Frailty is a Poor Predictor of Postoperative Morbidity and Mortality After Ruptured Abdominal Aortic Aneurysm

Abstract

Frailty has gained prominence as a predictor of postoperative outcomes across a number of surgical specialties, vascular surgery included. The role of frailty is less defined in the acute surgical setting. We assessed the prognostic value of frailty for patients undergoing surgery for a ruptured abdominal aortic aneurysm (rAAA). A single-institution retrospective chart review of all patients undergoing surgical intervention for rAAA between January 1, 2011 and November 27, 2019 was performed. Frailty was assessed for each patient using the modified frailty index (mFI), a validated frailty metric based on the Canadian Study of Health and Aging. Frailty was defined as an mFI ≥0.27. The performance of the mFI was compared to that of the Vascular Study Group of New England (VSGNE) rAAA mortality risk score. Chi square, Fisher's exact, and t tests, were used to evaluate for associations between frailty and in-hospital outcomes. Univariate and multivariate logistic regression were used to obtain odds ratios for in-hospital mortality. A receiver operating characteristic (ROC) curve was generated to compare the predictive value of the mFI and VSGNE score for in-hospital mortality. Sixty patients were identified during the study period with an in-hospital mortality rate of 37%. Twenty-one patients were deemed frail by mFI metric and included all patients with known myocardial infarction, stroke with a neurologic deficit or dependent functional status, however the mortality rate did not differ significantly based on frailty status (33% nonfrail vs. 43% frail, P= 0.47). Frailty status was not significantly different for patients with acute kidney injury (10% nonfrail vs. 10% frail), prolonged intubation (13% vs. 5%), abdominal compartment syndrome (8% vs. 10%), and Type I or Type III endoleak (8% vs. 19%). On multivariate analysis controlling for systolic blood pressure <70 mm Hg, suprarenal aortic control, and creatinine >2.0 mg/dl, the mFI produced an adjusted odds ratio (aOR) of 0.7 (95% confidence interval [CI]: 0.2-3.0). The ROC curve for the mFI produced an area under the curve (AUC) of 0.55 (P= 0.55) for in-hospital mortality while that of the VSGNE score produced an AUC of 0.69 (P= 0.02). The mFI did not significantly predict in-hospital outcomes after rAAA in this cohort. This suggests that the baseline health status of a patient with rAAA may play a less significant role in their postoperative prognosis than their acuity on presentation.