BACKGROUND: The aim of this meta-analysis was to evaluate the role of androgens or androgen-modulating agents on the probability of pregnancy achievement in poor responders undergoing IVF.METHODS: Medline, EMBASE, CENTRAL, Scopus and Web of Science databases were searched for the identification of randomized controlled trials evaluating the administration of testosterone, dehydroepiandrosterone (DHEA), aromatase inhibitors, recombinant luteinizing hormone (rLH) and recombinant human chorionic gonadotrophin (rhCG) before or during ovarian stimulation of poor responders.RESULTS: In two trials involving 163 patients, pretreatment with transdermal testosterone was associated with an increase in clinical pregnancy [risk difference (RD): +15%, 95% confidence interval (CI): +3 to +26%] and live birth rates (RD: +11%, 95% CI: +0.3 to +22%) in poor responders undergoing ovarian stimulation for IVF. No significant differences in clinical pregnancy and live birth rates were observed between patients who received DHEA and those who did not. Similarly, (i) the use of aromatase inhibitors, (ii) addition of rLH and (iii) addition of rhCG in poor responders stimulated with rFSH for IVF were not associated with increased clinical pregnancy rates. In the only eligible study that provided data, live birth rate was increased in patients who received rLH when compared with those who did not (RD: +19%, 95% CI:+1 to +36%).CONCLUSIONS: Based on the limited available evidence, transdermal testosterone pretreatment seems to increase clinical pregnancy and live birth rates in poor responders undergoing ovarian stimulation for IVF. There is insufficient data to support a beneficial role of rLH, hCG, DHEA or letrozole administration in the probability of pregnancy in poor responders undergoing ovarian stimulation for IVF.