Predictors of Acute Respiratory Distress Syndrome in Patients with Paraquat Intoxication

• DOI: 10.1371/journal.pone.0082695
• Publication Date: 2013-12-11T16:31:36Z
• Authors: Cheng-Hao Weng; Ching-Chih Hu; Ja-Liang Lin; Dan-Tzu Lin-Tan; Ching-Wei Hsu; Tzung-Hai Yen
• Keywords: Adult; Male; Paraquat; Respiratory Distress Syndrome; Herbicides; Science; Q; R; Middle Aged; Prognosis; 3. Good health; 03 medical and health sciences; 0302 clinical medicine; Medicine; Humans; Female; Research Article; Retrospective Studies

Paraquat poisoning is characterized by acute lung injury, pulmonary fibrosis, respiratory failure, and multi-organ failure, resulting in a high rate of mortality and morbidity. The objectives of this study were to identify predictors of acute respiratory distress syndrome (ARDS) in cases of paraquat poisoning and determine the association between these parameters.In total, 187 patients were referred for management of intentional paraquat ingestion between 2000 and 2010. Demographic, clinical, and laboratory data were recorded. Sequential organ failure assessment (SOFA) and Acute Kidney Injury Network (AKIN) scores were collected, and predictors of ARDS were analyzed.The overall mortality rate for the entire population was 54% (101/187). Furthermore, the mortality rate was higher in the ARDS patients than in the non-ARDS patients (80% vs. 43.80%, P<0.001). Additionally, the ARDS patients not only had higher AKIN48-h scores (P<0.009), SOFA48-h scores (P<0.001), and time to ARDS/nadir PaO2 (P=0.008) but also suffered from lower nadir PaO2 (P<0.001), nadir AaDO2 (P<0.001), and nadir eGFR (P=0.001) compared to those in the non-ARDS patients. Moreover, pneumomediastinum episodes were more frequent in the ARDS patients than in the non-ARDS patients (P<0.001). A multivariate Cox regression model revealed that blood paraquat concentrations (P<0.001), SOFA48-h scores (P=0.001), and steroid and cyclophosphamide pulse therapies (P=0.024) were significant predictors of ARDS. The cumulative survival rates differed significantly (P<0.001) between patients with SOFA48-h scores <3 and SOFA48-h scores ≥3, with a sensitivity of 95.8%, specificity of 58.4%, and overall correctness of 67.6%. Finally, the area under the receiver operating characteristic (AUROC) analysis showed that SOFA48-h scores (P<0.001) had a better discriminatory power than blood paraquat concentrations (P=0.01) for predicting ARDS.The analytical results indicate that SOFA48-h scores, blood paraquat concentrations, and steroid and cyclophosphamide pulse therapies are significantly associated with ARDS complications after paraquat intoxication.