ADVERTISEMENT
Lessons Learned in Research: An Attempt to Study the Effects of Magnetic Therapy
Scant information is available in the English language scientific literature to support the extraordinary growth of the use of complementary therapies for treatment of health problems. This article discusses the difficulties encountered in attempting to investigate one of the available complementary treatments - magnetic therapy - and its effect on wound healing.
The magnets used for this study do not produce thermal or ionizing effects in tissue; therefore, they are not considered medical devices. The magnetic field generated by the magnets is fixed in terms of intensity at any measured point, in contrast to the changing fields associated with electrical power sources. Because these magnets are not regulated by the FDA or other such agency, they may be purchased without a physician's prescription and used without the supervision of a healthcare professional.
Literature Review
In 1993, one-third of Americans spent $13.7 billion on complementary medical treatments.1 By 1997, these numbers had grown - one-half of all Americans used some form of complementary healthcare at a cost of $36 billion out-of-pocket per year.2 As a result, in 1999, the federal government established the National Center for Complementary and Alternative Medicine with a $50 million budget within the National Institutes for Health (NIH). The Center's mission is to seek data on the efficacy of alternative/complementary healthcare and to establish an information clearinghouse.2
A 1992 report to the NIH on bioelectromagnetics identified wound healing as one of the new or unconventional uses for nonthermal, nonionizing electromagnetic (EM) fields.3 Using a static magnetic field (SMF) for therapeutic purposes is generally referred to as static magnetic field therapy and involves exposing the body or specific body part to a fixed magnetic field through the use of magnetized device.4 Although the effectiveness of pulsed electromagnetic stimulation in wound healing is fairly well documented,5-8 only one case report9 and one study10 using SMF therapy were found; both had positive findings. The chronic wound population involved with SMF therapy in this study was comprised of individuals with diabetic foot ulcers. The hypothesis of the study was that people with diabetes who receive SMF therapy in addition to standard treatment (sharp debridement, twice-daily dressing changes, antibiotics when needed, offloading the ulcer) would have accelerated healing of their foot ulcers compared to people with diabetes who receive the standard treatment and placebo.
Methods
The authors anticipated recruiting 70 subjects from patients receiving treatment at a wound care center (WCC). In addition to having a diabetic foot ulcer, subjects had to be 18 years of age or older, able to read and understand English, and living in an environment where they had control over application and removal of the insole of the shoe (ie, home setting, subject or family member to assume responsibility for insole).
The sample size estimate was based upon one-way analysis of variance (ANOVA) and the additional effect magnets were expected to have over the standardized healing rate protocol established by the WCC. The authors estimated that by enrolling 70 subjects (35 per group), would provide 80% power with a p = 0.05.
The subjects understood they would receive a "treated insole" that would be worn in addition to the standard treatment for their foot ulcers. After obtaining written informed consent, the subjects were randomly assigned to the experimental (magnetic insole) or control group (placebo insole). The subject, data collectors, and physicians did not know whether a specific insole was placebo or magnetized in an attempt to double-blind the study. The insoles chosen were commercially available (Magstep) and purchased from Nikken Wellness Technology for Living in Irvine, Calif. The specific model has magnets specially arranged within the insole material to provide an alternating magnetic field pattern. Each magnetic domain is triangular in shape; the domains are arranged in a square pattern having two north and two south magnetic poles in the square (see Figure 1). The manufacturer's literature specifies the peak field strength for any given domain to be approximately 400 gauss.
The pedorthotist made the placebo insoles and prepared all the insoles for each subject, making any necessary modifications and stitching the insole into a piece of stockinet. The stockinet concealed the insoles and also was used to hold the insole in place. Verbal and written instructions were given to participants on how to apply the insole; they were to apply the insole after their evening dressing change (all were on twice-daily dressing changes) and wear the insole for 12 hours each night (or for length of sleep time) for the entire 8 weeks of the study. Wear time was limited to 12 hours to avoid any offloading issues during weightbearing hours. Subjects also were instructed to keep the dressing as thin as possible so the insole would be in direct contact with the plantar surface of the foot.
Subjects were to be seen every other week at the WCC. At each visit, the WCC staff recorded length, width, and depth of the wound in millimeters according to their standard policy which included measuring the wounds before debridement, using the longest measurement as the length and the perpendicular measurement to that as the width. The deepest point was measured for depth. Wound healing was to be measured by time to complete healing or comparison of wound volumes from entry to week eight.
Results
Fifty-six subjects were enrolled over 15 months; only 37 completed the study. Three subjects withdrew because they did not like having something on their foot at night; the insole either bothered them or was too hot or too tight. One withdrew because his physician did not want him to continue. Fifteen others were dropped from the study for a variety of reasons (see Figure 2).
On completion of the data collection, 12 (32%) of the 37 subjects were 100% healed. Of the remaining 25, five (13.5%) had wounds larger at the end of 16 weeks than the first visit, two (5%) were about the same size, and the other 18 (49%) were smaller.
The data were reviewed by two different statisticians - both concluded that the data could not be analyzed in any meaningful way because of the small sample size and the large variations in the data. Table 1 shows the demographic and basic medical characteristics of the participants for each group.
Discussion
Clearly, the authors were quite disappointed to have no usable data and no answer to the hypothesis for the study. The following discussion reflects the problem areas identified in retrospect with the hope that others will be prevented from making similar mistakes.
Enrollment and retention. Based on the number of patients with diabetic foot ulcers routinely seen at WCC and the addition of a $25 incentive, recruitment was not expected to be a problem. In reality, however, recruitment and retention both were problems. After 15 months, the authors had enrolled only 56 subjects and retained only 37 of them. The primary investigator was relying on the WCC staff, who continued with their regular full-time responsibilities, to recruit subjects. Such an expectation was unrealistic. In retrospect, a larger stipend to subjects may have been helpful, but two other considerations would be more important. If the primary investigator cannot physically be present, a study coordinator whose primary focus would be the study - recruitment, obtaining consent, following-up with subjects and monitoring compliance - should be available on site. In addition, a multisite study would help ensure larger numbers over a shorter period of time.
Complementary product (magnetic insoles). Most complementary products, including magnets, are available over-the-counter and from network marketing distributors. The insoles purchased for this study were specified by the manufacturer to have a maximum field strength of 400 gauss. After the purchase of insoles and enrollment of several subjects in the study, the clinical engineer involved in the study developed a gauss meter to obtain survey measurements of the magnetic domains in some of the insoles. Measurements were taken at the surface of the insole at the center of the magnetic domain. Thirteen insoles were measured with the lowest mean strength being 242 gauss and the highest mean strength 350 gauss. Although no literature indicates what gauss level is needed for wound healing, variations this large might affect outcomes. The authors recommend that future researchers thoroughly investigate the chosen products before beginning the study.
Blinding and mechanics. The double-blind experiment was chosen because it is considered the most rigorous form of scientific investigation.11 Also, double-blind controlled experiments offer the best design for protection against observation bias.11
Placebo insoles were not available from the commercial company. The pedorthotist made the placebo insoles that matched the magnetic insole in size, contour, and texture, but not color. Because the magnetic insole was dark brown and the placebo light beige, in an attempt to keep the study blinded, both were covered with moleskin and sown into a long piece of stockinet (one size only). No one could tell the difference between the insoles by sight or touch.
Two mechanical problems that were not well addressed in the study design were wound dressing and insole application. Measured magnetic field intensity decreases as distance from the source increases. This change follows the inverse square law - ie, field strength decreases as the square of distance. Doubling the distance from the source results in field strength of one-fourth the intensity measured at the original distance.
On other words, the farther the magnet from the wound or the plantar surface of the foot, the less the penetration of the magnetic field into the tissues. The wound dressing potentially lifted the magnet from the planter surface of the foot, adding to the distance already created by the two layers of material used for blinding. Even though subjects were instructed to keep their wound dressing wraps to a minimum, the authors were not specific and did not require return demonstrations of the dressing wrap technique. The dressing and the variation in thickness that undoubtedly occurred between dressing changes likely impacted the effect of the magnet.
Another problem of proximity of magnet to plantar surface was present in the application system. Subjects were instructed to wear the insoles during sleep time because of offloading requirements, applying the magnet by means of the stockinet into which the magnet was stitched. Practically speaking, using the stockinet was an easy for the subject and avoided use of adhesives on the skin. However, the fit of the stockinet likely varied depending on the size of the leg and wear on the fabric. In retrospect, ensuring that the magnet fit snugly against the foot could not be guaranteed.
In summary, the combination of the lower-than-expected strength of the magnet, the layers of material on the magnet (moleskin and stockinet), the thickness of the dressing, and the variations of fit with the stockinet all magnified the issue of inconsistency in the magnetic field penetration of the tissues and, consequently, the therapeutic effect. These issues would need to be resolved in another study design. A search for a company that would provide both placebo and magnet products identical in appearance would eliminate one of these problems.
Measurement of wound healing. Because the wound measurement method was established policy and part of the daily expectations of the WCC staff and the WCC outcomes were based on their measurements with reporting to the national headquarters, no measures were taken to ensure standardization of wound measurement among staff members. In retrospect, some significant variations in wound measurements from visit to visit left a question regarding valid and reliable wound measurements.
Many of the wounds did not progress in an orderly fashion. In an attempt to gain more information about healing rates, the authors extended the data collection time to a total of eight visits or as close to 16 weeks as possible. For those totally healed, time can be used as the unit of measure, but this becomes more complicated with participants who did not heal completely. The authors intended to compare wound volumes (percentage of change), but two problems arose. The authors thought they could compare percent of wound volume change every 2 weeks. However, visit dates were altered with some frequency - sometimes visits were 1 week apart and other times 4 weeks or more. The second issue had to do with wound volumes changing significantly from visit to visit (see Table 2). This may have been due to different measurement techniques among data collectors, but the greater variation was likely due to debridement. No plan for dealing with major debridement was in place, which sparked several questions: Should the subject whose wound went from 55 mm3 at visit 7 to 2,260 mm3 at visit 8 be treated differently than the others? Should this be considered a "new" wound that should start over with 2,260 mm3 as visit one? If so, what constitutes a major debridement versus a minor debridement - ie, how much of a change in wound volume between visits is a "major" debridement? If not, does comparison of beginning wound volume to debrided wound volume tell anything about the treatment being studied? Is it valid data?
Better planning is needed to address these issues. A computerized system to measure wound areas or volumes would eliminate interrater variations in wound measurement. Greater numbers of subjects and a more longitudinal study would be desirable. The larger sample size should provide a larger number of totally healed wounds and the greater number of subjects would, hopefully, through randomization, equal out the number of major debridements between the two groups during the study.
Magnetic Therapy Questions
For those interested in magnetic therapy, a number of basic questions need answers. Commercially available magnetic therapy products have two types of field distribution: fixed or uniform field and alternating field. Is one better than the other? What is the ideal strength of the magnet? What size should the magnet be in relation to the wound size, and where should the magnet be placed in relation to the wound? How long should the magnet be worn? Although people with diabetic foot ulcers present great healing challenges, choosing this population allowed the authors, from a practical standpoint, to standardize the magnet size and surface area covered by the magnet.
Funding. Another barrier not easily overcome is that of funding for nursing research. Many of the suggestions made in this article require increased funding: securing an on-site study coordinator, increasing money for recruitment incentive, mechanizing or computerizing wound measurement, designing a more longitudinal study, and expanding the study to multiple sites to increase the number of participants. Even though the potential impact of magnetic therapy on the healthcare system could be great, the "big" money traditionally goes to drug research and to major centers with a known research history. Where will funding for less grandiose efforts originate? Should professional organizations be designing studies for which the members could serve as data collectors? If important questions are addressed and clinicians work together, will they be recognized as worthy of funding?
Study validity and ethics. Because the double-blind controlled study is the most valued in terms of guiding clinical practice, the authors attempted to double-blind the study. The Institutional Review Board (IRB) allowed the study to be explained to potential subjects using the words "treated" versus "untreated" insole. Interestingly, in an attempt to expand the study to another WCC to increase number of subjects, the IRB required the authors to inform the subjects that they would receive a "magnetic" versus placebo insole. The authors chose not to expand the study to this center because a major change in protocol would have been required.
The differing opinions on amount of disclosure of the two IRBs involved raised an interesting question that will need to be addressed, as it may affect the ability to blind the study. Ethically, should the subjects be informed that the study is testing magnetic therapy? Brody stated in a New York Times article, "unless patients and their evaluators are blind to who is receiving real magnet therapy and who is getting a sham treatment, it will not be possible to attribute any benefit to the magnets."12 But even if it is ethically acceptable not to disclose the magnetic property of the treatment, the question becomes: Is it really possible to blind a magnetic study? Magnets are familiar enough to the general public that some study participants, whether or not they were informed of the involvement of magnets, might be curious enough to do their own simple test to determine if their treatment device was magnetized. What does this do to the validity of any study using SMF therapy, as well as any product that cannot be blinded (eg, aromatherapy)? As stated by van Rijswijk, "because different evidence for different problems exists and different evidence for different problems is needed, it has long been recognized that ?one size does not fit all'."13
Conclusion
Studies on magnetic therapy, as well as other complementary treatments, are needed. If these therapies are proven to produce positive outcomes, the potential benefit to the patients and healthcare system could be enormous. Research on effectiveness should not be dismissed because study designs are challenging or because blinding is not possible. Rather, greater effort and forethought must be exerted to ensure that the endeavor is scientifically and statistically sound.
Acknowledgment
The authors would like to thank Sue Mason, MSN, RN, Supervisor WCC, and Mike Burstynski, pedorthotist, for their assistance with this study.
1. Dunn JM. Consider the alternatives. The Rotarian. 1997;20-23,60.
2. Strasen L. The silent health care revolution: the rising demand for complementary medicine. Nursing Economics. 1999;17(5):246-251,256.
3. Rubik B. Becker RO, Flower RG. Hazlewood DF, Liboff AR, Walleczek J. Bioelectromagnetics applications in medicine. A report to the National Institutes of Health on alternative medical systems and practices in the United States. Prepared under the auspices of the Workshop on Alternative Medicine, Chantilly, Va; 1992.
4. Roffey LE. Why magnetic therapy works. Massage. 1993;44:34-38.
5. Bergstrom N, Bennett MA, Carlson CE, et al. Clinical Practice Guideline Number 15: Treatment of Pressure Ulcers. Rockville, Md: U.S. Department of Health and Human Services. Agency for Health Care Policy and Research; 1994. AHCPR Publication 95-0652.
6. Baker LL, Rubayi S, Villar F, DeMuth S. Effect of electrical stimulation waveform on healing of ulcers in human beings with spinal cord injury. Wound Repair and Regeneration. 1996;4 (1):21-28.
7. Buswell W. Treating a pressure sore with PEME. Journal of Wound Care. 1992;1(2):14-18.
8. Isakov E, Ring H, Mendelevich I, et al. Electromagnetic stimulation of stump wound in diabetic amputees. Journal of Rehabilitation Sciences. 1996;9(2):46-48.
9. Szor J, Topp R. Use of magnet therapy to heal an abdominal wound: a case study. 1998;44 (5):24-29.
10. Man D, Man B, Plosker H. The influence of permanent magnetic field therapy on wound healing in suction lipectomy patients: a double-blind study. Plast Reconstr Surg. 1999;104 (7), 2261-2268.
11. Protney LG, Watkins MP. Foundations of Clinical Research Applications to Practice, 2nd ed. Upper Saddle River, NJ: Prentice Hall Health; 2000.
12. Brody Jane E. Less pain: is it in the magnets or the mind? The New York Times. Personal Health. November 28, 2000.
13. van Rijswijk L. (1999). Clinical practice guidelines: moving into the 21st century. Ostomy/Wound Management. 1999;45(1A Suppl):47S-53S.
