Relationship Between Admission Blood Glucose And Clinical Outcome Among Patients With Acute Stroke

ObjectiveTo investigate the case-fatality rate and disability rate during hospitalization among patients with acute stroke and the association between admission blood glucose(BG) and in-hospital death and disability, so as to provide basis for decreasing case-fatality rate and disability rate of stroke patients.MethodsA total of 2 769 acute ischemic stroke and 1 807 hemorrhagic stroke patients confirmed by a computed tomography scan or magnetic resonance imaging in six hospitals in Tongliao City, Inner Mongolia, from January, 2003 to June, 2006 were included in the present study. BG and blood pressure (BP) were examined within the first 24h of hospital admission. Data on demographic characteristics, life style risk factors, medical history, clinical laboratory tests and imaging data were obtained using a standard questionnaire administered by trained staff. If a patient survived the acute stroke or died in the hospital, the study neurologists conducted a comprehensive clinical evaluation including death and disability (Modified Rankin's scale, MRs≥3) at discharge. Statistical analysis were conducted using SPSS15.0 statistical software. Clinical characteristics and risk factors of ischemic stroke and hemorrhagic stroke patients were analyzed. Risk factors between different kinds of clinical outcomes were compared for both ischemic and hemorrhagic stroke. Odds ratios (ORs) and 95% Confidence Interval (95% CI) of clinical outcomes among patients with different BG levels were calculated by using logistic model.ResultsThe rates of cigarette smoking, diabetes history and dyslipidemia were higher in ischemic stroke patients than those in hemorrhagic stroke patients(cigarette smoking: 26.1% vs. 21.7%, diabetes history: 11.4% vs. 2.2%; dyslipidemia: 53.9% vs. 44.4% ), P<0.05, whereas, the rate of hypertension history was higher in acute hemorrhagic stroke patients than that in ischemic stroke patients (59.2% vs. 54.0%), P<0.05, and patients with hemorrhagic stroke had higher systolic and diastolic BP and BG levels than those with ischemic (systolic BP: 173.7±34.0 vs. 151.6±28.2 mmHg; diastolic BP: 104.8±19.0 vs. 92.2±16.5 mmHg; BG: 7.06±2.97 vs. 6.65±3.08 mmol/L), all P<0.05.The in-hospital case-fatality rate was higher for acute hemorrhagic stroke (5.6%) than it was for acute ischemic stroke (1.6%). However, the disability rate was higher for acute ischemic stroke (45.5%) than it was for acute hemorrhagic stroke (38.2%).Among acute ischemic stroke patients, the rates of cigarette smoking, alcohol drinking and diabetes history, and BG level were significantly different between patients of death, disability and MRs<3. The rate of diabetes history and BG level were higher for dead patients than those were for disabled patients and those with MRs<3. The rates of cigarette smoking and alcohol drinking were higher for those disabled than those with MRs<3.Among acute hemorrhagic stroke patients, the rates of Mongol ethnicity, hypertension history, age, BP and BG level were significantly different between patients of death, disability and MRs<3. The rate of Mongol ethnicity is lower for dead patients than it was for those disabled, whereas, systolic BP, diastolic BP, and BG levels were higher for dead patients than those were for disabled patients and those with MRs<3. The rate of hypertension history was higher for those disabled than those with MRs<3.After adjustment for age, gender, ethnicity, alcohol drinking, cigarette smoking, systolic BP, diastolic BP, history of hypertension, diabetes and dyslipidemia, compared with those with normal BG (BG<6.1 mmol/L), OR (95% CI) of in-hospital death was 3.85(0.85, 17.37) for acute ischemic stroke patients with BG 6.1-6.9 mmol/L (P>0.05); ORs of in-hospital death were 15.44 (4.45, 53.57), 9.51 (2.75, 32.86) and 20.94 (6.32, 69.46) for acute ischemic stroke patients with BG 7.0-7.7, 7.8-11.0,≥11.1 mmol/L, respectively (all P<0.05). There was a trend that risk of in-hospital death was increasing with elevated BG levels (P<0.05). ORs (95% CI) of disability were 1.04 (0.72, 1.50), 1.18 (0.76, 1.84), 1.17 (0.81, 1.68) and 0.73 (0.44, 1.22) for acute ischemic stroke patients with BG 6.1-6.9, 7.0-7.7, 7.8-11.0,≥11.1 mmol/L, respectively; ORs (95% CI) of combined outcome were 1.10 (0.76, 1.58), 1.45 (0.95, 2.21), 1.40 (0.99, 1.99) and 1.22 (0.79, 1.89), respectively. There were not significant associations between BG and disability and combined outcome among ischemic stroke patients (P>0.05).After adjustment for other factors, compared with those with normal BG (BG<6.1 mmol/L), OR (95%CI) of in-hospital death was 0.86 (0.23, 3.15) for acute hemorrhagic stroke patients with BG 6.1-6.9 mmol/L (P>0.05); ORs (95% CI) of in-hospital death were 3.38 (1.25, 9.13), 3.95 (1.71, 9.09) and 6.80 (2.65, 17.45) for acute hemorrhagic stroke patients with BG 7.0-7.7, 7.8-11.0,≥11.1 mmol/L, respectively (all P<0.05). There was also a trend that risk of in-hospital death was increasing with elevated BG levels (P<0.05). ORs (95% CI) of disability were 1.17 (0.71, 1.95), 0.96 (0.51, 1.82), 1.20 (0.74, 1.94) and 1.28 (0.64, 2.56) for acute hemorrhagic stroke patients with BG 6.1-6.9, 7.0-7.7, 7.8-11.0,≥11.1 mmol/L, respectively (P>0.05). Compared with those with normal BG (BG<6.1 mmol/L), OR (95% CI) of combined outcome was 2.16 (1.21, 3.86) for acute hemorrhagic stroke patients with BG≥11.1 mmol/L (P<0.05). There was a tendency of increased risk of in-hospital poor outcome with elevated BG levels.ConclusionsAdmission BG was significantly and positively associated with risk of death in acute ischemic and hemorrhagic stroke patients during hospitalization. The present study suggested that BG should be controlled actively in acute phase of post-stroke.