ObjectiveIntra-uterine insemination (IUI) is widely used for different indications. The aim of the present study was to evaluate the efficiency of intra-uterine insemination as a function of indication and origin of the inseminated spermatozoa.Study designThe retrospective study involved 827 first attempts of IUI in 827 couples between January 2011 and July 2017 in the Toulouse university hospital. Of these, 642 used fresh sperm from the husband, 40 frozen sperm from the husband and 145 frozen sperm from donors. The measured outcome parameter was live birth rate per attempt.ResultsWhen comparing couples lacking functional gametes (due to male or female causes), to couples who could potentially conceive spontaneously, i.e. subfertile, the latter were found to have a significantly lower live birth rate (18% vs 26%; P < 0.05). Even when adjusted for demographic parameters, which differed significantly between the 2 groups (female age, percentage of women suffering from primary infertility, BMI, number of inseminated motile spermatozoa and stimulation duration), this difference remained statistically significant (OR = 0.639 [0.425-0.961]; P = 0.0316).ConclusionWhen compared to couples lacking functional gametes, subfertile couples have poor IUI outcomes, suggesting a hidden cause of infertility, despite no apparent differences in ovarian reserve, tubal potency, results of ovarian stimulation and normal conventional sperm parameters. Further studies are required to better characterise and identify this subgroup of women with poor IUI outcomes.