Feature ArticleContinuing Medical Education

Evaluation of a noninvasive cardiac output monitor in

mechanically ventilated children*
Astrid Botte, MD; Francis Leclerc, MD, PhD; Yvon Riou, MD, PhD; Ahmed Sadik, MD; Vronique Neve, MD;
Tameur Rakza, MD; Adelaide Richard, MD
LEARNING OBJECTIVES
On completion of this article, the reader should be able to:
1. Describe the strengths and weaknesses of noninvasive cardiac output assessment.
2. Recall the available modalities for the assessment of noninvasive cardiac output in the pediatric intensive care unit.
3. Summarize the differences between the assessment of noninvasive cardiac output by the CO2 partial rebreathing technique
versus thermodilution.
All authors have disclosed that they have no financial relationships with or interests in any commercial companies pertaining
to this educational activity.
Lippincott CME Institute, Inc., has identified and resolved all faculty conflicts of interest regarding this educational activity.
Visit the Pediatric Critical Care Medicine Web site (www.pccmjournal.org) for information on obtaining continuing medical
education credit.

Objective: To compare measurements of cardiac output (CO)

and cardiac index (CI) obtained by a recently developed noninvasive continuous cardiac output system, NICO (CONICO), and transthoracic Doppler echocardiography (COTTE) in mechanically ventilated children.
Design and Setting: Prospective study in a university-affiliated
tertiary pediatric intensive care unit.
Patients: A total of 21 mechanically ventilated children, weighing >15 kg, in stable respiratory and hemodynamic condition.
Measurements: Sets of three successive measurements of CO
with the NICO system and transthoracic Doppler echocardiography were obtained. BlandAltman analysis was used to compare
the agreement between the two methods.
Results: The mean SD CO values were 4.06 1.43 L/min for
CONICO and 4.67 1.78 L/min for COTTE. Bias SD between the
two methods was 0.61 0.94 L/min. The variability of the
difference between the two methods increased as the magnitude
of the CO measurement increased. Similar results were obtained
for cardiac index: 4.01 1.40 Lmin1m2 for CINICO and 4.59

1.48 Lmin1m2 for CITTE. BlandAltman analysis revealed a

nonuniform relationship between CI difference and the magnitude
(y 0.299 0.0655 mean). The variability of the differences
did not increase as the magnitude of the CO measurement increased (SD of estimate was 0.827 Lmin1m2). With both CONICO
and CINICO, each measurement was highly repeatable, with coefficient of variation of only 2.88% 2.31%. Repeatability with
Doppler echocardiography was 7.02% 4.33%.
Conclusions: The NICO system is a new device that measures
CO easily and automatically in mechanically ventilated children
weighing >15 kg. CO values obtained with this technique were in
agreement with those obtained with Doppler echocardiography in
children in respiratory and hemodynamic stable condition. The
NICO system needs further investigation in children in unstable
respiratory and hemodynamic condition. (Pediatr Crit Care Med
2006; 7:231236)
KEY WORDS: cardiac output; CO2 partial rebreathing; Doppler
echocardiography; child

ecause the heart is one of the

most common organs to fail
during critical illness, care of
critically ill patients with cardiocirculatory failure would be improved

by the availability of continuous accurate

measurement of cardiac output (CO) (1
4). However, CO measurements are not
frequently used in children. The reasons
are due to technical and size constraints

*See also p. 284.

Resident (AB), Professor of Pediatrics and Director (FL), Physician (AS, TR), Pediatric Intensive
Care Unit, Physician (YR), Physician [Part Time] (VN),
Department of Respiratory Physiology, Resident
(AR), Department of Pediatric Cardiology, University

Hospital of Lille, Lille Cedex, France.

Copyright 2006 by the Society of Critical Care
Medicine and the World Federation of Pediatric Intensive and Critical Care Societies

Pediatr Crit Care Med 2006 Vol. 7, No. 3

DOI: 10.1097/01.PCC.0000216691.69976.E4

and to questionable usefulness and reliability of the various techniques for obtaining the measurement. Furthermore,
industrial interest is generally directed
toward adult critical care medicine (5).
Recently, a device using the CO2 partial
rebreathing technique for CO measurement has been commercially developed
(NICO, Respironics, Wallingford, CT).
This is an automated, noninvasive
method that uses the indirect Fick principle (6). This NICO monitor has been
231

tested in animals (6, 7) and in adults

(8 14). Although the accuracy and clinical applicability of the NICO has not been
fully evaluated, it seems that the performance of the NICO is comparable with
continuous thermodilution in adult patients (10 12, 14) and that this technique
may be used to monitor central hemodynamics in critically ill patients. In children, only one study has evaluated the
accuracy of noninvasive CO measurement using the NICO compared with
measurement obtained via thermodilution (15), and the study concluded that
noninvasive measurement of CO using
partial rebreathing technology may provide clinically acceptable assessment of
CO in hemodynamically stable children.
However, in this study, the majority of
the 37 children had relatively healthy
lungs and underwent cardiac catheterization to evaluate their hemodynamic function. So, the aim of this study was to
assess, in mechanically ventilated critically ill children in stable respiratory and
hemodynamic conditions (i.e., during the
recovery phase of their acute disease), the
accuracy of CO measurement obtained by
the NICO (CONICO) compared with measurement of CO using transthoracic
Doppler echocardiography (COTTE). The
Doppler echocardiography was preferentially chosen because the technique is
noninvasive and harmless to the children.

METHODS
This study was approved by the local hospital institutional review board, and parents
provided informed consent before measurements. All children admitted to our university-affiliated pediatric intensive care unit from
January 2005 to July 2005 and mechanically
ventilated were eligible for the study if they
weighed 15 kg and were in stable respiratory
and hemodynamic condition. Patients with intracardiac shunt or receiving vasoactive drugs
were excluded. The 15-kg minimum weight

requirement was chosen because the NICO

monitor was approved by the U.S. Food and
Drug Administration at a minimum tidal volume of 200 mL. A total of 21 children were
included over this period of 6 months during
which 136 children were mechanically ventilated. Measurements of CO using both technologies (NICO and TTE) were performed by a
different investigator who was blinded to the
results of the other technique.
COTTE measurement was performed by a
cardiologist or a trained intensivist using a
Doppler echography system (GE Vivid 7 Pro,
GE Medical Systems, Milwaukee, WI) and a
Pedoff transducer (2-MHz carrier frequency).
During the baseline period, ventilator settings (peak inspiratory pressure, mean airway
pressure, positive end-expiratory pressure, inspiratory and expiratory tidal volume, respiratory rate, and FIO2) and respiratory variables
calculated by the NICO system (peak inspiratory pressure, mean airway pressure, positive
end-expiratory pressure, inspiratory tidal volume, expiratory tidal volume, respiratory rate,
and dynamic compliance and resistances of
the respiratory system) were collected. After a
period of 15 mins allowing patient stabilization, an arterial blood gas analysis was performed and PaO2, PaCO2, FIO2, and hemoglobin
concentrations and patient weight were fed
into the NICO system (see APPENDIX for
further details of NICO methodology (16, 17)).
During the NICO data collection period, which
lasted 30 mins, CONICO was continuously
monitored, but only the last three values just
before the TTE measurements were collected.
Three sets of COTTE were performed, and for
each set, blood flow velocity and echocardiographic determination of the ascending aortic
internal diameter were measured. Mean values
of CONICO and COTTE were used for data analysis, and cardiac index (CI) was calculated by
dividing CO by body surface area.
Statistical Analysis. Results are expressed
as mean SD or range. Mean values were
compared by Students t-test. The intrasubject
reliability of the two methods was evaluated by
an intraclass coefficient correlation (18) and
coefficient of variation (CV). Correlation between the two methods was assessed by linear
regression analysis. Agreement between

CONICO and COTTE measurements and between CINICO and CITTE measurements was
evaluated according to the BlandAltman
analysis after adjusting for the nonuniform
relationship between the difference and the
mean value and after modeling the variability
in the SD of the differences as a function of the
level of measurement when required (19, 20).

RESULTS
Patients. A total of 21 children, ranging in age from 2.9 to 16.8 yrs (median
age, 7.89 yrs), were enrolled in the study.
Median weight and body surface area for
the group were, respectively, 25 kg
(range, 15 60 kg) and 0.93 m2 (0.64
1.65 m2). Nine girls and 12 boys were
studied. Nine patients had primary lung
disease such as pneumonia (n 2), sepsis (n 3), and acute respiratory distress
syndrome (n 4). Eleven patients had
neurologic disease such as encephalopathy (n 2), status epilepticus (n 1),
meningitis (n 2), neuromuscular disease (n 5; two having aspiration pneumonia), and head trauma (n 1). One
patient had cardiovascular disease (postcorrection of tetralogy of Fallot). Ventilator settings and respiratory and hemodynamic variables remained unchanged
during the study period (Table 1). No
significant changes in oxygen saturation
or heart rate with time were observed
Reliability of Measurements. Both
CONICO and CINICO measurements were
highly repeatable. Intrasubject variability
was small: intraclass coefficient correlation of CONICO was 0.99 and mean CV of
CONICO was 2.88% 2.31%. This repeatability was lower with Doppler echocardiography: intraclass coefficient correlation of COTTE measurement was 0.97 and
CV of COTTE was 7.02% 4.33%. Surprisingly, this variability in COTTE was
mainly due to variability in the mean
aortic blood flow velocity measurements
(CV, 6.46% 4.65%) and not to variabil-

Table 1. Ventilator settings and respiratory and hemodynamic parameters

Ventilator Settings
PIP, cm H2O
PEEPt, cm H2O
VT, mL/kg
FIO2, %
Air leaks, %

23.2 7.3
4.0 2.5
11.8 3.9
27.4 7.8
3.8 7.2

Respiratory Parameters
PetCO2, mm Hg
VCO2, mLmin1kg1
Re dyn, cm H2OL1sec1
Crs dyn, mLcm H2O1kg1

Hemodynamic Parameters
33.9 7.1
4.0 1.7
27.3 13.8
0.75 0.61

HR, beats/min
SBP, mm Hg
Body temperature, C

118 2
102 17
37.4 0.9

PIP, peak inspiratory pressure; PEEPt, ventilator positive end-expiratory pressure; VT, expiratory tidal volume; PetCO2, expiratory end-tidal CO2 partial
pressure; VCO2, expiratory CO2 production; Re dyn, expiratory dynamic airway resistances; Crs dyn, dynamic compliance; HR, heart rate; SBP, systolic blood
pressure. Values are expressed as mean SD.

232

Pediatr Crit Care Med 2006 Vol. 7, No. 3

ity in the aortic cross sectional area (CV,

1.86% 1.28%).
No difference in end-tidal CO2 (PetCO2)
values was observed before and after the
partial rebreathing period (33.92 7.13
vs. 33.95 7.39 mm Hg). Alveolararterial CO2 content difference ranged
between 4.70 and 15.2 mm Hg (mean,
4.01 8.46 mm Hg).
Doppler Echocardiography vs. CO2
Partial Rebreathing CO. The mean SD
CO values were 4.67 1.78 L/min
(range, 1.829.96 L/min) for COTTE and
4.06 1.43 L/min for CONICO (range,
1.76 7.40 L/min). The coefficient of correlation between the two methods was
0.85 (COTTE 1.07 CONICO 0.33; p
.01). However, the BlandAltman analysis shows a difference between the two
methods, with a bias SD of 0.61
0.94 L/min (Fig. 1). CO difference did not
increase with the mean CO. However, the
variability of the difference increased as
the magnitude of the CO measurement
increased. The SD of the residuals was
modeled as a function of the magnitude
of CO to obtain the limits of agreements
(19). For a mean CO of 1.78 L/min, limits
of agreement (95% confidence interval)
were 1.40 and 0.17 L/min. For a mean
CO of 8.68 L/min, limits of agreement
were 4.51 and 3.28 L/min.

Similar results were obtained for CI:

the mean ( SD) CI values were 4.59
1.48 Lmin1m2 (range, 2.86 8.96
Lmin1m2) for CITTE and 4.01 1.40
L/min for CI NICO (range, 2.23 8.51
L/min). The coefficient of correlation between the two methods was 0.85 (CITTE
0.91 CINICO 0.96; p .01). Bland
Altman analysis reveals a nonuniform relationship between CI difference and the
magnitude (Fig. 2). The difference between the CINICO and CITTE was regressed
on their average (x). The regression equation was y 0.0655 0.299. The
variability of the difference did not increase as the magnitude of the CO measurement increased (SD of estimate was
0.827 L/min).

DISCUSSION
In this population of 21 mechanically
ventilated children in stable respiratory
and hemodynamic condition, the main
findings were the following: 1) there was
a good agreement between CONICO and
COTTE or CINICO and CITTE measurements, according to the criteria defined
by Critchley and Critchley (20), and 2)
this noninvasive measurement of CO using the NICO monitor was easy to use by
the intensivist.

Figure 1. BlandAltman representation of the agreement between the two techniques. The solid line
represents the mean difference between the measurements of cardiac output (CO) by the noninvasive
continuous cardiac output system (CONICO) and transthoracic Doppler echocardiography (COTTE)
(bias), and the dotted lines define the limits of agreement (95% confidence interval). CO difference
between the two methods did not increase with the mean CO (mean difference 0.61 0.94
L/min). However, the variability of the difference increased as the magnitude of the CO measurement
increased. The SD of the residuals was modeled as a function of the magnitude of CO to obtain the
limits of agreements.

Pediatr Crit Care Med 2006 Vol. 7, No. 3

A method-comparison study requires

certainty of the nature and accuracy of
the reference method to judge whether
differences are caused by errors induced
by the new technique under investigation
or to inaccuracies of the reference
method defined as the gold standard (21).
There are known limitations to each
technique of measuring CO. As a result,
although thermodilution is still considered by some authors as the reference
method, others agree that there is no
gold standard for measuring CO (22, 23)
except the direct-oxygen Fick method (6).
In most studies, thermodilution was chosen as the reference method, although its
limitations are well known (24, 25). In
our study, Doppler echocardiography was
preferred as the reference method because it is noninvasive and harmless to
children. Alverson et al. (26), comparing
direct-oxygen Fick method and pulsedDoppler CO measurements during cardiac catheterization in 33 neonates and
children, found excellent agreement between the two methods. Importantly, the
causes of errors in CO measurements by
using Doppler echocardiography have to
be kept in mind; they include optimal
measurement of the area under the
Doppler flow velocity signal, the angle of
insonation, and the correct measurement
of the aortic cross-sectional area (5).
Chew and Poelaert (5) concluded, based
on a 20-yr literature review, that Doppler
CO measurements are acceptably reproducible in children, although the Doppler
CO technique has been less well investigated in children than in adults. From 12
articles in which 344 children were studied, the interobserver and intraobserver
repeatability of Doppler CO measurements, defined as the CV, were respectively 3.121.7% and 2.522.0%, with a
bias of 10% in the majority of studies
(5). These results were similar to those
reported in the meta-analysis including
25 comparative studies of CO measurement techniques (animals, adults and
children): the CV was 12.5112.5%
(mean, 32.5%), which is larger than in
our study (20).
Compared with the direct-oxygen Fick
method, thermodilution had a bias of 2.3
1.6 L/min, whereas CO2 partial rebreathing technique had a bias of 0.2
1.1 L/min in healthy adults (25) and
0.17 3.13 L/min in critically ill adult
patients (27). Several studies comparing
the partial rebreathing technique with
the thermodilution technique have suggested that the NICO monitor yields er233

Figure 2. BlandAltman representation of the agreement between the two techniques. The solid line
represents the difference between cardiac index (CI) as measured by the noninvasive continuous
cardiac output system (CINICO) and transthoracic Doppler echocardiography (CITTE) (bias), and the
dotted lines define the limits of agreement (95% confidence interval). BlandAltman analysis reveals
a nonuniform relationship between CI difference and the magnitude. The difference between the
CINICO and CITTE was regressed on their mean. The regression equation was y 0.299 0.0655
mean. The variability of the differences did not increase as the magnitude of the CI measurement
increased (SD of estimate was 0.827 Lmin1m2).

rors that are acceptable for clinical application in adults (6, 10, 12, 14, 28 34). In
children, only one clinical study, including patients with relatively healthy lungs
and near-normal cardiovascular function,
has confirmed the accuracy of the CO
determination using the partial rebreathing technique, as compared with thermodilution (15).
If we accept the idea that there is no
reference method, each method thus has
an inherent variability for which Critchley and Critchley (20) further expanded
on the method of Bland and Altman by
quantifying the acceptable limits of
agreement, given a known error in the
method used as the reference. As an
example, these authors recommended
that limits of agreement should not exceed 1 L/min (or 20% of the mean CO
value) and reported that acceptable range
for these variables, bias, and limits of
agreement should be, respectively, 1
L/min, 1 L/min, and 20% (20). In our
study, BlandAltman analysis demonstrated a bias and limits of agreement
comparable with those of previous studies comparing the NICO with thermodilution or Doppler technique either in
adults (6, 10 12, 14, 28 34) or children
(15) (Table 2).
In most studies, like in ours, the partial CO2 rebreathing technique slightly
234

underestimated CO as compared with the

reference technique (8, 10, 11, 1315, 28,
31, 33, 35, 36) (Table 2). However, five

studies showed a slight tendency for the

NICO technique to overestimate CO obtained with the reference method, from
10 to 200 mL/min (12, 29, 30, 32, 34).
The NICO monitor calculates only the
nonshunted pulmonary capillary blood
flow (PCBF) and adds a correction factor
for the intrapulmonary shunt flow for
estimating CO; this estimation of the intrapulmonary shunt may lead to inaccuracies and underestimation of the total
CO. In contrast, Doppler echocardiography CO measurement is not affected by
the intrapulmonary shunt (11). As a result, the bias magnitude may be influenced by the population studied.
Our study has two technical limitations: our patients weighed 15 kg and
were in stable respiratory and hemodynamic condition. The weight limitation is
due to important instrumental dead
space of the rebreathing valve (32 mL)
and to the smallest VT (200 mL) imposed
by the NICO system. Thus, using the
NICO system in children weighing 15
kg is impossible, unless VT is increased,
like in the study of Levy et al. (15). In fact,
in this study, tidal volume was large (16.5
5.4 mL/kg), whereas three studies in
adults have demonstrated that tidal volume magnitude influenced CO values
when using partial rebreathing technol-

Table 2. Summary of data from reviewed articles comparing cardiac output (CO) measurements
obtained with the noninvasive continuous cardiac output (NICO) system with those obtained with a
reference method
Reference
34
29
30
12
32
7
14
39
10
28
31
8
35
40
15
13
11
Our study
33

Bias
0.18
0.1
0.05
0.04
0.01
0.01
0.04
0.07
0.07
0.09
0.13
0.16
0.17
0.21
0.27
0.54
0.58
0.61
0.90

SD

of Bias

1.04
0.6
0.70
1.07
0.4
0.69
0.82
0.70
0.91
1.00
0.46
0.90
1.45
0.71
1.49
0.92
0.9
0.94
2.71

Limits of Agreement

Patients

Age

1.78 to 2.34
1.10 to 1.30
1.35 to 1.45
2.10 to 2.20
0.79 to 0.81
1.39 to 1.37
1.68 to 1.76
2.1 to 0.70
1.90 to 1.75
1.91 to 2.09
1.05 to 0.79
1.96 to 1.64
5.97 to 5.63
1.64 to 1.22
3.25 to 2.71
2.38 to 1.30
2.38 to 1.22a

25
20
41
68
12
11
15
6
29
25
32
30
22
46
37
39
28
21
12

Adults
Adults
Adults
Adults
Adults
Pigs
Adults
Dogs
Adults
Adults
Adults
Adults
Adults
Adults
Children
Adults
Adults
Children
Adults

4.52 to 6.32

Methods
NICO/TDCO
NICO/TDCO
NICO/TDCO
NICO/TDCO
NICO/TDCO
NICO/CCO
NICO/TDCO
NICO/TDCO
NICO/TDCO
NICO/TDCO
NICO/TDCO
NICO/TDCO
NICO/TDCO
NICO/CCO
NICO/TDCO
NICO/TOE
NICO/CCO
NICO/TTE
NICO/TDCO

TDCO, discontinuous thermodilution; CCO, continuous thermodilution; TTE, transthoracic

echography; TOE, transesophageal echography.
a
In our study, the variability of the differences increased as the magnitude of the CO measurement
increased. For a mean CO of 1.78 L/min, limits of agreement were 1.40 and 0.17 L/min. For a mean
CO of 8.68 L/min, limits of agreement were 4.51 and 3.28 L/min. Mean bias, SD of bias, and limits of
agreement are shown for each study. Subject groups, number of patients in each study, and methods
compared are also shown. Data published in abstracts are not mentioned in this table.

Pediatr Crit Care Med 2006 Vol. 7, No. 3

ogy; this is probably due to pulmonary

dead space (VD/VT) ratio and alveolar ventilation modifications (28, 29, 34). At the
present time, measuring CO with the
NICO system must be considered as unfeasible in children of 15 kg. Another
limitation of our study is that the measurements were performed only in children who were in stable respiratory and
hemodynamic condition (i.e., with relatively normal CO).
Because the NICO system measures
effective pulmonary blood flow, it was
important to minimize all factors, such
as abnormalities in ventilationperfusion
matching, that could increase the difference between pulmonary blood flow and
systemic CO. However, Jaffe (6) considers
that if the true shunt is 10%, this translates to only a 2% error in CO measured
by the NICO system.
Further studies are necessary to evaluate the accuracy and reproducibility of
the NICO system under various unstable
hemodynamic and respiratory conditions, particularly in children with low
and high CO. In fact, several authors
found less agreement between NICO and
thermodilution techniques at higher CO
(9, 10). In the same manner, Levy et al.
(15) showed that in children, CONICO underestimated CO measured by thermodilution at lower average COs and overestimated CO measured by thermodilution
at higher COs. Finally, this study did not
verify the fundamental equation of the
NICO, which supposes constant VD/VT,
constant CO, and constant mixed venous
CO2 during the rebreathing period. Yem
et al. (37), using a mathematical model of
the cardiorespiratory system, found that
systematic errors in CO measurement by
using the NICO system may be due to
three important mechanisms: recirculation, alveolar-proximal airway PCO2 and
CO2 flux differences, and inadequate rebreathing times. However, this model
was criticized because it bared little resemblance to what was implemented in
the presently commercially available
NICO system (38 40). Odenstedt et al.
(14), using a mechanical lung model, observed that increased respiratory rate led
to a substantial underestimation of VCO2,
which clearly did not introduce substantial bias in CO measurements. This may
be due to the measurement of VCO2 and
PetCO2 with the same gas monitor, minimizing the effect of the CO2 measurement error.
Pediatr Crit Care Med 2006 Vol. 7, No. 3

CONCLUSION
In conclusion, the NICO system is a
new device based on the partial CO2 rebreathing technique, which measures CO
easily and automatically. In our mechanically ventilated children weighing 15
kg, CO values obtained with this technique were in agreement with those obtained with Doppler echocardiography.
However, the NICO system needs further
clinical investigation in children in unstable hemodynamic and respiratory condition.

13.

14.

15.

REFERENCES
1. Proulx F, Gauthier M, Nadeau D, et al: Timing and predictors of death in pediatric patients with multiple organ system failure.
Crit Care Med 1994; 22:10251031
2. Shekerdemian L, Bohn D: Cardiovascular effects of mechanical ventilation. Arch Dis
Child 1999; 80:475 480
3. Egan JR, Festa M, Cole AD, et al: Clinical
assessment of cardiac performance in infants
and children following cardiac surgery. Intensive Care Med 2005; 31:568 573
4. Tibby SM, Hatherill M, Marsh MJ, et al: Clinicians abilities to estimate cardiac index in
ventilated children and infants. Arch Dis
Child 1997; 77:516 518
5. Chew MS, Poelaert J: Accuracy and repeatability of pediatric cardiac output measurement using Doppler: 20-year review of the
literature. Intensive Care Med 2003; 29:
1889 1894
6. Jaffe MB: Partial CO2 rebreathing cardiac
output: Operating principles of the NICO system. J Clin Monit Comput 1999; 15:387 401
7. Maxwell RA, Gibson JB, Slade JB, et al: Noninvasive cardiac output by partial CO2 rebreathing after severe chest trauma.
J Trauma 2001; 51:849 853
8. Nilsson LB, Eldrup N, Berthelsen PG: Lack of
agreement between thermodilution and carbon dioxide-rebreathing cardiac output. Acta
Anaesthesiol Scand 2001; 45:680 685
9. Van Heerden PV, Baker S, Lim SI, et al:
Clinical evaluation of the non-invasive cardiac output (NICO) monitor in the intensive
care unit. Anaesth Intensive Care 2000; 28:
427 430
10. Murias GE, Villagra A, Vatua S, et al: Evaluation of a noninvasive method for cardiac
output measurement in critical care patients. Intensive Care Med 2002; 28:
1470 1474
11. Kotake Y, Moriyama K, Innami Y, et al: Performance of noninvasive partial CO2 rebreathing cardiac output and continuous
thermodilution cardiac output in patients
undergoing aortic reconstruction surgery.
Anesthesiology 2003; 99:283288
12. Botero M, Kirby D, Lobato EB, et al: Measurement of cardiac output before and after
cardiopulmonary bypass: Comparison among

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

aortic transit-time ultrasound, thermodilution and noninvasive partial CO2 rebreathing. J Cardiothorac Vasc Anesth 2004; 18:
563572
Bein B, Hanne P, Hanss R, et al: Effect of
xenon anesthesia on accuracy of cardiac output measurement using partial CO2 rebreathing. Anaesthesia 2004; 59:1104 1110
Odenstedt H, Stenqvist O, Lundin S: Clinical
evaluation of a partial CO2 rebreathing technique for cardiac output monitoring in critically ill patients. Acta Anaesthesiol Scand
2002; 46:152159
Levy RJ, Chiavacci RM, Nicolson SC, et al: An
evaluation of a noninvasive cardiac output
measurement using partial carbon dioxide
rebreathing in children. Anesth Analg 2004;
99:16421647
Nunn JF. Applied Respiratory Physiology.
Fourth Edition. Oxford, Butterworth-Heinemann, 1993
NICO Cardiopulmonary Management System, Manuel de lutilisateur (moniteur cardiopulmonaire, modle 7300). Fevrier 2002,
Novametrix Medical Systems
Armitage P, Berry G. Intraclass correlation.
In: Statistical Methods in Medical Research.
Oxford, Blackwell Scientific Publications,
1994, pp 273276
Bland JM, Altman DG: Measuring agreement
in method comparison studies. Stat Methods
Med Res 1999; 8:135160
Critchley LA, Critchley JA: A meta-analysis of
studies using bias and precision statistics to
compare cardiac output measurement techniques. J Clin Monit Comput 1999; 15:8591
Barker SJ: Blood volume measurement: The
next intraoperative monitor ? Anesthesiology
1998; 89:1310 1312
Tibby SM, Murdoch IA: Monitoring cardiac
function in intensive care. Arch Dis Child
2003; 88:46 52
Stetz CW, Miller RG, Kelly GE, et al: Reliability of the thermodilution method in the
determination of cardiac output in clinical
practice. Am Rev Respir Dis 1982; 126:
10011004
Synder JV, Powner DJ: Effects of mechanical
ventilation on the measurement of cardiac
output by thermodilution. Crit Care Med
1982; 10:677 682
Bazaral MG, Petre J, Novoa R: Errors in thermodilution cardiac output measurements
caused by rapid pulmonary artery temperature decreases after cardiopulmonary bypass.
Anesthesiology 1992; 77:3137
Alverson DC, Eldridge M, Dillon T, et al:
Noninvasive pulsed Doppler determination of
cardiac output in neonates and children.
J Pediatr 1982; 101:46 50
Dhingra VK, Fenwick JC, Walley KR, et al:
Lack of agreement between thermodilution
and Fick cardiac output in critically ill patients. Chest 2002; 122:990 997
Tachibana K, Imanaka H, Takeuchi M, et al:
Noninvasive cardiac output measurement
using partial carbon dioxide rebreathing is
less accurate at settings of reduced minute

235

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

236

ventilation and when spontaneous breathing

is present. Anesthesiology 2003; 98:830 837
de Abreu MG, Geiger S, Winkler T, et al:
Evaluation of a new device for noninvasive
measurement of nonshunted pulmonary capillary blood flow in patients with acute lung
injury. Intensive Care Med 2002; 28:318 323
Binder JC, Parkin WG: Non-invasive cardiac
output determination: Comparison of a new
partial rebreathing technique with thermodilution. Anaesth Intensive Care 2001; 29:
19 23
Neuhauser C, Muller M, Brau M, et al: Partial
CO2 rebreathing technique versus thermodilution: Measurement of cardiac output before and after operations with extra-corporal
circulation. Anaesthesist 2002; 51:625 633
Rocco M, Spadetta G, Morelli A, et al: A
comparative evaluation of thermodilution
and partial CO2 rebreathing techniques for
cardiac output assessment in critically ill patients during assisted ventilation. Intensive
Care Med 2004; 30:82 87
Valiatti JL, Amaral JL: Comparison between
cardiac output values measured by thermodilution and partial carbon dioxide rebreathing
in patients with acute lung injury. Sao Paulo
Med J 2004; 122:233238
Tachibana K, Imanaka H, Miyano H, et al:
Effect of ventilatory settings on accuracy of
cardiac output measurement using partial
CO2 rebreathing. Anesthesiology 2002; 96:
96 102
Mielck F, Buhre W Hanekop G, et al: Comparison of continuous cardiac output measurements in patients after cardiac surgery.
J Cardiothorac Vasc Anesth 2003; 17:
211216
Espersen K, Jensen EW, Rosenborg D, et al:
Comparison of cardiac output measurement
techniques: Thermodilution, Doppler, CO2rebreathing and the direct Fick method. Acta
Anaesthesiol Scand 1995; 39:245251
Yem JS, Tang Y, Turner MJ, et al: Sources of
error in noninvasive pulmonary blood flow
measurements by partial rebreathing. Anesthesiology 2003; 98:881 887
Orr JA, Kuck K, Brewer LM: Noninvasive
cardiac output monitor algorithms are more
sophisticated and perform better than indicated in modeling paper. Anesthesiology
2003; 99:14611462
Haryadi DG, Orr JA, Kuck K, et al: Partial
CO2 rebreathing indirect Fick technique for
non-invasive measurement of cardiac output. J Clin Monit Comput 2000; 16:361374

40. Tsujimoto S, Arimura Y, Kuroda N, et al:

Introduction and clinical evaluation of a new
non-invasive cardiac output monitor (NICO)
based on Fick partial CO 2 rebreathing
method. Masui 2001; 50:799 804

APPENDIX
The partial CO2 rebreathing combines
measurements obtained during a nonrebreathing period (nonrebr) with the ones
obtained during a subsequent rebreathing period (rebr). CO is then computed
with an alternate form of the Fick equation that has been termed the differential
Fick partial rebreathing method. It uses
the change in CO2 elimination (VCO2) and
the change in end-tidal CO2 in response
to a change in ventilation to calculate a
CO value. By combining the CO2 Fick
equations from the baseline (QPCBF
VCO2nonrebr/[CvCO2nonrebr CaCO2nonrebr])
and rebreathing periods (Q PCBF
VCO2rebr/[CvCO2rebr CaCO2rebr]) and by
assuming no significant change of pulmonary blood flow during the measurement period, pulmonary blood flow
(QPCBF) can be expressed as follows
QPCBF VCO2nonrebr VCO2rebr
/[CvCO2nonrbre CaCO2nonbre)
CvCO22rebr CaCO22rebr.

[1]

The method assumes that the venous CO2

concentration remains relatively constant throughout the rebreathing and the
nonrebreathing periods
QPCBF VCO2nonrebr VCO2rebr
/[CaCO2rebr CaCO2nonrbre) VCO2/CaCO
[2]
in which VCO2 and CaCO2 are the
change in CO2 elimination in mL/min
and the change in alveolar blood content
in mL/mL blood between baseline (nonrebr) and rebreathing (rebr) periods. As
the difference between CaCO2 and endtidal CO2 partial pressure (PetCO2) re-

mains constant, the difference in PetCO2

reflects the difference in CaCO2. Using the
PaCO2 and CO2 dissociation curve, CaCO2
is estimated by the following equation:
CaCO2 (6.957 Hb 94.864) log
(1.0 0.1933 PaCO2), where Hb is
hemoglobin.
Because the partial rebreathing
method measures only QPCBF and not the
total cardiac output, Q, that represents
the sum of QPCBF and the shunt flow Qs
(Q QPCBF Qs), the NICO uses arterial
blood saturation measured by pulse
oximetry (SpO2) and FIO2 to estimate the
intrapulmonary shunt flow (Qs), based on
Nunns iso-plots that describe the relationship between arterial oxygen tension
and FIO2 for different levels of intrapulmonary shunt (16).
The NICO monitor continuously measures airway flow, pressure, and CO2. It
uses a rebreathing valve with large-bore
tubing combined with a dual CO2/flow
sensor inserted between the patient and
the Y piece of the ventilator circuit. During the baseline mode (nonrebr), gas flow
is straight through the valve (32 mL of
dead space and resistance of 6 cm
H2OL1sec1). During the rebreathing
period (rebr), flow is diverted through the
expandable loop, which adds from 150 to
450 mL of dead space, depending on the
degree of its expansion. The partial rebreathing valve is pneumatically controlled using solenoid valves and a pump
fail-safe designed in such a way that disconnection or pump or valve failures
cause the valve to be restored to the baseline mode. Between each rebreathing period, which lasts 35 secs, a stabilization
period (baseline mode), which lasts 85
secs, allows the CO2 stores of the patient
to recover (17).
The use of the NICO system is contraindicated in patients who cannot tolerate
a change of up to several torr in the
PaCO2, such as in children with severe
hypercarbia, pulmonary hypertension, or
raised intracranial pressure.

Pediatr Crit Care Med 2006 Vol. 7, No. 3