St. John’s Wort Reduces Digoxin Levels

Abstract & commentary

Source: Johne A, et al. Pharmacokinetic interaction of digoxin with an herbal extract from St. John’s wort (Hypericum perforatum). Clin Pharmacol Ther 1999;66:338-345.

Due to the increasing popularity of alternative or natural treatments, St. John’s wort (Hypericum perforatum) has become widely used in the treatment of depression. Despite being largely available as an over-the-counter drug, very little is known about the pharmacokinetics of its ingredients and/or its drug interactions. With a single-blind, placebo-controlled parallel design, the investigators studied the interaction between hypericum extract LI160 (the putative active ingredient in St. John’s wort) and digoxin. The LI160 preparation contained 92 mcg of hypericin per 300 mg tablet of dried hypericum extract.

All subjects, who were instructed not to smoke or consume alcohol, coffee, tea, cola beverages, or drugs, received a loading dose of digoxin 0.25 mg twice daily for two days then once daily thereafter. After the achievement of steady state for digoxin on day five, healthy volunteers who were 22-33 years of age received digoxin (0.25 mg/d) either with placebo (n = 12) or with 900 mg/d LI160 (n = 13) for another 10 days. Digoxin concentration profiles on day five were compared with day six (single-dose interaction) and day 15 (10th day of co-medication).

The effect of a single dose of hypericum extract on digoxin kinetics did not achieve statistical significance, although the digoxin levels were slightly higher than the placebo group. After 10 days of co-medication, trough levels of digoxin were reduced by ~33% in the hypericum group. Maximum levels (Cmax) and the area under the curve of digoxin were similarly affected (reductions of 28% and 26%, respectively). These parameters were also significantly lower compared to those for the placebo arm. Cotreatment with hypericum did not affect the elimination half-life of digoxin, suggesting that a mechanism other than hepatic enzyme induction. The effect on digoxin seemed to be time dependent (maximum reduction not present until the 10th day). This fact, combined with the fact that the single-dose effect of hypericum co-administration did not reduce but rather raised digoxin levels, suggests that the mechanism was not due to an impairment of absorption by physicochemical binding of digoxin and hypericum in the gut.

Comment by Michael F. Barber, pharmD

The current study is important for several reasons. First, the reduction of digoxin concentrations by hypericum is clinically significant since patients may lose efficacy of digoxin when St. John’s wort (SJW) is taken concomitantly. Patients who claim that they have been adherent to their digoxin regimen may present with lower serum digoxin concentrations despite the fact that their dosage has not been changed. Alternatively, patients who have achieved therapeutic levels of digoxin while taking SJW may develop digoxin toxicity once they discontinue SJW.

Another important point about this finding is that this pharmacokinetic study illustrates the importance of studying both the acute and chronic effects of drug interactions. The acute effect of this combination, although not statistically significant, may have suggested to clinicians that SJW may raise serum digoxin levels, whereas the chronic administration showed an impressive reduction in serum digoxin levels. The findings of this study also suggest that SJW does not induce hepatic metabolism of digoxin. Marangell1 has previously discussed that SJW is an inducer of CYP3A4, and can lower serum levels of substrates of that enzyme. However, hepatic metabolism of digoxin constitutes a relatively minor pathway. A large portion of digoxin metabolism takes place in the intestine. The intestinal degradation of digoxin may be mediated by P-glycoprotein, a product of the multiple-drug-resistance gene MDR1, which transports digoxin into the gut from the blood. Thus, the mechanism by which SJW reduces serum digoxin levels may be via induction of the P-glycoprotein transporter. This is consistent with several other CYP3A4-inducing drugs, such as rifampicin, which also induce the expression of P-glycoprotein.

Finally, the current study is yet another example of a drug interaction involving a non-prescription product. With the current increase in popularity of alternative, over-the-counter medications (particularly herbal "remedies"), clinicians should be aware of patients taking such products as well as the possible interactions between the products and prescription medications.


1. Marangell LB. St. John’s wort may alter drug levels. Psychiatric Medicine in Primary Care 1999;1(3):17-18.