Prognostic indicators for intrauterine insemination (IUI): statistical model for IUI success.

A retrospective analysis of 260 completed intrauterine insemination (IUI) cycles was used in an attempt to identify significant variables predictive of treatment success. Couples received a maximum of three IUI cycles for the treatment of anovulation, cervical factors or unexplained infertility. Male factor problems were largely excluded by pretreatment screening. The overall pregnancy rate was 19.6% per completed cycle, the miscarriage rate 15.6%, the multiple pregnancy rate 23.5% and the cancellation rate 19%. Logistic regression identified four significant IUI variables [follicle number (P < 0.005), endometrial thickness (P < 0.005), duration of infertility (P < 0.01) and progressive motility (P < 0.05)] which were the most predictive of IUI success. The chance of conceiving when only one follicle was produced was only 7.6%, whereas with two follicles this chance increased to 26%. These variables were incorporated into a statistical model to allow the prediction of the chance of success in subsequent cycles. We conclude that careful patient selection criteria coupled with successful ovarian stimulation is the model for IUI success.