Poor glycemic control and posterior circulation ischemic stroke

Junya Kuroda; Ryu Matsuo; Yuko Yamaguchi; Noriko Sato; Masahiro Kamouchi; Jun Hata; Yoshinobu Wakisaka; Tetsuro Ago; Takanari Kitazono; Fukuoka Stroke Registry Investigators

doi:10.1212/CPJ.0000000000000608

Poor glycemic control and posterior circulation ischemic stroke

Neurol Clin Pract. 2019 Apr;9(2):129-139. doi: 10.1212/CPJ.0000000000000608.

Authors

Junya Kuroda¹, Ryu Matsuo¹, Yuko Yamaguchi¹, Noriko Sato¹, Masahiro Kamouchi¹, Jun Hata¹, Yoshinobu Wakisaka¹, Tetsuro Ago¹, Takanari Kitazono¹; Fukuoka Stroke Registry Investigators

Affiliation

¹ Department of Medicine and Clinical Science (JK, RM, NS, YW, TA, TK), Graduate School of Medical Sciences, Kyushu University, Fukuoka; Cerebrovascular Division (JK), Cerebrovascular and Neurology Center, National Hospital Organization Fukuoka Higashi Medical Center, Koga; Department of Health Care Administration and Management (RM, YY, NS, MK), Graduate School of Medical Sciences, Kyushu University, Fukuoka; Center for Cohort Studies (MK, JH, TK), Graduate School of Medical Sciences, Kyushu University, Fukuoka; and Department of Epidemiology and Public Health (JH), Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan.

Abstract

Background: This study aimed at determining whether diabetes or glucose metabolism is associated with ischemic stroke in the posterior circulation.

Methods: We included 10,245 patients with acute ischemic stroke (mean age 72.7 ± 12.5 years, men 59.5%) who were enrolled in a multicenter hospital-based stroke registry in Fukuoka, Japan, between June 2007 and August 2016. Posterior circulation ischemic stroke (PCIS) was defined as brain infarction in the territory of the posterior cerebral artery and vertebro-basilar arteries. We investigated the associations between diabetes or glycemic parameters, including plasma glucose concentrations, hemoglobin A1c, and the homeostatic model assessment of insulin resistance (HOMA-IR), and PCIS using logistic regression analysis. To improve covariate imbalance, we further evaluated associations after propensity score matching using 1:1 nearest neighbor matching and inverse probability weighting.

Results: Diabetes was significantly associated with PCIS even after adjusting for multiple confounding factors (odds ratio-OR [95% confidence interval], 1.37 [1.25-1.50]). Similarly, fasting (1.07 [1.02-1.12]/SD), casual plasma glucose (1.16 [1.11-1.20]/SD) concentrations, and hemoglobin A1c (1.12 [1.08-1.17]/SD), but not HOMA-IR (1.02 [0.97-1.07]/SD), were associated with PCIS. These associations were maintained in patients with ischemic stroke because of thrombotic etiology and were unchanged even after the propensity score matching methods. In patients with diabetes, the ORs of PCIS further increased with an increase in hemoglobin A1c and the presence of microvascular complications.

Conclusions: Poor glycemic control may be associated with an increased risk of thrombotic infarction that occurs preferentially in the posterior circulation of the brain.