Secondary Logo

Journal Logo

Original Papers

Factors determining end-expiratory alveolar pressure after cardiac surgery

Fletcher, R.; Foster, P.; Thornton, S.

Author Information
European Journal of Anaesthesiology: June 1999 - Volume 16 - Issue 6 - p 396-399



An intrinsic positive end-expiratory pressure (iPEEP) during coronary artery bypass grafting (CABG) causes problems. The increase in end-expiratory alveolar pressure causes raised peak and mean airway pressures. Incomplete emptying of the lung causes it to bulge, threatening internal mammary artery grafts and interfering with wiring of the sternum. Cardiac output may be depressed, particularly when the sternum is closed. However, most ventilators on anaesthetic machines are not capable of creating the end-expiratory no-flow period necessary to measure iPEEP, and the problem may therefore remain unrecognised.

The incidence of iPEEP, and its variation with sternal opening and closure, was documented in 38 patients undergoing CABG in a Swedish hospital [1]. Median iPEEP before surgery was 2.0 (range 0.5-13) cm H2O. Intrinsic positive end-expiratory pressure was virtually abolished by sternotomy, when the median was 1 (0-2) cm H2O. Following cardiopulmonary bypass (CPB), median iPEEP remained clinically unimportant at 1 (0-8) cm H2O, but when the sternum was closed it increased significantly to 5 (3.5-22) cm H2O. It was shown that iPEEP was most likely to occur in patients with obstructive spirometry and an increased body mass index (BMI). Three interlinked mechanisms were thought to be responsible. First, the accumulation of lung water following CPB [2] causes an increased airway resistance. Second, lung volume is reduced after sternal closure [3]. Third, the relation between expiratory resistance and lung volume is nonlinear [4], which can lead to large increases in expiratory resistance when lung volume is reduced by sternal closure; end-expiratory resistances in excess of 100 cm H2O I−1 s−1 were seen in some patients.

This paper reports two new studies. Study I was planned to document the duration of raised iPEEP into the post-operative period. Once patients with iPEEP >5 cm H2O after sternal closure had been identified, measurements were to continue into the post-operative period until iPEEP was less than 5 cm H2O. Patients were recruited into this study and simultaneously into another which required ventilation at 10 breaths min−1. Thus patients in study I were ventilated at this rate. In the event, iPEEP exceeded 5 cm H2O in only two patients, and in only one did it persist into the post-operative period. It was thought that the lesser ventilatory rate and greater tidal volumes used in study I might be responsible for the low incidence of iPEEP. Study II was therefore planned, with randomization to ventilatory rates of 10 or 20 breaths min−1.


The same anaesthetist (RF) administered all the anaesthetics. Successive patients for elective CAGB were recruited whenever one of the other authors was also able to be present. Patients with clinical heart failure were excluded. Informed consent, pre-operative spirometry and a smoking history (smoker, non-smoker, ex-smoker; 'pack years') were obtained. Premedication was with lorazepam 2-3.5 mg. Anaesthesia was induced with fentanyl 1.7-3.0 mg followed by pancuronium. Ventilation was with 50% O2, 50% N2 with 0.5-1% enflurane. Tidal volume was adjusted to give an end-tidal PCO2 of 3.5-4 kPa. The ventilator (Servo 900C; Siemens-Elema, Stockholm) was set to give inspiratory and pause times 25 and 10% of the cycle. Intrinsic positive end-expiratory pressure was measured by holding down the 'expiratory pause hold' button on the ventilator until the indicated airway pressure was constant. The first measurement was made before the start of surgery, and the second immediately after sternal wiring was complete. During CPB, the lungs were allowed to deflate and were not ventilated again until shortly before cessation of CPB.

Flow and pressure signals from the ventilator were recorded on a Gould recorder. End-expiratory resistance (Rendexp), static compliance (Crs) and their product, an 'end-expiratory time constant', intended to reflect conditions at end-expiration, were calculated from the recordings. Static compliance was calculated from tidal volume divided by end-inspiratory pause pressure minus iPEEP. End-expiratory resistance was calculated from end-expiratory flow divided by iPEEP.

Statistical analysis

We used multiple linear regression to identify factors associated with raised iPEEP, the Mann-Whitney U-test for between-group comparisons, and the Wilcoxon test for within-group comparisons. A probability of less than 0.05 was accepted as indicating significance. Results are expressed as median (range).

Study I

Twenty-four patients (21 male) undergoing CABG, were ventilated at 10 breaths min−1. Their median age was 62 years (38-73), height 171 cm (152-183) and weight 81 (57-114) kg.

Study II

Twenty-two patients undergoing CABG (16 male) were randomized to ventilation rates of either 10 or 20 breaths min−1. In all other respects, study II was identical to study I. The median age was 61 years (45-76), height 172 (143-190) cm and weight 75.0 kg (51-110).


Study (I), ventilation at 10 breaths min−1

Median iPEEP was 0.0 cm H2O (0.0-8.0) before surgery and increased (P<0.05) to 1.0 (0.0-7.0) cm H2O after surgery. Two of the 24 patients developed iPEEP>5 cm H2O after sternal closure.

Study (II), randomized to ventilation at 10 or 20 breaths min−1

One obese patient (randomized to 20 breaths min−1) became hypoxic after CPB and the protocol was abandoned. In the remaining 21 patients, there were no significant differences in age, weight, height or spirometry between the two groups (Table 1). In the 11 patients ventilated at 10 breaths min−1, median iPEEP increased from 0.5 cm H2O (0.0-1.5) before surgery to 1.0 (0.0-6.5) cm H2O after surgery (P<0.05). In the 10 patients ventilated at 20 breaths min−1, median iPEEP was 1.0 cm H2O (0.5-5.0) before surgery and 3.0 (0.0-7.5) cm H2O after surgery, not a significant increase. This post-sternal closure value at 20 breaths min−1 was not significantly different from that at 10 breaths min−1. Three patients developed iPEEP in excess of 5 cm H2O after surgery; two ventilated at 10 breaths min−1 and one at 20 breaths min−1.

Table 1
Table 1:
Comparison of the two groups of patients in study II: medians (range). There were four females ventilated at 10 breaths min−1 and two at 20 breaths min−1

Lung mechanics

Table 2 shows lung mechanics for the two groups before and after surgery. Rendexp and the expiratory time constant were both significantly increased after surgery compared with before (P<0.05). However, ventilatory rate had no effect upon any variable.

Table 2
Table 2:
Lung mechanics (medians and ranges) at the two ventilatory rates. End-expiratory resistance and the end-expiratory time constant were increased after surgery (P<0.05). There were however, no significant differences in any variable between the two ventilatory rates

Pre-operative determinants of raised iPEEP before surgery. The data from studies I and II were pooled. With iPEEP before surgery as the dependent variable, independent variables were entered into multiple linear regressions: height and weight (or BMI), age, sex, smoking status, 'pack years', ventilatory rate, and the spirometry variables VC and VC as percentage of predicted, plus FEV1 and FEV1 as percentage of predicted (FEV1%). The following were significant: FEV1% (P<0.05), BMI or weight (P<0.05); rate (P<0.05). R-square was 0.28.

Pre-operative determinants of raised iPEEP after surgery

Intrinsic positive end-expiratory pressure after surgery was correlated to iPEEP before surgery (r2 = 0.38, P<0.001). With iPEEP after surgery as the dependent variable, and ignoring iPEEP before surgery, the variables were again entered into multiple linear regression. As before, FEV1% was the only significant spirometric variable, and the others were discarded. Sex, smoking history and pack years were not significant in any combination, and were also discarded. Table 3 shows that BMI (or height and weight), age, FEV1% and ventilatory rate were all determinants of iPEEP after surgery.

Table 3
Table 3:
Results of entering various combinations of variables into multiple linear regression, where iPEEP after sternal closure is the dependent variable

Comparison with Swedish results

Patients from Manchester ventilated at 20 breaths min−1 and all the Swedish patients were pooled and multiple regressions run as above with the inclusion of a new variable, for country. In combination with FEV1% (P<0.005), age and BMI (P<0.05) this variable proved not to be significant (P=0.9).


Following a previous investigation [1], study I was designed to measure the duration of raised iPEEP after cardiac surgery. As before [1], iPEEP increased after sternal closure, but clinically significant levels were seen in only two patients. As study I and the previous investigation were performed at different ventilatory rates, study II was designed to test the hypothesis that ventilatory rate affects the development of iPEEP. It confirmed that increased iPEEP after surgery was due to increased expiratory resistance, but that ventilatory rate did not affect resistance. Further, the Mann-Whitney U-test showed no significant differences in iPEEP after sternal closure between patients ventilated at 10 breaths min−1 and those ventilated at 20 breaths min−1.

Because the protocols of studies I and II were identical apart from choice of ventilatory rate, and the patients were recruited from the same waiting list, the data were pooled and analysed by multiple linear regression. Ventilatory rate now proved to be a significant determinant, as did age, BMI and pre-operative spirometry. We therefore assume that the non-significance of ventilatory rate in comparative tests is due to its effect being obscured by the other variables. Perhaps surprisingly, no smoking history variable was correlated with iPEEP.

The greater iPEEP at 20 breaths min−1 is thus simply due to decreased expiratory time; there was no measurable effect of rate on resistance, as has been observed in animals [5]. Finally, to answer the question of whether there is a 'Swedish effect' on iPEEP, brought about by a different anaesthetic or CPB management, would ideally require some patients to have been ventilated at 10 breaths min−1 in the Swedish study. But pooling Manchester patients who were ventilated at 20 breaths min−1 with all the Swedish patients, no effect of country is discernible, whereas spirometry, age and BMI remain important determinants of iPEEP.

We conclude that ventilatory rate, although not affecting compliance or resistance, has a discernible effect on iPEEP after cardiac surgery, as do age, BMI and pre-existing airway obstruction. Intrinsic positive end-expiratory pressure should be suspected in cardiac surgery whenever airway pressures are raised, particularly after sternal closure, or whenever the lung bulges into the sternotomy wound. However, the possible effects of perfusion technique and lung management during CPB on lung water and therefore expiratory obstruction, have not been explored.


1 Fletcher R. Raised end-expiratory pressures during cardiac surgery. Br J Anaesth 1994; 72: 629-632.
2 Boldt J, King D, Scheld HH, Hempelmann G. Lung management during cardiopulmonary bypass: influence on extravascular lung water. J Cardiothorac Vasc Anesth 1990; 4: 73-79.
3 Jonmarker C, Nordström L, Werner O. Changes in functional residual capacity during cardiac surgery. Br J Anaesth 1986; 58: 428-432.
4 Bouhuys A, Jonson B. Alveolar pressure, airflow rate, and lung inflation in man. J Appl Physiol 1967; 68: 1086-1100.
5 Svantesson C, John J, Taskar V, Evander E, Jonson B. Respiratory mechanics in rabbits ventilated with different tidal volumes. Resp Physiol 1996; 106: 307-316.

VENTILATION, airway pressure, auto-PEEP, intrinsic-PEEP; SURGERY, cardiac

© 1999 European Society of Anaesthesiology