RESUMEN
OBJECTIVE: To analyse populational trends and perioperative complications following conservative surgery versus oophorectomy in women <50 years of age with ovarian torsion. DESIGN: Population-based retrospective observational study. SETTING: Nationwide Inpatient Sample in the USA (2001-2015). POPULATION: In all, 89 177 ovarian torsions including 20 597 (23.1%) conservative surgeries and 68 580 (76.9%) oophorectomies. METHODS: (1) Trend analysis to assess utilisation of conservative surgery over time, (2) multivariable binary logistic regression to identify independent factors associated with conservative surgery and (3) inverse probability of treatment weighting with a generalised estimating equation to analyze perioperative complications. MAIN OUTCOME MEASURES: Trends, characteristics and complications related to conservative surgery. RESULTS: Performance of conservative surgery increased from 18.9 to 25.1% between 2001 and 2015 (32.8% relative increase, P = 0.001) but decreased steadily after age 15, and sharply declined after age 35 (P < 0.001). On multivariable analysis, younger age exhibited the largest effect size for conservative surgery among the independent factors (adjusted odds ratios 3.39-7.96, P < 0.001). In the weighted model, conservative surgery was associated with an approximately 30% decreased risk of perioperative complications overall (10.0% versus 13.6%, odds ratio 0.73, 95% confidence interval 0.62-0.85, P < 0.001) and was not associated with venous thromboembolism (0.2 versus 0.3%, P = 0.457) or sepsis (0.4 versus 0.3%, P = 0.638). CONCLUSION: There has been an increasing utilisation of conservative surgery for ovarian torsion in the USA in recent years. Our study suggests that conservative surgery for ovarian torsion may not be associated with increased perioperative complications. TWEETABLE ABSTRACT: Conservative surgery for ovarian torsion may not be associated with increased perioperative complications.