[Predictive value of 3 methods in severity evaluation and prognosis of acute paraquat poisoning].

OBJECTIVE: To evaluate the predictive value of 3 methods, namely plasma paraquat concentration, severity index of paraquat poisoning (SIPP), and acute physiology and chronic health evaluation (APACHE II) in severity evaluation and prognosis of acute paraquat poisoning. METHODS: A total of 73 acute paraquat poisoning patients with oral administration were collected. Paraquat concentration in the plasma and other parameters on admission for SIPP and APACHE II were taken from medical records. According to the clinical outcome in the hospital or 7 days after the discharge, discrimination and calibration test were performed to evaluate the prognosis of the 3 methods. RESULTS: Discrimination of the 3 methods was greater than 0.8, and the area under the receiver operator curve for SIPP (0.938) was greater than paraquat concentration in the plasma (0.857) and APACHE II (0.801) with statistical significance (z=2.429, 2.021; P=0.015, 0.043). Difference in plasma paraquat concentration (0.857) and APACHE II (0.801) had no statistical significance (z=0.755, P=0.450). Hosmer-Lemeshow good fit test suggested better calibration value with statistical significance for the 3 methods (P>0.05). CONCLUSION: The 3 methods are valid in the severity evaluation and prognosis of acute paraquat poisoning. SIPP is the most preferred method, followed by paraquat concentration on admission. When there is no facility to measure paraqut concentration, APACHE II can be used as a reference for the death risk in acute paraquat poisoning.