Background and aims: The effect of known diabetes on in-hospital mortality from ischemic stroke is still debated whereas the role of unknown diabetes is virtually unexplored. This study evaluates the impact of known and unknown diabetes on in-hospital mortality from ischemic stroke.
Methods and results: We have retrospectively evaluated the records of 286 consecutive cases of ischemic stroke hospitalized from January 1998 to December 2000 at the Department of Internal Medicine of the General Hospital located in the western area of Naples. Fasting plasma glucose level < 7 mmol/L identified non diabetic subjects. Known diabetes mellitus was diagnosed by history of diabetes and/or hypoglycemic therapy, unknown diabetes was defined as a random plasma glucose level > or = 11 mmol/L and/or in-hospital fasting glucose > or = 7 mmol/L on two or more occasions. Severity of stroke was defined using the Canadian Neurological Score (CNS). According to these criteria, 144 subjects were non diabetics, 99 had known diabetes and 43 had unknown diabetes. Subjects with known diabetes showed a higher prevalence of female sex, hypertension and increased triglyceride levels as compared with non diabetic subjects (p < 0.01). Subjects with unknown diabetes were older (p < 0.01) and showed a more severe CNS (3.4 +/- 2.7) than non diabetic and diabetic subjects (5.8 +/- 2.6 and 5.8 +/- 2.6, respectively; p < 0.01). In-hospital mortality was significantly higher in the unknown diabetic group (44%) as compared with known diabetic (15%) and non diabetic groups (12%) (p < 0.001). This finding was independent of neurological deficit, age, atrial fibrillation and history of previous stroke.
Conclusions: Our study shows that unknown diabetes, more than known diabetes, is a strong risk factor for in-hospital mortality in subjects with acute ischemic stroke.