Ovarian hyperstimulation syndrome: pathophysiology, staging, prediction and prevention

Objective To identify, appraise and summarize the current evidence regarding the pathophysiology, staging, prediction and prevention of ovarian hyperstimulation syndrome (OHSS). Methods Two comprehensive systematic reviews were carried out: one examined methods of predicting either high ovarian response or OHSS and the other examined interventions aimed at reducing the occurrence of OHSS. Additionally, we describe the related pathophysiology and staging criteria. Results Seven studies examining methods of predicting OHSS and eight more examining methods of predicting high ovarian response to controlled ovarian stimulation were included. Current evidence shows that the best methods of predicting high response are antral follicle count and anti-Mullerian hormone levels, and that a high ovarian response (examined by the number of large follicles, estradiol concentration or the number of retrieved oocytes) is the best method of predicting the occurrence of OHSS. Ninety-seven randomized controlled trials examining the effect of several interventions for reducing the occurrence of OHSS were included. There was high-quality evidence that replacing human chorionic gonadotropin by gonadotropin-releasing hormone agonists or recombinant luteinizing hormone, and moderate-quality evidence that antagonist protocols, dopamine agonists and mild stimulation, reduce the occurrence of OHSS. The evidence for the effect of the other interventions was of low/very low quality. Additionally, we identified and described 12 different staging criteria. Conclusions There are useful predictive tools and several preventive interventions aimed at reducing the occurrence of OHSS. Acknowledging and understanding them are of crucial importance for planning the treatment of, and, ultimately, eliminating, OHSS while maintaining high pregnancy rates. Copyright (C) 2014 ISUOG. Published by John Wiley & Sons Ltd.