Oral Presentation The Annual Scientific Meeting of the Endocrine Society of Australia and the Society for Reproductive Biology 2014

A prospective cohort study investigating relative hyperglycaemia as a determinant of mortality or Intensive Care Unit admission: Introducing the stress hyperglycaemia ratio (#64)

Gregory W Roberts 1 2 , Stephen J Quinn 2 , Nyoli Valentine 3 , Tariq Alhawassi 1 , Hazel O'Dea 1 , Stephen N Stranks 2 4 , Matthew P Doogue 2 4 , Morton G Burt 2 4
  1. Pharmacy Department, Repatriation General Hospital, Adelaide, SA, Australia
  2. Faculty of Health Science, Flinders University, Adelaide, SA, Australia
  3. Sturt Fleurieu General Practice Education and Training, Adelaide, SA, Australia
  4. Southern Adelaide Diabetes and Endocrine Services, Repatriation General Hospital, Adelaide, SA, Australia

Background: Hyperglycaemia on admission to hospital is associated with increased morbidity and mortality. This association is strongest in patients without known diabetes. We hypothesized that relative hyperglycaemia would be more strongly associated with mortality and morbidity than absolute hyperglycaemia.
Methods: Estimated average glucose concentration was calculated from glycosylated haemoglobin in 2290 patients acutely admitted to Flinders Medical Centre, Adelaide. Relative hyperglycaemia was defined by the stress hyperglycaemia ratio (SHR), calculated by dividing admission glucose by estimated average glucose. The odds ratios for death or Intensive Care Unit (ICU) admission per unit change in SHR and admission glucose concentration were compared in a multivariable logistic regression analysis.
Results: In univariable analyses, SHR (odds ratio = 1.23 per 0.1, p<0.001) and glucose (odds ratio = 1.18 per mmol/L, p<0.001) were positively associated with in-hospital death or ICU admission. However, in multivariable analysis, the association between in-hospital death or ICU admission and SHR was maintained (odds ratio = 1.20 per 0.1, p<0.001), but attenuated for glucose (odds ratio = 1.03 per mmol/L, p=0.31). In contrast to glucose (p=0.002), the interaction between diabetes and SHR was not significant (p=0.35), indicating that the association between SHR and outcome was not affected by diabetic status. In patients with glucose ≤10 mmol/L, the odds ratio for in-hospital death or ICU admission was 3.9 (95% confidence intervals 2.3-6.8) in the fifth quintile and 2.4 (1.4-4.2) in the fourth quintile relative to the lowest SHR quintile.
Conclusions: SHR is a better predictor of in-hospital death or ICU admission than admission glucose. SHR is associated with in-hospital death or ICU admission in patients with and without diabetes. SHR can identify patients with mild hyperglycaemia below the usual threshold for hypoglycaemic therapy at increased risk of adverse outcomes.

Funding: College of Pharmacy, King Saud University, Saudi Arabia; Novo Nordisk Australasia.