A novel surgical risk predictor combining frailty and hypoalbuminemia - a cohort study of 9.8 million patients from the ACS-NSQIP database.

INTRODUCTION

The functional decline seen in frail patients is associated with significant morbidity and mortality. The modified frailty index 5 (mFI-5) score is an accepted risk predictor score in surgery. Hypoalbuminemia has been correlated with poor postoperative outcomes.There exists, however, a gap in the literature regarding the combined assessment of frailty and hypoalbuminemia and the predictive power of this combined assessment. This retrospective cohort study aimed to investigate the association of preoperative albumin and frailty, as assessed with the mFI-5 score, and its ability to predict surgical outcomes.

METHODS

We queried the ACS-NSQIP database (2008-2021) to identify all surgical patients. Perioperative data, including demographics and preoperative laboratory values, including albumin, were collected. The predictive power of the mFI-5 and hypoalbuminemia (Alb) independently and in combination (mFI-5+Alb), was assessed using multivariable linear and logistic regression models 30-day outcomes were assessed including mortality, length of hospital stay, reoperation, medical and surgical complications, and discharge destination.

RESULTS

A total of 9 782 973 patients were identified, of whom 4 927 520 (50.4%) were nonfrail (mFI=0), 3 266 636 had a frailty score of 1 (33.4%), 1 373 968 a score of 2 (14.0%), 188 821 a score of 3 (1.9%), and 26 006 a score greater or equal to 4 (0.3%). Albumin levels were available for 4 570 473 patients (46.7%), of whom 848 315 (18.6%) had hypoalbuminemia. The combined assessment (mFI-5+Alb) was found to be a more accurate risk predictor than each factor independently for all outcomes. A weak negative correlation between serum albumin levels and mFI scores was established (Spearman R : -0.2; <0.0001).

CONCLUSIONS

Combined assessment of frailty and albumin was the strongest risk predictor. Therefore, for patients undergoing surgery, we recommend consideration of both serum albumin and frailty in order to optimally determine perioperative planning, including multidisciplinary care mobilization and prehabilitation and posthabilitation.