Pharmacokinetic interaction of digoxin with an herbal extract from St John's wort (Hypericum perforatum)

Objective: Extracts of St John's wort (Hypericum perforatum) are widely use d in the treatment of depression, often as an over-the-counter drug. In con trast to its frequent use, knowledge about the pharmacokinetics of ingredie nts and drug interactions of St John's wort is poor. We studied the interac tion between hypericum extract LI160 and digoxin. Methods: The pharmacokinetics of digoxin were investigated in a single-blin d, placebo-controlled parallel study. After the achievement of steady state for digoxin on day 5, healthy volunteers received digoxin (0.25 mg/d) eith er with placebo (n = 12) or with 900 mg/d LI160 (n = 13) for another 10 day s. Digoxin concentration profiles on day 5 were compared with day 6 (single -dose interaction) and day 15 (tenth day of co-medication). Results: There was a highly significant combined-day-and-group effect for d igoxin area under the plasma concentration-time curve [AUC(0-24); P = .0001 ], peak concentration in plasma (C-max; P = .0001), and plasma drug concent ration at the end of a dosing interval (P = .0003) by two-way ANOVA. No sta tistically significant change was observed after the first dose of hypericu m extract [AUC(0-24) at day 6 of 18.1 +/- 2.9 mu g . h/L and 17.7 +/- 3.0 m u g . h/L, mean +/- SD for placebo and hypericum group, respectively]. Howe ver, 10 days of treatment with hypericum extract resulted in a decrease of digoxin AUC(0-24) by 25% (day 15, 17.2 +/- 4.0 mu g . h/L and 12.9 +/- 2.3 mu g . h/L; P = .0035). Furthermore, comparison with the parallel placebo g roup after multiple dosing showed a reduction in trough concentrations and C-max of 33% (P = .0023) and 26% (P = .0095), respectively The effect becam e increasingly pronounced until the tenth day of co-medication. Conclusion: As with grapefruit juice, a food product, physicians should als o be aware of potential drug-herb interactions. The interaction of St John' s wort extract with digoxin kinetics was time dependent. The mechanism invo lved may be induction of the P-glycoprotein drug transporter.