The Administrative Risk Analysis Index (RAI-A) Does Not Predict Reintervention or Mortality Within 30 Days in Patients Who Undergo Elective Open or Endovascular Abdominal Aortic Aneurysm Repair.

OBJECTIVE Frailty scores are increasingly utilized to predict postoperative complications. The purpose of this study is to determine whether the administrative risk analysis index (RAI-A) can be used to predict reintervention or mortality within 30 days in patients who undergo elective open or endovascular abdominal aortic aneurysm (AAA) repair. METHODS The American College of Surgeons National Surgical Quality Improvement Program database was used to query data from elective open or endovascular aortic aneurysm repairs from 2011 to 2018. The administrative risk analysis index (RAI-A) score was calculated for each patient using two approaches (conservative versus liberal) due to discrepancies in NSQIP data categorization. Multivariable regression analysis was performed to determine whether there were statistical or clinical significance for incremental increases of RAI-A for both the open and endovascular repair group. Outcome measures were re-intervention or death within 30 days. RESULTS Data from 4106 and 11,733 patients who underwent open and endovascular repair, respectively, were included in the analysis. The number of reinterventions within 30 days was 9.1% (375 out of 4106 patients) in the open repair group and 4.0% (463 out of 11,685 patients) in the endovascular group. Thirty-day mortality was 4.7% (192 out of 4106 patients) in the open repair group, and 0.9% (109 out of 11,685 patients) in the endovascular group. In the conservative calculation of RAI-A scores, the open and endovascular repair groups had median RAI-A scores of 7 (mean 8.31) and 9 (mean 9.51), respectively. There was no significant association between RAI-A scores and outcome measures in either group. For predicting 30 d reintervention, the C statistic was 0.535 (OR 1.02) for the open repair group and 0.532 (OR 1.02) for endovascular repair. For predicting 30-day mortality, the C statistic was 0.626 (OR 1.07) in the open repair group and 0.701 (OR 1.09) in the endovascular repair group. In the liberal calculation of RAI-A scores, the open and endovascular repair groups had median RAI-A scores of 6 (mean 6.19) and 7 (mean 7.65), respectively. There was no significant association between RAI-A scores and outcome measures in either group. For predicting 30 d reintervention, the C statistic was 0.527 (OR 1.02) for open repair and 0.529 (OR 1.02) for endovascular repair. For predicting 30-day mortality, the C statistic was 0.625 (OR 1.07) in the open repair group and 0.695 (OR 1.08) in the endovascular repair group. CONCLUSIONS The RAI-A is not useful in predicting 30 d reintervention or mortality in patients who undergo elective open or endovascular AAA repair.