A Simplified Frailty Index to Predict Outcomes After Radical Cystectomy

Background: Traditional surgical risk indices are notoriously inaccurate, especially for the elderly, so there has been a trend to evaluate frailty instead. Objective: To describe a simplified five-item frailty index for evaluating radical cystectomy outcomes and compare its predictive ability with other risk assessment tools for a total patient cohort and a subgroup of patients aged ≥65. yr. Design, setting, and participants: The National Surgical Quality Improvement Program (NSQIP) database from 2007 to 2015 was queried for radical cystectomy cases for bladder cancer. A simplified five-item frailty index (sFI) was created based on previously described measures of frailty. Outcome measurement and statistical analysis: Correlation of the sFI with postoperative outcomes was investigated using multivariate logistic regression analyses. The predictive ability of the sFI was compared to the extended 11-item frailty index, NSQIP risk calculator, and the American Society of Anesthesiologists (ASA) score by comparing the area under the receiver operating characteristic curve. Subgroup analysis was performed for those aged ≥65 yr to evaluate the accuracy of the index in the older age group, for which traditional risk indices are notoriously inaccurate. Results and limitations: Of 5516 cases identified, 10.7% experienced a Clavien grade ≥3 complication within 30 d of surgery. Individuals with an sFI of 3+ had a greater likelihood of experiencing a major complication (odds ratio 3.22, 95% confidence interval 2.01-5.17). The sFI outperformed the ASA score in predicting major complications and had a similar predictive ability to the more complex 11-item index and NSQIP risk calculator. There was also a significant correlation between the sFI and discharge destination. These relationships were consistent in the subgroup of patients aged ≥65. yr. The study is limited by the database, which only records 30-d outcomes and does not allow evaluation of long-term sequelae. Conclusions: There is a strong correlation between frailty assessed via the sFI and radical cystectomy outcomes, including in the elderly. This tool can be used in the clinical setting to counsel patients and aid decision-making. Patient summary: This study demonstrated that a simple five-item frailty index can be used to assess preoperative risk in patients undergoing radical cystectomy. Current indices used to assess surgical risk in clinical practice are notoriously inaccurate and either rely on subjective judgements or are too complex for routine use. This paper describes a simple, five-item frailty index that is correlated with radical cystectomy outcomes.