Prediction of zinc phosphide-induced hepatotoxicity and cardiotoxicity from clinical, laboratory, and radiological indicators.

Zinc phosphide (ZnP) is a phosphine-generating pesticide. Serious hepatotoxicity or cardiotoxicity might develop late in initially stable patients. The current prospective study aimed to predict ZnP-induced hepatotoxicity and cardiotoxicity. This prospective cross-sectional study included 150 patients admitted to the Alexandria Poison Center (APC) for over 6 months (from August 2018 to January 2019). The recorded patients' data included personal data, poisoning, medical history, clinical assessment using the poisoning severity score (PSS), investigations, and the outcome. The mean age of the patients was 23.36 ± 13.53 years. Females constituted 68.7%, and 76% of the patients ingested ZnP deliberately. Only two cases that ingested the highest amount of ZnP had radio-opaque shadows in the abdominal plain X-ray. The peak of clinical manifestations was in 12 h, whereas the peak liver transaminases (alanine aminotransferase (ALT), aspartate aminotransferase (AST)) and cardiac enzymes (creatine phosphokinase (CPK), creatine kinase-muscle/brain (CK-MB), troponin) were in 24 h. No fatalities were reported, 56.7% are completely cured, 37.3% are discharged on consent, and 6% are discharged with elevated liver and/or cardiac enzymes. The receiver operating characteristic (ROC) curve was applied. Persistent elevation of liver enzymes after 72 h could be predicted if the ingested amount > 3.5 (accuracy 86.2%), time till hospitalization > 5 h (accuracy 85.9%), PSS > 1 with (accuracy 58.9%), AST > 50 U/L (accuracy 86.2%), or ALT > 82 U/L (accuracy 86.3%). Elevation of cardiac enzymes could be predicted if the ingested amount > 2.5 sachets (accuracy 99.5%), time till hospitalization > 5 h (accuracy 99.7%), PSS > 1 with (accuracy 61.2%), CPK > 260 U/L (accuracy 99.9%), CK-MB > 6 ng/mL (accuracy 99.7%), or troponin > 0.9 ng/mL (accuracy 99.8%).