Tens for Chronic Low Back Pain

N Engl J Med 1990; 323:1423-1425November 15, 1990DOI: 10.1056/NEJM199011153232015

To the Editor:

Well-controlled studies of various physical-therapy treatments are essential to establish their clinical effectiveness. Deyo and colleagues (June 7 issue)1 have prudently focused their examination on transcutaneous electrical nerve stimulation (TENS), a treatment for which allowed Medicare charges in 1987 were more than $40 million.2 After concluding that active TENS was no better than a placebo, the authors suggested possible reasons for these negative results and then systematically ruled them out. It would seem appropriate, however, to examine additional factors that may have contributed significantly to the observed results.

Details of the protocols used suggest that optimal TENS treatments may not have been employed, despite the patients' being convinced that the units were functioning properly. For example, specific characteristics of electrical stimulation have been determined to produce superior pain relief (e.g., frequency close to 60 Hz and amplitude producing distinct paresthesia within the painful area for "conventional" low-intensity TENS), whereas others have been found to worsen pain in some instances (e.g., frequency above 80 Hz, as was used in the present study).3 4 5 Furthermore, the efficacy of "modulated-Pulse Rate" TENS in relation to the half dozen other common methods of stimulation has yet to be established.

Placebo effects have long been recognized as being associated with a wide range of medical and surgical interventions,6 and certainly with TENS.3 It should be recognized that in another TENS study that incorporated a strong suggestion of treatment potency and even included a strategy for focusing patients' attention, placebo effects of sham TENS were also large and equivalent to those of active TENS.7 Similar results are not seen when instructions to patients are given under more neutral or permissive conditions. At what point does a sham treatment become an active psychological treatment?

Finally, it was encouraging that significant improvements were reported for nearly every outcome measure in all four study groups for up to two months after treatment. Such results appeared to be based on a regimen involving the use of hot packs, home-based application of electric heating pads, and instructions to patients concerning lifting, standing, and resting positions, which all subjects followed. Interestingly, the use of similar educational techniques was a significant treatment factor in another study that claimed that TENS was of no benefit in managing chronic low back pain.8 In the present investigation, might not it have been possible that the positive effects of common physical-therapy treatments were sufficient to mask or minimize the effects of active TENS? In contrast, outcomes of greater magnitude were produced by an exercise program that was effective when patients complied with it.

John O. Barr, L.P.T., Ph.D.
University of Iowa College of Medicine, Iowa City, IA 52242

8 References

To the Editor:

The report by Deyo et al. was well written and the results carefully analyzed, but unfortunately the primary technique was faulty. The placement of the TENS electrodes over the area of the most severe pain is not anatomically correct for pain relief. The electrodes should be placed over the neuroanatomical sites. Evidence that the electrodes are functioning properly is demonstrated by muscle movement. If there is no muscle activity indicating stimulation by electrodes over the nerve site, the placement is incorrect. Furthermore, giving identical instructions to subjects when the causes of their pain and pathology are not the same also confounds results. In addition, it is very difficult for patients to apply the electrodes to the proper anatomical site without assistance.

My colleagues and I have obtained excellent results in the treatment of many patients with acute and chronic pain, paralysis due to stroke, head injury, and spasticity. The electrical stimulation of the muscles is itself physiotherapeutic. Pain is relieved and spasticity and motor weakness are reduced by the release of endorphins.1 TENS on the extremities or back is recorded in the brain and does increase cerebral activity.2

In conclusion, I believe that the instructions and application of TENS therapy were inadequate in the study by Deyo et al., and therefore that the conclusions are invalid. The results as statistically presented in the paper on the application of TENS for chronic back pain were based on improper methods and wrongly condemn a very useful technique for the treatment of pain and spasticity.

Arthur Winter, M.D.
New Jersey Neurological Institute, Livingston, NJ 07039

2 References

To the Editor:

The study by Deyo et al. purports to assess the value of electrodes in the treatment of musculoskeletal low back pain. Their ill-defined placement of the TENS electrodes ("initially placed over the area of most severe pain, and . . . then moved as necessary to optimize pain relief"), however, is inconsistent with a large body of literature defining proper electrode placement.1 2 3 4Specifically, Travell and Simons described musculoskeletal low back pain commonly referred from at least 13 well-defined myofascial trigger points.5 No mention was made in the study by Deyo et al. of any attempt to locate the myofascial trigger points that were the sources of pain among their subjects in order to guide proper TENS placement. We find this disturbing in that it is just such sources that appear most amenable to TENS therapy with proper placement.5 In our experience with more than 500 patients with musculoskeletal pain referred from myofascial trigger points documented thermographically6 and by physical palpation,5 proper placement of TENS electrodes appears to be the critical factor in achieving pain relief with this technique. In their study design, Deyo et al. appear to have failed to take this factor into adequate account; the TENS component of their study therefore must be regarded as seriously flawed.

David E. Conwill, M.D., M.P.H.
University of Mississippi Medical Center, Jackson, MS 39216

Billy H. Cook, M.D., Ph.D.
Outpatient Center, Pleasant View, TN 37146

6 References

To the Editor:

The study by Deyo and colleagues addresses the therapeutic efficacy of TENS for chronic low back pain. However, the analgesic efficacy of TENS while actually in use does not seem to have been assessed directly. In view of this, the rather short time that TENS was applied each day (45 minutes three times a day), and the limitations imposed by the study design on the optimization of TENS settings, the investigation cannot really be said to have added much to the debate about the analgesic capabilities of TENS. After a careful review, Woolf* concluded that

The fact that placebo effect contributes to the pain relief should be regarded as a bonus, but there is little doubt that there is, in addition, a genuine alteration in sensibility produced by TENS. TENS has, however, limited efficacy in the sense that while it can eliminate moderate or mild pain it is less effective against severe pain.

The study by Deyo et al. adds weight and emphasis to the notion that TENS has little specific therapeutic value, but few reputable pain clinics would claim otherwise, and I believe that Woolf's assessment may still be acceptable. It is beyond dispute that some patients with chronic pain find relief with long-term ambulant use of TENS. Like many things in medicine, TENS has a place. This study contributes to the definition of that place with respect to low back pain but should not be taken as evidence that TENS has no place at all.

Alan Merry, F.F.A.R.A.C.S
Green Lane Hospital, Auckland 3, New Zealand

1 Reference

To the Editor:

I am writing to express my concern about the impact of the publication of the study conducted by Dr. Deyo and his colleagues at the Seattle Veterans Affairs Medical Center on the effectiveness of TENS devices in the treatment of lower back pain. Because of the excellent reputation of theJournal, publication of this study resulted in articles in major newspapers across the country. The negative tone of these articles left the impression that TENS is ineffective in treating pain. Not only has this affected the public's perception of the effectiveness of TENS devices, but the barrage of publicity also caused one large insurance company to cancel coverage for TENS therapy, with three more instructed to discontinue payments.

TENS therapy is effective. For 20 years physicians and therapists have prescribed TENS to hundreds of thousands of patients with pain as an effective alternative to addictive narcotics and unnecessary surgical procedures. This clinical experience is backed by a 1989 report from the Office of the Inspector General,* which found that nearly 80 percent of the patients using TENS find relief from pain. It is only one of many reports and studies that confirm the efficacy of TENS therapy. Even Deyo et al. said that "perhaps TENS is efficacious for other pain syndromes."

Greg A. Todd
Medical Designs, Inc., Westerville, OH 43081

1 Reference

The above letters were referred to the authors of the article in question, who offer the following reply:

To the Editor: Drs. Barr, Winter, Conwill and Cook, and Merry all express concern about the technique of TENS employed in our Study. However, each identifies a somewhat different detail of the technique as being problematic and has a somewhat different solution. These views highlight the idiosyncratic nature of clinical protocols for the use of TENS. In the absence of rigorously controlled clinical research to address these issues, it is not surprising that the technical details of TENS application vary widely among practitioners.

The stimulation frequency, amplitude, duration and frequency of use, and electrode placement in our study were well within the range generally recognized as appropriate in TENS therapy.1 2 3 4 5Previously reported randomized trials of TENS for chronic low back pain all used electrodes over the center of pain, with use ranging from two times a week for 30 minutes to two treatments a day for 20 minutes.3 4 5 Two of these trials purported to show a benefit of TENS,4 , 5 and one reported relief lasting up to seven hours after periods of stimulation. Our subjects were advised to use the TENS units "at least three times a day for 45-minute periods." The instruction manual given to each patient indicated that there was no limit to the frequency of treatment in one day, but that the unit should be turned off for at least five minutes after every hour of stimulation.4 When TENS was used in the acupuncture-like mode, muscle activity was usually visible. The duration of use and electrode placement were highly individualized for each patient, and two modes of stimulation were attempted, with the patient selecting that which offered the best relief.

Patients often reported that it was difficult to apply electrodes by themselves. Fortunately, most had a spouse who could assist in this process at home. We agree with Dr. Barr that education of patients, local application of heat, and a therapist's interest and concern may have valuable therapeutic effects. Our study suggests, in agreement with Dr. Barr, that TENS adds little to these basic therapeutic interventions. Though Drs. Conwill and Cook feel strongly about the importance of myofascial trigger points, an expert panel convened by the Institute of Medicine could not even agree about their existence,6 and previous randomized trials have not used such points for electrode placement.3 4 5

The Inspector General's report was prompted by complaints from Medicare beneficiaries concerning aggressive and possibly fraudulent marketing techniques. Mr. Todd has reported the results inaccurately. Nearly 80 percent of beneficiaries who purchased TENS units found them beneficial, but the survey also included patients who had tried TENS, not found it useful, and not purchased a unit. In fact, 126 patients were contacted by the survey. Nine percent had possibly fraudulent claims, since beneficiaries denied ever receiving a TENS unit. Thirty subjects had rented TENS units but had not purchased them, and 80 percent of these found that the TENS units were not helpful. Thus, of the 115 persons who had actually used TENS units, only 63 percent reported that they were helped. This figure is similar to the percentage of our subjects who indicated at the end of treatment that they would like to continue TENS therapy (68 percent for true TENS and 56 percent for sham TENS).

We believe that TENS therapy was given a rigorous and fair test in our study. We hope that those who advocate the use of TENS in other stimulation modes or for other conditions will conduct equally rigorous randomized trials to provide support for their views.

Richard A. Deyo, M.D., M.P.H.
Seattle Veterans Affairs Medical Center, Seattle, WA 98108