Recently reported research found that transcutaneous electrical nerve stimulation (TENS) was significantly helpful in reducing chronic pain intensity and disability while improving perceived health status for up to 1 year. However, these outcomes were comparable to those produced by sham/placebo TENS. Does this mean that TENS is worthless as a pain-treatment modality, or is there a need to reconsider the role and value of meaningful placebos in pain management?
TENS is an easy to use, noninvasive, analgesic intervention that may result in less pain, increased functionality, and decreased use of pain medication. Although TENS has been in use since the early 1970s, short-term results of this therapy have been inconclusive for treating chronic pain, and long-term randomized placebo-controlled studies with treatment periods of more than 3 months had not been executed. Therefore, researchers in The Netherlands designed a study to explore the long-term (1 year) time course of TENS treatment effects compared with a sham-TENS placebo [Oosterhof et al. 2012].
Writing in the September edition of Pain Practice they report a randomized placebo-controlled trial enrolling 163 patients with chronic pain who had been referred to a multidisciplinary pain center at a university hospital. Patients were being treated for peripheral neuropathic pain, osteoarthritis and related disorders, or injury of bone and soft tissue, and usual care for their conditions had failed in the past to satisfactorily ameliorate their pain.
Prior to randomization, all patients had their pain medication optimized by the attending anesthesiologist or pain practitioner; however, at baseline, the mean pain intensity among all patients was still moderate (about 62mm on a 100mm scale). Identical TENS devices were used for the actual and sham treatments, and self-applied by subjects for several hours each day. Sham units showed a fake output reading on the LCD display, but no current was delivered to the electrodes. The main endpoints of interest were the proportion of patients satisfied with treatment results and willing to continue treatment at the end of 1 year, pain intensity, pain disability, and perceived health status.
Results indicated that, throughout the time-course of the study, there was no significant difference (P=0.79) in the proportion of patients satisfied with actual vs. sham-TENS therapy; at the end of the year, an intention-to-treat analysis showed that 30% (24/81) of patients in the TENS group and 23% (19/82) of the sham-TENS group were satisfied with treatment results. These 43 patients in both groups experienced a mean overall improvement of 62.7%, and there were no significantly different between groups (P=0.74). Also, for patients still satisfied after 1 year, there were no differences in pain intensity or disability, perceived health status, or pain medication use between the TENS and sham-TENS groups; although, these measures in both groups had improved significantly from baseline.
COMMENTARY: Limitations & Possibilities
As noted above, TENS therapy has been in existence for nearly 4 decades. The TENS unit is a small, portable device delivering mild electrical current to nerves through electrodes connected to the skin at or near the sites of pain.
The approach has been used for various types of pain, although exact mechanisms of its analgesic effects are still under examination. Basic science studies suggest that TENS activates endogenous pain-control chemicals (eg, endorphins, enkephalins, dynorphins, GABA, serotonin) and their receptors. Along with that, the high and low electrical frequencies produced by a TENS unit inhibit pain signals along affected nerves, ostensibly blocking the impulses from reaching pain-perception areas in the brain.
In the Oosterhof et al.  study, for those patients who appeared to respond to a TENS intervention and were satisfied with the therapy, there were stable improvements in pain and other measures over the 1-year period, whether they were assigned real or sham TENS units. However, as the authors concede, there was no third group for comparison receiving standard care without TENS to control for regression to the mean or the natural course of chronic pain, so they could not claim to have found a true placebo effect induced by the sham-TENS procedure.
During the year, 44 patients in the TENS group were lost to followup, with 18 dropping out due to dissatisfaction with treatment results. In the sham-TENS group, there were only 32 lost to followup, with 14 dissatisfied with results. Also, 93% of subjects in the actual TENS group but only 70% in the sham-TENS group believed they had received a real TENS unit, which should have decreased favorable placebo effects but did not.
The intervention was relatively safe, with the only adverse effect being that roughly half of all patients experienced skin problems caused by the electrodes, but only 4 (2 in each group) discontinued for this reason. Apparently, this was not due to actual electric current and the researchers noted that this problem can normally be overcome by changing the type of electrode.
The researchers note quite importantly that among those patients still satisfied with either actual or sham TENS after 1 year there was an average decrease of more than 50% in pain intensity, which is both statistically and clinically significant. Even among those who stopped treatment earlier, there was about a 28% improvement in pain, which may be clinically noteworthy since this went beyond the pain relief afforded them by usual medical care.
As one limitation, there may not have been a sufficient number of subjects completing the trial. The study design required 35% of patients in the TENS group and 15% in the sham-TENS group being successfully treated after 1 year to have 80% power for detecting significant difference between groups. Since these targets were not achieved, the lack of differences between groups might have been statistically a false negative, or Type II error.
Aside from that, in this long-term study by Oosterhof and colleagues  it appeared that TENS might have functioned similar to placebo or, conversely, the sham/placebo treatment may have had genuine medical efficacy. As the researchers note, this may support the contention that “placebo effects are genuine psychobiological events, which can be robust in both laboratory and clinical settings.”
Along these lines, the researchers observe that, in neuropathic pain trials, placebos have had durable long-term effects that pose difficulties in distinguishing between treatment and placebo effects. Furthermore, other researchers [eg, Quessy and Rowbotham 2008] have noted that this problem is magnified in trials in which long treatment periods under blinded conditions are required and in which the analysis assigns trial dropouts as treatment failures, as in the Oosterhof et al. study.
TENS for Chronic Pain
Other research has investigated TENS for chronic pain. A Cochrane Systematic Review by Nnoaham and Kumbang  compared “no treatment” controls with sham-TENS or active TENS using different electrical frequencies. Of 124 studies identified in their searches, only 25 RCTs (N=1,281) could be evaluated. At that, there was such a high degree of heterogeneity (differences and inconsistencies) across studies that a data meta-analysis was not possible.
Overall, in 13 of 22 sham/placebo controlled studies there was a positive analgesic outcome in favor of active TENS treatments. For multiple-dose treatment comparison studies, 8 of 15 were considered to be in favor of TENS therapy. Results were based on short-term low-volume TENS treatment: The duration of treatment was <4 weeks in about 80% of the studies, and in 70% of the trials treatment occurred <10 hours per week with 60% of the participants having <10 total sessions of TENS. The authors note that these limitations may explain why some of the studies failed to detect any differences between active TENS and sham controls.
Nnoaham and Kumbang conclude that published literature on TENS lacks the methodological rigor or robust reporting needed to make confident assessments of this therapy for chronic pain management. However, they assert that, even if the effect of TENS on chronic pain is a weak one, its potential to augment the effect of other pain treatment modalities should be explored. Indeed, in the much longer-term Oosterhof et al. study, TENS therapy was additive to analgesic effects of pain medications that all patients were taking.
TENS for Low-Back Pain
We have previously discussed in an UPDATE [here] evidence reviews and guidelines that found TENS ineffective for treating chronic low-back pain; although the evidence had some strong limitations that challenged its internal and external validity. Similarly, in 2008, a Cochrane Systematic Review by Khadilkar et al. examined 4 high quality RCTs of TENS for chronic back pain (585 patients) and found conflicting and inconsistent evidence, in comparison with placebo, to support the use of TENS. However, these studies were so disparate in design (clinical heterogeneity) that the researchers could not do a data meta-analysis and had to rely on qualitative observations that are prone to bias.
In response to the lack of clear and convincing evidence in support of TENS for chronic back pain, last summer the U.S. Centers for Medicare and Medicaid Services (CMS) announced it would no longer cover most uses of TENS for this pain condition [see MedPage Today article here]. They noted that reimbursement for TENS in treating low-back pain in particular will be available only when patients are participating in a randomized, controlled trial of the technology’s clinical effectiveness.
In the announcement they wrote, “TENS is not reasonable and necessary for the treatment of [chronic low back pain].” The CMS had conducted a review in the wake of a 2010 report by an American Academy of Neurology panel that, on the basis of only 5 trials, found the treatment was not effective [discussed in the abovementioned UPDATE]. Although, the CMS also acknowledged that some individual studies have shown that TENS can reduce pain and improve patients' physical function.
While the CMS plans to withhold coverage of TENS for chronic back pain, it will continue to fund RCTs of TENS for 3 years. The trials must directly address TENS’ clinical efficacy and be designed and powered to yield clear-cut answers. At present, the CMS emphasized that Medicare will still reimburse for TENS prescribed for treatment-resistant pain conditions other than low back pain, such as for chronic or severe postoperative pain.
When is a “Placebo” of Meaningful Value?
There are some interesting and remarkable parallels of research outcomes for TENS, as described above, with acupuncture for chronic pain in regard to the prominent influence of placebo effects as a possibly important component of therapeutic efficacy. Acupuncture was most recently discussed in an UPDATE [here], which noted that its benefits for treating musculoskeletal pain, osteoarthritis, and chronic headache were only weakly better than placebo, but that both placebo and true acupuncture conferred moderate benefits in reducing pain that went beyond and above the usual medical care patients had been receiving.
Both acupuncture and TENS have been available for a long time and there has been much research conducted to examine each modality; yet, the quality of most research has been surprisingly low and generally precludes an unbiased pooling of studies for more rigorous analyses. Despite this, it appears that both modalities have potential for offering select groups of patients nonpharmacologic options for pain relief and other health benefits; albeit, perhaps due largely to placebo components of the overall effects. And, as the Oosterhof et al. study demonstrates, the duration of those favorable effects can be long-lasting.
TENS is noninvasive and acupuncture is minimally invasive, both have relatively favorable safety profiles, and their benefit-to-cost ratios may be advantageous compared with some other therapies. Either therapy may be abruptly discontinued without adverse effects and, as the Oosterhof et al. study suggests, patients who do not benefit readily stop the treatment.
So, perhaps what is needed is a reconceptualization of the role of placebos in pain management and, when it comes to TENS or acupuncture, a recognition that a completely inert placebo version of these therapies may not exist.
As Andrew Avins, MD, suggested in an editorial discussed in our recent UPDATE on acupuncture, it may be time for an examination of why so many healthcare professionals feel threatened by the existence of placebo effects and, instead, consider how those effects might be harnessed for better pain care. Capitalizing on placebo effects may be perceived as bad science by some, but it may be welcomed as good medicine by patients who benefit.
Disclosure: We have no vested interests in or support from any manufacturers of electrical nerve stimulation devices, including TENS. Our only interest here is in arriving at sound practice decisions in pain management based on appropriately credible, reliable, and valid analyses of clinical evidence. — SBL
> Khadilkar A, Odebiyi DO, Brosseau L, Wells GA. Transcutaneous electrical nerve stimulation (TENS) versus placebo for chronic low-back pain. Cochrane Database of Systematic Reviews. 2008;4(CD003008) [abstract].
> Nnoaham KE, Kumbang J. Transcutaneous electrical nerve stimulation (TENS) for chronic pain. Cochrane Database of Systematic Reviews. 2008;3(CD003222) [abstract here].
> Oosterhof J, Wilder-Smith OH, de Boo T, et al. The Long-Term Outcome of Transcutaneous Electrical Nerve Stimulation in the Treatment for Patients with Chronic Pain: A Randomized, Placebo-Controlled Trial. Pain Practice. 2012(Sep);12(7):513-522 [abstract here].
> Quessy SN, Rowbotham MC. Placebo response in neuropathic pain trials. PAIN. 2008;138(3);479-483 [abstract here].
Don’t Miss Out. Stay Up-to-Date on Pain-Topics UPDATES!
Register [here] to receive a once-weekly e-Notification of new postings.