STS AVR Composite Advances Performance Measurement in Cardiac Surgery

David M. Shahian, MD, Chair, STS Workforce on National Databases

STS News, Fall 2012 -- Beginning next year, participants in the Adult Cardiac Surgery Database (ACSD) component of the STS National Database will have an opportunity to publicly report their scores on the newly developed STS aortic valve replacement (AVR) composite metric, in addition to their STS CABG composite scores, on a voluntary basis.

Although CABG surgery remains the most frequently performed cardiac surgical procedure in the United States, the STS Quality Measurement Task Force developed the STS AVR composite performance metric in response to a notable decline in the frequency of CABG surgery and a corresponding increase in the relative percentage of valve surgery cases. By providing information for both procedures, the overall performance of a cardiac surgical practice will be more comprehensively assessed.

CABG Composite Score
Historically, CABG mortality was the sole metric of cardiac surgical performance. However, in the past 20 years, overall CABG mortality rates have fallen dramatically, approaching 1% in some states, and the spread in mortality rates among programs has narrowed. Statistically, this makes it difficult to distinguish quality differences based only on mortality.

There has also been increasing recognition that mortality alone is inadequate to assess the overall quality of cardiac operations. Consider two patients who survive CABG, one of whom has a perfectly uncomplicated hospitalization while the other suffers acute renal failure and requires lifelong dialysis. Judged only by mortality, these two patients have had similar outcomes, but in reality, one has sustained a devastating, life-altering adverse outcome.

To more comprehensively assess performance quality, STS developed the four-domain CABG composite score in 2007. The STS CABG composite score incorporates not only risk-adjusted operative mortality, but also 10 other individually National Quality Forum-endorsed measures, including:

• occurrence of any of five major complications (stroke, renal failure, sternal infection, reoperation, or prolonged ventilation); 
• use of an internal mammary artery graft; and 
• administration of four perioperative medications.

Results for this composite measure have been reported to ACSD participants on a semi-annual basis using both numerical scores and a star rating system, and this metric subsequently received NQF endorsement in January 2011. Approximately 75% of ACSD participants receive two stars (average rating), 10%–15% receive one star (below average performance), and 10%–15% score three stars (above average performance).

Beginning in September 2010, ACSD participants were given the opportunity to publicly release their results on the Consumer Reports and STS websites. Presently, almost half of ACSD participants take advantage of this opportunity.

AVR Composite Score
Like its CABG predecessor, the new AVR composite includes two outcomes domains—risk-adjusted mortality and risk-adjusted morbidity. However, as there is no analogue to internal mammary artery use in valve surgery, and because the appropriate medications are less well defined for valve procedures, process domains are not included in the AVR composite.

ACSD participants received results of their AVR composite scores in May 2012. Similar to the CABG composite measure, the AVR composite score will be provided on a semi-annual basis, and participants will have an opportunity to publicly report their results on a voluntary basis beginning in 2013. Due to the smaller sample sizes compared with CABG, and because there are only two as opposed to four quality domains, pilot studies suggest that the percentage of both one- and three-star AVR programs will be smaller than for CABG.

Performance measurement is rapidly evolving nationally. STS will continue to be the leader in these initiatives by providing the very best metrics upon which to assess the quality of a cardiothoracic surgical practice. Adding the results of another commonly performed procedure, AVR, will enhance these efforts.

To learn more about the Society’s public reporting initiatives, go to or contact Bianca Reyes at breyes [at] sts [dot] org or (312) 202-5839.