Transcutaneous Electrical Nerve Stimulator (TENS) meets primary coverage criteria for effectiveness and is covered for treatment of chronic intractable pain unresponsive to conservative medical therapy.  This treatment is not covered for the treatment of TMJ disorders, headache or generalized pain not originating from a defined anatomical source.  TENS is not appropriate for pain of internal, visceral origin (including that from the head, chest, abdomen or pelvis).  

 

The TENS unit is used on a trial basis for a minimum of one month, not to exceed two months.  The trial period is monitored by a physician for effectiveness.  The trial is paid as a rental, with that allowance applied toward the purchase of the device, if the device is effective.

 

CPT 64550 (application of surface transcutaneous electrical stimulator) is billed by the physician for instruction in the use of the TENS unit.  This is allowed no more than twice.  

 

Services not meeting the above requirements are not covered based on benefit certificate primary coverage criteria that there be scientific evidence of effectiveness.

 

For contracts without primary coverage criteria, services not meeting the above requirements will be considered not medically necessary.  Medically unnecessary services are an exclusion in the member benefit contract.

 

 

 

 

 

 

 

 

 

Review of literature through 1998, of TENS for the treatment of chronic and postoperative pain, concluded that the evidence did not clearly show that the effects of TENS exceeded placebo effects. An updated literature search in October 2002 identified several Cochrane reviews of TENS.  One of the reviews, last amended in June 2000, addressed chronic pain due to a variety of conditions (e.g., osteoarthritis of the knee, rheumatoid arthritis of the wrist, pancreatic, myofascial trigger points, chronic back pain, temporomandibular joint pain, and a variety of nocioceptive and neuropathic causes of pain) (Carroll, 2002).  A total of 19 randomized trials were judged as meeting study selection criteria, but due to heterogeneity of methods and inability to extract sufficient dichotomous pain outcomes data, it was concluded that meta-analysis was not possible, and the review of evidence was inconclusive. The trials reviewed did not indicate which stimulation parameters were most likely to provide pain relief, or answer questions about long-term effectiveness. The authors suggested a need for large, multicenter, randomized, controlled trials of TENS in chronic pain.

 

In a 2004 literature review update, two additional Cochrane Reviews were identified along with several randomized controlled trials (RCTs) on the use of TENS. Neither the Cochrane Reviews nor any of the RCTs identified were sufficient to alter the previous conclusions. The authors of the Cochrane reviews concluded that the evidence was inadequate to draw conclusions about the effects of TENS.

 

The update in 2007 examined the 10 Cochrane reviews on TENS that had been published over the previous 7 years. Overall, evidence for the efficacy of TENS for different pain conditions was weak. Two Cochrane reviews concluded that there is limited and inconsistent evidence for the use of TENS as an isolated treatment for low back pain, and that results in patients with rheumatoid arthritis of the hand were conflicting.  Efficacy of TENS for neck pain, headache, shoulder pain after stroke, and dementia were considered inconclusive in 3 other reviews.  One Cochrane review of 9 small, controlled trials found high-frequency TENS to be effective for the treatment of dysmenorrhea (Proctor, 2002).  Another review found TENS and acupuncture-like TENS to be more effective than placebo for the treatment of knee osteoarthritis, but indicated that due to heterogeneity of the included studies, more well-designed trials with adequate numbers of participants were needed to conclude effectiveness (Osiri, 2002).

 

Two new systematic Cochrane reviews have been initiated to assess the use of TENS for cancer pain and acute pain. Factors that may influence efficacy, such as the type of pain, the type of TENS used, duration of treatment, and whether the study measures acute or chronic outcomes, will be addressed. Recent literature suggests that TENS may alleviate acute pain. For example, one double-blind randomized, sham-controlled trial found that during emergency transport of 101 patients, TENS reduced post-traumatic hip pain with a change in visual analog scale (VAS) from 89 to 59, whereas the sham-stimulated group remained relatively unchanged (86 to 79). Confirmation of these results is needed.

 

A search of the MEDLINE database from January 2007 to January 2008 identified several recent studies and 2 new meta-analyses on the use of TENS.

 

A randomized, controlled trial from Taiwan assessed the efficacy of TENS applied at bilateral acupuncture points (Li4: midpoint between the first and second carpal bones, and Sp 6: 5 cm above the medial malleolus) in 100 women in the first stage of labor (5% dropout due to precipitous labor) (Chao, 2007).   Additional medication was allowed, but was not reported. At the end of 30 minutes of treatment, the median VAS score improved in the active TENS group from 8 to 4.5 (10 mm scale), the VAS of the placebo group changed from 8 to 7. Pain relief of greater than 3 points was obtained in 62% of patients in the active group and 14% of patients in the sham group. The median duration of relief from pain following TENS acupuncture was 75 minutes.

 

An industry-sponsored meta-analysis by Johnson and Martinson included 38 randomized controlled comparisons (1,227 patients from 29 publications) of trans- or percutaneous electrical nerve stimulation (ENS) for chronic musculoskeletal pain, using any stimulation parameters on any location (e.g., back, neck, hip, knee) (Johnson, 2007).  The data were converted to a percentage improvement in VAS scores, then transformed into standardized mean differences (a continuous measure that adjusts for variability in different outcome measures). Based on the combined standardized difference, the authors concluded that TENS provided pain relief “nearly three times” the pain relief provided by placebo. There are a number of sources of bias in the analysis that seriously limit interpretation of the results. First, the heterogeneity of the individual study results (I2=82%) raises questions about the appropriateness of combining these studies in a meta-analysis (see previous discussion regarding the decision to not combine studies for the 2000 Cochrane review on chronic pain). Further limiting interpretation is the transformation of data to standardized effect size, which appears to have led to discrepant effect sizes of otherwise similar results. For example, comparison of the untransformed and transformed data shows that while two of the included trials (Deyo et al. 1990, and Machin et al. 1988), found similar percentage point differences in VAS between active and control groups (5% and 8%, respectively), the standardized effect sizes are not equivalent. Positive standardized effect sizes from data that are not statistically or clinically significant (e.g., 47% vs. 42% change from baseline in Deyo et al.) also raises concerns about the appropriateness of the data transformation. Inclusion of poor-quality studies is an additional concern, since several of the studies with the greatest effect sizes reported drop-out rates exceeding 25%. Furthermore, bias for publication of small positive studies may not have been adequately addressed, since the “Fail-safe N” method used to assess publication bias is problematic.

 

Another major limitation in interpretation of this meta-analysis is the absence of information about whether ENS results in a clinically meaningful improvement. For example, there was no discussion of the magnitude of the combined change in VAS scores or of the proportion of patients who achieved clinically meaningful improvements. Examination of the data indicates that there was less than a 15% difference between the ENS and placebo groups (with an average difference of 4%) for 13 of the 38 (34%) comparisons. The small effect observed in many of these small studies raises further questions about the contribution of publication bias to the meta-analysis. Also at issue is the relative contribution of percutaneous ENS (PENS), since meta-regression found PENS to be more effective than TENS. Given the substantial uncertainty regarding the appropriateness of the studies included and how the data were transformed, combined with questions regarding the clinical significance of the results, results from this meta-analysis are considered inconclusive.

 

Data regarding clinical significance were provided in a meta-analysis by Bjordal et al. on the short-term efficacy of physical interventions for osteoarthritic knee pain (Bjordal, 2007).  Included in the review were 11 studies (259 subjects on active therapy) using TENS, acupuncture-like TENS (AL-TENS), or interferential stimulation; 9 of the 11 studies were included in the meta-analysis reviewed above. Combined data revealed a 19 mm improvement in VAS over placebo (a “slight improvement”), with a confidence interval ranging from 10 mm (a “minimal perceptible improvement”) to 28 mm (above the 20 mm threshold of an “important improvement”). These results are similar to an earlier Cochrane review (overlap of 6 studies) on the use of TENS or AL-TENS for osteoarthritis of the knee (Osiri, 2002).  The inclusion of 2 studies on interferential stimulation (with an unweighted average improvement in VAS of 34 mm over placebo) may also have increased the magnitude of the effect. Considering that the potential for publication bias is high when combining a number of small studies in a meta-analysis (particularly when the effect is small), evidence of short-term relief of chronic musculoskeletal pain remains weak. Results from these positive meta-analyses must also be balanced against other systematic reviews of musculoskeletal pain syndromes that found mixed and inconclusive results.

 

A recent randomized sham-controlled trial (n=163) reported that although no differences in VAS pain scores were observed, more patients were satisfied following 10 days (10-12 hours/day) of TENS (58%) than following use of a sham device (43%) (Oosterhof, 2006).  Analysis of the results by type of pain (osteoarthritis-related, neuropathic, or bone/soft tissue/visceral) in a subsequent report showed no difference in patient satisfaction for the group with osteoarthritis and related disorders (39% vs. 31%, n = 31, 26) or in patients with neuropathic pain (63% vs. 48%, n = 16, 25), and greater satisfaction with TENS in the group of patients with injury of bone and soft tissue or visceral pain (74% vs. 48%, n = 34, 31).  The nearly 50% patient satisfaction rating in the sham control group suggests a strong nonspecific effect with this treatment protocol.

 

Overall, evidence for the use of TENS remains inconclusive. Due to the inconsistency and uncertainty in the TENS literature, the policy is unchanged.

 

 

According to the Agency for Healthcare Research and Quality (AHRQ) Guidelines Clearinghouse, the Department of Veterans Affairs published a guideline on management of low back pain or sciatica in the primary care setting. The report stated that “Evidence is insufficient to recommend transcutaneous electrical nerve stimulation (TENS) in the treatment of patients with acute low back pain.”

 

The American Geriatrics Society produced a guideline in 1998, stating that “… transcutaneous nerve stimulation may be helpful for some patients, but they are expensive and have not been shown to have greater benefit than placebo controls in the management of chronic pain.” (Ferrell, 2009)

 

The American Medical Directors Association created a guideline in 1999 on management of pain for elderly patients in the long-term care setting. Among complementary therapies, transcutaneous electrical nerve stimulation is one for which “Although no scientific evidence supports the effectiveness of these therapies in elderly patients in the long-term care setting, they may be beneficial to some individuals.”

 

The Department of Defense, Veterans Health Administration, published clinical guidelines for the management of postoperative pain in May 2002. These guidelines indicate that TENS may be useful for postoperative pain relief for a variety of procedures and sites. Except for postoperative abdominal pain and pain from cholecystectomy, all of the recommendations are consensus based. For postoperative abdominal pain and pain from cholecystectomy, the recommendations are based on at least 1 RCT and general agreement that TENS is acceptable.

 

The AHRQ Guidelines Clearinghouse also lists several other guidelines that indicate TENS may be used for management of pain. However, none of these guidelines lists TENS as a major recommendation.

 

The American Pain Society and American College of Physicians published guidelines on therapies for acute and low back pain in 2007 (Chou, 2007). No recommendations for TENS were made; evidence was found to be insufficient to assess the efficacy of TENS in comparison with other interventions.

 

The European Federation of Neurological Societies published 2007 guidelines on neurostimulation for neuropathic pain Cruccu, 2007).  The task force was not able to arrive at conclusive recommendations, with only about 200 patients with different diseases, in studies using different parameters and comparators, and with variable results. The task force concluded that standard high-frequency TENS is possibly (level C) better than placebo, and probably (level B) worse than acupuncture-like or any other kind of electrical stimulation.

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

The authors present both intention-to-treat (ITT) and per-protocol analyses, but only the ITT will be discussed. The reduction in the number of migraine days (run-in compared with 3 month) was 2.06 (95% CI, -0.54 to -3.58) for the TENS group versus 0.32 (-0.63 to +1.27) for the sham group; this difference did not quite reach statistical significance (p=0.054). The proportion of responders (≥50% reduction in the number of migraine days/month) was 38% (95% CI, 22% to 55%) in the TENS groups versus 12% (95% CI, 1.0% to 23%) in the sham group (p=0.014). The number of migraine attacks from the run-in period to 3rd month was significantly lower for the active TENS group (decrease of 0.82 in the TENS groups versus 0.15 in the sham group, p=0.044). Number of headache days also was decreased in the TENS group compared with sham (decrease of 2.51 versus 0.15, p=0.041). Patients in the active TENS group reported a 36.6% reduced number of acute antimigraine drugs taken compared with the 0.5% reduction in the sham group (p=0.008).Severity of migraine days did not significantly differ between groups.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

A 2014 Cochrane review on non-pharmacologic interventions for chronic pain in individuals with spinal cord injury identified 1 RCT on TENS (Boldt, 2014). This study had a high risk of bias, and no conclusion could be drawn on the effectiveness of TENS compared with sham for reducing chronic pain in this population.

 

The largest RCT on TENS after surgery was published by Rakel and colleagues (Rakel, 2014). This double-blind study compared TENS once or twice daily for 6 weeks versus sham TENS versus standard care to reduce pain during rehabilitation in 317 patients who had undergone total knee arthroplasty (TKA). The primary outcome was pain intensity during range of motion (ROM) and during walking, measured by a 21-point numeric rating scale (NRS) on postoperative day 1 and week 6. Secondary outcomes were pain intensity at rest, hyperalgesia, and function. Intent-to-treat analysis showed that patients who used TENS during exercises had less pain when compared with standard care in the near postoperative period, but there was no significant difference in subjective pain when compared with patients who used sham TENS. There was also no significant difference between the active and control groups when tested at 6 weeks. This study, which found no benefit of TENS over placebo/sham, had good methodologic quality and a low risk of bias.

 

A patient-blinded study of TENS after abdominal surgery (n=55) found that application of TENS for 1 hour per day resulted in a significant reduction in pain, particularly at rest, measured both during and immediately after treatment compared with sham TENS (Tokuda, 2014). Pulmonary function (vital capacity, cough peak flow) was also significantly better in the active TENS arm. Another assessor-blinded study of TENS in 74 living kidney donors found a modest reduction in pain at rest and during the measurement of pulmonary function when measured 1 day postoperatively (Galli, 2015).