Conversion in off-pump coronary artery bypass grafting: an analysis of predictors and outcomes.

BACKGROUND: The incidence, predictive factors, and outcomes related to conversion from off-pump coronary artery bypass (OPCAB) to on-pump coronary artery bypass grafting (ONCAB) have not been well defined. We sought to determine the incidence of conversion, predictive factors, and any associated adverse consequences. METHODS: From January 2000 through June 2002, 1,644 patients underwent nonemergent OPCAB with 61 patients requiring conversion from OPCAB to ONCAB. These groups were retrospectively compared by univariate and multivariate regression analysis. The converted group was then computer matched 1:3, to a cohort of ONCAB patients to determine differences in outcomes. RESULTS: The overall conversion rate was 3.71%. Converted patients compared with a computer-matched ONCAB patients had a higher incidence of operative mortality (18.0% versus 2.7%, p < 0.001). Urgently converted patients had a higher incidence of postoperative cardiac arrest (25% versus 1.1%, p < 0.001), multisystem organ failure (10.7% versus 0.6%, p < 0.001), vascular complications (7.1% versus 1.1%, p = 0.03), and perioperative myocardial infarction (10.7% versus 1.1%, p = 0.02). Predictive factors for conversion were surgeon early in OPCAB experience (odds ratio [OR] 4.4), previous CABG (OR 2.8), and congestive heart failure (OR 2.0). The need for urgent-emergent conversion was highly predictive for operative mortality (OR 7.3) compared with elective conversion. CONCLUSIONS: Patients undergoing urgent-emergent but not elective conversion from OPCAB to ONCAB had a significantly higher risk of mortality and morbidity compared with patients whose procedure was initially ONCAB. Variables predictive of conversion included previous CABG, congestive heart failure, and surgeons early in OPCAB experience.

On-Pump Beating Heart Surgery Offers an Alternative for Unstable Patients Undergoing Coronary Artery Bypass Grafting.

Abstract Background: Cardiac surgery has expanded the available approaches to aortocoronary artery bypass grafting to include approaches from minimally invasive surgery to full sternotomy. The heart can be arrested, left beating, or assisted with a right ventricular assist device or cardiopulmonary bypass pump. We have examined the 4 surgical modes that we use routinely in our large multisurgeon practice to determine our selection biases and the outcomes of the different techniques. Methods: Of the 4733 coronary artery bypass grafting (CABG) patients we studied from January 2000 through December 2002, 2332 (49.3%) operations were done on-pump on the arrested heart, 1908 (40.3%) were performed off-pump, 364 (7.7%) were performed on-pump on the beating heart, and 129 (2.7%) were performed with right heart assist. The preoperative risk factors, operative variables, and postoperative outcomes of the groups were analyzed. Results: Patients selected for on-pump beating heart procedures tended to be sicker with the highest predicted risk of death. We also selected patients who were in cardiogenic shock, in resuscitation, in emergent or salvage status, on dialysis, and with preoperative intra-aortic balloon pump (IABP) use for on-pump beating heart procedures at higher than expected rates. Patients with renal failure with or without dialysis, and those having a history of cerebrovascular accident tended not to be chosen for on-pump arrested heart procedures. Off-pump beating heart procedures were avoided for patients with cardiogenic shock or resuscitation, in emergent or salvage status, and with preoperative IABP use. The mortality rate in these patients was slightly worse in the on-pump beating heart group (4.4%) than in the on-pump arrested heart (3.5%) and off-pump (2.3%) groups (analysis of variance [ANOVA], P =.04). Atrial fibrillation occurred more frequently in both the on-pump beating heart (20.1%) and on-pump arrested heart (23.8%) groups (ANOVA, P <.001). The on-pump groups had higher rates of blood product use and reoperation for bleeding and a prolonged ventilation rate, compared with the other procedures. On-pump patients had a statistically longer length of stay than either off-pump or right heart-assisted patients ( P <.05) and required longer times on the ventilator and in the intensive care unit. Conclusions: Normothermic cardiopulmonary bypass with a beating heart is safe and efficacious and may be the method of choice for patients in cardiogenic shock, requiring resuscitation, or with previous CABG surgery, recent myocardial infarction, a low ejection fraction, or unstable arrhythmias.