Effect of stress hyperglycemia on the prognosis in patients with severe cerebral vascular diseases

Abstract

Objective To investigate the effects of stress hyperglycemia on prognosis in patients with severe cerebral vascular diseases. Methods A retrospective analysis was conducted. 416 patients with severe cerebral vascular diseases confirmed by radiological imaging admitted to intensive care unit (ICU) of Guangdong General Hospital from December 2013 to June 2015 were enrolled. According to the values of randomise blood glucose (RBG) and glycosylated hemoglobin (HbA1c) and diabetes history, the patients were divided into euglycemia group (RBG < 11.1 mmol/L, HbA1c < 0.065, without diabetes history), diabetes group (RBG ≥ 11.1 mmol/L, HbA1c ≥ 0.065, with diabetes history), and stress hyperglycemia group (RBG ≥ 11.1 mmol/L, HbA1c < 0.065, without diabetes history). The nosocomial infection rate, the length of ICU stay and 28-day mortality were compared among the three groups. Survival analysis was performed using Kaplan-Meier method, and multivariate Cox proportional hazard model was used to estimate the risk of death. Results Among 416 patients, there were 40 cases with stress hyperglycemia, 46 with diabetes and 330 with euglycemia, with the incidence of stress hyperglycemia of 10.81% (40/370). The nosocomial infection rates in the stress hyperglycemia group and diabetes group were significantly higher than those of the euglycemia group [55.00% (22/40), 52.17% (24/46) vs. 18.79% (62/330), both P 0.05). The 28-day mortality rate in stress hyperglycemia group was significantly higher than that of diabetes group and euglycemia group [47.50% (19/40) vs. 26.09% (12/46), 10.30% (34/330), P < 0.05 and P < 0.01]. It was showed by Kaplan-Meier survival analysis that 28-day cumulative survival rate in stress hyperglycemia group was significantly lower than that of euglycemia group and diabetes group (log-rank = 6.148, P = 0.043). It was showed by Cox death risk analysis that stress hyperglycemia was the risk factor of death in patients with severe cerebral vascular disease [hazard ratio (HR) = 1.53, 95% confidence interval (95%CI) = 1.04-1.26, P = 0.001]. Conclusion The patients with stress hyperglycemia may have a higher 28-day mortality and a poorer prognosis compared with those with diabetes and normal blood glucose in severe cerebral vascular diseases. Key words: Severe cerebral vascular disease; Stress hyperglycemia; Diabetes; Prognosis