Abstract Study question Are live birth rates after IUI with donor sperm (IUI-D) and controlled ovarian stimulation comparable between women with a spontaneous LH peak vs those without? Summary answer Biochemical, clinical, ongoing pregnancy rates and live birth rates were higher among women without an LH peak. What is known already It is common clinical practice to trigger ovulation in IUI cycles once specific criteria are met; if a natural LH surge appears, adjusting the IUI timing may become necessary. Pregnancy rates seem to be slightly better when IUI is scheduled in relation to the presence or absence of an LH peak in non-stimulated cycles. In IUI with stimulated cycles, however, there is no consensus in the medical literature regarding the best moment to program the IUI, due to different inclusion criteria, different IUI timing and definition of LH peak among studies. Study design, size, duration Retrospective cohort study of 9,657 IUI-D cycles performed between 2012 and 2019 in one fertility center. IUI-D without LH peak (n = 6,679) versus IUI-D with LH peak (n = 2,978) were compared. Differences in pregnancy outcomes between study groups were evaluated using a Pearson’s Chi2 test. A p < 0.05 was considered statistically significant. Participants/materials, setting, methods The definition used to define an LH peak is > 10UI/L in the last follicular control. In cases without an LH peak, when at least one dominant follicle reached 17mm, ovulation was triggered with human chorionic gonadotropin in the following 24h, and IUI-D was performed 38h after triggering. In cases with an LH peak, ovulation was triggered the 6h following the detection, and IUI-D was also performed 38h later. Main results and the role of chance The women BMI and age were comparable between groups, with a mean±SD of 35.2±4.8 years old, and 24.3±4.7 for BMI. Other characteristics such as number of previous inseminations, type of stimulation drug, initial dose, total dose, stimulation length and number of follicles > 16mm in the last follicular control were also comparable. As expected, the LH level at the last follicular control was different between groups, with a mean of 5.1UI/L in the no-LH peak and 21.4IU/L in the LH peak group. The group without an LH peak had higher biochemical, clinical, ongoing and live birth rates compared to the group with LH peak: 27.7% vs. 20.7%; 19.5% vs. 15.5%; 17.7% vs. 13.7%; 16.3% vs. 12.6%, respectively (p-value<0.001). Limitations, reasons for caution The main limitation of the study is its retrospective nature. Also, a definition of LH peak based in absolute values was used; a definition based in relative values may lead to different results. Wider implications of the findings: A definition of LH peak based on absolute numbers is imprecise, and the cut-off of 10UI/L does not allow a good scheduling for IUI. A LH peak based on relative values could improve the detection of patients starting ovulation and the accuracy in programming IUI. Trial registration number Not applicable