Abstract P108: Assessing Utility of Drug-Eluting Stents in Coronary Patients Aged 85 Years and Older

Abstract

Background: Very old (VO) adults (≥85 years) are intrinsically susceptible to coronary heart disease (CHD) and are often treated with coronary stents. While current evidence supports use of drug-eluting stents (DES), generalizability to VO remains controversial, especially given high age-associated bleeding risks from long-term dual antiplatelet therapy. Methods: We identified all VO CHD patients who received stents between April 1, 2003 and September 30, 2006 at all non-federal hospitals in Massachusetts (MA) (N=1619), and completed 2-year follow-up on the entire cohort. Patients were classified as DES-treated if stents were all drug-eluting and BMS-treated if stents were all bare-metal (those receiving DES and BMS were excluded). Mortality rates were determined from vital statistics records. New myocardial infarction (MI) and subsequent bleeding (requiring hospitalization) were determined using ICD-9 codes. Repeat target vessel revascularization (TVR) was determined from the MA database. Risk-adjusted cumulative incidence was estimated using propensity score matching based on 57 clinical, procedural, hospital, and insurance variables. Results: During the study period, 1145 and 474 VO patients received DES and BMS respectively. Unadjusted 2-year mortality rates were 23.8% vs. 35.0% (DES vs. BMS, p<0.0001). Risk-adjusted (propensity score matching [665 DES:343 BMS]) 2-year DES vs. BMS rates were 25.4% vs. 32.4% (hazard ratio [HR] 0.69, 95% confidence interval [CI] 0.51, 0.93; p=0.01) for mortality and 9.8 vs.16.9% (HR 0.51, 95% CI 0.35, 0.76; p=0.001) for MI. Risk-adjusted 2-year rates of bleeding (13.6% vs. 12.5%, HR 1.08, 95% CI 0.73, 1.59; p=0.72) and TVR (5.7% vs. 8.7%, HR 0.62, 95% CI 0.38, 1.02; p=0.06) were similar in patients treated with DES vs. BMS. Conclusions: In a large, unrestricted, state dataset of VO CHD patients with mandated follow-up, DES were associated with reduced 2-year mortality and MI compared with a matched BMS subset, without an increased hazard of bleeding.