Predictors of in-hospital mortality: after a methanol poisoning outbreak in Istanbul.

Abstract

Background: Methanol poisoning remains a major public health problem, particularly during outbreaks related to illicit alcohol consumption, and is associated with high mortality. Early identification of patients at high risk of death is critical to guide timely triage and aggressive management in the emergency department (ED).

Objectives: To identify clinical and laboratory predictors of in-hospital mortality among patients with methanol poisoning during an outbreak and to evaluate the prognostic performance of key parameters using receiver operating characteristic (ROC) curve analysis and logistic regression.

Methods: This retrospective observational cohort study was conducted in the ED of a tertiary-care hospital in Istanbul, Türkiye, during a methanol poisoning outbreak between December 1, 2024, and January 31, 2025. Adult patients (≥ 18 years) diagnosed with methanol poisoning were included. Demographic data, clinical findings, laboratory results, arterial blood gas parameters, and treatments were collected. The primary outcome was in-hospital mortality. ROC curve analyses and univariate and multivariable logistic regression models were performed.

Results: A total of 55 patients were included (92.7% male; median age 46.0 years [IQR 38.5-55.0]). In-hospital mortality occurred in 25 patients (45.5%). Non-survivors had significantly lower arterial pH, bicarbonate, and base excess values and higher lactate levels and anion gap compared with survivors (all p < 0.001). Arterial pH demonstrated excellent prognostic performance (AUC 0.969), with an optimal cut-off value of ≤ 6.89 (92.0% sensitivity, 96.7% specificity). In multivariable analysis, arterial pH remained an independent predictor of mortality, with each 0.1-unit decrease associated with a 2.78-fold increase in the odds of death. In a model excluding arterial blood gas parameters, higher lactate levels and lower Glasgow Coma Scale scores were independently associated with mortality.

Conclusions: During methanol poisoning outbreaks, arterial pH is the strongest predictor of in-hospital mortality. Serum lactate and neurological status provide additional prognostic value when arterial blood gas analysis is unavailable. These findings support the use of arterial pH as a decision-support tool for early escalation of care, including emergent hemodialysis and intensive care admission, and suggest that serum lactate and GCS may serve as actionable triage parameters in resource-limited or high-burden surge settings.