PURPOSE: Investigating whether pre-ovulatory follicular fluid (FF) levels of selected proteins differ between women who do or do not develop severe ovarian hyperstimulation syndrome (OHSS) and evaluate whether they potentially could guide a "freeze-all" strategy.

METHODS: FF was collected during a randomized controlled trial comparing OHSS in antagonist versus agonist protocol including 1050 women in their first assisted reproductive technology (ART) cycle during year 2009-2013. The present sub-study is a matched case-control study comparing FF levels of soluble urokinase plasminogen activator receptor (suPAR), C-reactive protein, placental growth factor, vascular endothelial growth factor, and angiopoietins 1 and 2 in OHSS cases (n = 25, severe OHSS, and ≥ 15 oocytes), high-risk controls (n = 25, no OHSS, and ≥ 15 oocytes), and low-risk controls (n = 25, no OHSS, and 5-8 oocytes).

RESULTS: FF level of suPAR differed significantly between the three groups (p = 0.018) with mean (SD) levels of 2.3 (0.4) μg/L, 2.6 (0.8) μg/L, and 2.8 (0.6) μg/L in OHSS cases, high-risk controls, and low-risk controls, respectively. Receiver operating characteristic curve analysis demonstrated that suPAR levels could predict severe OHSS (AUC 0.678; 95% CI 0.553-0.803) with a sensitivity of 64% and a specificity of 66%. None of the other investigated proteins differed between the three groups or between OHSS cases and combined controls.

CONCLUSION: The pre-ovulatory FF level of suPAR was significantly lower in women developing severe OHSS, indicating that the plasminogen activator system could be involved in the pathophysiology of OHSS. However, suPAR did not provide a satisfying predictive value for the prediction of OHSS.