Validated analytical methods suitable for determining hyperforin in plasma after administration of alcoholic Hypericum perforatum extracts containing hyperforin are described. After oral administration of 300 mg/kg Hypericum extract (WS 5572, containing 5% hyperforin) to rats maximum plasma levels of approximately 370 ng/ml (approx. 690 nM) were reached after 3 h, as quantified by a HPLC and UV detection method. Estimated half-life and clearance values were 6 h and 70 ml/min/kg respectively. Since therapeutic doses of Hypericum extracts are much lower than that used in rats, a more sensitive LC/MS/MS method was developed. The lower limit of quantification of this method was 1 ng/ml. Using this method, plasma levels of hyperforin could be followed for up to 24 h in healthy volunteers after administration of film coated tablets containing 300 mg hypericum extracts representing 14.8 mg hyperforin. The maximum plasma levels of approximately 150 ng/ml (approx. 280 nM) were reached 3.5 h after administration. Half-life and mean residence time were 9 and 12 h respectively. Hyperforin pharmacokinetics were linear up to 600 mg of the extract. Increasing the doses to 900 or 1200 mg of extract resulted in lower Cmax and AUC values than those expected from linear extrapolation of data from lower doses. Plasma concentration curves in volunteers fitted well in an open two-compartment model. In a repeated dose study, no accumulation of hyperforin in plasma was observed. Using the observed AUC values from the repeated dose study, the estimated steady state plasma concentrations of hyperforin after 3 x 300 mg/day of the extract, i.e., after normal therapeutic dose regimen, was approximately 100 ng/ml (approx. 180 nM).

A crude hydroalcoholic extract from Hamamelis virginiana bark was subjected to ultrafiltration (UF) with a cut-off limit of 3 kDa to obtain a higher and a lower molecular weight fraction. Characterisation of the fractions was attempted with TLC, HPLC, acidic hydrolysis, and chromatography over Sephadex LH-20. The UF-concentrate was shown to consist mainly of oligomeric to polymeric proanthocyanidins (PA). This fraction was found to exhibit significant antiviral activity against Herpes simplex virus type 1 (HSV-1). In addition, the UV-concentrate displayed radical scavenging properties, inhibited alpha-glucosidase as well as human leukocyte elastase (HLE), and exhibited strong antiphlogistic effects in the croton oil ear edema test in the mouse. With the exception of the antioxidant potential and the inhibition of HLE-action the lower molecular fraction possessed weaker activities and contained mainly hamamelitannin, catechin, and further, unidentified constituents.

The bioavailability of ginkgolides A, B and bilobalide was studied in rats after single oral administration of 30, 55 and 100 mg/kg Ginkgo extract EGb 761. The plasma levels of the terpene lactones were measured by a specific GC/MS method. The pharmacokinetics of the mentioned substances were found to be dose-linear. For the lowest dose maximum concentrations were 68, 40 and 159 ng/ml and half-lives 1.7, 2.0 and 2.2 h for ginkgolide A, B and bilobalide, respectively. Clearance values ranged from 24.2 to 37.6 ml/min/kg.

To be effective, herbal medicinal products are expected to meet comparable standards concerning the assessment of efficacy, safety and biopharmaceutical quality as chemically defined synthetic drugs as food supplements. However, these requirements are often not fulfilled, particularly regarding the characterization of biopharmaceutical properties such as in-vitro dissolution and in-vivo bioavailability. With respect to the relevance of biopharmaceutical quality of herbal medicinal products, two different Ginkgo biloba brands (test product: Ginkgo biloba capsules; reference product: Ginkgold) were analysed for dissolution rates and bioavailability of the most relevant active ingredients. Dissolution rates at pH 1 and 4.5 were determined according to the USP 23. The relative bioavailability of ginkgolide A, ginkgolide B and bilobalide was investigated after single oral administration of 120 mg Ginkgo biloba extract as tablets or capsules. Bioavailability data (area under the curve and peak concentration in plasma) were clearly different and did not show bioequivalence of test and reference products. The slow in-vitro dissolution of the test product resulted in a large decrease in bioavailability. These results indicate for the first time that the pharmaceutical properties of a herbal medicinal product have a significant impact on the rate and extent of drug absorption, and very likely on efficacy in humans.

No abstract