Predictors of use and outcomes of mechanical valve replacement in the United States (2008–2017)

Abstract

BACKGROUND: Contemporary nationwide data on the use, predictors, and outcomes of mechanical valve replacement in patients less than 70 years of age are limited. METHODS AND RESULTS: We identified hospitalizations for aortic valve replacement (AVR) or mitral valve replacement (MVR) in the Nationwide Inpatient Sample between January 1, 2008, and December 31, 2017. The study’s end points included predic-tors of mechanical valve replacement and risk-adjusted in-hospital mortality. Among 253 100 hospitalizations for AVR, the use rate of mechanical prosthesis decreased from 45.3% in 2008 to 17.0% in 2017. Among 284 962 hospitalizations for MVR, mechanical prosthesis use decreased from 59.5% in 2008 to 29.2% in 2017 (P for trend<0.001). In multilogistic regression analyses, female sex, prior sternotomy, prior defibrillator, and South/West geographic location were predictive of mechanical valve use. The presence of bicuspid valve was a negative predictor of mechanical AVR (odds ratio [OR], 0.68; 95% CI, 0.66– 0.69; P<0.001), whereas mitral stenosis was associated with higher mechanical MVR (OR, 1.28; 95% CI, 1.22–1.33; P<0.001). Unadjusted in-hospital mortality decreased over time with AVR but not with MVR, regardless of prosthesis choice. Using years 2008 and 2009 as a reference, risk-adjusted mortality also decreased over time with AVR but did not decrease after MVR. CONCLUSIONS: There is a substantial decline in the use of mechanical valve replacement among patients aged ≤70 years in the United States. Long-term durability data on bioprosthetic valve replacement are needed to better define the future role of mechanical valves in this age group.