Diagnosis of mental illness does not affect postoperative outcomes in patients undergoing endovascular aortic aneurysm repair in the VA healthcare system

Abstract

ObjectiveMental illness can be a debilitating chronic disease associated with a higher likelihood of preexisting medical comorbidities and postoperative morbidity and mortality. Given the relative prevalence of mental health disorders among the veteran population, we sought to examine postoperative outcomes in patients undergoing endovascular aortic aneurysm repair (EVAR). MethodsRetrospective review of a single institution Veterans Administration Hospital operative database was used to identify patients who underwent EVAR from January 2010 to December 2021. Patients’ demographics, comorbidities, medications, and intraoperative variables were collected. In addition, mental illness status was evaluated to stratify patients based on preexisting anxiety, depression, posttraumatic stress disorder, substance abuse disorder, or major psychiatric illness. The study’s primary outcomes were postoperative complications, mortality, and follow-up rates. Secondary outcomes included hospital length of stay, readmission rates, and intervention rates. ResultsA total of 241 patients underwent infrarenal EVARs at our institution. One hundred forty patients (58.1%) were diagnosed with mental illness, whereas 101 (41.9%) had no prior diagnosis of mental illness. Of the 241 patients, 65.7% had a history of substance abuse disorder, 38.6% depression, 29.3% post-traumatic stress disorder, 19.3% anxiety, and 3.6% major psychiatric illness. There was no statistical difference in the number of medical comorbidities, race, smoking status, or medications compared with patients without mental illness. We found no statistical difference in access type, wound infection rates, hypogastric coiling, estimated blood loss, and operating time. χ2 analysis demonstrated a statistically significant lower overall postoperative complication rate (28.6% vs 32.7%; P = .05) and decreased loss to follow-up (8.6% vs 15.8%; P = .05) among patients with a preexisting mental illness diagnosis. There were no statistically significant differences in readmission rate, length of stay, or 30-day mortality. When stratified by type of mental illness, binary logistic regression demonstrated no statistically significant differences in primary outcomes of postoperative complications, readmission rates, loss to follow-up, and 1-year mortality. Cox proportional hazards modeling demonstrated no significant difference in cumulative survival in patients diagnosed with a mental illness (0.56; 95% confidence interval, 0.29-0.107; P = .08). ConclusionsThere was no association between the presence of a prior mental health diagnosis and adverse outcomes following EVAR. Preceding mental illness did not correlate with an increased rate of complications, readmission, length of stay, or 30-day mortality in a veteran population. Lower loss to follow-up rates in patients with mental illness may reflect overall Veterans Health Administration expansion in resources and surveillance of these at-risk individuals. Further research is needed to assess the association between postoperative outcomes and mental illness.