What are the best semen parameters to predict pregnancy in intrauterine insemination cycles?

Background/Aim: Intrauterine insemination (IUI) is widely used for sub-fertile couples. The optimal method of evaluating semen analysis as a predictor of pregnancy is still not concluded. The aim of this study was to determine the best semen analysis parameter to predict pregnancy in IUI treatments and to evaluate when the sperm is insufficient to IUI, and the couple has to proceed with IVF. Materials and Methods: We evaluated all patients who underwent ovulation induction with IUI. The data were analyzed in different combinations to build the best model. Results: The semen analysis parameters that were found as the best predictors for pregnancy include volume, concentration, motility and morphology from the sperm analysis during the infertility evaluation. A cut-off of total motile normal morphology sperm count (TMNC) <4.8 × 10 6 from these parameters yielded very low probability of pregnancy and in those cases it is recommended to refer patients to IVF. Conclusion: The model includes the normal morphology in the formula of total motile sperm count more accurately and significantly predicts pregnancy rate than the data of sperm in the day of IUI. Semen analysis results with TMNC <4.8 × 10 6 should be considered as a threshold for referral to IVF.