Research suggests treatment improves sensation, but decision makers question the evidence
Published in the July 2007 issue of BioMechanics
By Jordana Bieze Foster
David A. Arnall, PT, PhD, is not what one would call an early adopter. A self-proclaimed professional skeptic, he is a believer in evidence-based medicine who, under most circumstances, puts no stock in fledgling therapies until their efficacy has been validated through rigorous scientific study.
The use of infrared light therapy for peripheral neuropathy is not what one would call evidence-based medicine-certainly not if one works for the Centers for Medicare and Medicaid Services. In an October decision memorandum, CMS ruled that the handful of studies were sufficient to conclude that the therapy was neither reasonable nor necessary for that indication. Most of the studies were retrospective or had other limitations; the single well-designed randomized, controlled trial found no significant benefit.
And yet Arnall, the skeptic, will tell anyone who will listen that he believes infrared light therapy can restore sensation and reduce pain in patients with diabetic peripheral neuropathy.
Indeed, this opinion has something to do with scientific evidence: Arnall conducted a 22-patient study in 2004 in Spain that found protective sensation significantly improved in neuropathic feet treated with pulsed infrared light therapy (PILT) compared with nonsignificant improvements or deterioration in untreated feet. In lieu of a placebo treatment in the study, one of each patient's feet was left untreated as a within-subject control. The results were published in the May 2006 issue of the European journal Acta Diabetologica.
But Arnall's opinion is also grounded in personal experience. Having lived with diabetes for 20 years, he had tried every available therapeutic option, including decompression surgery, to relieve his peripheral neuropathy with no luck. Infrared light therapy, he says, succeeded where all the other treatments had failed.
'My neuropathies were resolved,' said Arnall, who is a professor and chair of physical therapy at East Tennessee State University in Johnson City. 'Now I'm a believer, not only because of the positive outcomes of the Spanish diabetes patients, but also because of my own case. This is one modality that is spectacular.'
Arnall isn't alone in having had a positive experience with infrared light therapy for peripheral neuropathy. The CMS memo authors received 1315 comments during the initial public comment period, and all but three supported the use of the therapy. Five of those supporters were, like Arnall, clinicians who had seen positive effects on their own symptoms as well as those of their patients.
Although the scientific evidence supporting the use of infrared light therapy for peripheral neuropathy has thus far failed to convince CMS decision makers of its merits, the number of anecdotal success stories has kept the technology from dropping completely off practitioners' radar.
Given the severe consequences of untreated neuropathy-ulceration and, ultimately, amputation-and the lack of effective therapeutic alternatives, even some skeptics remain willing to believe that better research on infrared light therapy might conclusively document a positive effect. Also, a stimulated discussion at the American Physical Therapy Association's annual Combined Sections Meeting in February, when additional positive outcomes from a study were reported, served notice that the CMS memo may not be the final word on the therapy's potential efficacy.
Confusion about infrared light therapy, advocates say, begins with a lack of knowledge.
'Physicians are not exposed to light therapy during their training, so at least part of the reluctance of clinicians to look favorably on it is because they don't understand it well,' said Thomas J. Burke, PhD, director of research and clinical affairs for Anodyne Therapy and a coauthor of published studies on the company's infrared therapy technology. 'It's a novel field that will take time, especially in American medicine where everything is focused on pharmaceutical treatments, for physicians to get a handle on it.'
Even its strongest advocates, however, remain unsure of the exact mechanism by which the treatment works. Animal studies have shown that low-level laser irradiation improves microcirculation through vasodilation, and researchers believe that infrared light therapy triggers a similar mechanism, possibly involving the release of nitric oxide from hemoglobin. Restoring microcirculation is thought to improve sensation and reduce pain.
Further adding to the confusion about infrared light therapy are subtle, but possibly significant, differences that exist between devices used to deliver the therapy. The majority of studies have used monochromatic infrared energy (MIRE) devices (Anodyne Therapy, Tampa, FL) that deliver pulsed photoenergy at a wavelength of 890 nm via arrays of diodes contained in flexible pads strapped to the skin. Arnall's study involved PILT delivered by a RevitaMed (LymphaCare Healthlight Therapy, New York, NY) device. In this device, each therapy pad contains six rows of diodes emitting pulsed photoenergy at a wavelength of 880 nm (infrared light) alternated with five rows of diodes emitting light at 650 nm (visible red light).
The FDA has not cleared any phototherapy devices for treatment of peripheral neuropathy. Many devices have received approval for the temporary relief of muscle and/or joint pain and the temporary increase of local blood circulation, similar to the effects of an electric heating pad, but the agency has indicated that those approvals do not extend to the treatment of neuropathic pain.
The outpouring of impassioned but anecdotal evidence in support of infrared light therapy will likely carry as little weight with the FDA as it did with the CMS decision makers, whose opinions were based largely on the scientific strength of the evidence against it.
The only well-designed randomized, placebo-controlled trial of infrared light therapy, published in the December 2005 issue of Diabetes Care, found no significant difference in symptom improvement between patients with peripheral neuropathy whose feet were treated for four weeks with an active therapy device and those who were treated with a sham device.
In that study, researchers from the University of Tennessee in Memphis and Pulaski Physical Therapy in Pulaski, TN, tested the sensitivity of 39 subjects to 5.07 Semmes-Weinstein monofilaments at four sites on each lower extremity. After 12 treatment sessions administered over four weeks, the mean number of sites sensitive to the 5.07 monofilament improved from 0.57 (out of five) to 0.94 in the 35 lower extremities randomized to receive active MIRE treatment and from 0.86 to 1.42 in the 35 lower extremities that received placebo treatment. The change from baseline for each group was statistically significant, but the difference between groups was not.
An earlier study of infrared light therapy for peripheral neuropathy at the Joslin Center for Diabetes in Clearwater, FL, also involved a randomized, placebo-controlled design but the sham therapy was administered for only six of 12 treatment sessions, thus limiting the conclusions that can be drawn from the study's findings. During the first six sessions, each of 27 subjects received active MIRE treatment on one leg and a sham treatment on the other leg. All subjects were insensate to 5.07 monofilaments in at least two of five sites on the plantar surface of each foot, but were also subgrouped according to the subjects' ability to sense a 6.65 monofilament.
Interestingly, the researchers found that six sessions of MIRE treatment were associated with a significant decrease in the mean number of test sites insensate to a 5.07 monofilament, but only in those with less severe neuropathy at baseline. In those who could sense the 6.65 monofilament at all five test sites at baseline, mean number of sites insensate to a 5.07 monofilament decreased from 2.4 to 1.9 (a statistically significant change) in those who received active treatment and from 3.0 to 2.3 (not statistically significant) in those who received sham treatment. However, in those who were insensate to the 6.65 monofilament at baseline, the number of sites insensate to the 5.07 monofilament did not change significantly in either treatment group.
Those findings, published in the January 2004 issue of Diabetes Care, seemed to suggest that the efficacy of infrared light therapy is inversely related to the severity of a subject's peripheral neuropathy. The findings of other researchers, however, suggest that even those with very severe loss of sensation at baseline can benefit from the therapy.
Rather than reporting numbers of sites sensate or insensate to 5.07 monofilaments, Arnall et al documented sensation at each site in terms of the smallest monofilament that could be felt. In feet that received eight weeks (24 sessions) of PILT, mean peripheral protective sensation improved significantly at all four tested sites-the first, third, and fifth metatarsals, and the great toe-whereas mean sensation in the contralateral, untreated limbs did not change significantly from baseline (Table 1).
In contrast to the Florida findings, however, Arnall and colleagues found that subjects with severe baseline neuropathy did experience significant positive effects; in fact, improvement was more likely to be statistically significant at sites where subjects were insensate to 10 g of force (the amount required for a 5.07 monofilament to buckle) at baseline than in those with less severe neuropathy (Table 2).
Severity of neuropathy also did not appear to preclude a positive outcome in a 2006 retrospective review of medical records for 2239 subjects with peripheral neuropathy who had been treated with MIRE and physical therapy. The multicenter study, led by researchers from the University of Texas Health Sciences Center in San Antonio, found that the number of sites on both feet that were insensate to a 5.07 monofilament decreased significantly following treatment of varying durations, from a mean of 7.2 (out of 10) to a mean of 2.4. Significant improvement was also seen in the nearly 800 subjects who were insensate to 5.07 monofilaments at all 10 sites at baseline; following treatment, they were able to sense a mean of 6.2 sites.
Those findings, published in the March-April 2006 issue of the Journal of Diabetes and Its Complications, were consistent with those of an earlier retrospective study published in the March-April 2005 issue of the Journal of the American Podiatric Medical Association by three of the same authors.
The March-April 2006 findings also suggest that MIRE can help reduce the pain experienced by some patients with peripheral neuropathy. In 1563 subjects with neuropathic pain at baseline, initial mean pain levels of 7.2 on an 11-point visual analog scale were reduced to 2.4, a statistically significant improvement of 67%. Etiology of a subject's peripheral neuropathy did not significantly affect outcomes for either sensation or pain.
These data raise the possibility that the subjects in the Tennessee randomized controlled study may not have received enough treatments to see a significant between-group difference in outcomes. Although the 12 treatments administered were consistent with the manufacturer's guidelines, Burke said treatment durations in the two retrospective studies did vary depending on severity of symptoms.
A smaller-scale retrospective analysis of the medical charts of 30 patients with peripheral neuropathy, presented at the Combined Sections Meeting, also reported that the average number of MIRE treatments was 14. That study, conducted at Louisiana State University in Baton Rouge, found that the treatment was associated with significant reductions in pain and improvements in bilateral lower extremity sensation and balance.
The possibility of a placebo effect, however, also cannot be ruled out. All subjects in the Tennessee study had access to pamphlets on diabetic foot care, which may have resulted in lifestyle changes that could have positively affected sensation. Because the study did not include an untreated control group, this premise remains unclear.
'In our placebo group, they improved,' said T. Scott Newton, PT, DPT, owner of three physical therapy centers, including Pulaski Physical Therapy, and a coauthor of the Tennessee study. 'A lot of them swore they were on the active units and were very surprised to learn that they were not in the treatment group.'
Fortunately, more answers may be forthcoming. Researchers at the Mayo Clinic in Scottsdale, AZ, are currently recruiting patients with painful peripheral neuropathy for a randomized, placebo-controlled clinical trial of monochromatic near-infrared photoenergy therapy. Although the primary end point will be pain reduction as measured by a visual analog scale after four weeks of treatment, the researchers also plan to measure sensation using great toe vibration detection and to measure function using the Short-Form-8 Health Survey and Neuropathic Impairment Score.
In addition, Arnall has been in discussions with Judy Clifft, PT, the lead author of the Tennessee study, about a project that will combine aspects of each researcher's previous work.
All parties involved seem to agree that the needs of patients with peripheral neuropathy are too dire for infrared light therapy to be ruled out completely at this point. Even Newton, who chose not to purchase a MIRE device for his practice following the conclusion of the Tennessee study, plans to refer interested patients (who are willing to pay out of pocket) to other area therapists who offer the service.
Jordana Bieze Foster is a freelance writer based in Massachusetts and a former editor of BioMechanics magazine.