¹Department of Medicine, Indiana University School of Medicine, Indianapolis and Center for Drug Evaluation and Research and Office of the Commissioner, Food and Drug Administration, Rockville.

Correspondence: J. Christopher Gorski, PhD, Division of Clinical Pharmacology, Indiana University School of Medicine, Wishard Memorial Hospital, WD Myers Bldg, W7123, 1001 W 10th St, Indianapolis, IN 46202-2879, USA. E-mail: jcgorski@iupui.edu

^*Supported by National Institutes of Health grants T32GM08425 and M01-RR00750, Food and Drug Administration Cooperative Agreement FDT-001756, and a Merck Fellowship.

Received 9 July 2003; Accepted 16 September 2003.

Abstract

Background: Echinacea is a widely available over-the-counter herbal remedy. Tinctures of echinacea have been shown to inhibit cytochrome P450 (CYP) in vitro. The effect of echinacea (Echinacea purpurea root) on CYP activity in vivo was assessed by use of the CYP probe drugs caffeine (CYP1A2), tolbutamide (CYP2C9), dextromethorphan (CYP2D6), and midazolam (hepatic and intestinal CYP3A).

Methods: Twelve healthy subjects (6 men) completed this 2-period, open-label, fixed-schedule study. Caffeine, tolbutamide, dextromethorphan, and oral and intravenous midazolam were administered before and after a short course of echinacea (400 mg 4 times a day for 8 days) to determine in vivo CYP activities.

Results: Echinacea administration significantly increased the systemic clearance of midazolam by 34%, from 32 7 L/h to 43 16 L/h (P = .003; 90% confidence interval [CI], 116%-150%), and significantly reduced the midazolam area under the concentration-time curve by 23%, from 127 36 g h/L to 102 43 g h/L (P = .024; 90% CI, 63%-88%). In contrast, the oral clearance of midazolam was not significantly altered (P = .655; 90% CI, 75%-124%), 137 19 L/h compared with 146 71 L/h. The oral availability of midazolam after echinacea dosing was significantly increased (P = .028; 90% CI, 108%-153%), from 0.23 0.06 to 0.33 0.13. Hepatic availability (0.72 0.08 versus 0.61 0.16; P = .006; 90% CI, 73%-90%) and intestinal availability (0.33 0.11 versus 0.61 0.38; P = .015; 90% CI, 125%-203%) were significantly altered in opposite directions. Echinacea dosing significantly reduced the oral clearance of caffeine, from 6.6 3.8 L/h to 4.9 2.3 L/h (P = .049; 90% CI, 58%-96%). The oral clearance of tolbutamide was reduced by 11%, from 0.81 0.18 L/h to 0.72 0.19 L/h, but this change was not considered to be clinically relevant because the 90% CIs were within the 80% to 125% range. The oral clearance of dextromethorphan in 11 CYP2D6 extensive metabolizers was not affected by echinacea dosing (1289 414 L/h compared with 1281 483 L/h; P = .732; 90% CI, 89%-108%).

Conclusions: Echinacea (E purpurea root) reduced the oral clearance of substrates of CYP1A2 but not the oral clearance of substrates of CYP2C9 and CYP2D6. Echinacea selectively modulates the catalytic activity of CYP3A at hepatic and intestinal sites. The type of drug interaction observed between echinacea and other CYP3A substrates will be dependent on the relative extraction of drugs at hepatic and intestinal sites. Caution should be used when echinacea is coadministered with drugs dependent on CYP3A or CYP1A2 for their elimination.