5. Discussion
A good understanding of the mechanisms of drug-drug interactions is essential for assessing and minimizing clinical risks. Indeed many drug interactions are a result of inhibition or induction of CYP enzymes. This is especially true for many antineoplastic substances. However, interactions on P-glycoprotein and conjugation mechanisms should not be neglected. Additionally, pharmacodynamic interactions have to be considered.
Herbal medicinal drugs consist of multiple components. The complex nature of herbal drugs can provide broader information on multiple interaction mechanisms and the results may change due to environmental or manufacturing differences. More information is obtained than that with a single pure natural substance.
During the past decade, several potential mechanisms of interactions of valerian preparations, involving CYP enzymes, P-glycoprotein, and UGTs, have been studied in vitro [36, 38–42, 46]. Some of these studies pointed to a possible drug interaction potential by valerian extracts.However, critical assessment of these studies suggests that the clinical relevance of the findings is questionable due to various methodological limitations (Table 1). This is underlined by the available clinical interaction studies on CYP isoenzymes, which do not indicate a relevant drug interaction potential of valerian in healthy volunteers [35, 37].
In this context also the amiability of in vitro data to the situation in humans should be addressed. Especially the open questions of in vivo bioavailability and of the metabolism of extract components relevant for in vitro effects often prevent reliable extrapolations from in vitro to in vivo data. As in vitro and in vivo correlations usually are not available, results from in vitro interaction studies on herbal medical drugs may not simply be transferred to the in vivo situation. For valerian, and similarly also for Echinacea, Ginkgo, and hawthorn, clinically relevant interactions are lacking in in vivo studies, despite in vitro studies pointing to drug interactions in vivo. Since for herbal extracts a positive in vitro and in vivo correlation of data on drug bioavailability is rare, results from in vitro studies should be carefully interpreted [57].
Also the interpretation of metabolic studies in animals of different species should be critically evaluated, as often dosages far above those applied to humans are used and also regarding the enzyme variations in the species.
As for pharmacodynamic interactions of valerian, neither animal studies nor human data provide solid information for a possible risk [51, 52, 54]. Also the case reports are by no means convincing [55, 56]. At present therefore a relevant risk of pharmacodynamic interactions is not proven by valid clinical observations.
Altogether, this review could not identify studies showing a clinically relevant interaction effect of valerian. This confirms pharmacovigilance reviews of herbal medicinal products, which do not mention valerian at all [58–60] or claim valerian as safe [16, 27–31]. Block et al. [5] specifically pointed out that valerian is safe and efficient in patients undergoing cancer therapy. That therapeutic safety regarding interactions is high in valerian and is mentioned in the respective monograph of the HMPC [18] and also in a fact sheet of the Office of Dietary Supplements at the NIH [32]. In a data base for interactions of medicinal drugs (MedIQ), the rating “weak interaction” is given for CYP-isoenzymes, P-gp, and UGT, based on in vitro data. However, this rating is apparently the result of a merely formalistic approach and not the result from an analysis of the clinical relevance of these data.