Transcutaneous electrical nerve stimulation (TENS) for chronic pain - an overview of Cochrane Reviews
William Gibson,Benedict M Wand,Catherine Meads,Mark J Catley,Neil E O'Connell
DOI: https://doi.org/10.1002/14651858.CD011890.pub3
2019-04-03
Abstract:Background: Chronic pain, considered to be pain lasting more than three months, is a common and often difficult to treat condition that can significantly impact upon function and quality of life. Treatment typically includes pharmacological and non-pharmacological approaches. Transcutaneous electrical nerve stimulation (TENS) is an adjunct non-pharmacological treatment commonly recommended by clinicians and often used by people with pain. Objectives: To provide an overview of evidence from Cochrane Reviews of the effectiveness of TENS to reduce pain in adults with chronic pain (excluding headache or migraine).To provide an overview of evidence from Cochrane Reviews of the safety of TENS when used to reduce pain in adults with chronic pain (excluding headache or migraine).To identify possible sources of inconsistency in the approaches taken to evaluating the evidence related to TENS for chronic pain (excluding headache or migraine) in the Cochrane Library with a view to recommending strategies to improve consistency in methodology and reporting.To highlight areas of remaining uncertainty regarding the effectiveness of TENS for chronic pain (excluding headache or migraine) with a view to recommending strategies to reduce any uncertainty. Methods: Search methodsWe searched the Cochrane Database of Systematic Reviews (CDSR), in the Cochrane Library, across all years up to Issue 11 of 12, 2018.Selection of reviewsTwo authors independently screened the results of the electronic search by title and abstract against inclusion/exclusion criteria. We included all Cochrane Reviews of randomised controlled trials (RCTs) assessing the effectiveness of TENS in people with chronic pain. We included reviews if they investigated the following: TENS versus sham; TENS versus usual care or no treatment/waiting list control; TENS plus active intervention versus active intervention alone; comparisons between different types of TENS; or TENS delivered using different stimulation parameters.Data extraction and analysisTwo authors independently extracted relevant data, assessed review quality using the AMSTAR checklist and applied GRADE judgements where required to individual reviews. Our primary outcomes included pain intensity and nature/incidence of adverse effects; our secondary outcomes included disability, health-related quality of life, analgesic medication use and participant global impression of change. Main results: We included nine reviews investigating TENS use in people with defined chronic pain or in people with chronic conditions associated with ongoing pain. One review investigating TENS for phantom or stump-associated pain in people following amputation did not have any included studies. We therefore extracted data from eight reviews which represented 51 TENS-related RCTs representing 2895 TENS-comparison participants entered into the studies.The included reviews followed consistent methods and achieved overall high scores on the AMSTAR checklist. The evidence reported within each review was consistently rated as very low quality. Using review authors' assessment of risk of bias, there were significant methodological limitations in included studies; and for all reviews, sample sizes were consistently small (the majority of studies included fewer than 50 participants per group).Six of the eight reviews presented a narrative synthesis of included studies. Two reviews reported a pooled analysis.Primary and secondary outcomes One review reported a beneficial effect of TENS versus sham therapy at reducing pain intensity on a 0 to 10 scale (MD -1.58, 95% CI -2.08 to -1.09, P < 0.001, I² = 29%, P = 0.22, 5 studies, 207 participants). However the quality of the evidence was very low due to significant methodological limitations and imprecision. A second review investigating pain intensity performed a pooled analysis by combining studies that compared TENS to sham with studies that compared TENS to no intervention (SMD -0.85, 95% CI -1.36 to -0.34, P = 0.001, I² = 83%, P < 0.001). This pooled analysis was judged as offering very low quality evidence due to significant methodological limitations, large between-trial heterogeneity and imprecision. We considered the approach of combining sham and no intervention data to be problematic since we would predict these different comparisons may be estimating different true effects. All remaining reviews also reported pain intensity as an outcome measure; however the data were presented in narrative review form only.Due to methodological limitation and lack of useable data, we were unable to offer any meaningful report on the remaining primary outcome regarding nature/incidence of adverse effects, nor for the remaining secondary outcomes: disability, health-related quality of life, analgesic medication use and participant global impression of change for any comparisons.We found the included reviews had a number of inconsistencies when evaluating the evidence from TENS studies. Approaches to assessing risk of bias around the participant, personnel and outcome-assessor blinding were perhaps the most obvious area of difference across included reviews. We also found wide variability in terms of primary and secondary outcome measures, and inclusion/exclusion criteria for studies varied with respect to including studies which assessed immediate effects of single interventions. Authors' conclusions: We found the methodological quality of the reviews was good, but quality of the evidence within them was very low. We were therefore unable to conclude with any confidence that, in people with chronic pain, TENS is harmful, or beneficial for pain control, disability, health-related quality of life, use of pain relieving medicines, or global impression of change. We make recommendations with respect to future TENS study designs which may meaningfully reduce the uncertainty relating to the effectiveness of this treatment in people with chronic pain.