Good at One or Good at All? Variability of Coronary and Valve Operation Outcomes Within Centers

Johnston and colleagues evaluated morbidity and mortality following different cardiac operations at 18 institutions to determine if outcomes of coronary artery bypass grafting (CABG), aortic valve replacement, and mitral valve replacement were correlated within each institution. The large volume of CABG makes this procedure an obvious choice for large-scale measurement and reporting of outcomes, and the authors sought to determine if CABG outcomes correlated with, and could therefore be considered representative of, valve replacement outcomes. While the institutional observed-to-expected (OE) ratios for mortality were not correlated across the three procedures, the OE ratios for morbidity were. The authors suggest this could reflect greater dependence of morbidity on postoperative care and resources that are more broadly shared within an institution and greater dependence of early mortality on procedure-specific skills.