Objective: The objective of this study was the evaluation of rFSH versus purified urinary (u) FSH in a single IVF program involving all patients treated by a clinician and two embryologists and analysed by a multivariated approach. Design: A retrospective cohort analysis. Materials and Methods: The study included 425 consecutive cycles of IVF or ICSI, with at least one embryo transferred in women under the age of 40, down regulated under long or short protocol and stimulated to obtain high number of follicles. The only criterion for rFSH use was the coverage of medication from patient’s medical insurance. rFSH formulations were follitropin a or,. Factors analysed were: patient age, stimulation protocol, type of FSH, no. of eggs retrieved, no. of fertilized eggs, no. of transferred embryos. Cycle outcome factors were: increasing levels of serum hCG (pregnancy rate), heart beats (clinical pregnancy rate), implantation rate and delivery rate. The cycles were stratified in two groups, the rFSH group with 75 and the uFSH with 350 cycles. Univariated comparisons of all variables between the two groups were performed by ANOVA. Multivariated analysis was performed with all variables entering unconditionally into the model. Results: The mean age was similar in both groups, 33.4 in rFSH vs 32.7 years in uFSH. The mean no. of eggs retrieved and embryos obtained were 20.7 and 11.3 in rFSH vs 15.9 and 8.3 in uFSH groups (p=0.0001 and p=0.0000 respectively). The mean no of transferred embryos were similar, 4.3 in the rFSH vs 4.2 in the uFSH groups. In the rFSH group 37 cycles (49%) resulted in frozen spare embryos, significantly higher than those in uFSH 104 (30%), p=0.0001. Univariated analysis demonstrated similar outcome in both groups. Pregnancy, clinical pregnancy and delivery rates were 54.6%, 48.0% and 45.3% in rFSH vs 61.4%, 55.4% and 50.3% in uFSH groups. Miscarriages (9.3% in rFSH vs 12.0% in uFSH) and ectopics were similar in both groups. The rFSH group also presented a tendency towards lower implantation rate, 18.8 % vs 23.4% sacs/embryos (p=0.08) in the uFSH group. To eliminate confounding factors multivariated analysis was performed. Pregnancy was significantly correlated with the no. of availiable embryos (p=0.01) and the no. of transferred embryos (p=0.001). Delivery was significantly correlated with the age (p=0.004). Implantation rate was significantly correlated with the no. of availiable embryos (p=0.0001), negatively with the no. of transferred embryos (p=0.03) and negatively with the use of rFSH (p=0.03). Conclusion: The multivariated approach suggests that the use of rFSH in IVF correlates with higher no. of eggs retrieved, higher no. of embryos and higher no. of cycles with available frozen embryos. However, in terms of IVF outcome rFSH has no advantage over purified uFSH, possibly due to an underlying adverse effect of the use of rFSH on the endometrial receptivity.