Hepatocellular Drug-Induced Liver Injury Fulfilling Hy’s Law After Amoxicillin–Clavulanate Re-Exposure in an Elderly Saudi Man: A Case Report and Literature Review

Abstract

Background: Amoxicillin–clavulanate (AC) is the antibiotic most frequently implicated in idiosyncratic drug-induced liver injury (DILI) worldwide, and it characteristically produces a cholestatic or mixed pattern in older adults. A predominantly hepatocellular presentation is distinctly less common in this age group, and reports from the Arabian Peninsula are scarce. Case presentation: A 66-year-old Saudi man developed painless, progressive jaundice after a seven-day course of oral AC (500/125 mg three times daily) prescribed for a thumb laceration. He described a similar self-limiting jaundice after the same antibiotic approximately five years earlier, an episode that was never investigated and had not been entered in his allergy record, so the drug was prescribed again unknowingly. Liver biochemistry, documented as normal four months earlier, showed alanine aminotransferase (ALT) peaking at 587 U/L (14.3 × the upper limit of normal, ULN) and aspartate aminotransferase at 335 U/L, with total bilirubin 249 µmol/L (11.9 × ULN) and a 69% conjugated fraction, whereas alkaline phosphatase (ALP) never exceeded 245 U/L (1.9 × ULN). The R-value at the ALT peak was 19.3 using the same-day ALP, and 7.6 even when the peak ALT is paired with the highest ALP recorded at any point in the illness; the pattern was therefore hepatocellular (R > 5) throughout the injury phase. Hy’s law criteria were met, although the international normalised ratio remained 1.0 and no encephalopathy developed. Peripheral eosinophilia (peak 1.05 × 109/L) was documented. Hepatitis B surface antigen, hepatitis B core antibody, hepatitis C antibody and HIV serology were non-reactive; antinuclear, anti-smooth-muscle, anti-liver-kidney microsomal type 1 and antimitochondrial antibodies were negative and serum ceruloplasmin was normal; ultrasound with contrast-enhanced triphasic computed tomography excluded biliary obstruction and malignancy. Hepatitis A and E serology and cytomegalovirus/Epstein–Barr studies were unavailable at our institution, which we report as a limitation. AC was withdrawn and management was supportive. Liver biochemistry normalised completely over the following three months (ALT 25 U/L, ALP 103 U/L, and total bilirubin 22 µmol/L). Updated Roussel Uclaf Causality Assessment Method (RUCAM) scoring on the hepatocellular scale gave 7 points (probable); no credit was taken for the historical episode, since RUCAM requires documented enzyme re-elevation for a positive re-administration response. Conclusions: AC can cause hepatocellular, Hy’s-law-positive injury in older adults, not only the cholestatic phenotype emphasised in the literature. The decisive failure here was administrative rather than diagnostic: an adverse drug reaction that is never recorded will be repeated.