Relationship Between Trimethoprim‐Sulfamethoxazole‐Induced Hepatotoxicity and Cytochrome Protein 450 2C9 Using Food and Drug Administration Adverse Event Reporting System Database

Background Trimethoprim‐sulfamethoxazole (TMP‐SMX) is the second most frequently implicated drug for liver injury. A previous study reported that its onset may be due to the activation of cytochrome protein 450 (CYP) 2C9. However, few studies have measured the expression levels of CYP2C9 in clinical settings; thus, the association between TMP‐SMX–induced liver injury and CYP2C9 has not yet been revealed. This study investigated the reporting frequency of liver injury associated with TMP‐SMX alone and in combination with a CYP2C9 inducer or inhibitor. Methods Disproportionality analyses using the Food and Drug Administration adverse event reporting system database were performed to compare the reporting frequency (reporting odds ratio ROR; 95% confidence interval CI) of drug‐induced liver injuries. The reference group consisted of patients administered TMP‐SMX alone. Rifampicin was used as the CYP inducer, and fluconazole, metronidazole, and voriconazole as the inhibitors. Drug–drug interaction (DDI) signals were calculated based on the Ω shrinkage measure. Results The reporting frequency of liver injury increased when used in combination with rifampicin (ROR = 16.90, 95% CI = 13.50–21.15). The concomitant use of inhibitors did not increase the reporting frequency of liver injury (fluconazole, ROR = 0.675, 95% CI = 0.583–0.782; metronidazole, ROR = 0.672, 95% CI = 0.585–0.772; voriconazole, ROR = 1.435, 95% CI = 1.237–1.665). TMP‐SMX plus rifampicin only demonstrated DDI signals of liver injury in the Ω shrinkage measure (Ω = 2.878, 95% CI = 2.607–3.148). Conclusions The present study provides insight into the reporting patterns of TMP‐SMX–associated liver injury with the concomitant use of CYP inducers and inhibitors and suggests a potential safety signal involving increased CYP2C9 activity. These findings should be interpreted with caution because disproportionality analyses cannot establish causality, and further clinical and mechanistic studies are warranted.