Haemorrhagic shearing lesions associated with diffuse axonal injury: application of T2 star-weighted angiography sequence in the detection and clinical correlation.

PURPOSE: To assess the diagnostic value of T2 star-weighted angiography (SWAN) sequence for diffuse axonal injury (DAI) by virtue of correlation analysis between the number, volume and regional distribution of haemorrhagic lesions determined with SWAN sequence and clinical variables. METHODS: Twenty-eight DAI patients were included in our study and were divided into subgroups in compliance with dichotomized clinical variables separately. Global and regional number, volume and distribution of haemorrhagic lesions were compared between groups by non-parametric Mann-Whitney U test (two tailed) and independent samples t test. Spearman's rank correlation analysis was performed to compare the dichotomized clinical variables with the global and regional number and volume of lesions. RESULTS: Patients with lower Glasgow Coma Scale (GCS) score (≤8, n = 16) or prolonged coma (>4 days, n = 15) or abnormal pupillary light reflex (PLR, n = 15) had a greater global number (p ≤ 0.01) and apparent volume (p ≤ 0.01) of haemorrhagic lesions. In our study, haemorrhage extent in most brain regions, such as frontal white matter (FW), parietotemporaloccipital white matter (PTOW), corpus callosum (CC), thalamus (THAL), brain stem (BS), were greater in the lower GCS group, in the prolonged coma group and in the abnormal PLR group (p ≤ 0.05). Significant correlations were found between haemorrhage extent in global range and the dichotomized clinical variables (p ≤ 0.01). Correlations were also found between haemorrhage extent in most regions, such as FW, PTOW, CC, THAL and BS, and the dichotomized clinical variables (p ≤ 0.05). The number of involved regions was much more higher in the lower GCS group and the prolonged coma group (p < 0.0001). CONCLUSION: More accurate and objective assessment of injury can be obtained in DAI patients via SWAN sequence.