Frailty index as a predictive preoperative tool in the elder population undergoing major abdominal surgery: a prospective analysis of clinical utility

PURPOSE

The geriatric population has increased considerably in the last decades. Such increases come along with new challenges for surgical practitioners, who now face a risen number of frail patients in need of major operations. The value of frailty indexes in this setting has been discussed recently. This study assessed the modified Rockwood frailty index (mRFI) as a predictive tool for postoperative complications in older adults subjected to major abdominal operations and correlated it with other scores widely utilized for this purpose.

METHODS

We performed a prospective study utilizing the mRFI including all patients older than 65 years subjected to major abdominal surgery between May 2017 and May 2019 in a third-level academic center. A comparison between frail (mRFI >0.25) and non-frail patients (mRFI <0.25) was performed. We performed logistic regression to identify predictors of postoperative complications and 30-day mortality. We analyzed the correlation between mRFI and ACS-NSQIP, P-POSSUM, PMP, and Charlson score risk calculators.

RESULTS

One hundred forty patients were included in our study, of whom 49 (35%) were identified as frail. Frail patients demonstrated significantly prolonged hospital stay (p<.0001), ICU admission rates (p=0.004), hospital readmissions (p=0.007), and higher mortality rates (p=0.02). Our univariate analysis associated frailty (mRFI>0.25), ASA >III, increased age, and BMI with postoperative complications. In our multivariate analysis, frailty remained an independent predictor for postoperative complications (OR 6.38, 95% CI [2.45-16.58], p<0.001). Frailty was also associated with length of stay (LOS) regardless of the type of surgery (OR 3.35, 95% CI [0.37-6.33], p= 0.03). mRFI>0.25 demonstrated a sensitivity (Se) of 70% and specificity (Sp) 67% with area under the curve (AUC) 0.75 for perioperative complications, Se 69% and Sp 70% with AUC 0.74 for ICU admissions, and Se 83% and Sp 68% with AUC 0.83 for mortality.

CONCLUSION

Frail patients demonstrated significantly prolonged hospital stay, ICU admission rates, hospital readmissions, and higher mortality rates. mRFI is an independent predictor for perioperative complications with a Se of 70% and Sp 67% and AUC 0.75.