Stress-induced hyperglycemia and mortality in patients with traumatic brain injury without preexisting diabetes: A meta-analysis.
Level 3 - non-randomized controlled study
Meta-analysis of non-randomized observational cohort studies
PubMed 39319871 · doi:10.17305/bb.2024.10865
What was done
A systematic search was conducted in Medline, Web of Science, Embase, Wanfang, and CNKI databases up to May 15, 2024. Researchers included observational cohort studies comparing all-cause mortality between traumatic brain injury (TBI) patients without preexisting diabetes who had stress-induced hyperglycemia (SIH) and those who did not. Pooled risk ratios (RR) and 95% confidence intervals (CI) were calculated using a random-effects model.
What was found
Twelve cohort studies containing 15 datasets and 16,387 TBI patients met inclusion criteria. SIH was associated with significantly higher all-cause mortality (RR: 2.00, 95% CI: 1.72–2.33, P < 0.001) with mild heterogeneity (I² = 25%). The association remained statistically significant but attenuated in studies utilizing multivariate adjustments (RR: 1.76) relative to univariate analyses (RR: 2.69). Subgroup findings were consistent across study design, age, gender proportion, SIH definition, and follow-up duration.
Why it matters
This review establishes stress-induced hyperglycemia as a strong prognostic marker for increased mortality in non-diabetic TBI patients, highlighting the clinical relevance of acute post-trauma glucose monitoring.
Limits
All included data were observational, precluding causal inference. The attenuation of risk in multivariate-adjusted cohorts indicates potential residual confounding from initial injury severity. Abstract details omit exact glucose thresholds and follow-up lengths used across individual studies.
Cited by
- supports Elevations in blood sugar following a significant brain injury are associated with worse clinical outcomes.