Prediabetes and diabetes predict mortality risk after acute stroke

Takeaway

• Both pre-diabetes and diabetes may predict short-term (30-day) mortality after acute stroke but the association differed according to pre-stroke and stroke-related characteristics of patients. 

Why this matters

• Findings warrant further investigations to determine the mechanisms by which pre-stroke and stroke-related characteristics of the patient can modify the prognostic value of pre-stroke glycaemic status for stroke mortality.

Study design

• Retrospective study of 2076 patients with acute ischaemic stroke (AIS) and 586 patients with intracerebral haemorrhage (ICH).
• Glycaemic status assessed by haemoglobin A1c level and clinical severity assessed by the National Institutes of Health Stroke Scale (NIHSS).
• Main outcome: all-cause mortality at 30 days and 1 year after stroke onset.
• Funding: University of Bologna.

Key results

• Pre-diabetes and diabetes were associated with higher 30-day mortality after:
  • IS (total deaths, 280; normoglycaemic, 11.6%; pre-diabetes, 13.4%; diabetes, 15.5%; P=.125).
  • ICH (total deaths, 150; normoglycaemia, 24.1%; pre-diabetes, 25.1%; diabetes, 29.0%; P=.562).
• Among patients with IS, the association was limited to patients with pre-stroke disability and very severe stroke (P<.001).
• The risk for mortality risk was higher with:
  • pre-diabetes in patients with NIHSS score:
    • 25 (HR, 1.58; 95% CI, 1.07-2.35) and
    • 35 (HR, 2.12; 95% CI, 1.18-3.82).
  • diabetes in patients with NIHSS score:
    • 25 (HR, 1.67; 95% CI, 1.14-2.46) and
    • 35 (HR, 2.67; 95% CI, 1.53-4.68).
• In patients with ICH, the association was limited to women with pre-diabetes (HR, 1.93; 95% CI 1.15-3.24) and to men with diabetes (HR, 1.78; 95% CI, 1.02-3.12). 
• The risk for 1-year mortality was unrelated to the pre-stroke glycaemic status.

Limitations

• Retrospective design.