Predictors of Pregnancy in Women With Polycystic Ovary Syndrome

Abstract

Two professional medical societies used expert opinion as the basis for their recommendation that current first-line treatment of polycystic ovary syndrome (PCOS)-related infertility should be ovulation induction with the selective estrogen receptor modulator, clomiphene citrate (CC), for up to 6 ovulatory cycles, with the addition of the insulin sensitizer metformin only for glucose intolerance. This prospective, multicenter, randomized clinical trial was designed to determine which baseline patient characteristics among women with PCOS were predictive of success of fertility treatment in 4 clinically relevant prognostic models of ovulation, conception, pregnancy, and live birth. Live birth was the primary study outcome. Between 2002 and 2004, ovulation was induced in 626 infertile women with PCOS by either CC plus placebo (n = 209), metformin plus placebo (n = 208), or combination CC plus metformin (n = 209). The factors that were significant for prediction of treatment success in all 4 conception models were body mass index, baseline free androgen index, proinsulin level, and duration of attempting conception. History of a prior pregnancy loss was predictive of treatment success in the models for ovulation and conception, but was not predictive of pregnancy or live birth. A low hirsutism score (<8) was not predictive of ovulation success, but was predictive of conception, pregnancy, and live birth. Age had divergent effects on ovulation and pregnancy outcomes. Age 34 or younger was a predictive factor for a successful pregnancy and live birth, whereas age greater than 34 years was predictive only of successful ovulation. Smoking history was not a significant predictor. These findings may be useful to counsel women with PCOS on their likelihood for live birth and to select infertility therapy.