Secondary Logo

Journal Logo


Continuous cardiac output monitoring during haemodynamic changes in off-pump coronary artery bypass grafting surgery

Auler, J. O. C. Jr.1; Kim, S. M.1; Carmona, M. J. C.1; Malbouisson, L. M. S.1; de Oliveira, S. A.2

Author Information
European Journal of Anaesthesiology: October 2006 - Volume 23 - Issue 10 - p 890-892
doi: 10.1017/S0265021506211372
  • Free


A large number of manoeuvres and devices have been developed so that off-pump coronary artery bypass grafting (CABG) may be safely and reliably performed. The evolving surgical technique must be accompanied by adequacy of anaesthesia and intraoperative monitoring because haemodynamic instability may occur. Manipulation and rotation of the heart, application of stabilizing devices or myocardial ischaemia during temporary coronary flow occlusion may explain cardiac output deterioration. It is important to continuously monitor cardiac output to track and evaluate haemodynamic changes during off-pump CABG [1].

Yelderman [2] introduced in 1990 a method that used heat signals and changes in blood temperature to estimate cardiac output, and pulmonary catheters with a thermal filament were developed. The purpose of this study was to compare this method of continuous cardiac output measurement with that using intermittent bolus injection, in patients undergoing coronary artery bypasses on lateral, anterior and posterior ventricular walls during off-pump CABG.

The Ethics Committee of our institution approved this protocol, and it was carried out between September 2004 and June 2005. We enrolled 27 patients (18 male and 9 female) with coronary artery disease undergoing off-pump CABG. Their ages ranged from 48 to 79 yr (mean 66.3 ± 8.7) and mean body surface area 1.73 ± 0.18 m2, with minimal (n = 8), low (n = 13) or moderate (n = 6) surgical risk according to Higgins [3]. Those with unstable angina, poor ejection fraction, intracardiac shunt or valvular heart disease were not included. Two patients had single-vessel disease, 16 had double-vessel disease and 7 had triple-vessel disease.

After general anaesthesia was obtained, a 7.5-Fr pulmonary artery catheter with thermal filament (Edwards CCOmbo V catheter; Baxter Edwards Critical Care, Irvine, CA, USA) was inserted through the internal jugular vein and connected to the Vigilance monitor (Baxter Edwards Critical Care, Irvine, CA, USA). In order to optimize blood volume, all patients received 10 mL kg−1 of hydroxyethyl starch (Voluven, Fresenius Kabi, Germany) and crystalloid infusion with lactated Ringer's solution. The heart was exposed through a median sternotomy and suspended in the pericardial cradle. A single heavy suture was placed in the oblique sinus of the pericardium, and different degrees of traction on the suture facilitated exposure of all coronary branches [4].

Cardiac index measurements were made after heart positioning and establishment of the ‘Octopus-3’ suction stabilizer system (Medtronic Inc., Minneapolis, MN, USA) and after approximately 5 min, usually at the end of coronary distal anastomosis, before removing the Octopus-3 system. These data were corrected to baseline which was obtained before each procedure, when volaemic and haemodynamic adjustments were made. The measurement times were repeated for each coronary anastomosis in three settings: anterior wall, for anastomosis to left anterior descending and diagonal arteries, lateral wall, for marginal and circumflex arteries, and posterior wall, for right coronary and posterior descending arteries.

Continuous cardiac index was displayed on the Vigilance monitor. Triple bolus measurements of cardiac index were also performed, using dextrose 5% 10 mL at room temperature (<21°C). The mean value was recorded as intermittent cardiac index.

Accuracy of continuous cardiac index in comparison to intermittent measurements was analysed with the method described by Bland and Altman [5]. Bias and limits of agreement (bias ± 2 SD) between intermittent and continuous methods were measured and t-test used to examine any differences.

None of the patients had emergency surgery or reoperation. In the preoperative interview, 44.4% had diabetes and were under medication and 85.2% had hypertension. Preoperative beta-blocker was given to 70.4% and nitrate to 77.8% of the patients. Sixty-four distal coronary anastomosis (2.3 procedures per patient) were performed: 26 anastomosis to the left anterior descending, 15 to marginal, 3 to circumflex, 8 to diagonal and 12 to the right coronary artery.

The overall mean cardiac output was 2.97 ± 0.62 with continuous measurements compared with 2.40 ± 0.69 L min−1 m−2 with the intermittent technique. Bias, 95% confidence interval (CI), limits of agreement (bias ± 2 SD) and significance for differences are shown in Table 1. Using all data, we found that the mean difference was −0.623, with 95% CI −0.726 to −0.520. Thus, the continuous technique tends to be greater by between −2.076 and 0.830, differences which are unacceptable for clinical purposes. Separating data into three uniform settings and further into three distinct time points, we observed that the greatest differences occurred on the initial and final moments of lateral (bias −1.100 and 95% CIs [−1.449, −0.751] and −0.972 [−1.266, −0.678], respectively) and posterior walls (−0.933 [−1.220, −0.647] and −0.472 [−1.298, −0.585]).

Table 1
Table 1:
Comparison of continuous and intermittent cardiac index. Results of Bland–Altman analysis and paired t-test.

The main results of this study indicate that continuous cardiac output obtained from pulmonary artery catheters with thermal filament did not correlate with the intermittent bolus cardiac index during major haemodynamic changes that occur in off-pump cardiac surgery during exposure of different ventricular walls. In those situations where surgical manipulation of the heart resulted in less pronounced circulatory deterioration, as in anterior wall anastomosis, continuous cardiac index alterations were more accurate.

We measured the variables at two important time points: immediately after surgical positioning and after 5min. The first measurement identified the initial haemodynamic instability, and the second one could describe a rather stabilization of haemodynamic parameters in the given position.

Three settings were defined to group different anastomotic sites: lateral, anterior and posterior. Nevertheless, procedures within each setting differed greatly. The site of placement of the tissue stabilizer should have been considered, whether closer to ventricular base or apex, and considering how much mobilization of the heart was needed. For instance, some procedures to diagonal artery required slight lateralization of the heart, as opposed to maintenance of neutral position in anastomosis of left anterior descending artery. These two situations were classified as anterior setting.

The aim of this study was to verify whether continuous cardiac output monitoring might be useful to guide intervention due to major haemodynamic variations. Based on our results, we feel that this monitoring method should not be solely relied on during off-pump CABG.

J. O. C. Auler Jr.

S. M. Kim

M. J. C. Carmona

L. M. S. Malbouisson

S. A. de Oliveira

1Anesthesia and Surgical Critical Care Department, Heart Institute (InCor), University of Sao Paulo Medical School, Brazil

2Cardiothoracic Surgical Department, Heart Institute (InCor), University of Sao Paulo Medical School, Brazil


1. Couture P, Denault A, Limoges P et al. Mechanisms of hemodynamic changes during off-pump coronary artery bypass surgery. Can J Anaesth 2002; 49: 835–849.
2. Yelderman M. Continuous measurement of cardiac output with the use of stochastic system identification techniques. J Clin Monit 1990; 6: 322–332.
3. Higgins TL, Estafanous FG, Loop FD et al. Stratification of morbidity and mortality outcome by preoperative risk factors in coronary artery bypass patients. A clinical severity score. JAMA 1992; 267: 2344–2348.
4. D'Ancona G, Karamanoukian H, Lima R et al. Hemodynamic effects of elevation and stabilization of the heart during off-pump coronary surgery. J Card Surg 2000; 15: 385–391.
5. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–310.
© 2006 European Society of Anaesthesiology