Surgical Aortic Valve Replacement Outcomes in Non–Transcatheter Aortic Valve Replacement Centers: Implications for Tier-Based Systems of Care

Background: With wide expansion of transcatheter aortic valve replacement (TAVR) and dissemination of multidisciplinary-based approaches to care, societies are discussing the implementation of a tier system to valve centers. This study explores the impact of tier-based systems of care on surgical aortic valve replacement (SAVR) outcomes at institutions that perform SAVR only. Methods: Medicare beneficiaries undergoing SAVR procedures from 2012 to 2015 were included. The SAVR hospitals were stratified into either tier A, valve centers with a TAVR program; or tier B, valve centers without a TAVR program. Adjusted survival, assessed by multivariable Cox regression, controlled for program type and patient risk profile. Time-dependent analysis accounted for hospitals that initiated a TAVR program during the study period. Results: Overall, there were 562 tier A and 485 tier B SAVR hospitals. Tier A hospitals had significantly higher comorbidity burden compared with tier B hospitals (all P < .05) but had significantly lower rates of 30-day mortality (3.2% vs 4.1%) and 1-year mortality (8.1% vs 9.4%; both P < .05). After risk stratification, tier B hospitals had significantly worse 30-day mortality compared with tier A hospitals for all patient risk-profiles, except for the low-risk patients (P < .01). These findings persisted in the time-dependent analysis. Adjusted midterm survival was higher in tier A vs tier B hospitals. Conclusions: Low-risk patients can safely undergo SAVR in both tier level hospitals without compromising outcomes. Establishment of quality of care measures, especially in the SAVR-only hospitals, remains paramount and should be closely integrated when designing tier-based systems for aortic valve replacement care.