Preoperative frailty as the strongest predictor of postoperative adverse events burden in patients with operable non-small cell lung cancer: a retrospective analysis.

Abstract

Background: non-small cell lung cancer (NSCLC) is the most prevalent malignancy worldwide despite versatile screening programs. Therapy-related adverse events can be predicted with various tools including frailty. Frailty predictive power is less well studied in operable NSCLC.

Research design and methods: Retrospective analysis performed in NSCLC patients undergoing surgery in which ability of two preoperative frailty indexes mFI-5 and mFI-11 to predict the postoperative burden of adverse events was compared against conventional risk assessment tools such as American Society of Anesthesiologists (ASA), or the Revised Cardiac Risk Index (Lee score). Adverse events burden was categorized as any adverse event, any patient-related adverse event, any surgery-related adverse event, any administrative adverse event.

Results: In a sample of 98 patients with surgery for NSCLC, mFI-5 was the best predictor of adverse events burden (OR 36.34, p = 0.006 for any adverse event, 45.2, p = 0.002 for any patient- related adverse event, 23.1, p = 0.01 for any surgery-related adverse event, 12.26, p = 0.03 for any administrative adverse event.

Conclusions: Despite its sporadic use in this setting, preoperative frailty might be a more versatile predictor for postoperative adverse events in patients undergoing open surgery for NSCLC. Further studies with more complex approach for frailty are needed.