Insights into ovarian response with a fixed low dose FSH stimulation in an IUI programme: the PRORAILS study

Study question: Are patients' characteristics, such as anti-Müllerian hormone (AMH) and BMI, reliable factors to predict ovarian response in couples with unexplained subfertility undergoing IUI with ovarian hyperstimulation (IUI-OH)?

Summary answer: We observed no solid relationship between serum AMH and ovarian response.

What is known already: Ovarian stimulation during IUI treatment could lead to a higher chance of pregnancy, but also a higher incidence of multiple pregnancies, unless strict cancellation criteria are being used. Several factors could influence the result of the stimulation, such as age, BMI and hormonal status of the female. In IVF treatment, AMH has shown to be a useful predictor of ovarian stimulation to optimize the outcome; however, in a milder stimulation protocol, such as IUI, this has not been investigated.

Study design, size, duration: We performed a prospective cohort study and evaluated the first IUI stimulation cycle of 492 patients. The study was conducted between 2012 and 2017. Follow-up ended if patients were not pregnant after the first cycle. If pregnancy did occur, follow-up lasted until delivery.

Participants/materials, setting, methods: PRORAILS is a large multicentre nationwide cohort study executed in the Netherlands. Eligible women aged 18-43 years who were diagnosed with unexplained subfertility or mild male subfertility according to the Dutch guideline, with a regular indication for IUI-OH, were asked to participate. Ovarian response was assessed using a transvaginal ultrasound 5-7 days after initiation of the stimulation and was repeated according to the size of the leading follicles. Ovarian response was defined as optimal or suboptimal based on the total number of dominant follicles >15 mm. A successful stimulation was defined as the presence of two to three follicles >15 mm on the day of hCG administration. Serum AMH (µg/l) was measured by ELISA, and samples were taken on day 2, 3 or 4 of the menstrual cycle. Poisson regression was used to estimate the risk of a suboptimal ovarian response.

Main results and the role of chance: Of the 492 participants, the mean age was 33 years and the mean subfertility duration was 2.5 years. The median serum AMH was 2.1 (µg/l). The majority of patients had a suboptimal response: 326 women (66%), of whom 224 (45%) had a hypo response (defined as <two follicles sized > 15 mm) and 102 (21%) had a hyper response (defined as more than three follicles sized >15 mm). The lowest AMH category showed a trend towards a smaller risk of a suboptimal response (relative risk ratio 0.76 (95% CI 0.54, 1.06)), but this effect did not reach statistical significance. In the prediction models, BMI and serum basal FSH were significant predictors of a hypo response, while for hyper response the factors age, BMI and serum FSH were significant. A higher BMI showed a higher risk for hypo response, as did a higher FSH whereas a lower BMI and lower FSH showed a higher risk for hyper response. The addition of AMH to the models did not improve the predictive abilities.

Limitations, reasons for caution: Although the study was prospective, the main analyses were cross-sectional with characteristics measured at one time-point. The study was not powered to provide insight into predictors of pregnancy and live births and, therefore, the result for pregnancy should be interpreted with caution.

Wider implications of the findings: This was the first large multicentre study that investigated the characteristics of ovarian response categories using standardized methods and centrally analysed laboratory measures. PRORAILS is a nationwide study with 15 hospitals and, therefore, these results are generalizable to other hospitals in the Netherlands. This study provides high-quality outcomes advancing the subfertility research field. Future studies would benefit from a randomized design investigating the effectiveness of an individualized approach versus a fixed dose. Also, the relation between a good ovarian stimulation and pregnancy rate could be further investigated.

Study funding/competing interest(s): The PRORAILS study is sponsored by Merck B.V., Schiphol-Rijk, the Netherlands, an affiliation of Merck KGaA, Darmstadt, Germany (EMR700623_612). Merck KGaA, Darmstadt, Germany, reviewed the manuscript prior to submission. The opinions remain those of the authors. Merck KGaA, Darmstadt, Germany, had no influence on the use of medication in this study. The recombinant FSH was mostly provided by Merck B.V. or MSD. F.B. is a member of the external advisory board for Merck B.V., Schiphol-Rijk, the Netherlands, and has received a research grant from Merck B.V., Schiphol-Rijk. H.v.B. is an employee from IQVIA, which is a commercial data-analysing company, and received payment for her part in the article.

Trial registration number: NCT01662180.