Risk of stent thrombosis after sirolimus or paclitaxel eluting coronary stent implantation

Both sirolimus (SES) and paclitaxel eluting stents (PES) have been shown to reduce significantly both the incidence of binary angiographic restenosis and the need for new revascularization procedures [1]. We read with great interest the paper by Sidhu et al., in which the better angiographic behaviour of SES in comparison with PES was not translated into significant differences in terms of clinical events [2]. As paclitaxel is cytotoxic, whereas sirolimus is cytostatic, the risk of stent thrombosis between the two types of drug-eluting stents could be different. In fact, in a recent work, a trend to an increased risk of stent thrombosis after PES implantation was found, whereas the risk of stent thrombosis was similar with SES and bare-metal stents [3]. Stent thrombosis is an infrequent complication. Because of that, data from single trials are not sufficient to compare the incidence of stent thrombosis between two different types of stent. For the evaluation of such an infrequent entity, meta-analysis may increase increase power and precision and provides an overall estimate and range of effect. In order to help to clarify whether the risk of stent thrombosis is different between SES and PES, we have performed a meta-analysis from nine randomized trials that have compared SES and PES, including 5024 patients (2514 allocated to SES, 2510 allocated to PES). The trials included in the meta-analysis were: TAXI (n = 202), REALITY (n = 1353), SIRTAX (n = 1012), ISAR-DIABETES (n = 250), ISAR-DESIRE (n = 200), CORPAL (n = 652), ISAR-SMART-3 (n = 360), BASKET (n = 545) and ISAR-TEST (n = 450) [4–12]. Follow-up ranged from 6 to 12 months. In most trials, the Cypher stent (Cordis Corp., Miami Lakes, FL, USA) and the Taxus stent (Boston Sci., Natick, MA, USA), with polymeric-release of sirolimus and paclitaxel, respectively, were randomly compared in patients with native de novo lesions. In the ISAR-DESIRE study, only patients with in-stent restenosis after bare-metal stent implantation were included. In the ISAR-TEST, the Yukon stent, with nonpolymeric release of sirolimus, was compared with the Taxus stent. The risk ratio for stent thrombosis and its 95% confidence interval (CI) was calculated comparing SES with PES rates using raw data for each study and for the pooled population. The Q-test for heterogeneity and the fixed-effect model were used. There was no heterogeneity among the trials [Q-test for heterogeneity: χ2 = 5.41, d.f. = 6 (P = 0.49); I2 = 0%]. The overall risk of stent thrombosis in the overall population was 0.92% (n = 46 patients): 0.83% (21/2514) and 1.00% (25/2510) in patients allocated to SES and PES, respectively (risk ratio 1.17, 95% CI 0.67, 2.35; P = 0.57) (Figure 1). Figure 1 Comparison of the incidence of stent thrombosis in each of the nine randomized clinical studies included in the meta-analysis, as well as in the pooled population (fixed-effect model) Sirolimus and paclitaxel have different mechanisms of action (sirolimus is cytostatic, whereas paclitaxel is cytotoxic). Moreover, both stent platform and polymer are different in Taxus and Cypher stents, and whereas Taxus releases paclitaxel from a polymer, the Yukon stent has a nonpolymeric release of sirolimus. However, in view of our results, all these differences in stent design, type of polymer and type of drug do not seem to be translated into different risk of stent thrombosis. Given the high number of patients included in our meta-analysis (>5000), the possibility of existing but undetected differences in the risk of stent thrombosis between SES and PES is very low. In a previously published meta-analysis including trials that compared drug-eluting stents and bare-metal stents, we found a similar risk of stent thrombosis in PES and SES trials when the SCORE trial and the patients from the ASPECT study that did not receive thienopyridines were excluded (0.57% vs. 0.58%, P = 1.000) [13]. Drug-eluting stents have revolutionized cardiology worldwide, since they dramatically reduce the need for new revascularization procedures after percutaneous coronary interventions. Because of that, now that more diabetic patients are now being treated percutaneously,with longer lesions and smaller vessels, the absolute incidence of stent thrombosis in the era of drug-eluting stents will probably increase. However, we should bear in mind that the risk of stent thrombosis is related mainly to the appropriateness of antiplatelet therapy and the characteristics of the lesion, more than to the type of coronary stent.