Performance of optic nerve sheath diameter for predicting mortality in patients with traumatic brain injury: A systematic review and meta-analysis.

OBJECTIVE: This study explored the performance of optic nerve sheath diameter (ONSD) for predicting mortality in patients with traumatic brain injury (TBI). MATERIALS AND METHODS: PubMed, Web of Science, and China National Knowledge Infrastructure databases were searched until October 2025 for studies on the prediction of ONSD for mortality in TBI. The pooled odds ratios (ORs) with 95% confidence intervals (CIs) and area under the curve (AUC) were used. Subgroup analysis, sensitivity analysis, and Deek test were performed to assess ONSD's performance in mortality prediction. RESULTS: The systematic search yielded 6 articles involving 516 patients with TBI. A broad ONSD was associated with an increased mortality (OR = 21.86, 95% CI: 12.26-38.99, P < .001), and the AUC of ONSD for predicting mortality in patients with TBI was 0.87 (95% CI: 0.84-0.90), showing 86% sensitivity and 78% specificity. Subgroup analysis revealed that ages, gender, study design, sample size, publication location, and publication years had no important effects on heterogeneity. The pooled results were robust according to the sensitivity analysis. Deeks' test showed no evident publication bias in this meta-analysis (P = .19). CONCLUSION: This meta-analysis demonstrated that ONSD is a useful predictive method for mortality in patients with TBI.