Factors associated with severe effects following acute glufosinate poisoning

In acute glufosinate poisoning, sudden respiratory arrest and convulsion can occur after a latent period of 4-60 h. There is still no factor that accurately predicts the occurrence of these symptoms.

To elucidate the predictors of severe effects following acute glufosinate poisoning.

This study is a retrospective observational case series. The subjects were 16 patients who had acute glufosinate poisoning. They were divided into a group with respiratory arrest or convulsion during hospitalization (severe group) and a group without (non-severe group). The following characteristics (or predictors) were compared between the groups: age, sex, calculated amount of glufosinate (volume of ingested poison (glufosinate-containing herbicide) × glufosinate concentration of the product), time duration from poison ingestion to arrival at our hospital, use of gastric lavage, use of whole bowel irrigation, Glasgow Coma Scale, laboratory parameters, PaO₂/FiO₂ ratio (P/F ratio), shock index, and presence or absence of systemic inflammatory response syndrome (SIRS) on arrival.

The P/F ratio was significantly lower in the severe group than in the non-severe group (median, 287.5 vs. 409.0; P = 0.049). The receiver operating characteristic (ROC) curve was plotted for the predictor of increasing severity based on the P/F ratio. The area under the curve was 0.714, and the optimal cutoff point for increasing severity was 374.0. The sensitivity was 75.0%, specificity of 71.4%, and accuracy of 75.0%. The shock index was significantly higher (median, 0.52 vs. 0.41; P = 0.031). Significantly more patients had SIRS in the severe group than in the non-severe group (P = 0.015). Logistic regression analysis was performed with a backward elimination procedure. SIRS was selected as the independent predictor of increasing severity (odds ratio, 29.810; 95% confidence interval, 1.011-878.952; P = 0.049).

Severe effects following acute glufosinate poisoning were associated with two positive SIRS criteria. A low P/F ratio may be useful for predicting the occurrence of respiratory complications.