# Assessing frailty to predict surgical risk: a comparative study of three tools in older non-cardiac surgery patients

## TL;DR

This study compares three tools for predicting surgical risk in older patients and finds that the Clinical Frailty Scale is the most effective when combined with another standard measure.

## Contribution

The study evaluates and compares the predictive accuracy of three frailty tools in older surgical patients and identifies the best-performing combination for risk prediction.

## Key findings

- Frailty assessed with the Clinical Frailty Scale (CFS) was the strongest independent predictor of postoperative complications.
- Combining the CFS with the American Society of Anesthesiologists (ASA) classification improved predictive accuracy more than other combinations.
- All three frailty tools were significantly associated with higher complication rates in older non-cardiac surgery patients.

## Abstract

Frailty is a significant predictor of adverse outcomes in older surgical patients. this study, we aimed to evaluate the feasibility and predictive ability of the Clinical Frailty Scale (CFS), Modified Frailty Index-11 (mFI-11), and FRAIL scale for postoperative complications in older Thai patients who underwent intermediate- to high-risk non-cardiac surgery.

This prospective cohort study included 637 older patients (aged ≥ 60 years) scheduled for intermediate- to high-risk elective non-cardiac surgery. Frailty was assessed preoperatively using the CFS, mFI-11, and FRAIL scale. Postoperative complications were defined as Clavien–Dindo classification ≥ 2. Predictive performance was analyzed using logistic regression and the area under the receiver operating characteristic curve (AUC).

The mean age of participants was 70.5 years (standard deviation 7.68), and 48% were male. Frailty was significantly associated with higher rates of postoperative complications across all tools: CFS (44.9% vs. 22.2%), mFI-11 (57.8% vs. 26.9%), and FRAIL scale (56.3% vs. 26.0%) (all p < 0.001). In multivariable logistic regression, the CFS was the only independent predictor (odds ratio 2.39, 95% confidence interval [CI]: 1.42–4.00, p < 0.001). Area under the curve (AUC) values were 0.635 (95% CI: 0.5902–0.6794) (mFI-11), 0.632 (95% CI: 0.5881–0.6756) (FRAIL scale), and 0.619 (95% CI: 0.5742–0.6637) (CFS), compared with 0.657 (95% CI: 0.6152–0.6988) of the American Society of Anesthesiologists (ASA) classification. Combining frailty tools with ASA improved predictive accuracy, with CFS + ASA exhibiting the highest AUC (0.704, 95% CI: 0.660–0.748).

Frailty assessed with the CFS, mFI-11, and FRAIL scale was associated with postoperative complications, with the CFS demonstrating the strongest independent predictive value. Incorporating frailty screening into preoperative evaluation, especially combined with ASA classification, can improve risk stratification and perioperative care.

TCTR20210706002.