A Novel Preoperative Risk Score to Identify Patients at High Risk for Nonhome Discharge after Elective Open Abdominal Aortic Aneurysm Repair

Abstract

Background: Nonhome discharge (NHD) to a rehabilitation or skilled nursing facility after vascular surgery is poorly described despite its impact on patients. For home-dwelling patients undergoing elective surgery, the need for postoperative NHD can have meaningful implications on quality of life, long-term outcomes, and health-care spending. Understanding postsurgical NHD risk is essential to preoperative counseling and shared decision making. This is particularly true for the treatment of abdominal aortic aneurysms (AAAs) as the postoperative course can vary between open and endovascular surgery. We aimed to identify independent predictors of NHD following elective open abdominal aortic aneurysm repair (OAR), and to create a clinically useful preoperative risk score.

Methods: Elective OAR cases were queried from the Society for Vascular Surgery Vascular Quality Initiative from years 2013-2022. A risk score was created by splitting the data set into two-thirds for development and one-third for validation. A parsimonious stepwise hierarchical multivariable logistic regression controlling for hospital level variation was performed in the development dataset, and the beta-coefficients were used to assign points for a risk score. The score was then validated, and model performance assessed.

Results: Overall, 8,274 patients were included and 1,502 (18.2%) required NHD. At baseline, patients who required NHD were more likely to be ≥ 80 years old (23.6% vs. 6.5%), female (35.9% vs. 23.1%), not independently ambulatory (14.6% vs. 4.3%), anemic (24.4% vs. 13.9%), have chronic obstructive pulmonary disease (COPD, 41.6% vs. 30.7%), American Society of Anesthesiologists (ASA) class ≥4 (41.0% vs. 32.5%), and a supraceliac proximal clamp (9.8% vs. 5.7%; all P < 0.05). Multivariable analysis in the development group identified the following independent predictors of NHD: age ≥80 years, not independently ambulatory, proximal clamp location, hypogastric artery occlusion, anemia (Hb < 12 g/dL), COPD, female sex, hypertension, and ASA class ≥4. These were then used to create a 14-point risk score. Patients were stratified into three groups based upon their risk score: low risk (0-4 points; n = 4,966) with an NHD rate of 9.9%, moderate risk (5-6 points; n = 2,442) with an NHD rate of 25.5%, and high risk (≥7 points; n = 886) with an NHD rate of 44.6%. The risk score had good predictive ability with c-statistic = 0.73 for model development and c-statistic = 0.72 in the validation dataset.

Conclusions: This novel risk score can predict NHD following elective OAR using characteristics that can be identified preoperatively. Utilization of this score may allow for improved risk assessment, preoperative counseling, and shared decision making.