Pii: S1010-7940(00)00438-3

Background: The quantitative contribution of cardiopulmonary bypass (CPB) to respiratory dysfunction after cardiac surgery is not documented and the effect of the use of bilateral internal mammary artery (IMA) grafts is not clear. Methods: One hundred and seventy-®ve patients undergoing CABG with (CPB, n ˆ 150) and without (NOCPB, n ˆ 25) CPB were studied. PMN elastase (as a marker of the systemic in ̄ammatory response) and serial arterial oxygen (paO2) and carbon dioxide (paCO2) tension, alveolar arterial oxygen (AaO2) gradient and percent saturation were measured. The CPB group was subdivided into three groups by the number of IMA grafts used: 0IMA (n ˆ 12), 1IMA (n ˆ 82) and 2IMA (n ˆ 51). Results: The NOCPB group was younger, had signi®cantly better preoperative blood gases, received fewer grafts and had lower PMN elastase levels than the CPB group. In both groups maximum respiratory dysfunction occurred at 48 h (paO2, percentage saturation and Aa gradient all P , 0:001 versus baseline) with partial recovery by 5 days. The percentage decline and subsequent recovery in all blood gas parameters was near identical in the CPB and NOCPB groups. Amongst the three IMA groups the percentage changes in all blood gas parameters were similar, as was the duration of postoperative ventilation and time to discharge. There was no correlation between blood gas parameters at 48 h with age, CPB time, blood loss, duration of ventilation or peak PMN elastase level. Conclusions: Changes in postoperative gas exchange are similar in patients undergoing CABG with and without CPB even although PMN elastase levels indicate that CPB produces a more marked in ̄ammatory response. The use of 2IMA compared with 1IMA does not increase respiratory dysfunction. q 2000 Elsevier Science B.V. All rights reserved.