Success rates vary by ovulation induction protocol and body mass index: toward personalized medicine for intrauterine insemination

Obesity rates are increasing worldwide. Along with others, we have demonstrated that body mass index (BMI) influences fertility treatment success rates in a protocol-dependent fashion. Much less is known about intrauterine insemination (IUI) success because of the absence of IUI data in national reporting registries, such as the one maintained by the Society for Assisted Reproductive Technologies (SART). To investigate the effectiveness of ovulation induction with gonadotropins versus ovulation induction with oral agents and gonadotropins (aka “mini-stim”) in relation to BMI in patients undergoing IUI in a large, multi-center, retrospective analysis. We analyzed 67,662 treatment cycles from 33,867 patients undergoing IUI ovulation induction (OI) with gonadotropins alone versus oral agents and gonadotropin from 2009 to 2016. Discrete-time proportional-odds models with time-dependent covariates were used to model the odds of achieving ongoing pregnancy during a given cycle. Interaction terms between BMI and stimulation type were included in the models to test the hypothesis that the effect of BMI was modified by stimulation type. BMI was treated continuously and categorized according to NIH guidelines (<18.5, 18.5–25, 25–30, 30–35, >35). Models were adjusted for age, antral follicle count, medication type, sperm motility, clinic, diagnoses, and were weighted toward more recent data. Overall, OI with only gonadotropin agents had significantly higher odds of achieving ongoing pregnancy compared to OI with oral agents and gonadotropins (OR = 1.16, p < 0.05). However, we found that the impact of BMI on success rates also differed significantly with the type of IUI treatment protocol. The effect of BMI was nonlinear and peaked in the obesity categories of both protocols (p < 0.05, see Figure 1). At lower levels of BMI, there was no significant difference in the odds of having an ongoing pregnancy between stimulation types. However, there was a pronounced divergence in success rate as BMI increased; gonadotropin OI alone performed better than gonadotropin OI with oral agents. When subcategories of BMI were assessed, a statistically significant difference in success was observed in women with a BMI of 25–30. Additionally, gonadotropin OI alone was associated with higher odds of multiples than gonadotropin with oral agents (OR = 1.53, p < 0.05), independent of patient BMI. Oral OI agents when used in conjunction with gonadotropins appear to be less effective for obese women. We previously demonstrated oral agents alone result in lower pregnancy rates with both timed intercourse and IUI in obese women. However, the detrimental effect of increased BMI was not observed with gonadotropin alone. Here, we demonstrate that OI with only gonadotropin agents is associated with higher success rates than OI with oral agents and gonadotropins. This difference is most pronounced in higher BMI patients. Additionally, we demonstrated gonadotropin OI alone is related to a higher probability of multiples than oral and gonadotropin independent of BMI. These data suggest that the best IUI protocol for a given patient varies with respect to their BMI.