Analysis of Clinical Predictors for Mortality in Severe Organophosphate Poisoning

Abstract

Background: Severe organophosphate (OP) pesticide poisoning remains a major public-health problem in many low- and middle-income countries, with reported in-hospital mortality rates of 10–40 % despite antidotal therapy. Accurate early prediction of fatal outcome could optimise triage and critical-care resource allocation.

Methods: We conducted a prospective cohort study of 150 consecutive adults (≥ 18 y) with confirmed severe OP poisoning admitted to two tertiary toxicology centres between January 2024 and December 2024. Demographic, clinical and laboratory variables collected within 1 h of admission were analysed. Primary outcome was all-cause in-hospital death. Multivariate logistic regression, Kaplan–Meier survival analysis and receiver-operating-characteristic (ROC) curves were applied to identify independent predictors and evaluate existing scoring systems.

Results: Overall mortality was 20 % (30/150). Independent predictors of death were (i) time-to-presentation > 3 h (adjusted OR 3.4, 95 % CI 1.5–7.8), (ii) Glasgow Coma Scale ≤ 10 (OR 4.8, 2.0–11.4), (iii) serum butyryl-cholinesterase < 3 000 U/L (OR 6.1, 2.2–16.9), (iv) arterial pH < 7.25 (OR 5.2, 2.0–13.2) and (v) shock on admission (OR 10.2, 3.1–34.3). Kaplan–Meier curves showed significantly poorer 14-day survival for patients with low cholinesterase (log-rank p < 0.001; Fig. 1). The APACHE II score demonstrated the best discrimination (AUC 0.94; Fig. 2), followed by a modified SOFA-Lac (AUC 0.88) and Poisoning Severity Score (AUC 0.85).

Conclusion: Easily measurable bedside variables—particularly shock, profound acidaemia, low cholinesterase activity and delayed presentation—identify OP-poisoned patients at highest risk of death. Incorporating these factors into standard assessment, alongside APACHE II, could enhance early critical-care referral in resource-limited settings.