OBJECTIVE To study the effectiveness of different ovulation induction protocols in infertile women with polycystic ovary syndrome (PCOS) undergoing intrauterine insemination (IUI). METHODS Retrospective cohort study. METHODS Infertile women with PCOS undergoing IUI had ovulation induced with clomiphene citrate (CC), letrozole, or gonadotropins. METHODS Live birth and multiple pregnancy rates. RESULTS We performed 1068 IUI cycles in 765 couples. Live birth rates were comparable in CC-stimulated cycles (13.9%), letrozole-stimulated cycles (13.5%, OR 0.96 [95% CI, 0.63, 1.47]), and gonadotropins-stimulated cycles (13.2%, OR 0.94[95% CI, 0.62, 1.43]). Multiple pregnancy rates were 8.3%, 4.1% (OR 0.47 [95% CI, 0.09, 2.42]), and 3.3% (OR 0.34 [95% CI, 0.07, 1.95]) in CC, letrozole and gonadotropins stimulated cycles, respectively. Compared to CC, letrozole generated more often mono-follicular growth (75.9% versus 67.0%; OR 1.55 [95% CI, 1.11, 2.15]) but not more often after gonadotropins (72.9%, OR 1.17 [95% CI, 0.82, 1.66]. Cycles with multi-follicular growth did not result in statistically higher live birth rates than cycles with mono-follicular growth (15.8% vs. 12.7%, OR 1.29 [95% CI 0.89, 1.89]), but more often in multiple pregnancies (15.5% versus 0.8%, OR 22.4 [95% CI, 2.8, 181.6]). CONCLUSIONS In women with PCOS undergoing stimulated IUI, CC, letrozole and gonadotropins were equally effective and safe. Since multi-follicular growth increased the multiple pregnancy rates without increasing the overall live birth rate, ovulation induction would strictly aim for mono-follicular growth. Since letrozole had the highest mono-follicular growth rate, we recommend this drug as the treatment of first choice in infertile women undergoing ovulation induction and IUI.