Ten Year Outcomes of Percutaneous Coronary Intervention in a Low Volume Military Treatment Facility

Fentanes, Emilio; Wisenbaugh, Thomas
September 2013
Hawaii Journal of Medicine & Public Health;Sep2013 Supplement 4, Vol. 72 Issue 9, p51
Academic Journal
Background: Quality assurance of a percutaneous coronary intervention (PCI) program is particularly important when the center volume of procedures is low (<400/yr). Methods: We searched for predictors of 30-day and long-term incidence of stent thrombosis, myocardial infarction, need for repeat revascularization, and death from any cause for all PCIs performed at Tripler Army Medical Center from January 2002 through June 2012. The New York State Registry (NYSR) regression model was used to compute expected mortality rate based on patient risk factors. Review of electronic medical records, phone, and mail correspondence was used for follow-up. Long-term results were benchmarked against several large registries. Results: PCIs were performed 929 times in a total of 795 patients, for an average PCI volume of 88/yr. Follow up data was obtained on 99.8% of the patients at 30 days, with a median follow-up of 59 months. Eighteen deaths occurred prior to hospital discharge or during the first 30-days after PCI, for an unadjusted observed morality rate (OMR) of 2.26%. Based on the NYSR model our expected mortality was 2.19% (P=.88, NS). There was a higher incidence of acutely ischemic and unstable patients compared to NY State patients. Multivariate logistic regression identified independent predictors of death at 30 days: stent thrombosis (definite or probable, Odds Ratio 96), acute MI, hemodynamic instability (OR 47), emergent (OR 17) or salvage (OR 28) PCI, and need for pre-procedural balloon pumping (OR 27). The 30-day incidence of deinite or probable stent thrombosis was 2.6% and the cumulative Kaplan-Meier estimates were 3.0% at 6 months, 3.4% at 1 year and 4.2% at 3 years, all higher than benchmarks. Furthermore, since stent thrombosis was such a powerful risk factor for death at 30 days, we searched for and identified multivariate predictors of stent thrombosis: renal insufficiency (OR 7.15), emergent (OR 6.6) or salvage (OR 10.3) PCI, proximal LAD stenosis (OR 4.1), number of stents (OR 13.8), and operator (OR 3.7). Long-term survival Kaplan-Meier estimates were 94% at 1 year, 89% at 3 years, comparable to benchmarks. Discussion: Actual 30-day mortality was similar to expected mortality based on risk factors in the NYSR model, and long-term survival was comparable to that reported in large registries. Major adverse cardiac outcomes including stent thrombosis are known to be higher in low volume centers. In our facility, stent thrombosis was higher than predicted by patient-specific risk factors. Operator-related factors and the number of stents used per procedure may be modifiable, and have the potential of improving short-term outcomes.


Related Articles


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics