recently published a meta-analysis investigating the role of maternal obesity on the association between polycystic ovary syndrome (PCOS) and fetal and maternal outcomes. One of the outcomes was gestational diabetes mellitus (GDM), and they included 39 studies that recruited a total of 11 565 women with PCOS and 177 296 women without PCOS and compared them to assess the role of PCOS per se on the development of GDM. The fundamental question that the authors attempted to answer was if there was a higher risk for women with PCOS in terms of GDM and if this was independent of other GDM risk factors. The authors used univariate meta-regression as well as subgroup analysis of BMI matched and non-matched studies to conclude that indeed there was an independent predictive role of PCOS on the development of GDM. 1Our research team has been looking at incident GDM among high risk pregnant women, and as such, we read these findings with great interest. However, an important question remained unanswered, which is that, is obesity or insulin resistance the main driver of GDM in women with PCOS or not? We felt that the data provided by the authors could easily answer this question if we ran a non-linear meta-regression on the data from the 39 studies that were used in the analysis to explore the role of PCOS on GDM. The hypothesis was that this is primarily driven by insulin resistance, and therefore, as average BMI across each study increases, the PCOS effect should diminish since the insulin resistance differential should narrow.We ran a robust error (robust standard errors) inverse variance weighted meta-regression with log OR as the outcome and the overall mean BMI in each study as the covariate of interest. 2 We allowed for non-linearity using restricted cubic splines with three knots. Resultsshow that lean women with PCOS have 6 times the odds of developing GDM compared with lean women without PCOS, and as the BMI increases, the odds ratio diminishes which is consistent with our hypothesis that the effect is mediated through insulin resistance. In other words, women who suffer obesity without PCOS are not really different from women with PCOS in terms of risk for developing GDM (Figure 1). This is because lean women with PCOS are already insulin resistant, while obesity (in women without PCOS) drives the insulin resistance.The implications are that ultimately the risk factor for GDM is chronic insulin resistance in both women with and without PCOS. 3 This finding was not obvious because the authors did not carry out a comparison of women with and without PCOS by populations with varying obesity levels. Matched BMI analysis that the authors used to support their findings is in fact not appropriate because matching eliminates the effect of the factor completely. Meta regression that was carried out to negate the effect of BMI was not able to identify this effect as they used the BMI ratio between the groups as the predictor.Given that obesity drives insulin resistance and that lean women with PCOS are insulin resistant, t...