The prevalence of undiagnosed diabetes mellitus and the association of baseline glycemic control on mortality in the intensive care unit: A prospective observational study

Abstract Purpose Hyperglycemia is common during critical illness and can adversely affect clinical outcomes. We sought to determine the prevalence of undiagnosed diabetes among medical intensive care unit (MICU) patients with stress hyperglycemia and the association between baseline glycemic control and mortality. Materials and methods A prospective, observational cohort study was performed at a tertiary care MICU. Hemoglobin A1c (HbA1c) levels were obtained from any patient who developed hyperglycemia and all known diabetic patients. We assessed the prevalence of undiagnosed diabetes (defined by HbA1c) among patients with stress hyperglycemia, and the association between baseline glycemic control and mortality. Results We enrolled 299 patients. One hundred two (34.1%) had no history and 197 (65.9%) had a history of diabetes. Of the nondiabetic patients, 14 (13.7%) had an HbA1c of at least 6.5%. There was a significant difference in mortality between patients with HbA1c less than 6.5% and those with HbA1c of at least 6.5% (19.3% vs 11.7%, P = .038), despite similar Acute Physiology and Chronic Health Evaluation II scores. There was no significant difference in demographic characteristics between these groups. Multivariable logistic regression revealed lower HbA1c levels to be significantly associated with increased hospital mortality (odds ratio, 1.92; 95% confidence interval, 1.30-2.85; P = .001). Conclusion A significant number of MICU patients with stress hyperglycemia have undiagnosed diabetes. Hyperglycemia with lower baseline HbA1c was associated with increased mortality.