The Joint Annual Scientific Meetings of the Endocrine Society of Australia and the Society for Reproductive Biology 2018

Examination of stress hyperglycaemia and stroke outcomes using the stress hyperglycaemia ratio (#97)

Emily J Meyer 1 2 , James Sires 3 , Angela Chen 2 , Andrew Lee 4 , Cheyne Sullivan 5 , Tilenka Thynne 3 6 7 , Greg Roberts 3 5
  1. Discipline of Medicine, University of Adelaide, Adelaide, SA, Australia
  2. Endocrine and Metabolic Unit, Royal Adelaide Hospital, North Adelaide, SA, Australia
  3. Discipline of Medicine, Flinders University, Bedford Park, SA, Australia
  4. Stroke Unit, Flinders Medical Centre, Bedford Park, SA, Australia
  5. Pharmacy Department, Flinders Medical Centre, Bedford Park, SA, Australia
  6. Southern Adelaide Diabetes and Endocrine Services, Flinders Medical Centre, Bedford Park, SA, Australia
  7. Pharmacology Department, Flinders Medical Centre, Bedford Park, SA, Australia

Background
Stress hyperglycaemia in non-diabetic patients admitted with stroke is associated with increased morbidity and mortality1,2. The stress hyperglycaemia ratio (SHR), defined as admission glucose divided by estimated average glucose derived from HbA1c, measures relative hyperglycaemia3. Using SHR as a marker of relative hyperglycaemia is a superior biomarker for critical illness outcomes compared to glucose3.

Hypothesis
In patients admitted with acute stroke SHR may predict mortality and morbidity.

Methods
HbA1c and serum glucose was measured at time of hospital admission in patients with an acute stroke (n=443, 30.7% known diabetes). Endpoints were stroke severity at admission (either modified Rankin 5 at admission (mRS), GCS ≤10, or need for critical care) and poor outcome at discharge (either higher level of care, increased mRS, decreased GCS, stroke extension, haemorrhagic transformation or death). Outcomes were determined for quintiles (Q) classified for SHR and glucose.

Results
Odds ratios for Q5 (highest quintile) vs Q1 (lowest) were 2.94 (p<0.001) and 1.94 (p=0.039) for severity at admission for SHR and glucose respectively, and 2.16 (p=0.004) and 1.94 (p=0.039) for poor outcomes at discharge. Of those admitted with stroke severity (n=22), 25% of Q5 SHR had a poor outcome at discharge compared to 18% (n=18) for Q5 glucose. Identification of stress hyperglycaemia using SHR was independent of HbA1c (6.5+1.6 v 6.3+1.6, p=0.13 for Q1 and Q5 respectively), but not serum glucose (5.6+0.5 v 8.1+2.2, p<0.001). 

Conclusions
While both SHR and glucose were associated with stroke severity and poorer outcomes at discharge, SHR was independent of HbA1c. Multivariable analysis enabling direct comparison of SHR and glucose is pending and will be presented. In acute stroke SHR may be a useful prognostic marker and guide individualised glucose lowering therapy targets which may improve stroke outcomes.

  1. Capes SE, Hunt D, Malmberg K, Pathak P, Gerstein HC. Stress Hyperglycemia and Prognosis of Stroke in Nondiabetic and Diabetic Patients: A Systematic Overview. Stroke. 2001; 32(10):2426-32
  2. Dungan KM, Braithwaite SS, Preiser JC. Stress Hyperglycaemia. Lancet. 2009;373:1798-807
  3. Roberts GW, Quinn SJ, Valentine N, Alhawassi T, O’Dea H, Stranks SN, Burt MG, Doogue MP. Relative Hyperglycemia, a Marker of Critical Illness: Introducing the Stress Hyperglycemia Ratio. J Clin Endocrinol Metab. 2015;100(12):4490–4497.