Objective To investigate the predictive value of Rotterdam CT score combined with serum soluble cluster of differentiation antigen 40 ligand (sCD40L) and fibulin-5 for prognosis of patients with severe traumatic brain injury (sTBI).

Methods A total of 186 sTBI patients were divided into good prognosis group (n=104) and poor prognosis group (n=82). CT images were analyzed and Rotterdam CT scores were obtained; the enzyme-linked immunosorbent assay (ELISA) was used to detect serum sCD40L and fibulin-5 levels on the 1st, 3rd and 7th day of admission; the Spearman analysis was used to assess the correlations of serum sCD40L and fibulin-5 levels on the first day of admission with the Rotterdam CT score and Glasgow Coma Scale (GCS) score; the Multivariate Logistic regression analysis was conducted to explore the risk factors for poor prognosis in sTBI patients; the receiver operating characteristic (ROC) curve was performed to analyze the predictive value of Rotterdam CT score combined with serum sCD40L and fibulin-5 levels on the first day of admission for prognosis of sTBI patients.

Results On the 1st, 3rd and 7th day of admission, compared with the good prognosis group, the serum levels of sCD40L and fibulin-5 in the poor prognosis group were significantly elevated (P < 0.05); serum sCD40L and fibulin-5 levels were negatively correlated with GCS scores (r=-0.505, -0.421, P < 0.05) and positively correlated with Rotterdam CT score (r=0.495, 0.397, P < 0.05); sCD40L (OR=2.768, 95%CI, 1.537 to 4.983) and fibulin-5 (OR=2.539, 95%CI, 1.301 to 4.953) were independent risk factors for poor prognosis in sTBI patients (P < 0.001); the area under the curve (AUC) of Rotterdam CT score combined with serum sCD40L and fibulin-5 levels on the first day of admission for predicting poor prognosis in sTBI patients was 0.969 (95%CI, 0.933 to 0.989), with a sensitivity of 86.59% and a specificity of 94.23%; the diagnostic performance of Rotterdam CT score combined with sCD40L and fibulin-5 was superior to that of each individual indicator (Z=4.233, 4.274, 4.433, P < 0.001); the Bootstrap internal validation results showed that the predictive performance curve of the combined prediction model was consistent with the actual clinical occurrence curve.

Conclusion Rotterdam CT score combined with serum sCD40L and fibulin-5 has certain predictive value for poor prognosis in sTBI patients.