Quantifying the Added Value of 2 Common Frailty Measures for Predicting Adverse Outcomes After Elective Hysterectomy.

Abstract

Although frailty is predictive of adverse outcomes in predominantly male general and orthopedic surgical populations, the utility of American College of Surgeons National Surgical Quality Improvement Program-based frailty measurement for hysterectomy is unclear. The objective of this study was to measure the added contribution of the modified frailty index (mFI) and Risk Analysis Index (RAI) for predicting adverse outcomes after hysterectomy. A secondary analysis of the 2011 to 2014 American College of Surgeons National Surgical Quality Improvement database was conducted. Benign elective hysterectomy by any route was included. The primary outcome was readmission within 30 days of surgery. Secondary outcomes were major (Clavien-Dindo grade ≥3) and minor (grade 1-2) complications. The fraction of new prognostic information attributable to each frailty measure was estimated by the ratio of model likelihood-ratio χ 2 values compared with a baseline model, including American Society of Anesthesiologists classification, age, body mass index (BMI), smoking status, and surgical route. Among 70,649 cases, 3.0% (95% confidence interval [CI], 2.9-3.1) were readmitted within 30 days and 2.8% (95% CI, 2.7-2.9) and 5.2% (95% CI, 5.0-5.4) had major and minor complications, respectively. The RAI provided a greater fraction of new prognostic information than the mFI when predicting readmission (4.8 vs 2.7%) and major complications (4.8 vs 2.3%). Interaction analysis showed a stronger association of frailty and outcomes among individuals undergoing abdominal hysterectomy and with BMI of 40 of higher or less than 20. The RAI and mFI provided modest improvement in the ability to predict adverse outcomes, which limits its clinical utility. Surgeons may consider selective utilization among those individuals undergoing abdominal hysterectomy or with BMI of 40 of higher or less than 20.