BACKGROUND/AIMS: The usefulness of the diagnostic criteria of the International Consensus Meeting (criteria A) has been previously reported. However, these criteria are not clinically adaptable in Japan where allergic reaction is one of the major etiologies of drug-induced liver injury and thus it was revised and reported in the Digestive Disease Week-Japan of 2002 as DDW-J criteria (criteria B). It remains controversial whether the revised criteria can exclude drugs not causing liver injury. METHODOLOGY: Two new diagnostic criteria (criteria C and D) were designed to supplement the DDW-J criteria. Usefulness and limitations of the four criteria were retrospectively examined using cases of drug-induced liver injury experienced in 8 hospitals. RESULTS: It was confirmed that the sensitivity of criteria B is excellent for diagnosis of drug-induced liver injury. However, diagnostic criteria B were found to be disadvantageous in relation to specificity, while diagnostic criteria D were disadvantageous in relation to sensitivity. Sensitivity of diagnostic criteria C was a little superior to that of diagnostic criteria A. CONCLUSIONS: On the basis, the significant sensitivity of criteria B was confirmed again, however, modification should be done for increasing specificity. Criteria C appear to be the best for their sensitivity and specificity.
BACKGROUND/AIMS: The usefulness of the diagnostic criteria of the International Consensus Meeting (criteria A) has been previously reported. However, these criteria are not clinically adaptable in Japan where allergic reaction is one of the major etiologies of drug-induced liver injury and thus it was revised and reported in the Digestive Disease Week-Japan of 2002 as DDW-J criteria (criteria B). It remains controversial whether the revised criteria can exclude drugs not causing liver injury. METHODOLOGY: Two new diagnostic criteria (criteria C and D) were designed to supplement the DDW-J criteria. Usefulness and limitations of the four criteria were retrospectively examined using cases of drug-induced liver injury experienced in 8 hospitals. RESULTS: It was confirmed that the sensitivity of criteria B is excellent for diagnosis of drug-induced liver injury. However, diagnostic criteria B were found to be disadvantageous in relation to specificity, while diagnostic criteria D were disadvantageous in relation to sensitivity. Sensitivity of diagnostic criteria C was a little superior to that of diagnostic criteria A. CONCLUSIONS: On the basis, the significant sensitivity of criteria B was confirmed again, however, modification should be done for increasing specificity. Criteria C appear to be the best for their sensitivity and specificity.