IJCS | Volume 31, Nº2, March / April 2018

153 Andrade et al. Stent versus surgery: randomized trials Int J Cardiovasc Sci. 2018;31(2)152-162 Original Article references of reviews published on the subject. The date of January 2002 was chosen as the initial period since drug ‑ eluting stents began to be established as a therapeutic method after that. Clinical trials were included in the review if they were randomized, had compared surgery and coronary angioplasty, used drug-eluting stents, involved exclusively multivessel disease or left main coronary artery obstruction, had a minimum follow-up of 1 year, and were published in international journals with an impact factor > 2.0. We used the following terms in the search: coronary artery bypass surgery, coronary stents, and randomized controlled trials. Studies exclusively using balloon or bare-metal stents, or which assessed predominantly one-vessel disease were not included. Studies using drug-eluting and bare-metal stents 1,4 were included as studies of the drug‑eluting stent era. Works resulting fromobservational studies (registries) or only published asmeetings proceedingswere not considered. We identified 10 randomized studies that satisfied the requirements: LE MANS, 1 SYNTAX, 2-3 CARDia, 4 Boudriot et al., 5 PRECOMBAT, 6 VA CARDS, 7 FREEDOM, 8 BEST, 9 NOBLE, 10 and EXCEL. 11 Three authors (PJNA, BAAF, and JLAAF) evaluated the studies, which were all considered to be of high quality. The main outcomes of interest were mortality and stroke. The incidence of AMI was not evaluated because the definition of this event varied widely in the studies. We also did not evaluate the incidence of new revascularization, because the superiority of surgery on this outcome is well established. Mortality was divided into early mortality, mortality at 1 year, and late mortality. Early mortality was defined as death occurring up to 30 days after the procedure, including deaths occurring after randomization but before the procedure. This mortality was obtained from seven studies, whereas three studies did not provide this information. 2,4,7 Mortality at 1 year was defined as death occurring up to 1 year after the procedure, including early mortality. This mortality was obtained from nine studies, while one study did not provide such information. 9 Late mortality was defined as death recorded at the end of follow-up, after at least 3 years. This mortality was obtained from eight studies, six of which performed a follow-up for 5 years, one for 3 years 2 and one for 10 years. 1 We were unable to obtain this information from two studies. 5,7 For the incidence of stroke, we considered the events occurring up to 1 year after the procedure. In eight studies, we obtained the results up to 30 days and in one of them, 2 up to 1 year, while in one of the studies, this information was unavailable. 9 We evaluated separately the results of studies in the left main coronary artery and late mortality in the subgroup of patients with diabetes. We also performed analysis of combined major adverse cardiac and cerebrovascular events (MACCE) and assessed the variables age, gender, presence of diabetes, SYNTAX score, and compromised ejection fraction in subgroups based on data published in five trials. 2,4,6,8,9 Combined MACCE comprised death, AMI, and new revascularization in two of these trials, 6,9 and death, AMI, and stroke in the remaining ones. In order to aggregate the outcomes of mortality and stroke, as well as those of MACCE (in subgroups), we considered whenever possible the absolute number of events and the number of patients followedup. Otherwise, percentageswere transformed into absolute numbers. Statistical analysis We measured the relative risk and the risk difference after grouping the results of each outcome. In order to assess the statistical significance of the differences between the drug-eluting stent and the surgery groups, we performed a meta-analysis using the Mantel-Haenszel method, with a fixed-effect model. We calculated the heterogeneity of the studies using Cochran's Q test and the significance of the measure of the meta-analytic effect using the Z test. The differences between the results in the stent and CABG groups were considered significant if p < 0.05. The statistical analyses were performed using the program Review Manager (RevMan), version 5.3 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014). In order to represent the heterogeneity of the studies, we constructed Forest plots . We used the risk difference to plot these graphs since this is a more stable index. We refrained from using odds ratio or relative risk due to the inclusion of some clinical trials with zero or near zero events. Results Characteristics of the trials are shown in table 1 The studies included a total of 9268 patients (4642 in the stent group and 4626 in the CABG group). The mean age of the patients was 64 years, 75% were male, 51% were diabetic, 24% were smokers, 64% were hypertensive, and 31% had unstable angina. The mean ejection fraction (reported in seven studies) was 59%, the mean EuroSCORE (reported in five studies) was 2.9, and the mean SYNTAX