Association of multiple glycemic parameters at hospital admission with mortality and short-term outcomes in acutely poisoned patients

Abstract

The prognostic value of multiple glycemic parameters in poisoned patients was never assessed. We aim to explore the effects of glucose variability on short-term outcomes in nondiabetic and diabetic patients acutely poisoned with undifferentiated xenobiotics. We performed a prospective observational study in a tertiary center for toxicology in northeastern Romania. Over the course of 3 years, we included 1076 adults, older than 18 years, admitted for acute poisoning with a xenobiotic. The mortality rate was 4.1%. The admission blood glucose level (BGL) predicted mortality (OR 1.015, 95% CI 1.011–1.019, p < 0.001) and complications (OR 1.005, 95% CI 1.001–1.009, p 0.02). The mean glucose level (MGL) after admission (OR 1.007, 95% CI 1.000–1.013, p 0.034) and coefficient of glucose variability (CV) were predictive for complications (OR 40.58, 95% CI 1.35–1220.52, p 0.033), using the same multivariable model. The receiver operating characteristic curve (ROC) analysis revealed that BGL had good predictive value for in-hospital mortality (area under the curve (AUC) = 0.744, 95% CI = 0.648–0.841, p < 0.001), and complications (AUC = 0.618, 95% CI = 0.584–0.653, p < 0.001). In patients acutely poisoned with xenobiotics, the BGL, MGL and CV can be useful as mortality and short-outcome predictors.