• Gregory W Roberts, James S Krinsley, Jean-Charles Preiser, Stephen Quinn, Peter R Rule, Michael Brownlee, Guillermo E Umpierrez, and Irl B Hirsch.
• SA Pharmacy, Flinders Medical Centre, Bedford Park, SA 5042, Australia; College of Medicine and Public Health, Flinders University, Bedford Park, SA 5042, Australia. Electronic address: Greg.Roberts2@sa.gov.au.
• J Crit Care. 2024 Jun 1; 81: 154537154537.

IntroductionThe relationship between critical care mortality and combined impact of malglycemia remains undefined.MethodsWe assessed the risk-adjusted relationship (n = 4790) between hospital mortality with malglycemia, defined as hypergycemia (hours Glycemic Ratio ≥ 1.1, where GR is quotient of mean ICU blood glucose (BG) and estimated average BG), absolute hypoglycemia (hours BG < 70 mg/dL) and relative hypoglycemia (excursions GR < 0.7 in those with HbA1c ≥ 8%).ResultsEach malglycemia was independently associated with mortality - hyperglycemia (OR 1.0020/h, 95%CI 1.0009-1.0031, p = 0.0004), absolute hypoglycemia (OR 1.0616/h, 95%CI 1.0190-1.1061, p = 0.0043), and relative hypoglycemia (OR 1.2813/excursion, 95%CI 1.0704-1.5338, p = 0.0069). Absolute (7.4%) and relative hypoglycemia (6.7%) exposure dominated the first 24 h, decreasing thereafter. While hyperglycemia had lower risk association with mortality, it was persistently present across the length-of-stay (68-76% incidence daily), making it the dominant form of malglycemia. Relative contributions in the first five days from hyperglycemia, absolute hypoglycemia and relative hypoglycemia were 60%, 21% and 19% respectively.ConclusionsAbsolute and relative hypoglycemia occurred largely in the first 24 h. Relative to all hypoglycemia, the associated mortality from the seemingly less potent but consistently more prevalent hyperglycemia steadily accumulated with increasing length-of-stay. This has important implications for interpretation of study results.Copyright © 2023. Published by Elsevier Inc.

15 keywords...

hide…