1078. Effects of Neurovascular Comorbidities and In-Hospital Complications on Procedural Management of Unruptured Aneurysm

Authors: Pious D Patel; Silky Chotai, MD; Campbell Liles, BS; Heidi Chen, PhD; Shilin Zhao, PhD; Chevis Shannon; Matthew Fusco, MD; Rohan Chitale, MD (Nashville, TN)

Introduction:

Neurovascular comorbidities and complications can significantly affect aneurysm prognosis. For surgical and endovascular management of unruptured aneurysms, this study investigates the effects of comorbidities on in-hospital complications, and of comorbidity-adjusted complications on outcomes, in-hospital mortality, and length of stay (LOS).

Methods:

Unruptured aneurysm patients were identified unweighted from the 2012-2015 National Inpatient Sample (NIS) using ICD-9 procedural/diagnostic codes for aneurysm treatment (Clipping 395.1; Coiling 395.2, 397.2, 397.9) in non-subarachnoid hemorrhage patients (430). Patients with head trauma, AVM/AVF, and LOS less than one day were excluded. Variables were comorbidities (measured with the neurovascular comorbidities index (NCI)), in-hospital complications (divided into seizures, medical complications, and surgical complications), and sociodemographic factors. Outcomes were modified Rankin scale equivalent outcome (NIS-SOM), in-hospital mortality, and LOS. Multivariable logistic regression models were built for NIS-SOM and in-hospital mortality. Multivariable linear regression models were built for LOS.

Results:

7,398 patients were included (median age 58, 75% female, 66% White, 43% private insurance). NCI was higher in patients with each type of in-hospital complication (P<.001) and any in-hospital complication (P<.001). In multivariate regressions, NCI was significantly associated with poor NIS-SOM (P<.001), increased in-hospital mortality (P<.001), and longer LOS (P<.001). In-hospital complications were significantly associated with poor NIS-SOM (P<.001) and longer LOS (P<.001), but not in-hospital mortality (P=.71). The presence of a surgical complication strengthened the association between NCI and NIS-SOM. The presence of any in-hospital complication strengthened the association between NCI and LOS (P<.001).

Conclusion:

Among patients undergoing procedural management of unruptured aneurysms, we found that comorbidities are associated with higher occurrence of in-hospital complications. Comorbidities are strongly associated with poor outcomes, and the presence of in-hospital complications magnifies these effects for NIS-SOM outcome and LOS. The dynamic relationship between comorbidities and in-hospital complications should be further defined for effective preoperative risk discussions in unruptured aneurysm patients.