British Journal of Anaesthesia 106 (4): 451–3 (2011)
doi:10.1093/bja/aer056
EDITORIAL II
Minimally invasive cardiac output monitoring: what evidence
do we need?
M. Chikhani and I. K. Moppett *
Division of Anaesthesia and Intensive Care, University of Nottingham, Queen’s Medical Centre Campus, Nottingham University Hospitals
NHS Trust, Nottingham NG7 2UH, UK
* E-mail: iain.moppett@nottingham.ac.uk
Minimally invasive cardiac output monitoring is a hot topic.
Recent articles have been published in the British Journal of
Anaesthesia and elsewhere attesting to its safety, question-
ing its validity, and using minimally invasive monitoring in
clinical trials. Studies have demonstrated that less-invasive
cardiac output monitoring combined with a goal-directed
fluid administration protocol is associated with favourable
perioperative outcomes in surgical patients deemed to be
at high risk of morbidity and mortality.1 – 4 Recommendations
from the Enhanced Recovery after Surgery Group have
included specific reference to perioperative fluid optimization
with respect to goal-directed targets in high-risk patients.5 6
Despite these studies, questions remain about the role of
minimally invasive monitoring. What message can anaesthe-
tists take from these studies?
First, is this monitoring technique valid? Central to this ques-
tion is what the monitor is supposed to be measuring. In broad
terms, the monitors all aim to estimate stroke volume, either
as an absolute value or in terms of relative changes. In
addition, some use derived indices to predict whether stroke
volume is likely to increase in response to a fluid challenge.
In clinical practice, however, the clinician is really more inter-
ested in what the appropriate treatment should be for a par-
ticular patient, rather than stroke volume per se. There is no
single best method for validating minimally invasive cardiac
output monitoring and recent articles demonstrate the differ-
ent approaches. Reisner and colleagues7 have compared, in an
elegant study, different approaches to estimation of stroke
volume using graded lower body negative pressure (LBNP) to
simulate haemorrhagic hypovolaemia in healthy volunteers.
Their results are moderately reassuring. Each of the techniques
demonstrated a progressive reduction in stroke volume with
increasingly negative LBNP. On release, the stroke volume
returned to baseline with both bioimpedence and the rela-
tively newly described long-time interval analysis of the arter-
ial pressure waveform. Modelflow, the technique used by the
non-invasive Finometer, failed to return to baseline after
release of LBNP, suggesting that some part of its algorithm is
sensitive to a factor not related to stroke volume. However,
these results need to be taken in context. LBNP is only a
model for haemorrhagic hypovolaemia and the study was per-
formed in healthy volunteers. The clinical scenario is rather
more complicated, with vasoactive drugs, ongoing resuscita-
tion, and acute and chronic comorbidities all contributing.
Paarmann and colleagues8 have taken a more traditional
approach to validation; a new technique, in this case the
‘pressure recording analytical method’ (PRAM) is compared
in relatively stable clinical conditions with the gold standard
(thermodilution measurement of cardiac output). On the
basis of the Bland –Altman analysis of measured stroke
volume, they conclude that PRAM and thermodilution
cardiac output are not interchangeable. However, it may
again not be so simple. What is not reported is the response
to interventions, particularly fluids, which is clinically more
relevant. By and large, anaesthetists and critical care phys-
icians are much more interested in the dynamic response
to intervention than any absolute values. This study8 gives
us more information about the validity of PRAM but does
not answer the question of its clinical utility. Numerous pre-
vious studies have questioned the accuracy of different mini-
mally invasive techniques, but few have addressed whether
the monitors will provide appropriate information to guide
clinical management in the real world. Future studies could
usefully provide data on the sensitivity, specificity, and posi-
tive and negative predictive values of observed changes in
stroke volume. It is important to note that the studies
which report the degree of concordance between changes
in stroke volume do not report the degree of misclassification
between responders and non-responders to fluid challenge.
Ultimately, whatever the technique used, minimally inva-
sive cardiac output monitoring is simply a tool to assist the
anaesthetist in deciding on an appropriate course of action.
As with any equipment, we use there is variability in quality
but what really matters is the person using them, the same
is true for minimally invasive cardiac output monitoring.
The evidence base for use of this technology is less robust
than it first appears. Optimal volume loading is usually now
defined as a lack of a sustained response in stroke volume
451
BJA
to an adequate fluid challenge. This is predicated on the
Frank–Starling physiology of the heart; subjects nearer the
top of the curve will not be able to increase stroke volume
further. However, what is a sustained response and what is
an appropriate fluid challenge? A variety of bolus volumes
have been used in different studies with little evidence to
justify the choice. The increment in stroke volume needs to
be sufficient to be discernible from underlying variation in
the patient’s cardiac output and discernible from measure-
ment variation. It will of course be affected by the gradient
of the preload– stroke volume relationship. Some patients
will be classified as non-fluid responsive because their
response to a fluid challenge is below the observable
threshold. The consequence of these factors would suggest
that however the protocols are written, fluid optimization
remains something of an art. The clinician still needs to inter-
pret the cardiac output monitoring in the light of all the avail-
able data.
In an attempt to make this process more reliable, derived
indices are calculated by the monitors which make an esti-
mate of likely fluid responsiveness on the basis of heart –
lung interactions: pulse pressure variation (PPV), stroke
volume variation (SVV), and systolic pressure variation (SPV)
are the best described. Again the evidence base is perhaps
not as strong as it first seems. If tidal volume and therefore
intrathoracic pressures are not consistent then assessments
of SVV, PPV, and SPV are inaccurate.9 All of these indices
are reliant on change in intrathoracic pressure and therefore
a tidal volume of 8 ml kg21 has been suggested as a
minimum.10 A study using a pig model of acute lung injury
and haemorrhage adds support to this concept, although
care must be taken with extrapolation from animal models
to clinical practice.11 Low (6 ml kg21) tidal volume reduced
the discriminatory power of SVV and SPV, to diagnose hae-
morrhagic hypovolaemia, whereas PPV appeared less depen-
dent upon tidal volume (VT). Unfortunately, for
methodological reasons, the authors did not provide any
data on these indices as predictors of fluid responsiveness.
Patient position may also affect PPV and SVV, as demon-
strated by an increase in the optimal threshold value for
PPV and SVV from the supine to prone positions.12 The discri-
minatory ability of the indices did not change, however,
suggesting that the clinician can still use the indices but
with a different threshold. The obvious practical consequence
of these studies is that the clinician cannot take an isolated
value from a cardiac output monitor as an absolute indi-
cation of optimal intravascular volume. The data have to
be interpreted in the appropriate clinical context.
If the evidence base for the monitoring appears to be
slightly confusing, at least the monitoring appears to be
safe. In this issue of the British Journal of Anaesthesia,
Belda and colleagues13 report prospectively collected data
from more than 500 patients mainly undergoing femoral
arterial cannulation for transpulmonary thermodilution
measurement of cardiac output. Almost all complications
were transient with a catheter-related infection rate of
0.78%, comparable with routine intensive care monitoring.
452
Editorial II
However, one patient did develop a femoral thrombosis
requiring surgical embolectomy, so the catheters cannot be
viewed as risk free. In comparison, the complication rate of
following pulmonary artery catheter (PAC) insertion is
between 4% and 10%14 15 with PAC-related bacteraemia
occurring in 0.7–1.8%,16 and arrhythmia requiring treatment
in 3%.15 Complications after oesophageal Doppler placement
also seem to be low though not unknown. Inadvertent endo-
bronchial insertions causing incompetence of the cuffed tra-
cheal tube and aspiration pneumonitis has been reported.17
Despite these caveats, the evidence base for minimally
invasive cardiac output measurement improving patient
outcome as part of optimization protocols is reasonably
strong particularly for colorectal surgery.2 3 There are
several ongoing studies comparing minimally invasive
cardiac output monitoring-based protocols with standard
care which when reported may provide clearer evidence of
which patients may benefit from this approach.
Evidence-based manipulation of the cardiovascular
system can appear to be arbitrary in nature. Accurate admin-
istration of fluids according to a protocol may be difficult to
master and can be seen as subjective. The positive outcomes
associated with these goal-directed protocols may well stem
from the personalization of fluid administration whether or
not the endpoints nor their means to achieve them are
based on any particular science. It may be that the
thought process involved in the practical administration of
an optimal fluid regime is enough to make a difference.
One US study estimates that just less than one-quarter of
surgical procedures met criteria for minimally invasive moni-
toring.10 At present, there are few data to demonstrate
whether benefits seen in trials translate into patient or
healthcare benefits if used in this number of patients in
routine clinical practice.
Research will continue on the use of minimally invasive
cardiac output monitoring. In order to bring more clarity to
the field, we would suggest that there should be some frame-
work for future studies. Validation studies should ideally be
performed in relevant populations with a reasonable dispersal
of volume status. These studies should report not only absolute
relationships between measurement techniques but also the
degree of correct identification of response to interventions
and the ability to track changes in clinical conditions. The defi-
nition of responders and non-responders to fluid challenge
should be based on endpoints which have been used success-
fully in clinical studies. The choice of 10% increase in SV at 5
min after fluid challenge may have been arbitrary, but the
empirical evidence is that it seems to be associated with
benefit.1 – 4 Studies are required urgently to assess the practical
ability of clinicians to deliver the protocols used in research
studies. As with other practical skills, we need evidence of
how to teach its safe use, and also evidence of the inevitable
complications of its use in routine practice.
Meanwhile, minimally invasive cardiac output represents
another numerical trend monitor for anaesthetists and
intensivists to use, alongside medical history, clinical examin-
ation, situational awareness, and other monitoring
Editorial II
modalities in the delivery of personalized fluid regimes to
improve patient outcome.
Conflict of interest
I.K.M. has received honoraria from Schering-Plough in the
past 5 yr. He is an investigator in the OPTIMISE study which
receives equipment support from LiDCO Group Plc. He is the
chief investigator for a study using LiDCO in hip fractures
(no commercial support). He received consumables support
from Deltex Medical (CardioQ) for a study in 2004. He is a
member of the editorial board of the British Journal of Anaes-
thesia. M.C. is an investigator in the OPTIMISE study which
receives equipment support from LiDCO Group Plc.
References
1 Price JD, Sear JJW, Venn RRM. Perioperative fluid volume optimiz-
ation following proximal femoral fracture. Cochrane Database Syst
Rev 2004: CD003004. doi:10.1002/14651858.CD003004.pub2
2 Giglio MT, Marucci M, Testini M, Brienza N. Goal-directed haemo-
dynamic therapy and gastrointestinal complications in major
surgery: a meta-analysis of randomized controlled trials. Br J
Anaesth 2009; 105: 637 –46
3 Abbas SM, Hill AG. Systematic review of the literature for the use
of oesophageal Doppler monitor for fluid replacement in major
abdominal surgery. Anaesthesia 2007; 63: 44–51
4 Pearse R, Dawson D, Fawcett J, Rhodes A, Grounds RM, Bennett ED.
Early goal-directed therapy after major surgery reduces compli-
cations and duration of hospital stay. A randomised, controlled
trial [ISRCTN38797445]. Crit Care 2005; 9: R687–93
5 Lassen K, Soop M, Nygren J, et al. Consensus review of optimal
perioperative care in colorectal surgery: Enhanced Recovery
After Surgery (ERAS) Group recommendations. Arch Surg 2009;
144: 961– 9
6 Varadhan KK, Neal KR, Dejong CH, Fearon KC, Ljungqvist O,
Lobo DN. The enhanced recovery after surgery (ERAS) pathway
for patients undergoing major elective open colorectal surgery:
a meta-analysis of randomized controlled trials. Clin Nutr 2010;
29: 434– 40
BJA
7 Reisner AT, Xu D, Ryan KL, Convertino VA, Rickards CA,
Mukkamala R. Monitoring non-invasive cardiac output and
stroke volume during experimental human hypovolaemia and
resuscitation. Br J Anaesth 2011; 106: 23–30
8 Paarmann H, Groesdonk HV, Sedemund-Adib B, et al. Lack of
agreement between pulmonary arterial thermodilution
cardiac output and the pressure recording analytical method in
postoperative cardiac surgery patients. Br J Anaesth 2011; 106:
475– 81
9 Pinsky MR. Probing the limits of arterial pulse contour analysis
to predict preload responsiveness. Anesth Analg 2003; 96:
1245– 7
10 Maguire S, Rinehart J, Vakharia S, Cannesson M. Respiratory vari-
ation in pulse pressure and plethysmographic waveforms: intra-
operative applicability in a North American Academic Center.
Anesth Analg 2011; 112: 94 –6
11 da Silva Ramos FJ, de Oliveira EM, Park M, Schettino GPP,
Azevedo LCP. Heart– lung interactions with different ventilatory
settings during acute lung injury and hypovolaemia: an exper-
imental study. Br J Anaesth 2011; 106: 394–402
12 Biais M, Bernard O, Ha JC, Degryse C, Sztark F. Abilities of pulse
pressure variations and stroke volume variations to predict fluid
responsiveness in prone position during scoliosis surgery. Br J
Anaesth 2010; 104: 407–13
13 Belda FJ, Aguilar G, Teboul JL, et al. Complications related to
less-invasive haemodynamic monitoring. Br J Anaesth 2011;
106: 482–6
14 Binanay C, Califf RM, Hasselblad V, et al. Evaluation study of
congestive heart failure and pulmonary artery catheterization
effectiveness: the ESCAPE trial. J Am Med Assoc 2005; 294:
1625– 33
15 Harvey S, Harrison DA, Singer M, et al. Assessment of the clinical
effectiveness of pulmonary artery catheters in management of
patients in intensive care (PAC-Man): a randomised controlled
trial. Lancet 2005; 366: 472– 7
16 Hadian M, Pinsky MR. Evidence-based review of the use of the
pulmonary artery catheter: impact data and complications. Crit
Care 2006; 10 (Suppl. 3): S8
17 Chandan GS, Hull JM. Incorrectly placed oesophageal Doppler
probe. Anaesthesia 2004; 59: 723
453