Risk of ovarian torsion in patients with ovarian hyperstimulation syndrome

Abstract

Ovarian hyperstimulation syndrome (OHSS) is a serious complication of assisted reproductive technology (ART) for which hospital admission is required in severe cases to prevent life-threatening complications or for treatment of ovarian torsion. In OHSS, high ovarian volumes increase the risk of torsion, however population-based data evaluating incidence and complications of OHSS-related ovarian torsion are lacking. To examine the incidence of OHSS-related ovarian torsion and severe medical complications as well as to profile the characteristics of and interventions required for women with OHSS and ovarian torsion. A retrospective analysis of the Nationwide Inpatient Sample between 2001-2015 was performed. This database represents over 97% of the United States population. Hospital discharge records of women <50 years of age with OHSS and no other ovarian pathology were extracted and weighted. Women with and without ovarian torsion were compared. A classification-tree model was utilized to identify patterns of patient characteristics associated with ovarian torsion. There were 14,623 women admitted with OHSS during the study period, and 313 (2.1%, 95% confidence interval [CI] 1.9-2.4) had ovarian torsion. Of all women admitted with OHSS, 1330 (9.1%) women were pregnant, and pregnant women had a greater risk of torsion compared to non-pregnant women (3.2% vs. 2.0%, OR 1.6 95%CI 1.1-2.2, P=0.008). Severe medical complications, including ascites, electrolyte imbalance, hemorrhage, sepsis/shock, venous thromboembolism, stroke, myocardial infarction, pneumonia, pleural effusion, respiratory failure, intestinal obstruction, or kidney injury occurred in 8,322 (56.9%) women with OHSS overall. Women admitted with OHSS-related torsion had lower rates of medical complications (31.6% vs. 57.5%, OR 0.34 95%CI 0.25-0.44, P<0.001) and more rarely required paracentesis (5.8% vs. 31.1%, OR 0.14 95% CI 0.08-0.22, P<0.001). However, 11.5% of women with ovarian torsion underwent oophorectomy. There were no fatalities in the ovarian torsion group, while 9 (0.06%) deaths occurred among all OHSS patients. Median total charges for the index admission were higher in those with torsion compared to those without ($18,650 versus $10,943, net difference $7,707, P<0.001). Rates of ovarian torsion were lower in urban teaching centers (1.8%) compared to urban-non-teaching (2.6%) or rural (4.3%) hospitals (P<0.001), however oophorectomy rates were not different (P=0.123). Oophorectomy rates differed by region and were lowest in the West (0%) and highest in the Northeast (25%) (P<0.001). On classification tree analysis, the highest incidence of ovarian torsion (10%) was in women in urban non-teaching or rural hospitals located in the South after 2012. Of women admitted with OHSS, one in fifty had ovarian torsion, and oophorectomy, a devastating outcome for fertility-seeking patients, was performed in over 10%. Torsion occurred more frequently in pregnant patients and was associated with fewer medical complications, supporting the hypothesis that torsion in OHSS results from increased ovarian volumes rather than increased capillary permeability. Rates of torsion and oophorectomy also differed based on hospital teaching status and geographic region, respectively. As outpatient management of non-severe OHSS is increasing, there should be heightened awareness regarding the risk of torsion and prompt medical evaluation provided if torsion is suspected.