Preoperative pulmonary function tests do not predict outcome after coronary artery bypass

To evaluate the utility of preoperative pulmonary function tests in predicting postoperative complications and lengths of stay after coronary artery bypass grafting.

Medical records of 193 consecutive patients who underwent coronary artery bypass grafting from October 1993 to September 1994 were reviewed. Preoperative pulmonary function tests, comorbid conditions, smoking history, postoperative complications, and total days in the intensive care unit, hospital, and on mechanical ventilation were abstracted. Data were analyzed using linear regressions, analyses of variance, and nonpaired Student's t tests.

Pulmonary function tests were normal in 56 subjects (29%, group 1), mildly impaired in 72 (37%, group 2), and moderately impaired in 35 (18%, group 3). Thirty patients (16%) had no pulmonary function tests. Group 3 subjects were older (71) compared to groups 1 and 2 (63 and 65, P < 0.05). There was no major difference in comorbid conditions or smoking status among the groups. All patients had atelectasis postoperatively. The most frequent postoperative complications were pleural effusions (43%), pulmonary edema or congestive heart failure (28%), and atrial fibrillation (35%). The repeat surgery rate was 3.6%. The mean length of hospital stay was 10.1 +/- 0.6 days, with 1.5 +/- 0.1 days of mechanical ventilation and 2.8 +/- 0.2 days of intensive care unit stay. Overall, pulmonary function tests had no predictive value for postoperative pulmonary and nonpulmonary complications, nor for durations of mechanical ventilation and intensive care unit stay. There was a trend toward increased length of hospital stay in patients with impaired pulmonary function tests (group 18.6 +/- 0.6, group 2 9.6 +/- 0.8, group 312.7 +/- 2.3 days, P = 0.09) but this was consistent with random variation.

Preoperative pulmonary function tests were not useful in predicting postoperative outcomes in patients undergoing coronary artery bypass grafting.