Blood glucose and cerebrovascular disease in nondiabetic patients.

The objective of this study was to determine if elevations in blood glucose, in a range classified as impaired fasting glucose, are associated with a greater incidence of cerebrovascular disease in nondiabetic patients. Morning blood glucose determinations were evaluated with respect to subsequent stroke using records from 28 477 nondiabetic patients. Strokes and transient ischemic attacks (TIA) were identified from ICD-9 coding for a new event more than a year after baseline glucose determinations. Of the patients studied, 593 suffered stroke or TIA over a total risk analysis time of 100 982 years. Higher baseline morning glucose (100 to 126 mg/dL vs under 100 mg/dL) was associated with 31% more diagnoses (2.4% vs 1.8%, P < .001). Incidence rate was 5.3 per 1000 patient-years for those patients with glucose over 100 mg/dL and 3.9 per 1000 patient-years for those with glucose under 100 mg/dL (P <.001). Kaplan-Meier analysis showed that elevated baseline glucose was associated with a progressive, increased risk with time. A Cox proportional hazards model with adjustment for age, body mass index, sex, creatinine, lipids, smoking, and medications showed that elevated fasting glucose was associated with an increased hazard for a new event (hazard ratio 1.24, 95% CI 1.05-1.46, glucose over 100 mg/dL vs under 100 mg/dL). Thus, patients with higher baseline blood glucose levels in the absence of diabetes and after adjustment for covariants have a significantly greater risk for development of cerebrovascular disease.