Comparable outcomes for non-operative intracranial hemorrhage at Level III vs Level I/II trauma centers.

Abstract

Most patients with isolated, closed blunt intracranial hemorrhage (ICH) do not require neurosurgical intervention. However, outcome data for those managed entirely at Level III trauma centers remain incompletely characterized. Using the 2017-2020 Trauma Quality Improvement Program (TQIP), we compared clinical outcomes for non-operative ICH patients treated at Level I/II versus Level III trauma centers. The primary endpoint was composite in-hospital adverse event counts per patient, modeled using negative binomial regression and reported as adjusted incidence rate ratios (aIRRs). Secondary analyses included patients with ISS > 15, ICU admission by trauma center level, predictors of ICU admission, and unplanned ICU transfers. To address triage selection bias, we also compared time to ED discharge in a cohort of patients who met the same inclusion and exclusion criteria and were transferred from a Level III trauma center. A total of 24,132 patients were included after complete-case exclusion (Level I/II: 22,846; Level III: 1286). Composite adverse event counts were 0.067 events per patient at Level I/II centers and 0.042 at Level III centers. After adjustment, there was no statistically significant difference in composite adverse event counts between groups (aIRR 0.82, 95% CI 0.56-1.18). Mortality was lower at Level III centers (aOR 0.12, 95% CI 0.04-0.35). Composite adverse event count outcomes were similar in the ISS > 15 subgroup (aIRR 0.65, 95% CI 0.35-1.11). Among ICU admissions, the association between ICU admission and composite adverse event counts differed by trauma center level, with a lower relative effect of ICU admission at Level III centers compared to Level I/II centers (interaction aIRR 0.54, 95% CI 0.32-0.92). Exploratory analysis showed, after adjustment, significant factors associated with ICU admission were midline shift < 5 mm (aOR 1.44, 95% CI 1.29-1.61), lower GCS (aOR 0.84, 95% CI 0.82-0.86), alcohol use disorder (aOR 1.40, 95% CI 1.22-1.62), and anticoagulant use (aOR 1.41, 95% CI 1.27-1.59). Significant unplanned ICU admission included alcohol use disorder (aOR 5.04, 95% CI 2.74-8.90) and anticoagulation use (aOR 3.38, 95% CI 1.94-5.72). Following ED arrival, 94.2% of patients at Level III centers were admitted within 6 h, and 97.9% of transferred patients were transferred within 6 h (aOR 2.21, 95% CI 1.30-3.90). Selected patients with non-operative ICH can be managed at Level III trauma centers without increased complications or the need for neurosurgical intervention.