Objective: To determine the incidence and causes of drug-induced jaundice in a rural community.
Methods: A retrospective analysis of 800 patients presenting to a single-centre jaundice referral system serving a community of 400 000 over a period of 66 months (1998-2004). Standard criteria for drug-induced liver injury were applied to patients with a putative diagnosis of drug-induced jaundice. The incidence rates per prescription of drug-induced jaundice caused by co-amoxiclav and flucloxacillin were derived from local and national annual prescription rates.
Results: The incidence of drug-induced jaundice was 1.27 (confidence limits 0.85-1.8) per 100 000 per annum in a total of 28 patients (17 men, mean age 69 years). Antibiotics were the commonest cause of jaundice (n=21). Of these, co-amoxiclav (n=9) and flucloxacillin (n=7) caused the majority with an incidence rate per 100 000 prescriptions of 9.91 (4.6-18.0) and 3.60 (1.5-7.2), respectively. Co-amoxiclav-induced jaundice was observed more commonly in elderly males (age 65 years, M : F 7 : 2). In those patients with flucloxacillin or co-amoxiclav-induced jaundice, bilirubin ranged from 54 to 599 mumol/l (267 mumol/l) with a resolution of jaundice between 30 and 90 days. Counselling with regard to potential drug-induced liver injury and reporting of the adverse reaction had been performed in 1/28 patients.
Conclusions: 8.1% patients with no biliary obstruction and jaundice had a drug-induced and predominantly antibiotic-related aetiology particularly affecting an elderly population. We recommend that all patients receiving co-amoxiclav and flucloxacillin should be counselled before the therapy regarding the potential risk of jaundice and that an alternative antibiotic to co-amoxiclav is used if possible in men over the age of 60 years.