A randomised controlled trial comparing intrauterine-insemination or expectant management in couples with unexplained subfertility and a poor prognosis for natural conception: impact on coital frequency

Abstract Study question Is coital frequency (CF) influenced by intrauterine insemination with ovarian stimulation (IUI-OS) or expectant management in couples diagnosed with unexplained subfertility and a poor prognosis? Summary answer Couples allocated to expectant management had a higher CF than those allocated to IUI-OS. We found no evidence for associations with live birth. What is known already In couples with unexplained subfertility and a poor prognosis for natural conception, IUI-OS is generally first line treatment. Not much is known about the CF of these couples and to what extend they are still trying to conceive naturally and whether allocation to treatment or no treatment has an influence on CF. It is often postulated that the effect of IUI-OS might be derived from replacing coitus, especially in couples with longstanding subfertility i.e., over 24 months. Keeping a sex diary can provide insight on this. Study design, size, duration We performed a multicentre randomised-controlled-trial in couples with unexplained subfertility and a poor prognosis of conceiving naturally within one year. The couples were allocated in a 1:1 ratio to six months IUI-OS or six months expectant management. CF was assessed with diaries on moment of coitus in relation to the menstrual cycle. We intended to include 1091 couples but after almost 4 years, the study was stopped due to slow inclusion and lack of funding. Participants/materials, setting, methods We recruited 178 women; 86 women were assigned IUI-OS and 92 women expectant management. All participating Dutch-reading women were eligible for the online diaries (CASTOREDC). We defined the fertile window as 7 days before and 2 days after the estimated ovulation date for all cycles. We used the Mann Whitney U test for differences in CF and timing and logistic regression for the estimated association with live birth, adjusting for cycle number and treatment allocation. Main results and the role of chance After IUI-OS 28/86 women (33%) had a live birth and 12/92 (13%) after expectant management, yielding a relative risk of 2.5 (90%CI 1.49 to 4.17). Of the 178 recruited women 79 (44%) filled out at least one monthly diary on sex and/or menstruation dates, 35 after IUI-OS and 44 after EM. Of these10/35 (29%) had a live birth after IUI-OS and 3/44 (7%) after expectant management. In a total of 497 cycles, the 79 couples reported 2023 dates of coitus (average of 4.1 per cycle). The median CF was 3 (IQR: 0 to 7) in the IUI-OS group and 4 (1 to 6) in the expectant management group (p for difference: 0.08). The median CF that took place within the fertile window was 1 (0 to 3) in the IUI-OS group and 2 (0 to 3) in the expectant management group (p for difference: <0.01). No interaction was found between intercourse frequency or timing and treatment on live birth rate. The adjusted odds ratio for intercourse frequency was 0.93 (95%CI: 0.78-1.09). The adjusted odds ratio for intercourse timing was 0.94 (95%CI: 0.64-1.37). The odds ratio for allocated treatment was unaffected by these adjustments for intercourse frequency or timing. Limitations, reasons for caution Only 44% of trial population returned the CF and menstruation diaries which could have induced selection bias. We estimated the day of ovulation based on the date of the next menstrual period in the expectant management group of this pragmatic trial. Wider implications of the findings Even though the duration of subfertility was longstanding, couples in both groups still had coitus and the median CF was higher in the expectant management group. Trial registration number NTR5599