Professional Documents
Culture Documents
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
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
Index 295
Contributors
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
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 condence
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.
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.
Anita K. Simonds
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.
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
ERS Practical Handbook Noninvasive Ventilation 3
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
Figure 3. Timeline of developments in NIV from the 1980s to the present day. MD: muscular
dystrophy.
4 ERS Practical Handbook Noninvasive Ventilation
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.
ERS Practical Handbook Noninvasive Ventilation 5
Introduction
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”.
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!
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.
• 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
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.
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
ERS Practical Handbook Noninvasive Ventilation 11
The basics: equipment
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
12 ERS Practical Handbook Noninvasive Ventilation
The basics: equipment
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
ERS Practical Handbook Noninvasive Ventilation 13
The basics: equipment
a) b)
Pressure Pressure
Flow Flow
Volume Volume
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.
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)
Noninvasive interface
(non-vented mask)
Single limb
c)
Noninvasive interface
(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.
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
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,
1000 1000
800 800
The basics: equipment
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.
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.
ERS Practical Handbook Noninvasive Ventilation 21
The basics: equipment
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
ERS Practical Handbook Noninvasive Ventilation 23
The basics: equipment
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.
24 ERS Practical Handbook Noninvasive Ventilation
The basics: equipment
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.
• 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 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.
• Simonds AK, et al. (1995). Outcome of domiciliary nasal intermittent positive pres-
sure ventilation in restrictive and obstructive disorders. Thorax; 50: 604–609.
• 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.
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.
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.
The basics: equipment
27
The basics: equipment
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.
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.
ERS Practical Handbook Noninvasive Ventilation 29
The basics: equipment
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
30 ERS Practical Handbook Noninvasive Ventilation
Check compatibility of interface, ventilator and circuit. Identify where exhalation occurs. Choose correct headgear.
ARF CRF
Check for non-intentional leaks, asynchrony, Check for non-intentional leaks, asynchrony,
ventilator settings, circuit; adjust if necessary ventilator settings, circuit; 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.
The basics: equipment
31
The basics: equipment
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).
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
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.
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.
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).
36 ERS Practical Handbook Noninvasive Ventilation
The basics: equipment
Table 1. Advantages and disadvantages of HME filter and heated humidifier (HH) s ystems
during NIV
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
38 ERS Practical Handbook Noninvasive Ventilation
The basics: equipment
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.
ERS Practical Handbook Noninvasive Ventilation 39
The basics: equipment
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.
Mark W. Elliott
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 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
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.
44 ERS Practical Handbook Noninvasive Ventilation
The basics: patient selection
IMV success
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 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
invasive ventilation.
• 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)
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.
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.
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.
54 ERS Practical Handbook Noninvasive Ventilation
The basics: patient selection
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.
• 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.
• 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
Noninvasive ventilation: basic concepts, Hanover 2014. www.ers-education.org/
events/courses/noninvasive-ventilation-basic-concepts,-hanover-2014.aspx
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
ERS Practical Handbook Noninvasive Ventilation 57
58
Assess for underlying cause(s) of acute
Assess suitability for NIV
decompensation# and treat appropriately
NIV contraindicated
Suitable for trial of NIV
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.
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.
Respiratory failure
Hypoventilation
Respiratory muscle
weakness
Hypoxaemia Hypercapnia
1.0
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
62 ERS Practical Handbook Noninvasive Ventilation
The basics: patient selection
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.
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.
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
ERS Practical Handbook Noninvasive Ventilation 69
The basics: patient selection
observational studies) and reduction of mortality (data from four RCTs and one
prospective observational study).
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.
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
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
invasive ventilation.
• One death can be avoided for every 14 ACPO patients treated
with noninvasive positive pressure ventilation.
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
ERS Practical Handbook Noninvasive Ventilation 73
The basics: patient selection
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.
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,
74 ERS Practical Handbook Noninvasive Ventilation
The basics: patient selection
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).
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).
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.
ERS Practical Handbook Noninvasive Ventilation 77
The basics: patient selection
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
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.
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
transplantation. NIV has also been shown to facilitate extubation after cardiac
surgery in children with ARF due to heart disease.
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
82 ERS Practical Handbook Noninvasive Ventilation
Paediatric indications
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.
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.
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.
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.
a) b) c)
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.
ERS Practical Handbook Noninvasive Ventilation 87
Airway clearance and physiotherapy
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.
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%)
ERS Practical Handbook Noninvasive Ventilation 89
Airway clearance and physiotherapy
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.
Airway clearance
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
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,
with permission from the publisher.
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
ERS Practical Handbook Noninvasive Ventilation 95
Acute NIV monitoring
Table 2. Kelly–Matthay score for assessment of hypercapnic encephalopathy during NIV
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
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:
a) Unintentional
leaks
Ventilator
delivered volume
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
permission from the publisher.
100 ERS Practical Handbook Noninvasive Ventilation
Acute NIV monitoring
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.
• 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.
• 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.
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/
courses/noninvasive-ventilation-basic-concepts,-hanover-2014.aspx
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.
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.
O2 CO2 O2 CO2
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.
ERS Practical Handbook Noninvasive Ventilation 103
Acute NIV monitoring
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.
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
104 ERS Practical Handbook Noninvasive Ventilation
Acute NIV monitoring
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”.
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).
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.
ERS Practical Handbook Noninvasive Ventilation 107
Acute NIV monitoring
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
108 ERS Practical Handbook Noninvasive Ventilation
Acute NIV monitoring
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.
Alanna Hare
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.
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
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.
Causes of leaks
• Mouth-breathing with a nasal interface
• Incorrectly sized interface
• Overtightened straps or pressure sores
• Unusual facial anatomy
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.
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
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
Miquel Ferrer
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.
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
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.
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
Table 1. Pathophysiologic basis of weaning failure during transition from positive pressure
ventilation to spontaneous breathing
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
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
122 ERS Practical Handbook Noninvasive Ventilation
NIV and the ICU
Table 3. Summary estimates of the effect of NIV for weaning critically ill adults off invasive
ventilation
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
No Yes
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.
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
respiratory failure.
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
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
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.
132 ERS Practical Handbook Noninvasive Ventilation
NIV and the ICU
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.
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.
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,
ERS Practical Handbook Noninvasive Ventilation 137
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.
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.
Leo M.A. Heunks, Lisanne Roesthuis and Erik H.F.M. van der Heijden
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”).
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.
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.
ERS Practical Handbook Noninvasive Ventilation 143
NIV and the ICU
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).
144 ERS Practical Handbook Noninvasive Ventilation
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.
a) b)
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.
Anita K. Simonds
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.
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.
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
152 ERS Practical Handbook Noninvasive Ventilation
NIV and the ICU
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.
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.
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
ERS Practical Handbook Noninvasive Ventilation 157
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.
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
158 ERS Practical Handbook Noninvasive Ventilation
NIV and the ICU
Ventilation Aim Time off Time on RR SaO 2 Comfort start RR SaO 2 Comfort end
Physio Variance
Figure 1. Tracheostomy weaning chart. RR: respiratory rate; vent: ventilator; trachy:
tracheostomy.
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.
ERS Practical Handbook Noninvasive Ventilation 159
NIV and the ICU
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.
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.
• 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.
Anita K. Simonds
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.
(Continued)
165
166
Table 1. Continued
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.
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,
100
80
60
Survival %
40
20
0
15 20 25 30 35 40 45
Age at death years
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
168 ERS Practical Handbook Noninvasive Ventilation
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
ERS Practical Handbook Noninvasive Ventilation 169
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.
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
interface pressure on the face, designing masks to reduce pressure points and
disperse the pressure over wider areas, and rotating interfaces.
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.
• 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.
• Chatwin M, et al. (2013). Risk management protocol for gastrostomy and jeju-
nostomy insertion in ventilator dependent infants. Neuromuscul Disord; 23:
289–297.
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
Noninvasive ventilation: basic concepts, Hanover 2014. www.ers-education.org/
events/courses/noninvasive-ventilation-basic-concepts,-hanover-2014.aspx
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.
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
ERS Practical Handbook Noninvasive Ventilation 177
Long-term NIV
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.
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.
178 ERS Practical Handbook Noninvasive Ventilation
Long-term NIV
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”.
ERS Practical Handbook Noninvasive Ventilation 179
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.
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
respiratory failure.
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.
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
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.
ERS Practical Handbook Noninvasive Ventilation 183
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
184 ERS Practical Handbook Noninvasive Ventilation
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
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
ERS Practical Handbook Noninvasive Ventilation 185
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.
186 ERS Practical Handbook Noninvasive Ventilation
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.
ERS Practical Handbook Noninvasive Ventilation 187
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.
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.
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
Figure 1. Indication criteria for long-term NIV in COPD patients during stable disease.
192 ERS Practical Handbook Noninvasive Ventilation
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.
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.
• 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.
• 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.
• 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
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.
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)
100
SpO2 % 90
80
70
120
beats·min-1
Heart rate
100
80
60
10
PtcCO2 kPa
4
23:00 h
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).
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
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
EPAP EPAP
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
• 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.
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.
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.
ERS Practical Handbook Noninvasive Ventilation 205
Long-term NIV
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.
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.
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%.
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
212 ERS Practical Handbook Noninvasive Ventilation
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.
ERS Practical Handbook Noninvasive Ventilation 213
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.
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.
214 ERS Practical Handbook Noninvasive Ventilation
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.
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.
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
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.
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.
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.
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.
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.
• 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.
• 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.
• Hull J, et al. (2012). British Thoracic Society guideline for respiratory management
of children with neuromuscular weakness. Thorax; 67: Suppl. 1, i1–i40.
• 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.
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.
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.
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
228 ERS Practical Handbook Noninvasive Ventilation
Long-term NIV
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
230 ERS Practical Handbook Noninvasive Ventilation
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.
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.
• 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. (2014). Physiotherapy for Respiratory and Cardiac Problems: Adults
and Children. 5th Edn. London, Elsevier.
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.
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
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.
Table 1. Discomfort related to inappropriate ventilator settings as expressed by the patient,
and possible remedies
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.
238 ERS Practical Handbook Noninvasive Ventilation
Long-term NIV
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.
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,
ERS Practical Handbook Noninvasive Ventilation 239
Long-term NIV
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
240 ERS Practical Handbook Noninvasive Ventilation
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.
Lancet Neurol; 5: 140–147.
• Contal O, et al. (2012). Monitoring of noninvasive ventilation by built-in software
of home bilevel ventilators: a bench study. Chest; 141: 469–476.
• 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.
• 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.
• 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.
• 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/
courses/noninvasive-ventilation-basic-concepts,-hanover-2014.aspx
Alanna Hare
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.
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.
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.
248 ERS Practical Handbook Noninvasive Ventilation
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.
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.
ERS Practical Handbook Noninvasive Ventilation 249
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.
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.
Further reading
• Bushby K, et al. (2010). Diagnosis and management of Duchenne muscular dys-
trophy, part 1: diagnosis, and pharmacological and psychosocial management.
Lancet Neurol; 9: 77–93.
• Bushby K, et al. (2010). Diagnosis and management of Duchenne muscular dystro-
phy, part 2: implementation of multidisciplinary care. Lancet Neurol; 9: 177–189.
• 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.
• 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.
• 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.
“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.
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
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.
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
Table 3. When palliative NIV should be started and discontinued in the acute setting
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.
Joan Escarrabill
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.
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.
ERS Practical Handbook Noninvasive Ventilation 261
Discharge and community care
• 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.
262 ERS Practical Handbook Noninvasive Ventilation
Discharge and community care
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
ERS Practical Handbook Noninvasive Ventilation 263
Discharge and community care
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/
courses/noninvasive-ventilation-basic-concepts,-hanover-2014.aspx
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.
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
ERS Practical Handbook Noninvasive Ventilation 267
Discharge and community care
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).
268 ERS Practical Handbook Noninvasive Ventilation
Discharge and community care
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
ERS Practical Handbook Noninvasive Ventilation 269
Discharge and community care
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
270 ERS Practical Handbook Noninvasive Ventilation
Discharge and community care
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).
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. (2012). Monitoring of noninvasive ventilation by built-in software
of home bilevel ventilators: a bench study. Chest; 141: 469–476.
• 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.
• 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.
• 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.
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.
Table 1. Questionnaires for measurement of HRQoL most frequently used in patients receiving
long-term NIV
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,
278 ERS Practical Handbook Noninvasive Ventilation
Discharge and community care
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
Well-being
(SRI-WB)
SRI
Physical Functioning summary scale
(SRI-PF) (SRI-SS)
Anxieties
(SRI-AX)
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.
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)
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
284 ERS Practical Handbook Noninvasive Ventilation
Discharge and community care
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).
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.
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.
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
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.
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.
Specialist
Family
doctor
Figure 1. The team responsible for long-term ventilator-dependent patients in the community.
ERS Practical Handbook Noninvasive Ventilation 293
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.