Posttraumatic cerebral infarction (PTCI) is a severe secondary insult of traumatic brain injury (TBI). This study aimed to evaluate the characteristics and risk factors of PTCI after severe TBI (sTBI) and explore possible mechanism. This retrospective study included a cohort of 339 patients with sTBI; they were divided into the PTCI and non-PTCI groups. Clinical data and follow-up charts were reviewed for comparison. The logistic regression model was used for multivariate analysis to detect the risk factors of PTCI. The Glasgow Outcome Scale (GOS) and Barthel index (BI) for activities of daily living (ADL) were applied to evaluate their outcome. PTCI led to an increased mortality (43.5 % vs. 10.7 %, P < 0.001) and days of intensive care unit stay (14.3 days vs. 7.1 days, P < 0.001), decreased GOS (3.1 vs. 4.1, P < 0.001) and BI (25.0 vs. 77.9, P < 0.001). Increased infarction volume led to poor outcome assessed by GOS (r = -0.46, P < 0.0001) and BI for ADL (r = -0.36, P = 0.026) for surviving patients. Compared with non-PTCI patients, PTCI patients had a high incidence of midline shift (36.2 % vs. 20.7 %, P = 0.011) and posttraumatic vasospasm (PTV) (42.0 % vs. 27.4 %, P = 0.027). Daily prevalence of PTCI occurred in two peaks: one (73.9 %) was in the first 24 h after injury, while the other (18.8 %) was in the span of 43 to 60 h postinjury. In multivariate analysis, hyperthermia [adjusted odds ratio (OR), 3.11; P = 0.001] in the first 24 h, thrombocytopenia (OR, 27.08; P < 0.001), abnormal prothrombin time (OR, 7.66; P < 0.001) and traumatic subarachnoid hemorrhage (OR, 2.33; P = 0.022) were independent predictors for PTCI. PTCI deteriorates the outcome of sTBI patients. Mechanical compression and hemocoagulative disturbance serve as potential mechanisms mediating this pathophysiological process. PTV may also contribute to PTCI, but its association with PTCI is weak and needs further exploration. Early recognition and intervention of these factors might be beneficial for preventing PTCI.