Diffuse axonal injury (DAI) plays a major role after traumatic brain injury (TBI). Its imaging is based on computed tomography (CT) or magnetic resonance imaging (MRI). However, DAI is a histological diagnosis. Histopathological findings on survival after TBI are very rare. Hence, it is unclear whether the neuroradiological findings are of clinical relevance. Cerebral specimens were taken in 24 patients with TBI requiring surgery. The presence of histopathological evidence for DAI was evaluated. Specimens were taken from an extracranial brain prolapse (n = 2) and from peripheral parts of a brain contusion (n = 22). Histological findings were correlated to the clinical course and the neurological status. A clinical follow-up was carried out 6 months after the surgery using the Glasgow Outcome Score (GOS). The study was approved by the local ethics committee. Specimens taken were temporal (n = 11), frontal (n = 8), parietal (n = 4) and cerebellar (n = 1). The incidence of DAI within these specimens was 30% (7 with DAI, 17 without DAI). DAI was verifiable up to 3 days after trauma. There was no correlation between DAI and Marshall classification in CT. The period of coma was longer in subjects with DAI. There was no difference in GOS in the case of a verified DAI. These results enforce the prognostic and neuroradiologic relevance of DAI. However, it is debatable whether the pathomorphologic findings in CT or MRI represent the histological findings of DAI. We suggest a multicentre study for further clarification.