Acute glycemic variability and mortality of patients with acute stroke: a meta-analysis

Background Increased glucose fluctuation has been related to poor prognosis in patients with critical illnesses, while its prognostic role in patients with acute stroke remains unknown. The meta-analysis aimed to evaluate the association between the acute glycemic variation (GV) and mortality risk in patients with acute stroke. Methods Cohort studies were obtained by searching Medline, Web of Science, Embase, Wanfang and CNKI databases. A random-effect model which incorporates the intra-study heterogeneity was chosen to pool the results. Results Ten cohort studies with 1433 patients were included, and 280 (19.5%) of them died within 90 days of disease onset. Results of the meta-analyses showed that a higher acute GV was associated with an increased risk of early mortality in patients with acute stroke, as indicated by GV measured with the coefficient of variation of blood glucose (CVBG, odds ratio [OR]: 2.24, 95% CI 1.40 to 3.58, p < 0.001, I-2 = 73%), the standard deviation of blood glucose (SDBG, OR: 2.31, 95% CI 1.70 to 3.13, p < 0.001, I-2 = 50%), and the mean amplitude of glycemic excursion (OR: 3.57, 95% CI 1.44 to 8.85, p = 0.006, I-2 = 23%). For acute GV measured with CVBG and SDBG, subgroup analyses showed consistent results in patients with acute ischemic and hemorrhagic stroke, and for studies reporting 28-day and 90-day all-cause mortality (p for subgroup analysis all > 0.05). Conclusions Higher acute GV may be an independent risk factor of early mortality in patients with acute stroke.