Abstract
Background: Too high or low blood glucose levels after traumatic brain injury (TBI) negatively affect the prognosis of patients with TBI. This study aimed to examine the relationship between different levels of blood glucose in insulin therapy and Glasgow Outcome Score (GOS) in patients with TBI.Methods: This study was based on a randomized, dual-center, open-label clinical trial. A total of 208 patients who participated in the randomized controlled trial were followed up for 5 years. The information on disease, lab examination, insulin therapy, and operation of patients with TBI was collected. Also, the data on 5-year and 6-month GOS were collected as primary and secondary outcomes. The univariate analysis was used to detect variables that might contribute to outcomes, while the multivariate regression analysis was used to reveal the independent relationship between insulin therapy and outcomes. A generalized additive model (GAM) was used to investigate dose–response relationships between blood glucose levels and GOS. The results were presented as odds ratios (ORs) with their 95% confidence intervals (95% CIs). We further applied a two-piecewise linear regression model to examine the threshold effect of blood glucose level and GOS.Results: A total of 182 patients were involved in the final analysis. Compared with the non-intensive insulin therapy group, the 5-year GOS in the moderate-control intensive insulin therapy (IIT) group and the slight-control IIT group was 1.45 and 1.39 higher, respectively (both P < 0.05); the 5-year GOS in the strict-control IIT group was 0.23 higher (P > 0.05). GAM revealed that a bell-shaped relationship existed between blood glucose levels in insulin therapy and 5-year or 6-month GOS. The inflection points of the mean blood glucose level were 6.73 and 8.97 mmol/L considering 5-year GOS as the outcome and were 6.95 and 8.88 mmol/L considering 6-month GOS as the outcome. The multivariate analysis showed that the 5-year GOS in the medium-level group (≥6.73 and <8.97 mmol/L) increased by 0.83 [95% confidence interval (CI): 0.22–1.47] compared with that in the low-level group (<6.73 mmol/L). Also, the 5-year GOS in the medium-level group increased by 0.57 (95% CI: 0.02–1.08) compared with that in the high-level group (≥8.97 mmol/L).Conclusion: A proper blood glucose range in insulin therapy could improve the outcomes of patients with TBI; the range was 6.73–8.97 mmol/L according to our limited analysis.Clinical Trial Registration: ClinicalTrials.gov, NCT02161055. Registered 11 Jun 2014.