A decade of glyphosate–surfactant poisoning in India: clinical patterns and predictors of mortality

Abstract Background Glyphosate-surfactant herbicide (GlySH) poisoning is increasingly reported in low- and middle-income countries, yet data on clinical predictors of mortality remain limited, particularly from the Indian subcontinent. This study aimed to describe the clinical profile of GlySH poisoning and to identify predictors of in-hospital mortality in a tertiary-care setting in South India. Methods We conducted a retrospective observational study of consecutive adult patients with documented oral GlySH ingestion presenting to a tertiary-care emergency department between January 2009 and July 2019. Demographic, clinical, and laboratory variables were extracted from electronic medical records. In-hospital mortality was the primary outcome. Associations between clinical variables and mortality were assessed using Firth’s penalised likelihood logistic regression and Fisher’s exact test, with odds ratios (ORs) and 95% confidence intervals (CIs) reported. For predictors with complete separation, odds ratios were not estimable, and Fisher’s exact p-values are reported. Multivariable analysis was not performed due to the small number of outcome events. Results A total of 51 patients were included, of whom four (7.8%) died. Most patients were young adults (45.1% aged 25–44 years), male (76.5%), and had suicidal intent (64.7%). Shock (OR 14.67, 95% CI 1.50–143.74), altered sensorium (OR 14.63, 95% CI 1.34–159.23), elevated serum creatinine > 1.4 mg/dL ( p = 0.0009), and dialysis requirement ( p < 0.001) were associated with in-hospital mortality on univariate analysis. Elevated liver transaminases and respiratory distress were not significantly associated with mortality. Conclusion In this 10-year cohort, mortality following GlySH poisoning was closely associated with shock, altered sensorium, and early renal dysfunction, reflecting advanced systemic toxicity. These readily identifiable features may assist in risk stratification and triage of high-risk patients in the emergency setting. Given the small number of outcome events, these findings should be interpreted as hypothesis-generating, and larger multicentric prospective studies are required to validate these predictors.