Comparative effectiveness and safety of pharmacological interventions for secondary brain injury in adult traumatic brain injury: a network meta-analysis of randomized trials.
- 1Department of Neurology, The General Hospital of Northern Theater Command, Shenyang, China.
- 2Department of Neurosurgery, The General Hospital of Northern Theater Command, Shenyang, 110016, China.
- 3Department of Neurosurgery, The General Hospital of Northern Theater Command, Shenyang, 110016, China. [email protected].
Abstract
Background
Traumatic brain injury (TBI) remains a leading cause of death and disability. We compared pharmacological and hyperosmolar interventions intended to mitigate secondary brain injury in adults with TBI.Methods
We searched PubMed/MEDLINE, Embase, Cochrane CENTRAL, Web of Science from inception to 3 January 2026 without language restrictions. Eligible studies were randomized controlled trials enrolling adults with TBI that compared an eligible pharmacological or hyperosmolar intervention with placebo/standard care or with another eligible active intervention. We performed a frequentist random-effects network meta-analysis using odds ratios (ORs) with 95% confidence intervals (CIs); Placebo/Standard Care was the reference node. Certainty was assessed using a GRADE/CINeMA-informed approach. The protocol was registered in PROSPERO (CRD420261283150).Results
Thirty-four trials (n = 23,165) were included. In the mortality network (28 trials; n = 18,088), TXA (OR 0.90, 95% CI 0.82-0.99) and erythropoietin (OR 0.64, 95% CI 0.43-0.97) were associated with lower mortality versus Placebo/Standard Care, but the TXA estimate was dominated by the Clinical Randomisation of an Antifibrinolytic in Significant Head injury-3 (CRASH-3) trial and the erythropoietin estimate was based on two trials with important clinical heterogeneity. In the favorable neurological outcome network (19 trials; n = 8,259), statins had a large but very low-certainty estimate (OR 4.71, 95% CI 2.12-10.47) based on two small trials. TXA (OR 1.07, 95% CI 0.85-1.35) and hypertonic saline (OR 0.90, 95% CI 0.74-1.10) showed no clear functional benefit. Safety networks were sparse; TXA was not associated with thromboembolism (OR 0.98, 95% CI 0.74-1.29) or acute kidney injury (OR 1.18, 95% CI 0.88-1.58).Conclusions
TXA has the most actionable mortality signal, but the network estimate adds limited information beyond the CRASH-3 trial. The apparent erythropoietin and statin effects should be interpreted cautiously because of sparse evidence, heterogeneity, and imprecision. Current evidence does not support routine use of most other pharmacological interventions to improve long-term neurological recovery after TBI. Larger, severity-stratified trials using standardized outcomes are needed.Clinical trial number
Systematic review protocol registered in PROSPERO: CRD420261283150.
Related
- Clinical similarity, economic divergence: cost-effectiveness and population health implications of dietary, pharmacological, and exercise interventions in older adults with type 2 diabetes: a secondary analysis of a randomized controlled trial.
- Effectiveness of Non-Pharmacological Interventions for Fatigue in Long COVID: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.
- Comparative efficacy of non-pharmacological interventions for amnestic mild cognitive impairment: a network meta-analysis
- Effectiveness of non-pharmacological interventions on stress and anxiety management in pregnancy: a systematic review
- Effectiveness of Nursing Interventions on Stress and Functional Capacity Among Patients After Coronary Artery Bypass Grafting (CABG): A Randomized Controlled Trial
BibTeX
@article{qu2026comparative,
title = {{Comparative effectiveness and safety of pharmacological interventions for secondary brain injury in adult traumatic brain injury: a network meta-analysis of randomized trials.}},
author = {Hui-Ling Qu and Ying-Shu Li and Cheng-Guo Jiang and Xiao-Bin Zhang},
journal = {BMC neurology},
year = {2026},
month = aug,
doi = {10.1186/s12883-026-05216-6},
url = {https://europepmc.org/article/MED/42958079},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Comparative effectiveness and safety of pharmacological interventions for secondary brain injury in adult traumatic brain injury: a network meta-analysis of randomized trials.
AU - Hui-Ling Qu
AU - Ying-Shu Li
AU - Cheng-Guo Jiang
AU - Xiao-Bin Zhang
PY - 2026
DA - 2026/08/01
JO - BMC neurology
DO - 10.1186/s12883-026-05216-6
UR - https://europepmc.org/article/MED/42958079
AB - Background Traumatic brain injury (TBI) remains a leading cause of death and disability. We compared pharmacological and hyperosmolar interventions intended to mitigate secondary brain injury in adults with TBI.Methods We searched PubMed/MEDLINE, Embase, Cochrane CENTRAL, Web of Science from inception to 3 January 2026 without language restrictions. Eligible studies were randomized controlled trials enrolling adults with TBI that compared an eligible pharmacological or hyperosmolar intervention with placebo/standard care or with another eligible active intervention. We performed a frequentist random-effects network meta-analysis using odds ratios (ORs) with 95% confidence intervals (CIs); Placebo/Standard Care was the reference node. Certainty was assessed using a GRADE/CINeMA-informed approach. The protocol was registered in PROSPERO (CRD420261283150).Results Thirty-four trials (n = 23,165) were included. In the mortality network (28 trials; n = 18,088), TXA (OR 0.90, 95% CI 0.82-0.99) and erythropoietin (OR 0.64, 95% CI 0.43-0.97) were associated with lower mortality versus Placebo/Standard Care, but the TXA estimate was dominated by the Clinical Randomisation of an Antifibrinolytic in Significant Head injury-3 (CRASH-3) trial and the erythropoietin estimate was based on two trials with important clinical heterogeneity. In the favorable neurological outcome network (19 trials; n = 8,259), statins had a large but very low-certainty estimate (OR 4.71, 95% CI 2.12-10.47) based on two small trials. TXA (OR 1.07, 95% CI 0.85-1.35) and hypertonic saline (OR 0.90, 95% CI 0.74-1.10) showed no clear functional benefit. Safety networks were sparse; TXA was not associated with thromboembolism (OR 0.98, 95% CI 0.74-1.29) or acute kidney injury (OR 1.18, 95% CI 0.88-1.58).Conclusions TXA has the most actionable mortality signal, but the network estimate adds limited information beyond the CRASH-3 trial. The apparent erythropoietin and statin effects should be interpreted cautiously because of sparse evidence, heterogeneity, and imprecision. Current evidence does not support routine use of most other pharmacological interventions to improve long-term neurological recovery after TBI. Larger, severity-stratified trials using standardized outcomes are needed.Clinical trial number Systematic review protocol registered in PROSPERO: CRD420261283150.
SN - 1471-2377
ER -