[Gas exchange changes in the postoperative period of heart surgery]

We studied prospectively 59 open-heart surgical patients (CBPS) in order to evaluate postoperative arterial blood gases evolution and its predictive value of respiratory and non-respiratory post-surgical complications. Twenty-four hours after CPBS 28 over 59 patients showed left pleural effusion and/or left lower lobe atelectasis. 62.9% or pleural effusions were only blunted costophrenic angle. Chest x-ray film were normal in 38.9% (23/59) of patients. Forty-eight hours after CPBS only 5% (3/59) radiographs were normal and only 31% of pleural effusion were classified as minimal. Forty-eight hour radiographs worsened in 69.4% of the patients. (Table 1). During 48 hours period 71% of patients showed pleural effusion and 42% atelectasis. Only 1 patient showed an atelectasis up a third of hemithorax (3.5% of abnormal chest X-ray films). Twelve hours alveolo-arterial quotient (a/A) was decreased in 50/59 patients (0.51 +/- 0.16), more deeply at the second day. There was no relationship between CBP time and a/A at 12 or 48 hours. The normal chest X-ray film patients mean a/A was no different (0.54 +/- 0.17). The a/A at 48 hours was no different between patients with and without lower lobe atelectasis. Nine patients (15%) developed respiratory complications (RC) and 11 (19%) non-respiratory complications (NRC) (Table 2). There was no difference in CBP time (76.9 +/- 27.9 vs 88.1 +/- 27.7 min p = NS) nor aortic cross-clamp time (52.61 +/- 20.43 vs 59.57 +/- 19.39 min p = NS) between patients with and without RC. There were no differences in a/A at 12 hours (0.47, 0.51, 0.48 p = NS) and 48 hours (0.34, 0.32, 0.30 p = NS) between patients without complications, with RC and with NRC (Table 3). There was no correlation between 12 or 48 hours a/A and intensive care or hospital stay length. The absence of predictive value of hypoxemia could be explained from a different source of early a/A fall and important RC. It could mean that RC after CPBS are not specific of that sort of surgery nor involve mechanisms related to that special intraoperative circumstances, which is not the same for gas exchange alterations. We conclude that a/A deterioration is a very common finding after CBP and does not identify particularly risky patients.