Donor IUI is equally effective for heterosexual couples, single women and lesbians, but autologous IUI does worse.

study questionAre there differences in the clinical outcomes of IUI among different populational groups (heterosexual couples, single women and lesbian couples)? SUMMARY ANSWER: The outcome of donor IUI (D-IUI) is similar in all populational groups and better than that seen with autologous insemination. WHAT IS KNOWN ALREADY: A vast body of literature on clinical outcome is available for counselling heterosexual couples regarding decisions related to ART. The reproductive potential of single women, lesbian couples and heterosexual couples who need donor semen is assumed to be better, but there is a scarcity of data on their ART performance to actually confirm it. STUDY DESIGN, SIZE, DURATION: In this retrospective multicentric cohort study, a total of 7228 IUI treatment cycles performed in 3807 patients between January 2013 and December 2016 in 13 private clinics belonging to the same reproductive medicine group in Spain were included. Patients with previous IUI attempts were excluded from the study. Only 1.9% of cycles were lost to follow-up. PARTICIPANTS/MATERIALS, SETTING,

methodsA total of 5318 D-IUI cycles were performed in three different populational groups: heterosexual couples (D-HC, 1167 cycles), single women (SW, 2839 cycles) and lesbian couples (LC, 1312), while a total of 1910 autologous IUI cycles were performed in heterosexual couples (A-HC). This last one was considered the control group and was composed of cycles performed in couples with a male partner with sperm parameters equivalent to those requested from donors. In order to identify factors with an impact on clinical outcome, a multivariate logistic regression analysis was performed. Regarding live birth rate (LBR), mixed effect models were employed to control for the fact that different patients were submitted to different numbers of treatments. MAIN RESULTS AND THE ROLE OF CHANCE: Parameters that were significant to the primary outcome (LBR) according to the multivariate analysis were the populational group (D-HC, SW, LC and A-HC) to which the patient belonged, female age and a diagnosis of low ovarian reserve. At the age range of good prognosis (≤37 years), LBR was similar in all groups that underwent D-IUI (18.8% for D-HC, 16.5% for SW and 17.6% for LC) but was significantly lower in the autologous IUI (A-HC) group (11%). For all these significant findings, the strength of the association was confirmed by P values <0.001. From 38 years of age on, no significant differences were observed among the populational groups studied, and for all of them, LBR was below 7% from 40 years of age on. LIMITATIONS, REASONS FOR CAUTION: To the best of our knowledge, a smoking habit was the only known factor with a potential effect on ART outcome that could not be controlled for, due to the unavailability of this information in a significant percentage of the clinical files studied. Our study was not capable of precisely quantifying the impact of a diagnosis of low ovarian reserve on the LBR of both IUI and D-IUI, due to the number of cycles performed in patients with such diagnosis (n = 231, 3.2% of the total). WIDER IMPLICATIONS OF THE

findingsFor the first time, a comparison among D-HC, SW, LC and A-HC was performed in a study with a robust sample size and controlling for potential sources of bias. There is now sound evidence that equivalent clinical outcome is seen in the three groups treated with donor semen (D-HC, SW and LC). Specifically, regarding the comparison between SW and LC, our findings rule out differences in LBR proposed by previous publications, with very similar clinical outcomes within the same age ranges. At age ranges of good prognosis (≤37 years), reproductive performance of D-IUI is significantly better than that seen in heterosexual couples undergoing autologous IUI, even when only cases of optimal sperm quality are considered in this last group. This finding is in agreement with the concept that, as a group, A-HC are more prone to have female factor infertility, even when their infertility assessment finds no contraindication to IUI. Age affects all these groups equally, with none of them reaching a 7% LBR after the age of 40 years. Our findings will be useful for the counselling of patients from the different populations studied here about ART strategies. STUDY FUNDING/COMPETING INTEREST(S): None.