Skip navigation

Does TENS work?

Having one systematic review on a topic is great. Having two is terrific, but four would seem to be an embarrassment of riches. Or at least it would be an embarrassment if, having come at a topic from different perspectives, they reached conflicting conclusions. For transcutaneous electrical nerve stimulation (TENS) for treating pain, by and large they do not.

Is TENS much used?

According to the Canadian technology assessment paper [1] in which the authors surveyed TENS use across Canada, the answer would seem to be yes, it is widely used. After surveying 50 hospitals with 200 or more beds they estimated over 450,000 uses of TENS take place in Canadian hospitals each year with widespread use in physiotherapy and for acute pain (used by 93% of hospitals), labour and delivery (43%), and for chronic pain (96%).

TENS in acute pain

The Canadian study [1] included randomised and non-randomised studies, though it split them for descriptive analysis. They reviewed 39 studies in postoperative pain, dental pain, dysmenorrhoea and cervical pain. They described results as varied, but you have to read awfully closely to find much good about TENS. The review suggested that the authors of the original papers which were randomised reached positive conclusions about TENS in 19/34 papers, but a negative or balanced conclusion in 15/34.

In a review limited to randomised studies in acute postoperative pain [2], TENS was judged by the reviewers to be no better than placebo in 15 out of 17 randomised studies. Of 19 trials with pain outcomes which were not randomised, 17 of 19 the authors of the original papers had concluded that TENS had a beneficial effect. This is another good example of bias in non-randomised studies (see Bandolier 17 ).

TENS in labour pain

The Canadian review [1] summarised 6/9 randomised trials as reaching negative conclusions. This is a similar result to the second review [3], which examined eight reports, of which five were judged to have a negative result with TENS no better than placebo or sham-TENS. However, the three studies which were judged to be positive were positive only on weak outcomes like additional pain relieving measures and increased time to epidural local anaesthetic.

Additional analgesic interventions were significantly less likely with the use of TENS [3], with a number-needed-to-treat of 14 (95%CI 7 - 119) for one woman in labour to be spared an epidural or intramuscular injection. Of the four trials which reported this outcome, only the two smaller trials (23 women receiving TENS) were statistically significant, while the two larger studies (208 women receiving TENS) were not significantly different from placebo.

A randomised trial of 94 women in the first stage of labour published since the reviews were done [4] undermines even this possible level of benefit. It found no difference in analgesic requirement between active TENS and disabled TENS equipment, nor any difference on pain scores.

TENS in chronic pain

The Canadian review found 20 randomised trials, of which nine were definitely positive for TENS on some measure, but eight were negative for TENS.

A disappointing review limited to chronic low back pain [5] included just six papers. Two of the original studies were of electroacupuncture, which is not the same as TENS. One trial had only ten patients randomised, six patients to TENS and four to placebo. TENS was not significantly different from placebo. Electroacupuncture was significantly better than placebo, but in only two studies with 30 patients given electroacupuncture.

What is one to make of all this?

First of all there are methodological considerations. Adequate blinding of TENS is extremely difficult, so that most trials should best be regarded as open even if described a blinded. This must confer some degree of bias towards TENS.

Second, there was a tendency in all these reviews to point to an overall lack of methodological rigour in the original studies (but acknowledging that these trials are difficult). The trials with the best methods tended to produce negative results.

Then there is the issue of statistical validity. Put simply, many of these trials make a number of different measurements, only some of which show statistical benefit. So choosing just those measurements which are significant, and ignoring those that are not significant, can give a spurious weight to a review. This is especially true when statistical benefit is of dubious clinical value. Reviewers and readers should always make up their own minds, not just take a headline result chosen because of statistics.

So our reading of these reviews should be sceptical, especially when, as in acute postoperative pain, there are adequate alternatives. For labour pain there may just be an argument for good quality trials which examine the issue of delay or avoidance of interventions like epidurals or intramuscular opiates which carry some risk to mother or baby.

Babies and bathwater

Chronic pain is a different matter. Where the evidence is not clear cut, where some patients are seen to benefit, and where alternatives may not work for all patients, then carrying on using TENS until there is some clarification makes sense. That does put some heat on getting well-designed studies of sufficient power to provide practically useful answers underway.

In the meantime, those of you who see full-page adverts in the national newspapers full of happy souls extolling the virtues of TENS might like to refer the Advertising Standards Authority to these reviews.


  1. J Reeve, D Menon, P Corabian. Transcutaneous electrical nerve stimulation (TENS): a technology assessment. International Journal of Technology Assessment 1996 12:2: 299-324.
  2. D Carroll, M Tramèr, H McQuay, B Nye, A Moore. Randomization is important in studies with pain outcomes: systematic review of transcutaneous electrical nerve stimulation in acute postoperative pain. British Journal of Anaesthesia 1996 77: 798-803.
  3. D Carroll, M Tramèr, H McQuay, B Nye, A Moore. Transcutaneous electrical nerve stimulation in labour pain: a systematic review. British Journal of Obstetrics & Gynaecology 1997 104: 169-175.
  4. JM van der Ploeg, HAM Vervest, AL Liem, JH Schagen van Leeuwen. Transcutaneous nerve stimulation (TENS) during the first stage of labour: a randomized clinical trial. Pain 1996 68: 75-8.
  5. JG Gadsby, MW Flowerdew. The effectiveness of transcutaneous electrical nerve stimulation (TENS) and acupuncture-like transcutaneous electrical nerve stimulation (ALTENS) in the treatment of patients with chronic low back pain. Cochrane Library 1997 issue 1.

previous or next story in this issue