96 only BSIs patients were categorized into four groups according to HbA1c as < 6 % , ?#xA0;6 % but < 7 % , ?#xA0;7 % but < 8 % , or ?#xA0;8 % , respectively. The association of the four HbA1c groups with diffusion-weighted imaging (DWI) infarct volumes (DIV), National Institutes of Health Stroke Scale (NIHSS), and follow-up modified Rankin Scale (FmRS) scores were analyzed. Patients also were categorized into two groups according to HbA1c < 6 % or ?#xA0;6 % . Logistic regression analyses were performed to determine independent risk factors.
There was a significant correlation between HbA1c and DIV (Spearman ¦Ñ = 0.339, P = 0.001), NIHSS scores (¦Ñ = 0.292, P = 0.004) and FmRS scores (¦Ñ = 0.315, P = 0.002). The incidence of pons infarction was highest in BSIs and patients with HbA1c ?#xA0;6 % showed significantly more frequent isolated pontine infarction. Logistic regression analyses showed that only HbA1c was independently associated with larger DIV (P = 0.025) and FmRS scores (P = 0.026).
These results suggest that elevated HbA1c level may be a potential serologic marker in the evaluation of the severity and prognosis of acute BSIs. There is an urgent need to study control of diabetes mellitus (DM) before and after BSIs.