Laboratory-derived formulations of Echinacea angustifolia roots lack preventive and treatment efficacy against experimental rhinovirus infections of healthy volunteers.

Summary

Aim

To evaluate three chemically defined extracts from 2‐year‐old roots of Echinacea angustifolia for prophylaxis and treatment of experimental rhinovirus infections for efficacy relative to placebo.

Design

Double‐blind, randomised, seven‐arm, placebo‐controlled study.

Setting

A single US‐based academic medical centre. Subjects were recruited from the community surrounding the University of Virginia at Charlottesville, School of Medicine, USA. This centre also conducted the experimental rhinovirus administration and the distribution of blinded and coded study agents. Production of the three experimental formulations and extensive phytochemical analyses of each was conducted in Dr Bauer's laboratory at the Karl‐Franzens‐Universität in Graz, Austria.

Participants

A total of 437 volunteers were randomised to receive prophylaxis, starting 7 days prior to viral challenge, or treatment, begun on the day of the challenge with one of the three E. angustifolia preparations, or placebo. Healthy young volunteers were recruited from amongst the University of Virginia community and were qualified by susceptibility to rhinovirus type 39, as determined by serum‐neutralising antibody titre of 1:4 or less. Randomisation was to one of seven treatment regimens as described in ‘Intervention.’ A total of 18 subjects who had been randomised were excluded from viral challenge due to voluntary withdrawal or development of illness. Of the remaining 418 subjects challenged with virus, two withdrew prior to conclusion of the study and in 18 there was a viral antibody titre in serum or nasal‐lavage samples at day 0 greater than 1:4. The mean age of the remaining 399 subjects was 20.8 ± 3.3 years, with females representing 60% of the population. Caucasians represented the greatest proportion of subjects (79%), with 11% African–American, 9% Asian, 0.3% Native American and 0.8% of mixed race. Randomisation was balanced with respect to age and race, with the exception of one treatment group (60% ethanol extract) where women were over‐represented (75%) relative to placebo.

Intervention

Three separate extracts were prepared from the roots of 2‐year‐old E. angustifolia plants: (i) 60% ethanol, (ii) 20% ethanol and (iii) supercritical carbon dioxide. Each extract was analysed for content of individual and total alkamides, two caffeic acid derivatives, neutral monosaccharides and total polysaccharides. The phytochemical analyses were described in an extensive supplemental appendix to the primary manuscript. Briefly, the 60% ethanol extract contained 48.9% polysaccharides and 2.3% alkamides, the 20% ethanol extract contained 42% polysaccharides and 0.1% alkamides, and the supercritical carbon dioxide extract contained no polysaccharides and 72.8% alkamides. For active treatment groups, subjects used 1.5 ml of each respective tincture, all corresponding to 300 mg of E. angustifolia root, three times daily. Placebo consisted of a ‘mixture of alcoholic beverages,’ tap water and 250 ppm of denatonium benzoate preservative.

Seven possible treatment groups were designed based on a prophylaxis phase beginning 7 days prior to viral challenge (day −7 to day 0) and a treatment phase beginning on the day of viral challenge and continuing for 5 days (day 0 to day 5). While each of the treatment groups received one of the three extracts or placebo from day 0 to day 5, each of the ‘prophylaxis’ groups received one of the extracts for both phases (day −7 to day 5). A seventh group received placebo for both phases.

The viral challenge was accomplished on day 0 using nasal drops of an amount of rhinovirus type 39 corresponding to 100 50% tissue‐culture infectious doses (TCID50). Subjects were then confined to individual hotel rooms for the remainder of the study. No information was provided as to whether symptomatic treatments (i.e. H1‐histamine antagonists or α1‐adrenergic agonists) were withheld. There was also no discussion of exclusion criteria for subjects with a history of allergy or asthma.

The study was supported by a grant from the National Center for Complementary and Alternative Medicine of the US NIH. The corresponding author reported having served as a consultant for a number of pharmaceutical and consumer product companies.

Main outcome measures

The primary endpoint for the prophylaxis phase was the efficiency of viral infection in treated volunteers relative to placebo. The primary endpoint for the treatment phase was the total symptom score from infected subjects in each treatment group relative to those in the placebo group. While measurement of viral infection was an objective endpoint, symptom scoring was somewhat more subjective. Symptom scores, based on sneezing, rhinorrhea, nasal obstruction, sore throat and others, were rated by members of the study staff on a severity scale of 0 to 4 twice daily over the 5 days following infection (day 1 to day 5). The higher of the two daily symptom scores was used and the sum of the five daily scores expressed as the mean total symptom score (maximum score of 20). Nasal‐lavage specimens were collected daily from day 0 to day 5 and cultured. Day 0 collections were assessed for unsuspected viral infection in three cell lines as a means for exclusion, while post‐viral challenge lavage specimens were assessed for neutralizing antibody to rhinovirus type 39 by a standard microtitre assay.

Secondary analyses included evaluation of the study groups for incidence of clinical colds, viral titre in nasal secretions, impact on individual symptoms and effects on interleukin‐8 and polymorphonuclear leucocytes (PMNs) as indicators of intranasal inflammation. All P values were two‐sided and unadjusted for multiple tests.

Main results

None of the E. angustifolia prophylaxis or treatment regimens influenced mean infection frequency or mean total symptom score relative to those values in the placebo group. Infection frequency overall was very efficient, ranging from 81 to 92% across groups and producing mean total symptom scores of 12 to 19.2 (out of a maximum of 20). With regard to secondary measures, none of the E. angustifolia regimens altered the rate of clinical colds in infected individuals, nasal‐secretion weights or nasal‐lavage concentrations of interleukin‐8 or PMNs. Compliance was excellent, with more than 90% of patients taking at least 80% of their study agent. Blinding was reasonably effective with 41–50% of subjects in treatment groups believing they were given active medication, as compared with 36% in the placebo group.

During the prophylaxis phase, adverse events judged to be possibly related to the study medication were reported by four of 163 subjects (2.5%) in the E. angustifolia groups as compared with five of 274 (1.8%) in the placebo group. During the treatment phase adverse events were reported by 15 of 315 subjects (4.8%) in the E. angustifolia groups relative to four of 104 (3.8%) in the placebo group. Gastrointestinal side‐effects were the most common complaint.

Authors’ conclusion

‘The results of this study indicate that extracts of E. angustifolia root … do not have clinically significant effects on infection with a rhinovirus or on the clinical illness that results from it.’

Address

Ronald B Turner, University of Virginia School of Medicine, PO Box 800386, Charlottesville, VA 22908, USA, E‐mail: rbt2n@virginia.edu

Commentary

The outcome of this clinical trial generated an unusually high volume of international media attention and will likely stand as one of the most hotly debated topics in phytomedicine in 2005. The New England Journal of Medicine commissioned an accompanying editorial by a well‐known alternative medicine critic who suggested that the government funding agency of the study should ‘consider halting its search for active remedies through clinical trials of treatments with low plausibility’ to instead, ‘study the mechanisms behind errant social‐medical trends such as the alternative medicine movement.’ A more balanced and realistic commentary came from a former director of the US NIH who stated, ‘aficionados should take serious stock of the new study. But academics should not be too Echinacea smug in dismissing the herb as the fallen icon of medicine on the fringe.’

Indeed, the study was conducted by a well‐regarded clinical research team highly experienced in assessment of rhinoviral infections. The experimental virus challenge methodology minimises the heterogeneity of recruiting subjects with self‐described ‘colds’ and the model has been used previously with success to evaluate antiviral drugs. Moreover, the 22 pages of supplemental appendix material provided by Bauer stands as one of the most comprehensive examples of phytochemical analyses in support of a clinical trial. Yet there are some legitimate concerns with the trial, particularly as related to the study agents. The three Echinacea angustifolia formulations used in the trial were the product of laboratory extraction and do not correspond to any of the proprietary Echinacea products available worldwide. None of the three preparations were comprehensive for optimal recovery of all four chemical classes of constituents reported to possess antiviral and/or immunostimulating activity. Commercial formulations tested previously in Europe were derived from the aerial parts of the plant instead of the roots, but the most appropriate portion of the plant for extraction is still up for debate. Despite the outstanding attention to detail in the phytochemical analyses, no attempt was made to sample study subject serum to assess mean or peak plasma concentrations of purported active principles. Finally, the American Botanical Council has pointed out that the Echinacea dose employed in the trial (300 mg root equivalent, three times daily) is less than one‐third of that recommended in the WHO monograph on Echinacea and the recently developed draft monographs on Echinacea from the Canadian Natural Health Products Directorate. In the defence of the investigators, however, the limited amount of governmental funding provided forces one to choose between testing multiple dose levels or multiple botanical formulations; it is likely that equal criticism would be levelled had the investigators produced negative results with multiple doses of just one formulation.

The furore surrounding this negative clinical trial result is no surprise given that products made from any one of the Echinacea species are among the world's most widely selling botanical medicines and ranked recently as the most frequently used herbal product in the USA. The common cold reportedly affects 75% of US households each year, at an annual burden of US$40 billion in lost productivity and associated medical costs. Moreover, Echinacea is broadly touted as a treatment that improves the body's response to the common cold, as opposed to merely treating the symptoms of what is most often an infection by a rhinovirus (30–50% of cases) or coronavirus (10–15%). A number of in vitro studies are suggestive that Echinacea extracts have activities consistent with antiviral responses, but the concentrations of active constituents required for such activities are not commonly achieved in vivo following oral administration of Echinacea preparations. For example, the same research team conducted a limited bioavailability study with the 60% ethanol extract described in the study, and the primary alkamide compounds, achieved a plasma concentration of only 10.88 ng/ml. In fact, if the US funding agency is to be faulted, as it was in the New England Journal of Medicine editorial, it could more appropriately be for not requiring adequate preclinical pharmacology and phase I pharmacokinetic studies prior to funding expensive clinical trials. Currently, a highly reputable US botanical manufacturer is funded by the same agency to produce a comprehensive Echinacea formulation, yet three other clinical trials of other Echinacea formulations are ongoing.

Taken together, the study raises more questions than it answers. Was the study population appropriate? University students at a mean age of 21 are likely to already be at the peak of immune function, but one could argue equally that the population is under a high degree of stress and may not necessarily follow optimal nutritional practices. Controversy also continues as to the most appropriate Echinacea species for such studies. The authors used the most traditional plant for the current study, E. angustifolia, but most commercial Echinacea products are made from E. purpurea or E. pallida, as the latter are more amenable to commercial cultivation. A very recent report from the National Center for Natural Products Research at the University of Mississippi suggests that melanins from Echinacea species may be potent immunomodulatory substances that are left behind by modern extraction processes, a reminder that the perceived advantages of traditional whole‐herb preparations may have a scientific basis.

This is not the first negative trial of Echinacea to be published in a high‐impact medical journal , yet even a Cochrane monograph notes that many previous trials show some efficacy of Echinacea over placebo. Hence, devotees of the herb are unlikely to be swayed by the current results. These recent negative trials all share the common problem of inadequate dosing, emphasising further the need for basic science and pharmacokinetic studies to be used more judiciously to guide dosing for efficacy trials. In the meantime, we might best heed the advice derived from another trial conducted by the corresponding author: avoid hand‐to‐hand transmission of rhinovirus by frequent hand‐washing with an appropriate cleanser during the cold and ‘flu season.

Conflict of interest

None declared.

DJ Kroll, Research Triangle Institute, Research Triangle Park, North Carolina, USA

Author's Reply

FACT did not receive a reply from the authors.