Background: Traumatic brain injury (TBI) has been associated with an acute stress response mediated by the sympathoadrenomedullary axis, which can be assessed by measuring blood glucose level. Methods: This prospective observational study was conducted for a year in 2007 among 294 patients who had been treated for TBI in Hospital Kuala Lumpur. Patients fulfilling the set criteria were recruited into the study and data, including blood glucose level and Glasgow Outcome Score at 3-month follow-up, were collected. Results: 294 patients were included in the study: 50 females (17.0%) and 244 males (83.0%). The majority of cases were young adult patients (mean age of 34.2 years, SD 13.0). The mean blood glucose level during admission and post-surgery were 6.26 mmol/L (SD 1.30, n = 294) and 6.66 mmol/L (SD 1.44, n = 261), respectively. Specifically, the mean admission glucose level associated with mild TBI was 5.04 mmol/L (SD 0.71); moderate TBI, 5.78 mmol/L (SD 1.02); and severe TBI, 7.04 mmol/L (SD 1.18). The mean admission glucose level associated with a poor outcome in patients with isolated TBI was 6.98 mmol/L (SD 1.21). Patients with admission glucose of 5.56 mmol/L (SD 1.21) were more likely to have a favourable outcome. Conclusion: Mild, moderate, and severe TBI were associated with an increase in blood glucose levels during admission, and the mean increase in glucose levels is based on the severity of the isolated TBI. Surgical intervention did not cause further significant changes in blood glucose levels. Patients with isolated TBI and minimal increases in blood glucose levels were more likely to have a favourable outcome.