Prediction of Ongoing Pregnancy in an IVF Cycle Based on Clinical, Sonographic and Endocrine Parameters at Baseline

Abstract

To identify those baseline variables prior to stimulation that independently predict ongoing pregnancy in women undergoing IVF treatment. Analysis of a large, randomized, open-label, assessor-blind, multicenter, multinational trial (MERIT) comparing two gonadotropin preparations in IVF cycles with the long GnRH agonist protocol. Ongoing pregnancy data were collected from 731 ovulatory women with tubal factor, mild male factor, endometriosis stage I/II or unexplained infertility. Polycystic ovarian syndrome and poor response in previous cycles were exclusion criteria. All were downregulated with triptorelin 0.1 mg SC daily from the mid-luteal phase and randomised to either HP-hMG (MENOPUR, Ferring Pharmaceuticals) (n=363) or recombinant FSH (GONAL-F, Serono) (n=368). The starting dose was 225 IU SC for the first five days and thereafter dose adjustments of 75 IU were allowed not more frequently than every four days. Recombinant hCG 250 μg SC was administered when 3 or more follicles of 17 mm or greater were observed. Transfer of 1-2 embryos of pre-defined minimum quality criteria was done on Day 3. Ongoing pregnancy was defined as at least one viable fetus 10-11 weeks after embryo transfer. An analysis was made on the association between ongoing pregnancy and demographic (age, BMI, duration of menstrual cycle), infertility history (duration, previous treatments, previous pregnancies, cause of infertility), sonographic (antral follicles, ovarian volume, endometrial thickness) and endocrine (LH, FSH, progesterone, androstenedione, total testosterone, SHBG, FAI) data after downregulation, immediately before start of stimulation (day 1). The predictive value of each variable was summarised as an odds ratio (OR) of ongoing pregnancy with 95% confidence interval (CI). Significance testing was based on the Wald test. All variables with an indication of influence (regarded as a p-value < 0.1) in a univariate analysis were included in the multivariate analysis. The multivariate model was reduced stepwise and variables with p< 0.1 were kept in the final model. All results relate to the multivariate analysis. In the analysis of demographics and infertility history, age as well as primary cause and duration of infertility were significant predictors of ongoing pregnancy. Among the assessments made on day 1, endometrial thickness, serum total testosterone and duration of GnRH agonist before start of stimulation were significant predictors. Taking all variables before starting stimulation into account, age (p<0.05), primary cause of infertility (p<0.05), total testosterone (p<0.05) and endometrial thickness (p=0.07) were significant independent predictors. Women ≤ 29 and 30-34 years had increased chance of ongoing pregnancy compared to women 35-37 years of age, with an odds ratio of 2.71 (95% CI: 1.46-5.32) and 2.40 (95% CI: 1.33-4.62), respectively. Mild male or tubal factor resulted in decreased chance of ongoing pregnancy with an odds ratio of 0.40 (95% CI: 0.20-0.75) and 0.59 (95% CI: 0.40-0.88), respectively, compared to unexplained infertility. The odds were 0.54 (95% CI: 0.31-0.91) for total testosterone and 0.87 (95% CI: 0.73-1.01) for endometrial thickness. Age below 35 years, unexplained infertility and low total testosterone prior to starting stimulation are favorable prognostic factors for ongoing pregnancy in IVF cycles.