番茄社区

Hospital surgical volume may not be a meaningful surrogate for the quality of U.S. centers performing surgical aortic valve replacement (SAVR) and mitral valve replacement or repair, according to nationally-representative findings.

Between 2007 and 2011, there were 682 hospitals identified from the National Inpatient Sample -- a random 20% sample of an all-payer database of hospitalized patients in the U.S. -- as centers performing these procedures.

When broken down by tertiles, hospital volume among them was not associated with lower in-hospital risk-standardized mortality rates, Dharam Kumbhani, MD, SM, of University of Texas Southwestern Medical Center in Dallas, and colleagues reported online in .

In the top tertile of hospitals by SAVR volume, for example, over one-third could be considered low-performers (centers in the highest tertile for mortality rates), whereas one out of five in the lowest-volume tertile were high-performers. This volume-performance mismatch was observed with mitral valve replacements and repairs as well.

Judging by volume tertiles alone would mischaracterize hospitals as low- or high-performers 44.7% of the time, according to Kumbhani's group.

"Valve surgery quality improvement endeavors should focus on a more comprehensive assessment that includes risk-adjusted outcomes rather than hospital volume alone," the authors suggested, because "using a pure volume-based criterion as a surrogate to define quality of care has the potential to misclassify a substantial number of the hospitals performing these surgical procedures in the United States."

A typical hospital performed a median of 43 SAVRs, 13 mitral valve replacements, and nine mitral valve repairs per year. The risk-standardized in-hospital mortality rate was 4.8% for SAVR and coronary artery bypass grafting (CABG), 2.9% for isolated SAVR, 6.4% for mitral valve replacement, and 2.7% for mitral valve repair.

Patient characteristics and in-hospital mortality rates stayed stable from 2011 through the 2012-2014 period.

Kumbhani's group suggested that, at this point, valve surgeries might be mature enough that the volume-outcome relationship is less pronounced. In addition, individual operator volumes may be more important than institutional volume for these procedures, they added.

Their risk adjustment accounted for patient demographics and Elixhauser comorbidities.

The investigators acknowledged that their study was limited by not having all patient-level baseline data or procedural complications captured by the database.

In an accompanying editorial, Hartzell Schaff, MD, of the Mayo Clinic in Rochester, Minnesota, expressed other qualms with the study.

"Elixhauser comorbidity measures may have value in many areas, but these variables were selected as risk factors separate from the primary reason for hospitalization. In assessing patients for valvular heart surgical procedures, variables such as causes of valve disease, degree of functional disability (and frailty), left and right ventricular function, pulmonary vascular disease, extent of coronary artery disease, and prior cardiac interventions are important patient-related factors that affect outcome and often drive referral patterns," Schaff said.

"It is important to recognize that performance of a large volume of aortic and mitral valve procedures does not guarantee the best outcome as judged by early mortality," he added. "However, institutional and surgeon experience do affect surgical results, and future studies of thresholds of experience should consider outcomes more broadly than hospital mortality alone." These outcomes, he suggested, could include hospital quality, such as length of stay, wound infection rates, and patient satisfaction.

Ultimately, the editorialist said it is "reassuring to know that low-volume centers and, by extrapolation, low-volume surgeons, can perform standard aortic and mitral valve procedures with relatively low hospital mortality."