Document resource
Introduction Contrast-induced encephalopathy (CIE) is an uncommon but increasingly recognized complication of neurointerventional procedures, attributed to blood-brain barrier disruption and contrast neurotoxicity. Existing evidence derives largely from single-center series with heterogeneous diagnostic criteria and widely variable reported incidence. Population-level estimates of CIE incidence, independent clinical and predictors, and associated in-hospital burden remain poorly characterized.Methods We performed a retrospective cross-sectional analysis of the National Inpatient Sample (2019-2022), identifying adult hospitalizations involving neurointerventional procedures using ICD-10-PCS codes encompassing diagnostic angiography, endovascular thrombectomy, intracranial and carotid stenting, aneurysm embolization and flow diversion, angioplasty, and carotid endarterectomy. CIE was defined by co-occurring encephalopathy diagnosis codes and an adverse effect code for diagnostic agents (T50.8X5x). Multivariable complex-samples logistic regression identified independent predictors of CIE. Separate outcome models evaluated the association between CIE and in-hospital mortality, non-routine discharge, prolonged length of stay, mean length of stay, and total hospital charges, adjusting for demographics, comorbidities, and procedural variables. Procedure-specific CIE rates and unadjusted odds ratios were computed across neurointerventional categories.Results Among 170,717 unweighted neurointerventional hospitalizations (~853,585 weighted), 105 unweighted (525 weighted) met criteria for coded CIE, representing an incidence of 6.2 per 10,000. CIE hospitalizations were more likely to involve Black patients (25.3% vs 11.7%), non-elective admission (89.5% vs 58.2%), acute ischemic stroke presentation (52.4% vs 31.8%), and chronic kidney disease (37.1% vs 14.6%). Independent predictors of CIE included chronic kidney disease (aOR 3.10; 95% CI 2.01-4.80; P<.001), acute ischemic stroke (aOR 1.92; 95% CI 1.29-2.86; P=.001), congestive heart failure (aOR 1.85; 95% CI 1.15-2.98; P=.011), more than three coded procedures (aOR 1.79; 95% CI 1.20-2.66; P=.004), and overall procedure complexity (P=.009). Posterior circulation procedures carried the highest procedure-specific rates, including vertebral endovascular thrombectomy (36.6 per 10,000) and composite diagnostic angiography (10.0 per 10,000). After multivariable adjustment, CIE was associated with increased in-hospital mortality (aOR 2.62; 95% CI 1.46-4.70), non-routine discharge (aOR 5.73; 95% CI 3.09-10.65), and prolonged hospitalization greater than seven days (aOR 4.88; 95% CI 2.92-8.18), as well as longer adjusted mean length of stay (19.47 vs 10.93 days) and higher total charges ($433,918 vs $271,661; P<.001).Conclusion Coded CIE is rare but associated with significantly worse in-hospital outcomes across all measured endpoints. Chronic kidney disease and acute ischemic stroke were the most robust independent predictors, while posterior circulation procedures carried disproportionate procedure-specific risk. These findings provide the first nationally representative characterization of CIE epidemiology in a dedicated neurointerventional cohort and highlight patient- and procedure-level risk profiles that may inform risk stratification and peri-procedural monitoring strategies.Disclosures A. Terraciano: None. D. Vaishnav: None. K.P. Gupta: None. T. Vadset: None. A. Fortunell: None. D. Altschul: None.