Objective and method: We have previously successfully applied the Acute Physiology and Chronic Health Evaluation (APACHE) II system to assess the severity of patients with acute paraquat poisoning, and this article investigates further evidence of the usefulness of APACHE II system in predicting the in-hospital mortality of 64 patients with acute paraquat poisoning over a period of 12 years. The predictive factors including APACHE II score, plasma paraquat concentration, severity index of paraquat poisoning (SIPP), and estimated ingestion dosage of paraquat for evaluating the outcome in paraquat-poisoned patients were assessed.
Results: Overall mortality was 71.9%: 46 out of 64 patients died. Non-survivors (n = 46) had a higher APACHE II score (23.3 +/- 12.7) than survivors (n = 18) (6.1 +/- 4.2) (p < 0.001). The plasma paraquat concentration, SIPP, and estimated ingestion dosage of paraquat were significantly higher in non-survivors than in survivors (p < 0.05, in all comparisons). By multiple logistic regression analysis, only the APACHE II score and peak data of blood sugar in 24 h after admission were capable of predicting in-hospital mortality. By using the area under receiver operating characteristic curves (AURC), the APACHE II system yielded better discriminative power (AURC = 0.893) than SIPP (AURC = 0.674), plasma paraquat concentration (AURC = 0.676), and estimated ingestion dosage of paraquat (AURC = 0.673). An APACHE II score greater than 13 predicted in-hospital mortality with 67% sensitivity and 94% specificity.
Conclusions: The APACHE II score is a simple, reproducible, and practical tool for evaluating the severity of acute paraquat poisoning.