As a result of this, we generally suggest performing a trial of NIV in patients
with acute hypoxaemic respiratory failure with ARDS if the patients have a
PaO2/FIO2 ratio between 200 and 300 mmHg (26.6 and 39.9 kPa) (mild ARDS)
and good consciousness. In subgroups of patients with chest trauma, pneu-
monia or immunocompromise, similar criteria can be used to assess whether
NIV is suitable for the first-line management of acute hypoxaemic respiratory
failure. Among these patient subgroups, the chest trauma patients seem to have
better outcomes. No study has demonstrated an association between the mode
of ventilation or the interface used for NIV and failure of NIV. Bilevel positive
airway pressure ventilation and oronasal interfaces have commonly been used.
The selection of an appropriate mask and applying it without leakage are also
very important.
The settings for NIV in acute hypoxaemic respiratory failure patients may be
started with IPAP of 10 cmH2O and EPAP of 10 cmH2O. However, final settings
should be adjusted according to patient comfort and oxygenation. The target of
gas exchange is to keep PaO2 above 60 mmHg (8.0 kPa) and to optimise ventilation.
In case of hypotension or shock and high SAPS II score, NIV should be avoided,
especially in severe CAP. After a trial period of 1 h, all ARDS patients receiving NIV
must be reassessed. Endotracheal intubation with appropriate mechanical ven-
tilator settings should be provided if one of the following parameters is present:
worsening of chest radiography, decreased PaO2/FIO2 ratio, deteriorated conscious-
ness, decreased blood pressure or development of organ failure.
Further reading
• Agarwal R, et al. (2010). Role of noninvasive ventilation in acute lung injury/
acute respiratory distress syndrome: a proportion meta-analysis. Respir Care; 55:
1653–1660.
• Antonelli M, et al. (2000). Noninvasive ventilation for treatment of acute respira-
tory failure in patients undergoing solid organ transplantation: a randomized trial.
JAMA; 283: 235–241.

• Antonelli M, et al. (2007). A multiple-center survey on the use in clinical practice

of noninvasive ventilation as a first-line intervention for acute respiratory distress
syndrome. Crit Care Med; 35: 18–25.
• Brochard L, et al. (2014). Noninvasive ventilation for patients with hypoxemic
acute respiratory failure. Semin Respir Crit Care Med; 35: 492–500.
• Chiumello D, et al. (2013). Noninvasive ventilation in chest trauma: systematic
review and meta-analysis. Intensive Care Med; 39: 1171–1180.
• Delclaux C, et al. (2000). Treatment of acute hypoxemic nonhypercapnic respira-
tory insufficiency with continuous positive airway pressure delivered by a face
mask: a randomized controlled trial. JAMA; 284: 2352–2360.
• Depuydt PO, et al. (2004). Outcome in noninvasively and invasively ventilated
hematologic patients with acute respiratory failure. Chest; 126: 1299–1306.
• Ferrer M, et al. (2003). Noninvasive ventilation in severe hypoxemic respiratory
failure: a randomized clinical trial. Am J Respir Crit Care Med; 168: 1438–1444.
• Hernandez G, et al. (2010). Noninvasive ventilation reduces intubation in chest
trauma-related hypoxemia: a randomized clinical trial. Chest; 137: 74–80.
• Hilbert G, et al. (2001). Noninvasive ventilation in immunosuppressed patients
with pulmonary infiltrates, fever, and acute respiratory failure. N Engl J Med; 344:
481–487.
• Jolliet P, et al. (2001). Non-invasive pressure support ventilation in severe commu-
nity-acquired pneumonia. Intensive Care Med; 27: 812–821.
• Luo J, et al. (2014). Can non-invasive positive pressure ventilation prevent endo
tracheal intubation in acute lung injury/acute respiratory distress syndrome? A
meta-analysis. Respirology; 19: 1149–1157.
• Nava S, et al. (2011). Noninvasive ventilation for patients with acute lung injury or
acute respiratory distress syndrome. Respir Care; 56: 1583–1588.
• Nicolini A, et al. (2014). Early non-invasive ventilation treatment for respiratory
failure due to severe community-acquired pneumonia. Clin Respir J [In press; DOI:
10.1111/crj.12184].
• Rocker GM, et al. (1999). Noninvasive positive pressure ventilation: successful
outcome in patients with acute lung injury/ARDS. Chest; 115: 173–177.
• Thille AW, et al. (2013). Non-invasive ventilation for acute hypoxemic respiratory
failure: intubation rate and risk factors. Crit Care; 17: R269.
• Wrigge H, et al. (2003). Spontaneous breathing improves lung aeration in oleic
acid-induced lung injury. Anesthesiology; 99: 376–384.
Online resources
• Scala R (2014). NIV in hypoxemic respiratory failure (excluding acute pulmonary
edema). ERS Course Noninvasive ventilation: basic concepts, Hanover 2014.
www.ers-education.org/events/courses/noninvasive-ventilation-basic-concepts,-
hanover-2014.aspx

The patient with acute hypoxaemic

failure and cardiogenic pulmonary
oedema
João C. Winck and Luís F. Azevedo
Acute cardiogenic pulmonary oedema (ACPO) is defined as pulmonary oedema

due to increased capillary hydrostatic pressure secondary to elevated pulmonary
venous pressure. It results from an acute rise in left atrial or left ventricular fill-
ing pressure leading to increased pulmonary venous pressure, and consequently,
raised capillary pressure.
Common causes of ACPO include ischaemia with or without myocardial infarc-
tion, exacerbation of chronic systolic or diastolic heart failure, and dysfunction of
the mitral or aortic valve. However, a diversity of situations can lead to ACPO in
the absence of heart disease, including primary fluid overload (e.g. due to blood
transfusion) and renovascular causes such as renal artery stenosis or acute renal
failure with oliguria and subsequent fluid retention.
ACPO is not necessarily a straightforward diagnosis to make, as often there are con-
current exacerbating features such as a concurrent lower respiratory tract infection
or respiratory comorbidities (e.g. COPD). Nevertheless, a typical history of parox-
ysmal nocturnal dyspnoea or orthopnoea suggests ACPO. A combination of these
symptoms with clinical signs of jugular venous distension, inspiratory crackles (and
Key points
• Paroxysmal nocturnal dyspnoea or orthopnoea with clinical
signs of jugular venous distension, inspiratory crackles (and
often rhonchi) on lung examination, and radiographic signs
(increased heart size and central distribution of oedema)
suggest acute cardiogenic pulmonary oedema (ACPO).
• The standard medical therapy is an upright sitting position,
oxygen, diuretics and vasodilators.
• CPAP or bilevel positive airway pressure should be used in
patients not responding to standard medical therapy, to avoid
• One death can be avoided for every 14 ACPO patients treated
with noninvasive positive pressure ventilation.

When and how to start NIV in ACPO

• Respiratory rate >25 breaths·min−1; PaO2/FIO2 <200 mmHg (<26.6 kPa);
SpO2 <90%
• Use an oronasal mask
• Start with CPAP at 10 cmH2O
• If patients have not started to improve within 10–20 min then EPAP or
IPAP/EPAP may need to be increased
• Expect maximal benefit after 1 h
• A decrease in PaCO2 is expected after 1–2 h
• An increase in pH is expected 1–2 h after NIV initiation
• NIV is usually used for between 2 h and 24 h
• In cases of intolerance to CPAP, change to bilevel PAP (start with IPAP
15 cmH2O EPAP 5 cmH2O)
• Weaning from NIV if respiratory rate <25 breaths·min−1 and SpO2 >95%:
Step wise reduction (2 cmH2O) in IPAP/EPAP
Step wise reduction (10%) in FIO2
• Systolic blood pressure, heart rhythm and rate, SpO2 and urine output
should be monitored on a regular and frequent basis until the patient
has stabilised. Serial blood gases should be also performed (ideally via an
arterial line).
• Consider NIV failure if, after adequately titrated therapy: Glasgow Coma
Scale score >13, respiratory rate >40 breaths·min−1; PaO2<60 mmHg
(7.98 kPa), PaCO2 increase of 5 mmHg (0.67 kPa).
often rhonchi) on lung examination, radiographic signs (increased heart size and
central distribution of oedema), ECG changes (new ST segment and T wave changes),
and laboratory findings (brain natriuretic peptide level >500 pg·mL−1) confirms the
diagnosis. If available, a bedside echocardiogram should be performed, to confirm
the presence of elevated left ventricular filling pressures or raised left atrial pres-
sure, and assess the possible cause of ACPO (e.g. severe mitral regurgitation or
evidence of myocardial ischaemia with new regional wall abnormality).
ACPO is a rather common condition and may require mechanical v entilation,
leading to high in-hospital mortality. The use of NIV to treat ACPO was first described
by Poulton (1936) almost 80 years ago, and 17 years ago the first meta-analysis
by Pang et al. (1998) appeared, showing the efficacy of CPAP in the treatment of
ACPO. Since then, several RCTs comparing the use of CPAP and noninvasive posi-
tive pressure ventilation with standard medical therapy or with one another have
been published, and the role of NIV, and especially CPAP, in ACPO is becoming
more clearly defined.
According to the European Society of Cardiology ACPO guidelines, the patient
should be supported in an upright sitting position, and the key drugs are oxygen,
diuretics and vasodilators. Opiates and inotropes should be used more selectively.
Systolic blood pressure, heart rhythm and rate, SpO2 and urine output should be
monitored on a regular and frequent basis until the patient has stabilised.
For patients not responding to standard medical therapy, NIV can have an impor-
tant role. The first goal of NIV is to avoid endotracheal intubation (ETI). NIV
improves not only pulmonary congestion (by increasing functional residual capac-
ity, and improving respiratory mechanics and oxygenation), but also cardiac per-
formance (by reducing left ventricular afterload, venous return and cardiac preload
due to its effect of raising intrathoracic pressure).
The first trials on ACPO were performed with CPAP, but with technological pro-
gress new bilevel positive airway pressure (bilevel PAP) devices were introduced.
Although the latter were more effective at unloading the respiratory muscles than
CPAP, concerns about their safety followed an RCT showing higher myocardial
infarction rates in the bilevel PAP group. Further studies and meta-analyses dem-
onstrated that there was no such safety issue.
Evidence prior to the 3CPO (three interventions in cardiogenic pulmonary
oedema) trial
Our meta-analysis published in 2006 (Winck et al., 2006) showed that both CPAP
and bilevel PAP decreased ETI, but only CPAP reduced mortality. Moreover, there
was no difference in the risk of acute myocardial infarction between CPAP and
bilevel PAP. At this point some important questions arose, concerning efficacy in
some sub-populations and the ideal ventilator parameters.
Are hypercapnic patients better responders to bilevel PAP? In the subgroup analysis of
our meta-analysis, we could not find any difference in efficacy (in terms of intub
ation and mortality) of bilevel PAP versus CPAP considering patients with PaCO2
greater than or less than 50 mmHg (6.65 kPa).
Do patients respond quickly to bilevel PAP and to higher span pressures? In the meta-
analysis by Ho et al. (2006) the duration of NIV required until the pulmonary
oedema resolved was not different between CPAP and bilevel PAP. Moreover in
our meta-analysis, using PSV of more or less than 10 cmH2O did not change the
intubation and mortality rates.
Which is better: fixed or variable pressures? In the meta-analysis by Ho et al. (2006),
subgroup analysis of studies that used either fixed or titrated pressure during
bilevel PAP did not change the results in terms of mortality and intubation rates.
The 3CPO trial

Gray et al. (2008) published the largest trial to date (the 3CPO trial), including
more than 1000 patients, and found no effect of NIV on ETI and mortality. This
was in contrast with previous trials and meta-analyses. Some points in the study
by Gray et al. (2008) deserve comment as the patients included had less severe
disease than those included in previous trials, being less hypoxaemic and having
lower ETI and mortality rates (fig. 1b and fig. 2b). Indeed, these figures show that
the study by Gray et al. (2008) had the lowest ETI rate and the second lowest
mortality rate among all the trials analysed. Moreover, there were high rates of
patient crossover, with more than 17% of controls ending up using NIV, leading to
underestimation of the NIV effect. In addition, the devices used in the 3CPO trial
were not able to titrate FIO2 correctly.
Evidence post-3CPO
Incorporating the data of Gray et al. (2008) into the pooled analysis as published in
our meta-analysis, a significant 18% and 11% risk reduction for ETI and mortality,
a) Risk difference Risk difference

First author (year) Weight IV, random (95% CI) IV, random (95% CI)
Park (2001) 4.3% –0.07 (–0.50–0.37)

Takeda (1998) 5.7% –0.55 (–0.89–i–0.20)
Rasanen (1985) 6.9% –0.30 (–0.59–i–0.01)
Takeda (1997) 7.3% –0.33 (–0.61–i–0.06)
Bersten (1991) 9.0% –0.35 (–0.57–i–0.13)
Park (2004) 9.1% –0.35 (–0.56–i–0.13)
Crane (2004) 9.5% 0.15 (–0.05–0.35)
Lin (1995) 10.6% –0.20 (–0.37–i–0.03)
L’Her (2004) 10.9% –0.21 (–0.37–i–0.05)
Kelly (2002) 12.6% –0.06 (–0.17–0.04)
Gray (2008) 14.2% –0.00 (–0.03–0.02)
Total (95% CI) 100.0% –0.18 (–0.29–i–0.07)

Total events
Heterogeneity: Tau2=0.02; Chi2=50.34
df =10 (p<0.00001); I2=80% –1 –0.5 0 0.5 1
Test for overall effect: Z =3.31 (p=0.0009) Favours experimental Favours control
b) 0.8
0.7
0.6
Intervention group risk
0.5
0.4
0.3
0.2
0.1
Gray et al. (2008)
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Control group risk
Figure 1. a) Forest plots for the pooled analysis comparing CPAP and standard medical
therapy for the ETI outcome, as published in Winck et al. (2006). The meta-analysis was
performed using the random effects (random) model with the inverse variance (IV) weighting
method. A complete list of references can be found in the online data supplement. b) L’Abbé
plots representing the intervention group risk as a function of control group risk for the
outcome of ETI for the studies included in our meta-analysis, adding and highlighting
the results of Gray et al. (2008) (filled circle). Points in the plot have a size proportional to
the precision of the point estimates. The dotted line represents the unweighted regression
line through the data and the solid line represents the line of no effect (control group risk
equals intervention group risk). The study by Gray et al. (2008) has the lowest ETI rate
among the trials analysed. Reproduced and modified from Winck et al. (2006) with per-
mission from the publisher, and data from Gray et al. (2008).

a) Risk difference Risk difference

First author (year) Weight IV, random (95% CI) IV, random (95% CI)
Takeda (1998) 4.5% –0.55 (–0.88–i–0.21)

Rasanen (1985) 6.5% –0.15 (–0.40–0.10)
Park (2001) 6.5% 0.11 (–0.14–0.36)
Takeda (1997) 7.1% –0.13 (–0.37–0.11)
Bersten (1991) 7.6% –0.09 (–0.32–0.13)
Crane (2004) 8.3% –0.30 (–0.51–i–0.09)
L’Her (2004) 9.2% –0.03 (–0.21–0.16)
Kelly (2002) 9.8% –0.15 (–0.33–0.03)
Park (2004) 9.8% –0.19 (–0.37–i–0.02)
Lin (1995) 13.3% –0.04 (–0.16–0.08)
Gray (2008) 17.5% –0.00 (–0.05–0.04)
Total (95% CI) 100.0% –0.11 (–0.19–i–0.03)

Total events
Heterogeneity: Tau2=0.01; Chi2=25.12
df =10 (p=0.005); I2=60% –1 –0.5 0 0.5 1
Test for overall effect: Z =2.68 (p=0.007) Favours experimental Favours control
b) 0.7
0.6
Intervention group risk
0.5
0.4
0.3
0.2
Gray et al.
(2008)
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Control group risk
Figure 2. a) Forest plots for the pooled analysis comparing CPAP and standard medical ther-
apy for the mortality outcomes, as published in Winck et al. (2006). The meta-analysis was
performed using the random effects (random) model with the inverse variance (IV) weighting
method. A complete list of references can be found in the online data supplement. For Gray et
al. (2008), 7-day mortality was considered. b) L’Abbé plots representing the intervention group
risk as a function of control group risk for the mortality outcome for the studies included in our
meta-analysis, adding and highlighting the results of Gray et al. (2008) (filled circle). Points
in the plot have a size proportional to the precision of the point estimates. The dotted line rep-
resents the unweighted regression line through the data and the solid line represents the line
of no effect (control group risk equals intervention group risk). The study by Gray et al. (2008)
had the second lowest mortality rate among all the trials analysed. Reproduced and modified
from Winck et al. (2006) with permission from the publisher, and data from Gray et al. (2008).

respectively, is still evident (fig. 1b and fig. 2b). Indeed, a subsequent meta-analysis

by Weng et al. (2010), including the 3CPO trial, demonstrated a significant reduc-
tion of mortality rate with CPAP. Although discrepancies between meta-analyses
and large trials exist, this does not preclude their importance and shows their
relevance as tools for the analysis of evidence and its heterogeneity on a higher
level.
A more recent meta-analysis published by Mariani et al. (2011) showed that
at 1 h, both CPAP and bilevel PAP were associated with a nonsignificant trend
towards higher PaO2 values compared with standard medical therapy, and that
bilevel PAP significantly increased PaO2 compared with CPAP. This study shows that
noninvasive positive pressure ventilation is associated with more rapid and bet-
ter oxygenation than CPAP and confirms previous studies suggesting that either
form of NIV reduces the need for ETI and reduces mortality, although in this case
the significant mortality reduction of noninvasive positive pressure ventilation
was only apparent when considering the evidence coming from direct and indirect
comparisons in a network meta-analysis context. This study supports the idea that
one death can be avoided for every 14 ACPO patients treated with noninvasive
positive pressure ventilation; similarly, one death can be prevented for every nine
ACPO patients treated with CPAP. This means that 69 lives out of 1000 could be
saved with the addition of noninvasive positive pressure ventilation to standard
medical therapy.
Simplified CPAP systems for ACPO

Comparison of CPAP delivered by a simple Boussignac valve (Vygon, Ecouen,
France) versus a standard closed-circuit Drager CF800 CPAP system (Drager,
Lubeck, Germany) showed similar reductions in PaCO2 and respiratory rate, and
peripheral SaO2 increase, without any significant differences in disposition from the
emergency department or complication rate. Thus, this less expensive and more
transportable equipment seems to be a good alternative in this clinical setting.
CPAP in the pre-hospital setting

A recent meta-analysis by Goodacre et al. (2014) showed a reduction in the num-
ber of intubations and mortality in patients with ARF who received CPAP com-
pared with bilevel PAP in the pre-hospital setting.
Conclusions
• CPAP and bilevel PAP both significantly decrease ETI and mortality, although
the evidence for mortality reduction is stronger for CPAP.
• There is no superiority of bilevel PAP, so CPAP could be considered as the first-
line intervention in ACPO.
• Although some caution is still advised, there is no evidence of an increased risk
of acute myocardial infarction with any of the techniques.
• In case of intolerance or refractoriness to CPAP, bilevel PAP may be tried as it
seems to be faster for improving hypoxaemia.
• NIV should be widely available in coronary units and emergency rooms.
• NIV ventilators with waveform display and oxygen blending are preferable.
• CPAP valves (like the Boussignac valve) are easily available and may be useful
for pre-hospital treatment.
Further reading
• Chadda K, et al. (2002). Cardiac and respiratory effects of continuous positive air-
way pressure and noninvasive ventilation in acute cardiac pulmonary edema. Crit
Care Med; 30: 2457–2461.
• Goodacre S, et al. (2014). Prehospital noninvasive ventilation for acute respira-
tory failure: systematic review, network meta-analysis, and individual patient data
meta-analysis. Acad Emerg Med; 21: 960–970.
• Gray A, et al. (2008). Noninvasive ventilation in acute cardiogenic pulmonary
edema. N Engl J Med; 359: 142–151.
• Ho KM, et al. (2006). A comparison of continuous and bi-level positive airway
pressure non-invasive ventilation in patients with acute cardiogenic pulmonary
oedema: a meta-analysis. Crit Care; 10: R49.
• Leman P, et al. (2005). Simple lightweight disposable continuous positive airways
pressure mask to effectively treat acute pulmonary oedema: randomized con-
trolled trial. Emerg Med Australas; 17: 224–230.
• Lenique F, et al. (1997). Ventilatory and hemodynamic effects of continuous posi-
tive airway pressure in left heart failure. Am J Respir Crit Care Med; 155: 500–505.
• Mariani J, et al. (2011). Noninvasive ventilation in acute cardiogenic pulmonary
edema: a meta-analysis of randomized controlled trials. J Card Fail; 17: 850–859.
• McMurray JJ, et al. (2012). ESC Guidelines for the diagnosis and treatment of acute
and chronic heart failure 2012. Eur Heart J; 33: 1787–1847.
• Mehta S, et al. (1997). Randomized, prospective trial of bilevel versus continuous
positive airway pressure in acute pulmonary edema. Crit Care Med; 25: 620–628.
• Pang D, et al. (1998). The effect of positive pressure airway support on mortality
and the need for intubation in cardiogenic pulmonary edema: a systematic review.
Chest; 114: 1185–1192.
• Poulton PE (1936). Left-sided heart failure with pulmonary oedema: its treatment
with “the pulmonary plus pressure machine”. Lancet; 228: 981–983.
• Ware LB, et al. (2005). Acute pulmonary edema. N Engl J Med; 353: 2788–2796.
• Weng CL, et al. (2010). Meta-analysis: noninvasive ventilation in acute cardiogenic
pulmonary edema. Ann Intern Med; 152: 590–600.
• Winck JC, et al. (2006). Efficacy and safety of non-invasive ventilation in the treat-
ment of acute cardiogenic pulmonary edema – a systematic review and meta-
analysis. Crit Care; 10: R69.
Online resources
• Köhnlein T (2014). NIV in acute and chronic cardiac diseases. ERS Course
Noninvasive ventilation: basic concepts, Hanover. www.ers-education.org/events/
courses/noninvasive-ventilation-basic-concepts,-hanover-2014.aspx

Paediatric indications
Acute NIV in children, including

ventilator and interface choice
Alessandro Amaddeo and Brigitte Fauroux
ARF is one of the most common causes of ICU admission in children. Invasive
mechanical ventilation (IMV) is the classical treatment for severe ARF. However,
endotracheal intubation (ETI) is associated with upper airway trauma, lung injury
due to use of high pressures or high volumes, or ventilator associated lung infec-
tions. These side-effects are not observed with NIV, explaining why this ventilatory
support is increasingly used in children.
The efficacy of NIV for different causes of ARF has been widely proven in adults,
but less evidence is available in children. This may be explained by the greater
heterogeneity of the paediatric population and the lack of available and appro-
priate equipment. In recent years, the development of ventilators adjusted for
paediatric use and of interfaces designed for infants has enlarged the spectrum
of children treated with NIV. The main advantage of NIV is that it is, by definition,
noninvasive and that it can be used on demand. However, NIV is less efficient than
IMV because of unavoidable air leaks and the difficulty of “around the clock” use.
Importantly, the prompt recognition of early predictors of NIV failure is crucial for
the patient’s safety. This section gives a brief summary of the possible indications
and benefits of NIV in paediatric ARF, and highlights areas for future research.
Indications
Hypercapnic respiratory failure (type 2) ARF in children may result from various
pathophysiological mechanisms. The first mechanism is an imbalance between
the load imposed on the respiratory muscles and their capacity. When this imbal-
ance exceeds a certain threshold, hypoventilation occurs. Diseases that may be
responsible for this type of ARF, also known as type 2 respiratory failure, cause an
increase in the work of breathing due to upper or lower airway obstruction, or are
Key points
• NIV is increasingly used for ARF in children due to
improvements in ventilators and interfaces.
• Appropriate choice of the ventilator, the ventilator settings and
the interface is crucial for NIV success.
• NIV requires close monitoring to prevent any delay in
endotracheal intubation in case of NIV failure.

Table 1. Indications for NIV in paediatric ARF
Effective
Restrictive disease Lower airway obstruction
Neuromuscular diseases Bronchiolitis
Scoliosis (post-surgery) Asthma
Upper airway obstruction Parenchymal disease
Pierre Robin syndrome Community-acquired pneumonia
Treacher Collins syndrome Pneumonia in immunocompromised patients
Craniostenosis Acute exacerbation in CF
Pycnodysostosis Others
Achondroplasia Pulmonary cardiogenic oedema
Laryngomalacia ARF post-liver transplantation
Congenital or acquired Acute chest syndrome
laryngotracheal stenosis Post-operative respiratory failure
Other upper airway malformation Ventilator weaning
Storage diseases
Neck masses or tumours
Not effective
ARDS Sepsis
Interstitial lung disease
associated with respiratory muscle weakness such as neuromuscular disorders

(table 1). More rarely, disorders such as an acute brain injury (trauma, infection
or tumour) may depress neural ventilatory drive and cause acute hypercapnic
respiratory failure. The choice of ventilatory mode will depend on the underlying
pathophysiological mechanism. The aim of NIV will be:
• to unload the respiratory muscles in upper airway obstruction, CF and acute
viral bronchiolitis
• to assist or replace the respiratory muscles in neuromuscular disorders
In children with lower airway obstruction, such asthma or viral bronchiolitis, NIV
has been shown to improve gas exchange, to decrease the work of breathing
and to reduce the rate of intubation. In children with advanced CF lung disease
NIV has been shown to reduce the work of breathing, which translates into an
improvement in alveolar hypoventilation. As such, NIV is recommended as a first
line treatment for an acute hypercapnic respiratory exacerbation. In those with
upper airway obstruction, such as tracheo-laryngomalacia, stenosis or craniofacial
malformation, NIV may avoid tracheostomy and recannulation after extubation.
In severe restrictive scoliosis, NIV facilitates extubation after spinal surgery. NIV
may also be useful in moderate hypercapnic respiratory failure due to acute chest
syndrome in sickle cell disease, pneumonia and also during acute ARF after liver
transplantation. NIV has also been shown to facilitate extubation after cardiac
surgery in children with ARF due to heart disease.
Hypoxaemic respiratory failure (type 1) Hypoxaemic respiratory failure (also called

type 1 ARF) is characterised by a ventilation/perfusion mismatch resulting in
hypoxaemia and normocapnia or hypocapnia. In this case, increasing FIO2, rather
than enhancing alveolar ventilation, is the first line treatment. However, some
patients with type 1 ARF may still benefit from NIV, because the ventilation/perfu-
sion mismatch may be associated with airway obstruction, atelectasis, or respira-
tory muscle fatigue. As a result, NIV is effective in a large number of paediatric
causes of ARF (table 1).
NIV has little or no proven benefit in patients with ARDS due to parenchymal or
interstitial lung diseases, or when ARF is due to septic shock.
Choice of ventilator, settings and interface

Ventilators The availability of specific NIV modes on ICU ventilators and of inter-
faces adapted for children has allowed the development of NIV for ARF in children.
Two types of ventilator may be used:
1) ICU ventilators with a specific NIV mode, which use wall-supplied compressed
oxygen and air
2) Portable ventilators, which use room air and a servo-controlled air compressor
to deliver airflow and pressure to the patient
When using a portable ventilator oxygen supplementation can be provided by an
external source directly connected into the circuit or to the ventilator. ICU ven-
tilators are the best choice, because they have a better trigger and more precise
ventilation and monitoring. However, home ventilators designed for life support
have comparable performances and may deliver either CPAP, pressure- or volume-
targeted ventilation or a hybrid ventilation mode (pressure-targeted ventilation
with a preset delivered VT). The ventilator should be able to synchronise with
the child’s breathing pattern, which requires the ability to adjust the inspiratory
and expiratory trigger sensitivities, the pressurisation rate, the inspiratory time,
the backup respiratory rate and the VT. Most importantly, the ventilator should
be able to compensate for unintentional leaks, as leaks are the major cause of
patient–ventilator asynchrony. The ventilator should also allow monitoring of the
ventilation curves, the inspiratory and expiratory pressures, the flow rate, the VT,
the respiratory rate and the amount of unintentional leaks. Finally, alarms for
potentially hazardous situations such as power loss, high/low pressure, high/low
flow, leaks, patient disconnection, apnoea, and high/low estimated or measured
VT should be available.
Settings The ventilator settings depend strictly on the underlying disease. In

clinical practice pressure-targeted ventilation is the first choice for NIV, mainly
because it better compensates for unintentional leaks (table 2). For children with
upper airway obstruction or with a disease characterised by intrinsic PEEP (PEEPi),
such as bronchiolitis, CPAP is usually sufficient. Patients with restrictive disease,
or who require an EPAP of more than 10–12 cmH2O, need the addition of an IPAP
to achieve an adequate VT. The difference between inspiratory and expiratory pres-
sure should be at least 6–8 cmH2O to have a VT of 6–10 mL⋅kg−1 of ideal body
Table 2. Selection of ventilator setting according to the underlying condition
Condition Recommended ventilator settings

Upper airway obstruction and/or PEEPi CPAP
(e.g. upper airway obstruction, OSA
or bronchiolitis)
Requirement for CPAP >10–12 cmH2O Bilevel pressure ventilation with a
pressure difference between IPAP
and EPAP of ≥6–8 cmH2O to achieve
a VT of 6–10 mL⋅kg−1 of ideal body
weight
Restrictive lung disease Volume- or pressure-targeted ventilation
(e.g. neuromuscular or thoracic to achieve a VT of 10–15 mL⋅kg−1 of
scoliosis) ideal body weight
Zero or low EPAP
Backup rate (to avoid apnoea)
Lung disease (e.g. CF) Pressure-targeted ventilation with the
highest tolerated level of IPAP
Zero or low EPAP
Backup rate (to avoid apnoea)
weight. Patients with neuromuscular or lung disease who have no PEEPi may be
ventilated with a volume-targeted or hybrid mode to ensure adequate alveolar
ventilation, with the lowest possible EPAP. The correct settings for NIV represent a
major challenge. In the few clinical trials available, the NIV settings were generally
standardised, but studies adjusting NIV settings on an individual basis have shown
that the latter approach may be superior. Therefore, future studies should aim to
determine the optimal mode and settings for each type of ARF.
Interfaces Different types of interface may be used for NIV such as nasal masks,
nasal pillows (or nasal canula), oronasal masks, oral masks, mouthpieces or hel-
mets (see the section entitled “Choosing the interface”). Ideally, the best interface
should be:
• small (with minimal dead space)
• light-weight
• easy to fit and to remove
• have a headgear that confers stability preventing movement or dislocation of
the interface in order to minimise leaks
Full face or oronasal (facial) masks are generally preferred in the acute situation
because of frequent mouth breathing during ARF. The helmet is a very useful
interface in the acute setting because it has no direct contact with the patient’s
face, thus avoiding any skin injury. The choice of the interface is determined by:
• the patient’s age
• the facial morphology of the patient
• the ventilatory mode
Patients with upper airway obstruction or PEEPi may need CPAP. Therefore, use
of an interface with a manufactured leak, so-called “leak ventilation”, is a simple
and perfectly appropriate solution. Patients with neuromuscular or lung disease
are usually ventilated using a pressure- or volume-targeted mode without EPAP;
therefore, interfaces without manufactured leaks are required. In all cases the
interface associated with the best patient tolerance and comfort, as shown by the
absence of skin injury, pain, discomfort and leaks, should be used.
NIV in the ICU The choice of ventilator and interface in the ICU varies accord-
ing to the experience of the team and on the prompt availability of the equip-
ment. Physicians should be familiar with at least one or two ventilators and fully
understand the modes of ventilation available on those devices and how to use
them. Mild-to-moderate sedation may be useful for patients treated with NIV
in the ICU.
Indicators of NIV failure

In clinical practice, NIV is started in cases of mild-to-moderate ARF, at the point
at which ARF can be expected to be reversed or stabilised by a NIV and before the
progression to severe ARF requiring ETI and IMV. This window of opportunity may
be small and will vary according to the patient’s characteristics (e.g. the underly-
ing disease and age). The major difficulty is thus to define the optimal moment to
start NIV and to withdraw it without delaying an intubation when necessary. The
criteria used to initiate and stop NIV in paediatric studies have been mainly based
on clinical and gas exchange parameters. Future studies should aim to validate
these criteria for the different causes of ARF and patient profiles.
It has been shown that patients with type 1 ARF such as ARDS, sepsis, or with one
or more organ failures associated with ARF are more likely to fail NIV. Therefore,
careful selection of patients suitable for NIV is necessary. After NIV initiation, the
patient should be closely monitored to identify early signs of NIV failure to avoid
an unnecessary and deleterious delay of ETI. The parameters that have been
shown to be associated with NIV failure are a lack of improvement in respira-
tory rate, heart rate, carbon dioxide and oxygenation in the first hours after NIV
initiation (table 3). Indeed, a reduction in respiratory rate after 1–6 h is clearly
Table 3. Prognostic factors for NIV failure
Prognostic factors
Clinical parameters# No decrease in carbon dioxide tension
No decrease in oxygen requirements
No decrease in respiratory rate
No decrease in heart rate
Underlying conditions ARDS
Multiple organ failure
Septic shock
#: in the first 2–6 h after NIV initiation.

associated with NIV success. By contrast, an increase in FIO2, a FIO2 of >0.8 or a

decrease in PaO2/FIO2 after 1 h of NIV therapy are associated with a higher risk
of intubation. Moreover, a SpO2/FIO2 >190 at 1 h has been shown to be a reliable
predictor of early NIV failure in children.
Perspectives
Even if NIV is increasingly used for ARF in children, some issues remain controver-
sial. Since successful NIV depends on careful selection of the appropriate patient
at the correct time, we need more carefully conducted prospective studies in chil-
dren with various disorders in order to better identify which patients would ben-
efit most from NIV. Furthermore, even if great improvements have been made by
industry, the vast majority of ICU and portable ventilators were originally designed
for adults. Uncertainties remain about the performance of such ventilators when
dealing with paediatric patients, especially regarding trigger sensitivity and the
minimal volume and flow delivered. Finally, manufacturers should continue to
improve NIV interfaces, especially for younger children.
Further reading
• Abadesso C, et al. (2012). Non-invasive ventilation in acute respiratory failure in
children. Pediatr Rep; 4: e16.
• Cavari Y, et al. (2012). Non invasive positive pressure ventilation in infants with
respiratory failure. Pediatr Pulmonol; 47: 1019–1025.
• Dohna-Schwake C, et al. (2011). Non-invasive ventilation on a pediatric inten-
sive care unit: feasibility, efficacy, and predictors of success. Pediatr Pulmonol; 46:
1114–1120.
• Essouri S, et al. (2006). Noninvasive positive pressure ventilation: five years of
experience in a pediatric intensive care unit. Pediatr Crit Care Med; 7: 329–334.
• Fauroux B, et al. (2011). Why, when and how to propose noninvasive ventilation in
cystic fibrosis? Minerva Anestesiol; 77: 1108–1114.
• Fauroux B, et al. (2008). Performance of ventilators for noninvasive positive-pres-
sure ventilation in children. Eur Respir J; 31: 1300–1307.
• Gregoretti C, et al. (2010). Non-invasive ventilation in pediatric intensive care.
Minerva Pediatr; 62: 437–458.
• Hull J (2014). The value of non-invasive ventilation. Arch Dis Child; 99: 1050–1054.
• Leboulanger N, et al. (2013). Non-invasive positive-pressure ventilation in chil-
dren in otolaryngology. Eur Ann Otorhinolaryngol Head Neck Dis; 130: 73–77.
• Mayordomo-Colunga J, et al. (2009). Predictive factors of non invasive ventilation
failure in critically ill children: a prospective epidemiological study. Intensive Care
Med; 35: 527–536.
• Muñoz-Bonet JI, et al. (2010). Predictive factors for the outcome of noninva-
sive ventilation in pediatric acute respiratory failure. Pediatr Crit Care Med; 11:
675–680.

• Najaf-Zadeh A, et al. (2011). Noninvasive positive pressure ventilation for acute

respiratory failure in children: a concise review. Ann Intensive Care; 1: 15.
• Nava S, et al. (2009). Non-invasive ventilation in acute respiratory failure. Lancet;
374: 250–259.
• Niranjan V, et al. (1998). Noninvasive management of pediatric neuromuscular
ventilatory failure. Crit Care Med; 26: 2061–2065.
• Racca F, et al. (2015). Respiratory failure due to upper airway obstruction in chil-
dren: use of the helmet as bridge interface. Minerva Anestesiol; 81: 175–178.
• Wormald R, et al. (2009). Non-invasive ventilation in children with upper airway
obstruction. Int J Pediatr Otorhinolaryngol; 73: 551–554.
• Yañez LJ, et al. (2008). A prospective, randomized, controlled trial of noninvasive
ventilation in pediatric acute respiratory failure. Pediatr Crit Care Med; 9: 484–489.

Airway clearance and physiotherapy
Airway clearance methods and nebulised

therapy in acute NIV
Michelle Chatwin
Normal lungs produce around 100 mL of secretions per day. These secretions
protect the airway from inhaled irritants. Mucociliary activity, normal breath-
ing cycles and cough are the primary mechanisms of secretion removal. This is
because a cephalad airflow bias occurs with breathing. As we breathe in, the
airways expand, and as we breathe out, they narrow. Secretion movement is
also affected by other independent factors, such as secretion viscosity, elasticity,
depth of the liquid layer and the position of the airway. Physiologically, secre-
tion movement can be enhanced by generating higher expiratory airflow than
inspiratory airflow, aiming for a peak expiratory to inspiratory ratio greater than
1:1. By increasing VT, airway clearance can be enhanced. This uses the small
collateral channels of ventilation (channels of Lambert and Martin and pores of
Kohn). Resistance through these channels during breathing at normal VT is high
and, therefore, they remain closed. Taking a deep and long breath in opens these
small collateral channels of ventilation, allowing airflow to get behind secretions
and thus move them downstream.
Airway clearance techniques
During a common cold, respiratory muscle strength decreases by 10–15%,
even in healthy individuals. This can tip a patient who normally uses NIV to
have an increased ventilator requirement or to become NIV dependent. As
some airway clearance techniques require the patient to be able to take a deep
breath in and breathe out against resistance (e.g. positive expiratory pressure
with or without oscillation devices), these techniques are not suitable for the
Key points
• Airway clearance is an essential treatment in patients with
retained bronchopulmonary secretions.
• Secretion movement can be further enhanced by generating
higher expiratory airflow than inspiratory airflow.
• Airway clearance should be carried out in neuromuscular
disease patients when their SpO2 is <95% on room air.
• Aerosol drugs can be delivered in conjunction with NIV, if the
patient is unable to come off ventilatory support.

ventilator-dependent patient. Where possible, the airway clearance techniques

should be carried out in conjunction with NIV or any other device delivering
intermittent positive pressure. The patient should also be positioned in a com-
fortable supported position to decrease their work of breathing.
Conventional airway clearance techniques
Postural drainage (gravity-assisted positioning) and manual techniques may be
effective in clearing secretions, especially in patients who are unable to cooperate
with treatment (e.g. young children and patients who are intellectually impaired).
Manual techniques, including percussion, shaking and vibrations, can easily be
performed with the addition of NIV (fig. 1).
Breathing exercises
To further increase secretion removal by creating an expiratory airflow bias,
breathing exercises can be added, such as the active cycle of breathing technique
or autogenic drainage. However, this is only possible in patients who have pre-
served respiratory muscle function. As this is not the case in patients with weak
respiratory muscles or ventilator dependence, the simplest option is to modify
the active cycle of breathing technique. This technique consists of three compo-
nents: breathing control, thoracic expansion exercises and the forced expiration
technique. Breathing control, with normal VT and the ventilator's normal settings,
is followed by three or four thoracic expansion exercises (deep breaths empha-
sising inspiration, with an end-inspiratory hold followed by passive expiration),
which can be augmented by increasing the pressure or volume to provide a deeper
breath in. This may be followed by another period of breathing control and fur-
ther thoracic expansion exercises if required. The cycle concludes with the forced
expiration technique (or huff, as it is more commonly known). This is a medium-
sized breath in, followed by a fast breath out through an open glottis, using the
muscles of the chest and stomach to force the breath out, followed by a period of
a) b) c)
Continuum of treatment with the addition of ventilatory support or

increase of pressure or volume support
Figure 1. a) Percussion being performed while lying on the right side. Note the cupped
hands. Percussion is usually performed through a towel or thick clothing and not directly
onto the patient's skin. b) Percussion being performed in a supine position, targeting the
anterior aspects of the upper lobes of the right and left lungs. c) Shaking being performed
while lying on the right side, with the addition of NIV. These techniques can be performed
as a continuum of treatment for patients who require ventilatory support. Images courtesy
of the author.
breathing control. This can be performed by non-neuromuscular patients and can

facilitate further movement of secretions downstream. In patients with an ineffec-
tive cough, cough augmentation techniques can be performed.
Autogenic drainage can also be modified for the patient requiring NIV. Rather than
the patient breathing at different lung volumes, the therapist can hold them at
a low lung volume followed by a mid lung volume, “unstitching” and “collect-
ing” the secretions for the patient to then clear them with cough augmentation
techniques.
For further details on cough augmentation techniques, see the section e
ntitled
“Practicalities of and guide to cough augmentation and daytime mouthpiece
ventilation”.
Oscillatory devices
Oscillatory devices can also be used to assist secretion clearance in the acutely
unwell patient requiring NIV. Internal or external vibration of the chest is hypothe
sised to promote clearance of sputum from the peripheral bronchial tree. It is
hypothesised that the oscillations act as “mini coughs”. However, these devices
also produce higher expiratory than inspiratory airflows, which favour airway clear-
ance. One oscillatory technique is intrapulmonary percussive ventilation. This is
a modified method of intermittent positive pressure breathing, superimposing
high-frequency mini bursts of air (50–550 cycles·min−1) on the individual's intrin-
sic breathing pattern. This creates an internal vibration (percussion) within the
lungs. The device can be set up to provide ventilatory support and thus can assist
in moving secretions in the ventilator-dependent patient or those who fatigue
quickly.
In patients with obstructive lung disease who are not NIV dependent for 24 h a
day, the percussions should be set at a frequency >300 cycles·min−1 with a pres-
sure ranging from 10 to 20 cmH2O. These settings favour mucus transport over
ventilatory support. In patients who require ventilatory support, the frequency
of percussion is set slower (80–200 cycles·min−1) but the proximal airway pres-
sure may need to be as high as 40 cmH2O. As with all devices, the settings are
titrated to patient comfort and to achieve the goal of therapy. Indeed, the thera-
pist may choose to adjust the number of cycles·min−1, alternating between a high
frequency, which promotes the mobilisation of secretions, and a lower frequency
to allow the patient rest by targeting alveolar ventilation and secretions clearance.
High-frequency chest wall oscillation (HFCWO), or high-frequency chest wall
compression as it is also known, provides compression of the chest wall at fre-
quencies of 5–20 Hz. The compressive force is usually provided via an inflatable
jacket adjusted to fit snugly over the thorax. An air pulse generator then deliv-
ers intermittent positive airflow into the jacket. As the jacket expands, com-
pressing the chest wall, it produces a transient/oscillatory increase in airflow
in the airways, vibrating the secretions from the peripheral airways toward the
mouth. The recommended starting settings for this device are a frequency of
5 Hz building up to 10–15 Hz, and the jacket should be inflated to a pressure
that is comfortable but the oscillations should be felt by the patient within the
lungs and not just superficially. High levels of HFCWO can also be delivered via a
cuirass with NPV devices. These devices often have a dedicated secretion mode.
Once again, settings should be titrated to patient comfort; however, the oscil-
lations should be felt by the patient within the lungs and not just superficially.
Both of these devices can be used in conjunction with NIV. Treatments with
intrapulmonary percussive ventilation and HFCWO are usually carried out in
5-min stages or until the patient feels the need to cough. Treatment is then
ceased and, where required, cough augmentation techniques are performed
(see the section entitled “Practicalities of and guide to cough augmentation and
daytime mouthpiece ventilation”).
Oscillatory devices of this type have been shown to assist in the mobilisation
of secretions, decrease atelectasis on chest radiography, decrease the work of
breathing and decrease the sensation of breathlessness (ALS patients). However,
as there is the potential to mobilise large amounts of secretions, it is essential
to ensure that the patient is able to cough and clear these secretions effectively.
The newer mechanical insufflation/exsufflation devices also have the ability to
add vibrations onto the insufflation, exsufflation or both. At present, there is no
evidence to say whether this function mobilises more secretions. The interruption
of the expiratory flow with vibration may even decrease peak expiratory airflow
and may produce a poorer outcome, if the device is being used to enhance cough
augmentation.
Protocolised approach to the acutely unwell patient on NIV

Protocolised treatment sessions have been shown to facilitate extubation and pre-
vent intubation. They have also been shown to help prevent extubation failure in
patients who are at risk of extubation failure due to respiratory failure and secre-
tion encumbrance. In patients with no lung pathology, their SpO2 should be normal
on room air. When the SpO2 drops below 95%, the patient is experiencing atelec-
tasis as a result of secretion retention. It is essential to carry out airway clearance
techniques and cough augmentation techniques to reverse the a telectasis and
prevent sputum encumbrance. For diagrams of extubation and airway clearance
protocols, see figure 2.
Factors to assist airway clearance techniques

Humidification When secretions become tenacious, the addition of heated humidi-
fication within the NIV circuit can help. Care should be taken in patients who have
an overwhelming amount of oral secretions, as the additional water may exacer-
bate the excess oral secretions.
Hydration Good systemic hydration is required to improve viscoelasticity of the

sputum itself, while maintaining hydration of the airway surface liquid helps to
maintain efficient and effective movement of the cilia.
Inhaled therapies Airway obstruction can be reduced with the use of inhaled and
nebulised bronchodilators. Inhaled bronchodilators can also help to prevent or
reduce effort-induced bronchospasm, which can occur with airway clearance
techniques. Other inhaled therapies include mannitol, for improving sputum
clearance in CF and in bronchiectasis. In COPD, mucolytic agents such as carbo
cysteine may lead to a reduction in exacerbation frequency and total number of
days of disability. Inhaled hyperosmolar agents such as hypertonic saline (7%)
Extubation protocol Airway clearance protocol
Decrease O2 to room air on ventilator Decrease O2 to room air on ventilator

or pre-episode requirement or pre-episode requirement
Attach home ventilator to Optimise NIV settings if required

the endotracheal tube (increase IPAP if PaCO2 elevated,
(optimise settings if required) consider increasing EPAP if
atelectasis is present)
Perform airway clearance and cough

Perform airway clearance and cough
augmentation with mechanical
augmentation with mechanical
insufflation/exsufflation
insufflation/exsufflation when
SpO2<95%
Extubate onto home ventilator

(optimise settings further if required) Allow the patient to rest
Perform airway clearance and cough Repeat airway clearance and cough
augmentation when SpO2<95% augmentation when SpO2<95%
Wean by time off the ventilator Wean by time off the ventilator
rather than decreasing IPAP rather than decreasing IPAP
Figure 2. Protocolised approaches for extubation and airway clearance, in the management
of the neuromuscular patient.
are another option for altering sputum rheology and improving tracheobronchial
clearance. However, it is recommended to start with 3.5% hypertonic saline in
patients who are prone to bronchoconstriction or in whom the treatment may
produce an overwhelming amount of secretions. In patients with neuromuscular
disease it is essential to have an effective cough augmentation regime in place. In
patients who are colonised with bacteria such as Pseudomonas aeruginosa, inhaled
antibiotic therapy may also decrease exacerbation frequency. Ideal timings for
such inhaled medications are shown in figure 3.
If the patient is dependent on NIV and is unable to remove ventilator support, the
nebuliser can be given via the mouth with a mouthpiece while the patient is using
NIV via a nasal mask. An alternative option with a full face mask is a single-limb
NIV circuit where the nebuliser is placed as close to the patient as possible. If the
circuit includes a leak or expiratory valve, place the nebuliser between the leak or
valve and the full face mask. Deposition of the nebuliser may be improved with a
longer breath in and lower inspiratory pressures where possible, e.g. an IPAP of 10
or 15 cmH2O and an EPAP of 5 cmH2O for the duration of therapy. Care should
also be taken to replace any wet filters as this can affect ventilator performance.

Specified time prior to RhDNase

airway clearance (in patients with CF)
10 min pre-airway clearance Inhaled bronchodilators
Pre-airway clearance 7% or 3.5% hypertonic saline
Airway clearance
Post-airway clearance Inhaled antibiotics
Post-airway clearance Inhaled steroids
Figure 3. Timing of inhaled therapies around physiotherapy. The algorithm should be

individualised to the patient and obviously the patient will not necessarily take all of the
inhaled therapies. RhDNase: recombinant human deoxyribonuclease. Reproduced from
Rochester et al. (2012), with permission from the publisher.
Acknowledgements
This work was supported by the NIHR Respiratory Disease Biomedical Research
Unit at the Royal Brompton and Harefield NHS Foundation Trust and Imperial
College London.
Further reading
• Bott J, et al. (2009). Guidelines for the physiotherapy management of the adult,
medical, spontaneously breathing patient. Thorax; 64: Suppl. 1, i1–i51.
• Chatwin M (2008). How to use a mechanical insufflator–exsufflator “cough assist
machine”. Breathe; 4: 320–329.
• Hull J, et al. (2012). British Thoracic Society guideline for respiratory management
of children with neuromuscular weakness. Thorax; 67: Suppl. 1, i1–i40.
• Main E, et al., eds. (2015). Physiotherapy for Respiratory and Cardiac Problems:
Adults and Paediatrics. 5th Edn. London, Elsevier Limited.
• Rochester A, et al. (2012). Devices and techniques to aid physiotherapy in respira-
tory patients. ERS Buyers’ Guide; pp. 73–84.
• Schöni MH (1989). Autogenic drainage: a modern approach to physiotherapy in
cystic fibrosis. J R Soc Med; 82: Suppl. 16, 32–37.

Online resources
• Gonçalves M, et al. (2012). Cough assistance techniques in neuromuscular dis-
ease. ERS CME Online. www.ers-education.org/cmeonline/selection/cough-
assistance-techniques-in-neuromuscular-disease.aspx
• Hare A, et al. Acute and chronic neuromuscular disease modules. Skills-based
Simulator Training in Non-Invasive Ventilation. www.ers-education.org/e-learning/
simulators.aspx

Acute NIV monitoring
Monitoring choices in acute NIV
Raffaele Scala
NIV has been shown to be an effective therapy for patients with ARF of various
aetiologies. Avoiding endotracheal intubation (ETI) and the life-threatening com-
plications of conventional mechanical ventilation (CMV) (e.g. ventilator-associated
pneumonia) is the greatest advantage of using NIV for early ventilatory support.
This is also why NIV has been widely applied outside ICUs and high-dependency
units (HDUs), in settings with a lower capability for monitoring. Unfortunately, NIV
failure may occur in 5–60% of the treated cases, depending on numerous factors,
including the severity of ARF, the expertise of the team and the intensity of care
provided by the environment.
The rationale for monitoring during NIV is to assess whether this ventilatory sup-
port will be able to successfully avoid the need for ETI by:
• effectively unloading respiratory muscles
• correcting gas exchange abnormalities
• improving alveolar ventilation
Evaluation of the clinical and physiological effects of NIV in the first hours of treat-
ment is essential to detect patients who are likely to fail and should be quickly
Key points
• Monitoring is used to assess the effectiveness of acute NIV and
to identify patients who are likely to fail with this technique at
an early stage.
• Basic monitoring includes the evaluation of several clinical and
physiological parameters before and during NIV to assess the
effects of ventilation on gas exchange and the comfort of the
acute patient.
• Advanced monitoring takes into consideration issues specific
to acute NIV (e.g. leaks and patient–ventilator interaction) that
are likely to impact on the final outcome of NIV.
• The setting for monitoring should be chosen based on the
severity of the patient’s illness, the timing of NIV and the
likelihood of treatment failure.

Response Mask
Physiological Objective Subjective Fit
Comfort
Oximetry Respiratory rate Dyspnoea
Air leaks
Exhaled VT Blood pressure Comfort
Secretions
ABG Heart rate Mental alertness
Skin necrosis
Ventilatory monitoring
Respiratory pattern
Patient–ventilator interaction Respiratory muscle unloading
Sternocleidomastoid
Paradoxical abdominal motion
Availability of ETI
In case of NIV failure
Abdomen
Gastric distention
Activation with inspiration
Figure 1. Key points for monitoring of patients undergoing NIV for ARF. ABG: arterial blood
gas. Reproduced and modified from Umberto Meduri (1996) Clin Chest Med; 17: 513–553,
intubated and invasively ventilated in a high-intensity setting (ICU). In fact, the

majority of failures occur within the first 2 h treatment, although ∼20–25% of
patients may fail later after an initial response to NIV. Early identification of NIV
failure is of pivotal importance as unduly delaying CMV may be associated with
increased mortality. Furthermore, monitoring should also assess some intrinsic
issues associated with this ventilator technique (e.g. amount of leak, skin damage
and patient–ventilator interaction) that are likely to influence the overall success
of NIV in clinical practice (fig. 1 and table 1).
Clinical monitoring
Bedside clinical evaluation is essential particularly in the first hours of treatment
and includes monitoring of patient comfort, respiratory distress, cough efficiency
and neurological status. Poor tolerance, severe encephalopathy, marked dyspnoea
and inefficient removal of secretions are strong predictors of NIV failure.
The perception of the intensity of breathlessness, the use of accessory respiratory
muscles, paradoxical abdominal motion and active expiratory efforts are indirect
indexes of excessive work of breathing, and should be carefully assessed and moni-
tored. Of the various scales that quantify comfort, dyspnoea and respiratory distress,
the visual analogue scale has been validated in mechanically ventilated patients.
Cough efficiency may be measured by means of peak cough expiratory flow; values
<270 L⋅min−1 are considered at risk of retention of secretions and indicate the need
for manual and/or mechanical cough assistance, especially in neuromuscular
Table 1. Essential minimum level of monitoring for NIV during ARF
Monitoring Tool
Clinical
Dyspnoea Visual analogue scale
Compliance with NIV Visual analogue scale
Respiratory distress Use of accessory muscles, abdominal paradox
Cough efficiency Clinical, peak cough expiratory flow
Hypercapnic encephalopathy Kelly–Matthay score
Psychomotor agitation Richmond sedation–agitation scale
Delirium Intensive care delirium screening checklist
Physiological
Respiratory rate Clinical, plethysmography
ABG Pulse-oximetry, ABG analysis
Cardiovascular ECG, blood pressure, heart rate
Blood volume Fluid balance
NIV-related
Leaks Clinical, ventilator estimation
Interface comfort Clinical, visual analogue scale
Skin lesions Clinical
Breathing pattern Clinical, ventilator numerical data and curves
Patient–ventilator interaction Clinical, ventilator curves
patients. For further details, refer to the section on “Practicalities of and guide to
cough augmentation and daytime mouthpiece ventilation”.
Different alterations in neurological status may be found in acute NIV patients.
The Glasgow Coma Scale (GCS) is largely applied to assess sensorium level in ARF.
However, the GCS, which was initially calibrated for trauma-correlated neuro
logical dysfunctions, has not been validated for hypercapnic encephalopathy. The
Kelly–Matthay score is the best tool for measuring neurological alterations sec-
ondary to gas exchange abnormalities (table 2). The Richmond sedation–agitation
scale is the more suitable instrument for evaluating and monitoring the degree
of psychomotor agitation and the effects of sedation during NIV in poorly tolerant
patients (table 3). Delirium is a frequent complication in mechanically ventilated
patients and is correlated with NIV failure. The intensive care delirium screening
checklist is a reliable instrument for assessing delirium in NIV patients.
Physiological monitoring
Whenever NIV is applied for ARF, assessment of gas exchange, both at baseline
and after the first few hours of treatment, is mandatory to understand whether
NIV may be successfully continued or if CMV is required. Noninvasive continuous
monitoring of SpO2 should be used in all patients during NIV. Important limita-
tions of this technique need to be considered. From a technical point of view, the
Table 2. Kelly–Matthay score for assessment of hypercapnic encephalopathy during NIV
Grade 1 Alert, follows complex three-step command

Grade 2 Alert, follows simple commands
Grade 3 Lethargic, but arousable and follows simple commands
Grade 4 Stuporous (only intermittently follows simple commands even
with vigorous attempts to arouse the patient)
Grade 5 Comatose, brain stem intact
Grade 6 Comatose with brain stem dysfunction
Reproduced from Kelly et al. (1993) Chest; 104: 1818–1824, with permission from the publisher.
presence of poor peripheral perfusion, movement artefacts, haemoglobinopathies,

nail polish and hyperbilirubinaemia may affect the reliability of continuous
SpO2 monitoring. From the physiological point of view, the most critical issues are:
1) the variable correlation between SpO2 and PaO2 due to the shape and position
of the dissociation SpO2 curve
2) the lack of information about PaCO2
Table 3. Richmond sedation–agitation scale for evaluating psychomotor agitation and

effects of sedation during NIV
Score Term Description

+4 Combative Overtly combative or violent; poses an immediate
danger to staff
+3 Very agitated Pulls on or removes tube(s)/catheter(s), or behaves
aggressively towards staff
+2 Agitated Frequent non-purposeful movement or patient–
ventilator asynchrony
+1 Restless Anxious or apprehensive, but movements are not
aggressive or vigorous
0 Alert and calm
−1 Drowsy Not fully alert, but has sustained (>10 s) awakening,
with eye contact, in response to voice
−2 Light sedation Briefly (<10 s) awakens, with eye contact, in
response to voice
−3 Moderate sedation Any movement (but no eye contact) in response to
voice
−4 Deep sedation No response to voice, but any movement in response
to physical stimulation
−5 Unarousable No response to voice or physical stimulation
Reproduced from Sessler et al. (2002) Am J Respir Crit Care Med; 166: 1338–1344, with permission
from the publisher.

Accordingly, if patients have COPD or risk factors for hypercapnic ARF, a lower
target SpO2 (between 88% and 92%) is recommended. By contrast, for “pure
hypoxaemic” patients, the SpO2 target may be adjusted to 94–98%. The two
available techniques for noninvasive monitoring of PaCO2 changes (capnography
and transcutaneous measurement) have important drawbacks during NIV and,
therefore, have limited applications in clinical practice. However, newer trans-
cutaneous devices with a combined sensor for the measurement of both PtcCO2
and pulse oximetry have been shown to be useful for a noninvasive monitoring of
blood gases in ARF patients under NIV. The agreement between PtcCO2 and PaCO2
turns out to be accurate, especially in subjects with haemodynamic stability and
a moderate degree of hypercapnia (i.e. <70 mmHg (<9.33 kPa)). Moreover, the
opportunity to obtain the trend in PtcCO2 and pulse oximetry under NIV may be
helpful for the clinician in adjusting the ventilator settings according to suspected
patient–ventilator asynchronies and/or excessive unintentional leaks and/or
sleep respiratory disorders.
ABG analysis remains the gold standard to assess the severity of ARF and the
effects of NIV. Blood gases should be sampled at:
• baseline
• after 1 and 4 h of NIV
• then at regular intervals depending on the patient’s response
• with every change to the ventilator settings
• finally, on spontaneous breathing to choose the timing for weaning from NIV
Respiratory rate, both at baseline and after 1 and 4 h of NIV, is a strong predic-
tor of NIV outcome. The available methods for monitoring respiratory rate have
important limitations and the bedside clinical assessment remains the most
used tool.
Basic noninvasive cardiovascular monitoring (ECG, pulse rate and noninvasive
blood pressure) is recommended in patients undergoing NIV as the occurrence of
cardiovascular complications (i.e. severe arrhythmias, haemodynamic instability,
acute coronary syndromes and cardiovascular arrest) is correlated with NIV failure
and mortality.
A full panel of biochemical tests should be performed in all patients during NIV
as the development of extrapulmonary complications affects the outcome.
Hyperglycaemia is an independent predictor of NIV failure in COPD exacerbations;
therefore, frequent assessment of glucose level is recommended. Fluid balance
should be assessed daily in all patients as complications due to renal failure and
fluid overload (i.e. congestive heart failure) are likely to be correlated with NIV
failure.
NIV-related monitoring
Monitoring of interface- and ventilator-related issues is crucial to evaluate the
effectiveness (i.e. patient–ventilator interaction) and the complications of NIV (i.e.
tolerance, leaks and skin damage).
The majority of NIV complications are due to problems correlated with the inter-
face. The main determinants for the choice of the interface are: few leaks, good

comfort and less likelihood of skin damage/breakdown. During NIV delivered by

ventilators equipped with a single-limb circuit, we have to consider:
• intentional leaks (due to the presence of the exhalation system)

• unintentional leaks (between the mask and the face, and through the mouth
for patients using nasal ventilation)
Excessive unintentional leaks are strongly correlated with NIV failure as a con-
sequence of alveolar hypoventilation, discomfort, patient–ventilator asynchronies
and sleep fragmentation. A good balance between the amount of leak and level of
ventilatory support has to be achieved. This objective may be reached with close
monitoring of mask position, leaks and the patient’s comfort. A strategy based
on the rotation of different interfaces increases comfort and reduces NIV fail-
ure. Leaks may be grossly estimated at the bedside by placing the hands near
the mouth and around the interface profile to feel the amount of air dispersion
between the patient’s face and the mask (if any). However, this clinical evaluation
has low sensitivity and does not allow quantification of leaks. In some ventilators,
leaks are automatically estimated with special algorithms and displayed on the
screen of the ventilator.
It is important to choose a ventilator with reliable monitoring equipment.
Ventilator monitoring includes the assessment of numerical data with or without
a graphical curve display. The presence of a “rapid shallow-breathing pattern”
with a respiratory rate/expiratory VT ratio ≥105 is a predictor of NIV failure. The
accuracy in the monitoring of respiratory rate and expiratory V T during NIV is
dependent both on the type of ventilator and on the interaction between patient
and ventilator.
Expiratory VT assessment is feasible only with ventilators equipped with a
dual-limb circuit, where expiratory V T is obtained by subtracting leaks from
inspiratory V T.
In ventilators with a single-limb circuit there are two possibilities depending on
the type of exhalation system:
1) anti-rebreathing expiratory valve or “non-vented circuit”

2) intentional leak or “vented circuit”
In a non-vented circuit, the ventilator provides an inspiratory V T value that is

always an actual measurement of the volume delivered by the ventilator. The
values are computed at the beginning of inspiration, so that in the presence
of leaks, the leaks are considered as part of the delivered inspiratory V T. In this
case, the ventilator is unable to measure and provide an estimate of leaks and
expiratory V T.
In a vented circuit, the ventilator provides an estimation (and not a measure) of
expiratory VT that should be the real volume inspired by the patient without the
intentional leaks. In this case, the ventilator is able to provide an estimation of
leaks. The leak value displayed may be the total leak (intentional + unintentional)
or only the unintentional leaks depending on the algorithm used by the ventilator.
Unfortunately, in cases with a very large leak the estimation of leak, as well as the
estimation of expiratory VT, may become unreliable (fig. 2).
a) Unintentional
leaks
From the patient Expiratory VT
Ventilator
delivered volume
To the patient Inspiratory VT

Unintentional
leaks
Inspiratory VT = Ventilator delivered volume + Unintentional leaks
Expiratory VT = Inspiratory VT - Unintentional leaks
b) Unintentional
leaks
Non-rebreathing valve
Ventilator Expiratory VT
delivered volume
Inspiratory VT Unintentional
leaks
Inspiratory VT = Ventilator delivered volume + Unintentional leaks
Expiratory VT = ???#
c) Unintentional
leaks
Intentional leaks
Estimated
Ventilator expiratory VT
delivered volume
Inspiratory VT Unintentional
leaks
Inspiratory VT = Ventilator delivered volume + Intentional leaks + Unintentional leaks
Expiratory VT = Inspiratory VT - Intentional leaks
Figure 2. Differences in V T monitoring during NIV delivered by ventilators equipped with
a) a dual-limb circuit or a single-limb circuit with b) a “non-vented” or c) a “vented”
exhalation system. #: measurement of expiratory V T is unreliable with the large amount of
unintentional leaks in this type of ventilator.

When assessing respiratory rate during NIV, there may be a gap between the rate
of ventilator-assisted and patient-triggered breaths. In terms of the ventilator
respiratory rate, ineffective efforts and auto-triggering may cause underesti-
mation or overestimation of the patient’s effective breathing rate, respectively.
Even if patient–ventilator asynchronies may be suspected from a careful observation
of chest and abdomen movements during NIV, the interpretation of ventilator curves
is helpful to noninvasively assess patient–ventilator interaction (fig. 3). Correct
identification of the type of asynchrony is helpful in choosing the best strategy
to improve the degree of patient–ventilator interaction (e.g. choosing a different
interface to reduce leaks and/or changing the ventilator settings).
Invasive monitoring of trans-diaphragmatic pressure with an oesophageal
balloon is the best way to assess patient–ventilator interaction; however, this
method is not used in clinical practice because it requires advanced skills and it
is invasive.
In patients with severe sleep disordered breathing (i.e. OHS, overlap syndrome or
OSAS) the availability of sleep study techniques (i.e. cardiorespiratory monitoring
and polysomnography) may be useful for understanding and treating the causes of
night-time oxygen desaturations and/or patient–ventilator asynchronies.
Choices of location for monitoring
The patient’s needs for monitoring are the most important factors to consider
in when choosing the setting in which to start NIV. The greater is the severity of
ARF, and therefore the likelihood of NIV failure, the higher the monitoring capa-
bility of the setting needs to be (e.g. HDU or ICU). Treatment of patients with a
lower chance of successful with NIV (i.e. patients with severe pneumonia, ARDS
or asthma) requires:
• a high nurse-to-patient ratio together with expertise on NIV
• prompt availability of ETI and experience in the management of invasively
ventilated patients
Flow
Pressure
Figure 3. A typical patient–ventilator asynchrony pattern due to ineffective efforts during
NIV that may be easily identified by looking at the flow–pressure curves of the ventilator.
Arrows indicate the wasted efforts performed by the patient to successfully trigger the
ventilator. Reproduced from Vignaux et al. (2009) Intensive Care Med; 35: 840–846, with
Ward-based NIV without continuous monitoring should be used only for patients
with a greater likelihood of success, such as COPD exacerbations with mild degree
of respiratory acidosis without extrapulmonary complications.
Further reading
• Hill NS (2009). Where should noninvasive ventilation be delivered? Respir Care; 54:
62–70.
• Janssens JP, et al. (2011). Nocturnal monitoring of home non-invasive ventilation:

the contribution of simple tools such as pulse oximetry, capnography, built-in
ventilator software and autonomic markers of sleep fragmentation. Thorax; 66:
438–445.
• Kondili E, et al. (2010). Monitoring during acute ventilation. In: Elliott MW, et al.,
eds. Non-Invasive Ventilation and Weaning: Principles and practice. London,
Hodder Arnold; pp. 97–106.
• Nava S, et al. (2006). Time of non-invasive ventilation. Intensive Care Med; 32:
361–370.
• Nava S, et al. (2009). Interfaces and humidification for noninvasive mechanical

ventilation. Respir Care; 54: 71–84.
• Nava S, et al. (2009). Non-invasive ventilation in acute respiratory failure. Lancet;

374: 250–259.
• Nilsestuen JO, et al. (2005). Using ventilator graphics to identify patient-ventilator

asynchrony. Respir Care; 50: 202–234.
• O’Driscoll BR, et al. (2008). BTS guideline for emergency oxygen use in adult
patients. Thorax; 63: Suppl. 6, vi1–vi68.
• Ozyilmaz E, et al. (2014). Timing of noninvasive ventilation failure: causes, risk fac-
tors, and potential remedies. BMC Pulm Med; 14: 19.
• Scala R, et al. (2008). Ventilators for noninvasive ventilation to treat acute respira-
tory failure. Respir Care; 53: 1054–1080.
• Scala R, et al. (2010). How to start a patient on non-invasive ventilation. In: Elliott
MW, et al., eds. Non-Invasive Ventilation and Weaning: Principles and practice.
London, Hodder Arnold; pp. 70–83.
• Scala R (2011). Hypercapnic encephalopathy syndrome: a new frontier for non-

invasive ventilation? Respir Med; 105: 1109–1117.
• Tobin MJ (2001). Advances in mechanical ventilation. N Engl J Med; 344: 1986–

1996.
Online resources
• Scala R (2014). Monitoring of NIV in the acute setting. ERS Course Noninvasive
Ventilation: Basic Concepts, Hanover 2014. www.ers-education.org/events/

Starting and stopping acute NIV:

when and why?
Bernd Schönhofer
For patients with ARF, the treatment of choice was invasive mechanical ventilation
for many decades. NIV mainly via different kinds of masks has now been proven to
be an alternative and is increasingly applied. As a consequence, a range of national
and international guidelines on NIV in ARF have been published in the past few
years (Evans, 2001 and Schönhofer et al., 2008). Successful patient selection and
implementation depends on understanding the pathophysiology and mechanism
of action of NIV.
Pathophysiology and the rationale of how NIV works

The respiratory system consists of two different compartments: the lung, which is
responsible for gas exchange, and the respiratory pump, which stands for ventila-
tion. Based on this, mechanical ventilation has two aims (fig. 1).
Key points
• NIV is the first choice treatment, based a high level of evidence-
based medicine, in acute hypercapnic failure caused by acute
exacerbations of COPD with mild-to-moderate acidosis.
• In diseases with hypoxaemic respiratory failure based on a
complex pathophysiology, the rate of NIV failure is much
higher than in acute hypercapnic failure. NIV in hypoxaemic
respiratory failure should only be started if the team is
experienced in NIV and continuous monitoring of vital signs
is guaranteed.
• Arterial blood gases, breathing frequency, haemodynamic
status, level of dyspnoea and level of consciousness are variables
to monitor which indicate success or failure of NIV in ARF.
• It is important to assess clinical symptoms and parameters
1–2 h after initiation of NIV to identify the response.
• Irreversible NIV failure must be recognised without delay to
ensure invasive mechanical ventilation starts early enough.

Lung Respiratory pump
Hypoxaemic ARF Hypercapnic ARF
O2 CO2 O2 CO2
Oxygen Mechanical ventilation
Figure 1. Pathophysiology and principles of treatment.
1) The most important reason for acute hypercapnic failure is the imbalance
between respiratory muscle capacity and load. The failure of the respiratory
muscle pump leads to hypercapnia. In acute hypercapnic failure, NIV improves
ventilation and reduces the work of breathing.
2) Hypoxaemic respiratory failure is caused by an insufficiency of the lung paren-
chyma (e.g. in ARDS or pneumonia). Here, NIV increases transpulmonary
pressure, i.e. enlargement of the end-expiratory lung volume, and additional
PEEP recruits alveolar space.
Practical issues
Successful implementation requires practical essentials, which are dealt with in
more detail in other sections of this book.
Interface In acute hypercapnic failure NIV mainly is applied via different kinds of
masks (e.g. a full face mask) rather than the helmet, which is an alternative inter-
face in hypoxaemic respiratory failure.
Ventilator choice A broad spectrum of respirators are applied in NIV to treat ARF,
ranging from portable respirators, also used in home mechanical ventilation, to
high-tech ICU ventilators.
Medical team In the initial phase of NIV in the treatment of hypercapnic respiratory
failure there is an increased need for personnel, with a nurse-to-patient ratio of
up to 1:1. As the patient improves the need for personnel decreases, in contrast to
the situation for patients treated using invasive mechanical ventilation.
Training of medical team Comprehensive training programmes and practical experi
ence with NIV in the acute setting remain the most important determinants of
high quality treatment delivery and the successful implementation of NIV in
increasingly unwell groups of patients.
Location The preferred location to treat acute hypercapnic failure remains the ICU.
Depending on the clinical state of the patient, e.g. single organ respiratory insuffi-
ciency in a stable patient, NIV may also be performed in high-dependency respira-
tory units or even in normal respiratory wards.
Different indications for NIV in acute hypercapnic failure
NIV is used in a variety of diseases resulting in acute hypercapnic failure. The most
important disease is COPD.
Acute hypercapnic failure caused by COPD NIV is the treatment of choice in acute

hypercapnic respiratory failure due to acute exacerbation of COPD (for further
details see the section entitled “The patient with an acute hypercapnic exacer-
bation of COPD”), and this conclusion is supported by a high level of evidence.
ICU trials have demonstrated a reduction in the need for endotracheal intubation
and mechanical ventilation; indeed the largest study, to date, showed increased
survival, reduced complication rates and a decrease in the length of both ICU and
hospital stay (Brochard et al., 1995).
Where NIV can be successfully applied there are clear advantages, particularly in
terms of a reduction in infectious complications (e.g. nosocomial pneumonia) and
the length of ICU and hospital stay, with an attendant reduction in costs. NIV can
be introduced earlier in the natural history of the condition before mechanical
ventilation would normally be considered necessary.
With respect to prospective RCTs of NIV outside the ICU, either on general wards
or in the accident and emergency department, a multicentre RCT of NIV in acute
exacerbations of COPD undertaken on general respiratory wards demonstrated
that NIV reduced “treatment failure” (a proxy end-point for the need for intub
ation) from 27% to 15%. In-hospital mortality was also decreased from 20% to
10% (Plant et al., 2000). NIV was applied, by the usual ward team (who had
received basic training in NIV), using a bilevel device in spontaneous mode and
a simple stepwise protocol. This study showed that, with adequate staff training,
NIV can be applied successfully outside the ICU and that early initiation of NIV
on a general respiratory ward (pH <7.35 on admission to the ward) results in
a better outcome compared with providing no ventilatory support for acidotic
patients outside the ICU. More severely affected patients (pH <7.30 after initial
management) had a worse outcome compared with those in the ICU studies,
indicating that this simple, general ward-based approach is not appropriate in
this more severely unwell group, who are best managed in a high-dependency
setting with a more sophisticated ventilator individually adjusted to their
requirements and the availability of a higher nurse-to-patient ratio. The studies
described here indicate that early initiation of NIV is best (Peter et al., 2002;
Lightowler et al., 2003).
Non-COPD patients with hypercapnic failure Patients with both chest wall deform-
ity and neuromuscular disease may develop acute hypercapnic failure caused by
an overloaded or weak diaphragm. There is little published about the use of NIV
acutely in patients with chest wall deformity and neuromuscular disease. Broadly
speaking the principles established in patients with COPD can be extended to this
patient group, but with some important differences. There should be a very low
threshold for starting NIV and it is likely that it will be needed long-term during
sleep. Details of secretion clearance are discussed further in the section entitled
“Airway clearance methods and nebulised therapy in acute NIV”.
Obesity has reached epidemic proportions and it is projected that this will worsen.
The main findings of a recently published study by Lemyze et al. (2014) were that
NIV rarely failed to reverse ARF in morbidly obese patients, and that patients who
exhibited early NIV failure had a high severity score and a low bicarbonate level
at admission, and were likely to have hypoxaemic ARF caused by pneumonia.
Application of NIV to acute hypercapnic respiratory failure in OHS is discussed
further in the section entitled “Patients with acute hypercapnic respiratory failure
and OHS”.
Hypoxaemic respiratory failure

NIV has an established ability to improve gas exchange and reduce the work of
breathing in patients with hypoxaemic ARF. Compared with acute hypercapnic
respiratory failure the clinical efficacy and overall outcome of NIV in hypoxaemic
respiratory failure is less clear and depends very much on patient selection and the
assessment of the risks of the technique.
The potential risks include an insufficient reduction in the oxygen consumption of
the respiratory muscles, an insufficient and discontinuous opening of the alveo
lar space in cases of lung injury, and the risk of delayed intubation, which may
worsen outcome.
In a team with long experience with NIV in the acute setting, NIV significantly
reduced the risk for intubation, rate of septic shock and 90-day mortality rate
compared with standard treatment (Confalonieri et al., 1999). See also the sec-
tions entitled “The patient with acute hypoxaemic respiratory failure excluding
pulmonary oedema” and “The patient with acute hypoxaemic failure and cardio-
genic pulmonary oedema”.
Diseases which cause hypoxaemic respiratory failure are characterised by a
complex pathophysiology. The NIV failure rate in hypoxaemic respiratory failure
is 30–50% and even higher in patients with ARDS (Antonelli et al., 2001). That
means that NIV in ARDS patients should only be performed in specialised centres
with continuous monitoring of vital signs and the ability to rapidly escalate to inva-
sive ventilation if required.
When should NIV be started?

In general, patients who are likely to benefit from NIV can be selected by taking the
following considerations into account, in a stepwise fashion.
1) The indication for mechanical ventilation in general must be determined.

Important clinical signs are: respiratory distress; rapid shallow beathing;
tachypnoea; and accessory muscle use.
2) The use of NIV should be evidence based (e.g. acute hypercapnic exacerbation
of COPD or acute cardiogenic pulmonary oedema).
3) The patient should have no contraindications for NIV such as: the need for an
artificial airway for airway protection (e.g. major aspiration); inability to fit an
interface; a high severity of illness (e.g. respiratory arrest); an uncooperative
patient; or multisystem failure.
pH >7.35 7.20<pH<7.35 or pH <7.20

7.10<pH<7.20 + isolated + comorbidity
hypercapnic failure +
cooperative patient
Yes
Contraindications for NIV Intubation
No
No indication
for NIV NIV
Figure 2. pH-based algorithm for acute hypercapnic respiratory failure in COPD.
4) Patients should not suffer from diagnoses in which it has been shown that NIV
is not effective (e.g. severe, rapidly progressive ARDS).
5) The patient’s wishes and advance directives should also be considered; some
patients decide not to receive NIV.
6) In all cases decide in advance what you will do if NIV fails: is the patient suit-
able for intubation and invasive ventilation, or is NIV the ceiling of care?
Hypercapnic ARF caused by an acute exacerbation of COPD is the most common
indication for deployment of NIV. In figure 2 an algorithm for application of NIV is
shown, based on pH values and clinical signs.
When should NIV be stopped?
In essence NIV should be stopped if it is ineffective (i.e. it has failed), it is distress-
ing the patient, or if the patient no longer requires it.
In cases of NIV failure, clinical signs that are only equivocal on presentation
become more definitively predictive of failure if they persist after 2 h of NIV. Thus,
it is important to assess clinical symptoms and parameters 1–2 h after initiation
of NIV to identify the response. Arterial blood gas results, breathing frequency,
haemodynamic status, level of dyspnoea and vigilance are easy-to-monitor vari-
ables that indicate success or failure of NIV in ARF (table 1).
It is important to assess the risk of NIV failure. Risk factors for NIV failure are
shown in table 2. Subjects who are likely to fail NIV have a greater degree of severe
respiratory acidosis, a lower level of consciousness, are older, more hypoxaemic,
and have a higher breathing frequency on presentation (Confalonieri et al., 2005).
Table 1. Indications of early NIV failure in ARF (after 2 h NIV)
↑ PaCO2 ↑ Breathing frequency

↓ pH ↓ VT
↓ Oxygenation ↑ Heart rate
↑ Dyspnoea ↓ Cognitive function

Table 2. Risk factors for NIV failure
Tachypnoea: 35 breaths⋅min−1
Metabolic acidosis: pH <7.25
Excessive air leak
Agitation
Excessive secretions
Poor adherence to therapy
Acute hypoxaemic respiratory failure
Diagnosis of ARDS or pneumonia
Age >40 years
Hypotension: systolic blood pressure <90 mmHg (<11.97 kPa)
Low PaO2/FIO2
Simplified Acute Physiology Score II <34
No initial improvement of pH and PaCO2 within first 2 h of NIV
No improvement of oxygenation within the first hour of NIV
Persistent tachypnoea
Persistent hypercapnia
Berg et al. (2012) evaluated the ability of the rapid shallow breathing index (RSBI),
the ratio of breathing frequency (breaths⋅min−1) to VT, to predict NIV failure.
In the initial phase, unchanged parameters, e.g. acidosis and hypercapnia in
hypercapnic respiratory failure, may be acceptable if the clinical condition stabi-
lises or even improves during NIV.
Different causes of NIV failure
Recognition that NIV is failing is an important, but often overlooked, part of the
management of NIV. The reported NIV failure rate is 5–40% (Nava et al., 2004).
Some patients fail due to progression of the disease process, others because they
cannot tolerate NIV or it is suboptimally applied. In general, early failure relates
to equipment intolerance, and later failure is attributable to disease progression
despite NIV. Greater clinician experience and expertise with the application of NIV
are associated with a higher success rate.
It is crucial to understand the circumstances when NIV fails. NIV failure is likely if
there are associated complications or if the patient’s premorbid condition is poor. In
general, irreversible NIV failure must be recognised early. If the NIV failure is missed
in a patient with fair life expectancy, the delayed intubation, i.e. invasive mechan
ical ventilation, may worsen the prognosis. In figure 3 some clinically important
causes of NIV failure and the effects of invasive mechanical ventilation are shown.
Patients who fail because of problems with the application of or tolerance for the
interface will tend to fail early, and the substitution of a different interface will
improve the delivery of effective ventilation.
Massive Massive load Neurophysiological

hypersecretion on respiratory muscles weakness
Invasive mechanical ventilation
Lavage with Unloading Improvement of

hypersecretion respiratory muscles neurophysiological
situation
Aim: extubation after 2–3 days,

with/without consecutive NIV
Figure 3. Clinically important causes of NIV failure and the effects of invasive mechanical
ventilation separated according to the underlying clinical issues.
Other patients with hypercapnic failure will fail because they are difficult to ven-
tilate based on the underlying pathophysiology. Bronchial secretions and ineffec-
tive cough will reduce the efficiency of NIV in some patient groups, e.g. those with
neuromuscular disease. The combination of NIV and cough assistance techniques
is discussed in the section entitled “Airway clearance methods and nebulised ther-
apy in acute NIV”.
Patients may fail because of poor application of NIV and inappropriate ventilator
settings. The adequate response then is to recognise the problems and correct
them, rather than intubate the patient. For further details see the sections entitled
“Basic principles of ventilators” and “Matching mode and settings to the patient:
an introduction”.
When patients fail early there is still time for medical therapy to work and a reason
to buy some time with ventilatory support. When NIV failure occurs later, e.g. after
24–48 h, this will be despite medical therapy and there is less that is potentially
reversible. Late failure after initially successful NIV is a bad prognostic factor, with
over half the patients dying even with invasive ventilation (Moretti et al., 2000).
This is more likely to occur in patients with severe acidosis, poor functional status
and complications. In these patients, time should be usefully spent preparing the
patient and their family and moving to a more palliative approach, rather than
escalating therapy.
It should be kept in mind that while NIV is usually applied to correct ARF, if it
is used to palliate breathlessness in end-stage disease (e.g. in patients with
terminal malignancy and respiratory failure), the outcome to be assessed is
breathlessness and symptom relief, not arterial blood gas tensions. Therefore,
if dyspnoea is not relieved within a short period (minutes/hours) and fine tun-
ing of settings does not help, NIV should be withdrawn and attention directed
to other palliative measures such as opiates. See the section entitled “NIV in
palliative care and at the end of life”.
Finally, discontinuing NIV as the patient has recovered sufficiently to breathe spon-
taneously without ventilator support is discussed in the section entitled “NIV and
weaning”. NIV should first be stepped down during the day when the patient is awake,
and subsequently at night during sleep, according to arterial blood gas monitoring
(e.g. oximetry and transcutaneous carbon dioxide assessment) and symptom relief.
Crucial issues for implementation of NIV in ARF
Despite the high level of evidence underlining the impact of NIV for treatment of
acute hypercapnic failure, there are still crucial deficits both in terms of assess-
ing implementation and outcomes of this intervention in general clinical practice.
A recent audit from the UK raises significant concerns about the practice of
NIV in the “real” world (Roberts et al., 2011). Nearly one third of the patients
with the greatest evidence-base for NIV effectiveness did not receive NIV. Of
great concern, for a comparable pH, patients receiving NIV had a worse out-
come than those treated with conventional therapy. These data suggest poor
understanding of “when to start NIV”, highlighting the importance of training
and education.
Further reading
• Antonelli M, et al. (2001). Predictors of failure of noninvasive positive pressure
ventilation in patients with acute hypoxemic respiratory failure: a multi-center
study. Intensive Care Med; 27: 1718–1728.
• Berg KM, et al. (2012). The rapid shallow breathing index as a predictor of failure
of noninvasive ventilation for patients with acute respiratory failure. Respir Care;
57: 1548–1554.
• Brochard L, et al. (1995). Noninvasive ventilation for acute exacerbations of
chronic obstructive pulmonary disease. N Engl J Med; 333: 817–822.
• Confalonieri M, et al. (1999). Acute respiratory failure in patients with severe com-
munity-acquired pneumonia. A prospective randomized evaluation of noninvasive
ventilation. Am J Respir Crit Care Med; 160: 1585–1591.
• Confalonieri M, et al. (2005). A chart of failure risk for noninvasive ventilation in
patients with COPD exacerbation. Eur Respir J; 25: 348–355.
• Corrado A, et al. (2002). Respiratory intermediate care units: a European survey.
Eur Respir J; 20: 1343–1350.
• Demoule A, et al. (2006). Increased use of noninvasive ventilation in French inten-
sive care units. Intensive Care Med; 32: 1747–1755.
• Esteban A, et al. (2008). Evolution of mechanical ventilation in response to clinical
research. Am J Respir Crit Care Med; 177: 170–177.
• Evans TW (2001). International Consensus Conferences in Intensive Care Medicine:
non-invasive positive pressure ventilation in acute respiratory failure. Organised

jointly by the American Thoracic Society, the European Respiratory Society, the
European Society of Intensive Care Medicine, and the Société de Réanimation de
Langue Française, and approved by the ATS Board of Directors, December 2000.
Intensive Care Med; 27: 166–178.
• Hill NS (2009). Where should noninvasive ventilation be delivered? Respir Care;
54: 62–70.
• Lemyze M, et al. (2014). Determinants of noninvasive ventilation success or failure
in morbidly obese patients in acute respiratory failure. PLoS One; 9: e97563.
• Lightowler JV, et al. (2003). Non-invasive positive pressure ventilation to treat
respiratory failure resulting from exacerbations of chronic obstructive pulmonary
disease: Cochrane systematic review and meta-analysis. BMJ; 326: 185.
• Moretti M, et al. (2000). Incidence and causes of non-invasive mechanical ventila-
tion failure after initial success. Thorax; 55: 819–825.
• Nava S, et al. (1997). Human and financial costs of noninvasive mechanical ven-
tilation in patients affected by COPD and acute respiratory failure. Chest; 111:
1631–1638.
• Nava S, et al. (2004). Causes of failure of noninvasive mechanical ventilation.
Respir Care; 49: 295–303.
• Peter JV, et al. (2002). Noninvasive ventilation in acute respiratory failure – a meta-
analysis update. Crit Care Med; 30: 555–562.
• Plant PK, et al. (2000). Early use of non-invasive ventilation for acute exacerba-
tions of chronic obstructive pulmonary disease on general respiratory wards: a
multicentre randomised controlled trial. Lancet; 355: 1931–1935.
• Roberts CM, et al. (2011). Acidosis, non-invasive ventilation and mortality in hos-
pitalised COPD exacerbations. Thorax; 66: 43–48.
• Schönhofer B, et al. (2008). Nichtinvasive Beatmung als Therapie der akuten re
spiratorischen Insuffizienz [Non-invasive mechanical ventilation as treatment of
acute respiratory failure]. Pneumologie; 62: 449–479.

Problem-solving: case studies of NIV

problems and their management
Alanna Hare
Case study 1: persistent hypercapnia/hypoxaemia

A 65-year-old female with severe COPD attends the emergency department with
a 2-day history of increasing breathlessness and productive cough. She has been
using her inhalers at home without improvement. In the emergency department,
she is given nebulised bronchodilators, corticosteroids, antibiotics and controlled
oxygen therapy. After 1 h of therapy, her arterial blood gases on 24% inspired oxy-
gen are as follows: pH 7.31; PaCO2 9.2 kPa (69 mmHg); PaO2 7.2 kPa (54 mmHg);
HCO3− 31 mmol⋅L−1.
The doctors in the emergency department commence NIV and, 1 h later, ask you
to attend to review her. Her current settings are: IPAP 14 cmH2O; EPAP 4 cmH2O;
respiratory rate 12 breaths⋅min−1; with 1 L⋅min−1 oxygen entrained into the mask.
Practice point 1 When asked to review a patient who is using NIV, you
should examine the patient at the bedside. Bedside assessment should
include observation of the patient’s chest wall movement, auscultation of
the chest, review of mask fit and the presence of leaks, an assessment of
ventilator–patient synchrony, and review of the observation charts (includ-
ing oxygen saturation, pulse, blood pressure and temperature) and docu-
mentation of NIV usage. If the patient is able to talk, you may also ask the
patient about whether they are comfortable on the ventilator and, if not,
what their difficulties are.
Key points
• All patients who are commenced on NIV, particularly in the
acute setting, require regular monitoring and assessment.
• Most problems can be successfully managed using a simple
diagnostic approach involving careful patient examination and
observation at the bedside.
• Nevertheless, early recognition of NIV failure is vital and plans
should be set in place at the outset regarding escalation/
de-escalation of therapy in the event that the patient does
not improve or becomes more unwell while using NIV.

Your assessment reveals that the patient and ventilator are well-synchronised,
there is minimal leak from the well-fitted full face mask, and there is mild persis-
tent wheeze on auscultation of the chest. Oxygen saturations on NIV with settings
as above are 82%.
Practice point 2 Ensure all patients using NIV are also optimally medically
managed for their underlying condition. In the case of patients with an exac-
erbation of COPD, this should include nebulised bronchodilators, corticoster-
oids and antibiotics in the presence of signs/symptoms of infection. Arterial
blood gases should be repeated after 30–60 min on NIV to assess therapy.
A repeat of arterial blood gases on the aforementioned NIV settings reveals: pH

7.29; PaCO2 10.5 kPa (79 mmHg); PaO2 7.1 kPa (53 mmHg); HCO3− 30 mmol⋅L−1.
Practice point 3 Failure to improve hypercapnia in patients using NIV may

occur for a number of reasons. It is important to identify the underlying reason
behind the persistent hypercapnia, in order to address the problem correctly.
Causes of persistent hypercapnia/hypoxaemia

• Excessive leak from the mask
• Insufficient time on NIV
• Inappropriately high oxygenation
• Insufficient supplemental oxygen
• Carbon dioxide rebreathing
• Ventilator–patient asynchrony
• Pressure support too low (pressure support = IPAP−EPAP)
• Backup respiratory rate is too low
• EPAP is too low
Solutions
• Refit mask: ensure the mask is correctly fitted using the size gauge provided by
the manufacturer. In the early stages of NIV in acute exacerbations of COPD,
a full face mask is usually the interface of choice, as patients with ARF usually
mouth-breathe to bypass nasal resistance. Patients who mouth-breathe may
have excessive leakage if using a nasal interface.
• In the early stages of NIV, some patients may struggle to use the ventilator. Some
feel claustrophobic with the mask in place; others find the pressures difficult to
tolerate at first. Anxiety can also limit use. Many of these issues can be resolved by
spending time at the bedside with the patient, providing reassurance and explan
ation. The patient can hold the mask to their own face without using the straps and
headgear at first, in order to acclimatise. Dealing with dyssynchrony problems can
also improve compliance. Dyssynchrony is covered in more detail in case study 3.
• In patients with COPD, target oxygen saturations of 88–92% are sufficient.
Over-oxygenation will make it difficult to control hypercapnia. Remember to
ensure that controlled oxygen therapy is also used during periods off NIV.
• By contrast, insufficient oxygenation is dangerous and in the acute scenario,
supplemental oxygen may be necessary, at least until all other issues are

resolved and ventilation (and correction of hypoventilation) can be optimised.

If saturations cannot be maintained >88%, supplementary oxygen may be
required. Oxygen can be entrained into the NIV circuit via a T-piece or into a
port in the mask. A higher FIO2 is achieved when oxygen is entrained more prox-
imally. Delivered oxygen concentration is also affected by ventilator settings,
leaks and type of exhalation port. Some ventilators have an oxygen blender
within the ventilator, enabling more accurate delivery of FIO2 and higher oxygen
concentrations to be achieved.
• Rebreathing carbon dioxide will result in poor gas exchange that is not corrected
by ventilation. You should ensure the exhalation port is patent if this is used or
that a vented mask is in place if not. A minimum EPAP of 4 cmH2O is required
to flush carbon dioxide from the circuit in most bilevel single-circuit ventilators.
• Ventilator–patient dyssynchrony is discussed further in case study 3.
• Pressure support may be increased by increasing the IPAP setting in order to
improve gas exchange once the above issues have been resolved. Increase the
pressure by ~2-cmH2O increments at a time to the maximum level comfortably
tolerated by the patient. At the bedside, aim to reduce respiratory rate and work
of breathing, and increase oxygen saturations.
• The backup respiratory rate in most patients who are awake and alert will only
be utilised during sleep. Nevertheless, in drowsy patients or in those patients
who tolerate higher IPAP levels poorly, increasing the backup rate to just below
the patient’s spontaneous respiratory rate can improve gas exchange.
• Increasing EPAP to ~6–7 cmH2O may be helpful in the presence of pulmon
ary oedema or atelectasis. EPAP levels may need to be much higher than
this in patients with OSA. This is covered in more detail in the section enti-
tled “Patients with acute hypercapnic respiratory failure and OHS”. Note that
increasing EPAP without a corresponding increase in the IPAP setting will
reduce the effective pressure support (pressure support = IPAP–EPAP).
Case study 2: mask leak

A 43-year-old male with ALS is admitted with right lower lobe pneumonia. He is
fed via gastrostomy tube and is wheelchair-bound but has a good quality of life.
He has an advance directive in place that states that he does not wish to be resus-
citated or be intubated and ventilated. However, he has in the past used NIV when
unwell and is happy to use this again if required. He is given intravenous anti
biotics and fluids, and controlled oxygen therapy. His initial arterial blood gases in
the emergency department breathing 28% oxygen are as follows: pH 7.28; PaCO2
7.8 kPa (59 mmHg); PaO2 9.0 kPa (68 mmHg); HCO3− 27 mmol⋅L−1.
You decide to commence NIV at initial settings of: IPAP 14 cmH2O; EPAP 5 cmH2O;
backup rate 12 breaths⋅min−1; entraining 1–2 L⋅min−1 oxygen to achieve target
saturations of ≥94%.
Practice point 1 Patients with neuromuscular disease usually have normal

lungs and, therefore, higher target oxygen saturations are more appropriate
than in patients with COPD or other intrinsic lung diseases. Some patients
with severe respiratory muscle weakness may have difficulty in triggering
the ventilator, particularly during sleep. These patients are often dependent

on the ventilator’s set backup respiratory rate. The backup rate should
therefore be set just below the patient’s spontaneous waking respiratory
rate to ensure adequate ventilation during sleep.
You are called to review the patient after 30 min as the nursing staff report he is
not doing well on the ventilator. On your arrival, you can hear noisy ventilation,
and you can see that he is not synchronising well with the ventilator and looks
uncomfortable. You can feel air leaking when you place your hands around the
bottom perimeter of the mask.
Practice point 2 Patients with neuromuscular disease can sometimes pre-

sent problems with mask fitting as their muscle weakness means that
their jaws fall open, which can lead to leakage if the mask has been fit-
ted with the jaw closed. The same problem can occur in any patient when
asleep. It may help to fit the mask with the jaw relaxed and partly open.
Leaks can be detected by looking, listening and feeling. The patient may
appear uncomfortable and may not be well synchronised with the ventila-
tor. Dyssynchrony occurs because leaks impact on ventilator triggering and
cycling, causing prolonged inspiration and delayed cycling into expiration.
This is very uncomfortable for the patient and contributes to intolerance.
Leaks may be felt by placing the hands around the perimeter of the mask.
Leaks can also be seen visually on many ventilator display screens or data
downloads (fig. 1).
Causes of leaks
• Mouth-breathing with a nasal interface
• Incorrectly sized interface
• Overtightened straps or pressure sores
• Unusual facial anatomy
Figure 1. Leak as seen on the ventilator display screen.

Figure 2. Chin strap. Image © ResMed Limited. All rights reserved.
Solutions
• A full face mask should be used in patients with ARF who usually mouth-
breathe in order to bypass nasal resistance. An alternative is to use a chin-strap
(fig. 2) or cervical collar to prevent the jaw falling open but many patients find
these uncomfortable.
• The mask should always be fitted using the manufacturer’s sizing gauge to
avoid incorrect fitting.
• Leaks may be reduced by adjusting the head straps but care should be taken
not to overtighten the straps as this will lead to discomfort and pressure sores,
and lead to further leaks. Try taking the mask off the face and repositioning it
to reset the seal, before gently tightening the straps to the minimum tension
required to minimise leak.
• Some patients’ facial anatomy means that some models of mask will leak, even
when well-fitted. In these cases, a different model of full face mask may be
tried. It is unusual to require an individually customised mask but these are
available.
Case study 3: ventilator–patient dyssynchrony
A 70-year-old female with kyphoscoliois was admitted 2 days ago with fever and
purulent sputum, and has been treated with intravenous antibiotics. She has
become more drowsy and unwell in the past few hours, and arterial blood gases
performed on 4 L⋅min−1 oxygen via nasal cannulae reveals: pH 7.33; PaCO2 8.1 kPa
(61 mmHg); PaO2 7.2 kPa (54 mmHg); HCO3− 29 mmol⋅L−1.
Practice point 1 Patients with kyphoscoliosis develop hypoventilation during

rapid eye movement sleep before daytime respiratory failure is observed
but may present acutely in the context of intercurrent infection in type 2
respiratory failure. NIV is helpful in this context, and it is usually appropriate

to discuss long-term NIV with these patients, even if their daytime PaCO2
returns to normal once the acute illness has resolved, as these patients are
very likely to need domiciliary NIV within the next year. Note that patients
with chest wall disease often require high inspiratory pressures because
their chest wall is very noncompliant. IPAP levels of ≥25 cmH2O may be
necessary to achieve adequate ventilation.
NIV is commenced on the ward with initial settings as follows: IPAP 22 cmH2O;
EPAP 4 cmH2O; backup rate 14 breaths⋅min−1. You attend to review the patient,
and find her agitated and pulling the mask off her face. She tells you that she is not
going to continue using the ventilator as it is too uncomfortable.
Practice point 2 It is not unusual for patients to find NIV uncomfortable when
first applied. Time should be taken at the outset to explain the rationale
for its use and to titrate settings carefully at the bedside. A family member
or friend may be able to provide additional assurance. The most common
reason for poor tolerance of NIV in the acute phase is patient–ventilator
dyssynchrony as a result of leak and/or inappropriate ventilator settings.
The management of leaks was discussed earlier. Failure to correct hypoxia
and/or hypercapnia will also result in dyssynchrony, and was also covered
earlier. Assuming that leakage is controlled, comfort settings on the ventila-
tor can be adjusted to improve synchrony.
Causes of dyssynchrony
• Leak
• Anxiety
• Delayed or premature cycling to expiration
• Failure to trigger the ventilator
• Excessively rapid or slow rise times
Solutions
• Leak management is covered in case study 2.
• Delayed cycling to expiration occurs in the presence of leaks as the detected
flow remains above the threshold for cycling. This leads to prolonged inspir
ation and reduced time available for expiration. This can be avoided by ensuring
leaks are minimised and by setting a maximum inspiratory time (tImax). tImax
should be set just longer than the patient’s own inspiratory time to allow a
window of opportunity to cycle into expiration. Patients with chest wall disease
often benefit from longer inspiratory times to minimise the peak pressures that
their inspiratory muscles need to generate.
• Failure to trigger the ventilator can be a problem in some patients whose inspir-
atory flow rates may not reach the threshold required to trigger the ventilator
into inspiration. Trigger sensitivity can be increased to avoid this issue. Failure
to trigger may also occur in patients with COPD who have intrinsic PEEP, which
is worsened during acute exacerbations. Increasing the external PEEP (EPAP)
supplied by the ventilator can help supply the additional pressure required to

overcome intrinsic PEEP, preventing wasting of inspiratory effort to generate

inspiratory flow and making ventilator triggering easier.
• Rise time should also be adjusted to improve the patient’s comfort. Very dys
pnoeic patients with rapid respiratory rates require a fast rise time. Patients
with COPD also often benefit from a faster rise time to reduce inspiratory effort
and provide a longer expiratory time in order to reduce gas trapping. By contrast,
patients with chest wall disease benefit from a slower rise time. The rise time
should not be longer than the inspiratory time, or the full inspiratory pressure
will not be reached.
Further reading
• Aboussouan LS, et al. (2013). Respiratory support in patients with amyotrophic
lateral sclerosis. Respir Care; 58: 1555–1558.
• Bello G, et al. (2013). Noninvasive mechanical ventilation: practical advice. Curr
Opin Crit Care; 19: 1–8.
• British Thoracic Society, et al. (2008). Use of non-invasive ventilation in the
management of patients with chronic obstructive pulmonary disease admitted
to hospital with acute type II respiratory failure. www.brit-thoracic.org.uk/
document-library/clinical-information/niv/niv-guidelines/the-use-of-non-
invasive-ventilation-in-the-management-of-patients-with-copd-admitted-to-
hospital-with-acute-type-ii-respiratory-failure/
• Dwarakanath A, et al. (2013). Noninvasive ventilation in the management of acute
hypercapnic respiratory failure. Breathe; 9: 338–348.
• Hess DR (2013). Noninvasive ventilation for acute respiratory failure. Respir Care;
58: 950–972.
• Miller RG, et al. (2009) Practice parameter update: the care of the patient with
amyotrophic lateral sclerosis: drug, nutritional, and respiratory therapies (an
evidence-based review): report of the Quality Standards Subcommittee of the
American Academy of Neurology. Neurology; 73: 1218–1226.
• Shneerson JM, et al. (2002). Noninvasive ventilation for chest wall and
neuromuscular disorders. Eur Respir J; 20: 480–487.
Online resources
• Janssens JP (2014). Troubleshooting – what to do when NIV is not going well? ERS
Course Noninvasive ventilation: basic concepts, Hanover 2014. www.ers-education.
org/events/courses/noninvasive-ventilation-basic-concepts,-hanover-2014.aspx

NIV and the ICU
NIV and weaning
Miquel Ferrer
Mechanical ventilation, via an artificial airway, is a life-saving procedure, but it is

associated with multiple complications, of which the increased risk of nosocomial
pneumonia is the primary concern. Further complications include generalised
muscle weakness, possibly related to sepsis or the sedation or curarisation neces-
sary for invasive mechanical ventilation.
NIV has been used during withdrawal of mechanical ventilation to facilitate extub
ation in patients with difficult or prolonged weaning.
Weaning from invasive mechanical ventilation
The weaning process encompasses all measures that aim to achieve successful
disconnection of the patient from the ventilator. A statement on weaning from
mechanical ventilation made by the International Consensus Conference on
Intensive Care Medicine proposed a classification based on the difficulty and dur
ation of the weaning period.
1) Simple weaning describes patients who proceed from initiation of weaning to
extubation in a single attempt without difficulty.
2) Difficult weaning describes patients who fail the initial weaning attempt and
who require up to three spontaneous breathing trials (SBTs) or up to 7 days
from the first SBT to achieve successful weaning.
3) Prolonged weaning describes patients who fail at least three weaning attempts
or require >7 days of weaning after the first SBT.
Key points
• Patients with chronic airflow obstruction have an increased
risk of difficult or prolonged weaning, particularly when they
present with hypercapnic respiratory failure.
• The use of NIV to facilitate earlier extubation in these patients
is consistently effective in reducing periods of endotracheal
intubation, length of hospital stay and complication rates, and
improving survival.
• These benefits have been demonstrated in haemodynamically
stable patients with a normal level of consciousness and no
fever, who remain able to cough and expectorate.

NIV and the ICU
Discontinuation of invasive mechanical ventilation may be particularly difficult in

patients with chronic respiratory disorders. Persistent weaning failure is associ-
ated with prolonged mechanical ventilation, which is a major risk factor for noso-
comial pneumonia, and with increased morbidity and mortality, especially in
patients with pre-existing chronic respiratory failure.
Different approaches have been used to try and optimise the weaning process.
These include weaning protocols, automated systems, daily SBTs and PSV. In spite
of this, estimates indicate that 20–30% of patients cannot be extubated in the
first weaning attempt.
Prolonged weaning has been shown to be associated with increased mortality and
morbidity, while the differentiation between simple and difficult weaning had no
relevant clinical consequences (fig. 1). Hypercapnia at the end of a SBT strongly
predicts prolonged weaning and worse survival (fig. 2); its presence should lead to
implementation of measures that aim to reduce the need for prolonged mechan
ical ventilation and improve weaning outcome.
a) 25 b) 50
n=14
Ventilator-associated
40
p<0.001 p=0.005
pneumonia %
20
ICU stay days
30
15
n=13
20
n=12
10 10
c) 100
Simple weaning
90
Difficult weaning
80
Survival %
70 Prolonged weaning
p=0.019
60
50
0
0 10 20 30 40 50 60 70 80 90
Time days
Figure 1. a) Length of ICU stay, b) incidence of ventilator-associated pneumonia and

c) Kaplan–Meier survival curves for survival at 90 days from the onset of the weaning pro-
cess for patients with simple, difficult and prolonged weaning. The length of ICU stay and the
incidence of ventilator-associated pneumonia were higher, and the survival probability lower
in the prolonged weaning group. No differences were observed between patients with simple
and difficult weaning. Reproduced and modified from Sellares et al. (2011) Intensive Care
Med; 37: 775–784, with permission from the publisher.
NIV and the ICU
a) 60 b) 100
90
55
p=0.001
PaCO2 mmHg
80
Survival %
50 70
60 p=0.002
45 Normocapnia
50
Hypercapnia
40
0
Simple Difficult Prolonged 0 10 20 30 40 50 60 70 80 90
Time days
Weaning
Figure 2. Relationship between increased PaCO2 and outcomes during weaning. a) PaCO2 at
the end of the first SBT in patients with simple, difficult and prolonged weaning. PaCO2 was
higher in patients with prolonged weaning. b) Kaplan–Meier survival curves of survival at
90 days from the onset of the weaning process for patients with hypercapnia and normo
capnia at the end of the first SBT. The survival probability was lower for patients with
hypercapnia. Reproduced and modified from Sellares et al. (2011) Intensive Care Med; 37:
775–784, with permission from the publisher.
The main recommendations for weaning from mechanical ventilation made by

the International Consensus Conference on Intensive Care Medicine (Boles et al.,
2007) were as follows.
1) Patients should be categorised into the three groups (see earlier) based on the
difficulty and duration of the weaning.
2) Weaning should be considered as early as possible.
3) A SBT is the major diagnostic test to determine whether patients can be suc-
cessfully extubated.
4) The initial trial should last 30 min and consist of either T-tube breathing or
low levels of pressure support.
5) Pressure-support or assist–control ventilation modes should be favoured in
patients failing an initial trial.
6) NIV should be considered in selected patients to shorten the duration of
intubation, but should not be routinely used as a tool for extubation failure.
Rationale for the use of NIV during difficult and prolonged weaning
Respiratory mechanics Patients undergoing a SBT as part of a failed weaning
attempt will probably develop a rapid and shallow breathing pattern, with an
increased workload on the respiratory muscles. This is accompanied by a progres-
sive increase in dynamic lung elastance, intrinsic PEEP, inspiratory resistance and
work of breathing in COPD patients (table 1). Worsening hyperinflation means
that the respiratory muscles work at an increased mechanical disadvantage with a
consequent reduction in capacity. The development of acidosis can result in a fur-
ther reduction in capacity. In these patients, a high drive to breathe coupled with
the imbalance between the increased workload and reduced inspiratory muscle

NIV and the ICU
Table 1. Pathophysiologic basis of weaning failure during transition from positive pressure
ventilation to spontaneous breathing
Rapid and shallow breathing pattern

Increased workload of the respiratory muscles
Increased intrinsic PEEP
Increased elastance and resistance of the respiratory system
Increased work of breathing
Increased effective inspiratory impedance
Increased load/capacity balance
Inappropriate cardiovascular response
Increased venous return to the right ventricle
Increased negative deflections in intrathoracic pressure
Increased left ventricular afterload
Fall of mixed venous PvO2 and SvO2
Impaired neurological status
PvO2: mixed venous oxygen tension. Reproduced from Ferrer et al. (2002) Eur Respir J; 19:
959–965, with permission from the publisher.
strength causes respiratory distress and hypercapnia. These changes are not seen
in patients who successfully tolerate a SBT.
Cardiovascular response The switch from positive pressure ventilation to spontan
eous breathing may increase left ventricular afterload as a result of:
• an increase in the venous return
• large negative deflections in intrathoracic pressure due to the inspiratory
threshold load
An inappropriate cardiovascular response to these changes, involving left ventricu
lar dysfunction and an increased pulmonary artery occlusion pressure, may be
observed during weaning failure. This results in decreased SvO2 during spontan
eous breathing. By contrast, mixed venous oxygen saturation (SvO2) is unchanged
or increased during spontaneous breathing in successfully weaned patients.
Physiological effects of NIV In COPD exacerbations with acute hypercapnia, NIV is
effective in reducing the work of breathing and the large negative deflections in
intrathoracic pressure, with inspiratory positive pressure ventilation and external
PEEP having an additive effect in counterbalancing intrinsic PEEP (table 2). In
these patients, short-term improvements of hypoxaemia and hypercapnia with
NIV are due to increased alveolar ventilation secondary to a slower and deeper
breathing pattern. The ventilation–perfusion mismatch does not change.
In patients with chronic respiratory disorders not yet able to sustain totally spon-
taneous breathing, PSV delivered either through the endotracheal tube or non-
invasively after extubation is equally effective in reducing the work of breathing

NIV and the ICU
Table 2. Physiological effects of noninvasive positive pressure ventilation
Effect on respiratory mechanics

Decrease negative deflections of intrathoracic pressure
Decreased work of breathing
Additive effects of positive pressure ventilation and external PEEP
Effects on gas exchange
Improvement of hypoxaemia and hypercapnia secondary to slower and deeper
breathing pattern
No effects on ventilation–perfusion mismatch
Reproduced from Ferrer et al. (2014) Semin Respir Crit Care Med; 35: 507–518, with permission
from the publisher.
and improving arterial blood gases and respiratory pattern, when compared with
spontaneous breathing during a T-piece trial. Patients were shown to tolerate NIV
better than invasive PSV in this study.
Use of NIV during difficult and prolonged weaning
The primary end-point when NIV is used for patients with difficult or prolonged
weaning should be shortening weaning and avoiding re-intubation. Several RCTs
have assessed the role of NIV in this clinical setting. The most relevant trials are
described here.
• 50 intubated COPD patients with severe hypercapnic respiratory failure, who
had recovered from their exacerbation within 48 h of initiation of mechanical
ventilation but had failed a SBT using a T-piece, were randomly allocated to
receive: 1) extubation and noninvasive PSV; or 2) continuing intubation and
PSV. Patients receiving noninvasive PSV were ventilated for a shorter period,
had a shorter ICU stay, a lower incidence of nosocomial pneumonia and
improved 60-days survival. Noninvasive PSV was as effective as invasive PSV
for maintaining PaCO2 and arterial pH.
• 33 intubated hypercapnic respiratory failure patients with acute-on-chronic
respiratory failure after failure of a single weaning trial were randomly allo-
cated to: 1) remain intubated and receive PSV; or 2) undergo extubation
and noninvasive PSV. NIV reduced the average length of the endotracheal
mechanical ventilation period by 3 days. However, the total duration of ven-
tilatory support related to weaning was longer extubated patients receiving
noninvasive PSV. The incidence of complications associated with intubation
and weaning, length of stay and survival were not significantly different. 65
heart disease patients with post-surgery respiratory failure and chronic pul-
monary disease aggravation with difficult weaning were included in a study
with a similar design. In this study, extubation and NIV was shown to lower
the incidence of pneumonia and tracheostomy, without changing the length
of ICU stay or mortality.
• 208 intubated patients with chronic hypercapnic respiratory failure and dif-
ficult weaning were randomly allocated to receive: 1) conventional invasive
weaning; 2) extubation followed by standard oxygen therapy; or 3) extubation
NIV and the ICU
followed by NIV. Patients receiving extubation followed by NIV had significantly

lower weaning failure rates (which included post-extubation respiratory fail-
ure, re-intubation or death within the 7-day period following extubation) com-
pared with the other treatment arms. The duration of intubation was shorter
for both weaning strategies involving extubation. However, extubation followed
by NIV reduced both the duration of intubation and the risk of post-extubation
• 43 patients with prolonged weaning, 33 of whom had underlying chronic
respiratory disorders, who had failed SBTs on three consecutive days were
randomly allocated to: 1) extubation and NIV; or 2) a conventional wean-
ing approach with continuing intubation. Patients receiving NIV had shorter
periods of both endotracheal and total mechanical ventilation, shorter ICU
and hospital stays, a reduced incidence of nosocomial pneumonia and septic
shock/multiple organ failure, a reduced requirement for tracheostomy to facili-
tate weaning, and improved survival.
A systematic review and meta-analysis of the use of NIV to wean critically

ill adult patients from mechanical ventilation identified a total of 16 RCTs or
quasi-RCTs published or presented as abstracts. NIV had a consistently positive
effect on mortality and ventilator-associated pneumonia as well as other rele
vant outcomes, without increasing the risk of weaning failure or re-intubation
(table 3).
Table 3. Summary estimates of the effect of NIV for weaning critically ill adults off invasive
ventilation
Outcome Summary estimate (95% CI)

Mortality# 0.53 (0.36–0.80)
Ventilator-associated pneumonia# 0.25 (0.15–0.43)
Weaning failure# 0.63 (0.42–0.96)
Length of stay
ICU¶ −5.6 (−7.9– -3.3)
Hospital¶ −6.0 (−9.2– −2.9)
Duration of mechanical ventilation
Total¶ −5.6 (−9.5– −1.8)
Related to weaning¶ −0.3 (−2.1–1.6)
Endotracheal¶ −7.4 (−10.3– −4.6)
Adverse effects
Re-intubation# 0.65 (0.44–0.97)
Tracheostomy# 0.19 (0.08–0.47)
Arrhythmia# 0.89 (0.34–2.34)
#:relative risk; ¶: weighted mean difference in days. Reproduced from Burns et al. (2014) with

NIV and the ICU
Table 4. Summary estimates of the effect of NIV for weaning critically ill adults off
invasive ventilation, stratified according to studies that included exclusively COPD patients
or mixed populations with a variable proportion of these patients
Outcome Summary estimate (95% CI)

Mortality#
COPD patients 0.36 (0.24–0.56)
Mixed populations 0.81 (0.47–1.40)
Weaning failures#
Ventilator-associated pneumonia#
#: relative risk. Data from Burns et al. (2014).
Efficacy of NIV in patients with and without COPD

The benefits associated with use of NIV to facilitate weaning for several outcomes,
particularly mortality, were significantly greater in studies that included exclu-
sively patients with COPD (table 4). In contrast, studies in mixed populations with
variable proportions of COPD patients, including one in which COPD was an exclu-
sion criterion, did not find significant reductions in mortality associated with the
use of NIV.
Limitations of the use of NIV during weaning
The majority of published studies have been conducted in selected populations of
patients with underlying chronic airflow obstruction and hypercapnic respiratory
failure, who are haemodynamically stability, and have a normal level of conscious-
ness, no fever and a preserved cough reflex. Therefore, use of NIV in other popula-
tions cannot currently be recommended due to the lack of clinical studies.
It is necessary to distinguish between dependence on the ventilator and on
endotracheal intubation; therefore, at the start of the weaning process it is neces-
sary to evaluate the following: 1) the requirement for ventilatory support using
a SBT; and 2) whether or not the artificial airway is required. NIV can only be
expected to be a beneficial supportive technique for weaning when at least one
weaning attempt has failed (i.e. the patient is ventilator dependent) and when
there is no requirement for the artificial airway.
The main reasons for NIV failure are: lack of cooperation, excessive secretions,
severe strength–load imbalance and haemodynamic instability. All of these fac-
tors can be addressed by protection of the airways and proper medical therapy.
An estimated ∼30–35% of intubated COPD patients with hypercapnic respiratory
failure undergoing difficult or prolonged weaning are unlikely to benefit from NIV,
but further large-scale studies are needed to confirm this. An algorithm for the use
of NIV as part of a weaning protocol is shown in figure 3.
NIV and the ICU
Improvement in underlying indication for mechanical ventilation
Daily screening of respiratory function

PaCO2 >60 mmHg (>8 kPa) at FIO2 ≤0.4 and PEEP ≤5 cmH2O
Awake patient
Haemodynamically stable
No Yes
Mechanical ventilation SBT

and daily screening
No signs of poor tolerance Signs of poor tolerance
Contraindications for NIV No contraindications for NIV

Extubation
Chronic respiratory disease
Hypercapnia
Gradual withdrawal of
ventilatory support
Prolonged ventilation
Extubation + NIV
Consider tracheostomy
Figure 3. Proposed algorithm for the use of NIV as part of a protocol-driven weaning process.
Reproduced and modified from Esteban et al. (1998) Intensive Care Med; 24: 999–1008.
Acknowledgements
This work was supported by Centro de Investigación Biomédica en Red-
Enfermedades Respiratorias (CibeRes CB06/06/0028)-Instituto de Salud Carlos
III (ISCiii), 2009 SGR 911, and Institut d’Investigacions Biomèdiques August Pi I
Sunyer (IDIBAPS).
Further reading
• American Thoracic Society, et al. (2001). International Consensus Conferences in
Intensive Care Medicine: noninvasive positive pressure ventilation in acute respira-
tory failure. Am J Respir Crit Care Med; 163: 283–291.
• Boles JM, et al. (2007). Weaning from mechanical ventilation. Eur Respir J; 29:
1033–1056.
• Burns KE, et al. (2014). Noninvasive ventilation as a weaning strategy for mechan-
ical ventilation in adults with respiratory failure: a Cochrane systematic review.
CMAJ; 186: E112–E122.

NIV and the ICU
• Ferrer M, et al. (2003). Noninvasive ventilation during persistent weaning failure:

a randomized controlled trial. Am J Respir Crit Care Med; 168: 70–76.
• Ferreyra G, et al. (2011). Are guidelines for non-invasive ventilation during wean-
ing still valid? Minerva Anestesiol; 77: 921–926.
• Girault C, et al. (2011). Non-invasive ventilation and weaning in chronic hypercap-
nic respiratory failure patients: a randomized multicenter trial. Am J Respir Crit Care
Med; 184: 672–679.
• Nava S, et al. (1998). Noninvasive mechanical ventilation in the weaning of
patients with respiration failure due to chronic obstructive pulmonary disease.
A randomized, controlled trial. Ann Intern Med; 128: 721–728.
• Provost KA, et al. (2013). Complications associated with mechanical ventilation. In:
Tobin MJ, ed. Principles and Practice of Mechanical Ventilation. 3rd Edn. New York,
McGraw-Hill; pp. 973–994.
• Tobin MJ, et al. (2006). Discontinuation of mechanical ventilation. In: Tobin MJ, ed.
Principles and Practice of Mechanical Ventilation. 2nd Edn. New York, McGraw-
Hill; pp. 1185–1220.

NIV and the ICU
NIV to avoid re-intubation
Paolo Navalesi and Federico Longhini
Some patients experience respiratory failure requiring re-intubation after extub

ation even though they have fulfilled the criteria for weaning and successfully
passed the spontaneous breathing trial (SBT). Extubation failure can be defined
as an episode of respiratory failure requiring ventilatory support, occurring within
48 h (72 h in some studies) of deliberate or unplanned (accidental) extubation.
Rates of extubation failure as high as 23.5% have been reported.
Irrespective of the underlying disease, re-intubation following extubation fail-
ure is a major clinical problem. Compared with successfully extubated patients,
those experiencing post-extubation failure and re-intubation are older and
more severely ill, as indicated by a higher Acute Physiology and Chronic Health
Evaluation (APACHE) II score at the onset of the weaning trial. The cause of extub
ation failure and the time elapsed before re-intubation are both independent pre-
dictors of outcome.
The patients who experience extubation failure and re-intubation are characterised,
in contrast to those who are successfully extubated, by marked clinical deterioration
for equivalent levels of severity at extubation, which suggests a d irect effect of
these events on patient outcome. Any strategy capable of reducing the occurrence
of post-extubation respiratory failure or avoiding re-intubation in patients with
overt respiratory failure after extubation would, therefore, improve patient
outcome. Because NIV may play a role to accomplish both these aims, various
studies investigated its role for these purposes.
Here, we briefly summarise the use of NIV as a means to either prevent or treat
post-extubation respiratory failure. We refer only to the use of NIV, i.e. PEEP
plus inspiratory support, in ICU patients and do not consider the application
of CPAP to avert post-operative respiratory complications after major surgical
procedures.
Key points
• NIV may reduce the rate of post-extubation respiratory failure,
but only for patients at increased risk of extubation failure.
• NIV is not effective in treating overt post-extubation

NIV and the ICU
NIV to prevent post-extubation failure

In 1999, Jiang et al. first hypothesised that early application of NIV after extubation
might prevent the occurrence of post-extubation respiratory failure and, therefore,
positively affect patient outcome by decreasing the rate of extubation failure and
re-intubation. It is noteworthy that, at the time Jiang and colleagues designed
their study and started patient enrolment, the consequences and causes of extub
ation failure were not known. The authors enrolled 93 consecutive patients fol-
lowing planned or unplanned extubation, who were randomised, regardless of
age, underlying disorder and comorbidities, to either preventive NIV (treatment
group) or standard treatment (control group). They found no difference in the rate
of re-intubation between the two groups. More recently, another large RCT was
conducted by Su et al. (2012), who included all patients (n = 406) extubated after
passing the SBT, without selecting patients considered at risk of extubation failure.
Consistent with the previous study, compared with the standard treatment, NIV
did not result in any improvement of the clinical outcomes, i.e. rate of intubation
and ICU mortality (primary end-points). Notably, impaired clearance of secretions
was the main reason for extubation failure in more than one third of patients who
were re-intubated.
Two multicentre RCTs were conducted in the same period in Italy (Nava et al., 2005)
and Spain (Ferrer et al., 2006) to ascertain whether NIV could avert extubation
failure when applied immediately after deliberate extubation, in selected patients
at risk of post-extubation respiratory failure (table 1). The Italian trial enrolled 97
patients who were randomised 1 h after extubation, subsequent to a successful SBT,
to receive either NIV (≥8 h⋅day−1 for 2 days) or the sole standard treatment (oxy-
gen therapy). Compared with controls, patients in the NIV group had lower rates of
re-intubation, which resulted in a reduced risk of ICU mortality. The Spanish study
enrolled 162 patients, also considered at risk of extubation failure, although with
criteria different from those adopted in the previous investigation (table 1). Patients
were randomised to either NIV (almost continuously for 24 h after extubation) or
oxygen therapy only. The authors reported a significant reduction in the number
of patients experiencing post-extubation respiratory failure (primary end-point) and
ICU mortality. The differences in the rate of re-intubation, ICU and hospital lengths
of stay, and hospital mortality, however, were not significantly different between the
two groups. A post hoc analysis showed that NIV also improved hospital and 90-day
mortality in the NIV group, but only in the subgroup of patients who developed
hypercapnia during the SBT prior to extubation. In contrast, NIV did not produce any
benefits in the subgroup of nonhypercapnic patients. Subsequent to these observa-
tions, the same authors later enrolled a selected group of 106 patients with chronic
respiratory disorders developing hypercapnia during the SBT to investigate the
benefits of NIV, as opposed to oxygen therapy alone, after extubation. While ICU and
hospital death rates were not different between the treatment and control groups,
the 90-day survival rate was significantly higher in the NIV group. Consistent with
previous data, the incidence of respiratory failure after extubation was lower in the
NIV group than in the controls. Lengths of ICU and hospital stay, complications
related to mechanical ventilation, and rates of ventilator-associated pneumonia and
tracheostomy were not different between groups.
A recent small RCT (Ornico et al., 2013) assessed the beneficial effects of adding
NIV to oxygen therapy in a less selected group of patients for whom the only
Table 1. Risk factors for extubation failure in different studies
Jiang Nava Ferrer El-Solh Ferrer Su Vianello Ornico

et al. et al. et al. et al. et al. et al. et al. et al.
(1999) (2005) (2006) (2006)# (2009) (2012) (2011)# (2013)
>1 failure of weaning or extubation • •
Congestive heart failure •
Hypercapnia at extubation or CRD • • •
>1 comorbidity •

Weak cough • •
Stridor after extubation •
>65 years of age •
HF as the reason for intubation •
APACHE II >12 at extubation •
ARF requiring >72 h of IMV •
Obesity¶ •
Swallowing impairment •
Neuromuscular disease •
None • •
CRD: chronic respiratory disease; HF: heart failure; IMV: invasive mechanical ventilation. #: not an RCT (historical control group); ¶: BMI ≥35 kg⋅m−1.
NIV and the ICU
129
NIV and the ICU
inclusion criterion was mechanical ventilation for >72 h due to ARF, predomin
antly because of an acute exacerbation of COPD. In 40 patients who passed
the SBT and were extubated, NIV or standard treatment was randomly applied.
Despite the small number of patients, NIV was able to reduce the rates of re-
intubation and death; noteworthily, the difference in re-intubation rate was
maintained even when excluding COPD patients.
Two nonrandomised studies using historical control groups evaluated the use of
NIV to prevent extubation failure in very specific populations, i.e. morbidly obese
(El-Solh et al., 2006) and neuromuscular patients (Vianello et al., 2011), con
sidered highly at risk for extubation failure. In the obese patients, NIV resulted
in a reduction of the rate of post-extubation respiratory failure and ICU and hos
pital lengths of stay. The neuromuscular patients, who received mechanical cough
assistance through the noninvasive interface in addition to NIV, showed lower
rates of re-intubation and tracheostomy, and spent fewer days in the ICU, com-
pared with the historical control group.
As presented in table 1, there was little consistency in the criteria used to consider
patients “at risk” of extubation failure between different studies. While a beneficial
effect of NIV on preventing post-extubation respiratory failure appears to be evi-
dent (table 2), it remains to be clarified how to discriminate between the patients
who will benefit and those who will not.
NIV to treat post-extubation respiratory failure
Based on case series published in the early 1990s (Meduri et al., 1991), NIV has
been proposed as a means to treat patients with signs of incipient or even estab-
lished respiratory failure after extubation. Following these initial observations,
Hilbert et al. (1998) reported a promising study of the use of NIV to treat overt
post-extubation respiratory failure in COPD patients who developed hypercapnic
acidosis within 72 h of extubation. NIV reduced the need for re-intubation, dur
ation of ventilatory assistance and ICU length of stay, when compared with the
conventional treatment applied in matched subjects from an historical control
group of COPD patients.
In 2002, the first RCT (Keenan et al., 2002) investigated the role of NIV, as opposed
to standard treatment, to avert re-intubation in 358 patients who developed re
spiratory distress, as defined by tachypnoea and the use of respiratory accessory
muscles, within 48 h after extubation. As summarised in table 2, this single-
centre RCT did not find any beneficial effect of NIV on the rate of re-intubation,
ICU and hospital mortality, or lengths of ICU and hospital stay. It is worth mention-
ing that, contrary to the aforementioned nonrandomised study, the COPD patients
enrolled in this trial represented only 11% of the overall population, while the
vast majority of patients belonged to other diagnostic categories, primarily cardio
logical (35%) and respiratory conditions of different aetiologies (32%).
2 years later, a second RCT (Esteban et al., 2004) was performed in 37 centres
distributed in eight countries. This study enrolled 221 patients who developed
an episode of respiratory failure in the 48 h following extubation, as defined by at
least two of the following: hypoxaemia, respiratory acidosis, tachypnoea and signs
of respiratory distress. Patients were randomised to receive either NIV or standard
treatment. The trial confirmed the lack of benefit of NIV, compared with standard

Table 2. Outcome differences (intervention versus control group) in the RCTs
NIV to prevent post-extubation respiratory failure NIV to avoid re-intubation#

Jiang Nava Ferrer Ferrer Su Ornico Keenan Esteban
et al. et al. et al. et al. et al. et al. et al. et al.
(1999) (2005) (2006) (2009) (2012) (2013) (2002) (2004)
Post-extubation 16 versus 33* 15 versus 15 versus
respiratory failure 48* 13
Re-intubation 28 versus 8 versus 24* 11 versus 22 11 versus 19 10 versus 5 versus 39* 72 versus 69 48 versus 48

15 8
ICU length of stay 9 ± 6 11 ± 8 11 ± 13 17 ± 12 15 ± 11 18 (11–30)
days versus versus versus versus versus versus 18
12 ± 15 13 ± 11 10 ± 9 18 ± 12 19 ± 25 (11–26)
Hospital length of 23 ± 16 30 ± 23 29 ± 27 32 ± 25
stay days versus versus versus versus
25 ± 21 29 ± 18 24 ± 17 30 ± 28
ICU mortality 3 versus 14* 6 versus 8 1 versus 1 15 versus 24 25 versus 14*
Hospital mortality 12 versus 18 16 versus 23 11 versus 22 0 versus 22* 31 versus 31
90-day mortality 11 versus 31*
Data are presented as %, mean±sd or median (interquartile range). #: in patients with established post-extubation respiratory failure. *: p<0.05 for intervention
versus control group.
NIV and the ICU
131
NIV and the ICU
treatment, on the rate of re-intubation and ICU length of stay (table 2). It is note-
worthy that ICU mortality was higher in the treatment group, suggesting that NIV
could lead to harm because of delayed intubation. In fact, the median time from
respiratory failure to re-intubation was longer in the NIV group.
Based on the findings of these two RCTs, the use of NIV to avoid re-intubation in
patients with overt respiratory distress and/or respiratory failure consequent to failed
extubation is not advisable. However, because of the limited number of patients
studied and some limitations of these two trials, this statement not definitive.
Furthermore, before undergoing re-intubation, several controls experiencing respira
tory failure were successfully treated by rescue NIV in two of the aforementioned
trials (Ferrer et al., 2009; Esteban et al., 2004) and in one multicentre RCT performed
to evaluate NIV as a weaning strategy in patients with acute-on-chronic respiratory
failure (Girault et al., 2011). It should be noted that in the study by Esteban et al., the
use of NIV as “rescue” therapy had a much higher rate of success than in the initial
treatment group. In fact, further studies are necessary on this issue.
Conclusions
A flowchart depicting the role of NIV after extubation is presented in figure 1.
In summary, NIV can effectively avoid the risk of re-intubation by preventing
Extubation
Not present “At risk” Present

criteria#
Oxygen therapy Prophylactic NIV + oxygen¶
Not present Present

Respiratory failure
within 48 h+
Successful extubation Endotracheal intubation

Evaluate rescue NIV§
Figure 1. NIV and post-extubation respiratory failure. #: see table 1 for criteria. ¶: 24-48 h
after extubation, ≥8 h⋅day−1, PEEP 4–8 cmH2O, inspiratory support 8–12 cmH2O. +: two or
more of: clinical signs of respiratory muscle failure and increased respiratory effort, respira-
tory rate >25 breaths⋅min−1 for two consecutive hours, respiratory acidosis (pH <7.35, PaCO2
>6.0 kPa (>45 mmHg)), or hypoxaemia (SaO2 <90% or PaO2 <10.7 kPa (<80 mmHg) with
FIO2 >50%). §: if NIV was not previously administered.
NIV and the ICU
post-extubation respiratory failure in selected patients at risk of extubation fail-

ure. However, clear-cut criteria to select the patients that could benefit are lack-
ing, which makes further investigation necessary. Widespread and indiscriminate
application of NIV to avert re-intubation in the presence of established post-
extubation respiratory failure is currently not advisable.
Further reading
• El-Solh AA, et al. (2006). Noninvasive ventilation for prevention of post-extubation
respiratory failure in obese patients. Eur Respir J; 28: 588–595.
• Epstein SK (2010). Non-invasive ventilation for weaning and extubation failure. In:
Elliot M, et al., eds. Non-invasive Ventilation and Weaning. Principles and Practice.
London, Edward Arnold Ltd; pp. 530–542.
• Esteban A, et al. (2004). Noninvasive positive-pressure ventilation for respiratory
failure after extubation. N Engl J Med; 350: 2452–2460.
• Ferrer M, et al. (2006). Early noninvasive ventilation averts extubation failure in
patients at risk: a randomized trial. Am J Respir Crit Care Med; 173: 164–170.
• Ferrer M, et al. (2009). Non-invasive ventilation after extubation in hypercapnic
patients with chronic respiratory disorders: randomised controlled trial. Lancet;
374: 1082–1088.
• Ferrer M, et al. (2012). NIV in withdrawal from mechanical ventilation. In: Ferrer
M, et al., eds. New Developments in Mechanical Ventilation. Eur Respir Monogr;
55: 191–205.
• Girault C (2008). NIV for weaning from mechanical ventilation and post-extubation
ARF. In: Muir J-F, et al., eds. Noninvasive Ventilation. ERS Monogr; 41: 143–153.
• Girault C, et al. (2011). Noninvasive ventilation and weaning in patients with
chronic hypercapnic respiratory failure: a randomized multicenter trial. Am J Respir
Crit Care Med; 184: 672–679.
• Hilbert G, et al. (1998). Noninvasive pressure support ventilation in COPD patients
with postextubation hypercapnic respiratory insufficiency. Eur Respir J; 11:
1349–1353.
• Jiang JS, et al. (1999). Effect of early application of biphasic positive airway pres-
sure on the outcome of extubation in ventilator weaning. Respirology; 4: 161–165.
• Keenan SP, et al. (2002). Noninvasive positive-pressure ventilation for postextuba-
tion respiratory distress: a randomized controlled trial. JAMA; 287: 3238–3244.
• Meduri GU, et al. (1991). Noninvasive face mask mechanical ventilation in patients
with acute hypercapnic respiratory failure. Chest; 100: 445–454.
• Nava S, et al. (2005). Noninvasive ventilation to prevent respiratory failure after
extubation in high-risk patients. Crit Care Med; 33: 2465–2470.
• Nava S, et al. (2006). Time of non-invasive ventilation. Intensive Care Med; 32:
361–370.

NIV and the ICU
• Ornico SR, et al. (2013). Noninvasive ventilation immediately after extubation

improves weaning outcome after acute respiratory failure: a randomized con-
trolled trial. Crit Care; 17: R39.
• Su CL, et al. (2012). Preventive use of noninvasive ventilation after extubation: a
prospective, multicenter randomized controlled trial. Respir Care; 57: 204–210.
• Vianello A, et al. (2011). Prevention of extubation failure in high-risk patients with
neuromuscular disease. J Crit Care; 26: 517–524.

NIV and the ICU
NIV in the perioperative period
Yuda Sutherasan, Maria Vargas and Paolo Pelosi
Post-operative pulmonary complications (PPCs) increase morbidity and mor-

tality. There is no standard definition of a PPC, but they include post-operative
pneumonia, ARF, hypoxaemia, atelectasis, bronchospasm and fever. The reported
incidence of PPCs varies considerably, from 2% to 40%. PPCs are attributable
to anaesthetic, surgical and patient-related factors. Pulmonary function may be
modified by the operative procedure, general anaesthesia and/or host factors.
ARF requiring endotracheal intubation and invasive mechanical ventilation is an
independent predictor of high mortality. Consequently, NIV may help avoid the
complications associated with invasive ventilation. This section focuses on the
rationale, benefits and techniques of NIV during the perioperative period, espe-
cially in groups of patients who are at risk of PPCs.
Mechanisms of surgery-induced pulmonary complications
General anaesthesia has different effects on the respiratory system that begin
at induction and can persist for days after surgery. General anaesthesia reduces
functional residual capacity (FRC) by compression of the lung tissue and impairs
surfactant function, thus resulting in the occurrence of atelectasis in depend-
ent lung regions. General anaesthesia may contribute to atelectasis by affect-
ing the control of breathing and decreasing neural drive to the airway and chest
wall. General anaesthesia also reduces the muscle tone and impairs the ability to
cough. Surgical trauma adjacent to the diaphragm:
• alters the mechanical efficiency of the diaphragmatic, abdominal and thoracic
wall muscles
• induces pain
• inhibits phrenic and intercostal nerve activity
Key points
• Perioperative cases most likely to benefit from NIV are high-risk
patients with COPD, cardiac failure and obesity, and those
undergoing high-risk procedures, i.e. cardiac, thoraco-abdominal
and abdominal surgery.
• Physicians should consider absolute and relative
contraindications before initiating NIV in the
perioperative period.

NIV and the ICU
Atelectasis in the post-operative period results in alteration of ventilation/per-

fusion ratios leading to hypoxaemia and pneumonia. Other risk factors for PPCs
are older age, smoking, American Society of Anesthesiologists physical status >II,
COPD, OSAS and cardiovascular disease.
Application of NIV during perioperative period
NIV may be beneficial for high-risk patients who undergo high-risk surgery (i.e.
thoracic and abdominal surgery). NIV has been proposed as a preventive/prophy-
lactic or therapeutic/curative treatment for established ARF after surgery. The
aim of preventive/prophylactic NIV is to avoid the development of ARF in high-
risk patients.The aims of therapeutic/curative NIV are to treat ARF, and to avoid
endotracheal intubation and related complications.
The two main, commonly used noninvasive respiratory support techniques are:
• CPAP
• bilevel positive airway pressure ventilation (also known as PSV plus PEEP)
CPAP delivers constant positive airway pressure during both inspiration and expir
ation by the use of a flow generator or a portable compressor. In the context of
perioperative patients, CPAP increases intrathoracic pressure, maintains upper
airway patency, prevents alveolar collapse and atelectasis, improves FRC, and may
improve left ventricular function. In some patients (e.g. those with COPD), CPAP
decreases the work of breathing by counterbalancing auto-PEEP. Bilevel positive
airway pressure ventilation is delivered by a ventilator that provides two positive
levels of airway pressure. In this way, during inspiration, the patient receives a
pressure-supported breath. PSV improves patient–ventilator synchrony, comfort
and alveolar ventilation. During PSV, patients control not only their respiratory rate
but also their inspiratory and expiratory time. Therefore, NIV usually ensures bet-
ter alveolar ventilation and respiratory muscle unloading than CPAP.
High-risk patients
High-risk patients include those who are obese and those who have OSAS or
COPD/chronic pulmonary disease.
NIV may improve lung function and post-operative oxygenation in obese patients
undergoing bariatric surgery. In obesity, IPAP promotes lung inflation and produces
a dose-dependent effect on pulmonary restrictive syndromes. Moreover, CPAP
alleviates intrinsic PEEP and prevents end-expiratory lung collapse. Anaesthesia
and sedative drugs have been demonstrated to impair upper-airway muscle activ-
ity, particularly in OSAS and/or obesity. Nasal CPAP is recommended in OSAS
patients before and after surgery, immediately following extubation. Patients
previously diagnosed with OSAS should be advised to bring their CPAP machine
and interface into hospital for any surgical intervention, and discuss their CPAP
requirements in advance with the anaesthetist team.
The preventative use of NIV before and after lung surgery for COPD patients has
been shown to ameliorate lung function loss, and to reduce atelectasis and hos
pital length of stay. However, more recently, Lorut et al. (2014) did not demonstrate
any reduction in ARF, intubation rate or hospital stay in COPD patients randomised
to post-operative NIV or control following lung resection. Much must depend on
the severity of COPD, extent of the surgical procedure and comorbidities.
NIV and the ICU
Cardiac surgery
Preventive NIV After cardiac surgery, the incidence of diaphragmatic dysfunction is
higher than following thoracic or abdominal surgery, as a consequence of phrenic
nerve involvement, which can produce a decrease in vital capacity and FEV1 of up to
∼65% on the first post-operative days. NIV has been successfully applied in some
less invasive procedures not requiring general anaesthesia (i.e. percutaneous aortic
valve implant and transoesophageal echocardiography (TEE)). In the TEE procedure,
Guarracino et al. (2010) used homemade interfaces that allowed introduction of the
probe through the sealing port. NIV use attenuated impaired oxygenation, improved
radiological atelectasis and reduced PPCs after cardiac surgery. This approach is also
valuable in long-term home NIV recipients who require such procedures.
Therapeutic NIV The outcome of protocol-driven NIV in patients with ARF after
cardiac surgery showed that only ∼7% of patients could not tolerate NIV. Patients
who received NIV had a decreased incidence of re-intubation and an improvement
in oxygenation. Nevertheless, NIV patients suffered more frequent impairment of
sternal wound healing. Importantly the patients who failed NIV had a lower arterial
blood pH after 24 h of treatment and higher mortality. Close monitoring in an ICU/
high-dependency unit environment and a low threshold to re-intubate patients
where NIV is failing is clearly crucial.
Thoracic surgery
In thoracic surgery, PPCs are the leading cause of mortality following lobectomy
and pneumonectomy. Furthermore, invasive mechanical ventilation increases the
risk of bronchopleural fistula, pulmonary infection and bronchial stump disrup-
tion, which can raise mortality to 60–80%.
After lung resection, patients treated with nasal NIV who achieved an exhaled VT of
8–10 mL⋅kg−1 and a respiratory rate of <25 breaths⋅min−1 demonstrated a reduced
need for invasive mechanical ventilation and decreased mortality compared with
standard therapy, without obvious haemodynamic side-effects. Cardiac comorbid-
ities and a lack of initial response to NIV were independent factors associated with
NIV failure. NIV has been used in phrenic nerve dysfunction after lung transplant
ation. In a case series of 21 patients undergoing bilateral lung transplantation, 18
of the patients who developed ARF and received NIV avoided intubation and were
discharged from the ICU (Rocco et al., 2001).
Abdominal surgery
Preventive NIV Abdominal surgery can cause restrictive pulmonary defects by
elevating the diaphragm, resulting in a decreased FRC and leading to atelec
tasis. A meta-analysis including 10 studies demonstrated, albeit with very low-
quality evidence, that after abdominal surgery, post-operative CPAP reduces
post-operative atelectasis, pneumonia and the rate of re-intubation. However,
the effects of CPAP on mortality, hypoxia or the need for invasive ventilation are
uncertain (Ireland et al., 2014).
Therapeutic NIV A recent meta-analysis, including only RCTs in major surgery,
demonstrated that NIV decreased the length of ICU stay, post-operative re-
intubation rate and incidence of pneumonia when used for weaning and in the
post-operative period. NIV increased hospital survival in weaning and post-surgery,
NIV and the ICU
but not in post-ICU extubation (Glossop et al., 2012). The risk factors associated
with NIV failure in post-oesophagectomy patients were severe post-operative com-
plications (i.e. acute renal dysfunction and cardiac arrest) and PaO2/FIO2 <180 mmHg
(<23.9 kPa) after 2 h of noninvasive positive pressure ventilation. In post-operative
ARF after non-cardiac surgery, risk factors were nosocomial pneumonia and
Simplified Acute Physiology Score 2 ≥35.
Practical aspects of applying NIV during the perioperative period
Contraindications Although the previously mentioned studies have demonstrated
a clear role for NIV during the perioperative period, it is important to note that NIV
practitioners should carefully consider the contraindications to NIV application.
Table 1 summarises the indications and relative and absolute contraindications
for NIV during the perioperative period.
Table 1. Indications and contraindications for the application of NIV during the perioperative
period
Preventive indications
Patients at risk for PPCs or ARF (i.e. obese, OSAS, chronic lung diseases,
congestive heart failure)
Type of surgery (i.e. thoraco-abdominal surgery, abdominal surgery, cardiac
surgery and bariatric surgery)
Therapeutic indications
Post-operative ARF
Post-operative pulmonary oedema
Post-operative airway obstruction
Absolute contraindications
Impaired consciousness, severe agitation or encephalopathy
Cardiac or respiratory arrest or haemodynamic instability or unstable cardiac
arrhythmia
Acute myocardial infarction
Copious secretions
Upper gastrointestinal bleeding or haemoptysis
Facial trauma
Uncontrolled vomiting
Relative contraindications
Mildly decreased level of consciousness
Progressive severe respiratory failure
Uncooperative patients who can be calmed or comforted
Suspected acute coronary ischaemia
Pregnancy
Reproduced and modified from Jaber S, et al. (2010b) with permission from the publisher.

NIV and the ICU
Ventilator NIV practitioners often use standard, portable noninvasive ventila-

tors. The most common modes are CPAP and PSV. CPAP is the mode in which
the machine or device delivers constant positive pressure during both inspiration
and expiration, either using fixed and variable high-flow generators or by use of
a portable compressor. By contrast, NIV offers both PEEP and pressure support,
which are applied together. The patient’s spontaneous inspiratory effort triggers
the ventilator, which then provides a variable flow of gas that increases until the
preset pressure support level is reached. Once the selected pressure support level
is reached, the patient can continue to breathe until the inspiratory flow rate drops
below a threshold level (40–60% of the peak flow) when the ventilator will then
cycle to expiratory phase (for further details see the section entitled “Basic prin-
ciples of ventilators”). The VT delivered depends on the patient’s effort in combi-
nation with the pressure support level. The patient can control their respiratory
rate, and inspiratory and expiratory time. Some machines have an additional time
(backup) mode that can be set to deliver the pressure support level at prefixed
time interval as described in the section entitled “Basic principles of ventilators”.
A backup rate is important in sedated patients or those receiving opiate analgesia
post-procedures.
Interfaces the interface should be selected to fit according to patient comfort and
mask tolerance. Oronasal and nasal masks are more commonly used, although
there is little evidence supporting the advantage of any one type of interface over
others. In 99 patients with post-operative ARF treated with NIV using a helmet,
75% of patients avoided intubation (Redondo Calvo et al., 2012). A comparison of
the efficacy of NIV delivered by a helmet versus a face mask in patients with ARF
after abdominal surgery demonstrated helmet patients were intubated signifi-
cantly less often than mask patients. The main reason for NIV failure was interface
intolerance (Conti et al., 2007). The presence of a nasogastric tube often causes
leaks around an orofacial mask during NIV application. Some manufacturers have
invented specific interfaces limiting leaks around the mask. Furthermore, some
new NIV machines provide PSV with leak compensation aiming to minimise the
impact of leak in the circuit. We recommend maintaining the connection between
the nasogastric tube and drainage bag in order to detect early gastric insufflation
resulting from inappropriate NIV settings. Both IPAP and PEEP/EPAP settings may
need to be reduced in these circumstances.
Venue, set-up and monitoring Patients should be admitted to appropriately moni-
tored locations and usually managed in a head-elevated bed at above 30°. NIV
should be explained to the patient before applying the mask and ventilator.
Ventilator settings should be checked and set before connecting to the patient
using the following initial settings:
• PSV or NIV with a backup rate
• inspiratory trigger of −1– −2 L⋅min−1 or −1– −2 cmH2O
• PSV level of 3–5 cmH2O initially
• expiratory cycling of 40–60% or fixed inspiratory time of 1 s
• PEEP level of 3–5 cmH2O and FIO2 (to maintain optimal SpO2)
When NIV is applied, PSV and PEEP are gradually increased until dyspnoea is
relieved, and the patient achieves the optimal expiratory VT and respiratory rate
without patient–ventilator asynchrony. If a maximal pressure of >25 cmH2O
NIV and the ICU
(PSV+PEEP) is required without leakage or arterial blood gas tensions/pH fail

to improve, patients should be considered for invasive mechanical ventilation.
Humidification should be added as indicated, i.e. in cases with tenacious secre-
tions. Close monitoring of SaO2 and PtcCO2 is required, and arterial blood gases
should be checked within 1–2 h. Please see the discussion of high-flow oxygen
therapy in the section entitled “Supplemental oxygen and humidification” for
further treatment options.
Conclusion
Anaesthesia and surgery can cause pulmonary dysfunction resulting in PPCs
including respiratory failure, thus increasing morbidity and mortality. NIV has a
growing evidence base during the perioperative period, but is not a panacea. The
benefits are not only seen in the prevention of PPCs but also when used as a
therapeutic tool after post-operative ARF has developed. Post-operative patients
most likely to derive benefit are high-risk patients with COPD, heart failure and
obesity/OSA, as well as those undergoing high-risk surgical procedures (e.g. car-
diac, thoraco-abdominal and abdominal surgery). However, the learning curve for
NIV application is important. The education and experience of recovery room and
anaesthetist/ICU/surgery staff, and use in appropriate safe locations (such as the
ICU and high-dependency unit) are as important as identifying the right recipients
for NIV.
Further reading
• Cabrini L, et al. (2013). Non-invasive ventilation in cardiac surgery: a concise
review. Heart Lung Vessels; 5: 137–141.
• Canet J, et al. (2010). Postoperative pulmonary complications. Minerva Anestesiol;
76: 138–143.
• Cereda M, et al. (2013). Noninvasive respiratory support in the perioperative
period. Curr Opin Anaesthesiol; 26: 134–140.
• Chiumello D, et al. (2011). Non-invasive ventilation in postoperative patients:
a systematic review. Intensive Care Med; 37: 918–929.
• Conti G, et al. (2007). Noninvasive positive-pressure ventilation with different
interfaces in patients with respiratory failure after abdominal surgery: a matched-
control study. Respir Care; 52: 1463–1471.
• Glossop AJ, et al. (2012). Non-invasive ventilation for weaning, avoiding reintuba-
tion after extubation and in the postoperative period: a meta-analysis. Br J Anaesth;
109: 305–314.
• Guarracino F, et al. (2010). Non-invasive ventilation-aided transoesophageal echo-
cardiography in high-risk patients: a pilot study. Eur J Echocardiogr; 11: 554–556.
• Ireland CJ, et al. (2014). Continuous positive airway pressure (CPAP) during the
postoperative period for prevention of postoperative morbidity and mortality fol-
lowing major abdominal surgery. Cochrane Database Syst Rev; 8: CD008930.
• Jaber S, et al. (2010a). Postoperative noninvasive ventilation. Anesthesiology; 112:
453–461.

NIV and the ICU
• Jaber S, et al. (2010b). Role of non-invasive ventilation (NIV) in the perioperative

period. Best Pract Res Clin Anaesthesiol; 2010; 24: 253–265.
• Lorut C, et al. (2014). Early postoperative prophylactic noninvasive ventilation after
major lung resection in COPD patients: a randomized controlled trial. Intensive
Care Med; 40: 220–227.
• Redondo Calvo FJ, et al. (2012). Helmet noninvasive mechanical ventilation in
patients with acute postoperative respiratory failure. Respir Care; 57: 743–752.
• Rocco M, et al. (2001). Non-invasive pressure support ventilation in patients with
acute respiratory failure after bilateral lung transplantation. Intensive Care Med;
27: 1622–1626.
• Tusman G, et al. (2012). Atelectasis and perioperative pulmonary complications in
high-risk patients. Curr Opin Anaesthesiol; 25: 1–10.
• Warner DO (2000). Preventing postoperative pulmonary complications: the role of
the anesthesiologist. Anesthesiology; 92: 1467–1472.

NIV and the ICU
NIV for endoscopic procedures
Leo M.A. Heunks, Lisanne Roesthuis and Erik H.F.M. van der Heijden
Endoscopy is frequently performed for diagnostic or therapeutic purposes but is

an invasive procedure with associated risks. For instance, in patients with hypox
aemic or hypercapnic failure, endoscopy has a considerable risk of pulmonary
complications and, to a lesser degree, cardiac complications. In high-risk patients,
for whom noninvasive diagnostic tests are inconclusive, the clinician needs to
decide either to proceed with empirical treatment or endotracheal intubation to
allow endoscopy – for instance, bronchoscopy or transoesophageal echocardi-
ography. It should be noted that both options are unattractive. On one hand, in
immunocompromised patients, the differential diagnosis of pulmonary infiltrates
may be extensive, making it difficult to decide the reasonable empirical treatment.
On the other hand, endotracheal intubation results in high costs and imposes a
considerable risk for the patient. In the past decade, several studies have dem-
onstrated the feasibility of NIV as an alternative to endotracheal intubation to
facilitate endoscopy in hypoxaemic patients. NIV may reduce the risks associated
with endotracheal intubation and reduce healthcare costs, although the latter has
never been formally evaluated.
This section will discuss how NIV may facilitate endoscopy; in particular, flexible
bronchoscopy. We will not discuss the role of NIV as a rescue strategy when severe
hypoxaemia develops during bronchoscopy, as this will be discussed elsewhere
in this book (see section entitled “The patient with acute hypoxaemic respiratory
failure excluding pulmonary oedema”).
Risks of bronchoscopy in hypoxaemic patients

The risk of complications associated with flexible bronchoscopy in patients without
comorbidities is low. However, hypoxaemic patients and, to a lesser extent, hyper-
capnic patients may be at risk for complications during or after bronchoscopy,
Key points
• Hypoxaemia increases the risks of cardiopulmonary
complications during and after endoscopy.
• NIV facilitates endoscopy in hypoxaemic patients.
• We discuss the practical aspects of NIV during endoscopy, in
particular flexible bronchoscopy.

NIV and the ICU
in particular if this procedure is combined with bronchoalveolar lavage (BAL). In

one study, bronchoscopy in hypoxaemic patients (mean PaO2/FIO2 <200 mmHg
(<26.66 kPa)) was complicated by the need to increase FIO2 or to initiate NIV in
35% of the patients within 24 h after bronchoscopy, and led to endotracheal
intubation in 15% of patients. Clinicians should be aware of the risks associated
with bronchoscopy in hypoxaemic patients.
Benefits of NIV during bronchoscopy

Bronchoscopy may increase pulmonary shunt fraction through at least two
mechanisms.
1) With BAL, surfactant will be lost promoting lung collapse.
2) Strong suctioning through the bronchoscope may promote lung collapse.
In addition, the introduction of a bronchoscope in the central airways will increase
airway resistance and the work of breathing. As NIV both recruits lung tissue
through application of PEEP and reduces the work of breathing by application of
inspiratory support, there is a reasonable rationale to use NIV during bronchos-
copy in hypoxaemic patients. Indeed, several clinical studies have demonstrated
that NIV prevents oxygen desaturation during and after bronchoscopy.
Practical aspects of NIV during flexible bronchoscopy

Patient selection In general, hypoxaemic patients will benefit the most from NIV
during bronchoscopy. The reduction in PaO2 during bronchoscopy is ∼1.0 kPa
(∼7.5 mmHg) but after BAL PaO2 may decrease by 3.0 kPa (22.5 mmHg). In our
hospital, we use NIV to facilitate bronchoscopy when an experienced pulmon
ologist deems the patient to be a high-risk for pulmonary complications. In our
experience, a PaO2/FIO2 value <200 mmHg (<26.66 kPa) is a reasonable cut-off
value for the application of NIV, but clinical evaluation is equally important. No
safe lower threshold for oxygenation is defined but, in our experience, BAL can be
performed safely in patients with PaO2/FIO2 ∼140 mmHg (∼18.66 kPa) under NIV.
Bronchoscopy may be performed for diagnostic or therapeutic purposes. Under
both conditions, NIV should be considered when the patient is hypoxaemic or in
respiratory distress.
Contraindications are those related to NIV and are discussed elsewhere in this
book in detail. Specific to this procedure, if SpO2 cannot be increased to at least
90% despite appropriate PEEP and high FIO2, BAL should not be performed with-
out endotracheal intubation. If the reason for bronchoscopy were the presence
of atelectasis, it would be anticipated that oxygenation increases after bron-
choscopy and the clinician could consider proceeding despite low initial oxygen
saturation.
Environment and monitoring Despite the use of NIV, these patients are at risk
for cardiopulmonary complications during and after bronchoscopy. Therefore,
the procedure should be performed in an environment where personnel and
equipment is available for urgent endotracheal intubation, such as an ICU or
post-anaesthesia care unit. In addition, endoscopy should be performed by an
experienced clinician to limit duration of the procedure and enhance the yield of
the lavage.
NIV and the ICU
Analgesia/sedation Topical anaesthesia, for instance lidocaine, should be applied

to the upper airways and requires brief discontinuation of NIV. Whether or not
intravenous sedatives are administered depends on patient preference. The first-
choice sedative for this purpose is propofol, titrated to moderate sedation, either
as repeated i.v. bolus or continuous infusion, based on the intensivist’s prefer-
ence. The advantage of propofol is its short half-life and rather predictable effect.
However, no antidote is available. In addition, alfentanil is provided. Appropriate
continuous monitoring of peripheral oxygen saturation and heart rhythm is neces-
sary with the use of i.v. sedatives.
Interface Many types of interface have been used in the past. We have modified
a full face mask for this purpose. A synthetic cylinder (from a swivel connector)
was secured in the interface. A disposable cap sealed the cylinder to allow the
introduction of the bronchoscope without increasing the amount of leak (fig. 1).
Alternatively, a commercially available disposable “bronchoscopy elbow” with an
airtight diaphragm can be connected to specific interfaces, allowing the intro-
duction of the bronchoscope (figs 2 and 3). Other interfaces, such as the helmet
have been successfully used for this purpose as well. The final choice of interface
depends on clinician's experience and patient preference.
Application of NIV In patents unfamiliar with NIV, a period of habituation for 15–20 min
is recommended prior to endoscopy. The level of PEEP and pressure support should
be individually titrated during this time period. In general, PEEP of 5–8 cmH2O and
Figure 1. Full face mask modified for the use of a bronchoscope during NIV. A synthetic
cylinder was secured in the mask. A disposable cap (from a swivel connector) sealed the
cylinder and prevented air leakage during bronchoscopy. Reproduced from Heunks et al.
(2010) (© the author).
NIV and the ICU
a) b)
Figure 2. a) Side view and b) front view of a commercially available elbow connector that
allows introduction of the bronchoscope during NIV, when this connector is placed between
interface and ventilator tubing.
pressure support of 10–15 cmH2O are sufficient. Individual titration of pressure

should be performed based on SpO2 and respiratory distress. FIO2 is increased to
1.0 just before endoscopy and titrated to maintain SpO2 >95%.
It should be acknowledged that due to loss of surfactant and inflammation ven-
tilation/perfusion mismatch may last up to 12–24 h after lavage. In general, we
would continue NIV for at least 30 min after the end of bronchoscopy to allow
recruitment of lung tissue. Support and FIO2 are tapered guided by SpO2. In some
patients, support by NIV may be necessary for >12 h after bronchoscopy.
Endoscopy From practical perspective, bronchoscopy during NIV is not different
from unassisted endoscopy. Both oral and nasal introduction of the bronchoscope
are possible but, in general, oral introduction is easier due to the position of the
orifice for the bronchoscope in the interface (figs 1 and 3).
Adverse events
Bronchoscopy in hypoxaemic patients should be considered a high-risk proced
ure. Although NIV decreases the risks, in particular oxygen desaturation, clini-
cians should be well prepared for any adverse events related to bronchoscopy or
NIV. Gastric hyperinflation increases the risk of aspiration and should be reduced
by maintaining low inspiratory pressures, whenever possible. When hypoxaemia
develops despite an increase in PEEP and FIO2, the procedure should be discontin-
ued immediately. If severe or persistent hypoxaemia occurs, endotracheal intuba-
tion should be performed. Cardiac complications are rare and mostly secondary to
the development of hypoxaemia.
It should be acknowledged that clinically relevant complications are uncom-
mon with appropriate patient selection. In our clinic, patients with rather severe
hypoxaemia are deemed appropriate for bronchoscopy during NIV, but only under
meticulous monitoring of peripheral oxygen saturation and heart rhythm, in the
presence of a clinician trained for emergency endotracheal intubation.
NIV and the ICU
Figure 3. Oral introduction of the flexible video-bronchoscope through the dedicated

commercially available elbow.
A role for NIV in other endoscopic procedures?

Clinical experience regarding use of NIV to facilitate endoscopic procedures other
than bronchoscopy is rather limited. Very few studies have been reported in the
literature. We will, therefore, discuss this issue in less detail.
Severe orthopnoea in cardiac disease patients may increase the risks associated with
transoesophageal ultrasound as this usually requires the patient to be in supine or lat-
eral position. NIV has been used to facilitate transoesophageal ultrasound, for instance,
during percutaneous aortic valve replacement. Due to the relatively large diameter of
the ultrasound endoscope, the commercially available elbows (fig. 2) cannot be used.
However, with a surgical cutter, a vertical hole can be made in the soft part of the
interface, that allows leak free introduction of the endoscope. Very recently interfaces
have been introduced that allow introduction of large diameter endoscopes (fig. 4a).
Moreover, one type even allows application of NIV after the endoscope has been intro-
duced (fig. 4b). In general, some level of sedation will be required during the procedure.
We would prefer remifentanil due to its short half-life and favourable haemodynamic
characteristics. Close monitoring of vital signs is required in these high-risk patients.
Prolonged endoscopic procedures in gastroenterology, such as endoscopic retrograde
cholangiopancreatography (ERCP), usually require the use of sedatives. This may
increase the risk of pulmonary complications, in particular in patients with respira-
tory muscle weakness, COPD, obesity and heart failure. Despite the absence of high-
quality studies, the use of NIV during ERCP should be considered in these patients.
Although some experts have advocated the use of nasal interfaces for NIV with ERCP,
NIV and the ICU
a) b)
Connector for Sealed orifice

ventilator tubing for endoscope
Figure 4. a) Example of an interface with large orifice that allows the introduction of all
diameters of endoscope, including those used with transoesophageal ultrasound or ERCP
(VBM Endoscopy mask; VBM Medizintechnik, Sulz, Germany). b) Interface with a large
orifice that can be applied after oral introduction of the endoscope (Janus ventilation mask;
Biomedical Srl, Florence, Italy). Images used with permission of the suppliers.
the efficiency will be limited due to air leak with mouth opening. New types of inter-
faces allow introduction of the ERCP endoscope while using a full face mask (fig. 4).
The patient’s vital signs should be monitored closely during the procedure. Likewise,
NIV should be considered to facilitate percutaneous endoscopic gastrostomy in high-
risk patients, in particular patients with severe respiratory muscle weakness such as
ALS. Although the use of nasal masks has been advocated, the new full face masks
(fig. 4) are probably superior, due to reduced air leak. However, as clinical experience is
limited this should be performed only by professionals with considerable experience
in both NIV and endoscopy, and with close monitoring of vital signs.
Further reading
• American Society of Anesthesiologists Task Force on Sedation and Analgesia by
Non-Anesthesiologists (2002). Practice guidelines for sedation and analgesia by
non-anesthesiologists. Anesthesiology; 96: 1004–1017.
• Ambrosino N, et al. (2011). Unusual applications of noninvasive ventilation. Eur
Respir J; 38: 440–449.
• Cabrini L, et al. (2013). Non-invasive ventilation during upper endoscopies in adult
patients. A systematic review. Minerva Anestesiol; 79: 683–694.
• Cracco C, et al. (2013). Safety of performing fiberoptic bronchoscopy in critically ill
hypoxemic patients with acute respiratory failure. Intensive Care Med; 39: 45–52.
• Esquinas A, et al. (2013). Broncoscopia durante la ventilación mecánica no invasiva:
revisión de técnicas y procedimientos [Bronchoscopy during non-invasive mechanical
ventilation: a review of techniques and procedures]. Arch Bronconeumol; 49: 105–112.
• Heunks LM, et al. (2010). Non-invasive mechanical ventilation for diagnostic bron-
choscopy using a new face mask: an observational feasibility study. Intensive Care
Med; 36: 143–147.

NIV and the ICU
NIV in respiratory pandemics
Anita K. Simonds
The definition of a pandemic is “an epidemic occurring worldwide, or over a very

wide area, crossing international boundaries and usually affecting a large number
of people”. Pandemics tend to be viral in nature, the most important recent patho
gens being influenza virus, severe acute respiratory syndrome (SARS) coronavirus
(CoV) and Middle Eastern respiratory syndrome (MERS)-CoV. Clearly, new patho-
gens may arise at any time with rapid spreading potential.
The cornerstones of the development of a pandemic are that:
• the causative organism emerges in humans
• there is minimal or no population immunity
• it spreads easily from person to person
The latter prerequisite is met by airborne and droplet transmission as in influenza
virus, SARS-CoV and MERS-CoV, or transmission of a highly infectious pathogen
in bodily secretions, such as Ebola virus. There is likely to be crossover in that
additional spread may occur by direct contact or fomites (objects capable of trans-
mitting infection, such as clothing, furniture, door handles and toys). Important
factors that determine the impact of the organism are its transmissibility, as indi-
cated by the average reproduction number R (average number of people infected
by a single infectious person), and severity, as measured by case fatality rate. This
section will focus on airborne viruses, and the lessons learnt from respiratory
management in SARS and H1N1 influenza, and will give pointers for patient care
and infection control management in future outbreaks.
Key points
• NIV is unsuitable in rapidly progressive acute lung injury,
severe pneumonia and multiorgan failure in acute respiratory
pandemics.
• NIV may be helpful in early cases of acute respiratory infection
(e.g. influenza) to avoid the need for intubation and invasive
ventilation.
• NIV generates droplets that are an infection control risk. Full
personal protective equipment should be used by teams
experienced in both infection control and the application of NIV.

NIV and the ICU
Potential role of NIV in respiratory pandemic infections

Pneumonia and/or acute lung injury culminating in ARF are the most frequent
serious complications of respiratory pathogens. In a heterogeneous group of
patients with ARF treated with NIV, pneumonic consolidation was shown to be a
poor prognostic factor, and in community-acquired pneumonia, NIV failure may
be associated with higher mortality. However, subsequent studies and a meta-
analysis suggested NIV is reasonable as first-line therapy in subgroups of patients
with community-acquired pneumonia, particularly in those with COPD, although
should not prevent the rapid escalation to intubation and invasive ventilation in
those who are progressing rapidly or fail to respond. Furthermore, use of NIV in
a patient with extensive consolidation/bilateral whiteout, profound hypoxaemia
or multisystem failure is very unlikely to be successful and should be avoided.
A multicentre survey of NIV as a first-line intervention in ARDS in expert centres
showed NIV use avoided intubation in 54% of patients, with a Simplified Acute
Physiology Score II of >34 and an inability to improve FIO2/PaO2 after 1 h being key
predictors of failure. It should be noted that patients with more than two system
failures, haemodynamic instability and neurological disturbance were excluded
from NIV. These considerations raise the possibility that NIV could be used in
respiratory pandemics in limited situations shown in box 1.
Experience of NIV in SARS
SARS was first identified as being caused by coronavirus with high case severity
in 2003. NIV use in cases in Hong Kong, mainland China and Canada has been
reported predominantly in retrospective case series with NIV delivered by face
mask. Firm conclusions are therefore difficult to draw as milder patients or those
with single-organ failure were probably more likely to be allocated NIV and the
studies are small. Yam et al. (2005) showed early application of NIV resulted in
a reduction in mortality and need for endotracheal intubation, and Cheung et al.
(2004) found that in early ARDS cases, NIV reduced the need for intubation,
decreased ICU length of stay or enabled patients to avoid ICU admission. It should
be noted, however, that in a Canadian series, 19% of patients became critically ill
and for those requiring mechanical ventilation (invasive ventilation), mortality was
45%, with higher risk of death in patients >65 years of age or with comorbidities.
Experience of NIV in H1N1 influenza
Some of the lessons learnt from SARS were applied in the H1N1 influenza out-
break in 2009–2010. Here, up to 30% of hospitalised patients required admission
to an ICU, with rapid and progressive respiratory failure. Although guidelines from
most authorities, including the World Health Organization (WHO), US Centers for
Disease Control and Prevention, and European Centre for Disease Prevention and
Control, did not recommend NIV as first-line therapy, its role was explored in >20
published studies, which are summarised in a recent review (Esquinas et al., 2014).
Box 1. Potential uses of NIV

• Early use of NIV may reduce the need for intubation
• NIV may be used to wean patients from invasive ventilation
• NIV may be used for patients in whom it is a ceiling of ventilatory care
and/or to palliate symptoms

NIV and the ICU
The studies are disparate, include variable numbers of patients and only a small
proportion were prospective. NIV failure rate ranged from 0% to 100% and mortality
from 0% to 50%. The risk of NIV failure was increased in patients with refractory
hypoxaemia and multiorgan failure. The most favourable outcome from NIV was
seen in patients with:
• low initial Acute Physiology and Chronic Health Evaluation (APACHE) and
Sequential Organ Failure Assessment scores
• no requirement for vasopressor
• fewer radiological quadrants affected by consolidation
In a Canadian group, 82% of patients were mechanically ventilated on the first
day of ICU admission and of these, 33% received NIV. A high proportion of the
NIV recipients (85%) progressed to invasive ventilation, again suggesting that in a
rapidly deteriorating, severely ill patient group, invasive ventilation is the ventila-
tory management of choice. There are no systematic studies of the use of NIV to
wean patients with H1N1 from invasive ventilation, but logic would suggest it to
be of value here, particularly in those with comorbidities such as heart failure,
neuromuscular disease, OHS and COPD.
Pregnancy and H1N1 There are case reports of successful NIV use in pregnant
patients with H1N1, and Zhang et al. (2012) reported a case series on almost 400
pregnant patients with H1N1. Just over 60% of women required critical care and
47% mechanical ventilation. Of the 186 who were treated with mechanical ventila-
tion, 83 received NIV and in 38, this was successful. Mortality was lower in those who
initially received NIV compared with those who were initially intubated (p=0.006)
but on multivariate analysis, septic shock proved to be an important independent
prognostic factor for NIV failure. Higher APACHE score, evidence of liver damage and
central nervous system symptoms were associated with NIV failure compared with
those with successful NIV outcome on univariate analysis. It should be noted that
pregnant patients are a high-risk group overall and extreme caution is urged.
NIV in MERS-CoV and novel coronaviruses

There are a few anecdotal reports of CPAP/NIV use in patients with ARF due to
MERS-CoV where mostly it failed. The WHO has produced interim guidance for
management of novel coronaviruses (see Further reading), which is summarised
below.
• Recognise severe cases when severe respiratory distress may not be sufficiently
treated by oxygen alone, even when delivered at a high flow rate (intractable
hypoxaemia and high work of breathing).
• Mechanical ventilation should be instituted early in patients with increased
work of breathing or persistent hypoxaemia despite oxygen therapy.
• Consider NIV if local expertise is available, when immunosuppression is pre-
sent, or in mild ARDS without impaired consciousness or cardiovascular failure.
• Use a low-volume, low-pressure lung protection strategy with a target VT of
6 mL⋅kg−1, plateau airway pressure <30 cmH2O and SaO2 88–93%. Use ad
equate PEEP to reach the SaO2 target and allow permissive hypercapnia.
• Consider adjuncts in patients with severe ARDS: neuromuscular blockade,
prone positioning and possible extracorporeal membrane oxygenation.
• Employ conservative fluid management in ARDS patients without shock.
NIV and the ICU
Infection control risk and NIV

In the Toronto (Canada) and Singapore SARS outbreaks, healthcare workers com-
prised a significant proportion of the cases infected and ~20% of those admitted
to critical care. An analysis of these events showed that those at greater risk were
physicians and nurses performing intubation, and nurses caring for SARS patients.
Nurses caring for patients receiving NIV were at increased risk (relative risk 2.23,
95% CI 0.25–21.76) but this did not reach significance (p=0.5). Infection tended
to occur within 48 h of admission – perhaps before a diagnosis was made or effec-
tive personal protective equipment (PPE) was introduced – but was not related
to the level of training of physicians, or multiple or difficult procedures. Evidence
of droplet and direct contact spread came from an investigation of a nosocomial
outbreak of SARS involving 128 cases among healthcare workers, patients and
contacts in Toronto, where important risks were close proximity to the infected
patient and movement of staff between patients. An examination of the risk of
super-spreading cases (i.e. the development of three new cases within 2–10 days
of the admission of an index patient or a cluster of more than three new cases aris-
ing when there is no known index case) in Guangzhou (mainland China) and Hong
Kong, showed the following factors were key.
• Distance between beds of ≤1 m
• Lack of washing or changing facilities for staff
• Staff working while experiencing symptoms
• Resuscitation performed
Further factors related to the host were use of oxygen, nebuliser therapy and bilevel
NIV. Here, it is difficult to separate other confounding or compounding features,
such as the fact that sicker patients with higher viral loads are more likely to require
high-flow oxygen therapy and NIV compared with those with milder infections,
and patients requiring nebuliser therapy are likely to have asthma or COPD with
an exacerbation of cough causing further transmission risk by large droplet spread.
Aerosol or droplet transmission?
Coughs and sneezes can generate relatively large particles (>10 µm) that travel
short distances because of their higher mass, and contaminate the bedside or
local environment. Smaller droplets or aerosols will remain airborne for longer
periods and disseminate over greater distances. An aerosol is usually character-
ised by droplets of <5 µm. After SARS, some clinical procedures were designated
as aerosol-generating procedures. These included bronchoscopy, intubation of the
upper airway, invasive airway procedures such as suction, endotracheal intuba-
tion, resuscitation, and NIV. Certain other procedures, such as high-flow oxygen
and nebuliser therapy, were considered to be possible aerosol generators.
To understand the generation of droplets, Hui and co-workers (2006, 2009) car-
ried out a series of experiments analysing particle spread from NIV and oxygen
masks using smoke particles as a proxy for exhaled droplets from a high-fidelity
human simulator. However, smoke particles are much smaller (<1 µm) than re
spiratory droplets and human simulators may not closely reflect the behaviour of
sick patients.
In a study of human controls, those with flu-like symptoms and those with a
respiratory tract infections, NIV and physiotherapy were shown to produce large
NIV and the ICU
Practical management guide

Healthcare team Education in infection control is essential, including as
aspects of PPE. All staff applying NIV should be experienced in its use. Hand-
washing facilities and PPE should be fully available.
Equipment Bilevel ventilators are preferable, with use of a nonvented mask
with a high-efficiency filter (99.9997% filter efficiency) on the exhalation
port; a double-limb circuit may be advantageous. Oxygen should be entrained
via an FIO2 mixer or T-piece. Use an oronasal mask, total mask or helmet,
which should be secured in place before switching on the ventilator, and the
ventilator switched off before removing the interface or lifting it to provide
sips of fluid or mouth care.
Settings Initial low pressures (e.g. IPAP 10 cmH2O and EPAP 4 cmH2O) should
be employed to reduce the risk of barotrauma, with further titration to relieve
respiratory distress, and correct SpO2 and arterial blood gas tensions. CPAP
(e.g. 5–10 cmH2O) may be sufficient in previously fit patients but in those
with a high work of breathing, hypercapnia, or comorbidities such as OHS,
neuromuscular disease or COPD, NIV is the intervention of choice.
Decision-making In view of the points discussed earlier, patients failing to
respond to NIV should be transferred to invasive ventilation and so should be
managed in ward areas where this is possible (e.g. a high-dependency unit or
ICU). In patients in whom NIV is the ceiling of care, palliative care strategies
should be considered in advance and implemented. In such individuals, in
whom escalation to intensive care is not appropriate and NIV has been used
in attempt to palliate breathlessness, NIV should be withdrawn if no reduc-
tion in symptoms is seen and care refocused on symptom relief (e.g. opiates
and high-flow oxygen therapy).
Environment Negative-pressure rooms should be used if available but dis-
semination may be largely by large droplets and direct contact. There should
be a minimum distance of 1 m between beds in cohort-managed areas.
Health resource allocation The appropriate application of NIV may have useful
health resource implications in reducing the need for invasive ventilation in
large-scale pandemics.
droplets in infected patients (not controls), but not an aerosol. These droplets were
detected by an optical particle sizer immediately proximal to the NIV facemask but
had decreased significantly 1 m from bedside, in keeping with their large mass,
and were removed by placing a high-efficiency filter on the exhalation valve, i.e.
modifying the NIV circuit. High-flow oxygen therapy did not produce an aerosol or
droplets and the nebuliser produced an aerosol of bronchodilator but did not seem
to disseminate droplets from the patients. It is therefore accurate to designate NIV
as a droplet-generating procedure, not an aerosol-generating procedure.
What are the implications for infection control? The aforementioned results sug-
gest that full PPE should be used for NIV and physiotherapy. As the droplets are
large and many drop out within 1 m of the patient, the crucial importance of
NIV and the ICU
hand-washing and decontamination of nearby surfaces is evident, as transmis-

sion in these circumstances is likely to be from droplet spread or from fomites or
direct contact with the patient’s local environment. Modifying the circuit by filter-
ing the exhalation port reduces droplets, and sensible alternatives include using a
helmet with a separate expiratory limb, or a total or full face mask with a dual-limb
circuit. Nasal masks are not advisable because of mouth leaks, and a good mask fit
is essential to reduce peripheral mask leakage.
Multidrug-resistant TB
The principles discussed here are relevant to other highly infectious but nonpandemic
situations, such as management of cases of ARF due to pulmonary TB with NIV.
Further information on this topic is summarised in a review by Esquinas et al. (2014).
Further reading
• Ambrosino N, et al. (1995). Non-invasive mechanical ventilation in acute respira-
tory failure due to chronic obstructive pulmonary disease: correlates for success.
Thorax; 50: 755–757.
• Antonelli M, et al. (2007). A multiple-center survey on the use in clinical practice
of noninvasive ventilation as a first-line intervention for acute respiratory distress
syndrome. Crit Care Med; 35: 18–25.
• Centers for Disease Control and Prevention (2014). CDC Resources for Pandemic
Flu. www.cdc.gov/flu/pandemic-resources/index.htm
• Cheung TM, et al. (2004). Effectiveness of noninvasive positive pressure ventila-
tion in the treatment of acute respiratory failure in severe acute respiratory syn-
drome. Chest; 126: 845–850.
• Confalonieri M, et al. (1999). Acute respiratory failure in patients with severe com-
munity-acquired pneumonia. A propective randomized evaluation of noninvasive
ventilation. Am J Respir Crit Care Med; 160: 1585–1591.
• Esquinas AM, et al. (2014). Noninvasive mechanical ventilation in high-risk pul-
monary infections: a clinical review. Eur Respir Rev; 23: 427–438.
• European Centre for Disease Prevention and Control (2015). National
Pandemic Preparedness Plans. www.ecdc.europa.eu/en/healthtopics/pandemic_
preparedness/national_pandemic_preparedness_plans/Pages/national_
pandemic_preparedness_plans.aspx
• Fowler RA, et al. (2003). Critically ill patients with severe acute respiratory syn-
drome. JAMA; 290: 367–373.
• Fowler RA, et al. (2004). Transmission of severe acute respiratory syndrome during
intubation and mechanical ventilation. Am J Respir Crit Care Med; 169: 1198–1202.
• Hui DS, et al. (2006). Noninvasive positive-pressure ventilation: an experimental
model to assess air and particle dispersion. Chest; 130: 730–740.
• Hui DS, et al. (2009). Exhaled air dispersion distances during noninvasive ventila-
tion via different Respironics facemasks. Chest; 136: 998–1005.

NIV and the ICU
• Keenan SP, et al. (1997). Effect of noninvasive positive pressure ventilation on

mortality in patients admitted with acute respiratory failure: a meta-analaysis. Crit
Care Med; 25: 1685–1692.
• Kumar A, et al. (2010). Critically ill patients with 2009 influenza A(H1N1) infection
in Canada. JAMA; 302: 1872–1879.
• Simonds AK, et al. (2010). Evaluation of droplet dispersion during non-invasive
ventilation, oxygen therapy, nebuliser treatment and chest physiotherapy in clini-
cal practice: implications for the management of pandemic influenza and other
airborne infections. Health Technol Assess; 14: 131–172.
• Varia M, et al. (2003). Investigation of a nosocomial outbreak of severe acquired
respiratory syndrome (SARS) in Toronto, Canada. CMAJ; 169: 285–292.
• World Health Organization (2013). Interim Guidance Document. Clinical manage-
ment of severe acute respiratory infections when novel coronavirus is suspected:
what to do and what not to do. www.who.int/csr/disease/coronavirus_infections/
InterimGuidance_ClinicalManagement_NovelCoronavirus_11Feb13u.pdf
• Yam LY, et al. (2005). Non-invasive versus invasive mechanical ventilation for res-
piratory failure in severe acute respiratory syndrome. Chin Med J; 118: 1413–1421.
• Yu IT, et al. (2007). Why did outbreaks of severe acute respiratory syndrome occur
in some hospital wards but not in others? Clin Infect Dis; 44: 1017–1025.
• Zhang PJ, et al. (2012). Clinical features and risk factors for severe and critical preg-
nant women with 2009 pandemic H1N1 influenza infection in China. BMC Infect
Dis; 12: 29.

NIV and the ICU
Stepping up and down from NIV to

tracheostomy ventilation
Mark W. Elliott
Tracheostomy refers to the process by which a tube is inserted into the trachea
below the larynx. It can either be done as a surgical procedure or using a percutan
eous technique. Indications are shown in table 1.
For patients who have been ventilated invasively, a number of uncontrolled stud-
ies have suggested that early tracheostomy (within 2–10 days of intubation) is
advantageous. However, this has not been borne out in RCTs: mortality, duration
of mechanical ventilation, ICU and hospital stay, and the incidence of pneumonia
were no different in patients in whom tracheostomy was performed early (within
4–8 days of endotracheal intubation (ETI)) or late (after 14–16 days of ETI). For
patients who are likely to require ventilatory support in the long term (e.g. patients
with neuromuscular disease), the issue about whether to perform a tracheostomy is
particularly vexed, as once performed it may be difficult to decannulate the patient
and provide support with NIV. The ability of clinicians to predict which patients will
require prolonged ventilatory support is limited. The ability to clear secretions is
key in whether decannulation can be achieved; in a study by Bach et al. (1996),
a cough peak flow of <160 L⋅min−1 was associated with failure of removal of the
tracheostomy tube. However, this observation was not based upon controlled
trial data and was before the widespread availability of mechanical cough assist
devices. The possibility that it may not be possible to transfer the patient from
invasive ventilation to NIV needs to be balanced against the risks associated with
reintubation after failed extubation. In general, these RCTs support a policy of not
Key points
• Do not rush to perform tracheostomy in the endotracheally
intubated patient.
• There are a number of important choices to be made with
regard to tracheostomy tubes.
• Uncuffed dual cannula tubes are recommended for
long-term use.
• Respiratory distress in a tracheostomised patient should
be assumed to be due to the tracheostomy until proven
otherwise.

NIV and the ICU
Table 1. Indications for tracheostomy
Constant and fixed large airway obstruction

Malignancy
Post surgery
Subglottic stenosis associated with prolonged translaryngeal intubation
Intermittent and dynamic large airway obstruction
Laryngeal oedema
Various neurological conditions including:
Multisystem atrophy
Motor neurone disease
Stiff person syndrome
To facilitate secretion management in patients choking on excessive
oropharyngeal secretions in bulbar motor neurone disease or severe
neuromuscular disease
As part of weaning from prolonged invasive mechanical ventilation
For patients managed with NIV who either cannot be ventilated effectively
or are spending most of the day on ventilation
rushing to tracheostomy and trying to optimise respiratory function, in particular

trying to minimise the secretion burden, prior to extubation.
The most common reason for patients escalating from NIV to tracheostomy ven-
tilation is excessive upper airway secretions interfering with the delivery of NIV.
Increasingly, with modern interfaces patients can sustain NIV for 24 h per day,
sometimes supplemented with mouthpiece ventilation. NIV and tracheostomy
ventilation have been reported to be comparable in terms of survival, but there
are fewer complications with the noninvasive approach. Generally, tracheostomy
ventilation, as an escalation from NIV, can be considered when overnight ventila-
tion needs to be supplemented by significant periods of NIV during the day or
NIV tolerance/effectiveness is poor, e.g. in patients with neuromuscular disease.
Although it has been used in patients with chronic lung disease, this seems a less
common indication now, since the advent of NIV.
Choices
Once it has been decided that a tracheostomy is appropriate, there are a number
of choices to be made. These are listed in table 2.
Increasingly, for tracheostomies performed on patients in the ICU recovering
from an acute episode, the percutaneous technique is preferred. The major
advantage is that the procedure can be performed on the unit by the ICU staff,
without having to move the patient to the operating theatre or involve sur-
geons. Complication rates are generally lower with the percutaneous technique
(Freeman et al., 2000). For elective tracheostomies, the choice will depend upon
local factors. Generally speaking, the patient will require a short ICU stay after the
NIV and the ICU
Table 2. Choices and issues in tracheostomy care
Percutaneous versus surgical

Tube
Size: internal and external diameter
Length
Angled or curved
Uncuffed versus cuffed
Type of cuff
Single tube versus tube with removable inner tube
Fenestrated versus nonfenestrated
Subglottic suction port?
Speech during ventilation
Swallowing and aspiration
procedure and it may therefore be easiest to use the percutaneous technique, for
the above-mentioned reasons.
The dimensions of tracheostomy tubes are given by their inner and outer diam-
eters. The length and curvature are important in ensuring effective and comfort-
able fit. Occasionally, a tube with an adjustable flange is necessary in patients with
a very thick neck or unusual anatomy. Differences in length between tubes of the
same inner diameter but from different manufacturers are not commonly appre-
ciated, but may have important clinical implications. The fact that tracheostomy
tubes can be angled or curved can be used to improve the fit of the tube in the
trachea.
In the longer term, uncuffed tubes are better tolerated and associated with
fewer complications and better survival. However, initially a cuffed tube should
be placed, as this will allow more effective ventilation, which is important if the
patient becomes clinically unstable, particularly with oxygen desaturation, requir-
ing a period of “bagging”. Cuffs on tracheostomy tubes include high-volume low-
pressure cuffs, tight-to-shaft cuffs and foam cuffs. With an uncuffed tube, the
patient may find the sensation of air leaking through the upper airway uncom-
fortable and ventilator triggering into inspiration may be compromised. The cuff
should be deflated at the earliest opportunity during the day, in order to promote
audible speech (patients sometimes require prompting to use their voice, particu-
larly if they have been ventilated invasively for a significant period of time) and to
establish whether ventilation can be delivered effectively and comfortably with an
uncuffed tube. This should be done first during the day and then during ventilation
overnight. Once confirmed that ventilation can be delivered effectively, the tube
can subsequently be replaced with an uncuffed tube.
A dual cannula tracheostomy with a removable inner tube is advisable, particularly
for long-term ventilation, as cleaning of the inside of the tube and blockages due
to excessive secretions are easily dealt with. As a general rule, if there is a problem
NIV and the ICU
or the patient is uncomfortable, the inner tube should be removed and checked
for patency. Fenestrated tubes have the theoretical advantages of less resistance
during unassisted ventilation and of promoting speech, but in practice are seldom
required unless the tube fit is very tight. They are more likely to cause irritation to
the trachea and formation of granulation tissue. If a fenestrated tube is required,
a “pepperpot” type is recommended. If upper airway secretions are a problem, for
instance in patients with ALS, a tube with a subglottic suction cannula is useful,
as this allows intermittent aspiration of secretions that have accumulated above
the cuff.
Speech and safe swallowing are two important issues to be addressed. A number
of different sorts of speaking valves are available, but it is important to appreci-
ate that these may have an adverse physiological effect. For patients receiving
continuous ventilation, bilevel ventilation provides better speech duration than
assist-control ventilation. Tracheostomy is traditionally believed to compromise
swallowing by direct pressure on the oesophagus and by preventing laryngeal
elevation, an important part of the mechanism of protecting against aspiration.
However, in tachypnoeic individuals, effective ventilation may actually improve
swallowing efficiency. The possibility of aspiration should constantly be borne in
mind in tracheostomised patients, particularly if there are frequent respiratory
tract infections.
Weaning and decannulation

When a tracheostomy has been placed to facilitate weaning following ETI, a deci-
sion will need to be made as to whether it will be possible to remove the trache-
ostomy, with ventilatory support delivered noninvasively if needed, or whether the
patient can manage completely unassisted ventilation. There are no clinical trial
data to guide practice and no right or wrong ways to do this. In our respiratory
care unit, we usually abandon all weaning attempts for the first 24 h following
transfer from the ICU, unless the patient is already sustaining prolonged periods
of ventilator-free time. The focus is upon ensuring stability, developing a relation-
ship with the patient, ensuring that secretions are under control and checking that
aspiration is not occurring if the patient is eating or drinking. The latter is checked
using a simple bedside “dye” test: the patient is given some water containing food
dye, then if discoloured secretions are subsequently aspirated from the trachea,
aspiration is occurring and the patient should be “nil by mouth”. If the dye test
is passed, aspiration should still be borne in mind, with retesting if there is any
coughing during eating or drinking. Formal swallowing assessment is required in
more complex cases.
The patient is encouraged to have short periods off the ventilator, e.g. 5 min, four
times per day in the first instance, with monitoring of patient respiratory rate,
oxygen saturation and a simple assessment of respiratory distress before and
at the end of the period of ventilator-free time. See figure 1 for an example of a
weaning chart.
Depending on how this goes, the periods of ventilator-free time can be length-
ened, with the rate of progress determined by the clinical team. The aim should
be to wean the patient completely during the day but continue ventilatory support
overnight. Once it is established that the patient can sustain unassisted ventilation
NIV and the ICU
Date Days post ICU admission Days weaning
Ventilation Aim Time off Time on RR SaO 2 Comfort start RR SaO 2 Comfort end
Minutes off vent / trachy cap
Nursing Variance (including from ventilation plan)
Physio Variance
Dye test Monday Thursday Dietician review Monday Thursday

Breathlessness comfort score 0 = no visible respiratory distress 1 = mild distress 1 = severe distress
Figure 1. Tracheostomy weaning chart. RR: respiratory rate; vent: ventilator; trachy:
tracheostomy.
during the day, consideration can be given to withdrawal of ventilatory support

overnight. At the same time, periods with the tracheostomy tube capped off, so
that the patient is breathing normally through the upper airway as they would if
the tracheostomy tube was removed, are introduced, with monitoring before and
after as described. The aim is for the patient to be able to sustain adequate ventila-
tion for 24 h with the tube capped off. Once this is achieved, the tracheostomy can
be removed with confidence.
Sometimes the patient will require ongoing NIV overnight. NIV can be intro-
duced during the day when the tube is capped off and then similarly overnight.
Alternatively, if it is apparent that the patient tolerates NIV well and can sustain
reasonable periods of ventilator-free time, the tracheostomy can be removed,
with NIV delivered overnight and as required during the day. It should be stressed
that every patient is different and the approach needs to be tailored accordingly.
Respiratory distress should be assumed to be due to the tracheostomy until
proven otherwise. Tube patency should be confirmed and there should be a low
threshold for bronchoscopic inspection of the trachea.
Alternatives to tracheostomy
Mouthpiece ventilation has been described as an alternative for patients with
progressive neuromuscular disease requiring assisted ventilation during the day
as well as overnight, and this too can be used as part of the weaning process.
Further information about mouthpiece ventilation can be found in the section
“Practicalities of and guide to cough augmentation and daytime mouthpiece ven-
tilation”. Button tracheostomies may also be useful for some patients, as there is
less resistance to airflow through the trachea.
Complications
These may occur early, soon after tracheostomy has been placed, or late. They are
summarised in table 3.
Improving care for patients in hospital with a tracheostomy
The UK National Confidential Enquiry into Patient Outcome and Death has
made a number of key recommendations on improving tracheostomy care.
NIV and the ICU
Table 3. Complications of tracheostomies
Early (at the time of or soon after placement of the tracheostomy tube)
Minor and major bleeding
Pneumothorax
Accidental decannulation
Subcutaneous emphysema
Infection of the stoma
False placement
Tube obstruction due to mucus
Late
Tracheal stenosis
Granuloma
Vascular erosion
Tracheomalacia
Tracheo-oesophageal fistula
Tracheo-arterial fistula
Aspiration and pulmonary infection
Tube obstruction due to mucus
These have been previously discussed by Zhu et al. (2014), and include the
following points:
• Training for bedside staff should include routine care as well as resuscitation
procedures for tracheostomy patients. This should be supported by hospital-
wide guidance for tracheostomy care. Information about the type of tube, as
well as essential equipment, should be clearly available at the bedside.
• Multidisciplinary care pathways, which provide continuity of care between crit
ical care and ward clinicians, and facilitate decannulation and discharge plan-
ning, need to be established for all tracheostomy patients.
• Bedside staff caring for tracheostomy patients must be competent to recognise
and manage common airway complications.
• Unplanned and/or night-time discharge of a patient with a tracheostomy is
not recommended, particularly in patients with newly formed tracheostomy or
those recently weaned from respiratory support.
Routine chronic care

Cuff inflation pressure should be determined using a pressure gauge. In addition,
the volume of air instilled should be checked. It is possible with very long-term
use to gradually dilate the trachea, even with correct cuff pressures. The inner
tracheostomy tube should be cleaned regularly, the frequency determined by how
soiled it gets. There are no hard and fast rules as to how often the tracheostomy
NIV and the ICU
should be changed, where this should be done and by whom. Generally speaking,
monthly tube changes are adequate and in our unit these are done initially by spe-
cialist nurses, who when possible train carers to undertake this task in the home,
initially under supervision and then unsupervised once competence is confirmed.
In other units, patients attend hospital for the tube change, which is performed
by a doctor. A spare tracheostomy tube should always be readily available in case
of problems and the need to change or reinsert a tracheostomy tube urgently. It is
sensible to include a smaller tube in case there are difficulties in tube placement.
Carer burden
The presence of a tracheostomy can increase the strain on care givers and this
should be factored into decision making and care planning. It is not clear how
much this is due to the tracheostomy itself and how much due to the fact that a
tracheostomy is a marker for a high level of dependency and presence of complex
additional factors.
Further reading
• Bach JR, et al. (1996). Criteria for extubation and tracheostomy tube removal for
patients with ventilatory failure. A different approach to weaning. Chest; 110:
1566–1571.
• Benditt JO (2006). Full-time noninvasive ventilation: possible and desirable. Respir
Care; 51: 1005–1012.
• Blot F, et al. (2008). Early tracheotomy versus prolonged endotracheal intubation in
unselected severely ill ICU patients. Intensive Care Med; 34: 1779–1787.
• Freeman BD, et al. (2000). A meta-analysis of prospective trials comparing percuta-
neous and surgical tracheostomy in critically ill patients. Chest; 118: 1412–1418.
• Hess DR (2005). Tracheostomy tubes and related appliances. Respir Care; 50:
497–510.
• McKim DA, et al. (2013). Twenty-four hour noninvasive ventilation in Duchenne
muscular dystrophy: a safe alternative to tracheostomy. Can Respir J; 20: e5–e9.
• Prigent H, et al. (2003). Comparative effects of two ventilatory modes on speech in
tracheostomized patients with neuromuscular disease. Am J Respir Crit Care Med;
167: 114–119.
• Prigent H, et al. (2006). Characteristics of tracheostomy phonation valves. Eur
Respir J; 27: 992–996.
• Rossi Ferrario S, et al. (2001). Caregiver strain associated with tracheostomy in
chronic respiratory failure. Chest; 119: 1498–1502.
• Rumbak MJ, et al. (2004). A prospective, randomized, study comparing early per-
cutaneous dilational tracheotomy to prolonged translaryngeal intubation (delayed
tracheotomy) in critically ill medical patients. Crit Care Med; 32: 1689–1694.
• Soudon P, et al. (2008). A comparison of invasive versus noninvasive full-time
mechanical ventilation in Duchenne muscular dystrophy. Chron Respir Dis; 5: 87–93.

NIV and the ICU
• Terragni PP, et al. (2010). Early vs late tracheotomy for prevention of pneumonia
in mechanically ventilated adult ICU patients: a randomized controlled trial. JAMA;
303: 1483–1489.
• Terragni P, et al. (2012). Tracheostomy in mechanical ventilation. In: Ferrer M, et al.,
eds. New Developments in Mechanical Ventilation. ERS Monogr; 55: 206–216.
• Terzi N, et al. (2010). Impact of tracheostomy on swallowing performance in
Duchenne muscular dystrophy. Neuromuscul Disord; 20: 493–498.
• Toussaint M, et al. (2006). Diurnal ventilation via mouthpiece: survival in end-
stage Duchenne patients. Eur Respir J; 28: 549–555.
• Toussaint M, et al. (2007). Mechanical ventilation in Duchenne patients with
chronic respiratory insufficiency: clinical implications of 20 years published experi
ence. Chron Respir Dis; 4: 167–177.
• Trouillet JL, et al. (2011). Early percutaneous tracheotomy versus prolonged intub
ation of mechanically ventilated patients after cardiac surgery: a randomized trial.
Ann Intern Med; 154: 373–383.
• Young D, et al. (2013). Effect of early vs late tracheostomy placement on survival
in patients receiving mechanical ventilation: the TracMan randomized trial. JAMA;
309: 2121–2129.
• Zhu H, et al. (2014). Improving the quality of tracheostomy care. Breathe; 10:
286–294.

Long-term NIV
Chronic NIV in hereditary neuromuscular

disorders
Anita K. Simonds
As described in the introduction, long-term NIV has been used in neuromuscular

disorders for many decades. Following the use of NPV in the poliomyelitis epidem-
ics of the twentieth century, positive pressure mask ventilation was pioneered in
Duchenne muscular dystrophy patients in the 1980s and its use quickly spread
to other hereditary neuromuscular disorders. Improvements in mask design and
ventilator technology led to further widespread uptake, although there have
always been inequalities in the provision of NIV. Unlike COPD, there are few ran-
domised trials in neuromuscular disease, but many case series suggest a change
in outcome. General principles will be covered first and then detail provided for
the more common disorders.
Which patients are suitable and when should long-term NIV start?
Key considerations are to understand which patients are at high, moderate or low
risk of respiratory complications, and the likely natural history of the condition and
appropriate monitoring to assess progress. Table 1 shows the respiratory features
of a broad range of neuromuscular conditions to enable the risk of complications
to be assessed.
High-risk patients for respiratory decompensation in early childhood include
those with type 1 spinal muscular atrophy (SMA), X-linked myotubular myopathy,
infant-onset myotonic dystrophy and infantile nemaline myopathy. Respiratory
failure is inevitable in Duchenne muscular dystrophy, usually in the later teenage
years or early twenties, and type 2 SMA and congenital muscular dystrophy (CMD)
Key points
• NIV has increased survival in many inherited neuromuscular
conditions.
• NIV should be commenced when patients develop
symptomatic nocturnal hypoventilation.
• Cough augmentation with physiotherapy and cough assist
devices plus percutaneous gastrostomy feeding may reduce
the need for tracheostomy ventilation.
• New complications may arise in long-term NIV users.

164
Table 1. Identification of risk of respiratory complications
Long-term NIV
Condition Respiratory failure Secretion Recurrent Progression Disease-specific

clearance pneumonia features
difficulty
SMA
Type 1 All by 2 years Marked All Rapid All require full-time
respiratory support
Type 2 ∼40% in childhood Early ∼25% in first 5 years Slow
Type 3 Rare in childhood Rare in Rare in childhood Slow
childhood
SMA with respiratory All by 6 months Marked All Rapid in the first All require full-time
distress type 1 year, then respiratory support
(SMARD) slows
DMD/severe childhood After loss of ambulation After loss of Late Cardiomyopathy usually
onset limb-girdle ambulation occurs after respiratory
muscular dystrophy problems, but may
precede them
Facioscapulohumeral When onset <20 years With infantile With infantile onset Slow Severe infantile onset
muscular dystrophy onset type is frequently
associated with
sensorineural deafness
(Continued)

Table 1. Continued

difficulty
CMD
All types Any age depending on Any age Any age depending Slow
severity depending on severity
on severity
Ullrich 70% in adolescence Mild Infrequent Proximal contractures
with marked distal
laxity

Rigid spine muscular Early while ambulation Mild Infrequent Hypoventilation may
dystrophy preserved occur in ambulant
children with relatively
preserved vital capacity
Congenital myopathy
Central core Uncommon except in Uncommon Uncommon Slow Susceptible to malignant
severe recessive type hyperthermia
Minicore Early while ambulation
preserved
Nemaline Early in severe neonatal In severe form In severe form Slow
form, mild later onset form
may develop early while
ambulation preserved
Myotubular 85% in severe X-linked In severe form In severe form Slow Ophthalmoplegia, rare
form coagulopathy and liver
haemorrhage
Fibre type disproportion Depends on genotype Uncommon Uncommon
Long-term NIV
(Continued)
165
166
Table 1. Continued

Long-term NIV
difficulty
Myotonic dystrophy
Myotonic dystrophy 1 Common in severe Common Common in severe Initial Prominent learning
congenital onset, in severe congenital onset improvement, difficulty, somnolence,
usually improves congenital later slow central hypoventilation
onset deterioration
Myotonic dystrophy 2 Uncommon Uncommon Uncommon
Congenital myasthenic Often in neonatal period, Especially Possible if weakness Weakness may fluctuate,
syndromes may occur during during is severe and episodic apnoea in
intercurrent illnesses intercurrent persistent some Congenital stridor
illnesses in those with DOK7
mutations
Mitochondrial myopathy Common Possible Possible Acute
deterioration
possible
Charcot–Marie–Tooth With severe early onset, With severe With severe early Stridor, especially with
especially with GDAP1 early onset onset GDAP1 mutation
mutation
Pompe Infantile onset, may be Infantile onset Infantile onset Infantile rapid, Variable relationship
early in later onset while late onset between motor and
ambulation preserved slow respiratory progression
DMD: Duchenne muscular dystrophy. Reproduced from Hull et al. (2012) with permission from the publisher.

Long-term NIV
(Ullrich) patients may develop respiratory failure in childhood. Conditions which

are highly variable include Pompe disease, nemaline myopathy, limb girdle mus-
cular dystrophy (LGMD) and other myopathies. Most patients progress through a
sequence of nocturnal hypoventilation to diurnal ventilatory failure; the rate of
progression will depend on the natural history of the disorder, resulting respira-
tory and bulbar muscle weakness, and other factors such as scoliosis and general
nutrition.
Monitoring Assessment should be by clinical examination, history-taking with

reference to respiratory/sleep-related symptoms and chest infections, and
where possible, monitoring of pulmonary function (vital capacity and cough peak
flow). Noninvasive measurement of respiratory muscle strength, e.g. mouth
inspiratory and expiratory pressures and sniff inspiratory pressure, may be help-
ful, but in some cases is not more specific or sensitive in predicting decline
than simple vital capacity measurement. Overnight monitoring of respiration,
including measurements of SpO2 and PtcCO2, should be carried out in those with
sleep-related symptoms, a vital capacity of <70% predicted, failure to thrive or
recurrent chest infections. A number of consensus recommendations cover this
area of sequential monitoring (Finder et al., 2004; Birnkrant et al., 2007; Wang
et al., 2007).
Timing of the introduction of NIV NIV should be introduced at the time of symp-

tomatic nocturnal hypoventilation. Waiting until the development of daytime
hypercapnia exposes the individual to uncontrolled and unpredictable decompen-
sation at the time of a chest infection or other intercurrent event. One RCT showed
that once patients with congenital neuromuscular disorders developed nocturnal
hypoventilation in the presence of daytime normocapnia, 70% developed ventila-
tory decompensation within 12 months and 90% within 24 months (Ward et al.,
2005). Patients with daytime hypercapnia will inevitably have worse gas exchange
during sleep.
Clearly not all patients will benefit from ventilatory support, e.g. infants with
rofound type 1 SMA, and the goals of ventilatory support should always be
p
considered and agreed before initiation.
Outcomes
Duchenne muscular dystrophy Without ventilatory support, the outcome in
Duchenne muscular dystrophy is poor, with a FVC of <1 L, 5-year survival is 8%. A
RCT in patients with a FVC of between 20 and 50% predicted showed no survival
benefit or preservation of lung volumes when NIV was used preventively (Raphael
et al., 1994), but once nocturnal hypoventilation develops, the use of NIV extends
survival considerably with median survival increased into the late twenties (Bushby
et al., 2005) and many patients surviving into their thirties and forties, compared
with a median survival of ∼18–20 years without ventilatory support. Figure 1
shows survival in cohorts using tracheostomy ventilation and NIV. Undoubtedly,
other improvements occurred during this period, including better general medical
care, influenza and pneumococcal vaccination, physiotherapy, scoliosis surgery and
management of nutrition and cardiomyopathy, but the biggest impact was NIV. NIV
may cause a plateau in the progression of muscle weakness or slow the decline,

Long-term NIV
100
80
60
Survival %
40
20
0
15 20 25 30 35 40 45
Age at death years
NIV (from 1991)

Tracheostomy (1984–1991)
No ventilation, death after 1991
No ventilation, death before 1984
No ventilation, death between 1984 and 1991
Figure 1. Duchenne muscular dystrophy survival by intervention. Reproduced and modified
from Ishikawa et al. (2011) with permission from the publisher.
but patients with progressive disorders still become more ventilator dependent
over time and may benefit from:
• Daytime mouthpiece ventilation
• Use of mechanical insufflation–exsufflation devices
• Percutaneous gastrostomy (PEG) feeding
Echocardiogram and ECG (cardiac surveillance) should be carried out at least yearly
and cardiomyopathy actively treated; as a guide, treatment should be begun once
left ventricular fractional shortening has fallen below 28%.
Mouthpiece ventilation (fig. 3 in the section entitled “Long-term NIV case histories”,
see also the section entitled “Practicalities of and guide to cough augmentation
and daytime mouthpiece ventilation”) is sometimes called “sip ventilation” as the
sipping action generates a greater degree of negative pressure then inspiration.
Here, the positioning of a mouthpiece near the head allows the individual to
receive intermittent ventilatory support as required during the day and removes
the need for a mask. Breath stacking can be achieved. Khirani et al. (2014) surveyed
mouthpiece ventilation users and found the majority were highly ventilator
dependent and used the technique to reduce dyspnoea and fatigue during the day
and to facilitate speech and feeding. These ventilator top-ups can be helpful in
extending the period of noninvasive ventilatory support before invasive ventilation
is required. Toussaint et al. (2006) showed that mouthpiece ventilation reduced
symptoms, stabilised vital capacity, improved daytime PtcCO2 control and increased
survival in patients with Duchenne muscular dystrophy. However, practical problems
Long-term NIV
are fairly common, with troublesome alarms from the ventilator, and ineffective
triggering or auto-triggering of the ventilator. The use of a small mouthpiece with
a filter to create backpressure, and minimising the low pressure alarm can reduce
these problems.
Progression to tracheostomy ventilation occurs variably. For example, it is common
in Scandinavia and France but less so in the USA and UK. The combination of NIV,
mechanical insufflation–exsufflation and PEG feeding may reduce the need for
invasive ventilation but, in those individuals requiring NIV for long periods in the
daytime or with swallowing problems, tracheostomy ventilation may be preferable.
Becker muscular dystrophy Like Duchenne muscular dystrophy, this condition
is caused by defects in the dystrophin gene; however, Becker patients typically
remain ambulant for longer, and the condition may follow a more benign course.
A dilated cardiomyopathy is often evident and requires management before
the loss of ambulation and respiratory decompensation. NIV is often needed in early
adulthood, and cardiac surveillance remains crucial throughout the patient’s life.
Limb girdle muscular dystrophy LGMD is characterised by proximal weakness, and
the diagnostic group contains a wide range of conditions in which the underlying
molecular defect is being clarified. Type 1 LGMDs are autosomal dominant and Type
2 are autosomal recessive in inheritance. LGMD1B is a laminopathy and variable in
age of onset. LGMDC is a caveolinopathy often with childhood onset. LGMD2A is
a calpainopathy, LGMD2B is a dysferlinopathy, and LGMDl is an anoctaminopathy
and a common variant. In LGMD2 types C, E and F, known as sarcoglycanopathies,
cardiomyopathy is common. Cardiac surveillance is also important in LGMD2l and
LGMD1B. Creatinine phosphokinase is raised in some subtypes more than others
and so can be useful in distinguishing conditions, although DNA analysis is the
gold standard diagnostic test. Long-term survival with NIV in patients with LGMD
is fully possible.
Congenital muscular dystrophies This term refers to a variety of muscular conditions
characterised by dystrophy, which are evident at birth or shortly after, e.g. due
to presence of contractures or hypotonia. The molecular defects associated with
CMDs affect proteins in the extracellular matrix such as laminin 211, integrin,
collagen 6 or α-dystroglycan. Variants can be associated with cardiomyopathy,
learning disability or rigid spine syndrome and respiratory failure with a variable
age of onset. For myopathic CMDs that affect collagen 6, which include Ullrich
muscular dystrophy, a recent analysis by genetic subtype has clarified the time at
which respiratory compromise is likely (table 1).
In Ullrich muscular dystrophy, an early requirement for NIV in childhood is common
with an additional need for supplemental feeding because of swallowing difficulties,
and the child is unlikely to walk. Whereas in those with intermediate collagen VI
myopathic CMD respiratory failure occurs in later childhood or adulthood, and in
Bethlem myopathy subtype respiratory failure may not occur at all.
Spinal muscular atrophy Classically, SMA is divided into type 1, type 2 and type 3
on functional grounds.
• Type 1 patients never achieve sitting
• Type 2 patients are unable to walk
• Type 3 patients are able to walk, but might lose that capability later in life
Long-term NIV
Type 1 patients inevitable develop ventilatory failure in early infancy, type 2 usually
do so in childhood, and type 3 patients may never develop respiratory problems or
a small subgroup may experience respiratory complications in adulthood. Unlike
Duchenne muscular dystrophy and many other muscular dystrophies and myo-
pathies in which inspiratory and expiratory muscles weakness tends to progress
in tandem, individuals with SMA may have more pronounced expiratory muscle
weakness leading to weak cough and recurrent chest infection before the onset of
ventilatory failure. Cough assistance approaches are therefore particularly valuable
in SMA, and may be indicated before NIV is required. In some children, provision of
home NIV for physiotherapy purposes at the time of a chest infection has proved
helpful in preventing hospital admissions. In type 1 SMA outcome has improved
with the use of ventilatory support and nutritional supplementation. As discussed
elsewhere, the goals of therapy should be preset, and decisions made on an indi-
vidual basis.
Type 2 SMA patients often respond well to NIV and are less likely to develop bulbar
problems. Many of these children are now successfully transitioning to adulthood
(see the section entitled “Chronic NIV in children: indications, outcomes and tran-
sition”) with good medium- or longer-term prognosis. This will impact on adult
NIV service provision.
Congenital myopathies The features of central core, minicore and nemaline and

myotubular myopathies are shown in table 1. X-linked myotubular myopathy is
associated with severe weakness and with respiratory failure in early infancy. It
is noteworthy that central core patients are at risk of malignant hyperpyrexia.
Both nemaline myopathy and Pompe disease may be very variable, presenting in
an infantile form or in adulthood with diaphragm weakness. Alglucosidase-alfa
enzyme replacement therapy is available for patients with Pompe disease, but
does not usually avert the need for NIV. Cardiac involvement should be screened
for, but long-term survival with respiratory support is possible. Outcome data from
large studies are not available.
Myotonic dystrophy Patients with myotonic dystrophy develop hypercapnia as

a result of a combination of respiratory muscle weakness and decreased ven-
tilatory drive. They may also experience daytime somnolence as a feature of
myotonic dystrophy, and/or this may be as a consequence of sleep disordered
breathing. The balance of these contributing factors will vary between patients
and needs to be carefully teased out by measurement of pulmonary function,
respiratory muscle strength and a sleep study including monitoring of carbon
dioxide tension. Hypercapnia disproportionate to respiratory muscle weakness
indicates a more marked impact on control of breathing. Initiation of NIV for
sleep disordered breathing may be problematic if it is not symptomatic due to
a control of breathing disorder, or if the sleepiness is intrinsic due myotonic
dystrophy rather than sleep fragmentation. Management in these situations
has to be on an individual basis. Some patients with mild hypercapnia due pre-
dominantly to a control of breathing disorder may remain stable from some
time without ventilatory support, but NIV should be initiated wherever possible
if hypercapnia worsens. Cardiac surveillance is vital in view of the associated
cardiomyopathy and propensity to arrhythmias. Modafinil has been used for
intrinsic sleepiness.
Long-term NIV
Ventilator choices
Bilevel positive pressure ventilation is the commonest mode used long term in
neuromuscular patients, although volume-preset ventilators can be successfully
applied, and were the norm for this population before the introduction of bilevel
devices. Both have advantages. In bilevel modes, EPAP can be helpful in stabilising the
upper airway, and either reducing a tendency towards atelectasis or opening up atel
ectatic areas of lung. Volume-preset ventilators can be used to breath-stack, thereby
improving cough and increasing periods of spontaneous breathing. The combina-
tion of pressure support and volume assurance has been applied in neuromuscular
and chest wall disease patients. To date, studies have shown similar arterial blood
gas control overnight, but a crossover trial comparing standard bilevel pressure
support with IVAPS demonstrated overnight adherence was improved using IVAPS,
possibly indicating better patient–ventilator synchrony (Kelly et al., 2014). Volume-
assured bilevel pressure support could, therefore, be considered in patients in whom
standard pressure support proves suboptimal or is poorly tolerated.
Interfaces Interface choice is discussed in the section entitled “Choosing the
interface”. Particular considerations for long-term ventilation in neuromuscular
patients are the ease of taking the interface on and off, and the avoidance of acute
and chronic pressure effects in those who cannot easily remove or adjust the pos
ition of the mask themselves.
Ventilation via a mouthpiece is an option especially for daytime use and is
described above.
Mid-facial hypoplasia is a consequence of long-term mask use in children or young
adults who experience facial muscle weakness before the growth of the cranio
facial skeleton is complete. It should be avoided as far as possible by reducing
Box 1. Approach to surgery in patients on long-term NIV

• Preoperative assessment of pulmonary and cardiac function
• Optimisation of ventilation
• Discussion of the balance of risk of procedure versus risk of not carrying
out procedure
• Anaesthetic assessment of potential problems:
– Difficult intubation
– Scoliosis affecting epidural or spinal anaesthesia, or body position
during surgery
– Effect of anaesthetic drugs
– Risk of malignant hyperpyrexia, or hyperkalaemia
• Avoidance of uncontrolled sedation, with availability of ventilatory
support
• Avoidance uncontrolled oxygen therapy
• Patients may be extubated back onto NIV after the procedure, if they are
alert and mean SaO2 is normal on air (depending on nature of procedure)
• Cough assist devices and physiotherapy should be used proactively in the
post-operative period in all those with reduced cough efficiency
• Patients should be closely monitored in a high dependency unit or ICU
until they have returned to their usual level of ventilatory support

Long-term NIV
interface pressure on the face, designing masks to reduce pressure points and
disperse the pressure over wider areas, and rotating interfaces.
Management of intercurrent events/surgical procedures

Surgery Protocols and risk management guides for patients on long-term venti-
lation requiring intercurrent surgical procedures have been described (Birnkrant
et al., 2007; Chatwin et al., 2013). With detailed planning, such patients should
not be denied interventions that will aid their quality of life or even extend life,
and procedures such as PEG insertion can be safely carried out even in those with
unrecordable vital capacity. A multidisciplinary team familiar with the manage-
ment of these patients is essential (box 1).
The use of NIV during bronchoscopy is described in the section entitled “NIV for
endoscopic procedures”; surgical management of PEG placement is discussed by
Chatwin et al. (2013).
Pregnancy Most of the outcome data on pregnancy in patients with neuromuscu-
lar disorders on long-term ventilation come from case series or anecdotal reports,
but even from these it is clear that the use of NIV means that pregnancy has
become possible in some individuals in whom previously the risk was considered
too great. The physiological effects of pregnancy include:
• Increased ventilatory drive resulting in a higher work of breathing and lower
carbon dioxide tension (if that work of breathing can be sustained)
• Increased intra-abdominal pressure and elevation of the diaphragm, causing
pulmonary restriction and altering respiratory mechanics and cough efficacy
• Upper airway changes increasing the possibility of OSA
Each patient considering pregnancy should discuss feasibility with the multidiscip
linary team including a geneticist, pulmonologist, cardiologist, anaesthetist and
obstetrician (box 2).
It has been suggested that pregnancy is inadvisable in patients with a vital capacity of
<1 L. However, we have seen successful pregnancies in those with a vital capacity
of 800 mL, and there are reports of those with lower respiratory reserve with good
outcomes, although the risk undoubtedly increases. Rapidly deteriorating neuro-
muscular disease, severe cardiac disease, pulmonary hypertension and poor ven-
tilatory control remain contraindications to pregnancy. Progression of a scoliosis
during pregnancy is unlikely if the scoliosis was stable pre-pregnancy. Patients
with a normal sleep study pre-pregnancy may develop nocturnal hypoventilation
Box 2. Pre-pregnancy considerations

• Genetic risk to the infant (genetic counselling is essential)
• Risk of cardiorespiratory complications in the mother
• Risk of progression of scoliosis in pregnancy
• Obstetric complications (e.g. pelvic deformity)
• Anaesthetic problems (e.g. difficult intubation, problems with siting
epidural catheter, poor tolerance of anaesthetic agents or risk of
malignant hyperpyrexia, post-operative ventilatory failure, impaired
cough, weaning problems)

Long-term NIV
or obstructive sleep apnoea/hypopnoea during pregnancy via the physiological

mechanisms described above, and there should be a low threshold to introduce
NIV, and provide cough assistance in those with a cough peak flow <270 L⋅min−1.
CPAP may be sufficient in those with OSA, but if significant respiratory muscle
weakness is present NIV is usually preferable. Our practice is to mimic normal
changes in arterial blood gas tensions during pregnancy when on NIV overnight
with settings optimised to maintain a normal SpO2 and carbon dioxide tension.
Usually delivery of the baby is carried at ∼36 weeks in NIV users, to reduce the
impact of further pulmonary restriction on the baby, and the patient extubated
back onto NIV after caesarean section. If NIV was only required during pregnancy,
a further sleep study a few months post partum can be used to assess the long-
term requirement for this.
New symptoms in long-term survivors on NIV and palliative care
While it is clear that NIV has altered the natural history and extended survival
in a wide range of neuromuscular conditions, many individuals with progressive
conditions will become more dependent on ventilatory support over time. In
addition, symptom management with good palliative care may be needed many
years before death to optimise quality of life. New complications may arise,
which were never evident before the advent of ventilatory support such as renal
and bladder stones, back, shoulder joint and hip pain, and episodes of bowel
obstruction, in part related to poor abdominal muscle strength. Cases of volvulus
in Duchenne muscular dystrophy have been reported. These require careful
symptom management and/or surgical intervention on occasion. Cardiac com-
plications may also become a more dominant feature in Becker muscular dys-
trophy, Duchenne muscular dystrophy and myotonic dystrophy, and autonomic
dysfunction including arrhythmias may arise in those with SMA and CMDs.
Cardiac surveillance with echocardiogram and ECG at least annually therefore
remains important.
Further reading
• Birnkrant DJ (2006). New challenges in the management of prolonged survivors of
pediatric neuromuscular diseases: a pulmonologists perspective. Pediatr Pulmon;
41: 1113–1117.
• Birnkrant DJ, et al. (2007). American College of Chest Physicians Consensus
Statement on the respiratory and related management of patients with Duchenne
muscular dystrophy undergoing anesthesia or sedation. Chest; 132: 1977–1986.
• Bushby K, et al. (2005). The multidisciplinary management of Duchenne muscuar
dystrophy. Current Paediatrics; 15: 292–300.
Child; 96: 426–432.
• Chatwin M, et al. (2013). Risk management protocol for gastrostomy and jeju-
nostomy insertion in ventilator dependent infants. Neuromuscul Disord; 23:
289–297.

Long-term NIV
• DMD Pathfinders. (2014). Ventilation and Duchenne: Frequently asked questions.

Date last updated: October 21, 2014. https://dmdpathfinders.files.wordpress.
com/2014/11/dmd-ventilation-faq-booklet-small.pdf
• Dubowitz V (1995). Disorders of the lower motor neurone: the spinal muscular
atrophies. In: Dubowitz V, ed. Muscle Disorders in Childhood. 2nd Edn. London,
W.B Saunders Company Ltd; pp. 325–367.
• Finder JD, et al. (2004). Respiratory care of the patient with Duchenne muscular
dystrophy: ATS consensus statement. Am J Respir Crit Care Med; 170: 456–465.
• Foley AR, et al. (2013). Natural history of pulmonary function in collagen VI-related
myopathies. Brain; 136: 3625–3633.
• Hull J, et al. (2012). British Thoracic Society guideline for the respiratory manage-
ment of children with neuromuscular weakness. Thorax; 67: Suppl. 1, i1–i40.
• Ishikawa Y, et al. (2011). Duchenne muscular dystrophy: survival by cardio-respir-
atory interventions. Neuromuscul Disord; 21: 47–51.
• Kelly JL, et al. (2014). Randomised trial of ‘intelligent’ autotitrating ventilation ver-
sus standard pressure support non-invasive ventilation: impact on adherence and
physiological outcomes. Respirology; 19: 596–603.
• Khirani S, et al. (2014). Evaluation of mouthpiece ventilation in neuromuscular
disease. Respir Care; 59: 1329–1337.
• Ragette R, et al. (2002). Patterns and predictors of sleep disordered breathing in
primary myopathies. Thorax; 57: 724–728.
• Raphael JC, et al. (1994). Randomised trial of preventive nasal ventilation in
Duchenne muscular dystrophy. Lancet; 343: 1600–1604.
• Shneerson JM. (2010). Non-invasive ventilation in pregnancy. In: Elliott M, et al.,
eds. Non-invasive Ventilation and Weaning: Principles and Practice. 1st Edn. London,
Edward Arnold; pp. 496–498.
• Simonds AK (2007). Home non-invasive ventilation in restrictive disorders and
stable neuromuscular disease. In: Simonds AK, ed. Non-Invasive Respiratory
Support: a Practical handbook. 3rd Edn. London, Hodder Arnold; pp. 175–192.
• Vianello A, et al. (1994). Long-term nasal intermittent positive pressure ventilation
in advanced Duchenne’s muscular dystrophy. Chest; 105: 445–448.
• Wang CH, et al. (2007). Consensus statement for standard of care in spinal mus-
cular atrophy. J Child Neurol; 22: 1027–1049.
• Ward SA, et al. (2005). Randomised controlled trial of non-invasive ventilation
(NIV) for nocturnal hypoventilation in neuromuscular and chest wall disease
patients with daytime normocapnia. Thorax; 60: 1019–1024.

Long-term NIV
Online resources
• Hare A, et al. Chronic neuromuscular disease module. Skills-based Simulator
Training in Non-Invasive Ventilation. www.ers-education.org/e-learning/simulators.
aspx
• Hare N (2014). Secretion management in neuromuscular patients. ERS Course

Long-term NIV
Chronic NIV in motor neurone

disease/ALS
Joan Escarrabill
ALS is a neurological disease that can affect inspiratory and expiratory respira-
tory muscle strength. Respiratory muscle assessment is difficult, as these patients
may have different aptitudes for performing investigations requiring volitional
movements, particularly among bulbar patients. Furthermore, despite providing
prognostic information, the values recorded cannot predict the rate of decline of
respiratory muscle strength. Although respiratory support can improve symptoms
caused by the respiratory muscles, it does not affect the progression of ALS, which
continues inexorably and has previously led to therapeutic nihilism.
NIV improves the quality of life of ALS patients and increases survival, especially
in cases without bulbar involvement. There is uncertainty about the right time to
initiate ventilation in ALS patients. However, starting ventilation early could be
beneficial for survival, although most studies recognise a temporal bias. It is very
difficult to identify the common, objective starting point (homogeneous clinical
situations and degree of evolution) to determine the increase in survival related
to ventilation, but some studies have suggested that the benefits are improved
if ventilation is started early, especially if this occurs before the patient has
hypercapnia. It is not easy to monitor symptoms and signs to detect potential
respiratory impairment because respiratory muscle weakness (RMW) develops
insidiously, with subtle symptoms. Table 1 shows groups of symptoms, adapted
from National Institute for Health and Clinical Excellence (NICE) guidelines, that
can help to identify RMW. Clinical assessment should take into account the pres-
ence of suggestive signs, for example:
• increased respiratory rate • weak cough and/or sniff
• shallow breathing • use of accessory muscles of respiration
Key points
• The patient’s values and views are key for selecting the right
treatment and identifying the therapeutic ceiling.
• The main goal is to design a care package, starting with local
resources, that meets patients’ needs.
• It is essential to know the patient’s wishes in the advanced
stage of the disease so as to properly orient the decisions made
in critical situations.

Long-term NIV
Table 1. Symptoms to detect potential RMW
General symptoms
Fatigue
Poor concentration and/or memory
Poor appetite
Confusion
Hallucinations
Respiratory (lung)-related symptoms
Breathlessness
Orthopnoea
Recurrent chest infections
Coughing when eating
Sleep-related symptoms
Disturbed sleep
Non-refreshing sleep
Nightmares
Daytime sleepiness
Morning headaches
Non-RMW
Repeated unjustified falls
Swallowing disorders
Voice changes
Limb weakness
Assessment
The NICE guidelines provide useful guidance on the assessment of patients with
motor neurone disease (National Institute for Health and Clinical Excellence, 2010).
There should be systematic monitoring of respiratory function and assessment of
whether there is dysfunction of the upper airway (especially with regard to symp-
toms of dysphagia that could suggest bulbar involvement). Table 2 summarises
the key elements in the evaluation of lung function tests.
Criteria for NIV
Although there are no objective criteria, the presence of symptoms and the identifi-
cation of alveolar hypoventilation are criteria for starting NIV. Some studies suggest
that the effectiveness of the ventilation (i.e. correction of nocturnal desaturation)
might have prognostic value with regard to increased survival. The effectiveness of
the ventilation is related to the degree of bulbar involvement. Early work suggested
that NIV could not be easily applied in patients with bulbar involvement. While it
may be more difficult to initiate in some patients, there is still every reason to offer
Long-term NIV
Table 2. Indications of potential pulmonary involvement
Test Potential pulmonary involvement

SpO2 # SpO2 <94%
SpO2 <92% for those with lung disease
SpO2 >94% (or >92% for those with lung disease) but with sleep-
related respiratory symptoms: refer for sleep studies
Arterial PaCO2 >6 kPa (>45 mmHg): refer to a specialised service
blood gases#
PaCO2 <6 kPa (<45 mmHg), with symptoms (orthopnoea): refer to a
specialised service and/or for sleep studies
VC or FVC VC or FVC <50% predicted
VC or FVC <80% predicted with symptoms (orthopnoea) consider:
VC standing and lying
Sleep study
SNIP and/or SNIP or MIP <40 cmH2O
MIP¶
SNIP or MIP <65 cmH2O for males or 55 cmH2O for females, plus
any symptoms (particularly orthopnoea)
Decline >10 cmH2O over a 3-month period
VC: vital capacity; SNIP: sniff nasal inspiratory pressure; MIP: maximal inspiratory pressure.
#: at rest and breathing room air; ¶: assessment should be performed using the better value.
Reproduced and modified from National Institute for Health and Clinical Excellence (2010),
with permission from the publisher. This material was accurate at the time of going to press.
NIV in bulbar patients (especially those with mild-to-moderate bulbar weakness)

in view of the possible benefits in terms of quality of life, particularly amelioration
of sleep-related symptoms.
Tracheostomy ventilation
Ventilation via tracheostomy is an alternative for ALS patients with severe bulbar
weakness, aspiration and extreme ventilator dependency. The decision on trache-
ostomy depends more on views and wishes (of the patient, of the professionals
and of the social context) than on strictly technical criteria. Conversion from NIV
to tracheostomy ventilation usually involves expansion of the care package and
training of carers in tracheostomy management.
In some cases, the tracheostomy is carried out in the course of ARF. Decannulation
can be very difficult if cough flows are <160 L⋅min−1, even with the addition of a
cough insufflation–exsufflation device.
There is uncertainty about the possible benefits of using diaphragmatic pacemakers
in ALS, so at present, this is not evidence-based therapy. Trials of diaphragm pacing
in addition to NIV are ongoing.
Equipment
Patients with neuromuscular diseases (especially ALS) need a multitude of equip-
ment for support, as well as the ventilator. Table 3 summarises these needs.
Long-term NIV
Table 3. Devices needed by ALS patients
Medical devices directly related to respiratory function

Ventilator (assess the need for a spare ventilator)
Batteries
Devices related to managing secretions (e.g. suction machine, mechanical
insufflation–exsufflation device, bag valve mask)
Communication
Self-produced material (photos, icons, etc.)
Augmentative and alternative communication devices
Devices to make everyday activities easier
Transfers
Mobility (electric wheelchair)
Transportation (adapted vehicles)
Miscellaneous support apparatus (for dressing, eating etc.)
Adaptation of the home
Accessibility (accessible lifts)
Bed
Personal hygiene (adaptation of the bathroom)
The support needs directly related to respiration vary widely, and depend on the
situation of the patient and the context. Sometimes, sophistication can run con-
trary to practicality, so very complex communication apparatus can be difficult to
manage. In other cases, when the disease evolves very quickly, certain apparatus
becomes useless. It might be useful to organise “device banks” where the patient
can choose what best meets their needs and use it for as long as necessary. This
approach solves the problem of supply (it is immediate) and is less of a financial
burden than direct purchase.
Multidisciplinary intervention
ALS patients require interventions by many different professionals. The conven-
tional approach (successive consultations and referrals to other specialists) is not
convenient for ALS patients. Repeated examinations and lack of coordination must
be minimised.
The management of secretions is crucial in the care of ALS patients. An inability
to generate sufficiently high peak cough flow has been related to increased
mortality in ALS patients. Peak cough flows can be measured easily using portable
spirometers. Peak cough flows of <270 L⋅min−1 indicate a high risk of ineffective
coughing during acute periods. It is very important to train carers in how to
improve cough effectiveness either through manually assisted cough manoeuvres
or by means of a cough-assist apparatus for managing expectoration. Techniques
for cough augmentation are discussed in the section entitled “Practicalities of and
guide to cough augmentation and daytime mouthpiece ventilation”.
Long-term NIV
Percutaneous gastrostomy is indicated in patients at risk of aspiration (food

entering the lung passages). The right time to perform gastrostomy is hard to
decide. Sometimes, choking and aspiration are not obvious, and only weight loss is
observed resulting from a change in diet as a spontaneous response to problems
with swallowing. Whenever possible, percutaneous gastrostomy will be under-
taken in the best possible functional situation (ideally with a vital capacity >50%)
to avoid the risks associated with the intervention. Insertion of the gastrostomy
may be under radiological control (e.g. radiological insertion of gastrostomy tube)
or performed as a limited open procedure. In all cases, close monitoring of the
patient is required both pre- and post-operatively.
Hypersalivation can be an important problem and, if agents such as hyoscine der-
mal patches or glycopyrrolate are unsuccessful, in some cases, there may be a
need to propose radiotherapy of the submaxillary and parotid glands to control the
secretion of saliva. Palliative care input is crucial.
Psychological support for ALS patients and their carers is very important. The
needs are different at the time of diagnosis, during the evolution of the disease,
and at the time of taking end-of-life decisions, and require specific interventions.
In addition, providing emotional support for the patient and the family must be
incorporated into the skills of all the members of the care team.
Organising care
The way in which the different professionals caring for ALS patients are organised
is very important. Highly centralised networks are useful because they concen-
trate specialist knowledge and resources. However, they often make accessibility
difficult for patients and generate unnecessary journeys.
Working in a distributed network is very useful for patients if the roles of all the
professionals involved are shared properly, if there is easy access to information
and fluid communication, and it is clear which professional does the monitoring
and guarantees coordination (a case manager).
Care must be organised using the locally available resources to respond to the
patients’ needs. The design of the “care package” is variable depending on circum-
stances (available resources) and the patient’s preferences.
Caring for ALS patients must take into account the possibility of using intermedi-
ate care facilities, both for the direct care of patients with complex needs and as
temporary relief for the carer.
Home care can be very useful for managing acute situations and avoiding hospital
admissions. Information and communication technologies are a useful comple-
ment to the overall care package. It is very important to plan, as far as possible,
the response to acute situations. For an ALS patient, an unplanned visit to an
emergency service can be especially painful.
End stages of life and ethical dilemmas
Caring for patients in the advanced stages of life is difficult. The criteria for decid-
ing the best place to deal with these patients during the final period depend on
the patient’s own preferences and on the available resources. A recent study
showed that more than half of ALS patients die at home and the longer the
Long-term NIV
evolutionary course of the disease, the more likely it is that death will occur out-
side the hospital.
Knowing the patient’s values and advance directives are very important in taking
decisions about the therapeutic ceiling. In the case of patients using NIV, it is nec-
essary to decide on the indications or otherwise of a tracheostomy, and all cases
raise the question of continuing with ventilation. In any event, the debate must
not be focused on a dichotomy between treatment and non-treatment but must
always concentrate on reorienting the treatment, not withdrawing it, to give the
best possible response to the patient’s needs.
Further reading
• Bourke SC, et al. (2006). Effects of non-invasive ventilation on survival and quality
of life in patients with amyotrophic lateral sclerosis: a randomised controlled trial.
Lancet Neurol; 5: 140–147.
• Escarrabill J, et al. (2014). Place of death in patients with amyotrophic lateral scle-
rosis. Rev Port Pneumol; 20: 188–193.
• Farrero E, et al. (2005). Survival in amyotrophic lateral sclerosis with home
mechanical ventilation: the impact of systematic respiratory assessment and bul-
bar involvement. Chest; 127: 2132–2138.
• Georges M, et al. (2014). Noninvasive ventilation reduces energy expenditure in
amyotrophic lateral sclerosis. BMC Pulm Med; 14: 17.
• Gonzalez-Bermejo J, et al. (2013). Prognostic value of efficiently correcting noctur-
nal desaturations after one month of non-invasive ventilation in amyotrophic lat-
eral sclerosis: a retrospective monocentre observational cohort study. Amyotroph
Lateral Scler Frontotemporal Degener; 14: 373–379.
• National Institute for Health and Clinical Excellence (2010). CG 105 Motor neu-
rone disease: the use of non-invasive ventilation in the management of motor
neurone disease. London, NICE.
• Simonds AK (2003). Ethics and decision making in end stage lung disease. Thorax;
58: 272–277.

Long-term NIV
Chronic NIV in chest wall disorders
Chest wall disorders include disorders of the spine and its articulation (kyphoscoli
osis and ankylosing spondylitis), disorders of the sternum (pectus excavatum),
and disorders of the ribs (e.g. a flail chest or a, now uncommon, status post-
thoracoplasty). Obesity also affects the mechanics of the chest wall, as it adds
soft tissue to the rib cage and abdomen. Kyphoscoliosis severely affects respira-
tory function, while other chest wall disorders, such as ankylosing spondylitis and
pectus excavatum, rarely cause this problem. Kyphoscoliosis is classified as idio
pathic, secondary or congenital (table 1). Scoliosis (spinal curvature in the lateral
plane) is usually associated with kyphosis (spinal curvature in the sagittal plane),
while kyphosis occurs more frequently in isolation. In scoliosis, there is a relation-
ship between the severity of the scoliosis as measured by the angle of the curva-
ture in the spine, known as the Cobb angle (fig. 1), and the risk of development of
These disorders affect the respiratory system due to decreased compliance of
the chest wall. The stiff chest wall reduces the resting position of the respira-
tory system (functional residual capacity (FRC)) (fig. 2), resulting in tidal breathing
at a flatter, stiffer portion of the respiratory volume–pressure curve. This means
that a greater inspiratory effort is required for relatively small tidal breaths, which
Key points
• Benefits of chronic NIV in patients with chest wall disorders
have mainly been shown in uncontrolled trials. Despite this,
the benefits were convincing enough for chronic NIV to
become standard care in these patients.
• Different modes of NIV can be effective.
• The indication to start NIV is a combination of elevated
daytime PaCO2 (>6 kPa or 45 mmHg) and symptoms
reflecting hypoventilation or the development of its
associated complications.
• Adequate monitoring of gas exchange, as an outcome
parameter of effectiveness, is essential.

Long-term NIV
Table 1. Disorders causing chest wall deformities that lead to chronic ventilatory failure
(Kypho)scoliosis
Idiopathic
A multigene disorder, developing by childhood/adolescence
Secondary to:
Neuromuscular disease (muscular dystrophy, poliomyelitis, cerebral palsy,
Friedreich’s ataxia, Charcot–Marie–Tooth disease)
Vertebral disease (osteomyelitis/osteomalacia, Pott’s disease (tuberculous
spondylitis), neurofibromatosis, rickets, spina bifida)
Connective tissue disease (Marfan Syndrome, Ehlers–Danlos syndrome,
Morquio’s syndrome)
Post-thoracoplasty
Congenital
Due to spinal/vertebral abnormalities at birth
Ankylosing spondylitis
Aquired abnormalities of the thoracic cage
Thoracoplasty
Fibrothorax
Chest wall tumours
Abdominal processes, including morbid obesity and ascites
increases the work of breathing. However, respiratory muscle strength might

be decreased because of intrinsic respiratory muscle weakness (as in secondary
kyphoscoliosis due to neuromuscular diseases) or because the respiratory muscles
act at a mechanically disadvantageous position because of the altered chest wall
configuration. To prevent respiratory muscle fatigue patients adopt a pattern of
rapid shallow breathing with a low VT, leading to unequal ventilation of the lungs
and pulmonary microatelectasis. Respiratory failure eventually develops once the
output that can be generated by the respiratory muscles is insufficient to over-
come the additional load placed on the system.
Benefits
NIV has been used in patients with chest wall disorders and chronic hypoventila-
tion for decades. Despite this, evidence in terms of RCTs is quite thin. The most
recent Cochrane systematic review identified only one longer-term study in 26
patients. However, benefits have been shown in many reports and uncontrolled
studies. NIV improves sleep architecture, and nocturnal and daytime arterial blood
gases almost immediately after initiation. In addition, it might improve pulmonary
mechanics, as small improvements in lung volumes have been shown, and relieve
the work of breathing, as endurance time of the respiratory muscles may increase.
Long-term NIV
a) Posteroanterior b) Lateral
60°
70°
Figure 1. Drawings of the spine showing the lines constructed to measure the Cobb angle
of scoliosis and kyphosis. The degree of scoliosis can be assessed on the a) posteroanterior
radiograph and the degree of kyphosis on the b) lateral radiograph. The angle can be calculated
in two ways: either from the intersection of the lines parallel to the vertebrae (as shown in b)
for kyphosis) or from the intersection of lines perpendicular to these lines (as shown in a) for
scoliosis). Reproduced from Tzelepis et al. (2010), with permission from the publisher.
These physiological benefits translate into long-term clinical benefits, as NIV has
been reported to be effective in:
• reducing hypoventilation related symptoms and dyspnoea on exertion
• improving mental function
• reducing the number of days spent in hospital
• increasing health-related quality of life
Furthermore, long-term NIV has been shown to improve pulmonary haemo
dynamics. Survival in patients with chest wall disorders on long-term NIV is quite
good, with a 5-year survival rate reported between 76% and 90%. Although the
effect of NIV on survival has not been investigated in RCTs, there is indirect evi-
dence that (effective) NIV improves survival (fig. 3).
Many benefits of NIV have been shown, although mainly in uncontrolled trials.
Chronic NIV has become standard care in patients with chest wall disorders suf-
fering from symptoms of chronic hypoventilation.
Types of ventilatory support
Before the introduction of positive pressure ventilation (PPV), patients with chest
wall disorders were ventilated by NPV using a tank ventilator or cuirass. Although
NPV could be effective, it was often uncomfortable, especially in patients with
severe kyphoscoliosis who had to lie on their back and had limited mobility.
Furthermore, NPV often induced upper airway collapse and synchronisation
Long-term NIV
a) TLC b)
FRC TLC
Pressure Pressure
cmH2O -40 -20 20 40 -40 -20 20 40 cmH2O
FRC
0 0
Volume Volume
Lung Chest wall Respiratory system
Figure 2. Pressure–volume curves in a) a healthy subject and b) a patient with a kyphoscoliosis.

Schematic illustration of alterations in volume–pressure relationship of the lung, chest
wall and respiratory system in kyphoscoliosis. The chest wall is less compliant, leading to
a reduction in functional residual capacity (FRC). As a consequence, tidal breathing takes
place in a flatter portion of the respiratory system volume–pressure curve. Reproduced and
modified from Tzelepis et al. (2010), with permission from the publisher.
between the patient and ventilator was a common problem as the NPV ventilators
could only ventilate in a controlled mode. In the 1980s, PPV delivered via a mask
became available. PPV can be delivered with preset volumes or preset pressures.
Both preset volume and preset pressure ventilation have been shown to be effec-
tive. With preset pressure ventilation, a certain amount of inspiratory pressure is
necessary (usually between 15 and 20 cmH2O, titrated by changes achieved in
arterial blood gases), while PEEP is not essential, as lung hyperinflation is not
a feature of this disease. The EPAP is usually set at 2–4 cmH2O to overcome
the system dead space. The type of mask used is a matter of individual prefer-
ence (see the section entitled “Choosing the interface”); a nasal mask is often
sufficient.
Patient selection
High-risk patients (with normal (diurnal) arterial blood gases)
Patients with a low vital capacity (VC), those developing scoliosis at young age
and those with severe (kypho)scoliosis are at risk of developing nocturnal
hypoventilation, usually during their fourth or fifth decade of life. In a long-term
follow-up study, it was shown that a VC of <43% predicted at the age of skeletal
maturity reached, which was strongly and inversely related to a large scoliotic
angle, was a strong predictor of the development of respiratory failure over the
next 20 years. In some cases, especially in those with a severe (50–80°) thoracic
curve at the age of reaching skeletal maturity, increases in thoracic curvature
contribute to the development of respiratory failure. Other factors, such as the
effect of increasing age on VC, reduced diaphragmatic functioning, obesity or
smoking, might also contribute to the development of chronic respiratory failure.
These patients should be followed up and monitored for the occurrence of
symptoms of nocturnal hypoventilation, which can become apparent insidiously
or as an emergency during an acute illness. There is no evidence to start NIV
“prophylactically” when patients at risk still have normal diurnal arterial blood
gases. NIV is sometimes started when patients have normal daytime arterial blood
Long-term NIV
a) 100
90
80
70
Survival %
60
50
40
30
20
10
0
0 12 24 36 48 60 72 84 96 108 120
Follow-up months
Martí et aI. (2010) Buyse et aI. (2003) HMV Gustafson et aI. (2006) HMV
Duiverman et aI. (2006) Buyse et aI. (2003) LTOT Gustafson et aI. (2006) LTOT
Simonds et aI. (1995)
Figure 3. Survival curves from several studies showing that the survival in patients with
chest wall disorders treated with long-term home mechanical ventilation (HMV) is good and,
as shown by Buyse et al. (2003) and Gustafson et al. (2006), is substantially better with
HMV compared with long-term oxygen therapy (LTOT).
gases but an elevated PaCO2 during sleep. However, there is little evidence that
patients eventually benefit from this policy. The decision whether or not to start
NIV in this situation usually depends on:
• the severity of the physiological abnormalities detected during sleep, in par-
ticular the degree of hypoxia and hypercapnia
• the presence or absence of complications such as polycythaemia or a high pul-
monary artery pressure
In addition, if these patients encounter problems with sputum clearance this
might be another reason to start NIV. The only chest wall disease in which it has
been shown that early intervention may be of benefit is scoliosis associated with
poliomyelitis, as there is evidence that relief of excessive muscle activity may
reduce the rate of progression of the postpoliomyelitis syndrome.
Patients with abnormal diurnal blood gases
Patients with abnormal blood gases both at night and during the day may present
without symptoms of nocturnal hypoventilation or complications (e.g. polycythaemia
vera or pulmonary hypertension). In these patients, there is little evidence for start-
ing NIV as it has not been investigated whether this changes the clinical course of
the disease. A combination of elevated daytime PaCO2 (>6 kPa or 45 mmHg) and
symptoms reflecting hypoventilation or the development of its associated compli-
cations is stated in most guidelines as the indication to start NIV. In this case, NIV
has been shown to be of benefit.
Long-term NIV
Contraindications
Contraindications for NIV in chest wall disorders are, usually, the same as for other
indications:
• Inability to wear a mask
• Showing insufficient motivation
• In patients with a severely limited life expectancy NIV may be of limited benefit
and, therefore, a relative contraindication
Initiation and monitoring
Traditionally, NIV is initiated in a clinical setting, as in hospital, more advanced
monitoring is available and adjustments to NIV can be made more easily. As rever-
sal of nocturnal alveolar hypoventilation seems to be the basis of further clinical
benefits, monitoring of gas exchange (during both night and day) is essential. If
hypoventilation is insufficiently corrected, this has been shown to even negatively
affect survival.
However, more advanced and noninvasive monitoring equipment has become
available recent years enabling measurement of gas exchange with end-tidal or
transcutaneous measurement devices as well as use of telemonitoring devices.
In the Netherlands, a recent pilot showed that initiation of mechanical ventila-
tion at home titrated on PtcCO2 measurements and monitored by telemonitoring
is effective and safe (fig. 4). When confirmed in further studies, home initiation
might be an attractive alternative to hospital initiation as it is patient friendly and
cost-effective.
a) b)
SpO2
HR
PtcCO2
c)
Figure 4. Set-up and displayed results of the telemonitoring equipment at home. a) Setup:
ventilator, humidifier, transcutaneous monitor and laptop with mobile connection.
b) Display for results of the PtcCO2 monitoring, showing SpO2, heart rate (HR) and PtcCO2.
c) Results from by the ventilator showing inspiratory and expiratory pressures, breathing
frequencies and delivered VT. Results are saved on the laptop and sent electronically to the
hospital. Reproduced and modified from Hazenberg et al. (2014), with permission from
the publisher.
Long-term NIV
Summary
In patients with chest wall disorders suffering from the symptoms of chronic
hypoventilation, chronic NIV has many benefits. Although these benefits were
mainly shown in uncontrolled trials, the benefits were convincing enough for
chronic NIV to become standard care in these patients. RCTs, therefore, can no
longer be expected. In recent decades patients have mainly been ventilated using
PPV, with preset volume and preset pressure ventilation showing equal effec-
tiveness. The indication to start NIV is a combination of elevated daytime PaCO2
(>6 kPa or 45 mmHg) and symptoms reflecting hypoventilation or the develop-
ment of its associated complications. There is no evidence that starting earlier in
the course of the disease is of benefit. Recently, with the increasing availability of
noninvasive monitoring techniques, the possibility of initiating NIV in the home
settings has been investigated, and as promising results were obtained, home ini-
tiation might become an attractive and patients friendly alternative in the future.
Further reading
• Agre JC. (1994). Local muscle and total body fatigue. In: Halstead LS, et al., eds.
Post-Polio Syndrome. Philadelphia, Hanley and Belfus Inc., pp. 35–67.
• Buyse B, et al. (2003). Treatment of chronic respiratory failure in kyphoscoliosis:
oxygen or ventilation? Eur Respir J 22: 525–528.
• Cobb JR (1948). Outline for the study of scoliosis. Am Acad Orthop Surg Inst Course
Lect 5: 261–275.
• Duiverman ML, et al. (2006). Home mechanical ventilatory support in patients
with restrictive ventilatory disorders: a 48-year experience. Respir Med 100: 56–65.
• Ferris G, et al. (2000). Kyphoscoliosis ventilatory insufficiency: noninvasive man-
agement outcomes. Am J Phys Med Rehabil 79: 24–29.
• Gustafson T, et al. (2006). Survival of patients with kyphoscoliosis receiving
mechanical ventilation or oxygen at home. Chest 130: 1828–1833.
• Hazenberg A, et al. (2014). Initiation of home mechanical ventilation at home:
a randomised controlled trial of efficacy, feasibility and costs. Respir Med 108:
1387–1395.
• Hill NS, et al. (1992). Efficacy of nocturnal nasal ventilation in patients with restric-
tive thoracic disease. Am Rev Respir Dis 145: 365–371.
• Martí S, et al. (2010). Predictors of mortality in chest wall disease treated with
noninvasive home mechanical ventilation. Respir Med 104: 1843–1849.
• Pehrsson K, et al. (1991). Lung function in adult idiopathic scoliosis: a 20 year fol-
low up. Thorax 46: 474–478.
• Pehrsson K, et al. (1994). Quality of life in patients treated by home mechanical
ventilation due to restrictive ventilatory disorders. Respir Med 88: 21–26.
• Schönhofer B, et al. (1997). Comparison of two different modes for noninvasive
mechanical ventilation in chronic respiratory failure: volume versus pressure con-
trolled device. Eur Respir J 10: 184–191.

Long-term NIV
• Schönhofer B, et al. (2000). Effect of non-invasive mechanical ventilation on sleep

and nocturnal ventilation in patients with chronic respiratory failure. Thorax 55:
308–313.
• Schönhofer B, et al. (2001). Long term effects of non-invasive mechanical ventila-
tion on pulmonary haemodynamics in patients with chronic respiratory failure.
Thorax 56: 524–528.
• Schönhofer B, et al. (2001). Noninvasive mechanical ventilation improves endur-
ance performance in patients with chronic respiratory failure due to thoracic
restriction. Chest 119: 1371–1378.
sure ventilation in restrictive and obstructive disorders. Thorax 50: 604–609.
• Tzelepis GE, et al. (2010). The lung and chest wall diseases. In: Mason RJ, et al., eds.
Murray and Nadel’s Textbook of Respiratory Medicine. 5th Edn. Oxford, Elsevier;
pp. 2067–2083.
• Ward S, et al. (2005). Randomised controlled trial of non-invasive ventilation (NIV)
for nocturnal hypoventilation in neuromuscular and chest wall disease patients
with daytime normocapnia. Thorax 60: 1019–1024.

Long-term NIV
Chronic NIV in COPD
Wolfram Windisch and Jan H. Storre
Long-term NIV delivered by a nasal or a full-face mask is a well-established and

increasingly used therapeutic option to treat patients with chronic hypercapnic
respiratory failure that arises from different aetiologies. While the role of long-
term NIV in restrictive thoracic and neuromuscular patients as well as in those
with OHS is undisputed, there is still debate as to whether long-term NIV should
be established in patients with chronic hypercapnic respiratory failure that arises
from COPD. This discussion is based on the observation that several long-term
RCTs could not establish a convincing positive effect on long-term survival.
In most of these RCTs the ventilator settings were not sufficient to signifi-
cantly reduce elevated PaCO2 values; assisted forms of NIV using mean IPAP
settings ranging 10–18 cmH2O were applied in most of the trials. Therefore,
this technique has been shown to be ineffective in improving the physiology, i.e.
improving the chronic hypercapnia that provides evidence of chronic alveolar
hypoventilation.
Key points
• Whether or not long-term NIV should be used in patients with
chronic hypercapnic respiratory failure due to COPD has been
the focus of lengthy discussions over the past 20 years.
• New findings suggest that higher ventilator settings aimed at
achieving normocapnia are physiologically highly effective, and
this has been labelled as high-intensity NIV.
• The capability of NIV to significantly reduce elevated PaCO2
values has been suggested to be more evident in the following
conditions: 1) IPAP >18 cmH2O; 2) compliance >5 h of
ventilation per night; 3) baseline PaCO2 >55 mmHg (7.33 kPa).
• Increasingly the available data support the potential of long-term
NIV for improving lung function, exercise tolerance, health-related
quality of life and long-term survival, but NIV settings sufficient to
improve hypercapnia are mandatory to achieve this goal.

Long-term NIV
Following these trials, the most recent meta-analysis published by Struik et al.
(2014) concluded that there was still insufficient evidence to support the appli-
cation of routine NIV in stable COPD patients. However, the authors added that
higher IPAP levels, better compliance and higher baseline PaCO2 seemed to improve
the physiology and therefore reduce elevated PaCO2 levels. In particular, the con-
cept of using more aggressive forms of NIV that combine higher IPAP settings and
higher respiratory rates has been shown by several trials to provide several physio
logical and clinical benefits. This approach has been labelled high-intensity NIV,
while the conventional approach using lower settings as described above has been
labelled low-intensity NIV.
Rationale for NIV and physiological considerations
Chronic respiratory failure in COPD patients refers to either chronic pulmonary
failure or chronic ventilatory failure. Chronic pulmonary failure is characterised
by hypoxaemia and can be successfully treated by long-term oxygen therapy. By
contrast, hypercapnia is the characteristic hallmark of ventilatory failure.
In COPD patients, chronic ventilatory failure occurs as a consequence of an
increased load imposed on the respiratory muscles in addition to reduced inspira-
tory muscle strength. The latter is primarily ascribed to hyperinflation-induced dia-
phragm shortening, thus reducing the surface area of the diaphragm, i.e. the zone
of apposition. Elevating the lower part of the rib cage is then hindered. In addition,
higher lung volumes are acknowledged to reduce the pressure generation cap
ability of the diaphragm as a consequence of the physiological pressure–volume
relationship of the diaphragm. Importantly, cellular and molecular alterations of
the diaphragm muscle fibres in COPD patients are suggested to occur early in the
course of the disease.
Hypercapnic respiratory failure in COPD can occur acutely and chronically. During
acute exacerbation, acute development of hypercapnia is associated with respira-
tory acidosis. By contrast, if hypercapnic respiratory failure develops chronically
renal retention of bicarbonate aims to compensate for respiratory acidosis.
The physiological goal of long-term NIV in chronically hypercapnic COPD patients
is to augment reduced alveolar ventilation and, thereby, to reduce elevated lev-
els of PaCO2. This, in turn, should translate into a reduction in symptoms, most
importantly dyspnoea. Other symptoms of chronic ventilatory failure that should
be target for NIV are sleep disordered breathing, recurrent infections and symp-
toms related to hypercapnia-induced dilation of the vessels (morning headache,
vasodilatation visible in the conjunctiva of the eyes and, importantly, peripheral
oedema). Moreover, long-term NIV in COPD seeks to improve health-related qual-
ity of life (HRQoL), exacerbation rates and, at best, long-term survival.
Impact of the ventilatory strategy used for NIV?
As mentioned earlier, low ventilator settings used with assisted forms of ventila-
tion and IPAP settings of <18 cmH2O are physiologically rather ineffective and not
convincingly shown to improve outcome. In contrast, more aggressive forms of
NIV have been shown to be superior. Typically, ventilator settings are increased in
a stepwise approach to the individually tolerated maximum if necessary or to the
level necessary to achieve normocapnia. For that purpose, controlled ventilation

Long-term NIV
with high IPAP levels, typically ranging 20–30 cmH2O, is implemented in the

hospital setting and subsequently used at home. Even though high-intensity NIV
nearly always uses higher IPAP settings than the conventional approach of low-
intensity NIV, high-intensity NIV is not defined by a certain IPAP level but rather by
the effort to reduce elevated PaCO2 values. Therefore, the term high-intensity NIV
may include different modes and settings if aimed at PaCO2 reduction, but typically
the assist/control mode is chosen for high-intensity NIV.
Who are the candidates for long-term NIV?

During a stable disease state, COPD patients can be offered long-term NIV if they
have a well-established diagnosis of the underlying disease (COPD), if they have
symptoms of chronic hypercapnic respiratory failure and reduced HRQoL, if they
have evidence for chronic hypercapnic respiratory failure, and if all other conserva-
tive treatment strategies have already been established (fig. 1.). Importantly, base-
line PaCO2 values of >55 mmHg (7.33 kPa) are reportedly associated with higher
effectiveness of NIV. Regarding symptoms, peripheral oedema are not exclusively
due to cardiac failure, but may, like morning headache and visible vasodilatation
of the conjunctival vessels, also occur as the result of hypercapnia, since CO2 is
known to be a strong vasodilator.
Established diagnosis of COPD according to guidelines
Medication and long-term oxygen treatment according to guidelines already

established
Reduced HRQoL and symptoms of chronic hypoventilation:

Dyspnoea
Morning headache
Vasodilation of the conjunctival vessels
Peripheral oedema
Symptoms of sleep disordered breathing such as sleepiness, mental disturbances,
anxiety and depression
Chronic hypercapnic respiratory failure – one of the following:

Daytime PaCO2 ≥50 mmHg (6.66 kPa) with pH >7.35 and elevated bicarbonate levels
Nocturnal PaCO2 ≥55 mmHg (7.33 kPa)
Increase in PtcCO2 ≥10 mmHg (1.33 kPa) overnight
Consider long-term NIV
Figure 1. Indication criteria for long-term NIV in COPD patients during stable disease.
Long-term NIV
Hospitalised COPD patients with acute hypercapnic respiratory failure and

respiratory acidosis due to acute exacerbation, particularly if they had required
mechanical ventilation, have a high likelihood of hospital readmission and also
death, due to recurrent episodes of acute hypercapnic respiratory failure after
they have successfully treated in hospital. It seems reasonable to consider NIV
in these patients with the aim to avoid rehospitalisation and improve prognosis.
Recent research has indicated, however, that patients who remained hypercapnic
for 48 h after acute exacerbation and the requirement for acute NIV did not
benefit from long-term NIV in terms of readmission and prognosis. Nevertheless,
a close follow-up of these patients is recommended and long-term NIV should
be considered on an individual basis if hypercapnia persists for weeks or even
months.
The practical approach to long-term NIV in COPD patients
In general, NIV should be commenced in a stepwise approach in hospitalised patients
while monitoring blood gases and vital parameters. Although this might require an
individually adapted procedure, the single steps necessary for initiating NIV have a
typical order (box 1). Nevertheless, changing the order or using a modified approach
may sometimes increase success or may save time in hospital. Pitfalls and practical
advice for solving problems are provided in box 2. Finally, the effectiveness of NIV is
also dependent on the patients’ compliance with more substantial changes in PaCO2
observed in those patients with >5 h of ventilation per night.
Physiological effects of long-term NIV in COPD patients
NIV is physiologically only effective if ventilator settings are set appropriately. If a
reduction in PaCO2 during NIV is achieved this will be translated into a reduction in
PaCO2 during subsequent spontaneous breathing. This is primarily attributed to an
Box 1. High-intensity NIV: practical approach

• Use NIV in the daytime first, with the primary aim of establishing
tolerance.
• Start with assisted NIV and rather low IPAP levels typically ranging
between 12 and 16 cmH2O.
• Subsequently, carefully increase IPAP in a stepwise approach until
maximal tolerance is reached, usually between 20 and 30 cmH2O.
• Next, increase the respiratory rate just beyond the spontaneous rate to
establish controlled ventilation, but avoid excessively high respiratory
rate settings that cause dynamic hyperinflation.
• Then, set EPAP in order to avoid dynamic hyperinflation according to
subjective comfort (usually 3–6 cmH2O, but possibly higher in patients
with coexisting OSAS), and similarly, set the inspiration:expiration ratio
to 1:2 or lower.
• Finally, apply nocturnal NIV with individual adjustment of ventilator
settings according to subjective comfort and nocturnal gas exchange.
• Adjust ventilator settings according to subjective tolerance and nocturnal
monitoring of blood gases. Sometimes settings can be considerably
modified at the first follow-up visit in hospital after the patients has been
acclimatised to NIV at home for some weeks.

Long-term NIV
Box 2. Pitfalls and practical advice for problems encountered when setting up
NIV in COPD patients
• Tolerance of higher IPAP levels can last from minutes to several days or
even weeks. Do not push too hard. Further NIV modification might be
feasible at the first follow-up visit in hospital after having discharged the
patient for acclimatisation in the home environment.
• In cases of co-existing upper airway obstruction, higher EPAP levels might
be required. However, higher EPAP reduces the effective IPAP (defined as
IPAP minus EPAP); thus, avoid high EPAP levels unless necessary.
• For targeted controlled NIV, respiratory rates are typically set to
1 breath⋅min−1 higher than during spontaneous breathing; avoiding
excessively high respiratory rates that might cause dynamic
hyperinflation.
• Keep in mind that air leakage can never be totally avoided, but should
minimised.
• Try out several masks: for nocturnal NIV, oronasal masks are sometimes
preferable to avoid leakage via an opened mouth; for daytime NIV,
a nasal mask is often better tolerated.
• Several days in hospital are usually necessary to establish high-intensity NIV.
• Use humidification in case of dry mucous membranes.
• Gastrointestinal side-effects can be managed by medication, positioning
and adjustments (reduction) of the ventilator settings; in these cases,
pressure-limited NIV is generally better tolerated than volume-limited NIV.
• High-intensity NIV may induce a reduction in cardiac output. Care must
be taken in patients with pre-existing cardiac disease.
improved breathing pattern with an increase in VT. A reduction in PaCO2 produces

an increase in pH during the application of NIV. This, in turn, allows the kidney to
eliminate excess bicarbonate ions, which had been accumulated by the kidney to
compensate for respiratory acidosis prior to NIV.
In addition, NIV aimed at normalising PaCO2 is capable of improving oxygenation
and lung function. It has been suggested that NIV has an effect on the airways
in these cases. Possible mechanisms include a reduction in PaCO2-related airway
oedema or even stretching open of chronically fibrotic airways. This, however, is
speculative. In addition, a positive effect on respiratory muscle function or on the
respiratory breathing centre has yet not been convincingly demonstrated in COPD
patients; thus, the mechanisms by which NIV improves respiratory function dur-
ing subsequent spontaneous breathing still need to be established.
Clinical effects of long-term NIV in COPD patients
NIV has been shown to improve symptoms and HRQoL, whereas physical, func-
tional, psychological and social aspects of HRQoL are positively affected by NIV.
Again, ventilator settings sufficient to improve hypercapnia are needed to provide
subjective improvements. In addition, HRQoL needs to be measured by instru-
ments specific for severe chronic respiratory failure, such as the Severe Respiratory
Insufficiency Questionnaire or the Maugeri Foundation Respiratory Failure item
set (for further information see the section entitled “Assessing quality of life and
Long-term NIV
outcome of long-term NIV”). Clinical studies showing no HRQoL benefits or even

deteriorations in HRQoL following NIV should be interpreted with caution if only
generic and non-specific measurement tools have been used and/or if NIV was
physiologically ineffective as previously discussed.
Long-term NIV also has the potential to improve sleep quality and exercise per-
formance. NIV has been shown to augment the success of rehabilitation when
used nocturnally during rehabilitation, but has also been shown to improve
exercise capacity when used as an adjunct during physical activity. It is, however,
still unclear if long-term NIV can reduce exacerbation and hospital readmission
rates, even though some small studies have shown this. Nevertheless, patients
with acute hypercapnic respiratory failure requiring acute mechanical ventila-
tion in hospital do not seem to benefit from long-term NIV if PaCO2 returns to
normal after the exacerbation without NIV. However, this issue needs further
investigation.
Importantly, the most recent RCT by Köhnlein et al. (2014) showed that long-
term NIV was indeed capable of significantly and substantially improving long-
term survival in chronic hypercapnic COPD patients. This study contrasts with
all previous long-term trials, since earlier studies did not provide clear evidence
for survival benefits. As discussed earlier, the main difference was the ventilatory
strategy, which was successful in reducing elevated PaCO2 values in the trial by
Köhnlein et al. (2014). Therefore, long-term NIV has the potential to improve
the prognosis, but only if ventilation is physiologically effective. Nevertheless,
the selection criteria differed between the different RCTs and it remains unclear
how this impacts prognosis. Therefore, more data on how to best select patients
and on long-term outcome depending on different baseline characteristics are
needed to verify the most recent findings.
Conclusion
Increasingly, the available data support that long-term NIV is beneficial in chronic
hypercapnic COPD patients. For this purpose, however, ventilator settings capa-
ble of significantly reducing PaCO2 values need to be chosen. Typically this involves
using higher inspiratory pressures and respiratory rates than previously used;
this approach has been labelled high-intensity NIV. If PaCO2 is reduced, long-term
NIV has the potential to improve lung function, exercise tolerance, HRQoL and
long-term survival in COPD patients. However, criteria for patient selection need
to be further established, and it still remains unclear if readmission rates can be
reduced by NIV in those patients who remain hypercapnic after full recovery from
an exacerbation.
Further reading
• Consensus conference (1999). Clinical indications for noninvasive positive pres-
sure ventilation in chronic respiratory failure due to restrictive lung disease,
COPD, and nocturnal hypoventilation–a consensus conference report. Chest; 116:
521–534.

Long-term NIV
• Struik FM, et al. (2014). Nocturnal noninvasive positive pressure ventilation in sta-
ble COPD: a systematic review and individual patient data analysis. Respir Med;
108: 329–337.
• Windisch W (2011). Noninvasive positive pressure ventilation in COPD. Breathe;
8: 114–123.
• Windisch W (2013). Home mechanical ventilation. In: Tobin MJ, ed. Principles
and Practice of Mechanical Ventilation. 3rd edn. New York, McGraw Hill Medical;
pp. 683–697.
• Windisch W, et al. (2015). Nocturnal non-invasive positive pressure ventilation for
COPD. Expert Rev Respir Med; 9: 295–308.
Online resources
• Hare A, et al. Severe stable COPD module. Skills-based Simulator Training in
Non-Invasive Ventilation. www.ers-education.org/e-learning/simulators.aspx

Long-term NIV
Chronic NIV in OHS
Patrick B. Murphy and Nicholas Hart
Obese patients with chronic respiratory failure are an increasingly common

group receiving domiciliary NIV. There are three distinct patterns of sleep dis-
ordered breathing, which are readily distinguished by review of the respiratory
sleep studies, and these require differing approaches to management of the
sleep disordered breathing and chronic respiratory failure. The patterns observed
are:
1) Isolated severe OSA (fig. 1), where the post-apnoea hyperpnoeic response
is insufficient to clear the increased carbon dioxide load associated with the
obstructive event
2) Isolated OHS (fig. 2), where hypoventilation occurs in the absence of upper
airways obstruction
3) Combined OSA and OHS, where hypoventilation is superimposed on the back-
ground of upper airways obstruction (fig. 3)
Key points
• Chronic respiratory failure in obese patients is underdiagnosed
and screening for sleep disordered breathing in obese patients
presenting to other specialties is strongly advised.
• Elevated base excess level in obese patients without daytime
hypercapnia may represent an early stage of obesity-related
chronic respiratory failure.
• The optimum treatment for obesity-related chronic respiratory
failure, using nocturnal CPAP or NIV, will depend on the
contribution of upper airways obstruction and hypoventilation
to the sleep disordered breathing with auto-titrating volume-
targeted NIV having similar clinical efficacy to manual titration
of fixed level pressure support NIV.
• Step-down to CPAP therapy should be considered in patients
with predominantly upper airways obstruction in whom
chronic respiratory failure is reversed with NIV.

Long-term NIV
100
90
SpO2 %
80
70
120
beats·min-1
Heart rate
100
80
60
40
10
PtcCO2 kPa
8
6
4
23:00 h 01:00 h 03:00 h
Figure 1. Severe OSA. In this phenotype of sleep disordered breathing in obesity-related
chronic respiratory failure, periods of rapid repetitive oxygen desaturations with marked
heart rate variability are accompanied by an increase in PtcCO2, suggesting severe OSA
with inadequate post-apnoea ventilation (arrows) presumed to be during rapid eye
movement sleep.
There has been only a single RCT that has demonstrated an improved outcome
using home NIV compared with lifestyle advice, with clinical benefits observed
in sleep architecture and sleep disordered breathing. There are, however, several
uncontrolled studies supporting the therapeutic effect of either domiciliary NIV or
CPAP. These data, from observational and interventional trials comparing modes
of NIV, demonstrate a mortality and healthcare utilisation benefit in patients with
obesity-related chronic respiratory failure treated with NIV. Despite the increas-
ing incidence of this condition and the evidence to support improved long-term
outcomes with initiation of nocturnal respiratory support, underdiagnosis of
obesity-related chronic respiratory failure is common in clinical practice. Often
these patients are given a diagnosis of obstructive airway disease due to the clin
ical findings of expiratory wheeze, but this is, in general, not the result of intrinsic
airways disease but rather obesity-related early airways closure. This should be
carefully considered in those patients who do not have a history of asthma, have
a minimal pack-year smoking history and demonstrate a restrictive lung defect on
spirometry. Furthermore, it is not infrequent that such obese patients fail to be
initiated on home mechanical ventilation following the identification of chronic
respiratory failure, even following an episode of acute-on-chronic respiratory

failure.
A major contributor to the underdiagnosis of obesity-related respiratory failure
and overdiagnosis of other respiratory conditions is that chronic respiratory failure
in obese patients remains a diagnosis of exclusion. Currently, OHS is defined as
the combination of daytime hypercapnia (PaCO2 >6 kPa (>45 mmHg)) and sleep
disordered breathing in the absence of another cause of hypoventilation in patients
with a BMI >30 kg⋅m−2. Simple bedside tests, such as clinic pulse oximetry (SpO2)

Long-term NIV
100
SpO2 % 90
80
70
120
beats·min-1
Heart rate
100
80
60
10
PtcCO2 kPa
4
23:00 h
Figure 2. Lone OHS. In this phenotype of sleep disordered breathing in obesity-related

chronic respiratory failure, prolonged oxygen desaturation (red arrow), without evidence of
rapid repetitive oxygen desaturations (green arrow), is observed accompanied by an increase
in PtcCO2, indicating significant hypoventilation (blue arrow).
100
90
SpO2 %
80
70
120
beats·min-1
Heart rate
100
80
60
10
PtcCO2 kPa
8
6
4
23:00 h 01:00 h 03:00 h 05:00 h
Figure 3. Combined OSA and OHS. In this phenotype of sleep disordered breathing in
obesity-related chronic respiratory failure, rapid repetitive oxygen desaturations (green

arrow) with prolonged oxygen desaturation (red arrows) are observed accompanied by an
increase in PtcCO2 indicating significant hypoventilation (blue arrows).

Long-term NIV
and FVC, can facilitate the diagnosis of chronic respiratory failure. Specifically, in
obese patients with sleep disordered breathing, an FVC <3.5 L and a daytime SpO2
<95% in men, and an FVC <2.3 L and a daytime SpO2 <93% in women, should
prompt an arterial blood gas measurement. In addition, if the PaCO2 is <6 kPa
(<45 mmHg) but the base excess level is above 2 mmol⋅L−1, this may represent
an early stage of obesity-related chronic respiratory failure if all other causes of a
metabolic alkalosis have been accounted for. As part of the diagnostic pathway,
patients will undergo an overnight assessment to classify their sleep disordered
breathing, which can range from simple oximetry and capnometry to limited
respiratory polygraphy, or to full montage polysomnography (PSG). A number of
clinical guidelines recommend the use of full PSG to diagnose obesity-related
sleep disordered breathing, but, based on the current evidence, many centres
have established the diagnostic and therapeutic efficacy of respiratory-based
overnight monitoring.
The lack of any definitive RCT data means that the mode of respiratory support
employed depends on the characteristics of the sleep disordered breathing with
consideration given to the three different clinical phenotypes observed during the
overnight studies. CPAP has been shown to be equivalent to NIV in the medium-
term management of obese patients with moderate chronic respiratory failure who
have had a successful single night trial of CPAP. Specifically, in patients with a BMI
between 30 and 50 kg⋅m−2, with predominantly obstructive sleep apnoea and a
PaCO2 <7 kPa (<53 mmHg), an initial trial of CPAP is a clinically useful and cost-
effective strategy. Transition to NIV, however, is recommended in such patients if
nocturnal hypoventilation is demonstrated during the CPAP trial, shown by a rise
in overnight PtcCO2 with or without a fall in the early morning PaCO2. Patients with
combined OSA and OHS on the overnight study should be treated with NIV to man-
age both the upper airways obstruction and the nocturnal hypoventilation, with the
aim of normalising the PaCO2 and resetting neural respiratory drive as well as revers-
ing the associated cor pulmonale. Following adequate treatment, demonstrated
by a reversal of daytime hypercapnia and hypoxia, reassessment is recommended
with careful consideration of stepping down to CPAP therapy. In contrast, patients
with isolated OHS, i.e. without upper airways obstruction, should be established on
long-term home NIV.
A treatment algorithm for the management of obesity-related chronic respiratory
failure is shown in figure 4. EPAP or CPAP is titrated to abolish upper airways obstruc-
tion, which will offset the intrinsic PEEP and enhance the pulmonary mechanics
by placing the respiratory system at a more advantageous part of the pressure–
volume curve. IPAP is titrated to optimise alveolar ventilation and improve over-
night gas exchange. Evidence from post hoc analysis of previous clinical trials has
shown superior control of sleep disordered breathing and subsequent improved
clinical outcomes, in terms of daytime gas exchange, health-related quality of life
and subjective somnolence, in the patients treated with controlled rather than sup-
port ventilation. The evidence, therefore, favours a mandatory backup rate as part
of the NIV treatment strategy.
Advanced modes of PSV have been developed, which incorporate ventilator soft-
ware that calculates the estimated VT and modifies the delivered pressure sup-
port to reach a preset target volume. Such novel auto-titrating volume-targeted

Exclude co-existing airways disease and optimise management
Confirm diagnosis with overnight diagnostic assessment#
Exclude other causes of respiratory failure
Severe OSA without nocturnal hypoventilation Severe OSA with nocturnal hypoventilation Nocturnal hypoventilation in the absence of OSA
<1 kPa (<7.5 mmHg) rise in PtcCO2 overnight ≥1 kPa (≥7.5 mmHg) rise in PtcCO2 overnight ≥1 kPa (≥7.5 mmHg) rise in PtcCO2 overnight
PaCO2 <7 kPa (<53 mmHg) ≥1 kPa (≥7.5 mmHg) rise in evening to morning PaCO2 ≥1 kPa (≥7.5 mmHg) rise in evening to morning PaCO2
BMI 30–50 kg·m-2 AHI/ODI < 5 events·h-1
Establish interface and acclimatise patient with daytime therapy use

For overnight monitor with constant oxygen saturations, PtcCO2 and/or intermittent arterial blood gas analysis
Initial NIV settings Initial NIV settings
EPAP EPAP

Yes
Usual setting 8–12 cmH2O Usual settings 4–6 cmH2O
Start at 6 cmH2O and titrate by 2 cmH2O to: Start at 4 cmH2O and titrate by 2 cmH2O to:
CPAP trial (manual or Abolish upper airways obstruction Stop snoring
auto-titrating) Stop snoring/abdominal paradox Optimise triggering and offset PEEP
Uncontrolled nocturnal Stop repetitive desaturations
hypoventilation and/or IPAP
obstructive apnoeic events? IPAP Usual settings 18–24 cmH2O
Usual settings 18–24 cmH2O Start at 14 cmH2O and titrate by 2 cmH2O to:
Start at 16 cmH2O and titrate by 2 cmH2O to: Obtain adequate chest wall excursion
Obtain adequate chest wall excursion Obtain adequate VT (8–10 mL·kg-1 IBW)
Obtain adequate VT (8–10 mL·kg-1IBW)
No Other settings Other settings
Oxygen: entrained if hypoxia persists with adequate IPAP and EPAP Oxygen: entrained if hypoxia persists with adequate IPAP and EPAP
Backup rate: resting rate 2 breaths·min-1 Backup rate: resting rate 2 breaths·min-1
Inspiratory time: 1.2–1.5 s Inspiratory time: 1.2–1.5 s
Aim: SpO2 <90% for <10% of night, PtcCO2 fall or <0.5 kPa (<3.8 mmHg) rise, if using NIV <50% patient triggered breaths
Figure 4. Flow chart summarising initial setup of NIV for the management of different phenotypes of sleep disordered breathing in obesity-related chronic
respiratory failure. ODI: oxygen desaturation index; IBW: ideal body weight. #: overnight assessment can include oximetry and capnometry, limited respira-
tory polygraphy, or full PSG.
Long-term NIV
201
Long-term NIV
modes can compensate for the changes in respiratory load and neural respira-
tory drive that occur in different body positions and sleep stages and, therefore,
are potentially appealing. Initial data investigating these volume-targeted modes
suggest that they provided enhanced control of nocturnal hypoventilation but
at the cost of increased sleep disruption. However, the trial design of the ini-
tial studies investigating these modes did not provide an a priori setup strategy,
leading to differing delivered pressure support between the standard pressure
support and the volume-targeted pressure support. More recent data have dem-
onstrated equivalence of the volume-targeted modes compared with standard
pressure support modes.
A scheduled programme of follow-up is required with a particular focus on adher-
ence to home ventilatory support. An average of 4 h per night is recommended
to achieve a reduction in daytime carbon dioxide levels. Indeed, the reduction in
daytime carbon dioxide is correlated with improvements in other important out-
comes, including health-related quality of life, excessive daytime somnolence and
physical activity. An early clinical review within the first 2 months after initiation
of home ventilation provides an opportunity to optimise the interface application
and machine setup based on the patient reports, physiological measurements
and adherence time to ventilation. With patients showing initial improvement on
early review, subsequent follow-up at 3–6 months should be performed to con-
firm stability prior to extended follow-up or step-down of therapy. Furthermore,
obese patients with chronic respiratory failure and cor pulmonale, who initially
require daytime and nocturnal oxygen therapy in addition to NIV, need an oxygen
assessment at each scheduled appointment. Oxygen therapy can potentially be
deleterious in obese patients with respiratory failure if provided without venti-
latory support. Furthermore, it may not be required in the long term as hypox-
aemia improves with correction of sleep disordered breathing causing resolution
of pulmonary hypertension. In addition, patients showing clinical stability at
3–6 months should have their PaCO2 measured to determine if a transition from
NIV to CPAP can be considered based on normalisation of PaCO2 and the under
lying phenotype of sleep disordered breathing. Subsequent longer term follow-up
allows for adjustments of the ventilator settings in response to patient-reported
symptoms or an increase or decrease in weight, as well as for maintenance of the
ventilator equipment and interface renewal.
Further reading
• Fanfulla F (2010). Pathophysiology of respiratory failure in obesity. In: Elliott M,
et al., eds. Non-Invasive Ventilation and Weaning: Principles and Practice. 1st Edn.
London, Hodder Arnold; pp. 399–407.
• Mokhlesi B (2010). Obesity hypoventilation syndrome: a state-of-the art review.
Respir Care; 55: 1347–1365.
• Road R, et al. (2011). Section VII: HMV for patients with obesity hypoventilation syn-
drome. In: Home Mechanical Ventilation: A Canadian Thoracic Society Clinical Practical
Guideline. Canadian Thoracic Society; pp. 91–98. www.respiratoryguidelines.ca/
2011-cts-guideline-hmv

Long-term NIV
• Murphy PB, et al. (2014). Outcomes for obese patients with chronic respiratory
failure: results from observational and randomised controlled trials. Clin Sleep Med;
9: 349–356.
• Piper AJ, et al. (2011). Obesity hypoventilation syndrome: mechanisms and man-
agement. Am J Respir Crit Care Med; 183: 292–298.

Long-term NIV
Chronic NIV in bronchiectasis,

CF and interstitial lung disease
Amanda J. Piper
In contrast to the extensive literature regarding NIV use in restrictive chest wall
disorders and COPD, data from controlled trials regarding benefits and outcomes
of long-term NIV in patients with severe CF, bronchiectasis and especially inter-
stitial lung diseases (ILDs) are scarce. Most published data come from studies in
CF, with the principles of initiating and titrating therapy being generalised from
this patient group to the other lung conditions. Nevertheless, with careful patient
selection and appropriate monitoring, NIV can be a valuable clinical tool in the
management of these disorders.
Structural damage to the parenchyma and alterations in lung volumes in these dis-
eases leads to a high work of breathing. With progressive lung damage, a more rapid
and shallow breathing pattern develops. While this strategy reduces the load on
the respiratory muscles, it also affects gas exchange, eventually leading to alveo-
lar hypoventilation, hypoxaemia and carbon dioxide retention. Dyspnoea, exercise
intolerance and poor sleep arising as a consequence of progressive respiratory
insufficiency significantly impact on quality of life. Although there are significant
differences between the pathological processes producing lung damage and altered
mechanics in these disorders, the goals of NIV in each condition are the same:
• to stabilise the patient’s respiratory status
• to alleviate symptoms associated with sleep disordered breathing and
hypoventilation
• if possible, to slow the natural course of the disease progression
Key points
• In these progressive lung disorders, the goal of NIV is to
stabilise the patient’s respiratory status, relieve symptoms and
improve quality of life.
• Bilevel therapy is generally the first-choice NIV strategy in
those with hypoventilation.
• Sufficient pressure support needs to be applied to achieve the
goals of therapy.
• Monitoring of response to therapy is important to ensure
expected outcomes are achieved.

Long-term NIV
Table 1. Indications and possible outcomes from chronic NIV
Indication Goals and possible outcomes

To improve sleep Improved gas exchange during sleep
disordered breathing
Improved sleep quality
Reduced or stabilised diurnal carbon dioxide
Relief of sleep and hypoventilation symptoms
Improved quality of life
Rate of lung function decline stabilised
As a bridge to transplant Stabilised respiratory status

Widened transplant window
Prolonged survival
To improve tolerance Increased tolerance to longer treatment sessions

to airway clearance
Reduced therapy-related fatigue
treatment
Prevention of oxygen desaturation
To augment exercise Increased exercise intensity to achieve a training effect

training
Increased exercise endurance time
Reduced dyspnoea
The most common reasons to consider NIV are outlined in table 1.
Treating sleep disordered breathing

Poor-quality sleep is common in patients with CF, bronchiectasis and ILD, and
arises from a number of factors including: abnormal sleep breathing, poor gas
exchange, high work of breathing and cough.
Patients with more severe lung disease are more likely to experience sleep hypoxaemia
and hypoventilation. However, lung function poorly predicts sleep disordered breath-
ing. Symptoms of sleep disordered breathing are nonspecific, and may not be iden-
tified until after therapy is established and the patient’s clinical condition improves.
Upper-airway obstruction
• While OSA is rarely seen in teenagers or adults with CF, it has been occasionally
reported in the paediatric CF population, and should be considered in patients
with snoring and adenotonsillar hypertrophy.
• Where lung disease is complicated by tracheobronchomalacia, CPAP can splint
the airway open. This will reduce the work of breathing, improve expiratory flow
and assist in sputum clearance.
• OSA appears to be prevalent among ILD patients. In conjunction with sleep
hypoxaemia, OSA is thought to worsen symptoms of fatigue and contribute to
poor quality of life. Age and increasing BMI could promote OSA in patients with
bronchiectasis, although no such link has been established.
Long-term NIV
• CPAP is the therapy of choice if significant OSA is present. Pressure is titrated

to prevent upper-airway collapse during sleep.
• The goal is to improve nocturnal gas exchange, relieve symptoms such as
sleepiness and improve quality of life.
Sleep hypoxaemia and hypoventilation

• Patients with severe lung damage are predisposed to significant hypox-
aemia and hypoventilation, especially during rapid eye movement (REM)
sleep.
• Daytime hypercapnia is a common trigger to consider NIV. However, nocturnal
hypoventilation generally precedes this and is seen as a more appropriate time
to commence therapy.
• Nocturnal supplemental oxygen alone corrects hypoxaemia related to ventila-
tion–perfusion mismatch but will aggravate hypoventilation and promote fur-
ther diurnal carbon dioxide retention.
In order to identify hypoventilation during sleep, monitoring of nocturnal car-
bon dioxide is necessary. This may be achieved by measuring paired evening
to morning arterial or capillary blood gases, or through continuous monitoring
of PtcCO2 during sleep. End-tidal carbon dioxide measurements are not useful
in identifying sleep hypoventilation in patients with lung disease using oxygen
therapy.
By assisting inspiratory efforts and augmenting VT, NIV reduces the work of
breathing and prevents or limits nocturnal carbon dioxide retention. Patients with
ILD are more challenging to ventilate effectively due to lung stiffness from fibrotic
changes in the lung parenchyma and frequent nocturnal cough. Nevertheless, a
trial of NIV should be considered in selected individuals to determine whether any
benefits in terms of gas exchange, symptom relief or quality of life are achievable.
Bridging to transplant
NIV is commonly commenced during an acute exacerbation and considered for
home therapy when respiratory function fails to return to baseline levels or diurnal
carbon dioxide remains raised. NIV provides supportive therapy for those await-
ing transplants, with the aim of stabilising their condition in order to widen their
transplant window. Although NIV has also been used as a bridge to transplant
in ILD patients, survival times are significantly shorter than those reported for
patients with obstructive lung diseases.
Some centres routinely consider NIV when patients are listed for transplant,
although there is no evidence that this necessarily improves outcomes. Awake
carbon dioxide levels may not necessarily fall with NIV. However, there are sev-
eral reports documenting a stabilisation of lung function decline with long-term
NIV use.
Devices and settings

• Bilevel ventilation is usually the first choice of therapy due to its simplicity,
comfort and availability.
• A spontaneous mode of support is often sufficient as central respiratory drive
is usually not an issue.
Long-term NIV
• Inspiratory triggering and expiratory cycling sensitivities need to be appropri-

ately set to optimise synchrony between the patient and the ventilator.
• A spontaneous-timed mode with a sufficiently high backup rate to minimise
patient triggering can further reduce inspiratory effort, if required.
• Close attention to patient–ventilator synchrony is important. Attempts by the
patient to exhale while the machine is still maintaining high inspiratory pres-
sure is not only uncomfortable but will be detrimental to lung mechanics.
• Oxygen entrainment is generally needed to maintain SpO2 in the target range.
While there are some data around approaches to the set-up and titration of NIV
in patients with CF, this is not the case for patients with bronchiectasis or ILD.
Consequently, NIV is generally initiated in the same manner in all three condi-
tions, with daytime practice sessions undertaken to accustom the patient to ther-
apy. Settings are initially based on patient comfort, with EPAP set at 4–5 cmH2O
and IPAP set to the maximum tolerated. Further adjustments are then guided by
improvement in breathing pattern, gas exchange and comfort. The goals of titra-
tion are to normalise or at least reduce carbon dioxide levels and to relieve symp-
toms. In patients with CF, high inspiratory pressures are often required to achieve
effective ventilation. Once the patient is able to tolerate therapy comfortably dur-
ing the day, nocturnal use is commenced.
Polysomnography has been used to titrate and optimise settings. Large uninten-
tional leaks, inconsistent triggering or inadequate pressure support during REM
sleep can adversely affect gas exchange and sleep quality, ultimately impacting
on therapy compliance. These events cannot be identified from daytime clinical
evaluation alone and require nocturnal monitoring. Recent advances in home
ventilator software and internal monitoring capabilities mean that more detailed
data regarding the effectiveness of nocturnal ventilation can be obtained from the
ventilator itself. These data can be augmented by portable devices that monitor
PtcCO2 and SpO2.
An appropriately fitting interface that provides a good seal without excessive pres-
sure on the skin should be used. For sleep, a nasal mask may provide greater com-
fort and ease for coughing or expectorating secretions during the night. However,
in patients with CF, there is a high occurrence of sinus disease and nasal polyps,
which may increase nasal resistance, and make nasal masks less tolerable and
effective. Oronasal masks are an acceptable alternative.
As an adjunct to airway clearance

There is no evidence that the routine use of NIV during chest physiotherapy in stable
patients with CF or bronchiectasis offers clinical advantages over more traditional
airway clearance techniques (ACTs). However, in patients experiencing difficulties
tolerating ACTs due to breathlessness, fatigue or oxygen desaturation during treat-
ment, NIV integrated into therapy sessions can be beneficial. Patients with more
severe disease or respiratory muscle impairment are likely to benefit most.

• Although CPAP and pressure support can also effectively assist ACTs, bilevel
ventilation is the preferred technique, when available. It incorporates the bene
fits of CPAP (positive pressure during expiration to minimise dynamic airway
Long-term NIV
collapse) with the advantages of pressure support (unloading of the inspiratory

muscles and increasing VT).
• Initial concerns about NIV and CPAP potentially causing mucus impaction in
the peripheral airways or interfering with cough and expectoration have proven
to be unfounded.
Bilevel settings are adjusted both to achieve patient comfort and to maximise
the level of pressure support delivered. EPAP is usually set to 4–5 cmH2O, with
IPAP set to the maximum tolerated. A set level of pressure support may be used
throughout the therapy session. However, some patients will benefit from using a
slightly lower level of pressure support during the rest/breathing control periods of
treatment, with pressure support increased by 4–5 cmH2O during thoracic expan-
sion periods. NIV is removed for huffing, coughing and expectoration, and then
reintroduced to assist with recovery following coughing.
A mouthpiece or nasal mask is frequently chosen for airway clearance as these
make it easier to remove the interface for coughing and expectoration. Oxygen is
added to the circuit as needed to maintain SpO2 between 92% and 95%. ACTs are
further discussed in the section entitled “Airway clearance methods and nebulised
therapy in acute NIV”.
NIV-assisted exercise training

Patients with obstructive and restrictive lung disorders can present with severe
breathlessness and peripheral muscle deconditioning, which significantly impair
exercise tolerance. NIV-assisted exercise training can improve exercise tolerance
(table 1). Benefits are most likely to be seen in those with more severe lung dis-
ease. However, the response to NIV-assisted exercise is highly variable. A short
trial of NIV during exercise should be compared with exercise duration or intensity
during spontaneous breathing to determine whether benefits are achieved.

• Bilevel support is the therapy of choice for NIV-assisted exercise training.
• The specific ventilator used must have sufficient motor performance to sup-
ply adequate flow rates and pressurisation during peak exercise, along with
appropriate trigger and cycling sensitivities. Many home ventilators used for
nocturnal ventilation do not meet these specifications.
• Sufficiently high levels of pressure support need to be used to adequately
unload the inspiratory muscles. The level of pressure support required may vary
depending on whether cycling, treadmill walking or ground walking activities
are being performed.
• A mouthpiece or oronasal mask is better suited for NIV-assisted exercise train-
ing to accommodate mouth-breathing.
Additional considerations
Humidification High inspiratory pressures and leaks can produce airway mucosal
drying during NIV. This can be uncomfortable and may also impair mucociliary
clearance. The addition of heated humidification should be considered for patients
who already have thick, tenacious sputum. Patients with ILD and nocturnal
cough may also benefit from the use of humidification (see the section entitled
“Supplemental oxygen and humidification”).
Long-term NIV
Box 1. Common problems and barriers encountered when introducing NIV

• Intolerance of the interface
• Difficulty sleeping with NIV
• Inability to tolerate effective pressures
• Lack of perceived benefit
• Patient anxiety/fear
• Treatment is too constraining or burdensome
• Psychosocial factors
Serious complications Pneumothorax and pneumomediastinum occur spontan

eously in patients with structural lung diseases such CF and ILD. These complica-
tions may arise coincident to or as a consequence of NIV. NIV can be continued
once the pneumothorax is being adequately drained by a patent intercostal tube
and close monitoring has been instituted. With haemoptysis, especially major
bleeds, NIV should be ceased until the bleeding is controlled and the patient is
clinically stable. Therapy is recommenced once it is judged to be clinically safe to
do so. In both situations, it may be necessary to initially reduce inspiratory pres-
sures to gauge the patient’s response to the reinstitution of therapy. Patients
using NIV should be made aware of these potential problems.
Failure of therapy A number of problems and barriers to effective NIV use can arise
(box 1). Identifying the reason for poor tolerance or acceptance of therapy will aid
in introducing strategies to improve use.
Further reading
• Bright-Thomas RJ, et al. (2014). What is the role of noninvasive ventilation in cystic
fibrosis? Curr Opin Pulm Med; 20: 618–622.
• Contal O, et al. (2012). Monitoring of noninvasive ventilation by built-in software
of home bilevel ventilators: a bench study. Chest; 141: 469–476.
• Fauroux B (2010). Noninvasive ventilation in cystic fibrosis. Expert Rev Respir Med;
4: 39–46.
• Holland AE, et al. (2003). Non-invasive ventilation assists chest physiotherapy in
adults with acute exacerbations of cystic fibrosis. Thorax; 58: 880–884.
• Lima CA, et al. (2014). Effects of noninvasive ventilation on treadmill 6-min walk
distance and regional chest wall volumes in cystic fibrosis: randomized controlled
trial. Respir Med; 108: 1460–1468.
• Mermigkis C, et al. (2015). Obstructive sleep apnea should be treated in patients
with idiopathic pulmonary fibrosis. Sleep Breath; 19: 385–391.
• Moderno EV, et al. (2010). Effects of proportional assisted ventilation on exercise
performance in idiopathic pulmonary fibrosis patients. Respir Med; 104: 134–141.
• Moran F, et al. (2013). Non-invasive ventilation in cystic fibrosis. Cochrane Database
Syst Rev; 4: CD002769.

Long-term NIV
• Piper AJ, et al. (2009). Noninvasive ventilation as an adjunct to exercise training in

patients with chronic respiratory disease. Breathe; 5: 334–345.
• Vianello A, et al. (2014). Noninvasive ventilation in the event of acute respiratory
failure in patients with idiopathic pulmonary fibrosis. J Crit Care; 29: 562–567.
• Young AC, et al. (2008). Randomised placebo controlled trial of non-invasive ven-
tilation for hypercapnia in cystic fibrosis. Thorax; 63: 72–77.

Long-term NIV
Chronic NIV in heart failure patients:

ASV, NIV and CPAP
João C. Winck, Marta Drummond, Miguel Gonçalves and Tiago Pinto
Sleep disordered breathing (SDB), including OSA and central sleep apnoea (CSA), is
common and frequently undiagnosed in patients with congestive heart failure (CHF).
The prevalence of SDB is estimated to be as high as 73% among those with stable
CHF receiving optimised pharmacological therapy. The proportion of CSA and OSA
is variable and the two conditions may coexist, associated with Cheyne–Stokes
respiration (CSR); sometimes, there is an overnight shift from OSA to CSA.
CSR comprises a crescendo–decrescendo pattern of VT followed by central
apnoeas. OSA is characterised by periodic airway collapse during sleep, associ-
ated with increased respiratory effort, that causes repetitive episodes of oxygen
desaturation and arousals.
Key points
• Patients with congestive heart failure (CHF) can suffer from a
mixture of central sleep apnoea (CSA), OSA and Cheyne–Stokes
respiration (CSR).
• The proportion of each of the components may vary with time,
during the night and with different body positions.
• Sleep disordered breathing may significantly impact clinical
outcomes in CHF.
• CPAP therapy may be used in patients with chronic heart
failure and OSA, but is not generally effective in chronic heart
failure patients with CSA/CSR.
• Early trials and a meta-analysis showed ASV benefited heart
failure patients with CSA/CSR, but initial results from the
Serve-HF multicentre RCT of ASV in chronic systolic heart
failure patients with predominantly central sleep apnoea
showed an excess of mortality in the ASV group compared with
control. Consequently, ASV is no longer recommended in heart
failure patients with CSA and ejection fraction <45%.

Long-term NIV
There is evidence that central sleep apnoea with Cheyne–Stokes respiration (CSA/
CSR) is an indicator of higher morbidity and mortality in CHF patients. Moreover,
a recent meta-analysis showed that positive airway pressure modalities like ASV
reduce all-cause mortality in patients with chronic heart failure with concomitant
SDB. However, preliminary results from the Serve-HF (Treatment of Predominant
Central Sleep Apnoea by Adaptive Servo Ventilation in Patients With Heart Failure)
study suggest that ASV may increase the risk of sudden death in systolic heart failure
patients (those with left ventricular ejection fraction (LVEF) <45%), with predomi-
nantly central sleep apnoea, so it cannot currently be recommended in this situation.
Management of SDB in CHF
The first and most important step in the management of SDB in CHF should be
optimisation of CHF treatment in accordance with published guidelines (i.e. diuretics,
angiotensin-converting enzyme inhibitors and beta-blockers). Other conservative
measures effective in OSA, e.g. weight reduction, avoiding a supine position dur-
ing sleep and avoiding alcohol and sedatives before sleep, should also be applied.
Therefore, treatment of heart failure in its own right is likely to improve SDB.
The next step is to characterise the nature of SDB. At present, CPAP is recom-
mended in patients with OSA or predominantly obstructive sleep apnoea (>50%
obstructive events). Nocturnal CPAP in patients with OSA has been shown to
improve LVEF and quality of life.
Bilevel positive airway pressure ventilation may be more effective than CPAP in
some cases, but there have been no long-term trials and care should be taken. It
should be indicated for nonresponders to CPAP (as defined by the persistence of
nocturnal hypoventilation or CSA with AHI ≥15 events·h−1 on CPAP). This situation
may occur in patients with comorbidities, such as in individuals with COPD and
heart failure or those with OHS and heart failure. Bilevel positive airway pressure
ventilation is probably particularly relevant for comorbid patients in whom noc-
turnal hypercapnia is present (figures 1–3). Notably, it is unlikely to have a role in
patients with heart failure and CSA for whom episodic hypocapnia can be seen as
a consequence of the hyperventilation component of CSR.
Predominantly central sleep apnoea The Canadian Positive Airway Pressure Trial
included patients with CSA and heart failure who were randomised to either CPAP
or no CPAP. After 2 years of follow-up, the CPAP arm showed no benefit in terms
of the primary outcome of transplant-free survival, although CSA, ejection fraction
and exercise capacity were slightly enhanced. In fact, CPAP improves CSA/CSR in
only 50% of cases; so it is suggested that CPAP may improve survival if titrated to
achieve a therapeutic reduction in AHI, which is normally considered to be an AHI
<15 events·h−1.
ASV is a new positive airway pressure technology that adjusts the delivered pres-
sure support according to the ventilation of the patient. Since its introduction in
2001, a series of trials have been published showing that, compared with the other
positive airway pressure systems, ASV reduces AHI and systemic inflammation and
improves LVEF and quality of life in CHF patients with CSA/CSR, but none of these
studies was designed to examine long-term impact on mortality and morbidity.
Two RCTs are currently underway to assess the impact of ASV on long-term out-
comes of CHF patients with SDB. The first, Serve-HF, is a multicentre trial of ASV
Long-term NIV
Device flow 0
L·min-1
Thorax 0
Abdomen 0
SpO2 % 75
Pulse 100
beats·min-1 50
Device leak
25
L·min-1
Pressure -7.5
cmH2O
VT L 1
0.5
PtCO2 50
mmHg 40
11:45 h 11:46 h 11:47 h 11:48 h 11:49 h 11:50 h 11:51 h 11:52 h 11:53 h 11:54 h
Figure 1. Sleep study of a 64-year-old patient with chronic heart failure and nocturnal
hypoventilation. Noninvasive bilevel positive airway pressure ventilation in spontaneous-
timed titration mode was started to correct hypoventilation-related respiratory events and
the patient started to present CSR. Although inspiratory and expiratory ventilatory settings
were optimised and backup rate was adjusted, the patient maintained the same pathological
respiratory pattern.
versus standard therapy in systolic heart failure patients with ejection fraction
<45% and AHI >15 events·h−1 with events being predominantly central sleep
apnoea. The second is Advent-HF, which is recruiting heart failure patients with
OSA (AHI >15 events·h−1 with >50% obstructive events) who are not sleepy
(Epworth Sleepiness Scale score <10) and CSA (AHI >15 events·h−1 with >50%
CSA) with no preset criteria for sleepiness.
Importantly, preliminary results from the Serve-HF trial suggest that ASV may
increase the incidence of sudden death (despite controlling SDB and, in some
patients, providing symptom relief). Pending full results and further evaluation,
ASV cannot be recommended in patients with systolic heart failure and predomin
antly central sleep apnoea, and should be withdrawn in those that are using it. For
further information see: Serve-HF ASV safety alert FAQs (www.resmed.com/us/
en/serve-hf.html#Frequently-Asked-Questions).
These results do not apply to patients with heart failure and CSA whose ejection
fraction is >45%, or to heart failure patients with predominantly obstructive sleep
apnoea. Further advice for management of SDB in heart failure should follow from
sub-study analysis in the Serve-HF trial and from the Advent-HF trial, once completed.
Long-term NIV
Device flow
0
L·min-1
Thorax 0
Abdomen 0
SpO2 % 75
Pulse 100
beats·min-1 50
Device leak
20
L·min-1
Pressure
-10
cmH2O
VT L 1
0.5
PtCO2 50
40
mmHg
11:52h 11:53 h 11:54 h 11:55 h 11:56 h 11:57 h 11:58 h 11:59 h 12:00 h 12:01 h
Figure 2. In the same patient as figure 1, CSR events were not corrected with standard
bilevel positive airway pressure ventilation mode. ASV was applied (from the vertical line)
with an auto-backup rate. Maximum and minimum levels of EPAP were adjusted to assure
the optimal patency of the upper airway and correct obstructive events. Maximum and
minimum levels of pressure support were adjusted to the total correction of CSR events and
patient comfort. If the maximal value of pressure support was achieved and the patient
continued to present CSR, the level was increased by 2 cmH2O until total pattern correction.
PtcCO2 was monitored to evaluate safety/effectiveness of the ventilatory adjustments. After
ASV was optimised, the algorithm corrected all the events and the dysfunctional respiratory
pattern.
Titration of positive airway pressure systems

CPAP titration CPAP titration is started at 5 cmH2O. If obstructive apnoea–
hypopnea events persist, pressure is incremented in 1 cmH2O intervals to try and
eliminate them.
Bilevel positive airway pressure ventilation titration Kuzniar et al. (2007) provide the
following guidelines for bilevel positive airway pressure ventilation:
• Bilevel positive airway pressure ventilation titration is started with the EPAP set
at or near the CPAP level which abolished any obstructive apnoeas, hypopneas
and snoring.
• The IPAP is initially set 2–4 cmH2O above the EPAP. The EPAP is subsequently
titrated to eliminate residual apnoeas, while the IPAP is increased to eliminate
hypopnoeas and hypoventilation.
Long-term NIV
Device flow 0
L·min-1
Thorax 0
Abdomen 0
SpO2 % 75
Pulse 100
beats·min-1 50
Device leak
25
L·min-1
Pressure -20
-15
cmH2O
VT L 1
0.5
PtCO2 50
40
mmHg
16:04 h 16:06 h 16:08 h 16:10 h 16:12 h 16:14 h 16:16 h 16:18 h 16:20 h 16:22 h 16:24 h 16:26 h 16:28 h 16:30 h 16:32 h
Figure 3. Algorithm of ASV working to correct CSR, by increasing and decreasing the levels
of pressure support.
• The IPAP and/or EPAP are adjusted by 1 cmH2O at 1- to 2-min intervals while
trying to maintain an IPAP–EPAP difference of at least 2–3 cmH2O until stable
sleep and breathing is achieved. Maximum IPAP and EPAP should be no greater
than 20 and 15 cmH2O, respectively.
• For bilevel positive airway pressure ventilation in spontaneous-timed mode,
the respiratory backup rate is set at or slightly below the patient’s spontaneous
awake respiratory rate (usually 10–14 breaths⋅min−1).
Further reading
• Artzt M, et al. (2007). Suppression of central sleep apnea by continuous posi-
tive pressure airway pressure and transplant-free survival in heart failure: a post
hoc analysis of the Canadian Continuous Positive Airway Pressure for patients
with central sleep apnea and heart failure trial (CANPAP). Circulation; 115:
3173–3180.
• Aurora RN, et al. (2012). The treatment of central sleep apnea syndromes in
adults: practice parameters with an evidence-based literature review and meta-
analyses. Sleep; 35: 17–40.
• Bradley TD, et al. (2005). Continuous positive airway pressure for central sleep
apnea and heart failure. N Engl J Med; 353: 2025–2033.

Long-term NIV
• Ferreira S, et al. (2010). Prevalence and characteristics of sleep apnoea in patients

with stable heart failure: results from a heart failure clinic. BMC Pulm Med; 10:9.
DOI: 10.1186/1471-2466-10-9.
• Franklin KA, et al. (1997). Reversal of central sleep apnea with oxygen. Chest; 111:
163–166.
• Kaneko Y, et al. (2003). Cardiovascular effects of continuous positive airway pres-
sure in patinents with heart failure and obstructive sleep apnoea. N Engl J Med;
348: 1233–1241.
• Kuzniar TJ, et al. (2007). Moving beyond empiric continuous positive airway pres-
sure (CPAP) trials for central sleep apnea: a multi-modality titration study. Sleep
Breath; 11: 259–266.
• Mansfield DR, et al. (2004). Controlled trial of continuous positive airway pres-
sure in obstructive sleep apnoea and heart failure. Am J Respir Crit Care Med; 169:
361–366.
• Nakamura S, et al. (2015). Impact of sleep-disordered breathing and efficacy of
positive airway pressure on mortality in patients with chronic heart failure and
sleep-disordered breathing: a meta-analysis. Clin Res Cardiol; 104: 208–216.
• Pasquina P, et al. (2012). What does built-in software of home ventilators tell
us? An observational study of 150 patients on home ventilation. Respiration; 83:
293–299.
• Roebuck T, et al. (2004). Increased long-term mortality in heart failure due to sleep
apnoea is not yet proven. Eur Respir J; 23: 735–740.
• Sharma BK, et al. (2011). Sleep disordered breathing in patients with heart fail-
ure: pathophysiology and management. Curr Treat Options Cardiovasc Med; 13:
506–516.
• Sharma BK, et al. (2012). Adaptive servoventilation for treatment of sleep-
disordered breathing in heart failure: a systematic review and meta-analysis.
Chest; 142: 1211–1221.
• Teschler H, et al. (2001). Adaptive pressure support servo-ventilation: a novel
treatment for Cheyne-Stokes respiration in heart failure. Am J Respir Crit Care Med;
164: 614–619.

Long-term NIV
Chronic NIV in children: indications,

outcomes and transition
Hui-Leng Tan and Anita K. Simonds
In the past few decades, the number of children on long-term NIV at home
has increased exponentially. The factors that have contributed to this increase
include improved medical management of and, thus, survival in conditions
such as bronchopulmonary dysplasia and metabolic conditions; improvements
in NIV technology; and establishment of care pathways for the implementation
of home care packages. These have enabled and accompanied a profound
alteration in attitudes towards the role of home NIV, leading to its increased
use under a far wider range of conditions. Furthermore, studies in selected
patient groups where the use of home NIV has become more common, such
as those with neuromuscular disease, have shown convincing improvement in
outcomes.
Long-term positive pressure respiratory support can be delivered in the form
of CPAP or NIV (usually bilevel pressure support). Here, the indications and out-
comes of both are discussed, with particular emphasis on OSA and neuromuscu-
lar disease, due to their prevalence. The section concludes with a discussion of
transition, an area that, given the improving outcomes, is an increasingly impor-
tant aspect of management. Table 1 summarises the medical conditions in which
long-term CPAP and NIV are commonly used.
CPAP
The most common indication for home CPAP is in the treatment of OSA. Here,
promoting airway patency is the key mechanism of action as, technically, CPAP
is not a form of ventilation, as it does not augment minute ventilation. Paediatric
OSA is now recognised as a common condition, with reported prevalence ranging
from 1% to 5% depending on the study population and stringency of diagnostic
Key points
• The use of NIV, particularly in neuromuscular diseases such
as Duchenne muscular dystrophy, has dramatically improved
survival and changed the natural course of the disease.
• The number of children on NIV has increased significantly over
the past two decades, which has significant implications for
the planning of transition to adult care facilities.

Long-term NIV
Table 1. Common medical conditions in which long-term NIV is used
OSA
Airway malacia
Nocturnal hypoventilation
Neuromuscular weakness
Duchenne muscular dystrophy
Spinal muscular atrophy
Congenital myasthenia syndrome
Congenital myopathy
Congenital muscular dystrophies including Rigid spine syndrome
Chest wall deformity
Scoliosis
OHS
Restrictive and obstructive lung conditions
Chronic lung disease of prematurity
Interstitial lung disease
CF
Central alveolar hypoventilation syndromes
Congenital central hypoventilation syndrome
ROHHAD (rapid-onset obesity with hypothalamic dysfunction, hypoventilation
and autonomic dysregulation)
Apnoea of prematurity
Neurometabolic syndromes
Arnold Chiari malformation
Acquired
Brain tumours
Encephalitis
Cerebral infarction
Following head trauma/brain surgery
criteria. Untreated, it can result in significant morbidities, affecting cardiovascular

and neurocognitive outcomes and metabolic regulation, the prevention of which
necessitates appropriate intervention.
The aetiology of paediatric OSA can be broadly divided into factors that result in
reduced upper airway calibre and those that promote upper airway collapsibility.
Adenotonsillar hypertrophy is the most common cause of paediatric OSA. Other
anatomical factors include: a small or retro-positioned mandible, macroglossia
and midface hypoplasia. Thus, children with craniofacial syndromes such as
Long-term NIV
Pierre Robin sequence, achondroplasia, trisomy 21, mucopolysaccaridoses and

Beckwith–Wiedemann syndrome are all at risk of OSA. Obesity is another increas
ingly common risk factor.
The presence of upper airway inflammation and altered neurological reflexes of
the upper airway muscles are the most prominent drivers leading to increased
upper airway collapsibility. This explains why children with cerebral palsy, neuro
muscular disorders, myelomeningocoele and other causes of hypotonia are at
increased risk of OSA.
In children with adenotonsillar hypertrophy, adenotonsillectomy is the first-line
therapy for paediatric OSA and is an effective treatment. However, residual OSA
can still persist, particularly in children who are obese, have severe OSA pre-
surgery, or have other comorbidities such as craniofacial anomalies, chromosomal
defects and neuromuscular disease.
In children who do not have significantly enlarged tonsils or adenoids, or who
manifest residual OSA post-adenotonsillectomy, CPAP is recommended. The goals
of CPAP are:
• to maintain patency of the airway throughout the respiratory cycle
• to improve functional residual capacity
• to decrease the work of breathing
In the majority of children with OSA, CPAP is effective. However, if nocturnal carbon
dioxide is significantly raised, particularly in children with coexisting conditions
such as neuromuscular disease, craniofacial syndromes or obesity hypoventila-
tion, bilevel ventilation may sometimes be needed. A trial of bilevel ventilation
may also be considered in children who do not tolerate CPAP due to excessively
high PEEP requirements, as occasionally, it may be better tolerated.
Bilevel ventilation
Bilevel ventilation is used to treat children with chronic respiratory failure. This
is now the dominant mode of ventilation, although in early studies, several dec-
ades ago, volume-preset ventilators were predominantly used. Chronic respiratory
failure occurs when the central ventilatory drive and respiratory muscle capacity
are insufficient to support the respiratory load. This may be a result of decreased
respiratory muscle capacity with normal respiratory load (e.g. neuromuscular con-
ditions), normal respiratory muscle capacity with increase in respiratory load (e.g.
CF or interstitial lung disease) or abnormal central respiratory drive (e.g. congenital
central hypoventilation syndrome (CCHS)).
The main aim in starting bilevel ventilation is, therefore, to treat daytime or night-
time hypoventilation (fig. 1). However, in children with neuromuscular disease,
other advantages of bilevel ventilation include a reduction in the number of
respiratory exacerbations and the prevention of chest wall deformity in younger
children.
Neuromuscular disease An understanding of the natural history of the individual
neuromuscular diseases is important to determine which result in early ventila-
tory failure (e.g. spinal muscular atrophy (SMA) type 1, SMA with respiratory dis-
tress and X-linked myotubular myopathy); which have a slower, more progressive
course (e.g. Duchenne muscular dystrophy and congenital muscular dystrophy);
Long-term NIV
a)
10
PtcCO2 kPa
4
2
22:00 h 01:00 h 04:00 h
100
SpO2%
90
80
75
22:00 h 01:00 h 04:00 h
b)
15
PtcCO2 kPa
10
0
01:00 h 04:00 h 07:00 h
100
95
SpO2 %
90
85
80
01:00 h 04:00 h 07:00 h
Figure 1. a) Example of sleep disordered breathing in a child with SMA type 2. Note the
clusters of desaturations associated with an increase in carbon dioxide during periods of REM
sleep (arrows). b) Traces after NIV initiation.
and those in which ventilatory failure is rare (e.g. facioscapulohumeral muscular

dystrophy).
In the past, daytime hypercapnia was the main indication for starting NIV. A
seminal practice-changing paper was the RCT of neuromuscular disease patients
with nocturnal hypoventilation who had not yet developed daytime hypercapnia
(Ward et al., 2005). These patients were either started on NIV or monitored
(control group). In the control group, 70% and 90% required NIV within a
year and within 2 years, respectively, and several were acutely admitted for
ventilatory decompensation. This rate of progression, when combined with the
significantly improved health-related quality of life scores in the NIV group, has
led to a reappraisal of the timing of intervention and most centres now start NIV
once nocturnal hypoventilation develops, as this allows initiation in a planned,
controlled manner. Caveats of this trial are that the youngest patient was 8 years
Long-term NIV
of age and the aetiology of underlying conditions was mixed, but the principle
of starting NIV to correct nocturnal hypoventilation before the development of
diurnal ventilatory failure seems sound.
Better understanding of genotype–phenotype interactions has meant that per-
sonalised care plans can now be put in place. For example, among children with
congenital myasthenia syndrome, patients with COLQ mutations are now recog-
nised to have a progressive decline in their respiratory course and often need NIV
for nocturnal hypoventilation. In contrast, patients with RAPSN, CHRNE and CHAT
mutations are more prone to sudden apnoeas/crises. Their parents need to be
taught cardiopulmonary resuscitation including bag/mask ventilation and NIV is
often needed episodically, during these crises and when the children are tired.
British Thoracic Society guidelines recommend that assessment for sleep dis-
ordered breathing should be carried out at least annually in children who have
become non-ambulant because of progressive muscle weakness or who never
attained the ability to walk. This should also be the case in children who have a
vital capacity of <60% predicted, although it must be remembered that children
with rigid spine syndrome may have nocturnal hypoventilation despite near nor-
mal vital capacity. Children who have clinical signs of diaphragmatic weakness and
those who have symptoms of OSA or nocturnal hypoventilation should also be
assessed for sleep disordered breathing.
Peri-operative period While there is currently no evidence to suggest that starting

NIV in asymptomatic children who do not have nocturnal hypoventilation is of
benefit, it is often useful to familiarise children with NIV if elective surgery (scoliosis
surgery, orthopaedic surgery or insertion of gastrostomy) is planned, so that it can
be used in the peri-operative period.
Palliation Children with SMA type 1 have profound global weakness and respiratory
failure is the main cause of mortality. Previously, most did not survive past their sec-
ond birthday. Starting NIV in those who are less severely affected has been shown
to improve survival. In those who are severely affected (typically presenting before
3 months of age), a goal-targeted approach using NIV to palliate symptoms and
facilitate discharge may enable the child to spend some quality time with the family
at home and give the parents the opportunity to come to terms with the prognosis.
CCHS CCHS is a genetic condition caused by mutations in the PHOX2B gene.

The alveolar hypoventilation is most apparent in non-rapid eye movement (REM)
sleep, but can also be abnormal during REM sleep and wakefulness, to a milder
degree. Most children with CCHS are ventilated via tracheostomy in the first few
years of life. This ensures effective oxygenation and ventilation, and thus opti-
mal neurocognitive outcome, while minimising the risk of developing pulmon
ary hypertension. When they reach 6–7 years of age, children with the milder
phenotypes who only require ventilation when asleep may be changed over to
NIV. It is important to remember that these children do not display the normal
physiological responses to hypoxia and hypercapnia, and home oxygen saturation
monitoring is necessary. Some units also provide home carbon dioxide monitor-
ing. The American Thoracic Society recommends aiming for an end-tidal carbon
dioxide tension between 30 mmHg (4.00 kPa) and 50 mmHg (6.67 kPa) (ideally
35–40 mmHg (4.67–5.33 kPa)), and SpO2 ≥95%.
Long-term NIV
Figure 2. The morning after a successful NIV titration study.
Practical aspects of NIV initiation

Starting NIV in toddlers or preschoolers can be challenging. A multidisciplinary
team approach usually produces the best results and can mean that even a
screaming 2-year-old who initially will not have the mask anywhere near them
can, eventually, be gently persuaded to use NIV. This process can take weeks.
It is useful to include play specialists in the initial meeting who, through play,
make wearing the mask and “pretending to be an elephant/astronaut/Darth
Vader” fun and reduce anxiety. Typically, parents are then given the mask to go
home with, so the child can continue to play with it and familiarise themself
with it. Once the child is happy to wear the mask for brief periods of time when
awake, an inpatient admission can then be arranged to start NIV and titrate
pressures (fig. 2). Depending on the child, this may need to be over the course
of several nights. Do not underestimate the power of sticker charts and other
incentives – one of our greatest success stories hinged on the reward of a pink
bicycle. Parental involvement and education are crucial. Each child is differ-
ent and parents often know the strategy that will work best for their child. In
some children, it might be better for the parent to gently apply the mask after
the child has fallen asleep. While the majority of patients have NIV initiated in
hospital, in a carefully selected subgroup of stable patients, initiation in the
outpatient setting with intensive home monitoring has been shown to be feasi
ble, with equivalent outcomes in terms of improvement in gas exchange and
compliance.
In the more challenging cases, input from clinical psychologists with behav-
ioural modification strategies can sometimes result in breakthroughs. Time and
resources invested at the beginning can pay huge dividends later on. Once NIV
has been established as part of the daily routine, most children tolerate it sur-
prisingly well, particularly those who can feel a significant improvement in their
symptoms.
Long-term NIV
The American Academy of Sleep Medicine Noninvasive Positive Pressure Ventilation

Titration Task Force has published best practice recommendations (Berry et al.,
2010). They recommend a minimum starting IPAP of 8 cmH2O and EPAP of
4 cmH2O, with a gradual titration of pressures by 1–2 cmH2O. The maximum IPAP
recommended is 20 cmH2O in children younger than 12 years and 30 cmH2O for
those older than 12 years. The maximum PEEP recommended for children on CPAP
is 15 cmH2O for those younger than 12 years and 20 cmH2O for patients older
than 12 years. An open-minded approach adapted to the child is sensible, as no
guidelines are absolute.
Monitoring
It can be anticipated that pressure requirements will change over time and the
interface will need to be upsized and adjusted, due to the significant growth and
development that occurs in childhood. Most units repeat sleep studies at least
yearly, or more frequently to optimise settings if there are clinical indications.
Monitoring often needs to be more intensive around the time of puberty when the
child’s growth spurt can outstrip their lung growth and ventilatory requirements
can change more rapidly. Young children on NIV may develop flattening of the
midface or maxillary retrusion from the pressure of the mask on growing facial
structures and this should be monitored carefully.
Outcomes
The use of NIV in neuromuscular diseases such as Duchenne muscular dystrophy
has dramatically improved survival and changed the natural course of the dis-
ease. Outcomes of children on NIV for other causes of sleep disordered breathing
tend to depend on the prognosis of the underlying condition. In a recent review of
children on long-term NIV cared for at our hospital over the past 18 years, of the
449 children for whom follow-up data were available, 40% transitioned to adult
care, 24% died, 1% proceeded to tracheostomy ventilation and 9% discontinued
ventilatory support. The majority discontinued because of improvement in their
underlying condition, resulting in resolution of their sleep disordered breathing.
For example, children with chronic lung disease of prematurity, airway malacia,
etc. all improved with age; one CF patient had a lung transplant; and one child with
atypical nemaline rod myopathy required respiratory support until 5 years of age,
when the nocturnal hypoventilation resolved.
Transition to adult care

Increasing numbers of children on NIV are surviving into adulthood, which
has significant implications for the planning of transition and adult care facili-
ties. Well planned transition results in better clinical, educational and social
outcomes for the young person. It is important to remember that good transi-
tion is a process and not just a single event. There are several transition mod-
els, none of which has been shown to be clearly superior, so it is reasonable
to pragmatically adopt the model that best fits with how local healthcare is
organised. It is important to have close collaboration between children’s and
adult services. Ideally, there should be a period of preparation for the young
person and their family, information transfer with a comprehensive handover,
followed by monitoring of attendance until the young person is established in
adult services.
Long-term NIV
Successful transition depends on flexibility. To this end, it is important to keep in

mind the following points.
1) Treat the patients as the “young adults” they are: listen to what they have to
say; do not just talk to their parents. An offer to see them separately from their
parents for part of the consultation is useful.
2) Communicate at a level appropriate to their understanding.
3) Be prepared to negotiate; for example, one night off NIV so they can attend a
sleepover and not appear different from their peers may not be an unreason-
able compromise for consistent use on other nights.
4) Most teenagers appreciate a sense of humour.
5) Positive feedback and encouragement are always secretly welcome.
6) Know what you are doing: the most challenging questions often come from
adolescents, and they are acutely sensitive to doctors pretending to know the
answers.
7) A number of initiatives, such as Takin’ Charge (www.actionduchenne.org/
education/takin-charge), help young Duchenne patients start to take respon-
sibility for their health and wellbeing, and many young patients are experi-
enced networkers.
Further reading
• Amin R, et al. (2014). Pediatric long-term home mechanical ventilation: twenty
years of follow-up from one Canadian center. Pediatr Pulmonol; 49: 816–824.
• Berry RB, et al. (2010). Best clinical practices for the sleep center adjustment
of noninvasive positive pressure ventilation (NPPV) in stable chronic alveolar
hypoventilation syndromes. J Clin Sleep Med; 6: 491–509.
Child; 96: 426–432.
• Chatwin M, et al. (2013). Risk management protocol for gastrostomy and jejunos-
tomy insertion in ventilator dependent infants. Neuromuscul Disord; 23: 289–297.
• Fauroux B, et al. (2005). Facial side effects during noninvasive positive pressure
ventilation in children. Intensive Care Med; 31: 965–969.
• Kushida CA, et al. (2008). Clinical guidelines for the manual titration of positive airway
pressure in patients with obstructive sleep apnea. J Clin Sleep Med; 4: 157–171.
• Marcus CL, et al. (2012). Diagnosis and management of childhood obstructive
sleep apnea syndrome. Pediatrics; 130: 576–584.
• Michaud PA, et al. (2004). The adolescent with a chronic condition. Part II: health-
care provision. Arch Dis Child; 89: 943–949.
• Paulides FM, et al. (2012). Thirty years of home mechanical ventilation in children:
escalating need for pediatric intensive care beds. Intensive Care Med; 38: 847–852.

Long-term NIV
• Robb SA, et al. (2010). Respiratory management of congenital myasthenic syn-

dromes in childhood: workshop 8th December 2009, UCL Institute of Neurology,
London, UK. Neuromuscul Disord; 20: 833–838.
• Simonds AK, ed. (2007). Non-invasive respiratory support: a practical handbook.
3rd edn. London, Hodder Arnold.
• UK Department of Health, Child Health and Maternity Services Branch (2006).
Transition: getting it right for young people. Improving the transition of young peo-
ple with long term conditions from children’s to adult health services. London,
Department of Health.
• Wallis C, et al. (2011). Children on long-term ventilatory support: 10 years of pro-
gress. Arch Dis Child; 96: 998–1002.
• Ward S, et al. (2005). Randomised controlled trial of non-invasive ventilation (NIV)
for nocturnal hypoventilation in neuromuscular and chest wall disease patients
with daytime normocapnia. Thorax; 60: 1019–1024.
• Weese-Mayer DE, et al. (2010). An official ATS clinical policy statement: congenital
central hypoventilation syndrome: genetic basis, diagnosis, and management. Am
J Respir Crit Care Med; 181: 626–644.

Long-term NIV
Practicalities of and guide to cough

augmentation and daytime
mouthpiece ventilation
Michelle Chatwin
Cough is important for clearing mucus and debris from the upper airways into the
mouth, to be expectorated. An effective cough requires enough inspiratory mus-
cle strength to produce an inspiratory volume of around 2.3 L or 85–90% of TLC.
Intact bulbar function is required so that there is then rapid closure of the glottis
for approximately 0.2 s, followed by rapid opening of the glottis. This is coup
led with an abdominal contraction that produces a maximum expiratory pressure
>60 cmH2O. If one or more of these main components are impaired the cough
will be less effective. The expiratory airflow can be simply measured, using a peak
flow meter and mask (fig. 1), and is often referred to as peak cough flow (PCF).
Cough can be enhanced by performing assisted cough techniques, which should
be targeted towards whichever component of the cough is reduced. PCF can also
be used to evaluate whether assisted cough techniques increase cough strength.
A normal PCF is >400 L·min−1 and the cough is audible.
Patients who have decreased respiratory muscle strength will be unable to fully
expand the lungs to maximal capacity or compress them to residual volume, which
leads to a reduction in chest wall compliance. The reduction in chest wall compli-
ance is a result of shortening and stiffening of the unstretched tissue and respira-
tory muscles. The compliance of the lungs is also reduced by microatelectasis. This
reduction in lung compliance can lead to a decline in lung capacity and reduced
oxygen saturation on room air, and causes an impairment of the ability to cough.
Cough augmentation
An ineffective cough is usually limited to patients with respiratory muscle weak-
ness, for example patients with neuromuscular disease, spinal cord injury or
Key points
• Cough augmentation techniques are essential in patients with
neuromuscular disease.
• Peak cough flow can be used to assess cough efficacy.
• When peak cough flow is <160 L·min−1, mechanical
insufflation–exsufflation devices are recommended.
• Mouthpiece ventilation is an alternative to mask ventilation in
patients requiring daytime NIV.

Long-term NIV
Figure 1. Equipment for measurement of expiratory airflow: a low-flow mini EU scale peak
flow meter is attached to a full face mask. The patient is instructed to take a deep breath in
and cough into the face mask.
intensive care induced neuropathy. Evidence-based ways to enhance cough

strength in patients with respiratory muscle weakness are discussed in the fol-
lowing sections.
Inspiratory cough augmentation techniques
Inspiratory cough augmentation techniques will be effective in patients who
have a reduced PCF. When PCF drops below 270 L·min−1 mechanical and manual
assisted coughing are recommended. If the PCF is <160 L·min−1 patients are more
likely to benefit from a combination of techniques or a mechanical insufflation–
exsufflation device (cough machine).
Breath stacking The patient receives an assisted breath in and is told not to breathe
out. The manoeuvre is then repeated for a further one to four breaths. The aim
of this is to produce an augmented inspiratory vital capacity (VC) beyond that of
the patient's spontaneous VC. Once the patient achieves this, they are instructed
to cough, which results in an increased expiratory airflow and sheers secretions
towards the mouth. Devices that can be used to breath stack include a lung volume
recruitment bag (a resuscitation bag and mask with a one-way valve (Intersurgical
UK, Wokingham, UK)), a simple resuscitation bag or a volume-cycled ventilator
including mouthpiece ventilation. If the patient is unable to increase their VC beyond
their spontaneous VC using this technique they will have severe bulbar weakness
and assisted cough techniques will be limited or ineffective. When this technique is
measured objectively this value is called the maximum insufflation capacity (MIC).
Lung insufflation capacity This is the maximum passive lung insufflation volume
achieved. This technique can be performed using a manual resuscitator with a
closed expiratory port mimicking glottic closure, a mechanical insufflation–
exsufflation device, intermittent positive pressure breathing (IPPB), or a volume-
cycled ventilator at delivered volumes that approach predicted insufflation capacity.
The patient allows their lungs to be filled with a delivered breath. This delivered
breath aims to be deep enough to reach TLC. The patient is then instructed to
cough, which results in an increased expiratory airflow. In patients with complete
Long-term NIV
loss of glottic closure, where MIC no longer exceeds VC, lung insufflation can only
be provided by bypassing glottic function with a passive insufflation to achieve the
lung insufflation capacity.
Glossopharyngeal breathing For this technique a series of pumping strokes are
produced by the action of the lips, tongue, soft pallet, pharynx and larynx. Air is
held in the chest by the larynx, which acts as a valve when the mouth is opened
for the next breath. Expiration occurs via normal elastic recoil of the lungs and
rib cage. Glossopharyngeal breathing can be used to accumulate consecutive vol-
umes of air for lung volume recruitment. Patients can also use glossopharyngeal
breathing to increase their VC to achieve a higher PCF.
Expiratory cough augmentation techniques
Expiratory cough augmentation techniques will be effective in patients who have
a reduced PCF. These have been shown to be effective in patients with a PCF of
around 245 L·min−1. As stated earlier, if the PCF is <160 L·min−1 patients are more
likely to benefit from a combination of techniques or a mechanical insufflation–
exsufflation device.
Manually assisted cough A manually assisted cough (MAC) increases expira-
tory airflow by either compression of the chest wall or abdomen. Synchronous
compression of the abdomen causes a sudden increase in abdominal pressure.
This increased pressure causes the abdominal contents to push the diaphragm
upwards, increasing expiratory airflow. The increase in expiratory airflow assists in
moving airway secretions towards the mouth. MAC is a simple effective technique
that can be used anywhere. Some patients are able to use their arms to perform
their own MAC. A thoracoabdominal assisted cough will produce the greatest
increase in expiratory airflow (fig. 2).
Inspiratory and expiratory cough augmentation techniques
Breath stacking with a MAC By combining inspiratory and expiratory cough aug-
mentation techniques PCF can be further enhanced and this approach should be
used if use of the single technique does not improve cough efficacy. However, if
combining the techniques fails to significantly improve PCF or subjective cough
efficacy, patients are likely to benefit from a mechanical insufflation–exsufflation
device.
Mechanical insufflation–exsufflation Mechanical insufflation–exsufflation devices are
effective in patients with a low PCF (around 160 L·min−1). They use positive pres-
sure to deliver a maximal lung inhalation followed by an abrupt switch to negative
pressure to the upper airway. The rapid change from positive to negative pressure
aims to simulate the airflow changes that occur during a cough, thereby assist-
ing sputum clearance. However, in patients with severe bulbar weakness and with
spasticity of the upper airway (e.g. ALS) the negative pressure may induce collapse.
This phenomenon is not apparent in children with spinal muscular atrophy type 1
as they do not experience upper airway spasticity. Another potential complica-
tion of these devices, as with any device delivering positive pressure, is the risk
of pneumothorax. Although this risk seems relatively low, care should be taken
when acclimatising the patient to the device. Initially, it is wise to monitor oxygen
saturation and to investigate any reports of chest pain and oxygen desaturation
that are not a result of sputum mobilisation with a chest radiograph. Also, as the
Long-term NIV
Figure 2. Performance of a manual thoracoabdominal assisted cough. The respiratory

physiotherapist simultaneously performs a downward manoeuvre with the upper arm and
an inward and upward abdominal thrust with the lower arm, as the patient coughs aiming
to increase expiratory airflow. Image courtesy of the author.
aim of mechanical insufflation–exsufflation is to clear a large amount of secretions

from the upper airways, there is a risk of mobilising large mucus plugs into the cen-
tral airways where they may result in airways obstruction. Where appropriate, suit-
able emergency equipment (e.g. a resuscitation bag and suction) should be available
when using these devices. There has always been debate around what settings to
use. In patients who are extremely weak with a low or unrecordable PCF, high pres-
sures (greater than +40 to −40 cmH2O) are highly likely to be required. Moreover,
when patients are acutely unwell and weaker, it is likely that the pressures on the
device will need to be increased to further assist cough augmentation or combined
with a MAC. As with any medical device, the settings should be chosen for patient
comfort and then reassessed and adjusted to increase PCF on devices that have the
ability to measure this, or to increase cough audibility. Suggested initial settings are
shown in box 1. Commonly available devices include: CoughAssist and CoughAssist
E70 (Philips Respironics, Pittsburgh, PA, USA); NIPPY Clearway (B & D Electromedical,
Stratford-upon-Avon, UK) and Pegaso (Dimla Italia, Bologna, Italy). These devices can
produce expiratory airflows of >160 L·min−1 (value taken from marketing literature).
Historically, mechanical insufflations–exsufflation devices are often used in auto-
matic mode, enabling the device to be used in the home environment without a
trained professional. In this mode the device swings between a set negative and
positive pressure, and will hold in insufflation for a set period before switching to
exsufflation for a set time (the movement in manual mode should be one sweep-
ing movement with no pauses between insufflation and exsufflation), after which
it will pause. The patient learns to coordinate their cough to when the device
switches to exsufflation. When setting up the device tell the patient a deep breath
is coming and tell them to cough as they feel the negative pressure. Newer devices
have advanced triggering features that allow the patient to trigger the insuffla-
tion for a timed hold. Devices also now have internal software that enables the
Long-term NIV
Box 1. Set-up of mechanical insufflation–exsufflation

The main clinical goal of mechanical insufflation–exsufflation is to achieve
a rapid maximal insufflation followed immediately by a rapid exsufflation.
The settings should be aimed at this goal, which is evaluated by observing
the chest expansion and deflation. Some recommendations for initial
settings are shown below.
• In novice patients start with the insufflation pressure between 15 and
20 cmH2O, and increase the insufflation pressure to give a passive
inspiration to TLC.
• In patients on NIV start the insufflation pressure 5–10 cmH2O greater
than their IPAP or pressure support.
• Initially start in manual mode with a full face mask to allow the patient
to experience the deep breath in and increase the insufflation time as
required to achieve TLC. Tell the patient what is happening and only
provide insufflation for as long as required (1–2 s or longer if required).
• Initially start with the exsufflation (negative) pressure 5–10 cmH2O
greater than the positive pressure.
• Again, let the patient experience the insufflation and exsufflation pressure
in manual mode. Tell the patient what is happening and only provide an
insufflation for as long as required (1–2 s or longer if required), and do
the same for the exsufflation pressure. Reassess and see if you need to
increase the negative pressure to further enhance cough efficacy. You
may need to hold in exsufflation for longer. When using the device in
manual mode, you will need to move the switch in one fast movement to
ensure there is no pause between insufflation and exsufflation.
• Work out for how long you provided an insufflation and exsufflation, and
transcribe the settings to automatic mode. If you are unsure start with
an insufflation, exsufflation and pause time of 1 s, and adjust to patient
comfort.
• If the patient likes to trigger the cycles set the device up with the trigger
feature enabled.
therapist to work out natural timings for insufflation and exsufflation based on
the patients “natural” cough profile. However, to date, there is no evidence to
prove this is better than a therapist setting the insufflation and exsufflation times
manually. Patients may well need longer insufflation/exsufflation times than their
“natural” cough to enhance cough efficacy.
Daytime mouthpiece ventilation
Mouthpiece ventilation (MPV) is delivered to the patient via a small angled mouth-
piece or a little catheter (sip ventilation) rather than a mask (fig. 3).
Patients with neuromuscular disease and spinal cord injury are likely to bene
fit from periods of MPV instead of continuous mask ventilation during the day.
MPV increases survival, and is a safe alternative to tracheostomy and continuous
mask ventilation in patients with Duchenne muscular dystrophy. MPV is a useful
alternative for preventing mask-related pressure sores in patients who have high
Long-term NIV
Figure 3. A patient on MPV. The ventilator is attached to the back of the wheelchair. The
mouthpiece arm is attached to the wheelchair and the straw is situated in close proximity to
the patient's mouth. In patients with limited ventilator autonomy it is recommended that the
mouthpiece is secured around the neck of the patient. This ensures that if the patient moves
the mouthpiece moves with them rather than being out of reach from the patient.
ventilator acuity and can enhance communication. Mouthpiece ventilation can

also be used to assist the respiratory muscles during exercise in the ambulant
patient and to breath stack to assist coughing. However, clinicians are cautious
about using MPV as the technique may look insecure. There is also a lack of know
ledge about how to set it up. In addition to these barriers, there is a high preva-
lence of NIV centres using only pressure-cycled ventilators with limited access to
volume-cycled ventilators. There has also been a lack of commercially available
equipment to secure the mouthpiece to the wheelchair or bed.
Equipment required for MPV MPV requires a small, angled mouthpiece similar to a
nebuliser angled mouthpiece but without the hole or flexible straw. The mouth-
piece needs to have a small diameter at the mouth in order to prevent the ven-
tilator continually alarming low pressure when the patient is not receiving any
breathes from the ventilator. The mouthpiece is then attached via flexible tub-
ing to standard wide-bore tubing. You do not necessarily need an expiratory valve
(although one can be present if the ventilator you are using requires this) as the
patient will “sip” a breath in and then breathe out around the mouthpiece.
Mouthpiece ventilation is traditionally set-up with machines specifically designed
for volume-cycled ventilation. More recently ventilators have been marketed with
specific modes for MPV, which assist you in the set-up of the device. Suggested
starting settings for MPV are shown in box 2.
In the practice, a support on the shoulder is the preferred way to secure the
mouthpiece. This is because it follows the patients position, even if the patient
Long-term NIV
Box 2. Ventilator settings for MPV

• To initiate mouthpiece ventilation using a volume-cycled ventilator set
a V T of at least 700–1200 mL, but typically 1000 mL. This prevents the
ventilator alarming low V T when the patient is not taking “sips” from the
ventilator. Increase or decrease to patient comfort.
• Set the PEEP to 0 cmH2O.
• Set the inspiratory time to 1 s and increase until the patient is comfortable.
• Set the breaths per minute to the lowest possible setting.
• Set the inspiratory trigger to be sensitive enough to respond to the “sip”,
but not so sensitive that there is self-triggering. If the patient is unable to
take a sip from the ventilator then an appropriate backup rate may be set
to compensate.
• If the ventilator used continues to alarm despite these recommendations
you can increase the V T or insert a heat moisture exchanger (HME).
Insertion of the HME slightly increases the resistance within the tubing
and this tricks the low pressure alarm.
is falling on the side in their wheelchair. If the support is attached to the wheel-
chair, the patient cannot change their position during the ventilation time, but this
may be the only option. Conventional invasive mechanical ventilator arms can be
adapted so they can be secured to the patients wheelchair. Individual ventilator
companies have also designed specialist tubing and connectors to simplify the
equipment required to initiate mouthpiece ventilation.
Acknowledgements
This work was supported by the NIHR Respiratory Disease Biomedical Research
Unit at the Royal Brompton and Harefield NHS Foundation Trust and Imperial
College London.
Further reading
• AARC (1991). AARC (American Association for Respiratory Care) clinical practice
guideline. Postural drainage therapy. Respir Care; 36: 1418–1426.
• Bach JR, et al. (1993). Intermittent positive pressure ventilation via the mouth as
an alternative to tracheostomy for 257 ventilator users. Chest; 103: 174–182.
• Bott J, et al. (2009). Guidelines for the physiotherapy management of the adult,
medical, spontaneously breathing patient. Thorax; 64: Suppl. 1, i1–i51.
• Chatwin M (2008). How to use a mechanical insufflator–exsufflator “cough assist
machine”. Breathe; 4: 320–329.
• Main E, et al. (2014). Physiotherapy for Respiratory and Cardiac Problems: Adults
and Children. 5th Edn. London, Elsevier.

Long-term NIV
• McKim D, et al. (2013). Twenty-four hour noninvasive ventilation in Duchenne

muscular dystrophy: a safe alternative to tracheostomy. Can Respir J; 20: e5–e9.
• Rochester A, et al. (2012). Devices and techniques to aid physiotherapy in respira-
tory patients. ERS Buyers’ Guide; pp. 73–84.
Online resources
• Hare A, et al. Acute and chronic neuromuscular disease modules. Skills-
based Simulator Training in Non-Invasive Ventilation. www.ers-education
.org/e-learning/simulators.aspx
• Gonçalves M, et al. (2012). Cough assistance techniques in neuromuscular disease.
ERS CME online. www.ers-education.org/cmeonline/selection/cough-assistance-
techniques-in-neuromuscular-disease/video-introduction.aspx

Long-term NIV
Long-term NIV failure: causes and

problem solving
Jean-Paul Janssens, Dan Adler and Jesus Gonzalez-Bermejo
Long-term NIV is currently widely accepted as a first-line treatment for chronic

hypercapnic respiratory failure. However, certain conditions are required for the
efficacy and success of NIV:
• The underlying pathology must be an established indication for NIV
• Patient comfort must be ensured by optimal choice of interface, and air humidi
fication when required
• Ventilator settings must be adapted to the patient’s respiratory physiology and
perceived as comfortable
• Compliance must be sufficient to allow NIV to provide a significant physio
logical benefit
If these conditions are not fulfilled, NIV may fail to control chronic alveolar
hypoventilation. Here, we will define NIV failure as 1) immediate intolerance and
non-acceptance of NIV, 2) prolonged non-compliance with NIV, 3) failure to cor-
rect chronic hypercapnia and associated symptoms, or 4) the requirement for
invasive ventilation (tracheostomy) or death. We will also review elements which
are important to prevent NIV failure and, whenever possible, to improve and pur-
sue treatment after NIV failure.
Immediate NIV failure
There is a small group of patients for whom the idea of wearing a mask is psycho-
logically unacceptable (i.e. associated with apprehension, anxiety, phobic b
ehaviour
and claustrophobia). Tact and a very progressive approach by the NIV team are
important. Use of hypnosis has been reported in children for implementing NIV
or CPAP treatment for OSAS, and may also be offered to adults. Cognitive behav-
ioural therapy (CBT) is another option: in patients with OSAS and claustrophobia,
Key points
• Major causes of NIV failure are: poor tolerance of NIV, leaks,
inappropriate ventilator settings, an inappropriate indication
for NIV, progression of the underlying disorder, failure to
manage secretions and cognitive disorders.
• Tracheostomy is mainly considered when use of NIV exceeds
16–18 h per day or when management of secretions fails
under NIV.

Long-term NIV
CBT has been shown to improve adherence to CPAP. Whether CBT is also effective
in improving adherence to NIV has yet to be confirmed. Use of anxiolytics is a
double-edged sword in this population because of the risk of worsening nocturnal
hypoventilation if NIV fails, and must be proposed with appropriate surveillance.
Immediate NIV failure may also occur in specific groups of patients such as those
with bulbar-onset ALS (discussed later).
Interruption of NIV because of non-compliance
Compliance under NIV is usually considered insufficient when daily use of the ven-
tilator is <4 h·day−1. Below this threshold, it is unlikely that NIV can control chronic
respiratory failure (CRF) and normalise or significantly improve arterial blood gases
(ABGs). The most frequent causes of non-compliance are listed in box 1. Absence of
a perceived benefit of NIV is also a potential cause of non-compliance. Compliance
data provided by the ventilator software and the pattern of ventilator use may sug-
gest poor tolerance of NIV (fig. 1): this requires a systematic work-up of identify the
cause of intolerance and adjust treatment accordingly in order to prevent NIV failure.
Unintentional leaks Unintentional leaks are a major cause of patient discomfort
and thus poor compliance. They may cause xerophthalmia, conjunctival inflam-
mation and xerostomia because of the increased flow generated by the ventilator
to compensate for leaks. Patient awareness of leaks is highly variable: conscious-
ness of intermittent leaks may be perceived as a factor contributing to major
discomfort and sleep disruption even if the level of leaks reported by ventilator
software appears acceptable; conversely, tolerance of prolonged unintentional
leaks may be surprising in some patients despite leak flows compromising the
Box 1. Frequent causes of non-compliance with NIV

Patient discomfort resulting from:
• Unintentional leaks (around the mask or through the mouth)
• Leak-related xerophthalmia or conjunctival irritation
• Leak-related xerostomia
• Pain caused by the mask or headgear
• Skin lesions associated with the interface
• Nasal obstruction or congestion, chronic rhinitis
• Nasal mucosal irritation and dryness that is unresponsive to humidification
• Inappropriate ventilator settings
• Patient–ventilator asynchrony
• Inappropriate alarm settings
• Gastrointestinal distension
• Poor sleep quality
Psychological factors:
• Lack of motivation
• Lack of perceived benefit
• Apprehension, anxiety and claustrophobia
• Depression
• Cognitive disorders
• Lack of social support

Long-term NIV
12:00 h
09:00 h
06:00 h
03:00 h
Time
00:00 h
21:00 h
18:00 h
15:00 h
12:00 h
25 26 27 28 29 30 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Day
Figure 1. Compliance data provided by ventilator software in a subject with a poor tolerance
of NIV related to leaks and discomfort of interface. Reproduced and modified from Pasquina
et al. (2012) with permission from the publisher.
efficacy of NIV. Ventilator software may be helpful in quantifying leaks, although

reliability of the data provided is device dependent. Prevention and correction of
leaks requires expertise in choosing the most appropriate interface. Manufacturers
provide a wide choice of nasals prongs, nasal and facial masks. The NIV team must
be familiar with available interfaces, their pros and cons, and how to determine
the appropriate size. It is often necessary to proceed by “trial and error”, allowing
the patient to experience different interfaces over a period of days or weeks. For
further discussion, see the section entitled “Choosing the interface”.
Pain associated with the mask and/or headgear Pain associated with the mask or
headgear impairs quality of sleep and is a major cause of NIV intolerance. The
headgear must not be too tight, allowing the inner silicone layer of nasal and facial
masks to inflate, adhere to the skin and prevent leaks. Training must be provided
for carers to avoid pressure sores and incorrect positioning of the mask, or pain
related to the headgear.
Inappropriate ventilator settings Inappropriate ventilator settings may also lead to
poor tolerance of NIV and compromise compliance. Typical complaints are: levels of
pressure support are too high, expiratory discomfort, delayed or difficult inspiratory
triggering with unrewarded inspiratory efforts, delayed or premature cycling, or the
impression that “the ventilator goes too fast” (table 1). A targeted medical history
combined with data provided by ventilator software should in most cases lead to
identification of the underlying cause of discomfort and adjustment of ventilator
settings. If the underlying cause is unclear, polygraphy or polysomnography under
ventilation, combined with data provided by the ventilator software, will give an
excellent picture of patient–ventilator interactions and unwanted respiratory events.
Failure to correct chronic hypercapnia and associated symptoms
There is no clear-cut definition of failure to correct ABGs with NIV. The 1999
American College of Chest Physicians consensus conference report on NIV in CRF
Long-term NIV
Table 1. Discomfort related to inappropriate ventilator settings as expressed by the patient,
and possible remedies
Patient description Possible remedy

Too much air Rise time is too short
Pressure support is too high
Difficulty exhaling EPAP is too high
Delayed cycling (choose a higher percentage of peak
inspiratory flow)
Adapt tI to respiratory rate and desired tI/tE ratio
Difficulty in Adapt to intrinsic PEEP (adjust EPAP)
triggering the
ventilator and Consider dynamic hyperinflation induced by NIV and adjust
unrewarded settings accordingly
inspiratory Increase inspiratory trigger sensitivity (rarely necessary
efforts because most devices are very sensitive, there is a risk of
auto-triggering)
Pressurisation is Adapt cycling criterion (lower percentage of peak
too short inspiratory flow)
Increase tI and adapt tI/tE ratio
The ventilator goes Decrease backup rate
too fast
Adapt tI/tE ratio and tI in COPD (allow increased time for
exhaling)
tI: inspiratory time; tE: expiratory time.
provides recommendations for NIV in obstructive and restrictive disorders, but

does not mention any specific goal in terms of control of ABGs, or nocturnal SpO2.
Recent “best clinical practice” recommendations regarding NIV in CRF, published
by the American Academy of Sleep Medicine, state that “an acceptable goal for PCO2
is a value less than or equal to the awake PCO2” and that “supplemental oxygen
may be added in patients with an awake SpO2 <88% or when the pressure support
and respiratory rate have been optimised but SpO2 remains <90% for 5 min or
more” (Berry et al., 2010). A reasonable goal of NIV titration is to adjust ventilator
settings so that SpO2 is >90% for at least 90% of nocturnal recordings, and PtcCO2 is
<45–50 mmHg (6.00–6.65 kPa), but this is based only on expert opinion.
Appropriate monitoring of NIV is mandatory for the detection of residual daytime
or nocturnal hypoventilation, unwanted respiratory events, and leaks. Combining
clinical history, daytime ABGs, nocturnal pulse oximetry and data provided by the
ventilator software will detect most situations requiring adjustment of ventila-
tor settings and/or interfaces. However, the reliability of ventilator software varies
considerably from one device to another. Use of transcutaneous capnography is
a precious adjunct when available, especially in patients receiving supplemental
oxygen, provided that it is used by an experienced team, and that clinicians are
Long-term NIV
aware of possible and unpredictable errant values. It is used by many expert cen-
tres, and in some countries it is proposed by home care providers. User-friendliness
of the software and ease of use have markedly improved, but the devices remain
expensive and fragile.
Longitudinal observational data show that, on average, patients with restrictive dis-
orders normalise their PaCO2 values under NIV. In COPD, PaCO2 values obtained under
NIV depend on ventilation strategy: higher pressure support levels have been shown
to improve daytime and nocturnal levels of PaCO2; whether this suffices to improve
outcome is still a matter of debate. In an RCT of 201 COPD patients treated using ven-
tilatory support after an acute episode of hypercapnic respiratory failure (AHECOPD),
prolonged NIV (12 months) improved daytime ABGs and nocturnal PtcCO2, with a
trend for improvement in health-related quality of life, but failed to improve exacer-
bation rates, survival, pulmonary function tests, mood state (the presence of anxiety
or depression measured by the Hospital Anxiety and Depression scale), or dyspnoea
(Struik et al., 2014). Importantly, daytime PaCO2 also improved in control patients,
although to a lesser degree. Another large 12-month multicentre RCT compared NIV
with long-term oxygen therapy in stable hypercapnic COPD (median PaCO2: 7.0 kPa
(53 mmHg); interquartile range: 6.8–7.2 kPa (51–54 mmHg)), and showed a mar-
ginal benefit of NIV on survival at the expense of a decreased quality of life under
NIV. However, in this trial, pressure support levels were low (on average, the differ-
ence between IPAP and EPAP was 8 cmH2O), and NIV failed to significantly decrease
PaCO2 (McEvoy et al., 2009). In contrast, a recent multicentre RCT performed in a
highly selected population of stable severely hypercapnic COPD (PaCO2: 7.8±0.8 kPa)
showed that NIV targeted to reduce PaCO2 by at least 20% or to achieve PaCO2 values
<6.5 kPa (<49 mmHg) significantly improved survival and reduced emergency hos-
pital admissions when compared with standard care (Köhnlein et al., 2014). Patient
selection, timing of NIV, and ventilator settings all appear critical for the success of
long-term NIV in COPD. Current evidence suggests that long-term NIV should not
be initiated in hospital immediately after an AHECOPD. Patients recovering from
an AHECOPD should be reassessed for persistence of hypercapnia 2–6 weeks after
discharge (Elliott, 2014). Long-term NIV should be considered only in patients who
remain hypercapnic (PaCO2 >7 kPa (>53 mmHg)), and should be pursued only if opti-
mal ventilator adjustment decreases daytime PaCO2 by at least 20%, or achieves a
daytime PaCO2 under NIV <6.5 kPa (<49 mmHg). The possible haemodynamic reper-
cussions on cardiac output of the “high pressure ventilation” strategy should be
noted (Lukácsovits et al., 2012).
In all indications, when confronted with poor correction of ABGs and residual noc-
turnal hypoventilation, the following strategies should be considered:
• Identify and correct leaks and patient–ventilator asynchrony (which requires
either polygraphy or polysomnography).
• Adapt the interface (although a subject of debate because of a possible
increase in upper airway resistance in OSAS, facial masks decrease leaks and
may improve nocturnal ABGs).
• Consider increasing pressure support (or VT in volume-cycled modes) and
respiratory rate.
• In restrictive disorders, pressure support and backup rate should be increased,
according to patient tolerance, to optimise daytime ABGs and, when available,
nocturnal PtcCO2.
Long-term NIV
• As mentioned earlier, in COPD, higher levels of pressure support may also

improve daytime ABGs and nocturnal PtcCO2. The additional contribution of high
backup rates has recently been questioned (Murphy et al., 2012a). Importantly,
clinicians must be aware that high pressure ventilation may take time to imple-
ment, and prolong hospital stays.
• Consider volume-targeted pressure modes.
Volume-targeted modes are designed to continuously adapt pressure support in

order to maintain a VT set by the clinician. Although their systematic implementa-
tion is not justified in OHS or in COPD, volume targeting may be appropriate in
individual cases to improve control of nocturnal hypoventilation. Recently, devices
with automated titration of expiratory and inspiratory pressures, pressure support
and backup respiratory rate have been commercialised. However, the efficacy and
safety of these devices must be independently validated in clinical trials before
they can be recommended.
Indications for NIV associated with a higher mortality and failure rate
The probability of pursuing NIV (or of NIV failure) varies according to the underlying
disease. Obesity hypoventilation, most neuromuscular diseases (except for ALS)
and chest wall disorders are usually associated with excellent survival and a high
probability of pursuing NIV. For instance, Borel et al. (2013a) followed 107 obese
subjects under NIV for an average of 43±14 months: 13 died, with the major risk
factor being associated cardiovascular morbidity; and 14 interrupted NIV, seven of
whom switched to CPAP. Thus only 7% of these patients failed to pursue NIV. In
contrast, patients with bronchiectasis and COPD have a much lower survival and
probability of pursuing NIV, although, as mentioned earlier, the data concerning
COPD are heterogeneous and depend on the population studied, the ventilator
modes selected and their impact on PaCO2. For subjects with bronchiectasis and
CF, one of the critical issues is management of secretions, with the other being the
prognosis of the underlying disorder.
The critical issue of patient selection and timing for NIV in COPD has been previ-
ously discussed.
Duchenne muscular dystrophy Leger et al. (1994) reported that Duchenne muscular

dystrophy (DMD) patients had a low probability of pursuing NIV: most patients,
when becoming highly dependent on NIV, were switched to invasive positive
pressure ventilation via tracheostomy (TPPV). Today, marked improvements in
interfaces, the use of mouthpiece ventilation during the daytime and the use of
portable ventilators allow highly dependent patients with DMD (and other neuro-
muscular diseases) to even use NIV continuously with an acceptable quality of life.
Even among DMD patients over 30 years of age, TPPV is currently implemented in
<10% of cases in expert centres, the major reason for TPPV being management
of secretions.
Amyotrophic lateral sclerosis In ALS, the NIV failure rate is high in patients with
bulbar-onset disease, with poor tolerance of NIV, increased risk of broncho-
aspiration, difficult management of secretions and saliva, and glottic dysfunc-
tion leading to patient–ventilator asynchrony. When NIV is pursued, average
compliance is lower in bulbar-onset subjects than in non-bulbar patients. Also,
Long-term NIV
non-bulbar subjects tolerate higher pressures than bulbar-onset subjects.

Bourke et al. (2006) found no improvement of survival in patients with poor bul-
bar function put under NIV when compared with standard care; median survival
was 222 days, which was not significantly different from the usual care group,
although a few items on the quality of life scales did improve under NIV, including
the dyspnoea and symptoms scales. In fact, in patients with bulbar involvement,
intolerance of NIV may lead to immediate NIV failure in up to 40% of subjects.
Orthopnoea is a strong predictor of compliance and success of ventilation in ALS.
By contrast, difficulties in managing secretions in bulbar patients are predictive of
NIV failure. It is, however, the experience of the authors that NIV may be perfectly
well tolerated in bulbar patients and that having “bulbar-onset” ALS should not
be a reason for not considering or for discouraging NIV. In ALS, NIV failure leads
either to discontinuation of ventilator support and palliative care, or to trache-
ostomy (TPPV), if there is a strong willingness to pursue respiratory support on
behalf of the patient and their caregivers, because of the very high burden TPPV
puts on caregivers. Need for a tracheostomy is highly probable in subjects with
repeated episodes of low daytime SpO2 that cannot be normalised by mechanical
insufflation-exsufflation.
Steinert’s myotonic muscular mystrophy Patients with Steinert’s myotonic muscular

dystrophy (SMD) require a few comments. Although there is a consensus that NIV
is indicated when SMD patients develop CRF, adherence to NIV in SMD is often
low, mostly because of psychosocial issues (i.e. mild cognitive impairment), and
maybe also because of the muscular disorder per se. SMD patients may require
extra support from caregivers and healthcare professionals to achieve acceptable
compliance.
Palliative use An interesting RCT conducted by Nava et al. (2013) investigated

the palliative use of NIV in patients with terminal cancer. Of the 99 subjects
randomised to NIV, 11 “failed NIV” and discontinued treatment mainly because
of psychological reasons (anxiety, claustrophobia, sense of imminent death and
suffocation).
Ageing Ageing per se could be considered as a critical factor for prolonged NIV.

In fact, patients treated for CRF are frequently in the geriatric age span. The
Antadir observatory (www.antadir.com/fr/la-mediatheque) publishes yearly
data for patients treated at home for CRF (∼20 000 patients, excluding those
with sleep apnoea), and covers a 15-year observation period: 40% of all patients
are aged >75 years. To our knowledge, except for a small pilot study (Janssens
et al., 1998), no publication has specifically addressed the issue of ageing as a
limiting factor for NIV. In acute care settings, age is not considered a contra
indication for NIV. In a chronic setting, the presence of a significant cognitive
dysfunction (i.e. dementia) is the only major factor limiting the feasibility of
long-term NIV. The availability of home healthcare providers is also critical in
the very elderly.
Summary A higher NIV failure rate can be expected in subjects with bronchi
ectasis, in patients with COPD who remain hypercapnic under NIV, and in certain
subgroups of neuromuscular patients (e.g. those with SMD and ALS with bulbar
onset). Management of secretions is the most critical issue in these cases: when
Long-term NIV
not controlled by chest physiotherapy and optimal use of mechanical exsufflation-

insufflation devices, TPPV – if appropriate – may be the only option.
Tracheostomy after NIV failure
Although there are very few data on NIV failure leading to tracheostomy, some
indicators suggest that patients tend more and more often to remain on NIV
even when severely dependent, and that considering tracheostomy when NIV is
used for more than 16–18 h per day does not reflect current practice in many
centres. A Swiss observational study of 211 patients on home mechanical ven-
tilation over an 7-year period documented only two switches to TPPV because
of NIV failure (Janssens et al., 2003). According to the Antadir observatory,
although use of NIV has been increasing, home TPPV represents a marginal pro-
portion of patients under home mechanical ventilation. TPPV is of minor impor-
tance in COPD, parenchymal restrictive disorders, and sequelae of TB; however,
it represents ∼20% of neuromuscular patients with CRF (Antadir observatory
2011 data).
The authors of a Swedish report, covering a 10-year observation period of 1526
patients on home mechanical ventilation, estimated that ∼10% of NIV patients
“failed” NIV and were switched to TPPV. Again, the majority of these were neuro-
muscular patients (Laub et al., 2007).
OHS is currently the most frequent indication for NIV in Western Europe. We are
not aware of any published report of long-term NIV failure leading to TPPV in OHS,
and any occurrences are probably anecdotal. In fact, most OHS patients treated for
ARF by TPPV can eventually be weaned from their tracheostomy and put onto NIV.
In these patients, TPPV may be complicated by the thickness of subcutaneous tis-
sue between the skin and the trachea.
In summary, in Western Europe, among adults treated for CRF, NIV failure leading
to TPPV is a rare occurrence in most indications, except for neuromuscular dis-
eases. Failure to manage secretions in spite of mechanical cough assistance and
major glottis dysfunction are the major causes for a switch to TPPV.
Conclusions
NIV failure can be defined by non-compliance, failure to correct ABGs, the necessity
of a switch to invasive ventilation (TPPV) or death. Compliance is influenced by
multiple technical factors, many of which can be improved once correctly identified.
Correction of diurnal and nocturnal ABGs is a major end-point of NIV and may
require:
• increases in pressure support and backup rates,
• changing ventilator mode (from pressure to volume-cycled), or
• volume targeting, after identifying and treating technical problems (leaks,
patient–ventilator asynchrony, etc.)
NIV failure is seldom the only cause for deciding to implement TPPV: observational
data suggest that, even in highly dependent subjects, use of TPPV has markedly
decreased. Management of secretions is a critical factor and the most frequent
reason for implementing TPPV, most often in neuromuscular diseases. Finally, the
prognosis of patients under NIV depends on the underlying diagnosis: ALS, COPD
and bronchiectasis are globally associated with a worse prognosis.
Long-term NIV
Further reading
• American College of Chest Physicians (1999). Clinical indications for noninvasive
positive pressure ventilation in chronic respiratory failure due to restrictive lung
disease, COPD, and nocturnal hypoventilation – a consensus conference report.
Chest; 116: 521–534.
• Bach JR (2003). Noninvasive ventilation is more than mask ventilation. Chest; 123:
2156–2157.
• Bach JR, et al. (2004). Oximetry and indications for tracheotomy for amyotrophic
lateral sclerosis. Chest; 126: 1502–1507.
• Berry RB, et al. (2010). Best clinical practices for the sleep center adjustment
of noninvasive positive pressure ventilation (NPPV) in stable chronic alveolar
hypoventilation syndromes. J Clin Sleep Med; 6: 491–509.
• Borel JC, et al. (2013a). Comorbidities and mortality in hypercapnic obese under
domiciliary noninvasive ventilation. PLoS One; 8: e52006.
• Borel JC, et al. (2013b). Type of mask may impact on continuous positive airway
pressure adherence in apneic patients. PLoS One; 8: e64382.
• Bourke SC, et al. (2003). Noninvasive ventilation in ALS: indications and effect on
quality of life. Neurology; 61: 171–177.
• Bourke SC, et al. (2006). Effects of non-invasive ventilation on survival and quality
of life in patients with amyotrophic lateral sclerosis: a randomised controlled trial.
• Delord V, et al. (2013). Medical hypnosis as a tool to acclimatize children to non
invasive positive pressure ventilation: a pilot study. Chest; 144: 87–91.
Thorax; 65: 303–308.
• Ekkernkamp E, et al. (2014). Impact of intelligent volume-assured pressure sup-
port on sleep quality in stable hypercapnic chronic obstructive pulmonary disease
patients: a randomized, crossover study. Respiration; 88: 270–276.
• Elliott M (2014). Domiciliary NIV for COPD: where are we now? Lancet Respir Med;
2: 672–673.
• Farrero E, et al. (2005). Survival in amyotrophic lateral sclerosis with home
mechanical ventilation: the impact of systematic respiratory assessment and bul-
bar involvement. Chest; 127: 2132–2138.
• Gacouin A, et al. (1996). Long-term nasal intermittent positive pressure ventila-
tion (NIPPV) in sixteen consecutive patients with bronchiectasis: a retrospective
study. Eur Respir J; 9: 1246–1250.

Long-term NIV
• Gonzalez-Bermejo J, et al. (2012). Proposal for a systematic analysis of polygraphy

or polysomnography for identifying and scoring abnormal events occurring during
non-invasive ventilation. Thorax; 67: 546–552.
• Gonzalez-Bermejo J, et al. (2013). Prognostic value of efficiently correcting noctur-
nal desaturations after one month of non-invasive ventilation in amyotrophic lat-
eral sclerosis: a retrospective monocentre observational cohort study. Amyotroph
Lateral Scler Frontotemporal Degener; 14: 373–379.
• Heritier Barras AC, et al. (2013). Is tracheostomy still an option in amyotrophic lateral
sclerosis? Reflections of a multidisciplinary work group. Swiss Med Wkly; 143: w13830.
• Jackson M, et al. (1994). Long term non-invasive domiciliary assisted ventilation
for respiratory failure following thoracoplasty. Thorax; 49: 915–919.
• Janssens JP, et al. (1998). Non-invasive home ventilation in patients over 75
years of age: tolerance, compliance, and impact on quality of life. Respir Med; 92:
1311–1320.
• Janssens JP, et al. (1999). Ventilation non-invasive (VNI) a domicile: suivi à long
terme de 32 cas [Non-invasive home ventilation (NIHV): long-term survival of 32
cases]. Rev Mal Respir; 16: 511–520.
• Janssens JP, et al. (2003). Changing patterns in long-term noninvasive ventilation:
a 7-year prospective study in the Geneva Lake area. Chest; 123: 67–79.
• Janssens JP, et al. (2009). Impact of volume targeting on efficacy of bi-level non-
invasive ventilation and sleep in obesity-hypoventilation. Respir Med; 103: 165–172.
the contribution of simple tools such as pulse oximetry, capnography, built-in venti-
lator software and autonomic markers of sleep fragmentation. Thorax; 66: 438–445.
• Kelly JL, et al. (2014). Randomized trial of ‘intelligent’ autotitrating ventilation
v ersus standard pressure support non-invasive ventilation: impact on adherence
and physiological outcomes. Respirology; 19: 596–603.
• Khirani S, et al. (2013). Nocturnal mouthpiece ventilation and medical hypnosis to
treat severe obstructive sleep apnea in a child with cherubism. Pediatr Pulmonol;
48: 927–929.
• Kohler M, et al. (2005). Quality of life, physical disability, and respiratory
impairment in Duchenne muscular dystrophy. Am J Respir Crit Care Med; 172:
1032–1036.
• Laub M, et al. (2007). Survival of patients on home mechanical ventilation: a
nationwide prospective study. Respir Med; 101: 1074–1078.
• Leger P, et al. (1989). Home positive pressure ventilation via nasal mask for patients
with neuromusculoskeletal disorders. Eur Respir J; 2: Suppl. 7, 640s–644s.

Long-term NIV
• Leger P, et al. (1994). Nasal intermittent positive pressure ventilation. Long-term

follow-up in patients with severe chronic respiratory insufficiency. Chest; 105:
100–105.
• Lukácsovits J, et al. (2012). Physiological changes during low- and high-intensity
noninvasive ventilation. Eur Respir J; 39: 869–875.
• McEvoy RD, et al. (2009). Nocturnal non-invasive nasal ventilation in stable hyper-
capnic COPD: a randomised controlled trial. Thorax; 64: 561–566.
• McKim DA, et al. (2011). Home mechanical ventilation: a Canadian Thoracic
Society clinical practice guideline. Can Respir J; 18: 197–215.
• Means MK, et al. (2007). Graded exposure therapy for addressing claustrophobic
reactions to continuous positive airway pressure: a case series report. Behav Sleep
Med; 5: 105–116.
• Meyer TJ, et al. (1997). Air leaking through the mouth during nocturnal nasal ven-
tilation: effect on sleep quality. Sleep; 20: 561–569.
• Murphy PB, et al. (2012a). High pressure versus high intensity noninvasive ventila-
tion in stable hypercapnic chronic obstructive pulmonary disease: a randomized
crossover trial. Int J Chron Obstruct Pulmon Dis; 7: 811–818.
• Murphy PB, et al. (2012b). Volume targeted versus pressure support non-invasive
ventilation in patients with super obesity and chronic respiratory failure: a ran-
domised controlled trial. Thorax; 67: 727–734.
• Nava S, et al. (2013). Palliative use of non-invasive ventilation in end-of-life
patients with solid tumours: a randomised feasibility trial. Lancet Oncol; 14:
219–227.
• Oscroft NS, et al. (2014). Volume assured versus pressure preset non-invasive ven-
tilation for compensated ventilatory failure in COPD. Respir Med; 108: 1508–1515.
• Pasquina P, et al. (2012). What does built-in software of home ventilators tell
us? An observational study of 150 patients on home ventilation. Respiration; 83:
293–299.
• Perrin C, et al. (1997). Domiciliary nasal intermittent positive pressure ventila-
tion in severe COPD: effects on lung function and quality of life. Eur Respir J; 10:
2835–2839.
• Peysson S, et al. (2008). Factors predicting survival following noninvasive ventila-
tion in amyotrophic lateral sclerosis. Eur Neurol; 59: 164–171.
• Rabec C, et al. (2009). Evaluating noninvasive ventilation using a monitoring sys-
tem coupled to a ventilator: a bench-to-bedside study. Eur Respir J; 34: 902–913.
• Storre JH, et al. (2006). Average volume-assured pressure support in obesity
hypoventilation: a randomized crossover trial. Chest; 130: 815–821.

Long-term NIV
• Struik FM, et al. (2014). Nocturnal non-invasive ventilation in COPD patients with
prolonged hypercapnia after ventilatory support for acute respiratory failure: a ran-
domised, controlled, parallel-group study. Thorax; 69: 826–834.
• Teschler H, et al. (1999). Effect of mouth leak on effectiveness of nasal bilevel ven-
tilatory assistance and sleep architecture. Eur Respir J; 14: 1251–1257.
• Windisch W, et al. (2009). High-intensity non-invasive positive pressure ventila-
tion for stable hypercapnic COPD. Int J Med Sci; 6: 72–76.
Online resources
• Wijkstra P (2014). Monitoring of NIV in the chronic setting. ERS Course Noninvasive
ventilation: basic concepts, Hanover 2014. www.ers-education.org/events/

Long-term NIV
Long-term NIV case histories
Alanna Hare
Case study 1: obesity hypoventilation syndrome

A 42-year-old man with a BMI of 41 kg⋅m−2 who was commenced on NIV
6 months ago for OHS returns to your clinic for review. His current settings are:
IPAP 30 cmH2O; EPAP 14 cmH2O; backup rate 12 breaths⋅min−1. In the clinic, he
reports lethargy, morning headaches and unrefreshing sleep. He has been using
his NIV for around 2–3 h per night. Blood gas tensions in the clinic demonstrate
chronic type 2 respiratory failure: pH 7.35; PaCO2 9.8 kPa (74 mmHg); PaO2 8.2 kPa
(62 mmHg); HCO3− 36 mmol⋅L−1 in room air.
Practice point 1 Patients should be encouraged to use NIV for at least 4 h
overnight. Greater benefits are obtained with increased usage. Improvements
in clinical symptoms and blood gas tensions depend on adherence to NIV.
Most home ventilators are now equipped with advanced technology enabling
the practitioner to download a wide range of data from the ventilator itself.
Data available usually include parameters such as ventilator settings, usage
(compliance), estimated leak, estimated VT, respiratory frequency, minute
ventilation, residual apnoeas and ventilator–patient dyssynchrony. These
Key points
• Monitoring patients on long-term NIV after discharge is vital
to ensure optimal control of ventilation, and to identify and
manage problems early.
• Assessment should involve a review of symptoms, particularly
those of nocturnal hypoventilation, discussion of any problems
that have arisen with the ventilator or interface, and a review
of compliance. Measurement of daytime arterial or capillary
blood gas tensions should be performed and, in some patients,
overnight monitoring may also be required.
• A range of problems may occur, even in patients stably
established on NIV for some time, and provision must be made
for these patients to contact a member of the clinical team in
the event that such issues occur.

Long-term NIV
data are extremely valuable in the clinic when reviewing patients on home
ventilation. Medical history may be unreliable when assessing compliance
with ventilation and objective recordings are therefore valuable. They may
also suggest a particular pattern of use, which may provide clues as to why
the patient is not able to use their ventilator as much as the clinician would
like. Most ventilators have the facility to review compliance data on a night-
by-night and/or average nightly basis. Poor compliance should prompt you
to discuss the reasons behind this with the patient.
In the clinic, the patient tells you that he finds his mouth is very dry during the
night. He is so uncomfortable that he usually takes off his mask after only an hour
or so, and then usually falls asleep again before he can put it back on.
Practice point 2 Oral and nasal dryness can be problematic in some patients,
and may limit use of the ventilator. In patients using nasal masks, the prob-
lem can often be resolved by switching to a full face mask to reduce leakage
through the mouth. If this does not improve things, a humidifier can be pro-
vided. Heated humidifiers are more effective than pass-over versions but can
sometimes impact ventilator triggering, necessitating a review of trigger sensi-
tivity. Heat and moisture exchange systems are associated with increased work
of breathing and are not recommended for routine use. Humidification is not
required in all patients, but should always be used in those with high secretion
load, such as patients with CF and bronchiectasis, and during tracheostomy
ventilation. In the latter group, heated-wire humidifiers are recommended.
You supply the patient with a humidifier as he is already using a full face mask that
is well-fitted, with minimal leaks. You encourage him to increase his use of NIV
to at least 4 h per night and arrange to see him again in the clinic. Unfortunately,
when he returns for review, despite improved adherence and average nightly use
of 5–6 h, the patient continues to report symptoms of nocturnal hypoventilation,
with headaches, lethargy and poor quality sleep. Blood gas tensions in the clinic
demonstrate chronic type 2 respiratory failure: pH 7.36; PaCO2 6.7 kPa (50 mmHg);
PaO2 8.0 kPa (60 mmHg); HCO3− 35 mmol⋅L−1 in room air. You arrange for an over-
night respiratory sleep study. The sleep study is shown in figure 1.
Practice point 3 Optimisation of ventilation in patients with OHS can be

challenging. Nocturnal oxygen saturations, alongside daytime PaCO2 meas-
urements (and/or PtcCO2 monitoring) can help guide therapy. This study dem-
onstrates nocturnal hypoventilation (oxygen saturations <90% and elevated
PtcCO2) and obstructive events (repetitive episodic desaturations) with a high
number of apnoeas, despite ventilation. The patient’s ventilator settings need
to be adjusted.
• Repetitive oxygen desaturations suggest obstructive events and concur-
rent obstructive sleep apnoea–hypopnoea syndrome (OSAHS). 90% of
patients with OHS have SAHS. EPAP should be titrated to the level that
achieves resolution of obstructive events during sleep. The level of EPAP
required in patients with OHS depends on the severity of OSAHS. Usually,
Long-term NIV
an EPAP of ∼12–14 cmH2O is required in these patients, but may be higher

in the superobese. If higher EPAP levels are not well tolerated, moderate
EPAP levels (7–10 cmH2O) can be applied, and IPAP used to correct persis-
tent hypopnoeas and hypoventilation.
• Persistent desaturation requires an increase in the IPAP setting to correct
hypoventilation. IPAP levels should be titrated until oxygen saturations of
>90% are achieved or to the maximum tolerated level. Again, the IPAP
required may be quite high (>20 cmH2O). If higher IPAP levels are poorly tol-
erated, backup respiratory rate can be increased to just below the patient’s
spontaneous respiratory rate and/or inspiratory time can be increased.
• If saturations cannot be maintained >90%, supplementary oxygen may
be required.
• Optimum ventilation (correction of hypercapnia) should be ensured prior
to adding supplemental oxygen.
• It may take several nights to achieve correction of blood gas tensions.
The patient’s ventilator settings are already high, and he finds it very difficult to
tolerate the increased pressures and is unable to sleep at all when you increase
his IPAP settings.
100
80
60
SpO2 %
40
Desaturation episodes
20
0
150
beats·min-1
Pulse rate
100
50
0
15
10
PtcCO2
kPa
0
22:00 h 01:00 h 04:00 h
Figure 1. Sleep study on ventilation with settings of: IPAP 30 cmH2O; EPAP 14 cmH2O;
backup rate 12 breaths⋅min−1. Reproduced from Skills-based Simulator Training in Non-
Invasive Ventilation. © ResMed Limited. All rights reserved.
Long-term NIV
Practice point 4 In those patients who cannot tolerate further increases in
pressure, increasing the backup ventilatory rate may enable the use of a lower
IPAP. An alternative may be to consider the use of AVAPS. In this mode of
ventilation, a range of IPAP pressures is set, along with a target VT. The ven-
tilator then adjusts the pressure support delivered to the patient either on a
breath-by-breath basis or within each breath (intrabreath analysis), in order
to ensure the preset VT is achieved.
Case study 2: Duchenne muscular dystrophy

A 24-year-old man with Duchenne muscular dystrophy (DMD) who has been
using nocturnal NIV at home for the past 5 years attends your clinic. His cur-
rent settings are: IPAP 15 cmH2O; EPAP 5 cmH2O; backup rate 16 breaths⋅min−1.
He uses a full face mask. He has come to see you today because he is report-
ing increasing difficulty concentrating on his college course and is finding he is
increasingly breathless in the day. He is even finding it difficult to complete sen-
tences. He is concerned that his ventilator settings may need to be adjusted. You
perform an overnight sleep study but this demonstrates good control of overnight
ventilation (fig. 2).
100
80
60
SpO2 %
40
20
0
150
beats·min-1
Pulse rate
100
50
0
15
10
PtcCO2
kPa
0
22:00 h 01:00 h 04:00 h
Figure 2. Sleep study on ventilation with settings of: IPAP 15 cmH2O; EPAP 5 cmH2O; backup
rate 16 breaths⋅min−1. Reproduced from Skills-based Simulator Training in Non-Invasive
Ventilation. © ResMed Limited. All rights reserved.
Long-term NIV
Practice point 1 Even with modern standards of care and optimised ventilatory
support, most patients with DMD will become dependent on ventilation for
most of the 24-h period during their twenties and thirties, especially once vital
capacity falls below 500 mL. If gas exchange at night while using NIV is normal
but the patient is breathless in the day with daytime hypercapnia, daytime NIV
is recommended. Initially, it may be sufficient for the individual to use short top-
ups of NIV during the daytime, but as symptoms progress and NIV requirements
increase, the use of mouthpiece ventilation may be more appropriate (fig. 3).
You next see the patient 6 months later. He has had a percutaneous endo-
scopic gastrostomy (PEG) tube inserted in the intervening period, as he was
increasingly unable to meet his nutritional requirements due to swallow-
ing problems. Following the PEG tube insertion, a sleep study demonstrated
that his nocturnal hypoventilation was not well controlled and his NIV set-
tings were increased. His ventilator settings are now: IPAP 17 cmH2O; EPAP
6 cmH2O; backup rate 16 breaths⋅min−1. Over the past few weeks, he has
reported abdominal pain and distension throughout the night, and intermit-
tently during the day.
Practice point 2 Gastric distension can be particularly problematic in patients

with neuromuscular disease, although it may also occur with higher pres-
sures in other patient groups, as the ventilator more easily inflates the stom-
ach than the lungs. Advising the patient to sleep more upright may help but,
if unsuccessful, inspiratory pressure can be reduced and compensated for by
an increase in the backup rate. In patients with a PEG tube in situ, this can be
used to vent excess air to provide relief.
The patient’s mother also reports that he is finding it increasingly difficult to cough
and clear his secretions, which are very sticky and thick. They have previously been
shown how to perform manually assisted coughs, which they do on a daily basis,
but she feels as though this is no longer as effective in clearing her son’s chest.
Figure 3. Mouthpiece ventilation.

Long-term NIV
Practice point 3 Management of secretions is vital for the success of NIV in

DMD. Manually assisted cough can be carried out while the patient is using
NIV, to avoid early tiring. Assisted insufflation using an ambu-bag can be
used if manually assisted cough alone is not sufficient to clear secretions.
Mechanical insufflation–exsufflation may be required in patients with more
advanced respiratory muscle weakness, although this technique is less use-
ful in patients with moderate or severe bulbar weakness. In many patients,
a combination of these techniques is required. Further details of these tech-
niques can be found in the section entitled “Practicalities of and guide to
cough augmentation and daytime mouthpiece ventilation”.
Case study 3: chest wall disease

A 47-year-old lady with congenital kyphoscoliosis who has been using noctur-
nal NIV for the past 10 years attends for routine review, and tells you that she
has been feeling more tired and has noticed she cannot walk as far as previously
over the past few months. Her blood gas tensions reveal: pH 7.38; PaCO2 6.7 kPa
(50 mmHg); PaO2 6.8 kPa (51 mmHg); HCO3− 28 mmol⋅L−1.
Practice point 1 In some patients with congenital kyphoscoliosis, pulmonary

hypoplasia and abnormal development of the pulmonary vasculature, along
with increasing severity of the spinal curvature, mean that worsening hypoxia
with pulmonary hypertension and right heart failure are common with increas-
ing age. Careful reassessment is vital in these patients, and should include
overnight monitoring to ensure ventilator settings are appropriate and carbon
dioxide levels are controlled, echocardiography to assess for the presence of
pulmonary hypertension, pulmonary function tests to assess for additional lung
disease (e.g. coexistent obstructive airway disease) and consideration of imag-
ing to rule out other pathologies, such as chronic thromboembolic disease.
Despite optimisation of her ventilator settings, a repeat sleep study demon-

strates persistent hypoxia overnight. Echocardiography demonstrates moderate
pulmonary hypertension and further tests rule out any coexistent respiratory or
cardiac pathologies.
Practice point 2 If, despite optimisation of cardiorespiratory status and ventila-

tor settings, nocturnal hypoxia cannot be controlled, supplemental oxygen at
night may be added. A full assessment for long-term oxygen therapy should
also be performed and consideration given to the use of diurnal oxygen sup-
plementation, with ambulatory oxygen provided as appropriate.
Further reading
• Bushby K, et al. (2010). Diagnosis and management of Duchenne muscular dys-
trophy, part 1: diagnosis, and pharmacological and psychosocial management.
• Bushby K, et al. (2010). Diagnosis and management of Duchenne muscular dystro-
phy, part 2: implementation of multidisciplinary care. Lancet Neurol; 9: 177–189.

Long-term NIV
• Finder J, et al. (2004). Respiratory care of the patient with Duchenne muscular
dystrophy: ATS consensus statement . Am J Respir Crit Care Med; 170: 456–465.
• Glasser M (2014). What a difference a day makes. www.thoracic.org/professionals/
clinical-resources/sleep/sleep-fragments/2012-april.php Date last accessed:
November 12, 2014.
the contribution of simple tools such as pulse oximetry, capnography, built-in
ventilator software and autonomic markers of sleep fragmentation. Thorax; 66:
438–445.
• Kinnear WJM (2010). Scoliosis. In: Elliott M, et al., eds. Non-invasive ventilation
and weaning: principles and practice. London, Hodder Arnold; pp. 379–388.
• Mokhlesi B (2014). Sleep hypoventilation: a state-of-the-art overview. Sleep Med
Clin; 9: xv–xvi.

Symptom palliation
NIV in palliative care and at

the end of life
Anna Maria Cuomo
“He…closed his eyes and narrowed his thoughts until his world shrunk
down to a single breath, in and out, in and out.”
From Tuesdays with Morrie by Mitch Albom
Palliative care is usually defined as any interventions aimed to prevent and
relieve suffering by controlling symptoms and providing other support to
patients and families in order to maintain and improve their quality of liv-
ing during all stages of chronic life-threatening (or terminal) illness. A lot of
emphasis has been given to this problem in oncology, while little attention has
been paid by other specialists, including pulmonologists. Palliative care may be
administered during all stages of chronic illness and, therefore, it should not
Key points
• Palliative care is defined as any intervention that aims to
prevent and relieve suffering by controlling symptoms and
to provide other support to patients and families in order to
maintain and improve their quality of life.
• Dyspnoea is the main symptom to palliate; it may be abrupt or
be present at various times, and may last for years.
• In the chronic setting, NIV may improve dyspnoea, quality of
life and exercise tolerance, but there is no clear evidence that
it affects mortality rate; it may prolong the time to potential
tracheostomy.
• In the acute setting, NIV may be used to relieve dyspnoea
and potentially treat a reversible factor of ARF (e.g. pulmonary
oedema or infection), or as a “pure” palliative tool in the case
of an irreversible process.
• The major concern over the use of NIV in terminally ill patients
is that it may unduly delay death with an unreasonable
persistence of treatment.

Symptom palliation
be considered only in the terminal phase of a disease. Indeed, although with

the use of pharmacological and nonpharmacological approaches, we are able
to prolong the survival of patients with chronic respiratory diseases until very
advanced stages of their natural course, prolonging survival ad libitum is not
always a desirable goal for either the physician or the patient, in the patient-
centred management of diseases. With the introduction of NIV to treat acute
ARF of different aetiologies >30 years ago, classical outcome measures, such
as hospital mortality, need for endotracheal intubation, complications of inva-
sive ventilation and length of hospital stay, have been drastically improved. The
goal and usefulness of NIV in palliative care has other meanings and outcomes,
which are centred on relief of the patient’s symptoms, and quality of life and/or
quality of death.
Table 1 presents the definitions of some terms commonly used in the end-of-life
context.
Which are the symptoms to palliate?

During the time course of their chronic respiratory disease, patients usually experi
ence symptoms including fatigue, asthenia, pain, anxiety, depression, constipa-
tion and sleep disturbances, but by far the most common symptom at any stage
of the disease is dyspnoea. Shortness of breath may ensue abruptly, as in the case
of an acute exacerbation of a chronic illness, or be present at various intensities
as a “companion” over a time frame that may last years. It is clear, therefore, that
the aim of “palliative” NIV in these two situations may have different meanings,
such as, on one hand, to give relief from a possibly “terminal” dyspnoea and, on
the other hand, to allow the patient to improve their tolerance to exercise when
NIV is applied chronically. Indeed even in the “acute” scenario, two situations
Table 1. Definition of terms used in the end-of-life context
Euthanasia A doctor intentionally killing a person who is suffering

unbearably and hopelessly at the latter’s voluntary, explicit,
repeated, well-considered and informed request
Withholding A planned decision not to institute therapies that would
mechanical otherwise be warranted (i.e. intubation)
ventilation
Withdrawal of Discontinuation of treatments that have been started
mechanical (i.e. decreasing FIO2 to 21%, extubation or switching off the
ventilation ventilator)
NIV as the ceiling Elective decision that the patient will not undergo intubation,
of ventilatory with NIV as the maximal interventional treatment
care
NIV on demand NIV mainly applied upon the patient’s request to relieve
dyspnoea and/or respiratory distress
Principle of The use of opiates, sedation or similar with the intention of
double effect palliating symptoms of dyspnoea, pain or distress, which is
not likely to shorten life

Symptom palliation
can be distinguished according to whether the shortness of breath is related to

a reversible factor (e.g. pulmonary oedema, infection or drug overdose) or is a
manifestation of an irreversible process (e.g. progression of a malignant, cardiac,
respiratory or neurological disease). When dyspnoea is related to a reversible
event, medical treatment is combined with interventions that target the cause,
like NIV, which, in this case, is meant not only to treat breathlessness but also to
prolong life. All this should be considered when starting NIV with the main aim
of relieving symptoms.
The timing and intent of NIV application in palliative care
The meaning of NIV in palliative care, in the different contexts of chronic and
acute use, is summarised in table 2.
Chronic use The efficacy of long-term NIV use has been widely discussed in the
past few years. While there is agreement that chronic NIV may prolong survival
in restrictive respiratory disorders (i.e. neuromuscular diseases) and, therefore, its
aim is no longer palliative, data on COPD patients are more controversial.
The official American Thoracic Society Clinical Policy Statement on Palliative
Care stated that treatment should focus on both the psychological and phys
ical components of dyspnoea, and rely on visual or analogue dyspnoea scales
to assess the severity of symptoms and, therefore, the beginning of the treat-
ment. A reasonable indication to start chronic NIV may be when chronic mod-
erate hypercapnic respiratory failure (i.e. PaCO2 >7.3 kPa (>55 mmHg)) ensues
together with an impairment of exercise capacity due to dyspnoea. When these
conditions are met, a substantial number of randomised and nonrandomised
Table 2. The intent of NIV application in palliative care
Chronic setting
Reduce dyspnoea versus long-term oxygen therapy
Improve exercise capacity versus long-term oxygen therapy
Ameliorate quality of life
Potentiate the effect of rehabilitation
Postpone eventual tracheostomy
Acute setting
Ameliorate breathing pattern
Reduce dyspnoea
Reduce respiratory muscle efforts
Relieve the sensation of “hunger for air”
Ameliorate sleep
Give time to say goodbye to loved ones or make final arrangements
Minimise the effects of opiates

Symptom palliation
studies have shown that dyspnoea, quality of life and exercise tolerance may
be improved, despite that treatment did not affect the mortality rate and/or
arterial blood gases. Indeed, sleep disturbances are quite common in these
patients and some studies have shown an improved quality of sleep using
chronic NIV. Interestingly, some forms of NIV (e.g. mouthpiece ventilation)
may be able to postpone the need for tracheostomy, especially in neuromus-
cular patients.
No data are currently available on respiratory patients affected by other chronic
disorders (i.e. pulmonary fibrosis, occupational diseases and pulmonary
hypertension).
Acute use The use of NIV in this scenario has gained more popularity, especially
in the past few years. Table 2 illustrates the potential indications for starting and
discontinuing NIV in this context.
The Society of Critical Care Medicine suggested an approach based on explaining
the goal of care when NIV is applied at the end of life. When the aim of NIV is
to palliate dyspnoea, it was concluded that in this circumstance, NIV should
only be considered successful if it improves the patient’s symptoms of dys
pnoea or other distress without causing other troubling consequences. Indeed,
patients in this category should not be encouraged to tolerate discomfort asso-
ciated with NIV because the primary goal of the therapy is patient comfort.
Most importantly, there is no justification for providing NIV to patients who
are unable to communicate whether NIV has improved their symptoms. There
are also patients who are interested in maintaining cognition and the ability to
communicate with their loved ones, or who want to delay death until a family
member arrives to say goodbye, or just because they want to arrange economic
or business affairs.
A recent pilot RCT comparing palliative use of NIV with oxygen therapy in end-
stage patients with solid cancer and ARF showed that NIV was more effective than
oxygen in reducing dyspnoea and respiratory rate. Importantly, this was achieved
with less use of morphine and, therefore, the patients were likely to be conscious
if they needed to say goodbye to their relatives. Despite these encouraging results,
none of the treatments was able to completely eliminate the sense of breathless-
ness that was reduced with NIV from an average value of 6 to 4 on the Borg scale,
which is still “moderate dyspnoea”. It should be noted that the results of this study
might not be generalisable because all the institutions involved had experience in
NIV, and used sophisticated ventilators with NIV platforms and high-technology
interfaces, that might not be available everywhere.
At present, therefore, the “acute” use of NIV as a palliative tool in terminally ill
patients is recommended only in motivated patients, especially they are left free
to withdraw the treatment whenever they want and for any reason.
Pros and cons of using NIV in this context

As stated earlier, there is still a lot of debate about the use of NIV in this context.
For chronic use, the issue is relatively simple, as for any chronic treatment, the
patient is perfectly competent and able to decide whether or not to continue ven-
tilation. Poor tolerance is mainly related to the well-known side-effects of chronic

Symptom palliation
NIV (mask intolerance, oronasal symptoms and poor humidification) and to the
perceived lack of effectiveness of the treatment.
In acute settings, in the best scenario, besides the already described benefits,
NIV may also improve survival in a small subset of patients (about 20–30%),
so that they may even be discharged home. Sleep quality could also be tem-
porarily ameliorated and, therefore, quality of death improved. However, major
concerns were raised about unduly delaying death with an unreasonable per-
sistence of treatment, leading to false hope for the patients and relatives. NIV
is usually applied in protected environments like the ICU or the respiratory
high-dependency unit, which may be not the “ideal” place to die because of
environmental (i.e. noise, light, and number of clinicians and visitors) and “philo-
sophical” factors (i.e. the lack of confidence of the doctors in sedating these
patients).
How to set palliative NIV in the acute setting
This is a proposed 12-point practical approach to a patient with ARF or distress
towards the end of their life.
1) Explain to the patient and relatives the aim and scope of NIV.
2) Show the patient the ventilator and types of interfaces, and get them to touch
the mask and try to put it on their face.
3) Gently apply the interface on the patient’s face with your hands and then
switch the ventilator on using pressure support 5 cmH2O and CPAP 2 cmH2O,
in order to accustom the patient to NIV.
4) Set the oxygen (through the ventilator or the centralised hospital source) to
achieve an SaO2 >90%.
5) Reassure the patient, and ask them about their feelings and sensations.
6) Increase the inspiratory pressure very gradually in steps of 2 cmH2O, accord-
ing to the patient’s tolerance.
7) After 10–15 min of application, ask clearly if the patient wants to stop the
treatment because of discomfort.
8) Fix the interface gently and progressively tighten it up. Do not attempt to
attain a mask seal yet.
9) When feasible, set the inspiratory flow rate (or ramp) in order to get the best
sensation for the patient. You need to ask whether they feel as if they have
enough, too much or too little air.
10) Once everything is set, there is no need to schedule arterial blood gas samples
or monitoring, providing that the patient can get hold of you or your staff at
any moment to eventually stop the treatment.
11) Discuss the schedule of application (i.e. continuous, a set period of NIV every
few hours, etc.).
12) In the case of loss of consciousness or withdrawal of NIV, immediately initiate
opiates and/or sedatives.
Withdrawal of NIV in long-term ventilatory support patients: the chronic scenario
There are now many long-term users of NIV with chronic respiratory failure. These
individuals often choose to use NIV until the end of their life, but in some, NIV
may be withdrawn at the end of life as a matter of autonomy and choice, e.g. in
patients with ALS/motor neurone disease, coincident terminal malignant disease

Symptom palliation
or end-stage COPD. If the individual is 24-h or heavily ventilator dependent, the

addition of opiates or a combination of opiate and sedative infusion to control
symptoms when the ventilator is withdrawn is entirely consistent with the prin
ciple of double effect, i.e. the intention is purely to relieve respiratory and emotional
distress and not hasten death. Once the patient is unconscious from opiate/sedative
infusion, e.g. as demonstrated by loss of the eyelash reflex, the ventilator may be
discontinued. Some authorities add low-flow oxygen (1–2 L⋅min−1) to prevent
terminal hypoxic fitting and improve comfort, but this is optional. This plan can
be discussed with patients and families in advance, and of course families should
be given the option to participate in medically agreed end-of-life management as
much as they wish. The withdrawal plan can also be carried out in the venue of
choice of the patient – either in hospital, in a hospice or at home with suitable
backup support.
Conclusions
Applying or proposing NIV as a palliative tool has gained popularity in the past dec-
ade but is still very controversial. The scope of this treatment should be very clear
to the patient and relatives and to the hospital staff from the beginning. Long-
term NIV may ameliorate some patients’ (i.e. hypercapnic and dyspnoeic COPD)
dyspnoea and exercise tolerance but may also be associated with low compliance
to treatment.
In the acute setting, NIV is associated with a decrease in dyspnoea score and
less use of morphine but it is thought that in certain circumstances, it may
prolong the process of dying (table 3). Further clinical trials are needed to focus
better the philosophy, feasibility and effectiveness of NIV in the context of pal-
liative care.
Table 3. When palliative NIV should be started and discontinued in the acute setting
Consider starting NIV if one or more of the following is present

Tachypnoea (i.e. respiratory rate >25 breaths⋅min−1)
Considerable dyspnoea (i.e. Borg score >4)
Signs of increased respiratory muscle efforts (i.e. massive use of accessory
muscles and/or paradoxical abdominal movement)
Sensation of “hunger for air”
Orthopnoea during sleep
Full comprehension of the rationale and meaning of NIV in this context by the
patient and relatives
The patient refuses the use of any pharmacological treatment for dyspnoea
Consider discontinuing NIV if one or more of the following is present
The patient refuses to continue
Intolerance to the interface or ventilation
Impaired mental status (Kelly score >3)

Symptom palliation
Further reading
• Azoulay E, et al. (2011). Palliative noninvasive ventilation in patients with acute
respiratory failure. Intensive Care Med; 37: 1250–1257.
• Curtis JR (2008). Palliative and end-of-life care for patients with severe COPD. Eur
Respir J; 32: 796–803.
• Curtis JR, et al. (2007). Noninvasive positive pressure ventilation in critical and
palliative care setting: understanding the goals of therapy. Crit Care Med; 35:
932–939.
• Lanken PN, et al. (2008). An official American Thoracic Society clinical policy state-
ment: palliative care for patients with respiratory diseases and critical illnesses.
Am J Respir Crit Care Med; 177: 912–927.
• Nava S, et al. (2013). Palliative use of non-invasive ventilation in end-of-life
patients with solid tumours: a randomised feasibility trial. Lancet Oncol; 14:
219–227.

Discharge and community care
Discharging the ventilator-dependent

adult and child
Joan Escarrabill
Ventilatory dependence is usually fairly arbitrarily defined as the requirement for

ventilatory support for at least 6 h per day for more than 1 month.
Starting treatment, hospital discharge, changes in the care team, or the emer-
gence of new health problems are all key points in healthcare. These situations
are periods of increased risk to the patient and require specific interventions that
may be developed conceptually through transitional care models. These strategic
points may also cause anxiety, and dealing with them must take local circum-
stances into account. There are no general rules, but promoting self-care via infor-
mation and training, ensuring that ventilator alarms and clinical warning signs are
understood and acted upon effectively, and planning follow-ups are very impor-
tant considerations.
Starting home mechanical ventilation (HMV) is a perfect example for understanding
the importance of transitions. HMV can be started during an acute admission,
or from a more stable baseline on an inpatient or outpatient basis. In whichever
scenario, it signals an important change in the patient’s life.
The basic agenda items when starting HMV are:
• Global assessment of the patient.
• Planned interventions by the multidisciplinary team.
• Structured therapeutic education for the patient and their carers.
• A follow-up programme.
Plans for discharge (or for the start of treatment) must be made individually,
evaluating alternatives with the patient, family and the carer, and taking local
Key points
• Educating patients and carers in self-care and problem-solving
is essential.
• Follow-up and risk management require the participation of all
stakeholders.
• Home discharge or discharge to an intermediate care facility
should be the aim, according to patient preference and
practicalities.

circumstances into account. Starting HMV is usually only one aspect of the overall
“care package” that the patient may require.
Plans for discharge centred on the patient’s needs must include information,
consensus and deliberation, and the process can be summed up in the following
steps: viability (feasibility), education and training, evaluation of the needs, follow-
up and management of the risk, and decision-making.
The first step in evaluating a patient in whom HMV is considered is to assess
viability. Viability means taking into account not only the technical aspects (clini-
cal stability, time without the ventilator, etc.), but also social aspects such as the
arrangements in the home, the views and competence of the carers, and the tech-
nical backup service available. From a health service perspective, one must be
aware of the requirements for funding the overall package of care required, and
how these should be negotiated. Ventilator-dependent patients are typically also
dependent upon a carer; existing family members may or may not be in a posi-
tion to take on such roles, and recruitment and training of an extended care team
always increases the complexity of discharge planning.
Discharge to the patient’s own home is often the preferred option but safe, effec-
tive care is a prerequisite. Achieving this in highly ventilator-dependent individuals
with significant care needs can be challenging. Intermediate or longer term insti-
tutional care is sometimes required. While such institutions may provide lifelong
care, they may also facilitate training of a home care team and provide intermit-
tent respite care. The degree of dependence on the ventilator, clinical stability,
autonomy for day-to-day activities, treatment complexity, and the availability of
suitable intermediate facilities must all be taken into account.
Education and training

Discharge planning should commence as early as possible after hospital admission
and the time spent in hospital should be used to develop a structured programme
of therapeutic education. The aims of the educational programme are to ensure
clinical effectiveness, to promote patient autonomy, and to develop the treatment
in the framework of the greatest possible safety. The educational programme:
• Must be designed according to the needs of the patient and the carer.
• A systematic examination must be made of the true abilities of the patient and
the carer to manage the ventilator and the access points to the airway.
• At regular intervals, the skills that the patient and the carer are expected to
have must be reassessed.
Evaluation of needs
In evaluating the needs of patients using HMV five basic items must be taken into
account.
Equipment In addition to the ventilator, all the equipment accessories that the
patient will need must be assessed and confirmed with stakeholders.
• Oxygen: home oxygen concentrator, emergency cylinder, portable cylinders
and oxygen tubing.
• Management of secretions: cough insufflation–exsufflation devices, suction
machine and humidifier.
• Spare tracheostomy tubes, smaller tracheostomy tube for emergency use and
mask interfaces as required.
• Batteries and spare parts, including disposables (e.g. filters and ventilator
circuitry).
• Emergency equipment: bag valve mask.
• Optional requirements (e.g. nebuliser).
Physical adaptation of the home Patients may need specific advice about how to
adapt the home for installation of the apparatus (e.g. hoist and electric bed), and
for transfers and mobility in the home. It is very important to make these adapta-
tions before the patient returns home. Stairs and lifts are obstacles to take into
account. Planning for these changes is a priority, as structural problems often
cause delays to discharge. Occupational therapy team members are key players
in this area.
Mobility Mobility is an important problem for patients who need a wheelchair, and
especially for patients who have little ventilator-free time. Wheelchairs need adap-
tation to transport the ventilator. Adapted cars are not accessible to all patients
(both because of cost and specific needs). It is essential to know the local avail-
ability of suitable transport to enable the patient to leave the house, socialise and
attend follow-up appointments.
Adaptation to daily activities Personal hygiene, the type of bed required, and the
different kinds of help in dressing and eating must be taken into account. It must
be kept in mind that, in some patients with neuromuscular diseases, the needs for
assistive technologies for daily activities can change very quickly. Support groups
(or in some countries “apparatus banks”) can be an alternative for making avail-
able backup technology that will be used for a short period of time.
Communication Adequate communication is absolutely vital and all possible meas-
ures should be explored to maintain it. Short-term solutions, such as tablets/
mobile devices, letter boards, pictograms and photographs, can be very useful in
the hospital environment and can reduce the frightening impact of the environ-
ment in a vulnerable patient. There are many technical alternatives for improving
communication (“augmentative and alternative communication systems”) and
early speech therapist involvement is recommended. In progressive neuromuscular
conditions, one must be mindful that a deterioration in function can lead to a new
inability to use the current means of communication. Considering the next step in
disease progression in advance is important.
Follow-up and risk minimisation
Planning discharge and organising the follow-up of patients requires the inter-
vention of a number of professionals in different healthcare fields. Moreover, in
some countries there is a need for intervention by companies that specialise in
supplying material or in providing technical services in the patient’s home (e.g.
ventilator service and maintenance). It is essential to have one single person
responsible for coordinating the whole process. One of the responsibilities of
the process manager is to define the procedures for minimising the risks. “Zero
risk” does not exist, but education and good communication minimise the risks.
Aspects to keep in mind when plans for minimising risk are being drawn up are
summarised in box 1.
Box 1. Aspects to consider when defining procedures to minimise risk

Contact (“hot line”):
• Direct telephone contact with the hospital team (ideally with
communication routed through the case manager)
• Technical service telephone number
• Telephone numbers of the essential resources in the community:
ambulances, primary care and social support resources
Alarms:
• The operation of the alarms must be checked periodically
• Checks should include limits, sound level and the appropriate response
to each type of alarm
Connection (especially in tracheotomised patients):
• Appropriate response to accidental disconnection or blockage of the
airway
• Presence of the skills required to keep the accessories suitably clean
Electricity:
• Put in place a plan adapted to each environment to respond to failures in
the electricity supply (e.g. internal or external battery)
• Having essential spare parts available (e.g. replacement ventilator)
Shared information:
• It is very important to ensure that all the professionals who have to
attend the patient have access to the clinical information
The follow-up plan must also focus on the needs and education of the carer.
Failure on the part of the carer can disrupt the whole care plan. For this reason it
is very important to assess the need for temporary respite periods for the carers;
if the carer is struggling to cope in a stable situation, then a small deterioration
in the patient’s condition can lead to a crisis. Respite requirements will vary on
an individual basis and can span from a few hours during the day, to temporary
support during the night, or for complete holiday periods. The care plans must
also include the response to problems on the part of the carer (surgical operation,
illness, etc.) that could require the patient to be admitted temporarily to an inter-
mediate healthcare unit. Respite solutions must also, of course, be acceptable to
the patient.
Decision making
The process of decision making is very important at the start of the treatment.
However, the patient’s situation may change over time (depending on the clinical
situation or the context) and it is especially important to establish an open discus-
sion on the decisions to be taken in the final stages of life. Knowing the patient’s
preferences and values is very important in facing situations that cannot always
be predicted or prevented. Whether or not to perform a tracheostomy, to admit
to the ICU or to admit to a palliative care centre are decisions of huge importance
that must be taken calmly and planned in advance.
Sometimes, determining whether HMV is providing continued benefit can be
difficult in the final stages of life. This discussion is essential for guaranteeing
the autonomy of the patient (it might be useful to discuss it in the framework of
advance directives) and to avoid futile practices. Not performing a tracheostomy
or not treating the patient in an ICU is not a way of rationing treatment but, if in
accordance with the patient’s wishes, is a way of reorienting the treatment.
Further reading
• Coleman EA, et al. (2006). The care transitions intervention: results of a rand-
omized controlled trial. Arch Intern Med; 166: 1822–1828.
• McKim DA, et al. (2011). Home mechanical ventilation: a Canadian Thoracic
Society clinical practice guideline. Can Respir J; 18: 197–215.
• Wise MP, et al. (2011). Home mechanical ventilation. BMJ; 342: d1687.
• Epstein RM, et al. (2010). Why the nation needs a policy push on patient-centered
health care. Health Aff (Millwood); 29: 1489–1495.
• Vitacca M, et al. (2007). Home mechanical ventilation patients: a retrospective
survey to identify level of burden in real life. Monaldi Arch Chest Dis; 67: 142–147.
• Simonds AK. (2006). Risk management of the home ventilator dependent patient.
Thorax; 61: 369–371.
• Simonds AK. (2003). Ethics and decision making in end stage lung disease. Thorax;
58: 272–277.
Online resources
• Escarrabill J (2014). How to discharge patients with NIV? ERS Course Noninvasive
ventilation: basic concepts, Hanover 2014. www.ers-education.org/events/

Home monitoring and follow-up

of long-term NIV
Dan Adler, Claudio Rabec and Jean-Paul Janssens
Home NIV refers to the long-term (>3 months), daily application of mechan

ical ventilation in the home setting. Prescriptions for home NIV have markedly
increased as a result of the increasing prevalence of COPD and OHS. Major techno
logical improvements in ventilators and interfaces, with a progressive shift from
volume-cycled to less expensive, lighter and often more comfortable pressure-
cycled ventilators have also contributed to this trend. The use of portable ventila-
tors has therefore shifted from the ICU to the hospital ward and to the patient’s
home over the past decades. Monitoring and follow-up of home ventilation has
evolved accordingly. It should now include a stepwise approach to detecting
respiratory events of clinical relevance, and a systematic evaluation of side-effects
Key points
• Improvement of diurnal PaCO2 with NIV therapy assessed
by arterial blood gases measurement is the most important
monitoring variable in chronic respiratory failure, as it is
associated with improved outcome. As a rule of the thumb,
diurnal PaCO2 and nocturnal SpO2 in patients not using
supplemental oxygen are good markers of NIV efficacy as
long as built-in software shows an acceptable pattern of use
without significant leaks.
• Transcutaneous capnography is increasingly used in most NIV
centres. Indeed, detection of subtle sleep hypoventilation is
possible despite normal diurnal PaCO2 and nocturnal SpO2.
• Important data (pattern of use, leaks and VT) downloaded
from the ventilator are also frequently used to adapt ventilator
settings, although the reliability of these data is not equivalent
between all devices.
• Use of polygraphy or polysomnography should be limited to
patients not reaching predefined clinical goals with first-line
monitoring.

related to NIV in order to adapt ventilator settings and interface. It should also
comprise a comprehensive and disease-specific evaluation of health-related qual-
ity of life (HRQoL).
Overnight monitoring of home NIV
Several reports have shown that patients under home NIV may develop patient–
ventilator asynchrony, unrewarded inspiratory efforts, unintentional leaks, peri-
odic breathing and obstructive or central events. These complex interactions
between patient and ventilator may have a detrimental effect on quality of sleep
and control of nocturnal hypoventilation. A stepwise approach has therefore been
recommended to detect respiratory events of clinical relevance in order to adapt
ventilator settings and interfaces.
Pulse oximetry and transcutaneous capnography
Pulse oximetry remains an important and very simple tool to ensure that adequate
SpO2 is provided and to detect short recurrent or prolonged desaturations, although
low specificity of pulse oximetry traces under NIV remains a major drawback.
Indeed, pulse oximetry should be interpreted with caution during NIV, as recurrent
SpO2 drops can represent obstructive residual events, central events (not sufficiently
treated or even induced by NIV) and intermittent mask leaks. A prolonged SpO2
drop is suggestive of sleep hypoventilation but can also occur because of mask
leaks (figs 1 and 2). The sensitivity of overnight pulse oximetry alone is also insuf-
ficient to exclude sleep hypoventilation during NIV even if very strict criteria are
applied to the SpO2 analysis (mean nocturnal SpO2 <92% or SpO2 <92% during >10%
of recording time). This is why some experts recommend both SpO2 and transcuta-
neous capnography be systematically recorded as first-line NIV monitoring.
Transcutaneous capnography (PtcCO2) is helpful to discriminate between hypox-
aemia related to ventilation–perfusion mismatch and residual hypoventilation
under NIV. The technique has the advantage of being noninvasive and does
not cause sleep disruption. Newer devices are reliable, provide a more realis-
tic picture of the overnight carbon dioxide trend and replace repeated arterial
blood samples. However, they remain expensive and the quality of measure-
ments is user dependent. Good agreement between PtcCO2 and PaCO2 has been
demonstrated in patients with and without pressure support. The American
Association of Sleep Medicine has proposed diagnostic criteria to define
sleep hypoventilation (Berry et al., 2012): an increase in PaCO2 (or a surrogate
marker) to a value ≥55 mmHg (7.32 kPa) for ≥10 min, or an increase in PaCO2
≥10 mmHg (1.33 kPa) during sleep (in comparison to an awake, supine value)
to a value exceeding 50 mmHg (6.65 kPa) for ≥10 min.
Built-in software
Built-in home ventilator software provides informative data about compliance,
pattern of NIV use, mask leaks, VT and percentage of inspiratory or expiratory trig-
gering by the patient.
Pattern of use The pattern of ventilation is an indirect indicator of tolerance to

NIV and comfort; for instance, multiple interruptions during the night after short
periods of NIV or an erratic pattern of use over several days are suggestive of poor
adaptation to NIV. Conversely, a rapid increase in NIV use can suggest imminent
a)
Lights off
W
R
N1
N2
N3
22:00 h 00:00 h 02:00 h 04:00 h 06:00 h 08:00 h
b) 75
70
PtcCO2 mmHg
65
60
55
50
45
40
22:00 h 00:00 h 02:00 h 04:00 h 06:00 h 08:00 h
c) 104
102
100
SpO2 %
98
96
94
92
22:00 h 00:00 h 02:00 h 04:00 h 06:00 h 08:00 h
Figure 1. a) Hypnogram; b) PtcCO2; c) pulse oximetry. As important leaks had been ruled
out by downloading the ventilator report, a prolonged drop in SpO2 was suggestive of sleep
hypoventilation and confirmed by direct measurement of PtcCO2. Rapid eye movement (REM)
sleep hypoventilation was also ruled out by concomitant electroencaphalography. W: wake:
R: REM; N: non-REM.
exacerbation if patients are not ventilator dependent (fig. 3). Although telemoni-
toring is at an early stage, it can be expected that continuous analysis of pattern of
use, respiratory rate, VT and percentage of triggered cycles will be used in the near
future to predict an exacerbation. Today, a detailed report on compliance remains
an important tool for:
1) deciding whether or not to pursue ventilatory support
2) discussing alternate patterns of daily use of NIV in cases of poor tolerance
during the night (i.e. daytime sessions)
3) understanding insufficient impact of NIV on PaCO2 or symptoms
Tidal volume VT is often underestimated by home ventilators unless a math
ematical algorithm computing leak-related pressure losses through the tubing
is incorporated in the device. In a bench test study of seven home ventilators,
underestimation of VT increased with higher pressures but leaks had little affect.
In a similar bench study, higher leaks worsened underestimation of VT. In any
case, estimation of VT by built-in software varies from one device to another and
a) 60
50
Respiratory rate
40
breaths·min-1
30
20
10
0
22:30 h 23:00 h 23:30 h 00:00 h 00:30 h
b) 1.0
0.8
0.6
Leaks L·s-1
0.4
0.2
0
22:30 h 23:00 h 23:30 h 00:00 h 00:30 h
c) 19
16
13
V'E L·min-1
10
7
4
0
22:30 h 23:00 h 23:30 h 00:00 h 00:30 h
d) 90
80
SpO2 %
70
60
22:30 h 23:00 h 23:30 h 00:00 h 00:30 h
Figure 2. Characteristic PAP-Reslink (ResMed, San Diego, CA, USA) trace showing a) re
spiratory rate, b) unintentional leaks, c) minute ventilation (V′E) and d) SpO2. Significant
unintentional leaks (>0.4 L⋅s−1) are associated with a prolonged decrease in SpO2 directly
measured by the device’s pulse oximeter. Reproduced and modified from Rabec et al. (2009).
a) 12:00 h
08:00 h
04:00 h
Time
00:00 h
20:00 h
16:00 h
12:00 h
1
2
3
4
5
6
7
8
9
20
21
22
23
24
25
26
27
28
29
30
31
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
WT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F
August 2014 September 2014
b) 12
Daily use h
0
1
2
3
4
5
6
7
8
9
20
21
22
23
24
25
26
27
28
29
30
31
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
WT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F S SMTWT F
August 2014 September 2014
Figure 3. a) Pattern of daily use and b) total daily use in a patient with poor tolerance to
treatment. Note the frequent interruptions during treatment, days with no treatment and
insufficient daily use. 6 days before hospital admission for acute exacerbation of COPD, the
patient complained of worsening shortness of breath. Dyspnoea was effectively alleviated by
NIV, as demonstrated by a rapid increase in daily use over the previous week. Days of the
week are indicated by their initial letters.
independent validation studies are warranted for all commercially available venti-
lators. Physicians should be aware of differences in the estimation of VT between
home NIV devices and should not use this information as their only monitoring
criteria.
Leaks Leaks have a major impact on the quality of ventilation. Thus, measurement
of leaks is a very useful adjunct to the information downloaded from the built-in
software. Yet, bench testing has demonstrated important gaps in leak estimation
between commercially available ventilators. Leaks are reported either as an esti-
mation of unintentional leaks or as an estimation of total leaks (i.e. leaks through
mask exhalation valves plus unintentional leaks). Manufacturers have determined
an arbitrary threshold for leaks below which pressurisation is considered effective.
Physicians monitoring patients on home NIV should be aware of differences in the
estimation of leaks between ventilators. Adjustment of interface and ventilator set-
tings should aim to maintain the 95th percentile of leak values under this predeter-
mined threshold. Multichannel ventilator modules combining SpO2 measurement
with estimation of leaks and V′E have also been shown to be reliable and provide a
useful graphical illustration of the consequences of leaks on ventilation (fig. 2).
Respiratory rate and percentage of respiratory cycles triggered by the patient The spon-
taneous respiratory rate of the patient and percentage of cycles triggered by the
patient retrieved from ventilator software must also be interpreted with caution.
Indeed, a low percentage of spontaneous inspiratory triggering may occur when
Flow
Diaphram Pressure
EMG
Abdomen Thorax
Figure 4. Unrewarded inspiratory efforts with inappropriate ventilator settings. This 60-s
epoch shows unrewarded efforts identified by comparing the electromyography (EMG) trace
or that of the abdominal or thoracic belts with the flow or pressure traces. The patient is
unable to trigger the ventilator, which is on its backup respiratory rate. Reproduced from
Adler et al. (2012), with permission from the authors.
the patient has a backup respiratory rate above their spontaneous respiratory rate.
This situation is similar to controlled ventilation. However, in the presence of leaks
or conditions in which inspiratory efforts may not be perceived by the ventilator
(i.e. patients with severe neuromuscular disease, intrinsic PEEP in severe COPD
or upper-airway closure), a low percentage of spontaneous inspiratory triggering
may in fact reflect the inability of the ventilator to perceive the patient’s respira-
tory rate. This is illustrated in figure 4. It has been a long tradition to “capture” the
patient with a backup rate in neuromuscular diseases. In OHS, evidence suggests
that using a backup rate may prevent central and obstructive respiratory events.
In COPD, evidence for increasing the backup rate is controversial: high-intensity
NIV with high inspiratory pressures and a high backup rate improves physiologi-
cal end-points and survival; however, a recent RCT demonstrated no advantage of
associating a high backup rate with high pressure support on night-time ventilator
adherence or on any of the other measured physiological parameters (including
PaCO2) (Murphy et al., 2012). These data suggest that it is the high-pressure com-
ponent of the high-intensity NIV approach that plays the therapeutic role in the
management of hypercapnic respiratory failure, at least in the COPD population.
Whether it is preferable to treat COPD patients with a high or a low rate of spontan
eous triggering remains an unresolved question. As a monitoring tool, a high percent-
age of patient-triggered cycles only decreases the likelihood of unrewarded efforts.
Respiratory polygraphy or polysomnography
Although simple tools should be used as first-line monitoring tools, increas-
ing awareness of nocturnal respiratory events occurring under NIV has led to a
wider use of polygraphy and polysomnography (PSG) for NIV titration. However,
full-night PSG is time-consuming, costly and not always available, depending
on healthcare resources. For this purpose, unattended home PSG under NIV has
recently been proposed. The clinical relevance of a detailed analysis of respiratory
events under NIV has not been demonstrated so far. We can thus only assume
that correct identification and treatment of residual respiratory events (through
modification of NIV settings) will improve quality of life and quality of sleep. More
research is needed to assess the impact of such an approach to patient-centred
end-points. Until efficacy of alternative monitoring options is reliably established,
PSG performed under NIV in an experienced centre remains the gold standard to
monitor NIV. The first step in using polygraphy/PSG to monitor NIV is to under-
stand the mechanisms involved in nocturnal respiratory events in order to adapt
the interface or the settings appropriately.
Leaks A fall in positive pressure associated with an increase in flow signal and a
decrease in amplitude of thoracic and abdominal belt tracings is suggestive of
leaks. Some manufacturers are now able to display unintentional leaks provided
by the ventilator software on the polygraphy traces, which is very useful to confirm
that the observed respiratory event is related to leaks. During leaks, detection of
inspiratory efforts by the ventilator can be challenged with a switch to backup
respiratory rate in assist-controlled mode (fig. 4).
Obstructive and central events A reduction in flow amplitude during insufflation

while positive pressure is maintained suggests upper-airway closure or instabil-
ity. Phase opposition between thoracic and abdominal belt traces suggests per-
sistent respiratory efforts and an obstructive respiratory event. Progressive and
smooth reduction of flow amplitude while the pressure signal is unchanged, with
simultaneous reduction or disappearance of abdominal and thoracic belt signals,
is suggestive of central respiratory events. It can be very difficult to differenti-
ate central and obstructive respiratory events under NIV unless the spontaneous
mode is used, and complete cessation of flow and effort can easily be observed.
As oesophageal pressure measurement (the gold standard) is invasive and poorly
tolerated, alternative means have been proposed to quantify respiratory effort
such as pulse transit time (PTT) (i.e. the time is takes for a pulse wave to travel
between two arterial sites, usually quantified as the time elapsed between the
R wave of the ECG and 50% of the amplitude of the pulse wave measured with
a finger probe). The variation of PTT between inspiration and expiration reflects
the intensity of corresponding changes in intrathoracic pressure. It was recently
demonstrated that changes in PTT adequately reflect changes in oesophageal
pressure and that PTT is useful to differentiate obstructive from central respira-
tory events occurring under NIV.
Other respiratory events of unknown clinical relevance Unrewarded inspiratory

efforts (fig. 4) can be a consequence of multiple pathophysiological pathways. In
neuromuscular diseases, the presence of unrewarded efforts often reflects the
inability of the diaphragm to trigger the ventilator during sleep. This can be com-
pensated by increasing the trigger sensitivity or by using a backup respiratory
rate just above the patient’s spontaneous respiratory rate. In OHS, unrewarded
inspiratory efforts and patient–ventilator asynchrony may be associated with
upper-airway instability or closure. Setting EPAP appropriately to treat obstruc-
tive apnoea and hypopnoea should improve ventilator triggering. In COPD
patients, unrewarded efforts can be related to dynamic hyperinflation. In this
case, reducing pressure support, increasing EPAP to counteract intrinsic PEEP

and setting the expiratory trigger (cycling criterion) at a higher percentage of
the peak inspiratory flow under close monitoring of PtcCO2 can improve patients’
ability to trigger the ventilator. If auto-triggering is demonstrated (rapid succes-
sion of pressurisations at a rate >40 cycles⋅min−1 not related to the patient’s
efforts), mask fit can be adapted to prevent leaks and trigger sensitivity can be
reduced. Double triggering can be avoided by setting the expiratory trigger at a
lower percentage of the peak inspiratory flow or by increasing pressure support
and/or inspiratory time.
As a rule, resorting to polygraphy or PSG to optimise NIV settings is recommended
only if targeted PaCO2 reduction is not reached, if recurrent or prolonged SpO2
drops are still observed despite correction of unintentional leaks or if NIV is poorly
tolerated.
Assessment of quality of life and side-effects
NIV is time consuming, cost intensive and can be associated with numerous side-
effects. It is therefore important to determine whether improvements in physio
logical end-points and mortality are associated with an acceptable HRQoL and
few side-effects. Correlation between HRQoL and physiological parameters com-
monly used to monitor chronic respiratory failure is poor and many important
domains of life affected by respiratory failure remain unexplored by classical tools
used to assess efficacy of NIV. The Severe Respiratory Insufficiency (SRI) ques-
tionnaire is an instrument specifically designed for measuring HRQoL in patients
treated with NIV. Condition-specific questionnaires, in contrast to generic ques-
tionnaires, are more sensitive and responsive to changes induced by treatment.
The SRI questionnaire was developed and validated in a German multicentre trial
that included 226 patients treated with NIV. It covers seven domains:
1) respiratory complaints
2) physical function
3) symptoms and sleep
4) social relationships
5) anxiety
6) psychological well-being
7) social functioning
Responsiveness of the SRI questionnaire to changes induced by the introduction
of NIV is superior to that of the 36-item Short-Form Health Survey, a widely-
used generic questionnaire. Translations into most European languages are
already available. The minimal clinically important difference of the SRI ques-
tionnaire has not been defined. However its use, or that of another disease-
specific questionnaire for patients with chronic respiratory failure such as the
Maugeri Respiratory Foundation Questionnaire-28, provides a backbone for the
investigation of HRQoL in these patients.
NIV is invariably associated with side-effects that could counterbalance the bene
fits in HRQoL associated with NIV. Therefore, side-effects induced by NIV should
be sought and recorded in all planned visits. The most common side-effects
reported in a recent publication were facial soreness caused by the mask (26%),
dry throat (25%) and sleep disruption (20%). Commonly encountered side-effects

Table 1. Most common side-effects related to NIV and their pragmatic management
Mask-related side-effects
Discomfort Adjust mask fit and headgear
Change mask type
Skin erythema Loosen straps
Artificial skin
Claustrophobia Smaller mask
Progressive approach
Consider: cognitive and behavioural
therapy, hypnosis, mild short-acting
anxiolytic drugs to be used with
caution
Nose ulceration Loosen straps of headgear
Artificial skin
Provide different interfaces with
different pressure points to be used
on alternate days or daytime versus
night-time
Steroids increase risk of ulcerations
Acneiform rash Topical steroids or antibiotics
Pressure- or flow-related side-effects
Leaks Switch to oronasal mask
Adjust straps of headgear
Try a chin strap if a nasal mask is used
Reduce pressure support and/or rise
time slightly
Nasal/oral dryness Heated humidifier
Nasal saline
Emollients
Correct leaks
Nasal congestion Heated humidifier
Nasal steroids
Sinus or ear pain Reduce pressure
Nasal steroids
Consider ENT advice for Eustachian tube
dysfunction or sinusitis
Eye irritation Check mask fit
Readjust straps
Gastric insufflation Reduce pressure support and/or rise
time if not tolerated (level of EPAP
is critical)
ENT: ear, nose and throat.

and their pragmatic management are listed in table 1 (see also the section entitled
“Choosing the interface”). They are most often related to mask fit, leaks, ventilator
airflow or pressure. Major complications such as aspiration pneumonia, pneumo-
thorax and haemodynamic instability are rare events.
Further reading
• Adler D, et al. (2012). Polysomnography in stable COPD under non-invasive venti-
lation to reduce patient-ventilator asynchrony and morning breathlessness. Sleep
Breath; 16: 1081–1090.
• Berry RB (2010). Noninvasive positive pressure ventilation titration and treatment
initiation for chronic hypoventilation syndromes. Sleep Med Clin; 5: 485–505.
• Berry RB, et al. (2012). Rules for scoring respiratory events in sleep: update of the
2007 AASM Manual for the Scoring of Sleep and Associated Events. J Clin Sleep
Med; 8: 597–619.
• Clini E, et al. (2002). The Italian multicentre study on noninvasive ventilation in
chronic obstructive pulmonary disease patients. Eur Respir J; 20: 529–538.
• Contal O, et al. (2013). Impact of different backup respiratory rates on the efficacy
of noninvasive positive pressure ventilation in obesity hypoventilation syndrome:
a randomized trial. Chest; 143: 37–46.
• Contal O, et al. (2013). Pulse transit time as a measure of respiratory effort under
noninvasive ventilation. Eur Respir J; 41: 346–353.
• Crescimanno G, et al. (2014). Monitoring noninvasive ventilation in neuromus-
cular patients: feasibility of unattended home polysomnography and reliability of
sleep diaries. Sleep Med; 15: 336–341.
Thorax; 65: 303–308.
• Fanfulla F, et al. (2007). Effect of sleep on patient/ventilator asynchrony in patients
undergoing chronic non-invasive mechanical ventilation. Respir Med; 101:
1702–1707.
• Gonzalez-Bermejo J, et al. (2012). Proposal for a systematic analysis of polygraphy
or polysomnography for identifying and scoring abnormal events occurring during
non-invasive ventilation. Thorax; 67: 546–552.
• Guo YF, et al. (2003). Respiratory patterns during sleep in obesity-hypoventilation
patients treated with nocturnal pressure support: a preliminary report. Chest; 131:
1090–1099.
• Janssens J-P, et al. (2011). Nocturnal monitoring of home non-invasive ventilation:
the contribution of simple tools such as pulse oximetry, capnography, built-in venti-
lator software and autonomic markers of sleep fragmentation. Thorax; 66: 438–445.

• Janssens J-P, et al. (2003). Changing patterns in long-term noninvasive ventilation:

a 7-year prospective study in the Geneva Lake Area. Chest; 123: 67–79.
• Jones PW (1995). Issues concerning health-related quality of life in COPD. Chest;
107: Suppl., 187S–193S.
ment of severe stable chronic obstructive pulmonary disease: a prospective, multi
• Luján M, et al. (2013). Effect of leak and breathing pattern on the accuracy of tidal
volume estimation by commercial home ventilators: a bench study. Respir Care;
58: 770–777.
• Murphy P, et al. (2012). High pressure versus high intensity noninvasive ventila-
tion in stable hypercapnic chronic obstructive pulmonary disease: a randomized
crossover trial. Int J Chron Obstruct Pulmon Dis; 7: 811–818.
• Nardi J, et al. (2012). Nocturnal oximetry and transcutaneous carbon dioxide in
home-ventilated neuromuscular patients. Respir Care; 57: 1425–1430.
• Pasquina P, et al. (2012). What does built-in software of home ventilators
tell us? An observational study of 150 patients on home ventilation. Respiration;
83: 293–299.
• Pépin JL, et al. (2014). Scoring abnormal respiratory events on polysomnography
during noninvasive ventilation. Sleep Med Clin; 9: 327–339.
• Rabec C, et al. (2009). Evaluating noninvasive ventilation using a monitoring sys-
tem coupled to a ventilator: a bench-to-bedside study. Eur Respir J; 34: 902–913.
• Storre JH, et al. (2011). Transcutaneous monitoring as a replacement for arte-
rial PCO2 monitoring during nocturnal non-invasive ventilation. Respir Med; 105:
143–150.
• Teschler H, et al. (1999). Effect of mouth leak on effectiveness of nasal bilevel ven-
tilatory assistance and sleep architecture. Eur Respir J; 14: 1251–1257.
• Windisch W (2011). Noninvasive positive pressure ventilation in COPD. Breathe;
8: 114–123.
• Windisch W, et al. (2003). The Severe Respiratory Insufficiency (SRI) question-
naire a specific measure of health-related quality of life in patients receiving home
mechanical ventilation. J Clin Epidemiol; 56: 752–759.
• Windisch W, et al. (2008). Impact of home mechanical ventilation on health-
related quality of life. Eur Respir J; 32: 1328–1336.

Assessing quality of life and outcome

of long-term NIV
Sophie E. Huttmann and Wolfram Windisch
Long-term NIV is a well-established method to treat patients with chronic

hypercapnic respiratory failure. Here, it is most frequently used in patients with
restrictive thoracic disorders, neuromuscular diseases, OHS or COPD. Of course,
long-term NIV is aimed at improving long-term survival, which is accepted as
the main outcome parameter in medicine. Notably, several observational studies
and uncontrolled trials have suggested a fundamental effect of long-term NIV on
survival in patients with restrictive diseases when compared with historical con-
trols. In these patients, NIV has been shown to be capable of substantially and
sustainably improving hypercapnia resulting in 5-year survival rates of ∼70–90%,
depending on the underlying disease. This is particularly true for patients with
restrictive thoracic disorders and for those with certain neuromuscular disorders
such as Duchenne muscular dystrophy. Based on this observation, no RCT to
assess survival benefits has ever been performed in these patients and will also
presumably never be performed, as a survival benefit in these patients is undis-
puted. Accordingly, any RCT in these patients would raise ethical concerns.
Key points
• Health-related quality of life and long-term survival are
the two main outcome parameters in patients receiving
long-term NIV.
• Health-related quality of life needs to be measured
as specifically as possible if highly specific treatment
interventions such as NIV are assessed for their outcome.
• The Severe Respiratory Insufficiency Questionnaire is a
highly specific, well validated and internationally adapted
measurement tool designed to assess health-related quality
of life in patients with chronic respiratory failure undergoing
long-term NIV.
• Long-term NIV has been shown to improve both long-term
survival and health-related quality of life in patients with
chronic hypercapnic respiratory failure.

In contrast, several RCTs have failed to establish a clear survival benefit in patients
with chronically hypercapnic COPD and 5-year survival rates were reportedly
∼50%, depending on the study. Most of these studies have also failed to dem-
onstrate a physiological effect of NIV, as hypercapnia was reportedly only non
significantly and inconsistently improved following NIV establishment. This failing
of NIV to have an effect on physiology and outcome, however, has been attributed
to insufficient NIV techniques with relatively low ventilator settings being used. In
contrast, high-intensity NIV aimed at improving hypercapnia typically uses higher
ventilator settings than conventionally applied and this approach has clearly been
shown to improve several physiological parameters, including PaCO2. As a conse-
quence, the most recent RCT has shown that 1-year survival rates were substan-
tially improved following NIV aimed at improving hypercapnia. Therefore, there is
now increasing evidence to support that long-term NIV is also capable of improv-
ing long-term survival in COPD patients if physiologically effective NIV techniques
are being applied (see the section entitled “Chronic NIV in COPD”).
Irrespective of the underlying disease, most of the patients suffering from chronic
hypercapnic respiratory failure have incurable and often chronically progres-
sive diseases. Therefore, it is debatable whether long-term survival is the most
important outcome in these particular patients: prolonging life could also mean
prolonging suffering. Based on this consideration, the emerging question is if life
gained by NIV is worthwhile, and this has led to the question of how long-term
NIV impacts health-related quality of life (HRQoL).
What is health-related quality of life?
HRQoL is defined as a multidimensional psychological construct that describes the
subjectively experienced health status, which is typically based on different com-
ponents of health including physical, psychological, social and functional aspects.
In addition to these main aspects, HRQoL is influenced by individual experiences,
expectations, beliefs and perceptions of a person. Importantly, HRQoL is mainly
affected by chronic underlying diseases. Treatment interventions aimed at directly
treating the underlying disease or treating the consequences of the disease, such
as symptoms, are expected to improve HRQoL. However, more or less every treat-
ment strategy also has the potential to deteriorate HRQoL, for example, if the
treatment is time consuming or causes significant side-effects.
Therefore, the assessment of HRQoL has become steadily more essential both
in healthcare practice and in research in order to evaluate the human and finan-
cial benefits and burdens of modern medicine. Most importantly, this is valid for
patients with chronic and incurable disorders such as those with chronic respira-
tory failure.
Assessment tools for HRQoL

Questionnaires are the standard instruments for the assessment of HRQoL in
large research collectives. In general, there are two different types of question-
naires: generic and specific. Generic instruments can be applied in different dis-
eases and, therefore, allow comparative, cross-sectional HRQoL assessments.
Disadvantageously, responsiveness and sensitivity to specific medical interven-
tions are not sufficiently captured by generic instruments. In contrast, specific
instruments focus on particular diseases or conditions, which limit the application
Table 1. Questionnaires for measurement of HRQoL most frequently used in patients receiving
long-term NIV
Instrument Developer Subscales n Items n

Generic
Hospital Anxiety and Depression Zigmond et al. 2 14
Scale (HAD) (1983)
MOS 36-Item Short-Form Health Ware et al. 8 36
Survey (SF-36) (1983)
Specific
COPD
Chronic Respiratory Questionnaire Guyatt et al. 4 20
(CRQ) (1987)
St George’s Respiratory Jones et al. 3 76
Questionnaire (SGRQ) (1992)
Chronic respiratory failure and NIV
Maugeri Foundation Respiratory Carone et al. 3 28
Failure item set (MRF-28) (1999)
Severe Respiratory Insufficiency Windisch et al. 7 49
Questionnaire (SRI) (2003)
to the specific disease or condition for which the tools have been developed.
Specific instruments, however, are mandatory when specific treatment interven-
tions are longitudinally evaluated for their effect on HRQoL changes, as these
instruments are suggested to most exactly measure changes in HRQoL following
specific treatment interventions. HRQoL measurement tools frequently used in
patients receiving long-term NIV are listed in table 1.
Maugeri Foundation Respiratory item set
The MRF-28 was especially developed for patients with chronic respiratory fail-
ure receiving either long-term NIV or long-term oxygen therapy. Alongside former
questions from well validated questionnaires (SGRQ, CRQ and Sickness Impact
Profile), new items were included. It contains 28 questions (items) in three
domains:
• daily activities
• cognitive function
• invalidity
However, this instrument was primarily validated in patients with COPD. Therefore,
application in patients with restrictive thoracic or neuromuscular disorders
appears limited.
Severe Respiratory Insufficiency Questionnaire
Today, the SRI is the most frequently used instrument to assess HRQoL in patients
receiving long-term NIV. It is highly specific for the condition of severe chronic
respiratory failure, which is a target for NIV. It contains 49 items in seven domains,
which can be summarised into one Summary Scale (fig. 1). Scores are provided
for all seven subscales and the summary scale, ranging between 0 and 100, with
higher scores representing better HRQoL.
The SRI has been well validated by several trials. Even though it is highly specific
for the condition of chronic respiratory failure, the SRI has been shown to be valid
for a broad variety of diseases, including restrictive thoracic diseases, neuromus-
cular disorders, OHS and COPD. It has good psychometric properties regarding
reliability, validity and responsiveness. In particular, good psychometric properties
have been further established in COPD patients following the initial validation trial.
Furthermore, the SRI has been shown to score best in the assessment of HRQoL
in COPD patients compared with other specific questionnaires, including the CRQ
and the MRF-28 (table 1).
International adaptation of the SRI
This SRI was originally developed in German but has also been professionally
translated into several languages. The international adaptation includes transla-
tion and back-translation of the original German SRI by two independent pro-
fessional translators followed by further adaptation in case of incongruence or
discrepancies between the original and back-translated items. Some of the trans-
lated versions have also been validated by additional studies using the profession-
ally translated SRI. The current status of the intercultural adaption of the SRI is
provided in table 2.
HRQoL in long-term NIV patients
Several studies have investigated HRQoL in patients undergoing long-term NIV
and the underlying disease has been shown to impact absolute HRQoL scores.
In general, patients with COPD and those with thoracic restriction resulting
from post-TB sequelae have more HRQoL impairments than patients with other
restrictive thoracic or neuromuscular disorders. In addition, stable patients
Respiratory Complaints Social Functioning

(SRI-RC) (SRI-SF)
Well-being
(SRI-WB)
SRI
Physical Functioning summary scale
(SRI-PF) (SRI-SS)
Anxieties
(SRI-AX)
Attendant Symptoms and Sleep Social Relationships

(SRI-AS) (SRI-SR)
Figure 1. The multidimensional model of the SRI.

Table 2. Translation and intercultural adaption of the SRI
Language Translation Validation Clinical trials

German Original version Yes Published
Spanish Professionally translated Yes Published
English Professionally translated Yes Published
Norwegian Professionally translated Yes Ongoing
Dutch Professionally translated No Published
French Professionally translated No Ongoing
Swedish Professionally translated No Ongoing
Polish Professionally translated Ongoing No
Japanese Professionally translated No Ongoing
Danish Professionally translated No Ongoing
Portuguese Professionally translated Planned Ongoing
under long-term NIV reportedly have only minor impairments in psychosocial

components of HRQoL, comparable to that of patients with other chronic dis-
eases not receiving long-term NIV. In contrast, physical health, as subjectively
reported by the patients, might be significantly impaired, but this is not neces-
sarily translated into impairments of mental or psychosocial components of
HRQoL. In particular, mental health was reportedly well preserved in patients
with Duchenne muscular dystrophy receiving NIV despite severe physical
disability.
Longitudinal studies have also shown that HRQoL substantially improves fol-
lowing the establishment of long-term NIV in patients with chronic hypercap-
nic respiratory failure. These HRQoL benefits are found in all patient categories,
including those with COPD. For example, the Summary Scale scores of the SRI
comparably improved in patients with neuromuscular disorders, restrictive
thoracic diseases, OHS and COPD in a large multicentre trial, even though the
SRI was capable of discriminating the different disease categories by differing
results for the subscale scores. HRQoL benefits following initiation of NIV were
also established by RCTs in COPD patients. Therefore, there are now robust data
to conclude that HRQoL significantly improves not only in restrictive patients but
also in those with COPD. Studies showing no HRQoL benefit following NIV estab-
lishment in COPD patients must be interpreted with caution, as the evidence of
HRQoL benefits in COPD is dependent on specific HRQoL measures being used
and on ventilatory strategies sufficient to improve hypercapnia (see the section
entitled “Chronic NIV in COPD”).
Conclusion
HRQoL is an important outcome parameter in patients with chronic respiratory
failure receiving long-term NIV and is suggested to be at least comparably impor-
tant as long-term survival. Specific measurement tools are required to assess
the impact of NIV on HRQoL. Here, the SRI is currently the most frequently used
instrument, as this questionnaire is highly specific, well validated and internation-

ally available following professional translations into several languages. Finally,
long-term NIV has been shown to be capable of improving long-term survival
and HRQoL in patients with most restrictive diseases and COPD presenting with
chronic hypercapnic respiratory failure.
Further reading
• Carone M, et al. (1999). Analysis of factors that characterize health impairment in
patients with chronic respiratory failure. Eur Respir J; 13: 1293–1300.
• German Respiratory Society. The Severe Respiratory Insufficiency Questionnaire.
www.pneumologie.de/808.0.html
• Guyatt GH, et al. (1987). A measure of quality of life for clinical trials in chronic
lung disease. Thorax; 42: 773–778.
• Jones PW, et al. (1992). A self-complete measure for chronic airflow limitation.
The St George’s Respiratory Questionnaire. Am Rev Respir Dis; 145: 1321–1327.
• Struik FM, et al. (2013). The Severe Respiratory Insufficiency Questionnaire scored
best in the assessment of health-related quality of life in chronic obstructive
pulmonary disease. J Clin Epidemiol; 66: 1166–1174.
• Testa MA, et al. (1996). Assessment of quality-of-life outcomes. N Engl J Med; 334:
835–840.
• Ware JE Jr, et al. (1992). The MOS 36-item short-form health survey (SF-36).
I. Conceptual framework and item selection. Med Care; 30: 473–483.
• Windisch W, et al. (2003). The Severe Respiratory Insufficiency (SRI) Questionnaire:
a specific measure of health-related quality of life in patients receiving home
mechanical ventilation. J Clin Epidemiol; 56: 752–759.
• Windisch W, et al. (2008). The Severe Respiratory Insufficiency Questionnaire was
valid for patients with COPD. J Clin Epidemiol; 61: 848–853.
• Windisch W (2013). Home mechanical ventilation. In: Tobin M. Principles and
Practice of Mechanical Ventilation. 3rd Edn. New York, McGraw Hill Medical;
pp. 683–697.
• Zigmond AS, et al. (1983). The hospital anxiety and depression scale. Acta Psychiatr
Scand; 67: 361–370.

Patient and carer education, and

risk management
Joan Escarrabill
Individuals who have to cope with disease, especially a chronic disease, must
be able to obtain, process and understand the basic health information that will
enable them to take decisions. This is known as health literacy and requires know
ledge of health problems, practical skills to make proper use of the health devices
and the equipment required, intellectual competencies and initiative. The skills and
competencies of the patient and carer are related to the clinical outcome.
There are tools for assessing the degree of skills and competencies of the patient
and carer and attention must be given to clues that can help identify patients
with deficits in learning capacity. Fortunately, major problems are uncommon, but
cognitive defects can be part of the underlying health condition of the patient,
and if present require particular tactics to address them. The deficit indicators are
summarised in table 1.
Patient and carer education in home mechanical ventilation (HMV) is a patient-
centred learning process, which includes the process of adapting to the disease
and the needs of the patient. Some of the general characteristics of the educa-
tional approach are summarised in box 1.
HMV education is a goal-centred process based on competencies. The education
process has three stages: understanding the rationale and processes of mechan
ical ventilation, hospital discharge and follow-up.
Key points
• Patient and carer education are crucial: the degree of skills
and competencies of the patient and carer is directly related to
clinical outcomes.
• In the case of home mechanical ventilation, education is a
goal-centred process based on competencies.
• It is essential to supervise and verify the practical skills of
patient and caregiver.
• Zero risk does not exist but access to clinical information, a
good technical service and a structured education programme
minimise the risk.

Table 1. Indicators that could suggest deficits in the skills and competencies of the patient
(and of the carer)
Type of problem Manifestations

Difficulty in Difficulty in orientation in the hospital
processing
Difficulty in reading:
information
The companion usually reads the instructions
Excuses for not reading (“I haven’t got my glasses”)
Difficulty in filling in forms
Difficulty in managing activities:
Arranging consultations
Forgetting
Difficulty in oral comprehension:
Difficulty in understanding common languages
Hearing problems
Following Low adherence to treatment
instructions
Difficulty in following instructions on personal or dietary habits
Difficulty in preparing for a test or procedure
Repetition of mistakes in spite of precise explanations
(oral and written)
Taking decisions Patients who have difficulty in processing information tend
to postpone decisions (especially if reading is required)
Rationale of HMV and the process of delivery

Therapeutic education in this phase has one basic objective: to ensure that the
patient and carer both know how to use the apparatus. This education is essential
for guaranteeing the safety and effectiveness of the treatment. The step prior to
the educational process is to assess the capacities of the patient, and those of the
principal members of the care team at home.
The educational process must be started at the same time as HMV is proposed and
must cover the following points.
• Basic knowledge of the disease. It is very important to explain the expected
course of the disease and possible complications.
• Reasons why access to the airway and a particular equipment/apparatus are
indicated. The advantages and limitations of the chosen therapeutic option.
• Description of the basic elements of the apparatus that is used and the correct
names to describe them.
• Technical aspects including: managing the access to the airway; starting and
stopping the ventilator, and connecting circuitry correctly; checking the alarms
and understanding the triggers for the alarms; identification of the most impor-
tant problems, e.g. power failure, tracheostomy tube blockage or displacement,
sputum retention, or chest infection; maintenance of the equipment, e.g. basic

cleaning, regular servicing according to the manufacturers’ guidelines and the
procedure for dealing with ventilator failure.
• Managing problems connected with the treatment: aspiration of secretions,
humidification, use of oxygen during or in the periods without ventilation, con-
densation in the circuit, temperature of the room, etc.
• Adaptation of the home (e.g. safe operation of hoists or electric bed).
• Mobility and communication aids.
• Key elements of clinical safety: identification of safe and unsafe SpO2 levels,
ability to respond appropriately to alarms, use of a problem-solving approach.
Training in the areas listed above should take place at the learner’s pace. Once
practical tasks can be carried out safely and repeatedly, competency can be signed
off by experienced staff and this verification recorded.
Discharge
The crucial moment is the day of discharge or the point when the patient/carers
have to perform the ventilation alone. Key elements to address are listed below.
• Ensure that the patient (and/or the carer) know how to perform all the neces-
sary procedures for the HMV by themselves. It is essential for a trained profes-
sional to supervise and certify this practical skill.
• The people responsible for the process have been clearly identified. From
the health professionals’ point of view it is very useful to have a nurse case-
manager to coordinate everyone involved. From a practical point of view a
check-list helps to make sure of all the elements of the process.
• Provide written information about the key points and the contact addresses for
emergencies.
• Allowing the patient and carer to be responsible for the delivery of care for a few
days before discharge allows the home team to gradually assume this role in a
safe environment where they can enlist help and ask questions as required. This
will also aid confidence, and should smooth the transition from hospital to home.
Follow-up
The skills and competencies of patients and carers must be reassessed period
ically, both for knowledge and for the ability to carry out the procedures correctly.
It must never be taken for granted that the patient and the carer are doing things
right simply because of their long experience with the disease. The intensity of the
information and the therapeutic education from the professionals is often diluted
over time.
It is very important to identify new players in the healthcare process, for example
changes to the care team in primary care or in the companies that supply services,
or incorporation of informal carers into the patient’s environment.
Whenever there are significant therapeutic changes or new equipment is intro-
duced, the whole educational process (specific for the changes) must be re-started
as described in the section on starting the treatment.
The whole educational process must be adapted to the special needs of the
patients, especially in those who have difficulties in understanding (due to age or
Box 1. General characteristics of a therapeutic education programme

• It is an integral part of the treatment and cannot be separated from it.
• It involves the patient and their family environment (and consideration
must be given to the informal carers or the circle of family/friends who
are involved in caring for the patient).
• It is a continuous process, with a learning curve (which must be carefully
supervised) and with changes according to the evolution of the disease.
• The educational process must be structured and systematic, using a
variety of media.
• Therapeutic education is a networking activity that involves all the
professionals (it is crucial to avoid contradictory messages).
• All the professionals in the team have educational functions; trained
professionals are required to direct and supervise the education
process.
• Evaluation is essential, both of the individual impact and the overall
results of the education programme.
• The patient’s participation is very important, both in designing the
programmes and in evaluating them (by evaluating the patient’s
experience or outcomes that the patient views as important).
• In evaluating the impact of therapeutic education it is important to
acknowledge and value the points of view of all the players involved in
the treatment.
Reproduced and modified from WHO Regional Office for Europe (1998), Therapeutic
Patient Education: Report of a WHO working group, with permission from the
publisher (World Health Organization Regional Office for Europe, Copenhagen).
cognitive disorders), difficulties in communication (due to bulbar disorder or poor

understanding of the language), changes in context (e.g. divorce from the carer or
change of address) or the presence of important comorbidities.
The patient and carer should be helped in preparing for long or short journeys that
are outside the daily routine. Air travel can be especially complicated for patients
who require ventilation, but problems are not insuperable. The following is a good
guide to commercial airline travel in ventilated patients: http://muscle.ca/wp-
content/uploads/2012/11/Commercial_air_travel_report.pdf
Risk management
Zero risk does not exist, even in the ICU. Access to clinical information for everyone
involved, a good technical service and, most of all, structured therapeutic educa-
tion, minimise the risk.
The strategy for minimising risks must include the following items:
• Ventilator
• Access to the airway
• Identification of the medical problems
• Communication
• Resources in the community
• Proactivity with new medical problems
Ventilator The correct operation of the ventilator is the primary safety element to

bear in mind. It is critical to ensure periodic servicing of the ventilator, including
functioning of the alarms and the recommended operational parameters. It must
be kept in mind that in some apparatus the patients might manipulate the venti
lation parameters voluntarily or accidentally.
Patients with very limited ventilator-free time must have a spare ventilator. This
recommendation can be extended to patients who, while having reasonable venti-
lator-free time (over 6 h), live far away from the hospital or for whom the technical
service cannot guarantee an immediate response.
It is necessary to foresee possible failures in the electricity supply and assess require-
ments for the provision of a spare battery (in patients with little ventilator-free time,
those who suffer frequent power cuts, or those who live far from urban areas).
In tracheostomy patients a bag valve mask is essential for ventilating the patient
in the temporary absence of an electricity supply, assessing airway patency, and
aiding clearance of lower airway secretions.
Access to the airway Patients with noninvasive access to the airway must always
have at least one spare tracheostomy tube with a connector identical to the one
they usually use.
Patients with tracheostomy (and their carers) must know precisely how to act
if the tracheostomy cannula becomes blocked by mucus (humidification or aspira-
tion if necessary) or what to do in cases of spontaneous decannulation (they must
have a smaller cannula to insert temporarily).
The type of access to the airway will dictate the management of secretions. The
requirement for permanent mechanical cough assist devices, or provision/loan of
these devices during exacerbation should be assessed, especially in patients who
might develop sudden and unpredictable worsening of pulmonary function and
secretion retention (as in the case of ALS patients).
Identification of medical problems One essential element of the education pro-
gramme is training patients and carers to identify problem situations. The cir-
cumstances of each patient will dictate this advice. Carers must know the basic
resuscitation techniques. They must also have a clear understanding of the
patient’s preferences with respect to end-of-life decisions.
Communication When the HMV is started, the patient and the carer must know the
contact details of the specialist care team, the care facilities in the community, the
companies that perform technical servicing, and of medicalised transport resources.
A hotline is very important for immediate connection to the key person within the
care team (generally the nurse case-manager). Box 2 provides a summary checklist
for home ventilator users and carers.
Resources in the community HMV involves different types of care that are designed
according to the needs of the patient and the resources available in the commu-
nity. The availability of informal carers, complementary services at home (clean-
ing, hygiene, food supply, etc.), day centres and other community resources can be
very important in minimising risks. Availability of respite care for the patient and
family carers is a vital part of the care package and may help avoid carer burnout.
Box 2. Summary checklist for home ventilator users and carers

• Make sure that you have been fully trained to use the ventilator and
that you get training updates. Do you know who to contact to get
trained again? Have you received written instructions on how to use the
ventilator?
• Ensure the backup battery (internal or external battery) is fully charged
and functions correctly in the event of a power failure. Follow the
ventilator manufacturer’s instructions for battery management, e.g.
running the ventilator from battery once a month to allow an estimate
of the exact running times on specific patient settings, and test general
battery performance.
• Are you sure you know what to do in an emergency, for example in
ventilator failure or a power cut? If you are not sure, ask your clinical
team to explain again.
• Tell your electricity company that you are classified as priority in case of
an emergency, such as a power cut.
• Make sure the alarm is set loud enough so you can hear it wherever
you are in your home. Can you hear it over the top of other noises
(such as the TV, hi-fi, vacuum cleaner, washing machine)? Think about
maybe using a remote listening device similar to a baby monitor, if this
will help.
• If you go to hospital, take the ventilator and breathing system with you.
• If you are planning a holiday contact your home ventilation centre for
advice.
Reproduced from Medicines and Healthcare products Regulatory Agency (MHRA)
(2012), Tips on domiciliary ventilators used in life support, with permission from the
MHRA under delegation of authority from the Controller of MHSO.
Proactivity with new medical problems Patients with HMV may have new clinical
problems, not necessarily connected with the process that causes the need for
ventilatory support. One of the most frequent problems is the need for elective
surgery. For major surgery it is important to know in advance how the access to
the airway might be affected in the post-operative phase (e.g. in the case of facial,
dental or upper airway surgery) and the impact of the surgery on pulmonary func-
tion (e.g. abdominal surgery). A proactive preoperative approach is very important,
if possible, as is close post-operative monitoring by a team that is experienced in
HMV (see the section entitled “NIV in the perioperative period”).
Further reading
• Davis TC, et al. (2004). Health literacy: implications for family medicine. Fam Med;
36: 595–598.
• Dewalt DA, et al. (2005). The role of literacy in health and health care. Am Fam
Physician; 72: 387–388.
• Farré R, et al. (2006). Performance of mechanical ventilators at the patient’s home:
a multicentre quality control study. Thorax; 61: 400–404.

• Key P, et al. (2006). A citizen’s guide to chronic disease care. Geneva, Foundation
for Research and Training in Patient Education. Available from: www.desg.org/
desg/teaching-tools/a-citizens-guide-to-chronic-desease-care
• Norregaard O, et al. (2010). Patient and caregiver education. In: Elliott M, et al.,
eds. Non-invasive Ventilation and Weaning. London, Hodder Arnold.
• Safeer RS, et al. (2005). Health literacy: the gap between physicians and patients.
Am Fam Physician; 72: 463–468.
• Simonds AK (2006). Risk management of the home ventilator dependent patient.
Thorax; 61: 369–371.

NIV service set-up
Setting up and staffing your NIV unit
Joan Escarrabill
The delivery of an NIV service must take into account at least four scenarios:
1) Care for patients in the weaning stage after admission to the ICU.
2) Acute initiation and management of NIV in different hospital locations, e.g.
respiratory wards, emergency rooms and high-dependency units (HDUs).
3) Care of acute exacerbations or intercurrent medical procedures in patients
who receive NIV at home.
4) Assessment of NIV in stable patients and long-term follow-up.
The first two situations refer to acute care of critical patients so the organisation
will be closely linked to the ICU. The latter two cases refer to the care of patients
who need long-term NIV. Many hospitals have different care provisions for acute
patients and those receiving long-term NIV. Variability between countries and
within individual countries is great.
Complexity ceiling
In long-term NIV the service is organised with a view to the level of complexity of the
population catered for, and their ventilatory requirements. It is neither reasonable
nor beneficial for the patient or public to think that all health centres can provide
comprehensive NIV services. In general terms three kinds of unit can be described:
• Units capable of performing NIV and invasive ventilation in all patient groups,
with multidisciplinary teams and complete diagnostic and therapeutic
resources. They function as therapeutic reference or hub centres.
• Units capable of implementing NIV in noncomplex patients, but which do
not have comprehensive facilities and have to refer complex patients to the
reference centres (e.g. they refer patients with neuromuscular diseases who
require NIV).
• Local centres that occasionally deal with NIV patients but are not capable of
implementing long-term NIV.
Key points
• NIV is performed optimally by a multidisciplinary team.
• Technical and non-technical skills are equally important in
caring for NIV patients.

NIV service set-up
From this perspective, NIV units must work as a network, with a hub that can
solve the more complex problems and local centres that can take on part of the
care. The distribution of care and responsibilities between the network nodes will
depend on the patient’s situation and can also change over time depending on
these needs.
Professional skills
The professionals who work in an NIV unit must have technical skills related to
ventilation. However, in caring for patients with serious chronic diseases, non-
technical skills are very important (team work, participating in a deliberative model
for taking decisions that affect the patient, communication skills, empathy, etc.).
Box 1 summarises these skills.
NIV patients may need care from different professionals: doctors (pneumologists,
neurologists, and ear, nose and throat specialists), advanced practice nurses,
respiratory physiotherapists, nutritionists, speech therapists, etc. In addition, venti
lators are often supplied by specialist companies that also have professionals
available for patient care. The role of the case-manager nurse is very important,
especially in more complex cases. This is a bidirectional role. On one hand, it
involves coordinating the roles of all the professionals involved and helping the
patient navigate through the healthcare system, while on the other hand it identi-
fies the patient’s needs and seeks the most appropriate response.
Technical resources
Problems with availability of technical resources arise when NIV has to be per-
formed outside the ICU. The alternatives vary greatly but three things must be
borne in mind:
1) Whenever possible NIV must be performed in a stable physical unit (this
ensures availability of equipment, team building and the involvement of
trained professionals).
2) NIV patients must be supervised 24 h a day by trained staff (nurses or physio
therapists, with medical specialists on duty or in phone contact).
3) There must be a minimum monitoring level to assess the efficacy of the venti
lation (box 2).
When NIV is performed outside the ICU it is very important to have an established
route of access to invasive ventilation.
For the provision of long-term ventilation it is essential to have a reference clinical
team of a size appropriate to the number of patients to be treated. Technical care
is very variable, but cover must be guaranteed 24 h a day, with special care for
patients with little or no spontaneous breathing ability off the ventilator.
Care at home
Home care is crucial in NIV. Three situations can be identified in which home care
plays a very important role.
1) At the start of ventilation A home visit when ventilation starts is very impor-
tant in understanding the true home environment, both for aspects related
to safety and in order to recommend possible adaptations that could make

NIV service set-up
Box 1. Technical and non-technical skills of professionals who attend NIV patients
Technical skills
• Knowledge of the ventilators available in the community. There is no
need to be familiar with all the models, but professionals must have
sufficient knowledge to respond to patients’ needs.
• Knowledge of the various noninvasive interfaces/devices for access to
the airway.
• Skills in managing secretions.
• Knowledge of the basic elements for monitoring and evaluating the
efficacy of NIV.
Non-technical skills
• Teamwork: ability to maintain relationships of trust, to critically evaluate
one’s own activity and willingness to seek help.
• Knowledge of the resources in the community.
• The ability to integrate home NIV use during hospital admission.
• The ability to design care plans that respect the patient’s preferences
while promoting self-care.
• The ability to design personalised therapeutic education programmes for
patient and carer.
• Communication skills for active listening and for engaging and enthusing
support groups.
Information from Escarrabill et al. (2008).
treatment easier to manage. Accessibility (stairs, lift and doors), hoists and
bathroom fittings are aspects to be taken into account.
2) Permanent technical follow-up There must be a guaranteed permanent tech
nical service, both for periodic maintenance of the apparatus and to deal with
breakdowns or technical problems that may arise.
3) Home care during times of crisis The ability to attend to a patient at home is very
variable (it requires resources, trained staff and accessibility). However, a good
home care team can prevent visits to the emergency department and admis-
sions in NIV patients.
Telemedicine can be a facilitator of home care in transmitting signals and generat-
ing alarms, and also in communication between the patient and the care team.
It has been used widely in some countries (e.g. Spain and France) and rarely in
others. When a telemedicine service is offered, the capacity for responding to the
information that is generated must be clearly identified.
In some patients NIV is not viable at home, so alternatives must be found that do
not require hospitalisation in an acute facility. Intermediate care varies greatly from
country to country, but the training of the professionals involved is very important.
These training needs should be identified as soon as possible in the discharge plan-
ning process, otherwise discharge may be delayed.

NIV service set-up
Box 2. Suggestions for staffing an acute NIV service

• A named consultant with appropriate training should have overall
responsibility for the NIV service. This will usually be a consultant
respiratory physician. [D]
• NIV can be provided in a number of locations including the ICU, an HDU,
or a respiratory ward. However, each hospital should have a specific
designated area with an available cohort of staff with appropriate
experience, together with structures to ensure that patients requiring
NIV can be transferred to this area with the minimum of delay. [C]
• The clinical area in which a patient is treated with NIV will be influenced by
several factors including their clinical state, whether they will be intubated if
NIV fails, and the availability of beds. Taking into account the overall clinical
picture, patients with more severe acidosis (pH <7.30, H+ >50 nmol⋅L−1)
should be managed in a higher dependency area such as an HDU or ICU, as
should those in whom improvement in clinical state and arterial blood gas
tensions is not seen after 1–2 h of NIV on a respiratory ward. [C]
• Patients with acute hypercapnic respiratory failure from a cause where
the role of NIV is not yet clearly established (such as pneumonia, ARDS,
asthma) should only receive NIV in an HDU or ICU where facilities for
immediate tracheal intubation are available. [C]
• There should be a clear protocol for the on-call medical staff as to the
indications for NIV, how to initiate treatment, and who has continuing
responsibility for supervision of the patient. [D]
• Trained ICU staff, doctors, physiotherapists, lung function technicians,
and nurses can successfully set up and maintain NIV. When setting up
an acute NIV service, it is recommended that NIV be initiated and run by
nursing staff. [C]
• All staff involved in an acute NIV service should receive training
appropriate to their baseline knowledge and role in providing the service.
Training in NIV should be available for consultants in respiratory medicine
and should be included in all specialist registrar training programmes. [D]
• A training programme for the provision of an NIV service should provide
a combination of knowledge-based learning supported by clinical
experience in the workplace. [D]
• The use of NIV in ARF should be the subject of regular audit. In addition
to collection of data on patients receiving NIV, details of the number
of patients admitted with acute hypercapnic respiratory failure will be
required, together with the use of invasive ventilation in these patients. [D]
[C]: SIGN (Scottish Intercollegiate Guidelines Network) grading recommendation based
on a body of evidence including studies rated as 2+ (well conducted case-control or
cohort studies with a low risk of confounding or bias and a moderate probability that the
relationship is causal), directly applicable to the target population, and demonstrating
overall consistency of results; or extrapolated evidence from studies rated 2++ (high
quality systematic reviews of case–control or cohort studies, or high quality case–control
or cohort studies with a very low risk of confounding or bias and a high probability
that the relationship is causal). [D]: SIGN grading recommendation based on evidence
rated 3 (non-analytical studies, e.g. case reports, case series) or 4 (expert opinion); or
extrapolated evidence from studies rated 2+. Reproduced from British Thoracic Society
Standards of Care Committee (2002), with permission from the publisher.

NIV service set-up
Responses to crises and emergencies

For inpatients, emergencies will be handled under hospital procedures. In dealing
with NIV patients it is important to take into account not only technical problems
but also the personal circumstances of the patient and carer. Some important
points to bear in mind are:
• The availability of a call centre or hotline to solve technical problems 24 h a day
and ideally for contact with the care team for as much time as possible.
• Designing the response to clinical deterioration, both in the patient’s home and
during transfer to hospital. It is crucial to avoid using the emergency services
as far as possible.
• Special attention to the carer’s problems. Illness on the part of the carer or the
need for a break to unwind will require a specific intervention. Respite care
for the patient and family carers should always be part of long-term planning
arrangements.
Structured therapeutic education programmes are essential for responding to the

special needs of NIV patients.
Staffing
Acute NIV service The British Thoracic Society Standards of Care Committee sug-
gestions for staffing an acute NIV service are shown in box 2.
Long-term care team The team responsible for the long-term ventilator-depend-

ent patient in the community will have a coordinator (e.g. specialist nurse) and
involve a range of players (fig. 1). Outreach team workers from the hospital acute
team can act as useful problem-solving intermediaries and aid in the training of
Specialist
Family
doctor
Patient Community team

Home care General practitioner
Specialist
ventilation Other consultants
physiotherapist
team Nurses
Dietician
Occupational therapist
Specialist Physiotherapist
nurse Social care
Palliative care
Figure 1. The team responsible for long-term ventilator-dependent patients in the community.
NIV service set-up
community staff and family carers. The key is that the person delivering an aspect
of care is trained and competent to perform this specific task; hence, competency
training and education is more relevant than the designation of staff (e.g. nurse,
doctor, physiotherapist or carer) delivering it. Staff, patient and carer education
is discussed further in the section entitled “Patient and carer education, and risk
management”.
Further reading
• British Thoracic Society Standards of Care Committee (2002). Non-invasive venti-
lation in acute respiratory failure. Thorax; 57: 192–211.
• Chatwin M, et al. (2010). Analysis of home support and ventilator malfunction in
1,211 ventilator-dependent patients. Eur Respir J; 35: 310–316.
• Escarrabill J, et al. (2008). Training the home health team. In: Ambrosino N, et
al., eds. Ventilatory Support for Chronic Respiratory Failure. 1st Edn. New York,
Informa Health Care.
• Muir JF, et al. (2010). The chronic ventilator service. In: Elliott M, et al., eds.
Non-invasive Ventilation and Weaning. London, Hodder Arnold.

Index
A airway clearance see airway clearance
failure 57
abdominal surgery 137–138 interfaces 32
accessory respiratory muscles 94 monitoring see monitoring
acidosis 41–42, 292 paediatric 79–85
active cycle of breathing technique 87–88 service, staffing 292
active exhalation valve 15 stopping/starting 102–110
acute cardiogenic pulmonary oedema see also specic indications
(ACPO) 72–78 acute-on-chronic hypercapnic respiratory
acute hypoxaemic failure 72–78 failure 60–66, 61f
causes 72 interstitial lung diseases 64–65
clinical trials 74–77 non-COPD obstructive lung disease 61–64,
diagnosis 72–73 62f
medical therapy 73, 75f, 76f acute respiratory distress syndrome (ARDS) 67
NIV benets 73–74 children 81
NIV starting/settings 73 intubation avoidance 68
acute exacerbation of COPD (AECOPD) mild 70
41–48, 238 NIV failure 105
audit 46 NIV use 149
future care planning 46 acute respiratory failure (ARF)
high-ow oxygen therapy 39, 40 asthma 62–63
initial management 41–42 causes 61
interface choice 43–44 children 79–81
medical management 112 critical issues, NIV implementation 109
monitoring 42–43 hypercapnic see acute hypercapnic failure
NIV contraindications 42 hypoxaemic see acute hypoxaemic
NIV failure 44–45, 45f, 46 respiratory failure
NIV initiation 42 interfaces 30–32, 31f
NIV physiological effects 121, 191 monitoring 93–101, 94f, 95t
outcome predictors 41–42 NIV audit 292
patient deterioration 44–45 NIV failure 106–107
patient selection 41–42 obesity 105
readmission prevention 193 in OHS, NIV and 57–59, 58f
troubleshooting 45–46 perioperative period, prevention of 136
ventilator settings 44 practical issues 103–104
acute hypercapnic failure starting/stopping NIV 102–110
causes 103 ventilator choice 103
chest wall deformity 49–55, 104–105 adaptive servo ventilation (ASV) 6, 15,
COPD 104, 106, 106f 212–213, 215f
critical issues, NIV implementation 109 adenotonsillar hypertrophy 218, 219
interfaces 103 adenotonsillectomy 219
neuromuscular disease 49–55, 104–105 adult care transition 223–224
NIV indications 104–105 Advent-HF trial 213
non-COPD patients 104–105 aerosol-generating procedures 151
OHS 56–59 ageing 240
treatment location 104, 292 air travel 285
upper airway obstruction, children 80 airway clearance 86–92
acute hypoxaemic respiratory failure 67–71 assisting factors 89–90
acute cardiogenic pulmonary oedema humidication 89
72–78 hydration 89
alveolar damage 67 inhaled therapies 89–90, 91f
inammation 68 interfaces 208
meta-analyses/systemic reviews evidence manual techniques 87, 87f
69–70 neuromuscular patients 90f
NIV NIV as adjunct 207–208
benets 67 protocolised approach 89, 90f
outcomes 68–69 techniques 86–87, 87f
patient selection 70 see also individual techniques
physiological rationale 67–68 airway intubation 10
settings 70 alarms 17, 81, 263, 287
oxygenation improvements 67 alert cards 54
acute NIV 7 alfentanil 144
alveolar hypoventilation 177 behavioural modication strategies, children
American Thoracic Society Clinical Policy 222
Statement on Palliative Care 255 bilevel positive airway pressure (bilevel PAP;
amyotrophic lateral sclerosis (ALS) 176–181 bilevel ventilation)
assessment 177, 178t acute cardiogenic pulmonary oedema 74, 77
bulbar involvement 177–178 as airway clearance adjunct 207–208
care organisation 180 children 219–221, 220f
carer support 180 chronic heart failure 212, 213f, 214–215,
case study 113–115 214f
end stages of life 180–181 neuromuscular diseases 171, 219–221
equipment 178–179, 179t perioperative period 136
ethical dilemmas 180–181 respiratory pandemics 152
home care 180 biochemical tests 97
multidisciplinary intervention 179–180 breathing control 87
NIV criteria 177–178 breathing exercises 87–88
NIV failure 239–240 breath stacking 171, 227, 228
physiological support 180 bronchiectasis
survival increases 176 chronic NIV 204–210, 205t
tracheostomy ventilation 178 as airway clearance adjunct 207–208
ventilation initiation 176 bridging to transplant 206–207
analgesia/sedation, bronchoscopy 144 complications 209
antibiotics, neuromuscular disease 53 failure 209
anticipatory care plans 49 cystic brosis 63
anxiolytics 235 hypoventilation 206
apnoea/hypopnoea periods, ASV 15 mortality rates 239
apparatus (device) banks 179, 262 NIV-assisted exercise training 205t, 208
arterial blood gases (ABGs) NIV failure 239
acute exacerbation of COPD 43, 238 sleep hypoxaemia 206
acute respiratory failure 97 survival rate 5, 5f
chest wall disorders 185–186 upper-airway obstruction 205–206
COPD 239 bronchiolitis 63
correction 241 bronchoalveolar lavage (BAL) 143
failure to correct by NIV 236–239 bronchodilators 53, 89
arterial carbon dioxide tension (PaCO2) bronchoscopy
COPD 191, 192, 193–194, 238 adverse effects 145
noninvasive monitoring 97 analgesia/sedation 144
obesity-related chronic respiratory failure equipment 143
200 hypoxaemia, risks in 142–143
assist-control mode 21 interfaces 144, 144f, 145f, 146f
assisted insufflation 251 neuromuscular disease 51, 52f
assisted pressure-controlled ventilation (APCV; NIV 143–145
ASSPCV) 21 application 144–145
assist mode 21 benets 143
asthma 62–63, 80 contraindications 143
atelectasis 135 oral introduction 144f, 145, 146f
audit 46, 109, 292 PaO2 143
auto-expiratory positive airway pressure pulmonary shunt fraction 143
(auto-EPAP) 11, 14 bronchoscopy elbow 144, 145f, 146f
autogenic drainage 87–88 button tracheostomy 159
auto-triggering 12, 21, 21f, 272
average volume-assured pressure support
(AVAPS) 7, 249
C
call centre 293
Canadian Positive Airway Pressure Trial 212
B capnography 97
backup rate 11 carbocysteine 89
acute exacerbation of COPD 44 carbon dioxide removal, vented/non-vented
ALS 114 interface 30t
COPD 270 cardiac surgery 137
heart failure patients 215 cardiology, NIV in 6
neuromuscular disease 270 cardiomyopathy 50
OHS 270 cardiovascular monitoring, noninvasive,
perioperative period 139 ARF 97
bacterial infection 90 care delivery, future developments 7
bag valve mask 286 carers
bariatric surgery 136 communication 286, 287
Becker muscular dystrophy 169, 173 discharge planning 261
education 282–288 chronic/long-term NIV 190–196
follow-up plan 263 clinical effects 194–195
problems 293 hospital readmission 195
respite 263, 287, 293 monitoring 193
tracheostomy, burden of 161 patient selection/indications 192–193,
case fatality rate 148 192f
case-manager nurse 291 physiological considerations 191
case studies 111–117, 246–252 physiological effects 193
central core myopathy 51, 165t, 170 practical approach to 193
central respiratory events 271 problems/pitfalls 194
central sleep apnoea (CSA) rationale for 191
with Cheyne–Stokes respiration ventilatory strategy impact 191–192
(CSA/CSR) 212 chronic respiratory failure 191
heart failure and 211, 212–213 failure to trigger ventilator 116
cephalic mask see full face mask high-ow oxygen therapy systems 39
channels of Lambert 86 high-intensity NIV mode 20
channels of Martin 86 insufficient oxygenation 112–113
chest trauma patients 69–70 lung surgery 136
chest wall compliance 226 mortality rates 239
chest wall disorders 182–189, 183t NIV failure 97, 239
acute hypercapnic failure 49–55, 104–105 over-oxygenation 112
acute pneumothorax 53–54 post-extubation failure treatment 130
anticipatory care plans 54 quality of life 279, 280
case study 251 survival rates 5, 5f, 195, 277
decision making 54 target oxygen saturation 112–113
high-risk patients 185–186 unrewarded respiratory efforts 271–272
NIV 183–184 weaning 122
contraindications 187 chronic pulmonary failure, COPD 191
initiation 187, 188 chronic respiratory failure (CRF)
long-term 251 causes 61, 219
monitoring 187 children 219
patient selection 185–187 COPD 191
palliative care 54 interfaces 31f, 32
respiratory system effects 182–183 NIV withdrawal 257–258
sputum clearance problems 186 obesity-related 198, 200–202, 201f
survival rates 184, 186f chronic ventilatory failure, COPD 191
ventilatory support types 184–185 circuits, NIV 22–25
see also individual disorders Cobb angle 182, 184f
Cheyne–Stokes respiration (CSR) 6, 211 cognitive behavioural therapy (CBT) 234–235
children 79–85 cognitive defects 282, 283t
chronic NIV 217–225 common cold 86
indications 218t community care 260–288
monitoring 223 community resources 287
outcomes 223 compliance failure, NIV 235–236, 236f
controversial issues 84 compliance reports 246–247, 267
discontinued ventilatory support 223 congenital central hypoventilation syndrome
extubation 80–81 (CCHS) 221
interfaces 82–83 congenital muscular dystrophy (CMD) 163,
NIV 165t, 169, 173
failure indications 83–84, 83t congenital myasthenia syndrome 221
in ICU 83 congenital myopathies 165t, 170
indications 79–81, 80t congestive heart failure (CHF)
initiation 222–223, 222f comorbidities 212
nocturnal hypoventilation 5–6 sleep disordered breathing 211, 212–213
palliative NIV 221 continuous positive airway pressure (CPAP) 12
perioperative period 221 acute cardiogenic pulmonary oedema 73,
transition to adult care 223–224 75f, 76f, 77
ventilator choice 81 acute hypoxaemic respiratory failure 68
chin straps 32, 115, 115f central sleep apnoea and heart failure 212
chronic hypercapnic respiratory failure 61, 280 children 82, 217–219, 223
chronic NIV see long-term NIV chronic heart failure 214
chronic obstructive pulmonary disease (COPD) historical aspects 4
acute exacerbation see acute exacerbation nonresponders 212
of COPD (AECOPD) obesity 136
acute hypercapnic failure 104, 106, 106f, 191 obesity-related chronic respiratory failure
case study 111–113 200, 202
OHS 198, 200 long-term NIV 167–169, 168f, 249–251
OSA 4, 136, 206, 212, 217–219 medical care 51
perioperative period 136, 139 mouthpiece ventilation 239, 250, 250f
tracheobronchomalacia 205 new symptoms 173
contraindications, NIV 105 NIV and cough assist techniques 50
controlled mode 21 NIV failure 239
coordinator 293 respiratory failure 163
COPD see chronic obstructive pulmonary survival/outcomes 5, 223
disease (COPD) tracheostomy ventilation 169
coronaviruses, novel 150 transition to adult care 224
cough dye test 158
effective 226 dyspnoea
ineffective 108, 226–227 interfaces 22–23
particle/infection spread 151 palliative NIV 254–255
cough assist techniques
ALS 179
neuromuscular disease 50–51
E
echocardiogram 73, 168
spinal muscular atrophy 170 efficacy, of NIV 234
cough augmentation 226–233 electricity supply 263, 286, 287
expiratory techniques 228 electrocardiogram 168
inspiratory and expiratory techniques end of life 253–259, 254t
228–230 ALS 180–181
inspiratory techniques 227–228 decision making 263–264
cough efficiency measurement 94–95 see also palliative care
craniofacial syndromes, OSA 218–219 endoscopic procedures 142–147
creatinine, neuromuscular disease 51 interfaces 146, 147, 147f
custom-made masks 28, 115 risks 142
cycling 12 see also individual procedures
cystic brosis (CF) 63, 80 endoscopic retrograde
bronchiectasis 63 cholangiopancreatography (ERCP) 146–147
chronic NIV 204–210 endotracheal intubation
as airway clearance adjunct 207–208 historical aspects 2–3
bridging to transplant 206–207 neuromuscular disease, acute chest
complications 209 infections 51–53, 52f
failure 209 OHS, acute decompensated respiratory
exacerbation treatment 63 failure 57
hypoventilation 206 equipment 10–40
interfaces 207 exercise performance, COPD 195
NIV-assisted exercise training 205t, 208 expiration, delayed cycling to 116
NIV use 63 expiratory airow measurement 226, 227f
OSA 205 expiratory positive airway pressure (EPAP) 10–11
prognosis 239 acute exacerbation of COPD 44, 46, 113
sleep hypoxaemia 206 as airway clearance adjunct 208
upper-airway obstruction 205–206 bridging to transplant 207
bronchiectasis 207, 208
D chest wall disorders 185
delirium 95 children 82, 223
dermal dressings 32 chronic heart failure 214, 215
device (apparatus) banks 179, 262 cystic brosis 207, 208
diaphragm failure to trigger ventilator 116–117
COPD patients 191 interstitial lung diseases 207, 208
surgical trauma 135, 137 neuromuscular disorders 171
discharge 260–288 obesity-related chronic respiratory failure
follow-up 262–263 200
patient education 284 obstructive lung disease 11
plans 260–261 obstructive sleep apnoea/hypopnoea
risk minimisation 262–263 syndrome 247–248
disease transmission 30 OHS 57, 247–248
double triggering 272 OSA 21
dual-limb circuits 16f extubation 90f
dual-mode ventilation 19 failure 127
Duchenne muscular dystrophy
cardiac surveillance 168
case study 249–251
F
fall time 12
creatinine phosphokinase 51 exible bronchoscopy see bronchoscopy
ow alarms 17 home mechanical ventilation (HMV) 265, 266
ow amplitude 271 carer education 285, 286
uid balance assessment 97 communication 262
forced expiration technique (huff ) 87–88 daily activities, adaptation to 262
forced vital capacity (FVC), OHS 200 decision making 263–264
forehead spacer 33 elective surgery 287
full face mask 23, 27f, 29t equipment 261–262
acute exacerbation of COPD 44, 112 home care 290–291
ARF 32 times of crisis 291
bronchoscopy 144, 144f home visits 290–291
children 82 mobility 262
endoscopy 147, 147f needs evaluation 261–262
neuromuscular disease 115 NIV 265, 266
respiratory pandemics 152 patient education 261, 282, 285
functional residual capacity (FRC) delivery process 283–284
chest wall disorders 182–183, 185f discharge 284
general anaesthesia 135 follow-up 284–285
medical problems identication 286
G new equipment 284
rationale for 283–284
gas exchange assessment, ARF 95
gastric distension 250 stages 285
gastric hyperination 145 technical aspects 283–284
general anaesthesia 135 patient’s special needs, adapted to
Glasgow Coma Scale (GCS) 95 284–285
glossopharyngeal breathing 228 physical adaptation of home 262
gravity-assisted positioning 87 starting 260
technical follow-up 291
H viability assessment 261
home NIV 265, 266
H1N1 inuenza 149–150
hotline 263, 286, 293
pregnancy 150
huff (forced expiration technique) 87–88
haemoptysis 209
humidication 36–38
headgear/straps 115, 236
aims 37
healthcare workers, infection control risk 151
airway clearance 89
health literacy 282
bronchiectasis 208
health-related quality of life (HRQoL) 277
cystic brosis 208
assessment 272, 276–281
dry mouth/nose 247
tools 277–278
home ventilators 38
COPD 194–195
natural 37
long-term NIV patients 279–280
system types 37
heart failure patients, chronic NIV 211–216
hybrid mode 19
positive pressure systems titration
children 82
214–215
hypercapnia
heat and moisture exchanger (HME) systems
acute exacerbation of COPD 41–42
37, 38t
chronic, NIV failure to correct 236–239
heated humidiers 37, 38t, 247
long-term NIV 277
helmets 27f, 29t
myotonic dystrophy 170
acute exacerbation of COPD 44
persistent 111, 112
anti-asphyxiation valve 29
vessel dilation induction 191, 192
ARF 32
hypercapnic encephalopathy 95, 96t
children 82
hypercapnic respiratory failure
humidication systems 37
acute see acute hypercapnic failure
hypoxaemic respiratory failure 103–104
children 79–81
perioperative period 139
chronic 61, 280
respiratory pandemics 152
COPD 191
high-dependency unit 292
hypoventilation 60–61
high-ow oxygen therapy systems 38–40, 39f
medical team 103
hypercapnic exacerbation, COPD 39
weaning 122–123
neuromuscular disease 54
hyperglycaemia 97
NIV vs. 39
hypersalivation, ALS 180
high-frequency chest wall compression 88–89
hypertonic saline 89–90
high-frequency chest wall oscillation
hypnosis 234
(HFCWO) 88–89
hypotension 70
high-intensity NIV
hypoventilation
COPD 20, 191, 192, 195, 270
bronchiectasis 206
practical approach 193
cystic brosis 206
historical aspects, NIV 1–5, 4f, 18
hypercapnic respiratory failure 60–61 developments 7
kyphoscoliosis 115 disposable 30
nocturnal see nocturnal hypoventilation endoscopy 146, 147, 147f
hypoxaemia tting 32
bronchoscopy 142–143, 145 humidication systems 37–38
persistent 111, 112 hypoxaemic respiratory failure 103
hypoxaemic respiratory failure intolerance 45, 107, 234–235
acute see acute hypoxaemic respiratory leaks see leaks
failure long-term use 30
causes 60, 103 monitoring 97–98
children 80t, 81, 83, 83t neuromuscular disorders 171–172
NIV failure 83, 83t, 105 pain associated with 236
NIV indications 105 perioperative period 139
NIV risks 105 removal 32
hypoxia, kyphoscoliosis 251 respiratory pandemics 152
safety considerations 29–30
I side-effects 273t
technical compatibility 28, 28t, 30t
immunocompromised patients 69
indications for NIV 41–78 see also individual types
with higher mortality rate 239–241 interstitial lung diseases (ILDs)
paediatric 79–85 acute-on-chronic hypercapnic respiratory
infection control risk 151 failure 64–65
infection transmission, aerosol/droplet chronic NIV 204–210
151–152 as airway clearance adjunct 207–208
inuenza see H1N1 inuenza complications 209
inhaled therapies, airway clearance failure 209
89–90, 91f deterioration causes 64
inspiratory muscle strength, COPD 191 hypoventilation 206
inspiratory positive airway pressure (IPAP) 10 lung-protective ventilatory strategy 64
acute exacerbation of COPD 44, 46, 113 NIV-assisted exercise training 208
as airway clearance adjunct 208 OSA 205
bridging to transplant 207 palliative NIV 65
bronchiectasis 207, 208 sleep hypoxaemia 206
children 82, 223 upper-airway obstruction 205–206
chronic heart failure 214, 215 intrapulmonary percussive ventilation 88
COPD 20, 191, 192 intrinsic positive end-expiratory pressure
cystic brosis 207, 208 (iPEEP), children 82
interstitial lung disease 207, 208 invasive mechanical ventilation (IMV) 10
neuromuscular disorders 19 in NIV failure 45f, 108f
obesity 136 NIV vs. 10, 79
obesity-related chronic respiratory invasive positive-pressure ventilation via a
failure 200 tracheostomy (TPPV)
OHS 20, 57, 248 after NIV failure 241
OSA 21 ALS 240
thoracic cage disorders 19 Duchenne muscular dystrophy 239
inspiratory time (tl) 11, 57 historical aspects 2–3
institutional care 261, 291 indications 241
integral mask see full face mask OHS 241
intelligent volume-assured pressure support iron lung 1–3, 3f, 18
(IVAPS) 7, 171
intensive care delirium screening checklist 95 K
intensive care unit (ICU) 118–162, 292 Kelly–Matthay score 95, 96t
interface(s) 22–25, 24f kyphoscoliosis 182, 183t, 185f
accessories 32 case study 115–117, 251
acute exacerbation of COPD 43–44, 112 hypoventilation 115, 185
acute hypercapnic failure 103 long-term NIV 251
acute setting 30–32, 31f ventilator–patient dyssynchrony 115–117
advantages/disadvantages 28 kyphosis 182
airway clearance 208
bridging to transplant 207
bronchoscopy 144, 144f, 145f, 146f
L
late cycling 11
categories 26, 27f leaks
children 82–83 acute exacerbation of COPD 45–46
choosing 26–34, 28t bedside estimation 98
in chronic setting 31f, 32 case study 113–115
complications 97–98 causes 114
cycling, effects on 12 hypercapnic respiratory failure 103
humidication systems 37 NIV unit 293–294
intentional 23, 98 respiratory pandemics 152
neuromuscular disease 113–115, 114f training 103
nocturnal hypoventilation 266, 267f, 268f mental health 280
prevention/correction 236 Middle Eastern respiratory syndrome
reduction 32 coronavirus (MERS-CoV) 150
single-limb circuits 98 mid-facial hypoplasia 171–172
total 269 minicore myopathy 165t, 170
triggering 12 minitrach 53
unintentional 98, 269 minute volume 11
poor compliance 235–236 monitoring
ventilator estimation of 98, 99f, 268f, 269 acute NIV 93–117, 94f, 95t
leak ventilation, children 83 choices 93–101
left ventricular afterload 121 clinical 94–95
lidocaine 144 location choice 100–101
limb girdle muscular dystrophy (LGMD) 169 NIV-related monitoring 97–100
lining dressings 32 physiological 95–97
lobectomy 137 bedside evaluation 94, 111
long-term NIV 163–252 bronchoscopy 143
approaches to surgery 171 in chest wall disorders 187
care team 293–294, 293f at home 265–275
case studies 246–252 long-term NIV 265–275
complexity ceiling 289–290 children 223
failure 234–245 COPD 193
follow-up 265–275 neurological status alteration 95
home monitoring 265–275 neuromuscular disorders 167
outcomes 5, 5f perioperative period 139–140
assessment 276–281 puberty 223
quality of life assessment 276–281 rationale for 93
survival benets 276 ventilator software 266–270
see also individual diseases/disorders motor neurone disease see amyotrophic lateral
lower airway obstruction, children 80 sclerosis (ALS)
low-intensity NIV 191 mouthpiece(s) 27f, 29t
lung compliance reduction 226 airway clearance 208
lung insufflation capacity 227–228 mouthpiece ventilation 231
lung surgery, COPD 136 nebuliser delivery 90
lung transplantation 137 NIV-assisted exercise training 208
NIV bridging to 206–207 mouthpiece ventilation (MPV) 159,
lung volume recruitment bag 227 230–232, 231f
benets 230–231
M Duchenne muscular dystrophy 168, 239,
250, 250f
malignant hyperpyrexia 51
mannitol 89 equipment 231–232
manually assisted cough (MAC) 228, 229f, neuromuscular disease 230
251 securing mouthpiece 231–232
masks see interface(s) spinal cord injury 230
mask ventilation, historical aspects 3 starting setting 232
Maugeri Foundation Respiratory item set mucus plugs 229
(MRF-28) 278 multidrug-resistant tuberculosis 153
maximum inspiratory time (tlmax) 116 multiminicore disease 51
maximum insufflation capacity (MIC) 227 multipoint headgears 33
mechanical insufflation-exsufflation devices myotonic dystrophy 166t, 170, 173
89, 228–230 myotubular myopathy 165t, 170
airway spasticity 228–230
complications 228–229 N
Duchenne muscular dystrophy 251 nasal dryness 247
initial settings 229, 230 nasal masks 23, 23t, 24f, 27f, 29t
modes 229, 230 acute exacerbation of COPD 43
neuromuscular disease 51, 52f airway clearance 208
mechanical ventilation 102–103 chronic respiratory failure 32
clinical signs of need for 105 cystic brosis 207
novel coronaviruses 150 humidication systems 37
mechanism of action, NIV 102–103 perioperative period 139
medical team nasal pillows 27f, 43
nasal prongs 29t ARDS 105
nebuliser, via mouthpiece 90 bronchiectasis 239
negative-pressure rooms 152 causes 107–109, 108f
negative pressure ventilation (NPV) 1–3, 18, children 83–84, 83t
184–185 clinical signs/indications 106, 106t
nemaline myopathy 165t, 170 COPD 97, 239
neurological status alteration 95 Duchenne muscular dystrophy 239
neuromuscular disease/disorders early 107
acute chest infection management 50–51 early identication 94
acute hypercapnic failure 49–55, 104–105 H1N1 inuenza 150
acute pneumothorax 53–54 higher rate indications 239–241
acute respiratory care, issues impacting hypoxaemic respiratory failure 105
on 49–50 immediate 234–235
airway clearance 90f inappropriate ventilator settings 108
anticipatory care plans 49, 54 interface intolerance 107, 234–235
bulbar weakness 49 invasive mechanical ventilation effects 108f
chest infection prevention 53 late 107, 108
chronic NIV 163–175 long-term NIV 234–245
approaches to surgery 171, 172 missed 107
intercurrent events management 172–173 non-compliance 235–236
monitoring 167 noninvasive failure 45, 45f
outcomes 167–170 obese patients 239
patient selection 163–167 OHS 57
pregnancy 172–173 palliative NIV 240
starting/timing of 167 poor NIV application 108
ventilator choices 171–172 risk factors 106, 107t
decision making 54 Steinert’s myotonic muscular
dual-mode ventilation 19 dystrophy 240
endotracheal intubation 51–53, 52f technical 44
extubation 90f tracheostomy following 241
failure prevention 130 ventilation failure 45, 45f
general medical care 51 during weaning 124
mask tting problems 114 NIV service/unit
mask leaks 113–115, 114f care at home 290–291
new symptoms 173 complexity ceiling 289–290
NIV, issues impacting on 49–50 deliver scenarios 289
NIV initiation 220–221 emergencies/crises 293
nocturnal hypoventilation 220–221 on-call medical staff 292
palliative care 54, 173 professional skills 290, 291
personalised care plans 221 set-up 289–294
respiratory complications 164–166t staffing 292, 293–294
sleep disordered breathing assessment 221 acute service 292, 293
survival rates 5, 5f, 6, 276 long-term care team 293–294, 293f
tracheostomy 51–53 technical resources 289, 292
unrewarded respiratory efforts 271 types 289
upper limb weakness 50 nocturnal carbon dioxide monitoring 206
NIV (noninvasive ventilation) 10 nocturnal hypercapnia, heart failure 212
acute see acute NIV nocturnal hypoventilation
chronic/long-term see long-term NIV chest wall disorders 185
efficacy/success, conditions 234 children 5–6
failure see NIV failure chronic heart failure 213f
future developments 7 correction strategies 238–239
high-intensity see high-intensity NIV mask leaks 266, 267f, 268f
implementation in clinical practice 7 neuromuscular disease 167, 220–221
interfaces see interface(s) OHS 247
matching modes/settings to patient 18–25 non-compliance, NIV 235–236, 236f
mechanism of action 102–103 non-COPD obstructive lung disease 60–66
palliative see palliative NIV non-cystic brosis bronchiectasis 63–64
present trends 6 noninvasive ventilation see NIV (noninvasive
“real” world practice audit 109 ventilation)
stepping up/down from 109, 155–162 non-vented circuit 98, 99f
see also specic topics relating to NIV non-vented masks 28, 30t
NIV failure 234
acute exacerbation of COPD 44–45, 46
acute respiratory failure 106–107
O
obesity
ALS 239–240 ARF 105
early airways closure 198 oxygen therapy
extubation failure 130 high-ow see high-ow oxygen therapy
mortality rates 239 systems
NIV failure 239 obese patients 202
NIV in perioperative period 136 OHS 57
obesity hypoventilation syndrome (OHS) palliative NIV vs. 256
198–200 post-extubation failure prevention
acute decompensated respiratory failure 128–130
57–59, 58f
acute hypercapnic failure 56–59
AVAPS 7
P
paediatrics see children
bedside tests 198–200 pain, equipment-related 236
case studies 246–249 palliative care 6, 253–259
chronic/long-term NIV 197–203, 246–249 chest wall disorders 54
follow-up 202 neuromuscular disease 54, 173
lifestyle advice vs. 198 NIV in see palliative NIV
comorbidities 57 palliative NIV 253–259
isolated 197, 199f, 200 acute use 254–255, 255t, 256, 257,
malignant 56 258, 258t
NIV in 56–59, 58f breathlessness in end-stage disease 108
OSA and 197, 199f, 200 children 221
overnight assessment 200 chronic use 255–256, 255t, 256–257
sleep study 247, 248f failure 240
unrewarded respiratory efforts 271 intent of 255–256, 255t
ventilator modes 20 interstitial lung diseases 65
obesity-related chronic respiratory failure oxygen therapy vs. 256
198, 200–202, 201f pros/cons 256–257
obstructive bronchiolitis 63 quality of death 257
obstructive lung disease 88 respiratory pandemics 152
obstructive respiratory events 271 successful 256
obstructive sleep apnoea (OSA) symptoms to palliate 254–255
characteristics 211 timing 255–256
children 217–219 withdrawal 108–109, 257–258
aetiology 218–219 pandemics 148
anatomical factors 218–219 respiratory see respiratory pandemics
upper airway collapsibility 219 passive exhalation valve 23, 25
congestive heart failure 211 pathophysiology 102–103, 103f
craniofacial syndromes 218–219 patient education 261, 282–288
cystic brosis 205 home mechanical ventilation see home
heart disease and 6 mechanical ventilation (HMV)
interstitial lung diseases 205 patient selection 40–78, 105–106
isolated severe 197, 198f patient–ventilator asynchrony/dyssynchrony
OHS and 197, 199f, 200 21f, 100, 100f
ventilator modes 21 acute exacerbation of COPD 46, 112
obstructive sleep apnoea/hypopnoea bridging to transplant 207
syndrome (OSAHS) 247–248 case study 115–117
oesophageal pressure measurement 271 causes 116
oral dryness 247 failure to trigger ventilator 116–117
oral masks 27f neuromuscular disease 114
oronasal masks 23t, 27f, 29t solutions 116–117
acute exacerbation of COPD 43 patient–ventilator interaction assessment
ARF 30–32 100, 100f
children 82 peak cough ow (PCF) 94–95, 226, 227, 227f
cystic brosis 207 ALS 179
dyspnoea 22–23 percussion 87, 87f
humidication systems 37–38 percutaneous endoscopic gastrostomy (PEG)
NIV-assisted exercise training 208 147, 250
OHS 57 percutaneous gastrostomy, ALS 180
perioperative period 139 perioperative period 135–141
respiratory pandemics 152 ARF prevention 136
oscillatory devices 88–89 children 221
outreach team workers 293–294 high-risk patients 136
oxygen blender 35 interfaces 139
oxygen saturation, COPD 42, 112–113 NIV
oxygen supplementation see supplemental application 136, 138–140
oxygen contraindications 138, 138t
indications 138t weaning 121–122
monitoring 139–140 pressure-targeted ventilation, children 82
venue 139 principle of double effect 258
pH, acute exacerbation of COPD 41–42 principles of treatment 103f
phase opposition 271 problem-solving 111–117
physiotherapy 86–92 propofol 144
NIV and, in neuromuscular disease 50–51 proportional assist ventilation (PAV) 13
particle spread 151–152 puberty 223
pneumomediastinum 209 pulmonary hypertension, kyphoscoliosis 251
pneumonectomy 137 pulse oximetry
pneumonia 68–69, 149 acute NIV 97
persistent weaning failure 119, 119f ARF 95–97
pneumothorax 53–54, 209, 228–229 long-term NIV 266
polio 2, 5, 5f nocturnal hypoventilation 266, 267f, 268f
polygraphy OHS 198–200
home monitoring 265, 270–272 overnight 266
inappropriate ventilator settings 236 pulse transit time (PTT) 271
leak detection 270f, 271
polysomnography (PSG) 100
bridging to transplant 207
Q
quality of death 257
inappropriate ventilator settings 236 quality of life see health-related quality of life
long-term NIV 265 questionnaires 277–278, 278t
as monitoring tool 270–272
OHS 200
unattended home 271
R
rapid shallow breathing index (RSBI) 107
Pompe disease 166t, 170
rebreathing 15
pores of Kohn 86
rehabilitation, COPD 195
portable ventilators, children 81
re-intubation avoidance 127–134
positive end-expiratory pressure (PEEP)
remifentanil 146
acute hypoxaemic respiratory failure 67
respiratory cycles, patient triggered 269–270
bronchoscopy 144
respiratory failure
positive pressure ventilation (PPV) 10
acute see acute respiratory failure (ARF)
chest wall disorders 185
causes 60–61, 61f
post-extubation respiratory failure
chronic see chronic respiratory failure (CRF)
neuromuscular disease 130
type 1 see hypoxaemic respiratory failure
NIV and oxygen therapy 128–130
type 2 see hypercapnic respiratory failure
NIV to prevent 128–130, 129t, 131t, 132f
respiratory muscles
NIV to treat 130–132, 131t, 132f
chest wall disorders 183
obese patients 130
COPD, chronic/long-term NIV 194
post-operative pulmonary complications
failed weaning 120–121, 121t
(PPCs) 135
weakness, ALS 176, 177t
mechanisms 135
respiratory pandemics 148–154
risk factors 136
aerosol transmission 151–153
thoracic surgery 137
decision-making 152
postural drainage 87
droplet transmission 151–153
pregnancy
infection control 152–153
H1N1 inuenza 150
NIV equipment 152
neuromuscular disorders 172–173
NIV potential role 149
physiological effects 172
respiratory rate 269–270
vital capacity 172
estimation 100
premature cycling 11
NIV monitoring 97
preschoolers, NIV initiation 222
respite care 263, 287, 293
preset pressure ventilation, chest wall
resuscitation boxes 3
disorders 185
Richmond sedation–agitation scale 95, 96t
pressure alarms 17
rise time 12
pressure-controlled ventilation (PCV) 12–13
chest wall disease 117
pressure-preset NIV 19, 20f, 21f, 21t
COPD 117
ventilators 13, 14f, 14t
dyspnoea 117
pressure sores 32
failure to trigger ventilator 117
pressure support ventilation (PSV) 12, 21
risk management 285–287
COPD 238, 239
risk minimisation 262–263
obesity-related chronic respiratory failure
airway access 286
200–202
communication 286, 287
perioperative period 139
medical problems identication 286
spontaneous mode 12
proactivity with new medical problems 287
spontaneous/timed mode 12
strategy 285–287
S supplemental oxygen 35–36
acute exacerbation of COPD 112–113
sarcoglycanopathies 169 children 81
scoliosis 182, 185 entrainment site 36, 36f
secretion(s) 86 indications 35
NIV failure 108 inspiratory/expiratory pressure alteration
tracheostomy ventilation 156 36, 36f
secretion management 286 kyphoscoliosis 251
ALS 179, 240 set-up/practical tips 36
Duchenne muscular dystrophy 250–251 severe lung damage 206
secretion removal 86 support groups 262
self-care 260 symptom palliation 253–259
Serve-HF (Treatment of Predominant Central
Sleep Apnoea by Adaptive Servo Ventilation in
Patients with Heart Failure) study 212–213 T
settings, matching to patient 18–25 tank ventilation 1
severe acute respiratory syndrome (SARS) telemedicine 291
149, 151 telemonitoring 187, 187f, 267, 291
Severe Respiratory Insufficiency (SRI) terminal dyspnoea 254–255
questionnaire 272, 276, 278–279, 279f, thoracic expansion exercises 87
279t, 280 thoracic surgery 137
international adaptation 279, 280t thoracoabdominal assisted cough 228, 229f
validation 279 3CPO (three interventions in cardiogenic
shaking, airway clearance 87, 87f pulmonary oedema) trial 74, 75f, 76f
shallow breathing 183, 204 tidal breathing, chest wall disorders 182–183
shock 70 tidal volume (VT)
side-effects, NIV 272–274, 273t acute hypoxaemic respiratory failure 67
single-limb circuits 15, 16f, 23–25 estimation, home ventilators 267–269
leaks 98 expiratory assessment 98, 99f
sip ventilation see mouthpiece ventilation inspiratory value 98, 99f
sizing gauges 32 monitoring 98, 99f
sleep disordered breathing (SDB) ventilator settings 11
bronchiectasis 205–206 timed ventilation 13
children 223 toddlers, NIV initiation 222
congestive heart failure 211, 212–213 total face mask see full face mask
cystic brosis 205–206, 205t TPPV see invasive positive-pressure ventilation
interstitial lung diseases 205–206, 205t via a tracheostomy (TPPV)
myotonic dystrophy 170 tracheobronchial secretion 37
neuromuscular disease 221 tracheobronchomalacia 205
patterns 197, 198f, 199f tracheostomy 24f, 155–162
SMA type 2 220f after NIV failure 241
sleep hypoxaemia 206 ALS 178
sleep quality, COPD 195 alternatives 159
sleep studies 100 blocked 286
Duchenne muscular dystrophy 249, 249f carer burden 161
OHS 200 changing 160–161
OHS case study 247, 248f choices 156–158, 157t
sneezes, particle spread 151 cleaning 160
speaking valves 158 complications 159, 160t
spinal muscular atrophy (SMA) 164t, congenital central hypoventilation
169–170, 173 syndrome 221
NIV and cough assist techniques 50 cuffed tubes 157
NIV outcomes 169–170 cuff ination pressure 160
type 1, palliative NIV 221 decannulation 158–159
type 2 163, 170, 220f denition 155
S 2 see pulse oximetry
pO
dual cannula 157–158
spontaneous breathing trials (SBTs) 120 Duchenne muscular dystrophy 169
difficult weaning 118 early 155
hypercapnia at end of 119, 120f fenestrated tubes 158
spontaneous respiratory rate 269–270, 270f improved patient care 159–160
sputum expectoration manoeuvres, CF 63 indications 156t
starting, acute NIV 102–110 multidisciplinary care 160
Steinert’s myotonic muscular dystrophy 240 neuromuscular disease 51–53
stepping up/down from NIV to tracheostomy step down to NIV 53
109, 155–162 overnight NIV 159
stopping, acute NIV 102–110 pepperpot type 158
percutaneous technique 156–157
risk minimisation 286 perioperative period 139
routine chronic care 160–161 principles 10–17
speech and 157, 158 risk minimisation 286
swallowing 158 settings
tube dimensions 157 acute cardiogenic pulmonary oedema 73
uncuffed tubes 157 acute hypoxaemic respiratory failure 70
weaning 158–159, 159f children 81–82, 82t
transcutaneous capnography 237–238, 265 mouthpiece ventilation 232
overnight monitoring 266 OHS 247–248
transcutaneous carbon dioxide tension parameters 11f
(PtcCO2) 97 perioperative period 139–140
acute exacerbation of COPD 43 respiratory pandemics 152
COPD 238, 239 spare 286
monitoring 97 terminology 10–12
trans-diaphragmatic pressure monitoring 100 volume 11
transoesophageal echocardiography (TEE) 137 ventilatory dependence 260
transoesophageal ultrasound 146 patient discharge 260–264
transplantation see lung transplantation ventilatory failure, natural history 5
transport needs 262, 285 viral bronchiolitis 80
treatment principles 103f vital capacity
triggering 12 augmented inspiratory 227
troponin, Duchenne muscular dystrophy 51 chest wall disorders 185
type 1 acute respiratory failure see hypoxaemic pregnancy 172
respiratory failure volume-assured pressure support (VAPS)
type 2 respiratory failure see hypercapnic 13–15
respiratory failure volume-preset ventilation 13, 14f, 14t, 19,
20f, 21f, 21t
U children 82
neuromuscular disorders 171
Ullrich muscular dystrophy 165t, 169
unrewarded respiratory efforts 270f, 271–272 nocturnal hypoventilation 239
upper-airway obstruction 80, 205–206 obesity-related chronic respiratory failure
200–202
V volume-targeted ventilators see volume-preset
ventilation
vented masks 28, 29, 30t
vomiting, risk of 29
ventilation modes 19–22
matching to patient 18–25
ventilation/perfusion mismatch 145 W
ventilator(s) 10–17 weaning 118–126
battery management 287 classication 118
built-in software 266–270 COPD patients 122
choices difficult 118, 119f, 120–122
ARF 103 hypercapnic respiratory failure 122–123
children 81 from invasive mechanical ventilation
neuromuscular disorders 171–172 118–120
circuits 15, 16f NIV
compliance assessment 246–247, 267 cardiovascular response 121
data available from 246–247 clinical trials 122–123, 123t
failure to trigger 116–117 efficacy 124, 124t
humidication systems 38 failure 124
hybrid modes 19 limitations 124
inappropriate settings 236, 237f physiological effects 121–122, 122t
NIV failure 108 respiratory mechanics 120–121, 121t
intelligent 7 use algorithm 125f
intensive care unit use rationale 120–122
children 81 OHS 59
humidication 38 outcomes 119, 119f, 120f
intolerance indicators 266–267, 269f prolonged 118, 119, 119f, 120–122
modes 12–15, 19–22 recommendations 120
monitoring 98 simple 118, 119, 119f
mouthpiece ventilation 231 tracheostomy 158–159, 159f
NIV-assisted exercise training 208 wheelchairs 262
oxygen delivery 15–17
patient asynchrony/dyssynchrony see X
patient–ventilator asynchrony/ X-linked myotubular myopathy 170
dyssynchrony pattern of use
266–267, 269f

MCU 2015 ERS Practical Handbook of Noninvasive Ventilation

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MCU 2015 ERS Practical Handbook of Noninvasive Ventilation

Uploaded by

Copyright:

Available Formats

practical

© 2015 European Respiratory Society

Cover image: Juanmonino, iStockphoto

Design by Claire Turner, Lee Dodd, ERS

All material is copyright to the European Respiratory Society.

CONTACT, PERMISSIONS AND SALES REQUESTS:

NIV: past, present and future 1

Chapter 2 – Getting the basics right: equipment

Basic principles of ventilators 10

Matching mode and settings to the patient: an introduction 18

Choosing the interface 26

Supplemental oxygen and humidication 35

Chapter 3 – Getting the basics right: patient selection

The patient with an acute hypercapnic exacerbation of COPD 41

Patients with acute hypercapnic respiratory failure and 49

Patients with acute-on-chronic hypercapnic respiratory failure 60

The patient with acute hypoxaemic respiratory failure excluding 67

The patient with acute hypoxaemic failure and cardiogenic 72

Chapter 4 – Paediatric indications for NIV

Acute NIV in children, including ventilator and interface choice 79

Chapter 5 – Airway clearance and physiotherapy

Airway clearance methods and nebulised therapy in acute NIV 86

Chapter 6 – Monitoring progress in acute NIV

Monitoring choices in acute NIV 93

Starting and stopping acute NIV: when and why? 102

Problem-solving: case studies of NIV problems and their 111

NIV and weaning 118

NIV to avoid re-intubation 127

NIV in the perioperative period 135

NIV for endoscopic procedures 142

NIV in respiratory pandemics 148

Stepping up and down from NIV to tracheostomy ventilation 155

Chapter 8 – Long-term NIV

Chronic NIV in hereditary neuromuscular disorders 163

Chronic NIV in motor neurone disease/ALS 176

Chronic NIV in chest wall disorders 182

Chronic NIV in COPD 190

Chronic NIV in OHS 197

Chronic NIV in bronchiectasis, CF and interstitial lung 204

Chronic NIV in children: indications, outcomes and transition 217

Practicalities of and guide to cough augmentation and 226

Long-term NIV failure: causes and problem solving 234

Long-term NIV case histories 246

Chapter 9 – NIV for symptom palliation

NIV in palliative care and at the end of life 253

Chapter 10 – Discharge planning and community care

Discharging the ventilator-dependent adult and child 260

Home monitoring and follow-up of long-term NIV 265

Assessing quality of life and outcome of long-term NIV 276

Patient and carer education, and risk management 282

Chapter 11 – Setting up an NIV service

Setting up and staffing your NIV unit 289

Chief Editor Jens C. Callegari

Miquel Ferrer Leo M.A. Heunks

Bearing in mind the growing applications for noninvasive ventilation (NIV)

Simply visit http://erspublications.com/titles and add the ERS Practical

Also available from the ERS

ERS Handbook: Self-Assessment in Respiratory Medicine

The new and updated second edition of Self-Assessment in Respiratory

AHI Apnoea–hypopnoea index

NIV: past, present and future

The history of NIV is an intertwining chronicle of the development of negative

ERS Practical Handbook Noninvasive Ventilation 1

NIV in acute hypoxaemic respiratory failure,