[Downloaded free from http://www.annals.in on Thursday, October 27, 2016, IP: 115.178.239.

182]

Original
44 Sujatha etArticle
al. Thermodilution versus Pulse Contour Cardiac Output Annals of Cardiac Anaesthesia 2006; 9: 4448

Comparison of Cardiac Output in OPCAB: Bolus Thermodilution

Technique versus Pulse Contour Analysis
P Sujatha, Y Mehta, A Dhar, D Sarkar, ZS Meharwal, N Trehan
Department of Anaesthesiology and Critical Care and Department of Cardiac Surgery
Escorts Heart Institute and Research Centre, New Delhi

The study was designed to evaluate the clinical agreement between intermittent bolus thermodilution
technique and pulse contour analysis technique. Sixty patients with normal left ventricular function
undergoing elective off-pump coronary bypass surgery were included in this prospective study. In addition
to routine monitoring, a 7.5F pulmonary artery thermodilution catheter via right internal jugular vein
and a 4F arterial thermodilution catheter into femoral artery were also placed. Cardiac output measurements
were compared before induction, after induction, after sternotomy, during the various anastomoses, post-
protamine and post-sternal closure. Statistical analysis was performed using analysis of agreement to
assure bias distribution of differences between the two methods by using Bland and Altman analysis. The
cardiac output values obtained at preinduction, post-induction, and post-sternal closure time points showed
good agreement, whereas the values obtained during the various anastomoses showed significant differences
(p <0.05). Therefore it was concluded that pulse contour analysis cannot be relied upon completely whenever
there is a change in the position of heart or alteration in systemic vascular resistance. But the trends in
cardiac output were in complete agreement during the entire procedure. (Annals of Cardiac Anaesthesia
2006; 9: 4448)

Key words: Off-pump coronary artery bypass, Cardiac output, Thermodilution, Pulse contour
analysis

C ardiac output (CO) is one of the important

haemodynamic variables during and after
cardiac surgery. CO estimation using thermo-
dilution pulmonary artery catheter (PAC) remains
the current clinical standard1 despite its pitfalls2,3
and potential hazards.4 There has been consensus
regarding the indications for use of PAC despite
the absence of prospective randomized studies
showing improvement in patient outcome. 5
Continuous measurement of CO is preferable to
intermittent measurements, especially during
rapid changes in cardiovascular function. Many
experimental and clinical studies have
demonstrated excellent agreement between
intermittent bolus and continuous CO. 6 The
response time of continuous CO by pulmonary
thermodilution is slow and may not be ideally
Address for Correspondence: Dr. Yatin Mehta, Senior Consultant and Head,
Department of Anaesthesiology and Critical Care, Escorts Heart Institute
and Research Centre, Okhla Road, New Delhi, Phone : 011-26825000,
26825001, Fax: 011-51628442, 26825013. Email: yatinmehta@hotmail.com
suited to track the abrupt haemodynamic changes
that occur during off-pump coronary bypass
(OPCAB) Surgery. Continuous CO monitoring
devices using the pulse contour CO principle have
been described by Wesseling et al,7 which appears
to be accurate in measuring CO and in tracking its
changes. This method measures CO continuously
by measuring the area under the systolic portion
of the arterial pressure waveform. We studied the
results of measurements using the pulse contour
CO monitoring system and compared them with
the conventional pulmonary bolus thermodilution
method in cardiac surgery patients.
Meterials and Methods
After obtaining approval from the ethics
committee and written informed consent, we
studied 60 patients aged 40-75 years with normal
left ventricular and pulmonary functions
undergoing OPCAB surgery. Patients with

ACA-04-126- OA.p65 44 12/24/2005, 12:31 PM

[Downloaded free from http://www.annals.in on Thursday, October 27, 2016, IP: 115.178.239.182]
Annals of Cardiac Anaesthesia 2006; 9: 4448 Sujatha et al.
valvular heart disease, left ventricular ejection
fraction <30% and peripheral vascular disease were
excluded. Individual cardiac medications including
antihypertensive agents, nitroglycerin and beta
blockers were continued until the day of surgery.
Routine haemodynamic monitoring was
established, which included electrocardiography,
pulse oximetry, central venous and radial artery
catheterization. A 7.5F PAC (Baxter Health
Corporation, Edwards Critical care division, Irvine
CA, USA) was inserted into the right internal
jugular vein and a 4F arterial thermodilution
catheter (PiCCO, Pulsion Medical Systems,
Munich, Germany) was inserted via femoral artery
using Seldinger technique for clinical monitoring
of arterial pressure and continuous arterial
thermodilution CO measurements. All invasive
cannulations for monitoring were performed under
local anaesthesia. The arterial catheter was
connected to a computer for pulse contour analysis.
To calibrate this system, individual arterial
impedance to arterial pressure is calculated by
simultaneously determining the area under this
systolic portion of the arterial pulse wave and
measuring arterial thermodilution CO.
The mean of three subsequent measurements,
randomized within the respiratory cycle, was used.
They were performed by injection of 10 ml of iced
saline solution via a central venous catheter with
subsequent detection by the thermister embedded
in the wall of the arterial catheter. Thereafter a beat
to beat analysis of CO based on pulse pressure
waveform is obtained. Conventional pulmonary
bolus thermodilution CO was also measured and
compared with pulse contour CO values on 8
occasions. Before and after five minutes of
induction of anaesthesia(T1,T2), post sternotomy
(five minutes after the start of LIMA dissection, T3)
five minutes after commencement of left anterior
descending artery (T 4) left obtuse marginal/
diagonal, (T5) and right sided, (T6) (right coronary/
posterior descending artery) anastomoses, five
minutes after completion of protamine (T7) and post
sternal closure (T8).
Anaesthesia was induced with fentanyl citrate
5 g/Kg, midazolam 0.05 mg/Kg, titrated doses
of thiopentone sodium and endotracheal intubation
ACA-04-126- OA.p65 45
Thermodilution versus Pulse Contour Cardiac Output 45
was facilitated with vecuronium bromide 0.08 mg/
Kg. Anaesthesia was maintained with oxygen + air
+ isoflurane mixture, intermittent doses of fentanyl
1-2 g/Kg, midazolam 0.02 mg/Kg and
vecuronium bromide 0.04 mg/Kg. Heparin
2 mg/kg was administered prior to commencing
myocardial revascularisation to achieve an
activated clotting time of 250-300 seconds.
Myocardial revascularisation was performed with
the aid of octopus III (Medtronic Minneapolis,
USA) myocardial stabilizer. All patients were
studied during periods of haemodynamic stability
and without cardiac arrhythmias. Heparin was
neutralised with protamine at the completion of
revascularisation. Normothermia was maintained
in all patients intraoperatively with forced air
warmers, warm intravenous fluids and circulating
warm water mattress beneath the patient. On
completion of surgery, patients were shifted to
intensive care unit. Intraoperative intravenous fluid
infusion was individually adjusted for each patient
to maintain a cardiac index preferably > 2 L/min/
m2 with normal haemodynamic filling pressures
(central venous pressure 4-10 mm Hg and
pulmonary capillary wedge pressure 8-12 mm Hg)
and urine output 1-2 ml/Kg/hour. Haemoglobin
concentration of < 9 gm/dl and platelet < 60,000/
mm3 were the triggers for transfusion of packed
red cells and platelet concentrate respectively.
Statistical analysis was performed using analysis
of agreement to assure bias distribution of
differences between two methods by using Bland
and Altman analysis.8 Bias is defined as the average
difference between 2 measures. Precision is defined
as limits of agreement or the variation (standard
deviation) of the difference between 2
measurements. Additional descriptive analysis was
performed using SPSS 10.0 Chicago, USA software
package.
Table 1. Demographics
Age (years) 59.2 7.8
Male 45/60 (75%)
Number of distal anastomoses 3.4 0.2
BSA 1.74 0.16
Diabetes 20/60 (33.3%)
Hypertension 30/60 (50%)
BSA: body surface area
12/24/2005, 12:31 PM
[Downloaded free from http://www.annals.in on Thursday, October 27, 2016, IP: 115.178.239.182]

46 Sujatha et al. Thermodilution versus Pulse Contour Cardiac Output Annals of Cardiac Anaesthesia 2006; 9: 4448

Results error percentage (2SD/mean CO100) was 20%.

The mean error percentage at T2 and T8 were 22%
There were 15 females and 45 males with mean and 24% respectively. The error percentage during
age of 59 7 years, mean weight of 68 12 Kgs rest of time frames were more than 35%, which may
and height of 164 6 cms (table 1). Values are be considered above the levels of clinical acceptance
presented as mean standard deviation and p (table 3). The pattern observed in systemic vascular
value <0.05 was considered statistically significant. resistance measurements made with both methods,
Differences between methods during each time was the same as observed in CO measurements
points are given in table 2. There was no significant (table 4).
difference between CO values assessed by
intermittent bolus thermodilution and pulse Discussion
contour analysis of cardiac output at T1, T2 and T8
points but showed statistically significant Bolus intermittent thermodilution was chosen
difference during T3, T4,T5, T6, and T7 time points as the CO comparison standard because it is a
(p< 0.05). The mean difference between widespread clinical method of measuring CO.
thermodilution CO and pulse contour CO at Rapid changes in CO may not be detected with
preinduction (T 1) was 0.23 L/min (bias). The intermittent technique. Previous studies have
precision (2SD of bias) was 1L/min and the mean reported a small mean difference (bias) and
clinically acceptable limits of agreement (bias 2 x
Table 2. Analysis of comparison of different methods of SD).9,10 Transpulmonary thermodilution using
cardiac output (L/min) estimation pulse contour analysis has been shown to be an
Variable T1 T2 T3 T4 T5 T6 T7 T8
easy and reliable technique for assessing CO and
is increasingly being used for haemodynamic
TDCO 51 41 4.81 51 41 4.21 4.01 61.5
PICCO 5 0 41 3.41 41 31 3.51 3.51 5.81 assessment in the operating room and intensive
P NS NS <0.05 <.01 <0.05 <0.01 <0.01 NS care unit. 12 Our results comparing these two
T1- preinduction, T2- 5 minutes after induction, T3- 5 minutes after the start of left
internal mammory artery (LIMA) dissection T4- 5 minutes after commencement of
methods did not agree well under conditions
LIMA to left anterior descending artery anastomosis T5 5 minutes after where heart was displaced to reach the target site
commencement of lateral wall anastomosis, T6- 5 minutes after commencement of
right wall anastomosis, T7- 5 minutes after completion of protamine, T8- post-sternal of anastomoses.
closure
TDCO: Thermodilution cardice output, PICCO: pulsian continuous cardiac output
Here we have tried to evaluate the agreement
Table 3. Statistical analysis of comparison of between pulmonary bolus thermodilution CO and
thermodilution cardiac output and pulsian continuous pulse contour CO, which delivers beat to beat CO
cardiac output values leading to early detection of CO changes.
Time No.of CO (L/mt) Bias SD of L.O.A Mean error
Frame data pairs (L/mt) Bias (L/mt) (L/mt) %
The concept of arterial pulse contour analysis
1 60 4.940.97 0.23 0.5 1.0 20
2 60 4.070.86 0.29 0.45 0.9 22 for the determination of continuous CO has been
3 60 4.150.85 0.39 0.91 1.8 43
4 60 4.221.1 0.51 0.91 1.8 46 the subject of investigation for a number of years.
5 60 3.971.9 0.55 0.74 1.4 35
6 60 3.831.01 0.51 0.95 1.8 47
Basic algorithm for calculation of CO was put
7
8
60
60
5.11.4
5.1.3
1.04
.23
1.22
0.6
2.4
1.2
47
24
forward by Wesseling.7 According to this, stroke
CO: cardiac output; SD: standard deviation; LOA: limits of agreement (2SD), mean error % volume is computed by measuring the area under
LOA/mean CO100
systolic portion of arterial pressure waveform

Table 4. Comparison of systemic vascular resistance (dynes. sec. m2 cm-5) measured by 2 methods
Variable T1 T2 T3 T4 T5 T6 T7 T8

TDCO 1437254 1468255 1593374 1325356 1400360 1311328 1430270 1091317

PICCO 1550264 1638381 1872326 1598355 1766383 1690446 1080292 149 270

P NS NS <0.01 <0.01 <0.01 <0.01 <0.01 NS

T1- preinduction, T2- 5 minutes after induction, T3- 5 minutes after the start of LIMA dissection T4- 5 minutes after commencement of LIMA to LAD anastomosis T5
5 minutes after commencement of lateral wall anastomosis, T6- 5 minutes after commencement of right wall anastomosis, T7- 5 minutes after completion of
protamine, T8- post sternal closure
TDCO: thermodilution cardiac output, PICCO: pulsian continuous cardiac output

ACA-04-126- OA.p65 46 12/24/2005, 12:31 PM

[Downloaded free from http://www.annals.in on Thursday, October 27, 2016, IP: 115.178.239.182]
Annals of Cardiac Anaesthesia 2006; 9: 4448 Sujatha et al.
divided by aortic impedance. This value multiplied
by heart rate gives the pulse contour CO. To adjust
for aortic impedance, which differs from patient
to patient , a thermodilution measurement of CO
for calibration is performed. The new software
analyses the shape of entire pressure waveform in
addition to area under systolic portion of pressure
wave.11 It takes into account individual aortic
compliance and systemic vascular resistance based
on following consideration. During systole, more
blood is ejected from left ventricle than what
actually leaves aorta. During subsequent diastole
blood flows from aorta into arterial network at a
rate depending on aortic wall compliance, systemic
vascular resistance and blood pressure. Calibrating
procedure for pulse contour CO device necessitates
a transpulmonary measurement of CO. Ten ml of
isotonic saline at 7o to 9oC is injected into central
venous line, which is subsequently detected by a
thermistor embedded in the wall of the arterial
catheter. The mean of three consecutive
measurements randomized within the respiratory
cycle was used. CO is computed by multiplying
stroke volume by heart rate. A moving average of
the preceding 30 sec is calculated and displayed
on the monitor.
The difference in CO during left internal
mammary artery dissection could be due to the use
of sternal retractor which may cause compression
of subclavian artery or descending aorta leading
to a decrease in CO measured by pulse contour
analysis as compared to the normal value when
measured using thermodilution CO. The octopus
stabiliser used for optimal positioning during
References
1. Zhao X, Mashikian JS, Panzika P, et al. Comparison of
thermodilution bolus cardiac output and Doppler cardiac
output in the early post cardiopulmonary bypass period.
J Cardiothorac Vasc Anesth 2003; 17: 193-198
2. Wallace DC, Winslow EH. Effect of iced and room
temperature injectable on cardiac output measurements
in critically ill patients with low and high cardiac output.
Heart Lung 1993; 22: 55-631
3. Nishikawa T, Shuji Dohi. Errors in the measurement of
cardiac output by thermodilution. Can J Anaesth 1993; 40:
142-53
4. Connors AF, Seperoff T, Dawson NV, et al. The
ACA-04-126- OA.p65 47
Thermodilution versus Pulse Contour Cardiac Output 47
various anastomoses may be producing some sort
of descending thoracic aortic compression which
might be causing the discrepancy in the values. The
difference in the values were statistically significant
(p <0.05, table 2). The algorithm for computation
of CO by pulse contour CO includes three major
properties of arterial system compliance,
peripheral vascular resistance and impedance.
Peripheral vascular wave reflects interplay between
left ventricular output and capacitance of vascular
tree. Systemic vascular changes caused by
vasopressors and protamine can have more impact
on pulse contour CO than thermodilution bolus
CO. This also explains the discrepancy during
various anastomoses and post-protamine CO
measurements. During T1, T2, T8 where these factors
do not come into play, CO measurements showed
good agreement. Bland and altmann analysis
revealed a bias value between thermodilution CO
and pulse contour CO of 0.23 L/min, 0.29 L/min
and 0.23 L/min at T1, T2 and T8 time points, which
shows good agreement. Use of pulse contour CO
alone provides no information on pulmonary
arterial pressure, and mixed venous oxygen
saturation, which are useful variables in
management of OPCAB.12,13
Conclusion
We conclude that assesment of continuous CO
by pulse waveform analysis and by intermittent
thermodilution technique provided comparable
measurements during OPCAB surgery as long as
singnifcant changes in systemic vascular resistance
did not occur.
effectiveness of right heart catheterization in initial care
of the critically ill patients. JAMA 1996; 276: 889-897
5. Chernow B. Pulmonary artery flotation catheters. A
statement by the American College of Chest Physicians
and the American Thoracic Society. Chest 1997; 111: 261-
66
6. Yelderman M. Continuous cardiac output by
thermodilution. Int Anaesthesiology Clin 1993; 3: 127-140
7. Wesseling R, de Wit B, Ty Smith N. A simple device for
the continuous measurement of cardiac output. Adv
Cardiovas Phys 1983; 5: 16-52
8. Bland JM, Altman DG. Statistical methods for assessing
12/24/2005, 12:31 PM
[Downloaded free from http://www.annals.in on Thursday, October 27, 2016, IP: 115.178.239.182]
48 Sujatha et al. Thermodilution versus Pulse Contour Cardiac Output
agreement between two methods of clinical measurement.
Lancet 1986; 1: 307-310
9. Rodig G, Prasser C, Keyl C, et al. Continuous cardiac
output measurement pulse contour analysis vs
thermodilution technique in cardiac surgical patients. Br
J Anaesth 1999; 82: 525-30
10. Chassot PG, van derlinden P, Zaugg M, et al. Off Pump
coronary artery bypass surgery: physiology and
anaesthetic management. Br J Anaesth 2004; 92: 400-413
11. Goedje O, Hoeke K, Goetz AE, et al. Reliability of a new
algorithm for continuous cardiac output determination
ACA-04-126- OA.p65 48
Annals of Cardiac Anaesthesia 2006; 9: 4448
by pulse contour analysis during hemodynamic
instability. Crit Care Med 2002; 30: 52- 58
12. Sakka SG, Ruhl CC, Pfeiffer UJ, et al. Assessment of
cardiac preload and extravascular lung water by
transpulmonary thermodilution. Intensive Care Med 2000;
26: 180-187
13. Michard F, Perel A. Management of circulatory and
respiratory failure using less invasive hemodynamic
monitoring in Vincent JL (ed): Yearbook of intensive care
and emergency medicine, Springer, berlin, 2003, pp 508-
520
12/24/2005, 12:31 PM