An Analysis of the Modified Five-Item Frailty Index for Predicting Complications following Free Flap Breast Reconstruction.

[BACKGROUND] The modified five-item frailty index is a validated and effective tool for assessing risk in surgical candidates. The authors sought to compare the predictive ability of the modified five-item frailty index to established risk factors for complications in free flap breast reconstruction.

[METHODS] The 2012 to 2018 American College of Surgeons National Surgical Quality Improvement Program database was queried for free flap breast reconstructive procedures. Univariate and multivariate regression analysis models were used to assess how modified five-item frailty index and factors commonly used to risk stratify (age, body mass index, American Society of Anesthesiologists classification, and history of smoking) were associated with complications.

[RESULTS] Of the total 10,550 cases, 24.1 percent experienced complications. A high modified five-item frailty index score is associated with a higher overall rate of postoperative complications (p < 0.001). This significant trend was demonstrated in both surgical (p < 0.001) and medical (p < 0.001) complications. When controlling for other risk factors commonly used for risk stratification such as age, body mass index, American Society of Anesthesiologists classification, and history of smoking, the modified five-item frailty index was significantly associated with medical (OR, 1.75; 95 percent CI, 1.37 to 2.22; p = 0.001) and any complications (OR, 1.58; 95 percent CI, 1.29 to 1.93; p < 0.001) and had the largest effect size. Assessing for specific complications, the modified five-item frailty index is the strongest independent predictor of reoperation (OR, 1.41; 95 percent CI, 1.08 to 1.81; p = 0.009).

[CONCLUSION] The modified five-item frailty index is a useful predictor of postoperative outcomes in patients undergoing free flap breast reconstruction when compared to other historically considered risk factors for surgical complications.

[CLINICAL QUESTION/LEVEL OF EVIDENCE] Risk, III.