1. CRITICAL CARE
Third-generation FloTrac/Vigileo does not reliably track
changes in cardiac output induced by norepinephrine
in critically ill patients
X. Monnet1,2*, N. Anguel1,2, M. Jozwiak1,2, C. Richard1,2 and J.-L. Teboul1,2
1
Hoˆpitaux universitaires Paris-Sud, Hoˆpital de Biceˆtre, service de re´animation me´dicale, 78, rue du Ge´ne´ral Leclerc, Le Kremlin-Biceˆtre
F-94270, France
2
Universite´ Paris-Sud, Faculte´ de me´decine Paris-Sud, EA 4046, 63, rue Gabriel Pe´ri, Le Kremlin-Biceˆtre F-94270, France
* Corresponding author. E-mail: xavier.monnet@bct.aphp.fr
Editor’s key points
† FloTrac/Vigileo and other
devices are available to
measure cardiac output
by pressure waveform
analysis.
† The performance of the
third-generation FloTrac/
Vigileo in measuring
cardiac index (CI) was
assessed in critically ill
patients undergoing
volume expansion
or changes in
norepinephrine dose.
† This device was
moderately reliable for
tracking volume-induced
changes in CI, but poorly
reliable with
norepinephrine dose
titration.
Background. The ability of the third-generation FloTrac/Vigileo software to track changes in
cardiac index (CI) induced by volume expansion and norepinephrine in critically ill patients
is unknown.
Methods. In subjects with circulatory failure, we administered volume expansion (20
subjects) and increased (20 subjects) or decreased (20 subjects) the dose of
norepinephrine. We measured arterial pressure waveform-derived CI provided by the
third-generation FloTrac/Vigileo device (CIpw) and transpulmonary thermodilution CI (CItd)
before and after therapeutic interventions.
Results. Considering the pairs of measurements performed before and after all therapeutic
interventions (n¼60), a bias between the absolute values of CIpw and CItd was 0.26 (0.94)
litre min21
m22
and the percentage error was 54%. Changes in CIpw tracked changes in CItd
induced by volume expansion with moderate accuracy [n¼20, bias¼20.11 (0.54) litre
min21
m22
, r2
¼0.26, P¼0.02]. When changes in CItd were induced by norepinephrine
(n¼40), a bias between CIpw and CItd was 0.01 (0.41) litre min21
m22
(r2
¼0.11, P¼0.04).
The concordance rates between changes in CIpw and CItd induced by volume expansion
and norepinephrine were 73% and 60%, respectively. The bias between changes in CIpw
and CItd significantly correlated with changes in total systemic vascular resistance
(r2
¼0.41, P,0.0001).
Conclusions. The third-generation FloTrac/Vigileo device was moderately reliable for
tracking changes in CI induced by volume expansion and poorly reliable for tracking
changes in CI induced by norepinephrine.
Keywords: cardiac output, measurement; equipment, monitors; measurement techniques,
cardiac output; norepinephrine; shock
Accepted for publication: 24 November 2011
In recent years, efforts have been made to develop devices
that allow beat-to-beat estimation of cardiac output (CO).
Some of these systems compute CO from the arterial pres-
sure waveform, using the principle that stroke volume is
physiologically related to arterial pressure, aortic compliance,
and arterial tone.1
In particular, some ‘uncalibrated’ devices
estimate aortic compliance and arterial tone from an ana-
lysis of the geometric properties of arterial shape and from
some patient characteristic data. These devices are in con-
trast to some ‘calibrated’ systems, which add to this continu-
ous estimation of CO an external calibration by a reference
technique.
The uncalibrated systems have the great advantage of not
requiring a specific system for recalibrating CO measure-
ment. Nevertheless, their ability to track changes in CO has
been questioned,2 – 9
especially when arterial tone changes
to a large extent3
or during hyperdynamic states.7 8 10
In a
previous study, we suggested that the second-generation
FloTrac/Vigileow
device had a poor ability for tracking
changes in cardiac index (CI) induced by norepinephrine in
septic patients.11
We hypothesized that changes in arterial
compliance and arterial tone induced by the vasopressor dis-
torted the arterial pressure wave analysis made by the unca-
librated system. In a recent study, a third generation of this
British Journal of Anaesthesia 108 (4): 615–22 (2012)
Advance Access publication 19 January 2012 . doi:10.1093/bja/aer491
& The Author [2012]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
byguestonApril26,2013http://bja.oxfordjournals.org/Downloadedfrom

2. system has been demonstrated to be as precise, more accur-
ate, and less influenced by systemic vascular resistance (SVR)
than the previous version.12
The third-generation system was
not better than the second for detecting significant changes
in CI over time,12
but these changes were not induced by sys-
tematic therapeutic interventions. Moreover, a recent study
suggested that the third version of the FloTrac/Vigileo
device does not accurately track changes in CO induced
with phenylephrine and ephedrine in the intraoperative
setting.13
In the present study, we tested whether the third-
generation FloTrac/Vigileo device allows tracking trends in
CI induced either by volume expansion or by changes in
the dose of norepinephrine in critically ill patients.
Methods
Subjects
After approval by the Institutional Review Board of our insti-
tution, subjects were enrolled if a volume expansion (20 sub-
jects) or an increase (20 different subjects) or a decrease (20
different subjects) in the dose of norepinephrine was planned
by the attending physician. Subjects’ relatives were informed
about the study at the time of enrolment with the possibility
of refusing participation at that time. Subjects were informed
as soon as their mental status allowed, and the possibility
was given to withdraw their participation in the study.
All subjects had a catheter inserted into the internal
jugular vein and a catheter inserted into the femoral artery
(PV8215 monitoring kit, Pulsion Medical Systems, Munich,
Germany). The arterial line was divided into two branches
through a stopcock, one connected to a PiCCO2 device
(Pulsion Medical Systems) and the other connected to a
third-generation FloTrac/Vigileo device (Edwards Life-
sciences, Irvine, CA, USA).
Measurements and study design
Before each therapeutic intervention, we performed a first
set of haemodynamic measurements, including heart rate,
systemic arterial pressure, CI measured by transpulmonary
thermodilution (CItd), CI measured by pulse-wave analysis
by the third-generation FloTrac/Vigileo device (CIpw), and
SVR. We used the values of CIpw automatically displayed on
the screen of the FloTrac/Vigileo device averaged over a
20 s rolling period. The CItd was measured by the PiCCO2
device by injecting 15 ml of iced saline (,108C) through
the central venous line. The injection was performed in trip-
licate and the values of CItd were averaged. Immediately
before performing thermodilution boluses, the value of CIpw
Table 1 Subject characteristics at baseline. n¼60. Data are
expressed as mean (SD), median (25–75% inter-quartile) or n (%).
SAPS, Simplified Acute Physiologic Score; ARDS: acute respiratory
distress syndrome; MAP, mean arterial pressure; Cltd, cardiac index
measured by transpulmonary thermodilution; PaO2
/FIO2
, ratio of
the arterial oxygen tension over the inspired oxygen fraction
Age (yr) 64 (15)
Gender (M/F) 41/19
SAPS II 45 (11)
ARDS (n, %) 34 (57)
Mechanical ventilation (n, %) 60 (100)
Respiratory variables
Tidal volume (ml kg21
of predicted
body weight)
7 (2)
Respiratory rate (breaths min21
) 20 (4)
Total positive end-expiratory pressure
(cm H2O)
8 (2)
PaO2
/FIO2
(mm Hg) 220 (100)
Shock aetiology
Septic (n, %) 48 (80)
Haemorrhagic (n, %) 6 (10)
Drug poisoning (n, %) 6 (10)
CItd (litre min21
m22
) 3.4 (1.3)
Systemic vascular resistance
(dyn s cm25
)
938 (739–1194)
Vasopressors
Norepinephrine (n, %) 56 (93)
Dose of norepinephrine
(mg kg21
min21
)
0.16 (0.04–0.41)
Dobutamine (n, %) 2 (3)
Table 2 Evolution of haemodynamic parameters during therapeutic interventions. Data are expressed as mean (SD) or as median (25–75%
inter-quartile). *P,0.05 vs before intervention, †
P,0.05 vs volume expansion. MAP, mean arterial pressure; CItd, cardiac index measured by
thermodilution; CIpw, arterial pressure waveform-based cardiac index measured by the FloTrac/Vigileo device; SVR, systemic vascular resistance
Volume expansion (n520) Increase in norepinephrine (n520) Decrease in norepinephrine (n520)
Before After Before After Before After
Heart rate (beats min21
) 102 (23) 98 (23) 82 (16)†
88 (23) 87 (16)†
85 (17)
MAP (mm Hg) 75 (15) 83 (14)* 61 (13)†
82 (13)* 74 (62) 62 (13)*,†
CItd (litre min21
m22
) 3.3 (1.5) 3.8 (1.5) 3.1 (1.1) 3.3 (1.1) 3.7 (1.3) 3.5 (1.3)
CIpw (litre min21
m22
) 3.3 (1.3) 3.7 (1.3) 2.7 (0.6) 3.2 (0.8) 3.4 (0.9) 3.0 (0.9)
SVR (dyn s cm25
) 940 (861–1283) 927 (844–1169) 916 (677–1153) 1149 (929–1367)* 952 (696–1145) 846 (659–1064)*
Dose of norepinephrine
(mg kg21
min21
)
0.47 (0.31–0.84) 0.47 (0.31–0.84) 0.07 (0.02–0.19)†
0.21 (0.19–0.45)* 0.13 (0.07–0.20)†
0.03 (0.00–0.12)*,†
BJA Monnet et al.
616
byguestonApril26,2013http://bja.oxfordjournals.org/Downloadedfrom

3. was carried out. The CIpw was recorded immediately before
thermodilution to avoid interference between temperature
drift and accuracy of CIpw. The total SVR was calculated as
SVR¼mean arterial pressure×80/(CItd×body surface area).
After the first set of haemodynamic measurements was
completed, volume expansion was performed (500 ml of
saline over 30 min) or the dose of norepinephrine was
increased or decreased. All other treatments were un-
changed during the therapeutic interventions.
A second set of haemodynamic measurements was
carried out again after the therapeutic intervention (i.e. at
the end of fluid administration and 5 min after stabilization
of mean arterial pressure). This set included heart rate, sys-
temic arterial pressure, CIpw, CItd, and SVR.
Statistical analysis
All data were normally distributed (Kolmogorov–Smirnov
test) except the dose of norepinephrine and are expressed
as mean [standard deviation (SD)] or median [25–75% inter-
quartile range], as appropriate. Comparisons between values
recorded before and values after therapeutic interventions
were performed in both groups by paired Student’s t-test
or paired Wilcoxon’s test, as appropriate. Comparisons
between subjects receiving volume expansion, subjects in
whom the dose of norepinephrine was increased, and
subjects in whom the dose of norepinephrine was decreased
were performed by a two-tailed Student’s t-test or a Mann–
Whitney U-test, as appropriate. Correlations were assessed
by the Pearson coefficient and correlation coefficients were
compared using the Fisher transformation.14
This analysis
was also separately performed in subjects in whom SVR
changed in absolute value by more or less than 15% with
the therapeutic interventions.15
We compared the relative changes of CIpw with those of
CItd during the therapeutic intervention by the Bland and
Altman analysis (for absolute changes) and by linear regres-
sion analysis (for per cent changes). For assessing the trend-
ing ability of CIpw, we constructed a four-quadrant plot.16
This allowed calculation of the percentage of total data
points for which the direction changes of CIpw (increase or
decrease) were concordant with CItd. Since the least signifi-
cant change of CItd is 12% when three thermodilution mea-
surements are performed,17
we applied a 12% exclusion
limit. We also constructed a receiver operating characteristic
(ROC) curve to test the ability of changes in CIpw to detect an
increase in CItd ≥12% induced by volume expansion. We cal-
culated the percentage error of CIpw as 2× SD mean21
of
CItd.18
A P value of ,0.05 was considered significant. The
statistical analysis was performed using MedCalc8.1.0.0 soft-
ware (Mariakerke, Belgium).
Results
Subject characteristics
Subject characteristics at baseline are summarized in Table 1.
Circulatory failure was of septic origin in the majority of sub-
jects. All subjects received norepinephrine at baseline. Sub-
jects deemed as receiving volume expansion received
0 1 2 3 4 5 6 7
–2.0
–1.5
–1.0
–0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
(CIpw+CItd)/2 (litremin–1 m–2)
0.26
–1.63
2.15
0 1 2 3 4 5 6 7
CIpw–CItd(litremin–1m–2)
Fig 1 The Bland–Altman plot for the absolute values of CI
obtained by transpulmonary thermodilution (CItd) and by the
third-generation FloTrac/Vigileo device (CIpw) considering all
pairs of measurements performed during the study. n¼60;
straight line, bias; dashed line, +2SD/22SD limits of agreement.
Table 3 Comparison of changes in CI measured by pulse-wave analysis and transpulmonary thermodilution. Data are expressed as absolute
values. CItd, cardiac index measured by thermodilution; CIpw, arterial pressure waveform-based CI measured by the FloTrac/Vigileo device
Volume expansion
(n520)
Increase in norepinephrine
(n520)
Decrease in norepinephrine
(n520)
r2
between per cent changes in CItd and CIpw 0.26 2.5×1023
0.16
P-value for the correlation between per cent
changes in CItd and CIpw
0.02 0.81 0.07
Bias for the absolute changes in CIpw compared
with CItd (litre min21
m22
)
0.11 0.36 0.17
Upper limit of agreement for the absolute
changes in CIpw (litre min21
m22
)
0.98 1.98 0.92
Lower limit of agreement for the absolute
changes in CIpw (litre min21
m22
)
21.20 21.25 21.26
Third-generation FloTrac in critically ill patients BJA
617
byguestonApril26,2013http://bja.oxfordjournals.org/Downloadedfrom

4. norepinephrine at baseline and this dose was kept un-
changed during the study period (Table 2). The second set
of measurements was recorded 32 (4) min after the first
set in subjects receiving volume expansion and 35 (10) min
after the first set in subjects in whom the dose of norepin-
ephrine was changed.
–20 0 20 40 60 80 100
–20
0
20
40
60
80
100
r 2=0.26
P=0.02
–1.5 –1.0 –0.5 0.0 0.5 1.0 1.5 2.0
–2.5
–2.0
–1.5
–1.0
–0.5
0.0
0.5
1.0
1.5
2.0
2.5
–0.11
0.98
–80
–60
–40
–20
20
40
60
80
–80 –60 –40 0 20 40 60 80
C
Concordance=73%
(Exclusion zone 12%)
(Without exclusion zone=70%)
0
–20
(CIpw+CItd)/2 (litremin–1 m–2)
CItd (%)
CItd (%)
CIpw(%)
CIpw (%)
(CIpw–CItd)(litremin–1m–2)
–1.20
B
A
Fig 2 (A) The Bland–Altman plot for the changes in absolute values induced by volume expansion of CI measured by transpulmonary thermo-
dilution (CItd) and by arterial pressure waveform analysis by the third-generation FloTrac/Vigileo device (CIpw). (B) Correlation between the per
cent changes induced by volume expansion of CI measured by transpulmonary thermodilution (DCItd) and by arterial pressure waveform ana-
lysis by the third-generation FloTrac/Vigileo device (DCIpw). (C) Trending ability of the third-generation FloTrac/Vigileo device (DCIpw) against CI
measured by transpulmonary thermodilution (DCItd) during volume expansion based on four-quadrant concordance analysis. n¼20. The
Bland–Altman plots: straight line, bias; dashed line, +2SD/22SD limits of agreement. Correlation: dashed line, correlation line.
BJA Monnet et al.
618
byguestonApril26,2013http://bja.oxfordjournals.org/Downloadedfrom

5. Comparison between CItd and CIpw
Considering the pairs of measurements performed before
and after all therapeutic interventions (n¼60), the bias
between the absolute values of CIpw and CItd was 0.26
(0.94) litre min21
m22
and the percentage error was
54% (Fig. 1).
Comparison of CIpw with CItd in subjects receiving
volume expansion
In subjects receiving volume expansion, mean arterial pres-
sure, CItd, and CIpw significantly increased by 12 (9)%, 23
(23)%, and 19 (22)%, respectively. SVR did not significantly
decrease (Table 2).
The bias between absolute changes in CIpw and CItd
induced by volume expansion was 20.11 (0.54) litre min21
m22
. The coefficient of determination (r2
) between
fluid-induced per cent changes in CIpw and CItd was 0.26
(P¼0.02) (Table 3 and Fig. 2). The concordance rate
between changes in CIpw and CItd induced by volume expan-
sion was 73% (Fig. 2). After volume expansion, the bias
between the absolute values of CIpw and CItd was 20.15
(0.88) litre min21
m22
and the percentage error was 48%.
The area under the ROC curve constructed for the changes
in CIpw for detecting an increase in CItd ≥12% was not sig-
nificantly different from 0.5 (Fig. 3).
Comparison of CIpw with CItd in subjects with a
change in dose of norepinephrine
In subjects in whom the dose of norepinephrine was
increased, mean arterial pressure, SVR, CItd, and CIpw
significantly increased by 22 (25)%, 8 (28)%, 9 (21)%, and
20 (19)%, respectively (Table 2). In subjects in whom the
dose of norepinephrine was decreased, mean arterial pres-
sure, SVR, CItd, and CIpw significantly decreased by 15
(13)%, 8 (17)%, 9 (18)%, and 12 (12)%, respectively (Table 2).
Considering subjects in whom norepinephrine was
increased or decreased as one group (n¼40), the bias
between the absolute changes in CIpw and CItd induced by
norepinephrine decrease/increase was 0.11 (0.68) litre
min21
m22
. The coefficient of determination (r2
) between
the norepinephrine-induced per cent changes in CIpw and
in CItd was 0.11 (P¼0.04) (Fig. 4). The concordance rate
between the changes in CIpw and CItd induced by changing
the dose of norepinephrine was 60% (Fig. 4). After the de-
crease/increase of norepinephrine dose, the bias between
the absolute values CIpw and CItd was 20.30 (1.04) litre
min21
m22
and the percentage error was 61%. Results con-
cerning separate groups of subjects with an increase and a
decrease in the dose of norepinephrine are shown in Table 3.
Effects of changes in SVR on the agreement
of CIpw with CItd
Considering the aggregate of all therapeutic interventions,
changes in SVR ranged from 231% to +73%. The bias
between changes in CIpw and CItd significantly correlated
with changes in SVR (r2
¼0.41, P,0.0001).
In the subset of subjects in whom SVR (absolute value)
changed by less than 15% (n¼33), the bias between the ab-
solute changes in CIpw and CItd was 20.12 (0.43) litre min21
m22
. In these subjects, there was no significant correlation
between SVR and bias between CItd and CIpw (P¼0.29).
In the subset of patients in whom the SVR (in absolute
value) increased by more than 15% (n¼27), the bias
between the absolute changes in CIpw and CItd was 0.22
(0.89) litre min21
m22
. In these patients, the coefficient of
determination (r2
) between SVR and the per cent changes
in CItd and CIpw was 0.49 (P,0.0001).
Discussion
The third-generation FloTrac/Vigileo device was not reliable
for detecting trends in CI, especially when induced by nor-
epinephrine. The higher the total SVR, the higher was the
bias between CI measured by FloTrac/Vigileo and by trans-
pulmonary thermodilution.
Monitoring CO in critically ill patients is recommended
when shock persists despite adequate fluid resuscitation.19
A recent study also suggests that changes in arterial pressure
are unable to reliably monitor the changes in CI induced by
vasopressors,20
reinforcing the message that CO should be
measured in critically ill patients after initial fluid resuscita-
tion. Among the several techniques that are currently avail-
able for measuring CI, the estimation of CI computed from
arterial pressure waveform analysis has the advantage of pro-
viding beat-to-beat estimation of CI. In fact, three elements
influence the relationship between the shape of the periph-
eral arterial wave and stroke volume: arterial compliance,
0 20 40 60 80 100
0
20
40
60
80
100
100-specificity
Sensitivity
Fig 3 An ROC curve constructed for testing the ability of the
changes in CI by the third-generation FloTrac/Vigileo device to
detect an increase in CI obtained by transpulmonary thermodilu-
tion ≥12% induced by volume expansion.
Third-generation FloTrac in critically ill patients BJA
619
byguestonApril26,2013http://bja.oxfordjournals.org/Downloadedfrom

6. –60 –40 –20 0 20
–1.5 –1.0 –0.5 0.0 0.5 1.0 1.5 2.0
40 60 80 100
–60
–40
–20
0
20
40
60
80
100
0.11
- –1.27
1.48
0
0.11
-
1.48
0
-
0
r2=0.11
P=0.04
–2.5
–2.0
–1.5
–1.0
–0.5
0.0
0.5
1.0
1.5
2.0
2.5
20
40
60
80
C
Concordance=60%
(Exclusion zone 12%)
(Without exclusion zone=63%)
(CIpw+CItd)/2 (litremin–1 m–2)
CItd (%)
CIpw(%)
CIpw (%)
(CIpw–CItd)(litremin–1m–2)
B
A
–80
–60
–40
–20
–80 –60 –40 20 40 60 80–20 0
CItd (%)
Fig 4 (A) The Bland–Altman plot for changes in absolute values induced by changes in the dose of norepinephrine on CI measured by trans-
pulmonary thermodilution (CItd) and by arterial pressure waveform analysis by the third-generation FloTrac/Vigileo device (CIpw). (B) Correlation
between per cent changes induced by changes in the dose of norepinephrine on CI measured by transpulmonary thermodilution (DCItd) and by
arterial pressure waveform analysis by the third-generation FloTrac/Vigileo device (DCIpw). (C) Trending ability of the third-generation FloTrac/
Vigileo device (DCIpw) against CI measured by transpulmonary thermodilution (DCItd) during changes in the dose of norepinephrine based on
four-quadrant concordance analysis. n¼40. The Bland–Altman plots: straight line, bias; dashed line, +2SD/22SD limits of agreement. Correl-
ation: dashed line, correlation line.
BJA Monnet et al.
620
byguestonApril26,2013http://bja.oxfordjournals.org/Downloadedfrom

7. arterial vasomotor tone, and pulse-wave amplification phe-
nomenon.1
Thus, estimation of CI by arterial pressure wave-
form analysis is based upon a geometric analysis of the
shape of the arterial pressure curve which is then adjusted
with a factor taking into account the arterial compliance
and tone and the pulse-wave amplification phenomenon.
The devices currently available on the market are fundamen-
tally different. The ‘calibrated’ devices, namely the PiCCO
(Pulsion Medical Systems), EV1000 (Edwards Lifesciences),
and LidCOplus (LidCO) devices calibrate the estimation of
CO made from the arterial pressure curve by an external cali-
bration of CI by a reference method (transpulmonary thermo-
dilution for the PiCCO and EV1000 and lithium dilution for the
LidCOplus). This external calibration requires periodic injec-
tion of an indicator and the use of a specific material for
measuring dilution. In contrast, some ‘uncalibrated’ devices,
like the FloTrac/Vigileo (Edwards Lifesciences), LidCOrapid
(LidCO), or Pulsioflex (Pulsion Medical Systems), continuously
estimate arterial compliance and tone and pulse-wave amp-
lification from a complex geometric analysis of the arterial
wave and from some biometric data.
Reliability of uncalibrated pressure waveform analysis
for measuring CO has been demonstrated in several
studies.12 21 –24
Nevertheless, concerns have been raised
about the validity of the technique when the arterial
tone changes to a large extent,3
during hyperdynamic
states,7 8 10
and when vasopressors are administered, as
we recently showed with the second-generation FloTrac/
Vigileo device.11
To address this problem, a third version of
the FloTrac/Vigileo software has been developed. Compared
with the previous version, estimation of arterial compliance
and tone and pulse-wave amplification has been made
from a human database containing more recordings from
septic and liver transplant patients.12
In a recent study, in
septic shock patients, De Backer and colleagues12
found
that the third-generation FloTrac/Vigileo device is more ac-
curate, as precise as, and less influenced by total SVR than
the previous version. Consistent with this study, we found
that the percentage error for CIpw recorded for all pairs of
measurements was improved compared with results
obtained with the second-generation system11
(54% vs
61%, respectively).
In contrast to De Backer and colleagues, we focused on
the ability of the uncalibrated devices to follow changes in
CI induced by some systematic therapeutic changes. Even
though it was conducted in subjects with total SVR in the
same range as those in the study of De Backer and collea-
gues,12
the present study suggests that the most recent
version of the FloTrac/Vigileo system does not provide a sat-
isfactory track of therapy-induced changes in CO. First, even
though changes in CItd and CIpw were more significantly cor-
related when induced by volume expansion than by norepin-
ephrine, the third-generation FloTrac/Vigileo did not provide
a correct estimation of the response to a standardized fluid
challenge, as defined by an increase in CItd ≥15%. Second
and more importantly, our results suggest that the new
FloTrac/Vigileo device was not reliable for tracking trends in
CI induced by changing the dose of norepinephrine. More-
over, by showing that the accuracy of the third-generation
FloTrac/Vigileo device correlated with total SVR, we suggest
that its estimation of CI is still distorted by a change in vas-
cular tone. The ability to track norepinephrine-induced
changes in CI seems to be improved compared with the pre-
vious version of the system. Indeed, in our previous study,11
we did not find significant correlation between changes in
CIpw and CItd induced by norepinephrine changes, while
this correlation was significant in the present study
(r¼0.35, P,0.05). This suggests substantial but insufficient
improvements in the device software. Importantly, the
present results are in accordance with two recent studies
conducted in other clinical settings. In neurosurgical patients
requiring high-dose vasopressor support, Metzelder and col-
leagues25
showed that the introduction of the third-
generation FloTrac/Vigileo software algorithm did not
improve the insufficient precision for CI measurements
observed with the second software version. In a general
population of anaesthetized subjects, Meng and colleagues13
reported that the newest FloTrac/Vigileo generation accur-
ately tracked changes in CO when preload changed but did
not accurately track changes in CO induced with phenyleph-
rine and ephedrine.
We acknowledge some limitations to our study. First, we
could not perform a direct comparison between the second
and third generations of the FloTrac/Vigileo system since
the previous version is no longer provided by the manufac-
turer. Secondly, as a reference for measuring CI, we used
transpulmonary thermodilution rather than classical thermo-
dilution with the pulmonary artery catheter.26
Nevertheless,
the accuracy of transpulmonary thermodilution in measuring
CO has been repeatedly demonstrated.15 23 27 –33
Thirdly, we
did not test other systems that compute CI from the pressure
waveform without external calibration, so our conclusions
might not apply to these devices.
In conclusion, estimation of CI made from the third-
generation FloTrac/Vigileo device was only moderately reli-
able for detecting the changes in CI induced with volume ex-
pansion and did not allow tracking trends in CI provoked by
changing the dose of norepinephrine in critically ill patients.
As a clinical consequence, this study suggests that this
device should be reserved for patients who are not receiving
vasopressors.
Declaration of interest
J.-L.T. and X.M. are members of the Medical Advisory Board of
Pulsion Medical Systems.
Funding
This study was supported solely by institutional and depart-
mental sources.
References
1 van Lieshout JJ, Wesseling KH. Continuous cardiac output by
pulse contour analysis? Br J Anaesth 2001; 86: 467–9
Third-generation FloTrac in critically ill patients BJA
621
byguestonApril26,2013http://bja.oxfordjournals.org/Downloadedfrom

8. 2 Compton FD, Zukunft B, Hoffmann C, Zidek W, Schaefer JH. Per-
formance of a minimally invasive uncalibrated cardiac output
monitoring system (Flotrac/Vigileo) in haemodynamically un-
stable patients. Br J Anaesth 2008; 100: 451–6
3 Lorsomradee S, Cromheecke S, De Hert SG. Uncalibrated arterial
pulse contour analysis versus continuous thermodilution tech-
nique: effects of alterations in arterial waveform. J Cardiothorac
Vasc Anesth 2007; 21: 636–43
4 Opdam HI, Wan L, Bellomo R. A pilot assessment of the FloTrac
cardiac output monitoring system. Intensive Care Med 2007; 33:
344–9
5 Prasser C, Bele S, Keyl C, et al. Evaluation of a new arterial
pressure-based cardiac output device requiring no external cali-
bration. BMC Anesthesiol 2007; 7: 9
6 Sander M, Spies CD, Grubitzsch H, Foer A, Muller M, von
Heymann C. Comparison of uncalibrated arterial waveform ana-
lysis in cardiac surgery patients with thermodilution cardiac
output measurements. Crit Care 2006; 10: R164
7 Biancofiore G, Critchley LA, Lee A, et al. Evaluation of an uncali-
brated arterial pulse contour cardiac output monitoring system
in cirrhotic patients undergoing liver surgery. Br J Anaesth 2009;
102: 47–54
8 Della Rocca G, Costa MG, Chiarandini P, et al. Arterial pulse cardiac
output agreement with thermodilution in patients in hyperdy-
namic conditions. J Cardiothorac Vasc Anesth 2008; 22: 681–7
9 Biais M, Nouette-Gaulain K, Cottenceau V, Revel P, Sztark F. Unca-
librated pulse contour-derived stroke volume variation predicts
fluid responsiveness in mechanically ventilated patients undergo-
ing liver transplantation. Br J Anaesth 2008; 101: 761–8
10 Biais M, Nouette-Gaulain K, Cottenceau V, et al. Cardiac output
measurement in patients undergoing liver transplantation: pul-
monaryarterycatheter versus uncalibrated arterialpressurewave-
form analysis. Anesth Analg 2008; 106: 1480–6, table of contents
11 Monnet X, Anguel N, Naudin B, Jabot J, Richard C, Teboul JL. Ar-
terial pressure-based cardiac output in septic patients: different
accuracy of pulse contour and uncalibrated pressure waveform
devices. Crit Care 2010; 14: R109
12 De Backer D, Marx G, Tan A, et al. Arterial pressure-based cardiac
output monitoring: a multicenter validation of the third-
generation software in septic patients. Intensive Care Med
2011; 37: 233–40
13 Meng L, Phuong Tran N, Alexander BS, et al. The impact of phenyl-
ephrine, ephedrine, and increased preload on third-generation
Vigileo-FloTrac and esophageal Doppler cardiac output measure-
ments. Anesth Analg 2011; 113: 751–7
14 Fisher RA. Statistical Methods for Research Workers, 14th
Edn. Edinburgh, London: Oliver & Boyd, 1970
15 Hamzaoui O, Monnet X, Richard C, Osman D, Chemla D, Teboul JL.
Effects of changes in vascular tone on the agreement between
pulse contour and transpulmonary thermodilution cardiac
output measurements within an up to 6-hour calibration-free
period. Crit Care Med 2008; 36: 434–40
16 Critchley LA, Lee A, Ho AM. A critical review of the ability of con-
tinuous cardiac output monitors to measure trends in cardiac
output. Anesth Analg 2010; 111: 1180–92
17 Monnet X, Persichini R, Ktari M, Jozwiak M, Richard C, Teboul JL.
Precision of the transpulmonary thermodilution measurements.
Crit Care 2011; 15: R204
18 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: 85–91
19 Antonelli M, Levy M, Andrews PJ, et al. Hemodynamic monitoring
in shock and implications for management. International Con-
sensus Conference, Paris, France, 27–28 April 2006. Intensive
Care Med 2007; 33: 575–90
20 Monnet X, Letierce A, Hamzaoui O, et al. Arterial pressure allows
monitoring the changes in cardiac output induced by volume ex-
pansion but not by norepinephrine. Crit Care Med 2011; 39:
1394–9
21 Button D, Weibel L, Reuthebuch O, Genoni M, Zollinger A,
Hofer CK. Clinical evaluation of the FloTrac/Vigileo system and
two established continuous cardiac output monitoring devices
in patients undergoing cardiac surgery. Br J Anaesth 2007; 99:
329–36
22 Cannesson M, Attof Y, Rosamel P, Joseph P, Bastien O, Lehot JJ.
Comparison of FloTrac cardiac output monitoring system in
patients undergoing coronary artery bypass grafting with pul-
monary artery cardiac output measurements. Eur J Anaesthesiol
2007; 24: 832–9
23 de Waal EE, Kalkman CJ, Rex S, Buhre WF. Validation of a new ar-
terial pulse contour-based cardiac output device. Crit Care Med
2007; 35: 1904–9
24 Mayer J, Boldt J, Schollhorn T, Rohm KD, Mengistu AM,
Suttner S. Semi-invasive monitoring of cardiac output by a
new device using arterial pressure waveform analysis: a com-
parison with intermittent pulmonary artery thermodilution in
patients undergoing cardiac surgery. Br J Anaesth 2007; 98:
176–82
25 Metzelder S, Coburn M, Fries M, et al. Performance of cardiac
output measurement derived from arterial pressure waveform
analysis in patients requiring high-dose vasopressor therapy. Br
J Anaesth 2011; 106: 776–84
26 Richard C, Monnet X, Teboul JL. Pulmonary artery catheter mon-
itoring in 2011. Curr Opin Crit Care 2011; 17: 296–302
27 Bein B, Worthmann F, Tonner PH, et al. Comparison of esophageal
Doppler, pulse contour analysis, and real-time pulmonary artery
thermodilution for the continuous measurement of cardiac
output. J Cardiothorac Vasc Anesth 2004; 18: 185–9
28 Buhre W, Weyland A, Kazmaier S, et al. Comparison of
cardiac output assessed by pulse-contour analysis and thermodi-
lution in patients undergoing minimally invasive direct coronary
artery bypass grafting. J Cardiothorac Vasc Anesth 1999; 13:
437–40
29 Felbinger TW, Reuter DA, Eltzschig HK, Bayerlein J, Goetz AE.
Cardiac index measurements during rapid preload changes: a
comparison of pulmonary artery thermodilution with arterial
pulse contour analysis. J Clin Anesth 2005; 17: 241–8
30 Felbinger TW, Reuter DA, Eltzschig HK, Moerstedt K, Goedje O,
Goetz AE. Comparison of pulmonary arterial thermodilution and
arterial pulse contour analysis: evaluation of a new algorithm.
J Clin Anesth 2002; 14: 296–301
31 Godje O, Thiel C, Lamm P, et al. Less invasive, continuous hemo-
dynamic monitoring during minimally invasive coronary surgery.
Ann Thorac Surg 1999; 68: 1532–6
32 Rodig G, Prasser C, Keyl C, Liebold A, Hobbhahn J. Continuous
cardiac output measurement: pulse contour analysis vs thermo-
dilution technique in cardiac surgical patients. Br J Anaesth 1999;
82: 525–30
33 Zollner C, Haller M, Weis M, et al. Beat-to-beat measurement of
cardiac output by intravascular pulse contour analysis: a pro-
spective criterion standard study in patients after cardiac
surgery. J Cardiothorac Vasc Anesth 2000; 14: 125–9
BJA Monnet et al.
622
byguestonApril26,2013http://bja.oxfordjournals.org/Downloadedfrom