Belief in the healing power of magnets and magnetic fields has existed since the discovery of magnets several thousand years ago. In the late 18th century, Franz Anton Mesmer, an infamous charlatan, promoted the notion that he could heal with “animal magnetism.” In the 19th century magnetic healers were common – D.D. Palmer was a magnetic healer prior to founding chiropractic. Magnetic devices for everyday aches and pains have been increasingly popular recently, and today they are a multi-billion dollar industry.
Yet the scientific evidence does not, generally, support the use of magnets for specific indications, and the vast majority (if not totality) of claims made for magnetic devices in marketing are either false or unsupported and highly implausible. The media attention given to a recent study of static magnetic fields (SMF) in the treatment of inflammation brings up many important points regarding this disconnect.
A sciencedaily.com headline from January 7th proclaims: “Healing Value Of Magnets Demonstrated In Biomedical Engineering Study.” This headline is extremely misleading, especially the use of the word “healing.” The article is referring to a study by CE Morris and Thomas Skalak(1) published two months ago in the American Journal of Physiology – Acute Exposure to a Moderate Strength Magnetic Field Reduces Edema Formation in Rats. In the sciencedaily article Skalak is quoted as saying:
“We now hope to implement a series of steps, including private investment partners and eventually a major corporate partner, to realize these very widespread applications that will make a positive difference for human health.”
This optimism, however, is premature, and represents a significant problem with many popular therapies, the extrapolation from preliminary pre-clinical studies to clinical applications in humans. But before I look at this new study in more detail I will review some basic concepts for background.
First, it is important to recognize that not all magnets or magnetic fields are the same. The most significant difference is between pulsating magnetic fields and static magnetic fields. Electricity and magnetism are actually manifestations of the same fundamental force: electromagnetism. This was first recognized when it was discovered that a changing magnetic field can generate and electrical current, and a changing electrical current can generate a magnetic field. A pulsating magnetic field, therefore, is capable of generating an electrical current. Many aspect of cell function and communication involve electrical potentials or currents, and therefore it is plausible – at least from a physical point of view – for a pulsating magnetic field to affect electrical current in tissue and thereby manifest an effect. The best established clinical use of a pulsating magnetic field is in the healing of bone fractures. Although this effect is modest, the evidence so far supports the conclusion that there is a relevant biological response.
Static magnetic fields do not generate electrical current, and therefore any biological or medical evidence dealing with pulsating magnetic fields cannot be applied to SMF. An even more significant problem with SMF as a therapeutic modality is that the strengths of magnets commonly used are very low, probably too low to have any biological effect. Further, magnetic field strength decreases as the square of the distance or even as the cube of the distance from its origin (depending upon the shape and orientation of the magnets). So field strength decreases very quickly with distance.
Most magnetic health products list the magnetic field strength at the core of the magnet, but even 1 cm distance may decrease the field to negligible strength. Therefore, if a magnet is in a bandage or casing the field strength will be significantly decreased at the surface of the skin, and then even further decreased below the skin down to the target tissue. (David Ramey goes into this in more detail in this excellent review article.)
Assessing the biological plausibility of magnetic therapy reveals further difficulties. So far no proposed mechanisms for how a SMF would provide a therapeutic effect have been established. Some common claims include that SMF increases blood circulation, thereby providing more oxygen to the tissue and reducing inflammation, but again this has not been established and there is no known mechanism by which this would occur. When tissue lacks sufficient oxygen it sends out powerful chemical signals to increase blood flow. It is unlikely that a weak SMF would provide a stimulus for vasodilation (increasing blood vessel size) greater than what occurs naturally in the setting of low oxygen. Also, ironically, the claims made in the current study by Skalak is that SMF decreases blood flow to inflamed tissue.
Older claims, that SMF attracts the iron in blood, are easily dismissed because the form of iron in blood is not ferromagnetic. More recent claims that SMF affects the flow of ions in the blood, altering blood flow, are not plausible as the SMF strengths are too low to significantly affect the movement of ions. Still others claim that SMF decreases inflammation, and that is the subject of this recent study.
Skalak’s current study is an extension of prior work in which he demonstrated that, under a SMF, blood vessels that were already constricted would tend to dilate and those that were already dilated would tend to constrict. Therefore, a SMF must have a normalizing effect on blood vessel tone. This study has yet to be properly replicated, and one concern about such results is that they could simply be due to regression to the mean (a normalizing statistical effect).
Based upon this preliminary work, Morris and Skalak conducted a study in which they look at various SMF strengths a in a rat model of inflammation. Two different chemical irritants were used to produce local inflammation in a rat hind paw. The paws were then subjected to SMF of either 10mT (miliTesla), 70mT, or 400mT. What they found is that swelling due to histamine induced inflammation was decreased in the 10mT and 70mT groups, but not400mT, and not in the rats where inflammation was induced by lambda-carrageenan (CA). The effects were only seen if the SMF was applied immediately after inflammation was induced, not prior and not delayed or during peak edema. The authors conclude that this shows a dose and temporal dependent effect of SMF on inflammation, probably due to vasoconstriction reducing blood flow to the inflamed tissue. To their credit they measured the strength of the magnetic fields at the level of the tissue, not the core of the magnet, so field penetration was not an issue. (In fact the authors acknowledge that virtually all magnetic products currently on the market report core SMF strength, which is misleading.)
There are significant problems with this interpretation, however. First, there is not a consistent dose response effect, as the strongest field (400mT) showed no effect. The authors make the post-hoc analysis that there is an upper therapeutic threshold, but this was not predicted by the original hypothesis and there is no known mechanism for such a threshold. The more standard interpretation of such a result is a lack of consistent dose-response. Also, the effect was only seen in the histamine induced inflammation and not CA induced inflammation – again, not predicted by the working hypothesis of the study. If SMF works through vasoconstriction of dilated arteries, as the authors hypothesize, why would the mechanism of inflammation matter? Therefore, these results do not show a consistent or predicted pattern, and could easily be interpreted, if taken together, as a null effect.
The authors also looked at the effects of pharmacologically blocking L-type Ca(2+) channels or nitric oxide (NO) to see if either blocked the effects of the SMF. This is a clever and established way to infer mechanism of action, and they found that the former, but not the latter, did block the effect of decreased inflammation by a SMF. Therefore the authors hypothesize that the SMF may work its effect through calcium channels. But again, random effects cannot adequately be ruled out, given the large number of outcomes that were measured in total.
The study design and results are indeed interesting, and I think that such serious attempts at looking at various aspects of SMF effects, including possible mechanism of action, are constructive. However, these current results are only useful as preliminary data since the results require post-hoc analysis and were not specifically predicted ahead of time. In other words, no reliable conclusions can be drawn from this study until it is replicated to see if the pattern of results hold up. Is there really a 400mT therapeutic threshold, or was this simply a lack of dose-response, and therefore a negative result?
Also, there are significant limitations in applying this study to humans. As the authors recognize, SMF strength decreases significantly with distance. Rat paws are very small, and it is therefore practical to apply a strong SMF at tissue depth. A human joint or limb, however, is significantly larger requiring a much stronger SMF in order to maintain adequate strength at tissue depth. If these results represent a real physiological effect, they may not apply at all to humans or may simply be impractical.
Therefore it is premature to conclude from this study that SMF’s can be applied to treating inflammation in humans, or that there is any specific clinical effect. At best this study requires replication and may point the way toward further research.
The media reporting of this study has been misleading and highlights some of the difficulty in dealing with such research in areas where there is already a vigorous and largely unregulated market. Preliminary and problematic results, such as we have here, are presented by the media as if they were reliable clinical evidence that magnets can “heal.” Such studies are also used to market specific products that differ significantly from the magnets used in the study, and to make clinical claims that cannot be extrapolated from the available data.