Predictive Factors of Prognosis in Paraquat Poisoning

ABSTRACT

PURPOSE: Paraquat is widely used non-selective contact herbicide. In spite of efforts to improve the outcome in paraquat poisoning, the mortality rate still remains high. The purpose of this study is to assess the predictive factors of prognosis by investigating initial laboratory data on paraquat poisoning. METHODS: The author retrospectively analyzed the clinical features, laboratory data, and outcomes for 130 paraquatpoisoning patients treated from June 1997 to September 2003 at the emergency medical center of Chonnam National University Hospital. RESULTS: The results were as follows 1) The mortalities were significantly older than the survivors (49.3 years vs 37.6 years, p<0.05). Among the total 130 cases, the survivors numbered 50, and the deceased numbered 80. The mean age of all 130 cases was 44.8+/-16.9 years. 2) The white blood cell count and the levels of serum AST and ALT in the deceased were significantly higher than they were in the survivors. The levels of arterial pH, PaCO 2, base excess, HCO3 -, and serum potassium in the survivors were significantly higher than they were in the deceased. 3) A multivariate Cox-Regression analysis revealed that sex, serum potassium, and arterial HCO3 - were associated with the mortality rate. CONCILUSION The above data reveal that the ingested amount, the result of the urine sodium dithionite test, and the initial state of laboratory parameters, including the white blood cell count, ABGA (pH, PaCO2, HCO3 -, base excess), serum potassium, AST, and ALT, can be used to predict the outcome of paraquat poisoning.