Functional renal outcome in on-pump and off-pump coronary revascularization: a propensity-based analysis.

Background: Since the advent of off-pump coronary artery bypass grafting (OPCAB), comparisons have been made between it and on-pump coronary artery bypass grafting. Some observe a lesser incidence of acute renal failure requiring renal replacement therapy with OPCAB whereas others do not. The objective was to compare the occurrence of renal adverse outcome between on-pump coronary artery bypass grafting and OPCAB. Renal adverse outcome was defined as minor (20% increase in serum creatinine from preoperative) or major (composite end point of postoperative serum creatinine >200 mumol/L or postoperative mechanical renal support).

Methods: The study was based on 2,041 patients with no known preoperative renal disease having first-time isolated coronary artery bypass grafting of multiple coronary arteries between January 2001 and November 2003, at St. Mary's Hospital, Harefield Hospital, and Hammersmith Hospital, in West London; 1,224 patients had on-pump coronary artery bypass grafting and 817 patients had OPCAB. Selection bias for surgical technique was addressed by calculating the propensity score for each patient and using it as an independent variable for adjustment in the multivariate analysis. Univariate and multivariate ordered logistic regressions were used to identify factors associated with renal adverse outcome ordered as none, minor, and major.

Results: The number of grafts was 3.22 +/- 0.82 for the on-pump coronary artery bypass grafting group and 3.35 +/- 0.95 for the OPCAB group. On-pump coronary artery bypass grafting and increasing age were found to be the strongest independent predictors (p < 0.001) of renal adverse outcome. Other independent predictors included hypertension (p = 0.005), diabetes (p = 0.032), and preoperative serum creatinine (p = 0.001). A left ventricular ejection fraction of 0.30 to 0.49 (p = 0.099) and an ejection fraction of 0.50 or greater (p < 0.001) were associated with decreased risk compared with patients with an ejection fraction of less than 0.30. Interestingly, the use of non-left internal mammary arterial conduits significantly decreased the likelihood of renal adverse outcome (p = 0.034).

Conclusions: The results of this propensity-based study show that the OPCAB technique may reduce the risk for minor and major renal adverse outcome after coronary artery bypass grafting.