Objective: In coronary artery bypass grafting (CABG) without cardiopulmonary bypass (CPB) the inflammatory response is suggested to be minimized. Coronary anastomoses are performed during temporary coronary occlusion. Inflammatory response and myocardial ischaemia need to be studied in a randomized study comparing CABG in multivessel disease with versus without CPB. Methods: Following randomization 30 consecutive patients received CABG either with (n = 16) or without CPB (n = 14). Primary study endpoints were parameters of the inflammatory response (interleukin (IL)-6, interleukin-10, ICAM-1, P-selectin) and of myocardial injury (myoglobin, creatine kinase-MB (CK-MB), troponin I) (intraoperatively, 4, 8, 16, 24 and 48 h after surgery). The secondary endpoint was clinical outcome. Results: The incidence of major (death: CABG with CPB n = 1, not significant (n.s.)) and minor adverse events (wound infection: with CPB n = 2, without CPB n = I, n.s.; atrial fibrillation: with CPB n = 3, without CPB rt = 2, n.s.) was comparable between both groups. The release of IL-6 was comparable during 8 h of observation (n.s.). Immediately postoperatively IL-10 levels were higher in the operated group with CPB (211.7 +/- 181.9 ng/ml) than in operated patients without CPB (104.6 +/- 40.3 ng/ml, P = 0.0017). Thereafter no differences were found between both groups. A similar pattern of release was observed in serial measures of ICAM-1 and P-selectin, with no difference between both study groups (n.s.). Eight hours postoperativeIy the cumulative release of myoglobin was lower in operated patients without CPB (1829.7 +/- 1374.5 mu g/l) than in operated patients with CPB (4469.8 +/- 4525.7 mu g/l, P = 0.0152). Troponin I release was 300.7 +/- 470.5 mu g/l (48 h postoperatively) in patients without CPB and 552.9 +/- 527.8 mu g/l (P = 0.0213). CK-MB mass release was 323.5 +/- 221.2 mu g/l (24 h postoperatively) in operated patients without CPB and 1030.4 +/- 1410.3 mu g/l in operated patients with CPB (P = 0.0003). Conclusions: This prospective randomized study suggests that in low-risk patients the impact of surgical access on inflammatory response may mimic the influence of long cross-clamp and perfusion times on inflammatory response. Our findings indicate that multiregional warm ischaemia, caused by snaring of the diseased coronary artery, causes considerably less myocardial injury than global cold ischaemia induced by cardioplegic cardiac arrest. (C) 2000 Elsevier Science B.V. All rights reserved.
