Managing and Interpreting the Complexities of Botanical Research

Tagged:  

This paper is based on a keynote address delivered by the author at the Cork Institute of Technology & Irish Institute of Medical Herbalists Conference in Cork, Ireland, April 18, 2008.
Summary

The need to examine botanical medicines scientifically is a compelling requirement in modern evidence-based culture. However, not all botanical research is necessarily valid. Good quality research must consider and document the specific character of the botanical preparations being studied and assess the results in the proper context. The importance of accurately identifying the herbal preparation and content being studied cannot be overstated, as this information is necessary for the scientific interpretation of the results. The inherent limitations of preclinical research approaches should be explicitly acknowledged, especially when attempting to apply the findings to normal human use and/or clinical practice. Human clinical trials can provide some solid evidence for specific preparations, but conventional research methodologies do not necessarily apply to traditional herbal practice. In nearly every type of research on complex botanicals, utilizing 2 or more distinct forms of well-characterized preparations from a single herb is an especially effective and practical means of assessing their comparative value. Human studies on the pharmacokinetics and pharmacodynamics of different preparations from the same herb may be the most useful for herbal practitioners who prefer to develop individualized prescriptions.
Introduction

The scientific research protocols for determining clinical efficacy of drugs may be insufficient for botanicals for several reasons;( n1) however, good quality research to validate efficacy and compare outcomes of conventional medicines, botanicals, and/or their combinations is necessary for the widespread acceptance of botanical products.( n2) The question of what constitutes good quality research requires different answers for botanicals than for pharmaceutical drugs, largely due to the phytochemical complexity of botanicals versus the comparative chemical simplicity of drugs, most of which are single chemical entities.( n3)

Simply identifying research outcomes of botanical derivatives by the name of the mother herb may be the single greatest obstacle that researchers and reporters must overcome in regards to generalizations in botanical discussions. The varying content and proportions of active components in different types of extracts requires that they be seen as separate entities--children of the same mother but with their own particular identities. Due to the limited amount of scientific research on most botanical species, the utilization and consolidation of any and all published observations and research findings on an herb and its derivatives can at times be described as "grasping at straws." Commonly, the unfortunate result is the construction of a straw house.

There are 2 main questions regarding research results and their interpretation that will be addressed in this discussion. Since variations and uncertainties exist in regards to different types of preparations, what are appropriate means for characterizing these differences? And given the limitations of applying standard scientific protocols for drugs to botanicals when attempting to establish their safety and efficacy and provide accurate mechanistic rationales, what methods of research are achievable and appropriate to develop practical knowledge of modern and traditional botanicals?
Botanical Product Variations and Characterizations

Research that attempts to treat the study of botanicals as if they were drugs often fails to make important distinctions that clarify exactly what preparation is being studied and how the results can best be assessed from the perspective of multifactoral, versus linear, influences. There needs to be greater transparency in communicating the accurate identity of each preparation undergoing research. (See Table 1.)

Due to the vast complexity of the phytochemical content of herbs and most botanical preparations, utilizing the proper scientific taxonomy for the herb is of critical importance. Reliance on botanical experts to establish identity through plant morphology, organoleptic testing, and/or chromatography is also essential for assuring meaningful data from research. Results are scientifically valid only if the material studied can be positively identified and verified so as to be able to reproduce the research. Techniques for quantitative phytochemical assessment should be form-appropriate. For example, in comparing herb versus extract contents with phytochemical assays, the methods of solubilizing solid preparations may need to differ, such as requiring exhaustive Soxhlet extraction for certain herbs rather than merely sonication that is adequate for simply dissolving solid extracts.(*)

The key to accurate characterization is to verify, rather than to presume, knowledge of the preparation and its use. Adequate documentation involves preservation of a voucher specimen. Description of processes and/or solvent proportions is necessary, but in cases of research on proprietary products where this information remains undisclosed, at a minimum the lot number and expiration date must be given. In the context of in vivo studies, errors in designing research often occur in failing to assess the propriety of dosage regimens and the relative applicability of particular dosage forms that can be pertinent to their efficacy.

Fundamental aspects of preparation and dosing characteristics can influence outcomes and thus need to be specifically described in detail. Providing precise descriptions of content and use helps establish a sound basis for outcome comparisons, so research design and reporting need to take such factors into account. For example, do research reports address issues such as the relative dilution or concentration of extracts? Not only should the use of 'x' ml or 'y' mg of extract be noted, but also that these are derived from 'z' mg of herb. (Often when this is done, unfortunately the phrase "equivalent to," rather than "derived from," is used to express the relation between the amount of extract and original herb, suggesting an unfounded assumption about the activity of these quantities of the herb and its extract.) When comparing studies, one must assess dosage to see whether design variations are appropriate to account for the size or age of subjects, acute versus chronic therapy, prevention versus therapy, etc. If botanicals are used in combination, these concerns of documenting the preparation for each species still apply.

Without the particular qualifying features being explicitly discussed, meaningful conclusions and comparisons become implausible.( n4) The modern era of botanical research requires a sophisticated understanding of identification( n5) to avoid misapplication of research outcomes borrowed from studies of other preparations.( n6) The US Congress has directed the Office of Dietary Supplements at the National Institutes of Health to establish a validation process for analytical methods, analytical standards, and reference materials.( n7) Along with providing the identity and origin of the herb being studied, the application of proper processing and phytochemical nomenclature to accurately describe the final product being studied is demanded by those who rely on scientific criteria to evaluate relative merits of different preparations from the same botanical species origin.( n8)
Need for Comparative Studies to Assess Phytochemical Differences of Similar Botanicals

By determining and describing similarities and differences in defined variables, distinctive identities of particular botanical preparations can be more accurately discerned. Since studies of a single complex botanical agent yield isolated data that is relatively limited in extrapolation to other preparations, the comparison of outcomes from 2 or more well-characterized preparations undergoing the same means of assessment can greatly help in identifying those features that most influence the results. Research on what is simply called "echinacea" (Echinacea spp., Asteraceae) illustrates this point nicely. The various species of echinacea and their plant parts each have their own chemical fingerprints, and preparations made from the different parts of the echinacea plant or according to different dosage forms have been found to have different chemical properties (See Table 2.)
Limitations of in vitro Laboratory Research

Herbs remain a major source of compounds that are screened to develop modern drugs for neurological, microbial, inflammatory, and neoplastic pathologies. The production of laboratory data on botanical activity as preliminary research for human outcomes has commonly been intended for the development of such drugs. That is, the focus is on sequentially reducing the complexity of phytochemical bioactivity by selectively partitioning extracts into fractions and subfractions and then into isolates, ultimately developing semi-synthetics, thereby increasing chemical purity and potency rather than maintaining complexity and balance.

In vitro research can function as a means of demonstrating phytochemical differences between botanical preparations. However, a common error is to assume in vitro findings for complex mixtures ultimately apply to their in viva influence. (See Table 3.) Direct exposure in vitro to a complex mixture of bioactive therapeutic agents does not reflect systemic tissue exposure to altered and partially absorbed component combinations following oral dosing. In vivo pre-absorption digestion and/or metabolism by gut flora can chemically change phytochemicals, as does post-absorption catabolism and/or conjugation, resulting in reduced absolute concentration and/or bio-activation or inactivation. What is more, variable absorption can lead to significant, and even dramatic, systemic proportional differences between compounds or entire classes of phytochemicals, compared to the botanical content consumed.

In vitro studies of an isolated phytochemical may be a useful tool in assessing partial botanical activity when this bioactive component has an established pharmacokinetic profile and known tissue bioavailability following oral dosing. However, long duration of exposure and steady concentration in vitro misrepresent in viva fluctuations, as inevitable changes occur due to ongoing metabolism and excretion. The effects of a steady-state isolate concentration in vitro are often inappropriately correlated to its peak tissue or serum level achieved in vivo.

In regard to the target tissues, cells separated from the extremely dynamic human ecology of systemic biochemical interactions are relatively static templates in comparison to the in viva milieu. In vitro results for botanicals appear most applicable to in viva effects when they involve cells to which the complete phytochemical complex of a botanical can be applied directly, i.e., skin or mucosa and conditions involving these tissues. For example, extrapolation of in vitro antimicrobial activity is largely limited to local botanical applications.
In vivo and ex vivo Laboratory Research

Findings from pre-clinical laboratory studies are simply suggestive, or at best supportive, of systemic clinical effects of a particular botanical preparation. Very high doses are often used with lab animals to produce an overt measurable effect and are typically injected to avoid dosing variability. Animal studies provide stronger supportive evidence when oral botanical administration (including gastric intubation) is utilized and when dosage is proportional to human use. Such approaches better reflect normal clinical applications and human tissue exposure. While these issues also apply to synthetic drugs, the multiplicity of active agents in complex botanical preparations produces greater challenges for correlating human outcomes.

Animal physiology and/or metabolism may not allow accurate extrapolation to humans. Interspecies genetic variations and subsequent biochemical peculiarities alter pharmacokinetics and pharmacodynamics. For example, rats and rabbits do not have the same CYP isozymes as humans, though the isozymes from pigs, monkeys, and humans are similar.( n24) The strongest indictment against depending on botanical results exclusively from in vitro and animal CYP isozyme inhibition studies is the lack of validation in many comparable human oral studies.( n25)

Human ex vivo studies can be useful for verifying activity after oral dosing but are limited to accessible fluids and cells (blood, urine, bile, feces) or surgically-excised pathological tissue. For example, testing antimicrobial activity of urine collected after oral consumption of-a botanical urinary tract antiseptic would provide better direct evidence than testing the extract in vitro. These types of post-absorption human pharmacological mechanistic studies can provide useful data for relatively little expense, risk, and subject inconvenience.

Outside of empirical data established clinically in humans for a particular preparation, applying the findings of almost any other research (positive or negative) to humans can be challenged on the basis of inherent limitations and constraints of the methodology. Given this limitation, however, in vivo laboratory data clearly indicative of botanical toxic effects applicable to human physiology and usage patterns should be given relative credence unless contradicted by human studies. When it comes to clinical applications, the premise of primum non nocere ("first, do no harm") should be paramount.
Botanical Issues Illustrated in Human Clinical Studies

Though research data on botanicals from the same species are typically compiled under the heading of the herb from which they are derived, this does not mean that research conclusions regarding a particular botanical preparation should be applied to other dissimilar preparations. (See Table 4.) When several human studies examining different types of preparations of an herb confirm their efficacy for a given application, this is supportive evidence (not proof) for similar extracts and the herb itself, as long as dosages provide adequate delivery of effective components. On the other hand, when divergent results are found, the distinctions between preparations are compromised in a "meta-analysis" approach of compiling research from different botanical products that fails to take into account the variations in processing, contents, and established clinical applications. Consequently, these details must be examined.

Eighty-one English-language MEDLINE-indexed studies that were published between Jan. 1, 2000, and Feb. 9, 2004, were selected based on being randomized, controlled, single-botanical trials of 1 of the 5 most popular herbs: echinacea, garlic (Allium sativum, Liliaceae), ginkgo (Ginkgo biloba, Ginkgoaceae), saw palmetto (Serenoa repens, Arecaceae), and St. John's wart (Hypericum perforatum, Clusiaceae). When these studies were assessed for preparation characterization, only 40 (49%) identified the plant source by its scientific Latin binomial, and only 8 (10%) identified the plant part used. In addition, only 23 (28%) reported the processing and/or extraction method used. In 12 (15%), quantitative analysis was performed to determine actual contents and this information was published. In a randomized survey of 40 of the authors who had published no quantitative analysis, several cited editorial decisions for not doing so. While 3 authors noted lack of funding, 17 of the 23 principal investigators who responded to the survey indicated that it was because a claimed "standardized" extract was used. Of these 81 trials, 54 (67%) were assessed as good by Jadad scores (? 3), though obviously many of these were compromised by failure to adequately describe the preparation utilized. Quantitative analysis was performed more frequently in studies with lower Jadad scores (7/27, 26%) than those with good scores (5/54, 9%). The use of preparations claiming to be standardized in 36 trials (44% of studies) was not a definite advantage, since the claimed contents were not verified.( n26)

Label content and standardization claims have not always proven reliable. Investigators assessed content claims for 59 preparations sold as echinacea in August 2000 in the Denver metropolitan area. They reported that 9 products claimed standardization to specific polyphenolic markers (the caffeic acid derivatives echinacoside for E. angustifolia and E. pallida or cichoric acid for E. purpurea), though none had sufficient quantity to match the label claim. Seven had on average only 26% of the label claims, and 2 contained none.( n27) The failure of this report to specify plant parts compromises species identification based on single polyphenolic markers, as does the fact that phenolic "markers" in fresh echinacea juice preparations are rapidly degraded by enzymatic activity and therefore are absent. Nonetheless, some of the observations are worth noting.

In 10 of the products evaluated, the labels on 4 made no particular species claim, and 6 contained no measurable polyphenolic markers. Based on qualitative thin layer chromatography assays of the other 49 preparations from Echinacea species verified by their polyphenolic markers, species identification matched the label species claim for 21 of 30 (70%) of the non-standardized products. Only 10 of 19 (53%) of the products claiming standardization were actually verified as the appropriate species. Quantitative analysis showed that the non-standardized products contained on average over 2 times more of the species markers than the standardized samples.( n27) While the recently-implemented good manufacturing practices being required of all US herbal dietary supplement manufacturers will help greatly to resolve such problems with commercial products in the United States, it does not relieve researchers of the responsibility of independently verifying product content.

Failure to document content of bioactive phytochemicals in product-specific human studies makes them more difficult to interpret.( n25) For example, E. purpurea root extract given orally at 1.6 g/day for 8 days to 12 subjects reduced bioavailability of intravenous midazolam, apparently by inducing liver CYP 3A. However, no change in clearance was detected with oral midazolam, a result that suggests that this herbal extract may also inhibit intestinal CYP 3A. This root extract increased bioavailability of oral caffeine, suggestive of inhibition of CYP 1A2, but no inhibition was detected for CYP 2D6 substrate dextromethorphan.( n28) On the other hand, E. purpurea whole plant extract administered at 1.6 g/day orally for 28 days to 12 subjects had no significant effect on the CYP metabolism of oral midazolam (3A4), caffeine (1A2), or debrisoquin (2D6).( n29)

These apparent discrepancies may be theoretically explained by differences in the phytochemical content of the different preparations, with total alkamide and C12 diene alkamides in particular being in much higher concentration in the roots and C12 tetraene alkamides higher in the leaves.( n10) Inhibition of CYPs 3A4 and 2D6, but not 1A2, has been shown to be related to total E. purpurea alkamides and the C12 tetraene alkamide content with a whole plant extract and 7 other E. purpurea preparations in vitro.( n30) Therefore, the actual CYP effects of E. purpurea extracts and their alkamide compounds in viva are demonstrably different than those found in vitro. These components appear quite pertinent since echinacea alkamides are detected in the serum after oral administration, while the phenolic caffeic acid derivatives are not.( n11) Nonetheless, the polyphenolic influences on CYP isozymes in the gut mucosa may be just as significant, or moreso. Without documenting the research product content used in human studies to allow for comparisons of phytochemical complexes, assessing outcomes remains not merely challenging but an unrelenting guessing game.

As these examples illustrate, research can best establish meaningful and conclusive knowledge of similarities and differences between preparations from the same herb if it utilizes direct comparative pharmacological or therapeutic trials of well-characterized products. This is convincingly demonstrated by human studies that administered a wide variety of St. John's wort preparations. These studies showed the preparations' impact on inducing both the major drug metabolism isozyme CYP 3A4 and that the cellular efflux drug transporter P-glycoprotein is hyperforin-dose-dependent.( n31, n32) Unfortunately, such direct comparative human studies of different preparations from the same herb are rare. This makes it even more imperative that reviews of clinical studies identify and assess distinctive parameters that help determine relationships between preparations that render them more or less effective.

For example, confusion is regularly expressed about conclusions from research in regard to the efficacy of echinacea preparations used for treating upper respiratory tract infections related to colds and influenza. When comparing the preparations studied, it becomes clear that in single-species studies of echinacea preparations for the treatment of acute upper respiratory viral infections in adults, the liquid extracts of the fresh E. purpurea aerial plant or whole plant yield consistently beneficial outcomes.( n33) Simply drying the same batch of this herb by different methods changes the bioactive content. Freeze-drying preserves caffeic acid derivatives in fresh E. purpurea flowers significantly better than air-drying at 40°C, which is better than air-drying at 25 or 70°C.( n34) Similarly, freeze-drying preserves the alkamide content better than air-drying at 70°C in E. purpurea roots but not in its leaves.( n35)
General Aspects of Clinical Research Strategies and Botanical Outcome Interpretations

There are a number of issues that can compromise clinical study results with which those who design research and those who interpret it must contend, whatever the agent being tested. Botanicals may be even more vulnerable to such compromising factors due to the inherent complexities of their poly-phytochemical nature, the less overt impact of some of the phytochemical effects, and the receptivity of those using the intervention.

When conducting research using subjects with specific diagnoses, the researcher must ensure the appropriateness of the tested botanical preparation's form. The form of a preparation may or may not be conducive to a positive outcome, based on content, bioavailability, or patient psychosomal preference. Also, subjects' individual genetic, physiological, and pathological variations could alter their response to a botanical, despite having the same diagnosis. Short-term pharmacological studies, meanwhile, may not represent an actual pattern of use. Single-dose pharmacokinetic or pharmacodynamic assays may not replicate the response that occurs with repeated dosing, especially with mild botanical effects. For example, CYP isozyme induction typically requires over a week of exposure to a botanical. It is important to note that pharmacokinetic diversity due to genetic polymorphisms is more likely to have some impact on the activity of botanicals, since they contain multiple bioactive components whose metabolism can potentially be influenced. Also, the pharmacodynamic effects can depend on whether they were tested in healthy or non-healthy individuals.

Still other features of clinical research protocols are frequently overlooked that can impact interpretations of the data and/or conclusions about its value. These can be especially significant for assessing the relative value of botanicals. (See Table 5.) All attempts should be made to optimize the processes that allow for more reliable data collection and interpretation.

Over the last several decades, increases in the number of botanical clinical studies have led to progress in recognizing features of trial design that serve to better validate outcomes and improve interpretation of results. For establishing efficacy in healthcare interventions, randomized / controlled / blinded trials have become the preferred means of clinical validation, involving methodologies that can be effectively applied to botanicals. However, as has been repeatedly stated, without adequate identification and reporting on the preparation utilized, interpretations of the process and data are compromised. There remains an urgent need to accurately communicate details regarding the product studied, its content of known active constituents, and specification of the means used to establish these criteria, as well as the larger context behind a product's selection for study. These points have been made previously as expanded recommendations for consolidated standards of reporting trials (CONSORT) specific to botanicals.( n36)

"Mining the data" for analysis of a product's application to specific presentations found in patients is optimized when trials report a thorough intake history, physical, and lab work-up for each individual to help possibly correlate phenotypic data with outcomes. Inherent variations in effective doses within a patient population suggest utilizing an adjustable protocol or a trial follow-up comparing effects after dosage adjustment. This approach allows increasing dosage for those without efficacy and decreasing it in those with adverse effects to observe the impact of these changes on effectiveness and safety. Such research also more accurately mimics clinical practice. Uncontrolled clinical trials should be encouraged, since they can provide a useful first step in evaluating these parameters and assist in developing a more sophisticated controlled clinical study.( n37) Among the different types of possible clinical study designs, each has its own specific advantages and relative limitations.( n38)

For assessing the relative benefits of single-species botanical preparation options, It is highly desirable, if not imperative, especially in preliminary clinical studies with fewer cost demands, to design a trial with several arms that allows comparative assessment of outcomes. Whether this is done on the basis of variations in herbs, dosages, specific herb parts, or different solvent extraction products, the opportunity to observe and describe differences from adjusting these variables in an otherwise standard protocol is possibly the most helpful means of establishing parameters to optimize botanical selection. When the different preparations utilized are appropriately characterized on the basis of herb production, sourcing, and verification, phytochemical fingerprinting, and extract processing details, then valid comparisons and reasonable conclusions can be made more reliably.

Clinical research using conventional drugs as bioactive controls provides the opportunity for comparisons of adverse effects as well as therapeutic and/or pharmacological outcomes.( n39) For example, in evaluating botanical effects on drug metabolism, statistically significant effects do not necessarily result in clinically significant effects. Clinical trials on botanicals with positive drug controls also serve as a means of assessing potential research design flaws, as when the positive control arm fails. Herbal therapies typically differ mechanistically from conventional pharmaceutical approaches. Though botanical influence may not be as overt, adverse effects are also generally less frequent and/or less severe. These differences provide incentive not only for comparative trials with drugs to examine efficacy and safety but to also incorporate a combination arm in the trial to determine if concurrent use can enhance outcomes and/or reduce adverse effects, e.g., by lowering the effective drug dose. A number of recent studies such as those on anti-inflammatory and anti-hyperglycemic botanicals have been performed with patients using similar drugs as part of the inclusion criteria. This under-reported feature should be emphasized, rather than overlooked.

Intervening with botanicals in patients who are reliant on lifesaving and/or vital organ-sustaining drug therapy can be controversial. Where potential complementary effects appear to enhance therapeutic outcomes, safety issues remain a top priority. Pharmacodynamic botanical-drug interactions are reasonably predictable and can often be assessed in animal studies. Pharmacokinetic interactions are largely unpredictable unless specific mechanisms have previously been demonstrated through human research. To responsibly consider combining botanicals with drugs, the potential for both types of interactions and their clinical relevance should first be evaluated. In researching the concomitant use of botanicals and drugs, identification of the drugs, their doses, dosage adjustments, and durations of use or percentage of discontinuations should be detailed, if possible. Adverse effects and their amelioration should both be reported as significant secondary outcomes for these combination studies.

Clinical research typically fails to address the historical application of complex herbal products in an individualized holistic context. Studying botanical efficacy as the mean response of a group that shares a particular diagnosis or symptom follows the conventional medical therapeutic paradigm. In contrast, traditional herbal medicine systems evaluate and treat cell, organ and/or system functions of an individual, depending on their particular expressions of associated physiological deficiencies and/or excesses. Differentiating disease expressions beyond standard diagnostic categorization enables the selection of the most appropriate botanical options. The evaluation of botanical impact on particular disease expressions in clinical trials can be facilitated by identifying distinguishing symptomatic features in the exclusion or inclusion criteria of the studies.

Establishing the human pharmacological effects of a particular botanical preparation may be one of the most useful approaches to assist herbalists, since traditional prescribing criteria and blending of remedial agents goes beyond applying a botanical monotherapy to a conventionally diagnosed condition. As a means of identifying some of the potential impact of botanicals, the employment of human pharmacokinetic studies and ex vivo research, along with non-invasive evaluation techniques and monitoring changes in physiological function and/or performance, can effectively aid in evaluating and comparing botanical preparations and their clinical combinations for supportive clinical interventions.( n40)
Conclusions

The challenge and promise of botanical research was well-stated in a recent comparative "laboratory anti-inflammatory study for extracts from Echinacea spp. roots: "It is tempting to consider the diversity of these plant genera and the complexity of their constituents as barriers to understanding their potential health benefits. However, the range of variation in these plants, when systematically analyzed, provides a strong foundation on which to develop the strategies and tools needed to produce the most efficacious products for a growing body of consumers."( n41)

Only a few dozen botanicals have been thoroughly studied by modern scientific methods. Besides evaluating processing influences on botanical composition, pharmacology and selective therapeutic efficacy, the issues of safety and drug interactions have become important topics in the discussion on the need for improving approaches to research. Further human studies on phytochemical pharmacokinetics and the impact of botanicals on drug pharmacokinetics are central to assessing these features. The complexity of content and activity involving botanicals from the same genus or species make studies that compare their similarities and differences of paramount value, especially when evaluating their clinical impact. Possibly the best means to analyze and interpret outcomes is to be able to compare and contrast the effects of several well-characterized botanical preparations in the same study.

Herbal science applied to traditional holistic practice addresses patient individuality beyond simply monitoring the expression of similar pathological characteristics shared by a group. Every person is much more than his or her medical diagnosis. In an analogous fashion, botanicals and their complexity must be considered similarly distinct and deserve to be studied as such.
Glossary

For the purpose of this discussion, the following 3 words are to be interpreted as follows:

Herb: a living or dried plant or its complete medicinal part(s), including the insoluble fiber content, whether intact or fragmented, taken to improve health. (These plant forms of medicine are designated as "herbal drugs" in the European Pharmacopoeia.)

Botanical: an herb or one of its complex therapeutic derivatives, i.e., an extract or fraction that consists of a variety of component chemicals. (The extractives are designated as "herbal drug preparations" in the European Pharmacopoeia.)

Drug: an isolated bioactive compound, whether natural or synthetic, used for medicinal purposes irrespective of its pharmacopeial recognition. (For the purposes of this paper, herbs or other botanicals with official pharmacopeial status, whether available by prescription-only or over-the-counter, will not be incorporated under this term.)

(*) Soxhlet extraction utilizes a Soxhlet apparatus by which constituents can be extracted from solid matter by repeated treatment with distilled solvent. Sonication uses a device (sonicator) to apply ultrasound energy that agitates particles in solution to speed dissolution by breaking intermolecular interactions.
Table 1. Appropriate Botanical Characterization and Inadequate Means of Describing Herbal Preparations(n2)

Appropriate Characterization

a. Identified by scientific binomial
b. Described plant part used
c. Verified species identity
d. Established chemical profile
e. Specified preparation precisely
f. Detailed dosage exactly
g. Noted details of administration

Inadequate Description

Using common name only
Giving only herb or manufacturer's name
Accepting supplier's or manufacturer's claim of identity
Assuming chemical content or label claim
Calling it simply an "extract" or "tablet"
Listing "tablets" per dose or only daily dose
Not identifying when or how it is taken

Table 2. The Chemical Complexity of Different Echinacea Species, Plant Parts, and Preparations

a. Species phytochemical variations –

Extracts (80% methanol / 20% water) of 3-year-old echinacea roots yielded these total phenolic amounts and major/minor caffeic acid derivatives:

E. purpurea 23.2 mg/g (cichoric acid / caftaric acid), E. pallida 17.8 mg/g (echinacoside / cichoric acid, caftaric acid), E. angustifolia 10.5 mg/g (echinacoside / cynarin).( n9)

b. Alkamide content of plant portions –

Alkamides were highest in E. angustifolia roots and absent in E. pallida root. E. purpurea roots/rhizome are much higher in alkamides (especially C12 diene-diynes) than its leaves, while the aerial (above-ground) plant has proportionally more C12 tetraenes.( n10)

c. Absorption and bioavailability –

Alkamides from tablets of ethanolic extracts from E. purpurea and E. angustifolia are readily absorbed and identified in the plasma, whereas caffeic acid derivatives in the tablets are not.( n11) Maximum serum concentration of alkamides from these echinacea root extracts occurs more rapidly from 60% ethanol liquid extract (20 min.) than from tablets (30 min.)( n12)

d. Metabolism –

Echinacea alkamides are metabolized variably by human liver cytochrome P450 (CYP) microsomes in vitro, but metabolism rates differ depending oh individual structural chemistry and whether they are isolated or combined.( n13)

e. Herb components versus extract content –

Whereas Echinacea spp. (E. angustifolia, E. pallida, E. purpurea) water and ethanolic extracts lack melanin, melanin in the herb ( 5-10% of plant dry weight) has bioactive mechanistic implications. Echinacea melanin activates monocytes in vitro by binding to toll-like receptor 2, and after being fed to mice melanin increased immunoglobulin (Ig)A from Peyer's patch cells and interferon (IFN)-? from spleen cells ex vivo.( n14)
Table 3. Results of in vitro Research on Echinacea spp. Contrast with Results Observed in vivo

a. Complex extracts used in vitro do not match the post-digestion exposure from extracts used in vivo.

In a series of tests, different types of Echinacea spp. extracts tested in vitro failed to enhance mononuclear cell proliferation and macrophage activation until they had undergone a simulated digestion process.( n15)

b. Peculiar preparation or noncommercial solvents of in vitro samples can result in artifacts or atypical effects not found with ordinary use.

Increased T-cell proliferation and cytokine secretion in vitro is found with Echinacea spp. root cold infusions after refrigeration for 4 days (but not with hot water infusions or 50% ethanolic tinctures) due to microbial endotoxin contamination.( n16) These xenobiotic compounds and subsequent reactions would not occur with standard preparations used in vivo. A 90% aqueous phenol solvent is required to extract melanin from Echinacea spp. that enhances immune responses in vitro.( n14) Solvents employed in research but not used in producing commercial preparations can provide phytochemicals that are not available in commercial botanical extracts and thus would not contribute to the effect in vivo.

c. Some isolated compounds or subfractions used in vitro misrepresent post-absorption influence.

High molecular weight heteropolysaccharides isolated from 30% ethanolic extracts of E. angustifolia, E. pallida, and E. purpurea roots by ultrafiltration demonstrate mitogenic activity directly on spleen cells( n17) that would not ordinarily be exposed to such large compounds following oral administration. The phytochemical matrix as it exists in botanicals may also enhance or reduce the activity of isolated components.( n13)

d. Types of cells or tissues employed for in vitro research can limit expression of critical processes.

Healthy cells (normal physiology) can differ in critical responses to those of diseased specimens (pathophysiology). This was demonstrated when E. purpurea plant juice and root 50% ethanolic extract reduced cytokine/chemokine secretions from rhinovirus-infected epithelium but increased them in normal epithelial cells.( n18) Laboratory research utilizing normal tissue cells can thereby result in misleading implications if it is theoretically applied to the therapeutic application of the same preparation to cells in tissue involved in pathogenic processes (and vice versa).

e. Pharmacodynamic activity demonstrated through in vitro research can be different or opposite from human effects.

Separate extracts of E. purpurea root or aerial plant and E. angustifolia root greatly increased THP-1 cell cytokine gene expression in vitro, whereas human consumption of the combined extracts over 3 days decreased the cytokine gene expression of these cells (except for significantly increased IFN-? by day 12).( n19) Further, E. purpurea 55% ethanolic extract was found to be moderately active against influenza virus and equivocal with 70% ethanolic extract in vitro,( n20) but root tincture of E. purpurea with 55% ethanol was effective in flu-like infections in a human clinical trial.( n21) No antiviral effects were seen with the E. pallida root 55% or 70% alcoholic extracts in the in vitro research,( n20) yet efficacy was shown with E. pallida root tincture in viral respiratory tract infections in a clinical study.( n22) E. angustifolia 70% ethanolic root extract was moderately antiviral and equivocal with 55% ethanolic extract for rhinovirus type 14 in vitro,( n20) but E. angustifolia root tincture with 60% ethanol was ineffective for induced rhinovirus type 39 infection in a clinical study.( n23)(*)

[(*)Unfortunately, use of these 3 separate clinical trials for demonstrating that the outcomes are inconsistent with the in vitro results is marred by the differences between processing the root extracts and in the clinical research protocols.]
Table 4. Different Botanical Preparations Present Particular Concerns in Human Research

a. Powdered herb capsules/tablets

• Individual digestive capability varies, especially being reduced in elderly and those with gastrointestinal (GI) or febrile diseases.

• Unless an herb powder is appropriately assayed for bioactive markers, its content remains uncertain. Standard assays for extracts may not be adequate for greater herb complexity.

b. Liquid extracts

• Relative strength depends on potency of herbs and methods used for extractions.

• Rapid absorption may increase peak concentrations but reduce duration of activity.

• Compliance may be difficult when taste and/or flavor is unusual or strong.

• Herbal flavors are often difficult to mask or mimic with inert liquid placebos.

c. Solid standardized extracts

• Dissolution of material in GI environment/medium is usually untested.

• The rate and percentage of absorption of standardized content can vary between products depending on type and amount of coating, fillers, binders, adjuvants, etc.

d. Concentrated fractions

• Reduction in the phytochemical content compared to herb or simple extracts can alter parameters of absorption, metabolism, excretion, and ultimately the bioactivity found in the original extract or herb.
Table 5. Steps Toward a Better Assessment of Outcomes in Botanical Research
Good Research Practices

a. Check volume or pill count left.

b. Specify reasons for withdrawal.

c. List number/type of adverse effects.

d. Compare adverse effects with drug.

e. Compare/contrast responders and non-responders.

f. Document concurrent medication.

g. Emphasize value of co-medication.

h. Assess cost effectiveness.
Common Practices to Avoid

Assume dosage compliance.

Report only the number of withdrawals.

Fail to specify adverse effects or whether they occur.

Compare only to benefits of drug.

Assume homogeneity of responders and non-responders.

Ignore impact associated with drug use.

Presume neutral effect from drug combination(s).

Disregard comparative cost to patient.

PHOTO (COLOR): This illustration originally appeared in chapter 6, Herbal Medicine (by Mark Blumenthal), in the Reader's Digest Family Guide to Natural Medicine (1993) Used with permission of The Reader's Digest Association, Inc., Pleasantville, NY, www.rd.com. Illustration by Jean Francois Allaux.

PHOTO (COLOR): Echinacea Echinacea purpurea Photo © 2009 Steven Foster

PHOTO (COLOR): Echinacea Echinacea purpurea Photo © 2009 Steven Foster

PHOTO (COLOR)
References

(n1.) Jensen CB. Clinical trials of herbal and pharmaceutical products. Alt Compl Ther. February 1998;30-35.

(n2.) Bondurant S, Sox HC. Mainstream and alternative medicine: converging paths require common standards. Ann Intern Med. 2005; 142(2): 149-150.

(n3.) Tyler VE. Product definition deficiencies in clinical studies of herbal medicines. Sci Rev Alt Med. 2000;4(2): 17-21.

(n4.) Leung A. Scientific studies and reports in the herbal literature: what are we studying and reporting? HerbalGram. 2000;48:63-64.

(n5.) Cooper R. Contribution of pharmacognosy to clinical trials of botanicals and dietary supplements. J Air Comp Med. 2007; 13 (9): 1045 - 1046.

(n6.) Bucco G. Industry grapples with ethics of "borrowed" science. NFM's Nutr Sci News. January 1997; 18-20.

(n7.) Betz JM, Fisher KD, Saldanha LG, et al. The NIH analytical methods and reference materials program for dietary supplements. Anal Bioanal Chem. 2007;389:19-25.

(n8.) Kroll DJ, Shaw HS, Oberlies NH. Milk thistle nomenclature: why it matters in cancer research and pharmacokinetic studies. Integr Cancer Ther. 2007;6(2):110-119.

(n9.) Pellati F, Benvenuti S, Magro L, et al. Analysis of phenolic compounds and radical scavenging activity of Echinacea spp. J Pharm Biomed Anal. 2004;35:289-301.

(n10.) Perry NB, van Klink JW, Burgess EJ, et al. Alkamide levels in Echinacea purpurea: a rapid analytical method revealing differences among roots, rhizomes, stems, leaves and flowers. Planta Med. 1997;63:58-62.

(n11.) Matthias A, Addison RS, Penman KG, et al. Echinacea alkylamide disposition and pharmacokinetics in humans after tablet ingestion. Life Sci. 2005;77:2018-2029.

(n12.) Matthias A, Addison RS, Agnew LL, et al. Comparison of Echinacea alkylamide pharmacokinetics between liquid and tablet preparations. Phytomed. 2007;14:587-590.

(n13.) Matthias A, Gillam EMJ, Penman KG, et al. Cytochrome P450 enzyme-mediated degradation of Echinacea alkylamides in human liver microsomes. Chem-Biol Int. 2005; 155:62-70.

(n14.) Pugh ND, Balachandran P, Lata H, et al. Melanin: dietary mucosal immune modulator from Echinacea and other botanical supplements. Int Immunopharm. 2005;5:637-647.

(n15.) Rininger JA, Kickner S, Chigurupati P, et al. Immunopharmacological activity of Echinacea preparations following simulated digestion on murine macrophages and human peripheral blood mononuclear cell. J Leuko Biol. 2000;68:503-510.

(n16.) Senchina JDS, McCann DA, Asp JM, et al. Changes in immunomodulatory properties of Echinacea spp. root infusions and tinctures stored at 4 °C for four days. Clin Chim Acta. 2005;355:67-82.

(n17.) Willigmann I, Egert D, Bodinet C, et al. Chemical and immunological properties of the immunomodulatory active compounds from the roots of different Echinacea species. Planta Med. 1993;59(Suppl):A671-A672.

(n18.) Sharma M, Arnason JT, Butt A, et al. Echinacea extracts modulate the pattern of chemokine and cytokine secretion in rhinovirus-infected and uninfected epithelial cells. Phytother Res. 2006;20:147-152.

(n19.) Randolph RK, Gellenbeck K, Stonebrook K, et al. Regulation of human immune gene expression as influenced by a commercial blended echinacea product: preliminary studies. Exp Biol Med. 2003;228:1051-1056

(n20.) Hudson J, Vimalanathan S, Kang L, et al. Characterization of antiviral activities in Echinacea root preparations. Pharm Biol. 2005;43(9):790-796.

(n21.) Braunig B, Dorn M, Limburg E, et al. Echinacea purpurea radix for strengthening the immune response in flu-like infections. [German] Z Phytother. 1992;13:7-13.

(n22.) Dorn M, Knick E, Lewith G. Placebo-controlled, double-blind study of Echinacea pallida radix in upper respiratory tract infections. Comp Ther Med, 1997;5:40-42.

(n23.) Turner RB, Bauer R, Woelkart K, et al. An evaluation of Echinacea angustifolia in experimental rhinovirus infections. N Engl J Med. 2005;353:341-348.

(n24.) Butterweck V, Derendorf H, Gaus W, et al. Pharmacokinetic herb-drug interactions: are preventive screening necessary and appropriate? Planta Med. 2004;70:784-791.

(n25.) Venkataramanan R, Komoroski B, Strom S. In vitro and in vivo assessment of herb drug interactions. Life Sci. 2006;78:2105-2115.

(n26.) Wolsko PM, Solondz DK, Phillips RS, et al. Lack of herbal supplement characterization in published randomized controlled trials. Am J Med. 2005;118:1087-1093.

(n27.) Gilroy CM, Steiner JF, Byers T, et al. Echinacea and truth in labeling. Arch Int Med. 2003;163:699-704.

(n28.) Gorski JC, Huang S-M, Pinto A, et al. The effect of echinacea (Echinacea purpurea root) on cytochrome P450 activity in vivo. Clin Pharm Ther. 2004;75:89-100.

(n29.) Gurley BJ, Gardner SF, Hubbard MA, et al. In vivo assesment of botanical supplementation on human cytochrome P450 phenotypes: Citrus aurantium, Echinacea purpurea, milk thistle, and saw palmetto. Clin Pharm Ther. 2004.76:428-440.

(n30.) Modarai M, Gertsch J, Suter A, et al. Cytochrome P450 inhibitory action of Echinacea preparations differs widely and co-varies with alkylamide content. J Pharm Pharmacol. 2007;59(4):567-573.

(n31.) Godtel-Armbrust U, Metzger A, Kroll U, et al. Variability in PXR-mediated induction of CYP3A4 by commercial preparation and dry extracts of St. John's wort. Naun-Schmied Arch Pharmacol. 2007;375:377-382.

(n32.) Mueller SC, Uehleke B, Woehling H, et al. Effect of St John's wort dose and preparations on the pharmacokinetics of digoxin. Clin Pharm Ther. 2004;75 (6):546-557.

(n33.) Brinker F. Echinacea for upper respiratory viral infections: preparatory differences matter. Alt Med Alert. 2008; 11 (4):41-45.

(n34.) Kim H-O, Durance TD, Scaman CH, et al. Retention of caffeic acid derivatives in dried Echinacea purpurea. J Agric Food Chem. 2000;48:4182-4186.

(n35.) Kim H-O, Durance TD, Scaman CH, et al. Retention of alkamides in dried Echinacea purpurea. J Agric Food Chem. 2000;48:4187-4192.

(n36.) Gagnier JJ, Boon H, Rochon P, et al. Recommendations for reporting randomized controlled trials of herbal interventions: explanation and elaboration. J Clin Epidemiol. 2006;59:1134-1149.

(n37.) White A, Ernst E. The case for uncontrolled clinical trials: a starting point for the evidence base for CAM. Compl Ther Med. 2001;9:111-115.

(n38.) Gatchel RJ, Maddrey AM. Clinical outcome research in complementary and alternative medicine: an overview of experimental design and analysis. Alt Ther. 1998;4(5):36-42.

(n39.) Fonnebo V, Grimsgaard S, Walach H, et al. Researching complementary and alternative treatments - the gatekeepers are not at home. BMC Med Res Meth. 2007;7(7): 1-6.

(n40.) Bone K. Modern phytotherapy: integrating scientific data and methodology with traditional herbal practice. J Am Herb Guild. 2002;3(1):34-39.

(n41.) Birt DF, Widrlechner MP, LaLone CA, et al. Echinacea in infection. Am J Clin Nutr. 2008;(suppl):488S-492S.

~~~~~~~~

By Francis Brinker, ND

Francis Brinker, ND, is a clinical assistant professor with the Arizona Center for Integrative Medicine at the University of Arizona. Dr. Brinker has written numerous articles along with several books on botanical medicine, including Herb Contraindications and Drug Interactions 3rd ed. (2001) and Complex Herbs - Complete Medicines. HG (2004)

Share this with your friends