Vitamin K antagonists (VKAs) are well established for the prevention of stroke or embolism in atrial fibrillation (AF) patients. VKAs are underused in AF, because of concerns about bleeding risk, drug and food interactions, and laboratory monitoring . There is a need for anticoagulant agents that would overcome these obstacles by being effective, safe and convenient to use. Dabigatran is a new oral thrombin inhibitor. The Re-Ly study, supported by Boehringer Ingelheim, was a non-inferiority trial in which fixed doses of dabigatran were compared with adjusted-dose warfarin in 18 113 AF patients . Dabigatran was found to have a similar effect as warfarin in stroke prevention, but with a lower complication rate . One of the advantages of dabigatran as compared with VKA should be its lower rate of drug and food interactions. However, dabigatran absorption is dependent on the P-glycoprotein (P-gp) system in the gut, which is influenced by drugs and food components [3–5].
The aim of our cross-sectional study in AF patients was to assess the prescription frequency of P-gp-affecting drugs. The embolic risk was assessed by the CHADS2 score, an acronym for congestive heart failure, hypertension, age > 75 years, diabetes mellitus, and prior stroke or transient ischemic attack.
P-gp-affecting drugs were searched for in the literature and are listed in Table 1 [3–5]. Consecutive patients with AF hospitalized between December 2009 and January 2010 were included. The CHADS2 score was obtained, and the medication was registered and screened for P-gp-affecting drugs.
|Drug||Food component, herb|
|Amiodarone||Black pepper (Piper nigrum)|
|Amitriptyline||Ginkgo (Gingko biloba)|
|Amprenavir||Ginseng (Panax ginseng)|
|Bromocriptine||Seville orange (Citrus aurantium)|
|Carvedilol||St John’s wort (Hypericum perforatum)|
|Chlorpromazine||Valerian (Valeriana officinalis)|
Among 100 patients (47 females, mean age 74 ± 12 years), heart failure was present in 53, hypertension in 70, diabetes in 33, and prior stroke in 14. The mean CHADS2 score was 2.4 ± 1.4. VKAs were received by 54 patients: 2/8 with CHADS2 0, 8/20 with CHADS2 1, 15/27 with CHADS2 2, 15/24 with CHADS2 3, 10/14 with CHADS2 4, 2/5 with CHADS2 5, and 2/2 with CHADS2 6. Forty-two patients took at least one P-gp-affecting drug, and 5/42 took two P-gp-affecting drugs: simvastatin (n = 22), amiodarone (n = 8), vitamin E (n = 8), carvedilol (n = 4), diltiazem (n = 2), dipyridamole (n = 1), propranolol (n = 1), and verapamil (n = 1). Twenty-six of the 54 VKA-receiving patients took one (n = 23) or two (n = 3) P-gp-affecting drugs.
Interactions between dabigatran and P-gp-affecting drugs have only been studied in phase I trials in healthy volunteers . Verapamil and amiodarone elevated dabigatran concentrations by 50–60% and clarithromycin by 19% . Food–drug and herb–drug interactions with dabigatran are unknown.
We conclude that 42% of hospitalized AF patients and 48% of VKA-receiving patients take P-gp-affecting drugs. Although most of these drug–drug interactions should result in weak clinical consequences, more information on the relevance of drug and food interactions is needed before dabigatran is widely used for stroke prevention. It should be investigated whether a dose reduction of dabigatran should be suggested in cases of coprescription of P-gp-affecting drugs. As the present study does not provide any evidence, we suggest to reanalyse bleeding and embolic events in dabigatran-investigating trials according to presence or absence of comedication with P-gp-affecting drugs.