Correlates of late and very late thrombosis of drug eluting stents.

Background: Late and very late thrombosis of coronary drug-eluting stents (DES) has received much attention but essentially remains unpredictable. We sought to identify correlates of stent thrombosis (ST) developing >30 days after DES implantation.

Methods: We analyzed data from our single-center registry on 5,342 consecutive patients, who underwent a first DES implant between May 2003 and December 2006. The Academic Research Consortium definitions were applied to classify definite, probable, and possible ST. Cox regression analysis was performed to identify predictors of ST.

Results: Follow-up information was obtained at 6 months and at 1, 2, and 3 years after DES implantation in 97.2%, 95.2%, 92.4%, and 89.8% of patients, respectively. We identified 34 patients who developed definite and 5 with probable ST >30 days after the index stent procedure. The 3-year cumulative incidence of definite and definite + probable ST >30 days was 1.33% and 1.50%, respectively. By Cox multiple variable regression, predictors of definite + probable ST were age (hazard ratio [HR] 0.95, 95% CI 0.92-0.98, P < .001), current smoking (HR 2.55, 95% CI 1.29-5.07, P = .007), prior percutaneous coronary intervention (HR 2.68, 95% CI 1.42-5.05, P = .002), "off-label" DES indication (HR 3.10, 95% CI 1.10-8.75, P = .032), bifurcation stenting (HR 2.37, 95% CI 1.40-3.99, P = .001), and stenting an occluded vessel (HR 3.02, 95% CI 1.59-5.74, P < .001).

Conclusions: We identified several baseline characteristics, which, when combined, may identify patients at risk for late-occurring ST, particularly after off-label DES placement.