Brain MRI White Matter Hyperintensity and Microbleed Associations in Trauma Patients With CT-Detected Intracranial...

Objective The present study aimed to compare white matter hyperintensity (WMH) and white matter microbleed (WMM) proportions in trauma patients with intracranial hemorrhage (ICH) to published participants without ICH. When WMH or WMM proportions were associated with head trauma, these proportions were assessed to determine their associations with multiple undesirable outcomes. Methods The study population had ICH on CT, an admission Glasgow Coma Scale (GCS) score of 3-15, and blunt trauma. MRI was reviewed to determine the presence of WMH and WMM. The prevalence of non-trauma WMH and WMM by age was collated from the literature. The CT ICH mass effect (ME) scores reflect the presence or absence of lateral ventricular compression, basal cistern compression, and midline shift (0-3). The CT ICH Abbreviated Injury Scale (AIS) scores were obtained from the trauma registry (1-5). The total ICH score was determined by calculating AIS + ME + WMM (presence = 1; absence = 0) (range 1-9). An adverse outcome included hospital deaths, intensive care stay ≥2 days, or hospital stay >5 days. Multivariate analyses were used to determine independent associations. Results Of 1,545 patients, MRI was performed in 173 (mean age, 64.7 years). The WMH proportion was lower in patients with ICH (117/173; 67.6%) than in non-ICH age-matched counterparts (3,385/4,073; 83.1%; p < 0.0001). Mean ages with and without WMH were 72.4 and 48.6 years, respectively (p < 0.0001). The WMM proportion was greater in patients with ICH (96/173; 55.5%) than in non-ICH age-matched counterparts (473/3,216; 14.7%; p < 0.0001). Macrobleed on CT (ME + AIS scores) had univariate associations (p < 0.02 to < 0.001) with hospital deaths, intensive care and hospital lengths of stay, an adverse outcome, and failure to follow commands at discharge. Hospital mortality had an independent association with total ICH score (GCS 3-12, p > 0.1000; ME + AIS score, p > 0.1000; total ICH score, p = 0.0151). An ICU stay of ≥2 days was independently associated with total ICH score (GCS 3-12, p = 0.0004; ME + AIS score, p > 0.1000; total ICH score, p = 0.0004). Similarly, a hospital stay exceeding five days was significantly associated with total ICH score (GCS 3-12, p = 0.0004; ME + AIS score, p > 0.1000; total ICH score, p = 0.0071). An adverse outcome also showed an independent association with total ICH score (GCS 3-12, p = 0.0031; ME + AIS score, p > 0.1000; total ICH score, p = 0.0046). Additionally, failure to follow commands at three months was independently associated with total ICH score (GCS 3-12, p = 0.8283; ME + AIS score, p = 0.0948; total ICH score, p = 0.0443). The ME + AIS score did not show significant independent associations with these outcomes (p > 0.1000). Conclusions WMMs were indicative of traumatic axonal brain injury, whereas WMHs were associated with advancing age. Macrobleed on CT had univariate associations with multiple undesirable outcomes. The combination of macrobleeds identified on CT (ME + AIS) and microbleeds detected on MRI (WMM) showed independent associations with poor outcomes. However, CT-detected macrobleeds without microbleed status did not demonstrate an independent association. Findings from this study underscore the need for further investigation into the quantification of CT macrobleeds and MRI microbleeds and their correlation with both in-hospital and long-term clinical outcomes.