Concomitant aortic valve replacement and CABG surgery for mild, asymptomatic aortic stenosis

The optimal management of patients with mild, asymptomatic aortic stenosis that requires coronary artery bypass graft (CABG) surgery remains controversial. Concomitant aortic valve replacement (AVR) at the time of CABG surgery obviates the need for reoperative surgery with its increased technical risks for damage to the left internal mammary artery and patent vein grafts. It may also prevent the progression of aortic stenosis, which can result in hypertrophy, decreased ventricular function, and ultimately congestive heart failure. However, AVR at the time of CABG surgery increases morbidity and mortality and is unlikely to result in improved symptomatology. The patient is exposed to an increased risk of endocarditis and bleeding if warfarin (Coumadin) is necessary. In younger patients, delaying an AVR may allow a choice between a mechanical or bioprosthetic valve, thus avoiding lifelong warfarin therapy. Finally, a reoperation may still be necessary in any CABG patient because of progression of native disease or graft failure.

The current study by Smith and Peterson (page 31) uses the Society of Thoracic Surgeons National Database and Markov decision analysis in an attempt to determine whether CABG patients with mild or moderate aortic stenosis should undergo concomitant AVR.1 Their results indicate that CABG and AVR should be performed together in those asymptomatic patients whose resting peak aortic gradient exceeds 30 mm Hg. The addition of an AVR to CABG surgery increased the risk of stroke and renal failure. Finally, the operative mortality for an AVR following a CABG was higher than for either an initial CABG or CABG and AVR.

There are several limitations to this study. The population was assumed to be free of non-cardiac life-threatening morbidity and it was further assumed that all patients would receive a mechanical prosthesis and could take warfarin. The possibility of prosthetic valve endocarditis was not included in the model, nor was the possibility for a reoperative CABG due to ischemia from progression of native disease or graft failure. Serial echocardiograms were not always available, so that the rate of progression of aortic stenosis could not be accurately determined. Most studies suggest that the rate of progression of peak gradients is 5 to 10 mm Hg per year and aortic valve area is 0.8 to 1.0 cm2 per year.2 However, it may be difficult to predict for individual patients. The rate of progression may be increased in patients with serum cholesterol greater than 200 mg/dL,3 and in patients with bicuspid, rheumatic, and calcified valves.4 Recent studies have shown that calcified aortic leaflets may have an increased inflammatory response similar to the atherosclerotic process5 and that HMG-CoA reductase inhibitors (statins) may retard the progression of increased valvular gradients and decreased aortic valve area.6,7

Previous studies by Fiore and Odell document the findings of Smith and his coworkers, showing that subsequent AVR following a CABG is associated with increased morbidity and mortality.8,9 Patients are older and usually have progression of native and graft disease requiring concomitant coronary revas-cularization. Cross-clamp and cardiopulmonary bypass times are longer and it is more difficult to achieve good myocardial protection. There is more chance of embolization of atherosclerotic debris resulting in strokes and myocardial infarctions. Hoff and coworkers had no operative deaths in their series of AVR following a prior CABG.10 However, many of these procedures were done under elective circumstances, which is not always possible. Unlike the results from Smith’s series, they found no correlation between aortic valve gradients at the initial CABG and the need for a future AVR.

What then should be the guidelines for an AVR in asymptomatic patients with mild aortic stenosis requiring CABG surgery? I would agree with Smith and Peterson that patients less than 70 years of age with peak gradients greater than 30 mm Hg should have their valves replaced. In my own practice, I would also replace valves that are bicuspid, rheumatic, and heavily calcified. I also favor replacing valves in patients with reduced (≤ 35%) ejection fraction and peak gradients of 20 mm Hg or higher where the gradient is likely to increase as the ejection fraction improves after CABG surgery. Finally, I am more likely to replace valves when previous echocardiograms have been obtained which document that there has been a significant increase in gradients and a reduction in valve area over a 1- to 2-year period.

I do not favor replacing valves in asymptomatic patients when previous echos show stable gradients, in older patients with comorbid conditions where life expectancy will not exceed 5 years, and in those patients requiring emergent or urgent CABG for ongoing myocardial necrosis or ischemia where the extra cross-clamp and pump time may result in increased morbidity and mortality. Finally, concomitant AVR should not be done in any asymptomatic patient if they will not accept the increased risk of stroke and renal dysfunction at the time of CABG surgery.

Unfortunately, most of the data available to clinicians regarding the outcomes of CABG and AVR in asymptomatic patients with mild aortic stenosis come from studies comparing CABG patients with patients who ultimately undergo AVR and redo CABG. The best way to resolve this issue is to do a prospective study comparing outcomes in patients with symptomatic coronary disease and asymptomatic mild aortic stenosis who are randomly assigned to either a CABG only or CABG and AVR group. This would determine the true long-term morbidity and mortality in this difficult group of patients.