†These authors contributed equally.
Post-traumatic hydrocephalus is a common complication secondary to traumatic brain injury. It can cause cerebral metabolic impairment and dysfunction. Therefore, timely treatment with shunt implantation is necessary. However, the outcomes of shunt surgery in patients with post-traumatic hydrocephalus combined with disturbance of consciousness are doubtful. The objective was to develop a predictive model that uses the information available before surgery to predict the outcome of shunt implantation in such patients. Retrospectively collected data were used to develop a clinical prediction model. The model was derived from 59 patients using logistic regression analysis, and then it was evaluated by the area under the receiver operating characteristic curve and Hosmer-Lemshow test. A validation cohort verified the model. Four independent predictors were identified: age < 50 years, mild hydrocephalus, Glasgow Coma Scale scores 9-12 at the time of injury, and time interval from trauma to shunting < 3 months. We calculated the total score and defined the patients into three groups: low-probability (0-10 points), medium-probability (11-16 points), and high-probability (17-30 points). The rates of improved outcomes in the three groups were 14.3%, 52.6%, and 94.7%, respectively (P < 0.0001). The correlative rates of the validation cohort were 21.4%, 54.5%, and 85.7%. The prognostic model showed good discrimination (area under the receiver operating characteristic curve = 0.869) and calibration (Hosmer-Lemshow test, P = 0.391). The developed predictive model can identify patients with post-traumatic hydrocephalus combined with disturbance of consciousness who can benefit from shunt implantation. Therefore, our prognostic model can predict the outcomes of patients with post-traumatic hydrocephalus and disturbance of consciousness after shunt surgery.