1 Scientific RepoRts | 6:27345 | DOI: 10.1038/srep27345 www.nature.com/scientificreports early Invasive strategy for Unstable Angina: a New Meta-Analysis of old Clinical trials olivia Manfrini 1 , Beatrice Ricci 1 , Ada Dormi 2 , paolo emilio puddu 3 , edina Cenko 1 & Rafaele Bugiardini 1 Randomized controlled trials (RCTs) were conficting to support whether unstable angina versus non- ST-elevation myocardial infarction (UA/NSTEMI) patients best undergo early invasive or a conservative revascularization strategy. RCTs with cardiac biomarkers, in MEDLINE, EMBASE, and Cochrane Central Register of ControlledTrials from 1975–2013 were reviewed considering all cause mortality, recurrent non-fatal myocardial infarction (MI) and their combination. Follow-up lasted from 6–24 months and the use of routine invasive strategy up to its end was associated with a signifcantly lower composite of all-cause mortality and recurrent non-fatal MI (Relative Risk [RR] 0.79; 95% confdence interval [CI], 0.70–0.90) in UA/NSTEMI. In NSTEMI, by the invasive strategy, there was no beneft (RR 1.19; 95%CI, 1.03–1.38). In the shorter time period, from randomization to discharge, a routine invasive strategy was associated with signifcantly higher odds of the combined end-point among UA/NSTEMI (RR 1.29; 95% CI, 1.05–1.58) and NSTEMI (RR 1.82; 95% CI, 1.34–2.48) patients. Therefore, in trials recruiting a large number of UA patients, by routine invasive strategy the largest beneft was seen, whereas in NSTEMI patients death and non-fatal MI were not lowered. Routine invasive treatment in UA patients is accordingly supported by the present study. Two strategies have been used in managing patients with unstable angina/non-ST-elevation myocardial infarction (UA/NSTEMI). Patients may undergo an early invasive strategy of coronary angiography and revascularization by percutaneus coronary intervention (PCI) or a conservative, “ischemia guided” strategy in which hemodynamic procedures are performed only if there is evidence of recurrent ischemia or high risk features 1 . Clinical trials ofered conficting evidence to support one strategy over the other. As a result, there was consid- erable interest in summating the available information from large-scale clinical trials by using meta-analyses and systematic reviews that can provide a more robust estimate of the efect of a specifc therapy. Yet, even a number of meta-analyses have led to contradictory results regarding the efcacy of a routine use of an invasive strategy to reduce both nonfatal myocardial infarction (MI) and mortality 1–10 . Tere are a number of potential limitations of these studies, mainly heterogeneity, i.e. the extent to which diferent trials may give similar or diferent results 1–10 . Although statistical tests are routinely available to evaluate heterogeneity, physicians are not interested in this and rather look at clinical heterogeneity, i.e., specifc patho- physiologic causes that underlie heterogeneity across studies. In the current analysis, we explored the hypothesis that trials including a substantial number of patients with UA may provide evidence of a reduced rate of death and/or recurrent MI when treatment was by an early invasive strategy as compared to trials recruiting participants with just NSTEMI. To test this hypothesis we compared trials enrolling patients with both positive and negative biomarkers (UA/ NSTEMI) with those only recruiting participants with positive cardiac biomarkers (NSTEMI). Results Te literature search yielded 3896 hits. From these, 24 studies were selected for closer attention. Of these 16 studies were excluded according to explicit inclusion/exclusion criteria 11–26 (Fig. 1). Te eight remaining studies 1 Department of experimental, Diagnostics and Specialty Medicine, University of Bologna, Via Giuseppe Massarenti 9, 40138 Bologna, Italy. 2 Department of Medical and Surgical Sciences, University of Bologna, Via S. Giacomo 12, 40126 Bologna, Italy. 3 Department of cardiovascular, Respiratory, nephrological, Anesthesiological and Geriatric Sciences, Sapienza University of Rome, Viale del Policlinico 155, 00161 Rome, Italy. Correspondence and requests for materials should be addressed to P.E.P. (email: paoloemilio.puddu@uniroma1.it) Received: 12 November 2015 Accepted: 11 May 2016 Published: 07 June 2016 opeN