Over the last four decades, various urinary FSH (uFSH) products of different purity have been developed. In 1988 recombinant FSH (rFSH ) was prepared by transfecting Chinese hamster ovary cell lines with both FSH subunit genes. Both rFSH and uFSH are known to be effective in inducing ovulation in women with clomiphene-resistant polycystic ovary syndrome. Ovulation induction with FSH bears the risk of multiple follicle development, multiple pregnancies and ovarian hyperstimulation syndrome. The dose regimen used can affect the incidence of these complications. To compare in women with clomiphene-resistant polycystic ovary syndrome (PCOS) the safety and effectiveness in terms of ovulation, pregnancy, miscarriage, multiple pregnancy rate and ovarian hyperstimulation syndrome (OHSS) of 1) rFSH with uFSH and 2) different dose regimens of rFSH. The search strategy of the Menstrual Disorders and Subfertility review group was used to identify all relevant trials. Please see Review Group details. All relevant published RCT's were selected. Randomised controlled trials were eligible for inclusion if treatment consisted of recombinant FSH versus urinary FSH or recombinant FSH in different dose regimens, to induce ovulation in subfertile women with PCOS. A computerised MEDLINE and EMBASE search was used to identify randomised and non randomised controlled trials. The reference lists of all studies found were checked for relevant articles. Handsearching of bibliographies of relevant publications and reviews and abstracts of scientific meetings was performed. Serono Benelux BV and NV Organon, the manufacturers of follitropin alpha (Gonal F(R)) and follitropin beta (Puregon(R)) respectively, were asked for unpublished data and ongoing studies. Relevant data were extracted independently by two reviewers (NB, MW). Validity was assessed in terms of method of randomisation, completeness of follow-up, presence or absence of cross-over and co-intervention. All trials were screened and analysed according to predetermined quality criteria. 2X2 tables were generated for all the relevant outcomes. Odds ratios were generated using the Peto modified Mantel-Haenszel technique. Four randomised trials comparing rFSH versus uFSH were identified. No significant differences were demonstrated for the relevant outcomes. The odds ratio for ovulation rate was 1.19 (95% CI 0.78,1.80), for pregnancy rate 0.95 (95% CI 0.64,1.41), for miscarriage rate 1.26 (95% CI 0.59,2.70), for multiple pregnancy rate 0.44 (95% CI 0.16,1.21) and for OHSS 1.55 (95% CI 0.50,4.84). Similarly, in the only randomised trial that compared chronic low dose versus conventional regimen with rFSH no significant differences were found. At this moment there are not sufficient data to determine which of rFSH or uFSH is preferable for ovulation induction in women with PCOS.