Abstract
Background Researchers have shown that extracranial (EC) injuries can influence outcomes in patients with intracranial hemorrhage (ICH). Large data sets have noted that the presence of EC injury can have detrimental effects on ICH. However, the effects are not well quantified. EC injury has been historically treated as a categorical variable. Only one study besides the current one has attempted to calculate the degree of EC injury in affecting outcomes in ICH. We calculated a novel systemic variable for EC injury as the EC Injury Severity Score (EC-ISS) and aimed to assess the impact of EC injuries on ICH patient outcomes in blunt trauma. Objective Our primary objective was to examine the impact of EC injury on six undesirable outcomes in blunt ICH trauma: (1) mortality, (2) ICU stay ≥2 days, (3) hospital stay >5 days, (4) adverse outcome (AO; mortality, ICU stay ≥2 days, or hospital stay >5 days), (5) no commands at hospital discharge, and (6) no commands at three months. Secondary objectives included comparing EC-ISS and ICH Abbreviated Injury Scale scores (AIS) with ISS. Methods This was a retrospective analysis of adult trauma patients with ICH admitted to a level I trauma center in northeast Ohio from January 21 to July 21 during each year in 2018, 2019, and 2020 (the parent data set was obtained from a prior COVID-19 publication). The population had computed tomography (CT)-confirmed ICH, admission Glasgow Coma Scale (GCS) score of 3-15, and blunt trauma. CT ICH mass effect scores were rated as one each for lateral ventricular compression, basal cistern compression, and midline shift (0-3). The ICH CT score was mass effect score plus subarachnoid hemorrhage (SAH; 0-5). ICH AIS, ISS, and admission hypotension (systolic blood pressure <100 mmHg) were from the trauma registry (TR). EC-ISS was ISS minus the ICH AIS score squared. GCS deficit was 15 minus GCS. AOs were hospital death, ICU stay ≥2 days, or hospital stay >5 days. Admission glucose was from the electronic medical record. Results The study included 436 patients. Hospital death was independently associated with ICH mass effect (p<0.0001), age (p=0.0002), EC-ISS (p=0.0002), and ICH AIS (p=0.0010) but not with ISS ≥16 (p=0.8243). Intensive care days had independent associations with GCS (p<0.0001), admission glucose (p = 0.0003), CT score (p<0.0001), EC-ISS (p<0.0001), ICH AIS (p<0.0001), and hypotension (p = 0.0493), but not with ISS ≥16 (p = 0.5517). Hospital days had independent associations with GCS (p<0.0001), admission glucose (p<0.0001), CT score (p=0.0007), EC-ISS (p<0.0001), and ICH AIS (p<0.0001), but not with ISS ≥16 (p=0.2359). AO had independent associations with GCS (p<0.0001), admission glucose (p<0.0016), EC-ISS (p=0.0102), ICH AIS (p<0.0001), hypotension (p=0.0347), and CT score (p=0.0215), but not with ISS ≥16 (p=0.2011). Not following commands at discharge had independent associations with GCS (p=0.0006), EC-ISS (p=0.0459), ICH AIS (p=0.0002), and CT score (p=0.0201), but not with ISS ≥16 (p=0.9500). Not following commands at three months had independent associations with EC-ISS (p<0.0001), ICH AIS (p=0.0006), age (p<0.0001), GCS deficit + EC-ISS (p<0.0001), and hypotension (p=0.0078), but not with ISS ≥16 (p=0.4157). Conclusion ICH AIS and EC-ISS have independent associations with six undesirable ICH outcomes. ICH AIS and EC-ISS may better represent risk conditions in ICH patients when compared with ISS.