A clinical prediction score based on late ischemic and bleeding risk factors may help identify patients who may benefit or have harm from continuing to take dual antiplatelet therapy after undergoing PCI, according to a secondary analysis of a previously published study.
Lead researcher Robert W. Yeh, MD, MSc, of Beth Israel Deaconess Medical Center in Boston, and colleagues published their findings online in JAMA on March 29.
The researcher examined the DAPT (Dual Antiplatelet Therapy) trial, which was conducted in 11 countries from August 2009 to May 2014. Patients were enrolled after undergoing PCI with a drug-eluting stent or bare metal stent and treated with open-label thienopyridine plus aspirin for 12 months.
Patients were eligible to continue taking aspirin and remain in the study if they adhered to their therapy and were free from moderate or severe bleeding, MI, stroke, repeat coronary revascularization and stent thrombosis. They were then randomized to continue taking thienopyridine or placebo for 18 months.
Patients were excluded if they received long-term anticoagulation therapy, planned on undergoing surgery that required discontinuing antiplatelet therapy for more than 14 days or had a life expectancy of less than three years.
The researchers sought to stratify outcomes based on a multivariable risk score to identify patients who benefited the most and were harmed the most from continuing to take dual antiplatelet therapy more than a year after undergoing PCI. They identified 37 variables that were potentially associated with ischemic or bleeding events and compared clinical and procedural characteristic between patients experiencing events from 12 through 30 months with those who did not experience events.
Of the 11,648 patients who underwent PCI, 40.3 percent received everolimus-eluting stents, 22.9 percent received paclitaxel-eluting stents, 10.9 percent received zotarolimus-eluting stents, 9.6 percent received sirolimus-eluting stents, 14.4 percent received bare metal stents and 1.8 percent received more than one type of stent.
From months 12 through 30 following PCI, 3.0 percent of patients developed MI or stent thrombosis and 1.8 percent developed moderate or severe bleeding.
During the follow-up period, 33 patients had an ischemic and bleeding event. Patients who had an ischemic event had higher rates of cardiovascular risk factors, had higher rates of cardiovascular disease and were more likely to have been randomized to placebo compared with patients without an ischemic event. Patients who had a bleeding event were older, had a lower prevalence of smoking, had a higher prevalence of hypertension, prior congestive heart failure, renal insufficiency/failure, peripheral arterial disease, atrial fibrillation, prior stroke/transient ischemic attack, prior PCI and history of cancer and were more likely to have been randomized to continued thienopyridine compared with patients without a bleeding event.
A multivariable Cox regression analysis found that the following were significant predictors of both ischemic and bleeding events: randomized treatment group, peripheral arterial disease, hypertension and renal insufficiency/failure. The following variables predicted only the risk of ischemic events: history of PCI or MI prior to the index procedure, stent diameter less than 3 mm, MI at presentation, history of congestive heart failure or left ventricular ejection fraction lower than 30 percent, paclitaxel-eluting stent, vein graft stent, cigarette smoking within the year prior to index procedure and diabetes.
According to the researchers’ prediction rule, patients with a high score who continued taking thienopyridine had a significant reduction in ischemic events and smaller increases in bleeding compared with those with a low score.
The researchers cited a few limitations of the study, including that they did not have information on preexisting anemia, prior bleeding, and granular measures of atherosclerosis extent and severity. They also said the score was relevant only to patients who had similar characteristics to patients enrolled in this study.
“Although the development of the score was prespecified, the analysis should be considered exploratory,” the researchers wrote. “Thus, use of this prediction score should be cautious until further validation is performed, and optimal clinical and procedural care to reduce overall bleeding and ischemic risks should be practiced independent of a patient’s