Abstract 16595: The Impact of Aortic Valve Replacement on the Intermediate Survival of Patients With Severe Aortic Stenosis
Introduction: Recently more comorbid, elderly and frail patients with severe aortic stenosis are being eligible for aortic valve replacement (AVR) due to emerging new techniques. However, there is a lack of real life studies presenting impact of AVR in this population.
Hypothesis: AVR is of prognostic significance vs. medical treatment in patients with severe aortic stenosis (AS) adjusted for significant confounders and effect modifiers.
Methods: A prospective observational study of consenting patients >18 years of age under consideration for AVR at our tertiary teaching hospital. We used an explanatory strategy to investigate the relationship between AVR and survival. All other variables were of interest only as possible confounders or effect modifiers of this association. Maentel-Haenszel stratification analysis was performed to quantify confounders and to pinpoint effect modifiers using the Breslow and Day test of heterogeneity. Confounding was quantified by comparing the crude incidence rate ratio (IRR) with the adjusted Maentel-Haenszel IRR. Adjustment for multiple confounders was performed using the Cox proportional hazard regression model with a manual backward stepwise elimination procedure. The follow-up period was defined as time from AVR or from initial consultation (for non-operated patients) to date of death or study closing.
Results: Among 480 evaluated patients with severe AS, 389 had AVR and 91 were declined operative treatment. One-, three-, and five-year cumulative survival rates, respectively, were 95%, 87%, and 73% among operated patients, and 82%, 47%, and 27% among non-operated patients. Median survival time was 1604 days (95% CI, 1554-1655) in operated patients vs. 1090 days (95% CI, 954-1226) in non-operated patients ( p<0.001). The effect of operation on mortality depended on the interaction with diabetes, when adjusted for significant confounders (i.e. age, atrial fibrillation, NT-proBNP, hs-Troponin T, and NYHA classification). We found an effect of AVR on mortality in patients without diabetes (HR, 0.29; 95% CI, 0.19-0.468; p<0.001), but not among patients with diabetes.
Conclusions: AVR shows great prognostic effect in patients without diabetes.
Author Disclosures: A.I. Hussain: None. A. Auensen: None. C. Brunborg: None. S. Aakhus: None. A. Fiane: None. K.I. Pettersen: None. L. Gullestad: None.
- © 2016 by American Heart Association, Inc.