A nomogram to predict intracranial hypertension in moderate traumatic brain injury patients

Abstract

Abstract Objective: patients with moderate traumatic brain injury (TBI) are under the threat of intracranial hypertension(IHT), which is an important cause of death and unfavorable outcome of TBI patients. However, it is unclear which moderate TBI patients will develop IHT and when to receive ICP-lowering treatment or even invasive ICP monitoring after admission. The purpose of the present study was to develop and validate a prediction model that estimates the risk of IHT in moderate TBI patients with lower GCS (9-11) by admission data. Methods: baseline data collected on admission of 296 moderate TBI patients with GCS score of 9-11 were collected and analyzed. Multi-variable logistic regression modeling with backward stepwise elimination was used to develop a prediction model for IHT. The discrimination efficacy, calibration efficacy, and clinical utility of the prediction model were evaluated. Finally, the prediction model was validated in a separate cohort of 74 patients from 3 hospitals. Results: four independent prognostic factors for IHT were identified: GCS score of 9, Marshall diffuse injury type IV and nonevacuated mass lesion, ISS≥18and location of contusion (frontal and temporal contusion). A prediction model was established and shown as a nomogram. The C-statistic of the prediction model in internal validation was 84.30% (95% confidence interval [CI]: 0.794–0.892). External validation was performed in a separate cohort of 85 patients. The area under the curve for the prediction model was 82.70% (95% CI: 0.726~0.928). Conclusions: A prediction model based on patient parameters collected on admission was found to be highly sensitive in distinguishing moderate TBI patients with lower GCS score of 9-11 who would suffer IHT. The high discriminative ability of the prediction model supports its use in identifying moderate TBI patients with lower GCS score of 9-11 who need ICP-lowering therapy or invasive ICP monitoring.