Autologous Blood Therapy for Common Cold—–A Randomized, Double-Blind, Placebo-Controlled Trial

Complementary Therapies in Medicine (2009) 17, 257—261

available at www.sciencedirect.com

journal homepage: www.elsevierhealth.com/journals/ctim

Autologous blood therapy for common cold—–A randomized, double-blind, placebo-controlled trial

S. Hensler a,∗, P.W. Guendling a,b, M. Schmidt a, K. Jork a a Institute of General Practice, University of Frankfurt, Theodor-Stern-Kai 7, D-60596 Frankfurt, Germany b Master Course Complementary Medicine, Fresenius University of Applied Sciences, Idstein, Germany Available online 29 July 2009

KEYWORDS Summary Autologous blood Background: In Germany autologous blood therapy (ABT) is a widespread therapy for infectious therapy; diseases in complementary medicine. Clinical data for its use for common cold is lacking. Common cold; Methods: In a double-blind randomized controlled trial 139 patients with common cold were Primary care enrolled either to ABT (gluteal intramuscular reinjection of venous blood three times a week) or to placebo (sterile sodium chloride solution). Main criterion was time period of illness after initiation of treatment, measured by a modiﬁed symptom diary adapted from Jackson. Results: 58 and 56 patients completed therapy. In both groups illness duration was 7 days (5.0—10.0 for verum and 5.25—9.0 for placebo). Conclusions: This trial found no effect of ABT as treatment for common cold. Because of a rather highly selected patient sample another RCT on this topic is reasonable. Further research to analyse the effect of other doses or of autologous blood therapy in addition to homeopathic preparations or vitamin preparations is needed. © 2009 Elsevier Ltd. All rights reserved.

Background revealed only one randomized controlled trial (RCT) testing autologous blood therapy to prevent common cold in case In Europe particularly in Germany autologous blood therapy of elevated susceptibility to infection. This study found no is a widespread and well established therapy for infectious signiﬁcant effect, possibly due to an undersized sample.2 and allergic diseases in complementary medicine. It is used The role of autologous blood therapy for other indica- as treatment of allergic diseases (e.g. pollinosis), inﬂamma- tions has been subject to limited clinical trials. Schirmer tory and degenerative arthropathia or skin diseases1 and in et al.3 tested autologous blood therapy in combination with order to cure and prevent acute common cold especially in Formica rufa D6 for ankylosing spondilytis in a RCT and found case of elevated susceptibility to infection. However, clini- no statistical difference between treatment and placebo cal data for this use is rare. A detailed literature research in group. In another RCT treating atopic dermatitis Pittler et Medline and special databases for complementary medicine al.4 were able to show a positive effect of autologous blood (Cambase, RCCM) resulted in no clinical trial for autolo- therapy vs. placebo. A further RCT conducted by McGrath et gous blood therapy medicating acute common cold. It also al.5 found an effect of ultraviolet radiated autologous blood on walking distance of patients with intermittent claudication. A clinical study by Olwin et al.6 found an effect in the ∗ Corresponding author. Tel.: +49 69 6667031; fax: +49 69 6667031. treatment of herpes zoster. Pfeiffer et al.7 found an effect E-mail address: [email protected] (S. Hensler). in the treatment of recurrent spontaneous abortion.

Despite the lack of clinical data about the effectivity of practice with symptoms of common cold (cough, rhinitis, autologous blood therapy for common cold, there are only sore throat, fever etc.) which lasted not longer than 7 days few alternative therapy options in the treatment of com- and resulted in at least four points on a modified Jackson- mon cold with some evidence for effectivity. According to a Score (s.a.). Exclusion criteria were illness duration before review by Smith et al.8 there is no evidence for effectiveness treatment longer than 7 days, severe forms of respiratory of unspecific symptomatic therapy options such as antitus- infections (exacerbation of chronic obstructive pulmonary sives, mucolytics, expectorants and histamine antagonists. disease by infection, exacerbation of asthma by infection, Available results on effectiveness of zinc,9,10 echinacea11—13 streptococcal tonsillitis, peritonsillar infection, pneumonia, and inhalations with steam14 are too inconsistent to relay on. mastoiditis, etc.), other diseases with similar symptoms Only for Vitamin C does there seem to be some evidence for (e.g. allergic rhinitis), immunosuppressing diseases (e.g. a declining duration of illness. Simasek and Blendino15 and AIDS) or medication (e.g. Cyclosporin), dementia, and coag- the Cochrane-Review of Douglas et al.16 found that Vitamin ulopathy. C prophylaxis may modestly reduce the duration and severity of the common cold in the general population and may Randomisation reduce the incidence of the illness in persons exposed to physical and environmental stresses. After enrolling in the study, patients were allocated to Specific therapy with antibiotics is probably due to mostly intervention (autologous blood therapy) and control group viral etiology only marginally effective (common cold,17 (placebo) by a specially instructed practice nurse, who acute otitis,18 acute bronchitis,19 acute laryngitis,20 and contacted the study centre (Institute of General Practice, acute sinusitis21). Side effects of antibiotics outweigh the university of Frankfurt/Main) in order to ask the sequence benefit, especially for primary care a prudent use of antibi- of randomisation. In order to avoid an unbalanced allocation otics is strictly recommended.22 Exception is the acute of samples inside a practice, for each practice a blocked pharyngitis caused by Streptococcus pyogenes.23 From a randomisation list was prepared. Although effects on the public health point of view the widespread use of antibiotics effectiveness of autologous blood therapy are not known, gives rise to a growing number of resistant bacteria strains. an influence of illness duration before initiation of treat- Because on the one hand there is no really effective ther- ment is theoretically possible. Therefore randomisation was apy for the common cold and on the other hand there are stratified for 72 h and 4—7 days as illness duration before no sufficient data on the effectiveness of autologous blood initiation of treatment. therapy for the common cold, a need for a prospective study on this matter is evident. An effective therapy option could Test treatment be an alternative to the use of antibiotics in this indication. Considering the burden of illness (high rate of incidence and After randomisation patients were asked for three appoint- sick leave) an effective reduction of illness duration is rele- ments within one week to conduct the test treatment, vant for patient and employer. In order to provide basic data preferably on days 1, 2, and 4. Due to weekends and banking we aimed to test autologous blood therapy vs. placebo. holidays variations in sequence were allowed. Autologous blood therapy was realised by a venous blood sample (2 ml) Methods from a cubital vein and gluteal intramuscular reinjection. There were no homeopathic additives or vitamin preparations. The control group received 2 ml of sterile sodium Design and setting chloride solution instead of the blood sample. At each appointment the doctor was expected to monitor participat- A randomized double-blinded placebo-controlled study was ing patients for complications arising from original infection conducted in a primary care setting. Twenty practices of or test treatment. general practitioners (11 of them specialized in complementary medicine) in two cities (Frankfurt and Wetzlar) and Blinding their rural surrounding areas were involved in the study from 20 November 2003 until 28 February 2007 (four winter The doctors recruited and monitored patients but were seasons). blind to the therapy, because conduction of the blood ther- Medical practices were recruited out of a practice apy and separate documentation was the task of a picked network belonging to the university department of Frank- nurse in each practice. The nurses were specially instructed furt/Main (Institute of General Practice) and of a physician on site by two of the authors (SH + MS). In order to blind organisation (Hessian society of physicians for Naturopathy). treatment to the patient the placebo injection had to be Practices were not randomly selected. Some doctors were conducted slowly and with sodium chloride solution at room- practising as doctor of naturopathic medicine and experi- temperature. The syringe was prepared with tape or with enced in autologous blood therapy. The majority of doctors a paper cover in a separate practice room. Patients were were interested in naturopathy but were practising as usual required to turn around before applying intramuscular injec- general practitioners. tion of either verum or placebo.

Eligibility criteria for participants Finding of data

Patients (18—75 years of age) were asked by the practice The main criterion was time period of illness after initi- doctor to take part in the study when they consulted the ation of treatment. Illness was measured by a subjective Autologous blood therapy for common cold 259 rating scale adapted from Jackson.24 In this symptom diary, should be continued in case of injection-related complica- usually used for studies for common cold, patients moni- tion with harmless course (e.g. streptococcal tonsillitis, light tored the severity (0—3 points: absent, mild, moderate, or pneumonia) beside an antibiotic therapy. severe) of eight common cold symptoms (nasal discharge, nasal congestion, sneezing, cough, malaise, throat discom- Statistics fort, fever/chills, and headache). We modified the score adding the symptom ‘‘pain in the limbs’’. A score below Null-hypothesis was that the primary endpoint time period 3 points on three consecutive days was the cut-off crite- of illness after initiation of treatment was the same in the rion for illness. In case of a longer lasting illness an absolute intervention group (autologous blood therapy) and in the cut-off criterion for the end of monitoring was set at 35 days control group (placebo). We hypothesized that period of ill- after initiation of treatment, presuming that illness changed ness in the intervention group (autologous blood therapy) from acute into chronic state. is shorter than in the control group (placebo). Data analy- Side effects of therapy were registered in an additional ses were conducted intention-to-treat (according to group questionnaire to the patient, while complications and con- assignment by randomisation and not according to de facto comitant therapy (antibiotics, analgesics etc.) were asked in therapy) for all included patients. Data with non-normal a separate questionnaire by the doctor. Other parameters, distribution was tested with the Wilcoxon-Test, mean and especially blood parameters, were not analyzed. deviation data was figured as median and quartiles. Normal- The number of patients fulfilling inclusion criteria and distributed data was tested with the t-test (Student), mean invited to participate was asked post-hoc to the doctors and and deviation was figured as arithmetic mean and confidence could only be estimated. intervals). Sample size was calculated defining a reduction in time period of illness for 2 days as clinically relevant by a Interruption of treatment given probability of type 1 error = 5% and power = 90%. The mean illness duration for a common cold was estimated at Test treatment should be interrupted in case of severe 7 days and a standard deviation of 4 days.25 Assuming a nor- unexpected events (severe disease), admission to hospital mal distribution calculations were conducted with a t-test, or severe treatment-related adverse events (e.g. muscu- resulting in 70 cases for each group in a simple design and lar abscess as injection site complication). Test treatment 85 cases for a stratified design.

Table 1 Disposition of study objects. 260 S. Hensler et al.

Table 2 Baseline characteristics of treatment groups.

Autologous blood Placebo P value

n (analysed) 58 56 Age (SD) 43.7 (11.9) 39.6 (11.2) >0.05 Sex (F:M %) 77:23 79:21 >0.05 Smoker (%) 21.9 25 >0.05 Illness time before treatment (0—3 days:4—7 days) 29:29 31:25 >0.05 Intensity of symptoms (Jackson-Score) (SD) 13.6 (4.9) 13.1 (4.9) >0.05 SD: standard deviation.

Table 3 Main criterion and side-effects in treatment groups.

Autologous Placebo P valueb Difference (conﬁdence blood interval)a

N (completed) 58 56 Illness duration in days (median and quartiles) 7 (5.0—10.0) 7 (5.25—9.0) 0.81 0.0 (−2.0 to +1.0) Infection-related complications (light)c 2 4 0.61 0.0 (0.0—0.0) Infection-related complications (severe)d 00 Therapy-related complications (light)e 3 3 0.97 0.0 (0.0—0.0) Therapy-related complications (severe)f 00 a Difference (Hodges—Lehmann) and conﬁdence interval (P = 0.95) (moses). b Mann—Whitney. c With need for antibiotic therapy. d With need for hospital admission, pneumonia. e Initial aggravation or initial body-temperature rise. f Inﬂammation or abscess of injection site.

Ethics There was no difference in infect-related complication rates, which was small in both groups (see Table 3). The study passed the ethics commission of the university of Frankfurt/Main on the 17th November 2003. Discussion

Results Conclusions

Recruitment rates of the 20 participating practices dif- It was found in both groups an illness duration of 7 days, fered considerably. The majority of practices recruited 1—5 the study showed no advantage of autologous blood ther- patients, two practices recruited 10—20 patients and two apy vs. placebo in the treatment of acute common cold. We practices were able to recruit more than 20 patients. considered as clinically relevant a reduction of illness dura- After the winter season 2006/2007 (the 4th winter sea- tion for 2 days, because it could outweigh the burden of a son) reaching a number of 139 the decision was taken to time-consuming and interventional therapy and may reduce close patient recruitment and to forego stratified data anal- significantly the burden of illness for patient and society. ysis. Recruitment rate of patients asked to participate could Smaller reduction rates may also be considered as clinically be estimated only on a basis of doctor’s post-hoc informa- relevant but could be revealed only by larger sample sizes. tion and was very low (about 10%, see Table 1). 58 in verum In view of this highly selected and small patient sample group and 56 placebo group completed the study. The rate other confirming RCT for this indication are reasonable. In of discontinuing patients was rather high (about 17%) but did daily practice of complementary medicine autologous blood not differ between groups (12 and 13 patients respectively). therapy is rarely conducted as single therapy, homeopathic In both groups discontinuing patients were half women and preparations or vitamin preparations are usually added. So half men, average age was almost the same (38.4 vs. 39.4). we see the need of further research to analyse the effect of Baseline characteristics did not differ, only slight non- autologous blood therapy in addition to homeopathic prepa- significant differences were seen in age and smoking rate rations or vitamin preparations or in higher doses. (see Table 2). The percentage of female patients is markedly higher than male patients (78%:22%). The main criterion illness duration, analysed intention- Limitations to-treat, was the same for the two intervention groups (median 7, P = 0.81 (Wilcoxon-Test), difference: 0.0 (CI The low rate of recruited patients (about 10% of asked −2.0 to +1.0 Hodges—Lehmann)). Hypothesized advantage patients), the long time period of the study, the unbal- of autologous blood therapy vs. placebo was not found. anced recruitment rate per practice and the high rate of Autologous blood therapy for common cold 261 females indicate that the patient sample is highly selected apy in the treatment of patients suffering from peripheral which limits generalisation on normal primary care atten- arterial occlusive disease with intermittent claudication. Eur J ders (limited external validity). A qualitative interrogation Endovasc Surg 2002;23(5):381—7. of participating doctors revealed that difficult recruitment 6. Olwin JH, Ratajczak HV, House RV. Successful treatment of may be due to time-consuming treatment (three appoint- herpetic infections by autohemotherapy. J Altern Complement Med 1997;3(2):155—8. ments) without financial compensation, refusal of a possible 7. Pfeiffer KA, Sillem M, Daniel V, Kirschfink M, Süsal C, Run- placebo treatment by patients with high expectations in spe- nebaum B, et al. Activated autologous blood therapy in cialized practices and different person-related recruitment recurrent spontaneous abortion—–results of a pilot study. Hum 26 behaviour of participating doctors. Reprod 1998;13(2):491—7. The rate of discontinuing patients (especially for lost fol- 8. Smith SM, Schroeder K, Fahey T. Over-the-counter medications low up) is rather high, probably because patients with acute for acute cough in children and adults in ambulatory settings. common cold mostly need no further medical consultation. Cochrane Database Syst Rev 2008;(1):CD001831. Because the rate is equally distributed in intervention and 9. Marshall I. Zinc for the common cold. Cochrane Database Syst placebo group, the influence on findings seems to be accept- Rev 2007;(3):CD001364. able. 10. Caruso TJ, Prober CG, Gwaltney Jr JM. Treatment of naturally acquired common colds with zinc: a structured review. Clin The quality of blinding was dependent on the con- Infect Dis 2007;45(5):569—74. ducting practice nurse. Although the nurses were specially 11. Linde K, Barrett B, Wölkart K, Bauer R, Melchart D. Echi- instructed how to blind the therapy to doctor and patient, nacea for preventing and treating the common cold. Cochrane to conduct therapy knowing allocation is a source of bias. Database Syst Rev 2006;(1):CD000530. Nevertheless, such a bias usually favors the findings for the 12. Shah SA, Sander S, White CM, Rinaldi M, Coleman CI. Evalu- verum group, which we did not find. ation of echinacea for the prevention and treatment of the common cold: a meta-analysis. Lancet Infect Dis 2007;7(7): Strengths 473—80. 13. Tierra M. Echinacea: an effective alternative to antibiotics. J Herb Pharmacother 2007;7(2):79—89. The study is a double-blind RCT with a good allocation con- 14. Singh M. Heated, humidified air for the common cold. Cochrane cealment process and equally distributed baseline data. The Database Syst Rev 2006;3:CD001728. amount of bias can be considered small (good internal valid- 15. Simasek M, Blandino DA. Treatment of the common cold. Am ity). Fam Physician 2007;75(4):515—20. 16. Douglas RM, Hemilä H, Chalker E, Treacy B. Vitamin C for preventing and treating the common cold. Cochrane Database Syst Conflict of interest Rev 2007;(3):CD000980. 17. Arroll B, Kenealy T,The Cochrane Collaboration. Antibiotics for The author(s) declare that they have no conflict of interest. the Common Cold and Acute Purulent Rhinitis. John Wiley and Sons; 2007 [Review]. Acknowledgements 18. Glasziou PP, Del Mar CB, Sanders SL, Hayem M. Antibiotics for acute otitis media in children. Oxford: The Cochrane Library; 2003. Issue 1, (Review), Update Software. We wish to thank all participating patients and general prac- 19. Smucny J, Fahey T, Becker L, Glazier R. Antibiotics for titioners who participated in the study. Further we thank the acute bronchitis. Oxford: The Cochrane Library; 2003. Issue Karl und Veronica Carstens — Foundation in Essen, Germany 1, (Cochrane Review), Update Software. for supporting our project with an independent equipment 20. Schalen L, Eliasson I, Fex S, Kamme C, Schalen C. Acute grant. The foundation gave advice for the study design, laryngitis in adults: results of erythromycin treatment. Acta there were no further involvements. Otolaryngol Suppl 1992;492:55—7. 21. Young J, De Sutter A, Merenstein D, van Essen GA, Kaiser L, Varonen H, et al. Antibiotics for adults with clinically diagnosed References acute rhinosinusitis: a meta-analysis of individual patient data. Lancet 2008;371(9616):908—14. 1. Krebs H. Eigenbluttherapie, Methoden Indikationen Praxis. 22. Williams JW, Aguilar C, Makela M, Cornell J, Hollman DR, München Jena: Urban&Fischer; 1999. Chiquette E, et al. Antibiotics for acute maxillary sinusi- 2. Jobst D, Altiner A, Wegscheider K, Abholz HH. Do autologous tis. Oxford: The Cochrane Library; 2003. Issue 1, (Cochrane blood injections help against relapsing upper respiratory infec- Review), Update Software. tions? Traps on the way of doing a randomised trial. Z Allg Med 23. Turnidge J. Responsible prescribing for upper respiratory tract 2005;81:258—63. infections. Drugs 2001;61(14):2065—77. 3. Schirmer KP, Fritz M, Jäckel WH. Wirksamkeit von Formica 24. Jackson GG, Dowling HF, Spiesman IG, Boand AV. Transmission rufa und Eigenblut-Injektionen bei Patienten mit ankolysieren- of the common cold under controlled conditions. Arch Int Med der Spondylitis: eine randomisierte doppelblinde Studie. Z 1958;101:267—73. Rheumatol 2000;59:321—9. 25. Audera C, Patulny RV, Sander BH, Douglas RM. Mega-dose 4. Pittler MH, Armstrong NC, Cox A, Collier PM, Harz A, Ernst E. vitamin C in treatment of the common cold: a randomised Randomized, double-blind, placebo-controlled trial of autol- controlled trial. Med J Aust 2001;175(7):359—62. ogous blood therapy for atopic dermatitis. Br J Dermatol 26. Guendling PW, Hensler S, Schmidt M, Strehler-Mesenholl 2003;148:307—13. E. Requirements and barriers for practice-based research 5. McGrath C, Robb R, Lucas AJ, Stewart AH, Underwood CL, Hor- in naturopathy and complementary medicine: a qualita- ridge JK, et al. A randomised, double blind, placebo-controlled tive interview study. Focus Altern Complement Ther (FACT) study to determine the efficacy of immune modulation ther- 2007;12(Suppl. 1):25.