Influence of Failure to Rescue on Mortality After Transcatheter Aortic Valve Replacement

Abstract

Background

Mortality after transcatheter aortic valve replacement (TAVR) varies among centers. “Failure to rescue” (FTR) patients from post-TAVR complications may represent an unexplored opportunity for TAVR process improvement.

Methods

The Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy registry was queried for patients undergoing transfemoral TAVR between 2011 and 2016. Hospital FTR rate was derived from the ratio of observed-to-expected procedural mortality. Multivariable logistic regression models assessed the association between FTR and hospital mortality. Adjusted FTR rates were compared across tertiles of hospital mortality.

Results

The analysis included 61,804 patients (429 sites). Post-TAVR mortality at low-, middle-, and high-mortality hospitals was 1.8%, 3.3%, and 5.6% (P < .01), respectively. Risk-adjusted complication rates differed only slightly between tertiles (22.2% vs 24.5% vs 27.0%, P < .001). However, adjusted FTR rates were significantly worse in high- and medium-mortality hospitals than in low-mortality centers (14.6% vs 9.5% vs 5.4%, P < .001). This was true for all investigated complications, including conversion to open surgery (high-mortality: odds ratio [OR], 9.04 [95% CI, 4.12-19.83], P < .001; medium-mortality: OR 2.99 [95% CI, 1.48-6.07], P < .003), stroke (high-mortality: OR, 3.15 [95% CI, 1.97-5.04], P < .001; medium-mortality: OR, 1.67 [95% CI, 1.05-2.67], P < .032), and cardiac arrest (high-mortality: OR, 3.54 [95% CI, 2.57-4.87], P < .001; medium-mortality: OR, 1.67 [95% CI, 1.24-2.24], P < .001).

Conclusions

National TAVR mortality rates vary significantly across centers, despite comparable rates of postoperative complications. Patients at medium- and high-mortality centers face a disproportionately higher risk of death due to FTR. These findings highlight the need for a closer evaluation of post-TAVR care processes to address this disparity.