Predicting 30-day Mortality after Ruptured Abdominal Aortic Aneurysms: Validation of the Harborview Risk Score in a Single-Center Dutch Study Population

Abstract

Background: The Harborview Risk Score (HRS) was recently proposed as scoring tool to predict 30-day mortality in patients with ruptured abdominal aortic aneurysms (rAAAs). The HRS assigns 1 point for each of the following preoperative characteristics: age > 76 years, pH < 7.2, creatinine level > 2 mg/dL (> 176.8 μmol/L), and systolic blood pressure < 70 mm Hg, resulting in scores from 0 to 4. The 30-day mortality risk increases with every point. Primarily, we aimed to validate the HRS for the first time in a Dutch study population. A second objective was to identify other clinically relevant predictors for 30-day mortality after repair of rAAA.

Methods: Retrospective data from patients who underwent open repair or endovascular aortic repair for a rAAA between January 2009 and February 2022 were reviewed. Patients were grouped by HRS category (score 0-4). The 30-day mortality rate was calculated for each HRS category. Determinants for 30-day mortality were tested for significance and validated for HRS.

Results: In total, data from 135 patients were included. Open repair was performed in 95 patients and 40 patients underwent endovascular aortic repair. Univariate logistic regression identified pH < 7.2, systolic blood pressure < 70 mm Hg, female sex, performance status, and increase per HRS unit as significant determinants for 30-day mortality. After adjusting for sex and performance status in the multivariate analysis, the association between the HRS per-unit increase and 30-day mortality remained significant (odds ratio 2.532 (95% confidence interval: 1.437-4.461)). The 30-day mortality rate for HRS score 0 was 15.2%, while for HRS score 3 and 4 the mortality was 80% and 100% respectively.

Conclusions: The Harborview Risk Score was validated in this single-center Dutch population. Results were concordant with data presented in earlier studies. Therefore, the HRS seems accurate and accessible as preoperative tool. For now, the HRS should guide as an insightful tool to indicate the chances of postoperative mortality during the preoperative conversations in the emergency room, rather than as a decision-making tool whether to operate or not. Our results suggest that female sex and performance status are also relevant predictors that should be assessed in other populations to improve preoperative scoring systems.