Background
Persistent pain has been associated with disrupted cortical representation of the painful body area within the somatosensory cortex of patients with complex regional pain syndrome [
1,
2] and phantom limb pain [
3]. It has been proposed that this cortical disruption (or reorganisation) may play a role in pain maintenance and thus interventions aimed at normalising this disruption have been developed [
4].
Sensory discrimination training, where the bodily area receives a stimulus and the patient must attempt to correctly identify key aspects of the stimulus (e.g. the precise location of the stimulus) is such an intervention. One of the first studies to use sensory discrimination training for patients with chronic pain was performed by Flor et al. [
3]. In their study five phantom limb pain patients received electrical stimulation to eight adjacent but distinct parts of the residual limb (location) at different frequencies. Patients had to correctly identify the location and the frequency of the stimulation, and received feedback on their predictions from a therapist. Compared to a group of usual care controls, the sensory discrimination group had significantly greater improvements in pain and cortical reorganisation. Furthermore, the amount of improvement in pain, cortical reorganisation and skin sensory discrimination ability were all correlated with one another supporting the hypothesis that sensory discrimination training can improve the cortical representation of painful body areas and improve pain.
More recently, tactile acuity training, a form of sensory discrimination training, has been used to improve cortical representation in patients with chronic pain [
5]. Tactile acuity training is an adaptation of the two point discrimination test, a measure of the minimum distance that can be detected between two points on the skin. The smaller the two point discrimination the better the tactile acuity. In tactile acuity training two probes (a narrow and a thick probe) are used to stimulate adjacent areas on the skin and patients must identify the location and mode (narrow/thick) of the stimulus, whilst receiving guided feedback from a therapist. Tactile acuity training can improve pain and two-point discrimination performance in patients with CRPS compared to tactile stimulation alone [
5].
There is evidence to suggest that patients with CLBP have a distorted cortical representation of the sensorimotor area representing the lower back from studies directly measuring cortical representation using brain imaging [
6], and studies indirectly measuring cortical representation using two-point discrimination [
7,
8], lumbopelvic motor control tests [
9], laterality tests [
10] and graphaesthesia [
11]. Thus, it could be postulated that tactile acuity training may be beneficial for patients with CLBP.
A recent case series of patients with CLBP investigated the clinical effectiveness of a comprehensive sensorimotor retraining programme, a key component of which was tactile acuity training [
12]. Following a 10 week intervention all three participants showed clinically meaningful improvements in pain intensity, pain interference and disability. A key limitation of this work was its case series design, which lacked an adequate control. A randomised controlled trial (RCT) of tactile acuity training for patients with CLBP is needed to build on this work and more robustly investigate the effectiveness of this intervention.
Previous investigations of tactile acuity training have included twice daily delivery of the intervention, once by a clinician in the clinical setting and once by an informal carer, (friend/relative) over a three week period [
5]. This daily patient-therapist contact time would make the intervention impractical for use within the National Health Service (NHS) in the UK. Additionally, daily visits to the hospital setting would also place a significant burden on the patient. Self-management is an important component of the overall management strategy for people with long-term conditions and daily clinical visits may have a negative impact on a patient’s ability to self-manage, fostering feelings of dependency. Thus, in this pragmatic study the majority of the intervention was provided as part of a home training package. To comprehensively capture the experiences and perceptions of tactile acuity training of patients and their informal carer’s qualitative feedback was obtained via semi-structured focus groups. No such data exists within the literature and such information was considered important for the refinement of the intervention for future work.
The aim of this pilot, mixed-methods, RCT was to inform the development of a fully powered RCT by providing preliminary data on the effect of tactile acuity training for patients with CLBP compared to sham tactile acuity training for the primary outcome measures of pain and function. The secondary aim of this study was to obtain qualitative feedback from participants, via semi-structured focus groups, about their experiences and perceptions of the intervention to inform the refinement of the intervention and placebo intervention for a follow-up fully powered RCT.
Discussion
The aim of this pilot, mixed-methods, RCT was to provide preliminary data on the effect of tactile acuity training for patients with CLBP compared to sham tactile acuity training for the primary outcome measures of pain and function. The secondary aim of this study was to obtain qualitative feedback from participants, via semi-structured focus groups, about their experiences and perceptions of the intervention.
The main finding of this study was that individuals in the placebo group had a statistically better outcome for pain than the intervention group, while there was no difference between groups for functional improvement. The main secondary finding was that there were considerable barriers to delivering the home training component associated with the need of having a second person (and informal carer) to deliver the training and the large amount of time the training required.
It is unclear why the placebo group had a better outcome for pain compared to the intervention group. One explanation may be related to relaxation. During the focus groups some participants suggested the training was relaxing whereas others suggested it was
hard work and they felt
drained after it. It is reasonable to assume that the intervention, for which concentration and feedback was of key importance, was mentally challenging and thus “draining” in comparison to the relatively relaxing placebo intervention, which required no concentration. There is evidence to suggest that relaxation can have a positive effect on pain outcomes [
17,
18] and this may, at least in part, explain the improved outcomes in the placebo group.
This is the first study to provide data on the effects of tactile acuity training for CLBP using an RCT design. The findings contrast with a recent case-series of three participants with CLBP which reported clinically important improvements in pain and function following tactile acuity training [
12]. The reason for the contrasting findings may be related to the different methodologies used. In the case-series study, tactile acuity training was delivered as part of a comprehensive sensorimotor training package delivered over a minimum of 10 weeks. In the current study sensory discrimination training (tactile acuity training combined with graphaesthesia training) was used without a motor retraining component. Additionally, the case-series study had a comprehensive educational component and all participants were given the “Explain Pain” book. This may have helped patients to better understand the potential mechanisms of action and increased their expectancy, which may in turn have assisted with the quantity and quality of the training performed by the participants. In the current study the educational component explaining the mechanisms of action was relatively brief and simplistic. Without more detailed explanation patients may have been somewhat sceptical of the effects and thus may have had lower expectations. This is supported to some extent within the focus groups where patients implied a degree of scepticism about the intervention. Alternatively, the case-series design is more open to bias than the RCT design, which may explain the lack of effect in our study.
The findings of this study also contrast with those of Moseley et al. [
5] who reported statistically significant effects of tactile acuity training compared to sham in patients with complex regional pain syndrome (CRPS). Again, the difference in findings may be due to differences in methodologies, primarily the difference in clinical sample. The evidence for cortical disruption in CRPS is compelling [
19,
20] and there is also evidence to show that as the cortical disruption declines so too does pain, though no cause and effect has been established [
1,
2]. The evidence to implicate cortical disruption in CLBP is reasonably strong [
4,
6] though perhaps less so than for CRPS. Furthermore, recent evidence suggests that somatosensory cortical disruption may be more associated with neuropathic type pain [
21] which would explain why interventions targeting this disruption may be more beneficial for CRPS than CLBP.
In previous tactile acuity training studies [
10,
12] the intervention has been delivered on a daily basis by a therapist within a clinical setting. Such a treatment delivery system is not feasible within the current NHS. Thus the current study attempted to tailor the delivery system towards a predominantly home based non-therapist delivered system. Considering the lack of effectiveness shown in this study, adapting tactile acuity training in this manner to fit current clinical practice is not warranted. A recent prospective audit by Johnson et al. [
22] investigated the effect of graded motor imagery (GMI) for CRPS in clinical practice. The GMI used was based on techniques which had proven successful under RCT conditions but had been adapted to fit with usual clinical practice (e.g. patient-therapist contact time was reduced). The audit found no effect of GMI on pain and highlighted the challenge of translating results of a complex pain intervention requiring a high level of patient compliance from the RCT setting into clinical practice. The results of the current study mirror these findings, in that tactile acuity training was adapted to fit within usual clinical practice delivery systems and this did not translate into an effective intervention. Studies such as ours and that of Johnson et al. [
22] highlight the need for pragmatic trials to assess the clinical effectiveness of interventions to facilitate appropriate translation of research into practice. Furthermore, they emphasise the importance of developing interventions that can be easily applied in a typical healthcare setting.
Study limitations
Perhaps the most significant limitation of this work was that no measure of tactile acuity or cortical disruption was taken post treatment. Thus we cannot be confident that tactile acuity training resulted in tactile acuity improvements, which would indirectly indicate an improvement in somatosensory cortical representation, the mechanism by which this intervention purports to work. It is worth mentioning that post treatment 2 point discrimination scores were collected for two participants both of whom were in the intervention group. For these participants two-point discrimination reduced from 51 mm to 50 mm and from 63 mm to 61 mm. The magnitude of change is minimal and questions the effectiveness of this intervention to bring about changes in tactile acuity and thus cortical representation. Only four participants in total (two from each group) were accompanied by their informal carer to observe the physiotherapist delivering the intervention. When this is considered alongside the qualitative feedback from participants that the training manual and DVD were difficult to follow it further highlights the significant barrier that needing a second person to deliver the home intervention creates. While our quantitative measure of credibility suggested that participants, in general, found the home and hospital based training to be similar, and similarly credible (see Table
2), this was limited in scope and possibly lacking precision. These issues would suggest that a future study should closely examine fidelity in more detail.
Another limitation of this study was its small sample size, which reduces the generalisability of the results and may have led to the study being underpowered, increasing the risk of a type II error. (Given the magnitude and direction of effect this is unlikely to be biased against showing a significant benefit of the intervention.) Of the 46 individuals who initially showed interest for this study and were screened 18 (~40%) declined to participate through lack of time or lack of an informal carer, further illustrating the practical challenge presented by the need for an informal carer to deliver the home training. Finally, an intention-to-treat analysis was not performed and only participants who provided data pre and post intervention were included in the analysis.
A key strength of this study was its mixed methods approach, which has identified a number of relevant barriers to the delivery of this intervention mostly around the need for a second person to provide the intervention. Of particular interest were the potential psychosocial barriers of guilt and embarrassment felt by some of the participants because of the need to involve an informal carer. These findings have potential implications not just for this intervention but within wider health care in areas where informal carers are required to deliver medical type interventions e.g. partners delivering routine subcutaneous injections in the home. The literature in this area is scant [
23] and further investigation is warranted.
Clinical implications and future studies
There were a number of practical clinical implications identified in this study, primarily around the barrier presented by needing a second person to deliver the home training. This barrier suggests that this intervention, in its current form, may not be practical within the UK’s NHS and some automated version of the intervention may be warranted. Barker et al. [
24] investigated the effects of an automated sensory discrimination training device for patients with CLBP using the FairMed device. The device used vibrational stimulation rather than simple touch as used in tactile acuity training. The device was found to be similarly effective to TENS treatment although a high number of participants reported some kind of fault with the device (20/32), which may have negatively impacted on the treatment outcome when using the device. Automated devices such as the FairMed may be one approach by which sensory discrimination training techniques such as tactile acuity training may become more practical within the constraints of the NHS. Without dismissing the potential benefit of tactile acuity training, there is a need to investigate this intervention type using a more appropriate delivery system, as it may have been the latter which was responsible for the lack of effect of the active intervention in this study.
Conclusions
In this pilot RCT there was no statistically significant difference between tactile acuity training and sham tactile acuity training for three of the four main outcome measures (pain and disability, both in absolute terms and as percentage change from baseline). Furthermore, the one significant change between the groups was in favour of the placebo group. These preliminary findings suggest that tactile acuity training as delivered in this study was not an effective adjunct treatment to usual care physiotherapy. However, considering the small pilot nature of this study, the questions around the fidelity of the home training component and the absence of a measure of tactile acuity post treatment, no firm clinical recommendations can be made. Participants reported a number of barriers to the home delivery of this intervention with the need for a second person being the primary issue. Future work should consider the need for an automated device to make home delivery more feasible, and the clinical effectiveness of such a device should be investigated.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
CR conceived of the study, participated in its design and co-ordination, assisted with data collection, carried out the statistical/thematic analysis, contributed to the interpretation of the data and helped to draft the manuscript. BD was the primary collector of data, contributed to its interpretation, and helped to draft the manuscript. NH participated in the design and co-ordination of the study, assisted with data collection, contributed to the interpretation of the data and helped to draft the manuscript. DM participated in the design and co-ordination of the study, contributed to the interpretation of the data and helped to draft the manuscript. All authors read and approved the final manuscript.