Ann Ist Super Sanità 2012 | Vol. 48, No. 1: 105-109 105 DOI: 10.4415/ANN_12_01_17

Oscillococcinum for influenza treatment otes N Luigi Alberto Marrari(a), Laurence Terzan(b) and Gilles Chaufferin(b) (a) Laboratoires , Segrate, Milan, Italy r i ef (b)Laboratoires Boiron, Sainte-Foy-lès-Lyon, France B

Summary. The use of a complementary approach, and specifically of the popular medicine , for the treatment of influenza-like syndromes remains controversial. This brief paper analyses the currently available literature on this homeopathic preparation and the Cochrane Collaboration’s 2006 systematic review, along with other recent studies, in order to clarify certain fundamental aspects of its use in the treatment of influenza. In the light of the reported findings, and applying the rigorous criteria of evidence-based medicine, we suggest that this medicine should be placed in category “BI”. Key words: Oscillococcinum, efficacy, safety, evidence based medicine, randomized controlled trials.

Riassunto (Oscillococcinum nel trattamento dell’influenza). È ad oggi ancora controverso il ruolo e l’approccio della medicina complementare nel trattamento di sindromi simil-influenzali, ed in par- ticolare del noto medicinale omeopatico Oscillococcinum. Questa breve nota analizza la letteratura attualmente disponibile per questo medicinale, comprendente studi originali e la revisione sistema- tica della Cochrane Collaboration 2006, con lo scopo di chiarire alcuni aspetti fondamentali del suo uso. Alla luce delle prove fornite e in base ai rigorosi criteri della medicina basata sulle evidenze, si suggerisce che Oscillococcinum sia considerato nella categoria “BI”. Parole chiave: Oscillococcinum, efficacia, sicurezza, medicina basata sull’evidenza, studi controllati randomizzati.

INTRODUCTION conclusion is that the level of evidence and recommen- Oscillococcinum is a unique, original and patented ho- dation grade for this homeopathic medicine need to be meopathic medicine produced by Laboratoires Boiron. better defined. It is prepared as a Korsakovian dilution (200K) of a specific extract of duck liver and heart. For the prepara- tion of Korsakovian dilutions, Laboratoires Boiron use ANALYSIS OF THE LITERATURE a patented, fully automated machine designed to ensure Below we shall analyse in detail the 3 studies con- perfect reproducibility of dilutions. This homeopathic sidered (the Casanova trials of 1988, 1984 and 1992 medicine is officially recognised in Italy by AIFA (Italian have, for the sake of simplicity, been grouped togeth- Pharmaceuticals Agency) and registered in other coun- er as a single trial) in the Cochrane Review [5] and in tries such as France, where it has been authorised since the articles [1-3], the details of which are summarised 1944, and is sold in over 80 countries around the world. in Table 1. Oscillococcinum is a “homeopathic medicine”, in that Casanova and Gerard, 1988 [1]: the randomised it conforms to the definition introduced by Directive double-blind study involved 27 general practitioners 92/73/EEC and reiterated in Directive 2001/83/EC, distributed across France, who evaluated 300 patients adopted in Italy with Legislative Decree 185/95 and of both sexes (ranging in age from 25 to 65 years) pre- Decree n. 219 of 24 April 2006. It is used for the preven- senting symptoms of influenza such as body temper- tion and treatment of influenza and that cause ature ≥ 38 °C, shivering and muscle-skeletal pain. The influenza-like syndromes. studied patients received one tube of Oscillococcinum A careful analysis of the individual studies investi- or in the morning and one at night for gating the efficacy ofOscillococcinum [1- 3], along with one week of treatment. Body temperature was sig- the recommendations of a recent review [4] and an ar- nificantly decreased in the Oscillococcinum-treated ticle published online (www.farmacovigilanza.org), group (compared to the placebo group) as early as suggest that the conclusions of the recent Cochrane day 2 (evening body temperature in °C: 38.7 ± 0.52 vs Review by Vickers et al. (2006) need to be reconsidered 39.2 ± 0.96, p < 0.0001), with the greatest reduction in a different light [5]. This paper reports and discuss- observed on day 4 (evening body temperature in °C: es in detail the currently available scientific literature 37.2 ± 0.42 vs 38.1 ± 1.01, p < 0.0001). Furthermore, dealing with Oscillococcinum, clarifying certain funda- shivering and myalgia were also significantly reduced mental aspects of this pharmacological treatment. The in patients treated with Oscillococcinum. In particu-

Address for correspondence: Luigi Alberto Marrari, Laboratoires Boiron, Via Cassanese 100, 20090 Segrate, Milan, Italy. E-mail: [email protected]. 106 Luigi Alberto Marrari, Laurence Terzan and Gilles Chaufferin

Table 1 | Oscillococcinum randomized controlled trials otes Authors Study and Number of Conditions Treatment (s) Outcomes Treatment Key results N and year publication subjects (diagnosis) compliance (ref. n.) type r i ef B Casanova a, c 300 Influenza-like Oscillococcinum Temperature, Information not In the verum et al. 1988 syndrome (Anas barbariae shivering and available group: faster [1] 200K) 1 dose in myalgia temperature the morning and reduction, 1 in the evening significantly less for 3-4 days shivering and less myalgia after 4 days

Ferley a, b 478 Influenza-like Oscillococcinum Healing rate at 48 h Information not Clinical healing et al. 1989 syndrome (Anas barbariae after diagnosis available after 48 h [2] 200K) 5 doses, based on rectal and rate of one every 12 h temperature temperature and two of reduction better the following in the verum symptoms: group. headache, stiffness, lumbar pain, articular ache, shivering.

Papp a, b 372 Influenza-like Oscillococcinum Evaluation 95.7% in the Statistically et al. 1998 syndrome (Anas barbariae of symptoms verum group significant [3] 200K) 1 dose for during time and 89.1% in the reduction of 3 time/day for after treatment placebo group symptoms 3 days after 48 h in the verum group

(a): Double-blind placebo-controlled randomized ; (b) high-quality paper in medical literature; (c) paper published in non-indexed medical literature.

lar, myalgia disappeared on the 4th day in 70% of The patients in the study received one tube of patients treated with Oscillococcinum, compared to Oscillococcinum or placebo in the morning, and one 48% of patients in the control group (p < 0.0001). tube every 12 hours for 2 days. The proportion of The differences between the two groups were found cases resolved in the first 48 hours of treatment was to be statistically significant, and demonstrate the higher in the verum group compared to the placebo efficacy of Oscillococcinum in reducing the dura- group (17.1% vs 10.3%, p = 0.03). This correspond- tion of the influenza illness. The main findings of ed to a risk reduction of 1.67 (95% CI 1.1-2.7, p = this study show that Oscillococcinum is effective in 0.03). The results were better for the subgroup aged treating influenza syndrome, in that it more rapidly 12-29 years (25% vs 8.1%, p < 0.01), compared to relieves the characteristic symptoms of this illness the subgroup aged > 30 years (10.6% vs 8.4%, p = compared to placebo. The authors recommend 0.56), and for the subgroup with slight-moderate that further studies should be based on serologi- syndrome (24.6% vs 11.9%, p < 0.01), compared to cal data rather than on symptoms alone. This first the subgroup with severe syndrome (7.1% vs 8.2%, investigation was published in a non-indexed com- p = 0.8). A greater number of patients in the control plementary journal, but was followed by two other group resorted to other drugs for the treatment of publications of higher quality, which reported sim- fever and myalgia during the first 48 hours (50.2% ilar results. vs 40.7%, p = 0.04). The number of patients who Ferley et al., 1989 [2]: the randomised, double positively evaluated the treatment was higher in the blind study involved 478 patients of both sexes, treatment group (61.2% vs 49.3%, p = 0.02). Overall, over 12 years of age (237 in the Oscillococcinum this study showed that patients with influenza-like group and 241 in the placebo group) who, at the syndrome who were treated with Oscillococcinum time of enrolment in the study, presented rectal showed an improved (i.e. earlier) recovery rate than temperature above 38 °C and at least 2 symptoms patients receiving placebo. It is interesting that this such as headache, stiffness, lumbar pain, joint pain homeopathic medicine was more effective in pa- and shivering. Seventy-one% of patients were en- tients aged < 30 years, suggesting that the action of rolled during the height of the influenza season Oscillococcinum was more active against the influ- (as defined by the French Ministry of Health). enza . Oscillococcinum and influenza 107

Papp et al. 1998. [3]: the randomised, double ative risk of 0.93, that is to say that at 48 hours blind study involved 372 patients (188 treated with the difference in favour of Oscillococcinum is otes Oscillococcinum and 187 with placebo) of both sex- 0.93 compared to 1.0. N es, ranging in age from 12 to 60, who at the time The strongest result, according to the Cochrane

of enrolment presented rectal temperature ≥ 38 °C, authors, is the patients’ subjective assessment of r i ef muscle pains, headache, or at least one of the fol- the treatment. That is to say, whether they consid- B lowing symptoms: shivering, chest or periarticular ered the treatment to be effective, and whether this pain, spine pain, coughing, irritation of nasal mu- judgement differed between the verum and placebo cosa, feeling of malaise. Patients received 3 tubes groups. All three studies (for Casanova the 1984 tri- of Oscillococcinum or placebo each day (morn- al is cited, but the substance does not change, as the ing, noon and night) for 3 days. In particular, data Cochrane Review itself reports) were in this respect were collected concerning the patients’ condition in favour of the homeopathic medicine: the relative 48 hours after the onset of the influenza syndrome, risk was 0.60 (0.37-0.98), meaning that the propor- the speed of recovery from the symptoms present tion of patients treated with Oscillococcinum who at the time of enrolment, and the use of concomi- considered the treatment to be useless was 0.6, rela- tant therapies. The intensity of symptoms was con- tive to 1.0 for the placebo (difference of 40%). sidered on average and moderate, and efficacy was In addition to the Cochrane Review, other recent defined as a statistically significant abatement of studies have also dealt with the historical origins, symptoms. The results of the trial show a highly scientific basis and strength of evidence of comple- statistically significant difference between the two mentary therapies for influenza and viral infections groups, for what concerns disappearance of symp- of the upper airways [4-11]. The latter review [11] toms after 48 hours (19.2% in the Oscillococcinum states that “There are some positive findings suggest- group vs 17.1% in the placebo group) and improve- ing that Oscillococcinum may reduce the duration of ment in symptoms (43.7% vs 38.6% for placebo) influenza, but the effect size tends to be small.” What (Krauth test, p = 0.0028). Moreover, the frequency “small” actually means remains a matter of inter- of use of concomitant was slightly higher pretation. for the placebo group, as was also the use of mul- It is also important to note that the indexed ho- tiple medicines. Only 13.8% of the Oscillococcinum meopathic literature demonstrates an absence of group used two or three drugs (analgesics and anti- adverse events, given that the first and foremost re- rheumatics), against 19.6% in the placebo group. quirement for any medicine evaluated in clinical and Another parameter considered was the percentage epidemiological trials is safety [6, 10, 12, 13]. of patients able to return to work, which was higher in the Oscillococcinum group, both 2 days after the onset of the illness (16.3% against 9.3%) and after DISCUSSION 4 days, with highly significant differences. After 7 When assessing medical interventions, there is no days, these differences decreased until they became general consensus as to the quality criteria for clas- no longer statistically significant. This is unsurpris- sifying clinical data in terms of treatment outcomes, ing, considering that the illness in any case has a scientific strength and reliability, and this -is par duration of 5-10 days, even without treatment. This ticularly true for homeopathic medicines. In prac- clinical trial provides evidence that the treatment tice, there exists a hierarchy of methods, associated of influenza-like syndromes with Oscillococcinum with progressively better and hence more rigorous is beneficial, in terms of symptom relief and illness evidence-based medicine for aiding clinical deci- duration. sions. However, in this analysis we have relied exclu- Reviews and meta analysis: the Cochrane Review sively on the most rigorous “conventional” criteria on Oscillococcinum [5] concludes that: for evaluation. It is always important to note the 1) Oscillococcinum reduced the time needed for manner in which an existing publication has been recovery by 0.26 days (95% CI 0.47-0.05), rela- evaluated [12], and the relevance or importance of tive to an average duration of illness of 4.9 days. the journal in which it appears. Another quality cri- This effect could range from half a day to ap- terion concerns the methods used and described in proximately one hour; the paper. This is an indicator used for “weighing” 2) the number of days needed to return to work the reliability of results, and especially for “gener- was significantly reduced: 0.49 days less (95% ating” the conclusions. Finally, perhaps the most CI 0.89-0.08) compared to the control (average important criterion is the strength of evidence from of 4.1 days); randomized controlled trials (RCTs), including the 3) Oscillococcinum increased the likelihood of re- statistical power and difference size. covery within 48 hours of starting treatment: For what concerns the studies examined by the from the two studies which report this data, it Cochrane Review [5], the conclusions of the analy- can be seen that non-recovered patients were sis, as also reported in the abstract, are that: “Though slightly fewer in the verum group (339/416) than promising, the data were not strong enough to make a in the placebo group (365/418). The difference is general recommendation to use Oscillococcinum for statistically significant and corresponds to a rel- first-line treatment of influenza and influenza-like -syn 108 Luigi Alberto Marrari, Laurence Terzan and Gilles Chaufferin

dromes. Further research is warranted but the required ment of influenza, because of their adverse effects. sample sizes are large”. The recommendation in fa- The recommended treatment normally consists only otes vour or against the use of a treatment (i.e. if the data of bed rest, supplemented if needed with antipy- N are “strong enough”) depends essentially on evaluat- retics and analgesics. Yet, use of this homeopathic

r i ef ing its ratio of benefits to risks and the relevance of medicine produces a small but significant reduction B the considered outcomes. Since Oscillococcinum has in the length of the illness, which makes sense for the been on the market for over 80 years without any re- patient. A further consideration is that, since influ- ported AE, we can assert that this homeopathic med- enza syndromes affect millions of people, any reduc- icine has an excellent safety record with no ADRs of tion in the duration of illness and the time taken to class A, B, C, D, E, or F having ever been reported. resume work will cumulatively have a socially sig- Another source of confusion in the Cochrane nificant effect in absolute terms. Review [5], as also in a more recent review dealing The latest epidemiological views concur that evalu- with this subject [11], is that the final evaluation and ations of clinical trials should give greater weight to recommendations are based on evidence from differ- subjective aspects, meaning how patients themselves ent homeopathic medicine, prepared with different assess their state of health. From the patient’s per- source materials (e.g. virus or herbs, which spective of illness and health, being ill does not so are not contained in Oscillococcinum) or by differ- much consist in having an elevated body temperature, ent procedures (e.g. Centesimal Hahnemannian pro- as in how the person “experiences” the changes pro- cedure vs the Korsakovian procedure which is only voked by external aggression, in this case by a virus. used for Oscillococcinum). In other words, should normalising body tempera- The present work instead only considers and weighs ture be considered more important than the patient the evidence of randomised trials conducted using reporting that he or she “feels better”? The symptoms the original preparation. A system for grading the have to be considered as an expression of the normal level of available scientific data for or against use of activation of healing mechanisms, so that it might a given treatment for a specific medical condition is even be unreasonable to attempt to suppress them. that established by Natural Standard, an internation- The evaluation of any medical intervention should al expert panel founded to provide high-quality, evi- take into account not just therapeutic efficacy, but dence-based information about complementary and also other factors such as adverse events, costs and alternative therapies (www.naturalstandard.com). compliance: in short, an assessment of the ratio of Under this system, evidence is classified into 5 grades benefits to risks/costs. In this connection, it is neces- (Strong, Good, Unclear or conflicting, Fair negative, sary to repeat that Oscillococcinum has not shown, in and Strong negative). A treatment is considered to all the studies conducted to date, any adverse events. have a “good scientific foundation” if there is statisti- cally significant evidence of benefit from 1-2 properly randomised trials, backed up by supporting evidence CONCLUSIONS in basic science, animal studies, or theory. The first Based on the considerations outlined above and criterion (statistically significant evidence of benefit) the number of clinical studies currently available, is unquestionably satisfied by Oscillococcinum. With we believe Oscillococcinum should be assigned to respect to the second (theoretical foundation), recent recommendation class I (“generally proven”), since findings suggest that even high there is evidence derived from multiple controlled may contain some residual trace of the starting mate- randomised clinical trials and/or from systematic re- rial [14], or incorporate structural dishomogeneities views of randomised trials. In fact, this classification in the solvent which may retain some biological and requires studies with statistically significant evidence pharmacological activity [15, 16]. of benefit from 1-2 randomised trials, or evidence of Based on the evidence reported in the literature con- benefit from > 1 meta-analysis properly conducted cerning the efficacy of Oscillococcinum in the treat- in accordance with the Scottish Intercollegiate ment of influenza and influenza-like syndromes, we Guidelines Network and Natural Standards [17]. can assert that this medicine has shown evidence of Furthermore, given that Oscillococcinum complies efficacy in statistical terms, especially given the rigor- with the “strength of recommendations” statement ousness of the studies carried out. According to the that “There are some doubts as to whether the par- rigorous criteria of evidence based medicine (EBM), ticular procedure or intervention should always be the homeopathic medicine Oscillococcinum has dem- recommended, but it is believed that its use should be onstrated, in RCT studies, a statistically significant carefully considered”, this medicine should be raised difference in clinical efficacy compared to placebo, to category “B”. so that its classification as “weak” or “insufficient” or For each clinical situation and for each patient, the “not strong enough” is highly questionable, and con- risks must be weighed against the benefits, bearing in trasts with the precepts of EBM, which require setting mind the qualitative and quantitative effects of using aside subjective bias to consider only objective evalua- a particular medicine, and the likely progression and tions drawn from the available clinical evidence. outcome of the pathology. Any pharmacological in- Currently, antiviral drugs (such as neuraminidase tervention is justified only if the potential benefits are inhibitors) are not indicated in the first line treat- greater than the risks. The decision must be based on Oscillococcinum and influenza 109

an adequate understanding of the patient, the disease Conflict of interest statement and its natural history, as well as on a knowledge of The authors work for the Laboratoires Boiron. otes the treatment and its potential adverse effects, and in N this respect Oscillococcinum satisfies the requirements Received on 26 May 2011.

of effectiveness and safety. Accepted on 18 December 2011. r i ef B

References 1. Casanova P, Gerard R. Bilan de 3 années d’études randomisées 10. Trifirò G, Calogero G, Ippolito FM, Cosentino M, Giuliani multicentriques Oscillococcinum/placebo. Proposta Omeopatica R, Conforti A et al. Adverse drug events in emergency depart- 1988;6:14-7. ment population: a prospective Italian. Pharmacoepidemiol 2. Ferley JP, Zmirou D, D’Adhemar D, Balducci F. A controlled Drug Saf 2005;5:333-40. evaluation of a homoeopathic preparation in the treatment of 11. Ulbricht C, Chao W, Clark A, Conquer G, Cook D, Cormier T influenza-like syndromes.Br J Clin Pharmacol 1989;27:329-35. et al. Oscillococcinum (Anas barbariae hepatis et cordis extrac- 3. Papp R, Schuback G, Beck E, Burkard G, Bengel S, Lehr S et tum 200CK HPUS). Altern Complement Ther 2011;17:41-9. al. Oscillococcinum in patients with influenza-like syndromes: Available from: http://online.liebertpub.com/doi/pdf/10.1089/ a placebo controlled double-blind evaluation. A controlled act.2011.17107 DOI: 10.1089/act.2011.17107 trial. Br Homeopath J 1998;87:69-76. 12. Endrizzi C, Rossi E, Crudeli L, Garibaldi D. Harm in home- 4. Bellavite P, Marzotto M, Chirumbolo S, Conforti A. Advances opathy: aggravations, adverse drug events or medication errors? in and immunology: a review of clinical research. Homeopathy 2005;94:233-40. DOI: 10.1016/j.homp.2005.08.019 Front Biosci 2011;S3:1363-89. 13. Jose J, Rao PG, Kamath MS, Jimmy B. Drug safety reports on 5. Vickers AJ, Smith C. Homeopathic Oscillococcinum for pre- complementary and alternative medicines (ayurvedic and ho- venting and treating influenza and influenza-like syndrome meopathic medicines) by a spontaneous reporting program in a (Review). Cochrane Database Syst Rev 2006;3:CD001957. tertiary care hospital. J Altern Complement Med 2009;15:793-7. Available from: http://online.liebertpub.com/doi/pdf/10.1089/ 6. Bellavite P. Oscillococcinum e influenza. Storia, evidenze, ipotesi. acm.2008.0128 DOI: 10.1089=acm.2008.0128 Verona: Edizioni Libreria Cortina; 2008. 14. Chikramane PS, Suresh AK, Bellare JR, Kane SG. Extreme 7. Haidvogl M, Riley DS, Heger M, Brien S, Jong M, Fischer M homeopathic dilutions retain starting materials: A nano- et al. Homeopathic and conventional treatment for acute res- particulate perspective. Homeopathy 2010;99:231-42. DOI: piratory and ear complaints: a comparative study on outcome 10.1016/j.homp.2010.05.006 in the primary care setting. BMC Complement Altern Med 2007;7:7. 15. Roy R, Tiller W, Bell IR, Hoover MR. The structure of liq- uid water. Novel insights from materials research; potential 8. Riley D, Fischer M, Singh B, Haidvogl M, Heger M. home- opathy and conventional medicine: an outcomes study compar- relevance to homeopathy. Mat Res Innovat 2005;9:98-103. ing effectiveness in a primary care setting. J Altern Complement 16. Demangeat JL. NMR relaxation evidence for solute-in- Med 2001;7:149-59. duced nanosized superstructures in ultramolecular aqueous 9. Witt CM, Ludtke R, Willich SN. Homeopathic treatment of dilutions of silica-lactose. J Mol Liquids 2010;155:71-9. patients with chronic sinusitis: A prospective observational 17. Scottish Intercollegiate Guidelines Network. SIGN 50. A study with 8 years follow-up. BMC Ear Nose Throat Disord guideline developer’s handbook. Edinburgh: SIGN: January 2009;9:7. 2008.