Recent advances in understanding the relationship between long- and short-term weight change and fertility

Short-term weight change and assisted reproduction

Historically, studies on short-term weight loss interventions and fertility outcomes have focused on infertile obese women presenting for evaluation at fertility clinics. A systematic review of studies on this topic, published in 2014, nicely summarized the collective findings of the pioneering studies¹¹. Of the 11 articles included in the review, two were randomized controlled trials (RCTs), seven were cohort studies, and two were case reports. Eight of the studies implemented a dietary or lifestyle intervention program (or both) for weight loss, whereas three reported on a medical procedure for weight loss (for example, gastric bypass). The majority of studies (n = 9) included in the review were rated as having weak quality because of a less-than-rigorous study design, inappropriate control for confounders, high risk of selection bias, high rate of dropouts, or a combination of these. Moreover, the two RCTs on this topic, which were rated as having moderate quality, both had a total sample size of fewer than 50 women^12,13. Despite these limitations, most studies found that weight loss prior to assisted reproduction was associated with improved chances of pregnancy and live birth among obese infertile women¹¹. Furthermore, other benefits regularly noted included regularization of menstrual cycles, a reduction in the number of infertility treatment cycles required, and a decrease in cancellation rates. Thus, the conclusion of this review was that the current clinical recommendation of advising overweight or obese women to lose weight prior to assisted reproduction was supported by the evidence, although larger prospective RCTs were needed.

Since this first review, many more articles, including an additional systematic review and meta-analysis that came out in 2017¹⁴, have been published on this topic. This updated review aggregated data from six RCTs on diet and exercise interventions aimed at weight loss and pregnancy outcomes among female infertility patients. Although this meta-analysis found higher pregnancy rates among the intervention groups compared with the control groups (relative risk [RR] = 1.59, 95% confidence interval [CI] 1.01, 2.50), there was not a significant difference in live-birth rates (RR = 1.54, 95% CI 0.93, 2.56). Interestingly, data from studies pre-dating the Lifestyle Trial by Mutsaerts et al., published in 2016, suggested that diet and exercise interventions were superior to standard care but the inclusion of the results from the Lifestyle Trial changed this conclusion¹⁵. This is likely because the Lifestyle Trial was the biggest study to date and carried substantial weight in the meta-analysis. It also happened to be completely null.

The Lifestyle Trial was a large Dutch RCT that included 577 women randomly assigned to either a 6-month lifestyle intervention prior to 18 months of infertility treatment or prompt infertility treatment for 24 months¹⁵. This seminal study found no differences in the proportion of women achieving a clinical pregnancy or live birth between the groups 2 years after randomization. In fact, women in the control group had a significantly shorter median TTP resulting in live birth (5.2 months) compared with the intervention group (8.8 months), likely because the former proceeded directly to infertility treatment. The only findings in support of weight loss were that women in the intervention group had more ongoing pregnancies that resulted from natural conceptions and the number of infertility treatment cycles was lower in the intervention group compared with the control. The benefit of weight loss on natural conceptions was also shown, in a follow-up article, to be driven specifically by anovulatory obese women at baseline, as the effect was much less pronounced among ovulatory women¹⁶.

Although the findings of this influential RCT were mostly seen as discouraging, less than a year later, another large RCT of weight loss prior to IVF treatment among obese women was published with similar results¹⁷. Though not as large as the original Dutch study, this RCT was the second largest, including 317 obese infertile women from Sweden, Denmark, and Iceland who were randomly assigned to an intervention consisting of 12 weeks of a low-calorie liquid diet followed by 2 to 5 weeks of weight stabilization or a control group with IVF only. Both groups of women were followed through the completion of one IVF cycle. By the end of follow-up, although women in the intervention group had more spontaneous pregnancies leading to live birth than did the control group, there were no significant differences in the proportion of women achieving clinical pregnancy or live birth across the two groups.

Prior to these trials, most scientists and clinicians believed that short-term weight loss would improve reproductive outcomes in obese women with infertility. The counterintuitive results from these two large, well-conducted RCTs, however, pretty conclusively showed that short-term weight loss for obese infertile women does not remedy the outcomes of IVF cycles. The RCTs also nicely highlight the importance of high-quality randomized trials, particularly in settings where common sense suggests that the answer is already known. An additional, large multi-site RCT investigating a similar question is ongoing in the US. However, until its results are published, the best evidence remains that short-term weight loss does not appear beneficial for most infertile obese women undergoing infertility treatment. The one exception to this rule consists of overweight and obese anovulatory women for whom weight loss increased the likelihood of spontaneous pregnancy prior to undergoing fertility treatment. In this subgroup, weight loss may negate the need for costly medical intervention to achieve pregnancy and any potential complications that may be associated with infertility treatment.

Short-term weight change and natural conceptions

With regard to the benefits of short-term weight loss among overweight and obese women trying to conceive without medical assistance, the picture is less clear, as no RCTs have focused on this question. Although results from the two largest RCTs among women undergoing infertility treatment both suggested benefits of weight loss on achieving natural conception, it is important to remember that, in both trials, women in the control group received prompt infertility treatment and thus had less opportunity for unassisted pregnancy than did the intervention group.

In terms of observational evidence, the first studies on this topic focused on weight change between consecutive pregnancies and markers of fertility. In the first study, among 151,025 Swedish women, those with inter-pregnancy weight gains of 3 or more BMI units between their first two pregnancies had a 63% higher odds of stillbirth compared with those whose weight changed by less than 1 BMI unit¹⁸. A later study among 218,389 American women found that normal-weight mothers who became overweight or obese, overweight mothers who became obese, and obese mothers who stayed obese across the two pregnancies under study had increased risk of stillbirth compared with normal-weight mothers who stayed at normal weight¹⁹. With regard to effects on fecundity, among the 2,374 women who participated more than once in the Danish National Birth Cohort, those who were overweight or obese and lost or maintained weight between pregnancies had on average 5.50 (95% CI 1.35, 9.65) days shorter TTP for every 1 kg decrement in weight²⁰. Moreover, among women with a BMI of at least 18.5 kg/m², each 1 kg increment in weight gain was associated with 2.84 (95% CI 1.33, 4.35) days longer TTP.

The only other study on this topic did not focus on inter-pregnancy weight change but rather self-reported weight change in the year prior to pregnancy attempt. Among a preconception cohort of 629 women in the US, substantial weight loss in the past year (≥4 kg) was not associated with risk of pregnancy loss compared with maintaining a stable weight (RR = 0.99, 95% CI 0.56, 1.77)²¹. This null association persisted when the analysis was restricted to overweight and obese women (RR = 1.27, 95% CI 0.51, 3.14). However, substantial weight gain in the previous year was associated with a marginally significant increased risk of pregnancy loss (RR = 1.41, 95% CI 0.98, 2.02 for at least 4 kg gain) relative to maintaining a constant weight. These findings, taken together with the conclusions from the studies evaluating inter-pregnancy weight change, suggest the potential importance of preventing short-term weight gain, rather than solely focusing on weight loss, among reproductive-aged women in order to increase fecundity and decrease risk of pregnancy loss.