Off-Pump vs On-Pump CABG

Abstract & Commentary

By Michael H. Crawford, MD

Source: Hannan EL, et al. Off-pump versus on-pump coronary artery bypass graft surgery: Differences in short-term outcomes and in long-term mortality and need for subsequent revascularization. Circulation. 2007;116:1145-1152.

The effectiveness of off-pump coronary artery bypass graft surgery (OPCAB) is controversial. Thus, Hannan and colleagues from New York evaluated the short-term and long-term results of OPCAB with sternotomy vs on-pump CABG in New York State between 2001-2004. The data were derived from the New York State's Cardiac Surgery Reporting System (CSRS) which encompassed all non-federal hospitals certified to do cardiac surgery. New York's vital statistics were used to ascertain deaths, and the CSRS was used to identify repeat CABG. The companion registry for percutaneous coronary procedures was used to assess whether any CABG patient had subsequent percutaneous revascularization. The study population included 49,830 patients; 13,889 had OPCABG. Selection bias was reduced by an analysis of 25,530 patients who were in matched pairs receiving OPCAB and CABG.

Results: OPCAB patients were older, more likely to be women, more likely to have reduced left ventricular performance, as well as significant co-morbidities. OPCAB had a lower 30-day mortality (OR 0.81, CI .68-.97) and less perioperative complications. However, in the matched pairs, 3-year mortality was not different (10%), and OPCAB patients were more likely to have subsequent revascularization (10% vs 6%, P < .001). Less than 2% of patients were converted from OPCAB to CABG. No subgroup showed a mortality difference, but patients with circumflex disease, a calcified aorta, and renal failure were more likely to have subsequent revascularization after OPCAB. Hannan et al concluded that OPCAB has a lower 30-day morbidity and mortality, but is associated with more subsequent revascularization vs standard CABG.


OPCAB, with sternal incision, was heralded as a way to avoid some of the complication of CABG such as "pump head", stroke from aortic cross clamp debris, depressed left ventricular function from fibrillation arrest and atrial fibrillation induced by damage from atrial cannulation. Indeed previous studies have shown lower early morbidity and mortality with OPCAB. However, concerns have arisen about the completeness of revascularization, long-term patency and need for subsequent revascularization. This large observational study has confirmed all the above. Short-term complications and mortality rates are less with OPCAB, but long-term mortality is no different, and freedom from subsequent revascularization procedures, is higher with standard CABG.

Since this is a large database study, there is a paucity of more mechanistic data. For example, we cannot ascertain whether the increased frequency of subsequent revascularization with OPCAB is due to incomplete revascularization or reduced graft patency over time. There may be selection biases in non-randomized trails. One issue here is the application of OPCAB. Overall, 28% of CABG in New York was OPCAB during the 3 years of this study, but the rates per surgeon varied considerably (rarely to >50%). The investigators tested selection bias by examining a matched subgroup, and found similar results to the entire database. Strengths of this observational study include its very large size, and it includes everyone, so higher risk patients, often excluded from randomized trials, are included.

One thing the study does not address directly is who should be considered for OPCAB. Since most of the morbidity reduction was in stroke and respiratory failure, elderly females with a higher stroke risk, and those with preexisting pulmonary disease, may be good candidates for OPCAB. Also, those with calcified ascending aortas may do better with OPCAB. Those with complex coronary lesions, severe multivessel disease, calcified coronary arteries, and diabetes may be better candidates for standard CABG. Finally, the study does not address the issue of minimally-invasive CABG, since all the patients had sternotomy. Clearly, mini CABG should reap the benefits of being off-pump, but does the limited visibility and maneuverability effect the long term results?