Prevalence, incidence, and clinical resolution of insulin resistance in critically ill patients: an observational study

BACKGROUND: The primary objective of this study was to measure the prevalence, incidence, and resolution of insulin resistance (IR) in critically ill patients. A secondary objective was to explore the relationship between IR and inflammatory cytokines, coagulation abnormalities, and clinical outcomes. Design: Prospective observational study. METHODS: The setting was the medical/surgical intensive care unit (ICU). We enrolled consecutive patients within 24 hours of admission to the ICU. Blood samples were collected daily until discharge, death, or a maximum of 10 days, then sent for measurement of markers of IR, inflammation, and coagulation. Charts were reviewed retrospectively to determine clinical outcomes. The homeostasis model assessment method (HOMA) was used to determine IR; a score of > or = 4 represents insulin resistance. RESULTS: A total of 96 patients were enrolled. Upon admission, 64 (67%) patients had overt IR (glucose > 7 mmol/L or insulin use), 9 (9.4%) had non-overt IR (normal glucose but HOMA > 4), and 23 (24%) were insulin sensitive (IS; normal glucose and HOMA < 4). During the course of ICU stay, an additional 16 patients developed overt IR, while 10 (10%) remained IS. There were no significant differences in inflammatory markers, coagulation tests, and clinical outcomes between IR and IS patients. There was no significant correlation between HOMA and inflammatory markers and coagulation markers. In a multivariable regression model, only interleukin-6 levels were significantly associated with mortality. CONCLUSIONS: A high proportion of critically ill patients have IR. There may not be any significant relationship between IR and measures of inflammation, coagulation, and clinical outcomes in a heterogeneous population of critically ill patients.