Electrical stimulation of referred sensation area alleviates phantom limb pain

Abstract

Background:

Patients with brachial plexus avulsion (BPA) usually experience phantom sensations and phantom limb pain (PLP) in the deafferented limb. It has been suggested that evoking the sensation of touch in the deafferented limb by stimulating referred sensation areas (RSAs) on the cheek or shoulder might alleviate PLP. However, feasible rehabilitation techniques using this approach have not been reported.

Objective:

The present study sought to examine the analgesic effects of simple electrical stimulation of RSAs in BPA patients with PLP.

Methods:

Study 1: Electrical stimulation of RSAs for 60 minutes was conducted for six BPA patients suffering from PLP to examine short-term analgesic effects. Study 2: A single case design experiment was conducted with two BPA patients to investigate whether electrical stimulation of RSAs was more effective for alleviating PLP than control electrical stimulation (electrical stimulation of sites on side opposite to the RSAs), and to elucidate the long-term effects of electrical stimulation of RSAs.

Results:

Study 1: Electrical stimulation of RSAs evoked phantom touch sensations in the deafferented limb, and significantly alleviated PLP (p < 0.05). Study 2: PLP was alleviated more after electrical stimulation on RSAs compared with control electrical stimulation (p < 0.05). However, the analgesic effects of electrical stimulation on RSAs were observed only in the short term, not in the long term (p > 0.05).

Conclusions:

Electrical stimulation of RSAs not only evoked phantom touch sensation but also alleviated PLP in the short term. The results indicate that electrical stimulation of RSAs may provide a useful practical rehabilitation technique for PLP. Future studies will be required to clarify the mechanisms underlying immediate PLP alleviation via electrical stimulation of RSAs.

Keywords

Phantom limb pain electrical stimulation referred sensation brachial plexus avulsion

1 Introduction

Brachial plexus avulsion (BPA) is a type of peripheral nerve injury occurring after complete or partial avulsion of the lower cervical spinal roots (C5-Th1) of an upper limb. BPA leads to complete or partial loss of sensorimotor function in an upper limb (Shankar, Hansen, & Thomas, 2015). Although the upper limb is anatomically present, BPA patients usually experience phantom sensations and phantom limb pain (PLP) in their deafferented limb, similar to patients with an amputated limb (Flor, 2002). PLP in BPA patients is thought to arise from neuroplastic changes in the primary somatosensory and motor cortices (Dimou, Biggs, Tonkin, Hickie, & Lagopoulos, 2013; Giraux & Sirigu, 2003; Mercier & Sirigu, 2009; Qiu et al., 2014). Mirror visual feedback (MVF) is one of the most widely used rehabilitation techniques for alleviating PLP (Ramachandran & Altschuler, 2009). In MVF rehabilitation, patients move their intact limb while observing its reflection in a mirror, evoking an illusion of seeing movement of the (contralateral) deafferented limb (Ramachandran & Altschuler, 2009). Repeated exposure to MVF rehabilitation has been found to successfully restore kinesthesia of the deafferented limb, and consequently reduce painful cramping sensations in some BPA patients (Sumitani et al., 2008). Recent technological advancements have enabled MVF rehabilitation to be successfully conducted in virtual space, inducing more vivid kinesthesia, and consequently alleviating PLP (Ichinose et al., 2017). However, previous studies have suggested that rehabilitation for enhancing kinesthesia of the deafferented limb is not effective for all patients, particularly when PLP has no kinesthesia-related pain qualities, such as cramping (Giummarra, 2016; Osumi et al., 2019; Sumitani et al., 2008). We speculate that this may be because such rehabilitation techniques do not include sufficient procedures for enhancing phantom touch sensations in the deafferented limb. A previous study reported that phantom touch sensation alleviated PLP that was resistant to traditional MVF rehabilitation (Schmalzl, Ragno, & Ehrsson, 2013). In that study (Schmalzl et al., 2013), the experimenter simultaneously stroked a PLP patient’s stump and intact hand, while the patient watched a reflection of their intact hand in the mirror; this procedure was found to evoke phantom touch sensations. Additionally, methods using phantom touch procedures combined with MVF rehabilitation have been developed, and have been reported to evoke both touch sensation and kinesthesia in the deafferented limb, and alleviate PLP in BPA patients (Ichinose et al., 2017; Osumi et al., 2020). Although these rehabilitation techniques are promising, this approach might have limited acceptance because of the complexity of the procedure. Therefore, there is a need to develop more practical rehabilitation techniques that can evoke other types of phantom sensation and alleviate PLP.

In the current study, we focused on referred sensation for evoking phantom touch sensation in the deafferented limb, in accordance with a previous study (Schmalzl et al., 2013). Patients with PLP following BPA sometimes perceive phantom touch sensations in their deafferented hand when they are touched on the ipsilateral cheek or shoulder(Pourrier et al., 2010; Tsao, Finn, & Miller, 2016). This phantom sensation is known as a referred sensation, and specific sites on the body, including the ipsilateral cheek or shoulder, are known as referred sensation areas (RSAs), providing an important avenue for evoking referred sensation. Interestingly, a previous study of a BPA patient reported that stroking the RSAs evoked a phantom sensation of something moving along the deafferented limb, which the patient reported to be a pleasant sensation (Pazzaglia, Leemhuis, Giannini, & Haggard, 2019). Furthermore, transcutaneous electrical stimulation to RSAs in two amputees was reported to evoke referred sensations in the deafferented limb, and was found to alleviate PLP (Eugen Lontis, Yoshida, & Jensen, 2018; Lontis, Yoshida, & Jensen, 2019). Several lines of evidence suggest that stimulation of RSAs with transcutaneous electrical stimulation is sufficient for evoking phantom touch sensation in the deafferented limb, and alleviating PLP. However, in-depth case studies have not been reported. Considering transcutaneous electrical stimulation itself has been reported to reduce pain (Giuffrida, Simpson, & Halligan, 2010; Mulvey et al., 2013; Tilak et al., 2016), the effectiveness of phantom touch sensation induced by transcutaneous electrical stimulation is currently not clear. The present case study aimed to investigate whether inducing phantom touch sensation in the deafferented limb via transcutaneous electrical stimulation of RSAs alleviates PLP in BPA patients. In study 1, we conducted a preliminary investigation of the short-term analgesic effects of transcutaneous electrical stimulation to RSAs with BPA patients who were suffering from PLP and had experienced limited effectiveness of rehabilitation methods for restoring kinesthesia (e.g., MVF in virtual space) (Osumi et al., 2019). In study 2, we investigated long-term analgesic effects of electrical stimulation of RSAs with a single case experimental design. In two studies, we sought to elucidate the alleviating effects of electrical stimulation of RSAs in both the short and long term.

1.1 Study 1: The short-term analgesic effects of transcutaneous electrical stimulation on RSAs

We aimed to test a feasible rehabilitation technique for PLP patients who exhibited limited responses to rehabilitation for restoring kinesthesia in the deafferented limb. We hypothesized that inducing referred sensation using electrical stimulation of RSAs would cause patients to feel touch sensation in the deafferented limb, alleviating PLP.

2 Methods

2.1 Cases

Six patients with BPA participated in the present study. All patients perceived PLP (Table 1). The selection criteria were as follows: (1) presence of phantom upper-limb pain in the deafferented area; (2) limited response to rehabilitation for phantom limb kinesthesia using mirror therapy and virtual reality; (3) the main pain qualities were not kinesthesia-related pain, such as cramping pain. Even if patients reported having cramping pain, they were included in the BPA group if cramping pain was not their main pain quality. We explained the purpose and protocol of the present study to all patients, and obtained informed consent. The study protocol was in accordance with the Declaration of Helsinki. This study was approved by the ethics committee of Kio University Health Science Graduate School (approval number: R1-08).

Table 1
Clinical features of the participants

No Sex Age Disease duration Affected side Type of BPA Intercostal nerve transfer Referred sensation area (RSA) Consistency referred sensation and PLP Main pain quality

Case 1 Female 44 26 years Right Complete Done Shoulder Matched Aching, electric-shock

Case 2 Male 40 13 years Right Complete Done Neck Matched Stabbing, sharp

Case 3 Male 48 5 years Right Complete Done Shoulder Unmatched Aching, electric-shock

Case 4 Male 48 25 years Left Complete Done Shoulder Matched Heavy, tender

Case 5 Male 46 29 years Left Incomplete Not yet Cheek Matched Aching, burning

Case 6 Male 42 4 years Right Incomplete Not yet Shoulder Matched Throbbing

No	Sex	Age	Disease duration	Affected side	Type of BPA	Intercostal nerve transfer	Referred sensation area (RSA)	Consistency referred sensation and PLP	Main pain quality
Case 1	Female	44	26 years	Right	Complete	Done	Shoulder	Matched	Aching, electric-shock
Case 2	Male	40	13 years	Right	Complete	Done	Neck	Matched	Stabbing, sharp
Case 3	Male	48	5 years	Right	Complete	Done	Shoulder	Unmatched	Aching, electric-shock
Case 4	Male	48	25 years	Left	Complete	Done	Shoulder	Matched	Heavy, tender
Case 5	Male	46	29 years	Left	Incomplete	Not yet	Cheek	Matched	Aching, burning
Case 6	Male	42	4 years	Right	Incomplete	Not yet	Shoulder	Matched	Throbbing

2.2 Assessment of phantom limb pain

Patients’ PLP was assessed using the Short Form McGill Pain Questionnaire version-2 (SF-MPQ-2) immediately before and immediately after intervention. The SF-MPQ consists of 18 sensory items and four affective items rated on a scale of 0 (none) to 10 (worst possible) and provides valuable information regarding individual pain characteristics (Dworkin et al., 2009). Specifically, we were interested in the sensory qualities of pain; the SF-MPQ measures a total of 22 pain quality items: 1. throbbing, 2. shooting, 3. stabbing, 4. sharp, 5. cramping, 6. gnawing, 7. hot-burning, 8. aching, 9. heavy, 10. tender, 11. splitting, 12. tiring–exhausting, 13. sickening, 14. fearful, 15. punishing–cruel, 16. electric-shock, 17. cold-freezing, 18. piercing, 19. pain caused by light touch, 20. itching, 21. tingling or pins and needles, 22. numbness.

2.3 Detecting referred sensation area

The experimenter applied tapping stimulation with their index finger around the patient’s face, neck and shoulder. During stimulation, participants were asked to verbally report if they felt a touch sensation (i.e., referred sensation) in their deafferented hand. We sought to identify the RSA that matched each patient’s PLP area. In cases where more than one RSA matched with the PLP area, the RSA with the most vivid referred sensation was used for intervention. If we could not identify an RSA that matched the PLP area, the RSA that corresponded most closely to the PLP area was used for intervention. Regarding the relationship between the referred sensation induced by touch and electrical stimulation, high frequency electrical stimulation can activate large-diameter Aβ afferents, which are important fibers for tactile sensation. This neurophysiological compatibility has been reported to successfully identify common RSAs between identified tactile (mechanical) stimuli and electrical stimuli (Pazzaglia et al., 2019). In the present study, when electrical stimulation was applied to the RSAs identified by tactile (mechanical) stimuli, patients reported feeling referred sensation in the same area.

2.4 Transcutaneous electrical stimulation on referred sensation area

To conduct electrical stimulation on RSAs, commercially available standard 5×5 cm self-adhesive electrodes (Axelgaard Manufacturing Co., Ltd., USA) were attached across the RSAs. The electrodes may have been larger than necessary to stimulate RSAs, but this did not cause any problems. A previous study used electrodes of a similar size to enhance referred sensation in amputees (Eugen Lontis et al., 2018; Lontis et al., 2019). We placed two electrodes across a single RSA to maximize the referred sensation in accord with a previous study (Eugen Lontis et al., 2018; Lontis et al., 2019). We found no problem with RSAs being too small to cover the area of referred sensation.

Transcutaneous electrical stimulation was applied for 60 minutes using a TENS stimulator (Ito Co., Ltd., Japan). This stimulation time was determined on the basis of a previous study of PLP patients (Giuffrida et al., 2010; Mulvey et al., 2013). The present experiments involved a 60 min stimulation period without any break, and none of the cases opted to take a break when it was offered. Stimulation settings were continuous pulse pattern, pulse duration of 100μs, pulse frequency of 100 Hz, bipolar pulses. To set the intensity of stimulation, the intensity of the stimulus was gradually increased, and participants were asked to report when the intensity was “strong but comfortable”. The electrical stimulation was then delivered at this intensity. Because previous studies using electrical stimulation for PLP have reported various stimulus settings (Eugen Lontis et al., 2018; Giuffrida et al., 2010; Mulvey et al., 2013), we did not attempt to directly replicate the settings from these previous studies. Rather, we prioritized electrical stimulation that evoked tactile-like sensations in the phantom limb. High frequency (50–100 Hz) and low intensity electrical stimulation has previously been reported to evoke comfortable non-painful tactile-like sensations (Leonard, Goffaux, & Marchand, 2010). We determined appropriate settings as described above. There were no cases in which muscles were contracted during electrical stimulation.

2.5 Statistical analysis

Wilcoxon signed-rank tests were conducted to compare PLP before and immediately after TENS, with total SF-MPQ scores on 22 items in all six participants. SPSS ver. 24 (SPSS, Chicago, IL, USA) was used for statistical processing, and the level of significance was set to 5%.

3 Results

We successfully detected RSAs in areas consistent with PLP areas in five of the six cases. The RSAs were as follows; shoulder: case 1, 2, 4 and 6; neck: case 2: cheek: case 5. Five of the six cases reported feeling an electrical sensation in the PLP area; thus, SF-MPQ-2 total scores were decreased in these five cases (cases 1, 2, 4, 5, 6). However, for Case 3, the evoked sensation in the phantom limb did not reach the PLP area because we were not able to detect RSAs that were consistent with the PLP area. Case 3 reported feeling an evoked sensation by electrical stimulation on the lateral side in the deafferented hand, but suffered from PLP on the medial side in the deafferented hand. The total SF-MPQ score in Case 3 before electrical stimulation was similar to that after electrical stimulation. Wilcoxon signed-rank tests revealed that PLP was significantly alleviated after electrical stimulation, compared with before electrical stimulation (before stimulation: average = 66.33, SD =±40.92, after stimulation: average = 45.0, SD =±48.46; Z = –1.992, p = 0.046).

3.1 Study 2: Single case experimental design to reveal the effect of transcutaneous electrical stimulation to referred sensation area

We aimed to investigate whether electrical stimulation of RSAs was more effective for PLP than control electrical stimulation at sites on the side opposite to the RSAs using a single case experimental design. Additionally, we used a single case experimental design to test whether electrical stimulation of RSAs every day alleviated PLP in the long term.

4 Methods

4.1 Cases

Two participants agreed to participate in the present study (Cases 1, 5 in study 1). Four patients (Cases 2, 3, 4, 6) in study 1 did not participate in study 2 because they were not available to participate in the experiment every day.

4.2 Experimental design and procedure for transcutaneous electrical stimulation

The present study used a B-A-B-A design. The single case experiment design with repeated evaluation and intervention under multiple conditions made it possible to exclude other factors that might exacerbate or improve PLP as much as possible. During phase B, electrical stimulation was applied to RSAs (i.e., right shoulder in Case 1 and left cheek in Case 5; the RS + condition) with the same parameters as in Study 1. However, during phase A, electrical stimulation was applied to a site on the opposite side of the RSAs (i.e., the left shoulder in Case 1 and the right cheek in Case 5; the RS –condition). Both patients felt a phantom sensation on the PLP area during phase B but not phase A. The patients underwent transcutaneous electrical stimulation treatment every day in each phase, and PLP intensity was assessed before and after 60 minutes of treatment using an 11-point numerical rating scale (NRS: 0 = no pain; 10 = worst pain imaginable). Cases were instructed to evaluate their pain comprehensively because assessing each SF-MPQ item every day increased the patient’s load. If participants felt discomfort but not pain, they were instructed to give a rating of zero. An electrical stimulation session was conducted every day. Although we sought to conduct at least 1 week of sessions for each period, the number of experimental sessions was decided on the basis of availability of each patient in the process of the study. Regarding the difference in the number of sessions between phases for Case 5, the patient requested that the duration of phase A (RS–condition) be shortened during the process of phase A, because he did not feel analgesic effects of electrical stimulation delivered to the opposite side of the RSAs.

4.3 Data analysis

The effect size for electrical stimulation on RSAs was calculated using Tau-U analysis, a method for measuring data non-overlap between two conditions that is able to control for baseline trends. Thus, the analysis quantified the amount of PLP in the phase in which electrical stimulation was applied to the RSAs (phase B) compared with that in the phase in which electrical stimulation was applied to the opposite side of the RSAs (phase A) using Tau-U. Tau-U is analogous to Kendall’s Rank Correlation Coefficient and the Mann–Whitney U test (Parker, 2011). Tau-U is the only other non-overlap method that combines non-overlap and trend to control for the effects of “PLP alleviation” trends (whether linear, curvilinear, or mixed) in the baseline. This method is reported to be relatively impervious to the confounding effects of autocorrelation and to reliably detect medium-sized effects in short data sets (Parker, 2011). The analysis was used in a previous rehabilitation study for patients with chronic pain (Wurm, 2017). Tau-U calculations were performed using an online calculator (http://www.singlecaseresearch.org/calculators/tau-u). The online calculator available for Tau-U can be used to check for trends in the baseline, adjust for these trends when necessary, and compute contrasts between individual phase pairs (i.e., A vs B). Tau-U values were interpreted using published guidelines (Vannest, 2015). Accordingly, Tau-U values below 0.2 were considered small, values of 0.2–0.6 were considered medium, values of 0.6–0.8 were considered large, and values greater than 0.8 were considered to be very large. Additionally, baseline (phase A) correction was applied in accordance with published guidelines (Vannest, 2015). We corrected the baseline if Tau-U values for the baseline exceeded 0.20 and the trend occurred in the same direction as the aim of the intervention. The level of significance was set to 5%.

5 Results

5.1 Case 1

Case 1 suffered from electrical and aching PLP. The patient described her subjective experience of electrical stimulation of RSAs as “Electrical sensation mixed with my electrical pain; the pain was then alleviated”. The description meant that electrical stimulation of RSAs evoked sensation in the phantom hand, and the evoked sensation in the phantom limb was mixed with electrical pain in the phantom limb.

5.2 Pain intensity before treatment

Because the Tau-U value for the first and second phase A (RS –condition) before treatment exceeded 0.20, we conducted baseline correction. There were no significant differences in PLP intensity before electrical stimulation between phase B and A (0.18, p = 0.35) (Fig. 2a).

Fig. 1

Phantom limb pain assessed with the Short Form McGill Pain Questionnaire version-2 (SF-MPQ-2) for each patient. Numbers on the X axis and Y axis indicate item numbers of SF-MPQ-2 item numbers, and scores, respectively. The red and blue quadrangle indicates each score on the SF-MPQ-2 item before and after electrical stimulation. The filled red area in the illustration indicates the area of phantom limb pain.

Fig. 2

Phantom limb pain progress for Case 1 and 5 in each experimental session were plotted. RS+ indicates the phase in which electrical stimulation was applied to referred sensation areas, while RS–indicates the phase in which electrical stimulation was applied to the side opposite to the referred sensation area. The black and grey quadrangles indicate each phantom limb pain score before and after electrical stimulation. The red and blue dashed lines indicate mean intensity of phantom limb pain before and after electrical stimulation within each experimental phase. The X-axis and Y-axis indicate the session number and pain intensity (NRS), respectively. Electrical stimulation for 60 minutes per session was performed daily.

5.3 Pain intensity after treatment

Because Tau-U values for the first phase A (RS –condition) after treatment exceeded 0.20, we conducted baseline correction for first phase A. Tau-U analysis revealed that PLP intensity after treatment in phase B (RS + condition) was significantly lower than that after treatment in phase A (RS –condition) (0.49, p = 0.012). The effect size indicated that electrical stimulation of RSAs moderately alleviated PLP.

5.4 Case 5

Case 5 suffered from burning PLP and described his subjective experience of electrical stimulation of RSAs as follows: “The electrical sensation (sensation evoked by electrical stimulation in the phantom limb) changes my burning pain into cold pain. I disliked burning pain more than cold pain; TENS then alleviated my pain”. In addition, the patient reported: “an electrical sensation (sensation evoked by electrical stimulation in the phantom limb) arises when I experience a touch sensation in my deafferented hand; I feel like my hand belongs to me after electrical stimulation”.

5.5 Pain intensity before treatment

Because all Tau-U values at baseline did not exceed 0.20, we did not correct the baseline. We found no significant differences in PLP intensity before electrical stimulation between phase B and A (0.19, p = 0.66) (Fig. 2b).

5.6 Pain intensity after treatment

Because all Tau-U values for baseline did not exceed 0.20, we did correct the baseline. Tau-U analysis revealed that PLP intensity after treatment in phase B (RS + condition) was significantly lower than that after treatment in phase A (RS + condition) (0.66, p = 0.0051). The effect size indicated that electrical stimulation of RSAs largely alleviated PLP.

6 Discussion

We conducted electrical stimulation on RSAs in PLP patients following BPA who exhibited a limited response to rehabilitation for phantom limb kinesthesia using mirror therapy and virtual reality methods. Previous studies reported that rehabilitation for phantom limb movement tended to alleviate cramping and gnawing pain (Osumi et al., 2019; Sumitani et al., 2008). However, the main pain quality in the present cases was not cramping and gnawing pain, and mirror therapy and virtual reality did not alleviate their PLP. The present results revealed that phantom touch sensation (i.e., referred sensation) in the deafferented limb induced by electrical stimulation of RSAs alleviated PLP at the end of the 60 min stimulation period in five of six BPA cases. A subsequent single case design study confirmed that phantom touch sensation in the deafferented limb alleviated PLP at the end of the electrical stimulus intervention period, whereas control electrical stimulation of sites on the opposite side to the RSAs did not alleviate PLP. However, PLP alleviation via phantom touch sensation in the deafferented limb did not last into the next day.

The analgesic effects of transcutaneous electrical stimulation are reported to involve both peripheral and central mechanisms (Vance, Dailey, Rakel, & Sluka, 2014). Transcutaneous electrical stimulation of painful sites has been reported to reduce inflammation and alter excitability of peripheral nociceptors, leading to alleviation of pain (King et al., 2005). However, in the present study, patients did not receive electrical stimulation to the painful site, but to other sites (the cheek and shoulder). The shoulder and neck may be hard-wired to the peripheral nerves, unlike the face. However, only case 5 was stimulated on the face, exhibiting an analgesic effect that was the same as that in the other subjects. Thus, peripheral mechanisms of transcutaneous electrical stimulation are not in accord with the PLP alleviation we observed. Other studies reported that transcutaneous electrical stimulation altered the sensitization of spinal dorsal horn neurons (Flor, Denke, Schaefer, & Grusser, 2001) and the activation of descending pain modulation (DeSantana, Da Silva, De Resende, & Sluka, 2009) in accord with “gate control theory” (Johnson, 2014), causing alleviation of pain. These central mechanisms may underlie analgesia via electrical stimulation to remote sites. A previous study suggested that electrical stimulation of a remote site alleviated PLP via such central mechanisms (Giuffrida et al., 2010). From the perspective of central mechanisms, the analgesic effects of electrical stimulation on RSAs would be expected not to differ from that of contralateral RSAs, because the stimulus site in both conditions is remote from the painful site. However, the results of study 2 demonstrated that electrical stimulation of RSAs alleviated PLP to a greater extent than stimulation of contralateral RSAs, which cannot be explained by conventional theories of the peripheral and central mechanisms underlying the analgesic effects of electrical stimulation. Below, we describe a potential explanation of these analgesic mechanisms, focusing on the alleviation of pain by phantom touch sensations.

In study 2, electrical stimulation of RSAs caused patients to experience touch sensations in their deafferented limb, whereas electrical stimulation of sites contralateral to the RSAs did not. This result indicates that the experience of touch sensation in the deafferented limb is essential for PLP alleviation. The analgesic effect of touching a painful body part is a common everyday experience, and a previous study quantitatively demonstrated that touch sensations can have beneficial pain-reducing effects (Mancini, Nash, Iannetti, & Haggard, 2014). Analgesia via touch sensation is reported to occur at a cortical level (e.g., somatosensory areas) rather than at the spinal level (Inui, Tsuji, & Kakigi, 2006; Nahra & Plaghki, 2003). Considering that referred sensation in BPA patients is thought to arise from overlapping somatotopy in primary somatosensory cortex (Pourrier et al., 2010; Tsao et al., 2016), electrical stimulation of RSAs may affect the projected somatotopy corresponding to the deafferented limb, alleviating PLP. However, the present study did not produce neurophysiological data supporting this possibility. Future studies are needed to clarify the neural mechanisms underlying PLP alleviation with electrical stimulation of RSAs. Most of the patients in the current study reported that electrical stimulation of RSAs felt pleasant, suggesting that the pleasantness of the sensation might be an essential factor for PLP alleviation with electrical stimulation. Scores for the affective items of the SF-MPQ-2 (items 12, 13, 14, 15) were decreased in the present cases. Previous studies revealed that the analgesic effects of pleasant touch sensations depended on neural mechanisms involving affective and cognitive processing (Krahe, Drabek, Paloyelis, & Fotopoulou, 2016; Rolls, 2010). Therefore, pleasant touch sensation in the deafferented limb could activate top-down processes, alleviating PLP. However, it is possible that referred sensation does not lead to PLP alleviation in some cases. It will be important for future studies to elucidate the mechanisms underlying such cases.

The current findings can also be considered from another perspective. In the BPA patients in the current study, electrical stimulation of RSAs induced a subjective touch sensation in the deafferented limb. Such phantom touch sensation has been reported to restore a sense of ownership in the deafferented limb (Pazzaglia et al., 2019). In addition, several studies have reported that a distorted sense of ownership exacerbates pathological pain, and vice versa (Bekrater-Bodmann, Reinhard, Diers, Fuchs, & Flor, 2020; Schmalzl et al., 2013; Schmalzl et al., 2011). For example, PLP in amputees was reported to be higher in cases with a distorted sense of ownership regarding a limb prosthesis (Bekrater-Bodmann et al., 2020). In another line of evidence, synchronized visuo-tactile stimulation of RSAs in an amputee’s stump evoked a sense of ownership regarding a prosthesis, alleviating PLP (Schmalzl et al., 2011). In the current study, Case 5 reported “I felt that my hand belonged to me after electrical stimulation”, indicating that phantom touch sensation in the deafferented limb induced by electrical stimulation of RSAs evoked a sense of ownership of the deafferented limb. Considering these findings together, embodiment of the deafferented limb may also be an essential factor in PLP alleviation via electrical stimulation of RSAs. Additionally, combining a sensory discrimination procedure with electrical stimulation might enhance the sense of ownership, causing greater alleviation of PLP. A previous study reported that discriminating the frequency and location of non-painful electrical stimuli on amputees’ stumps successfully alleviated their PLP (Flor et al., 2001). Another study reported that sensory discrimination did not improve abnormal positioning of the phantom limb, but did alleviate PLP (Wakolbinger, Diers, Hruby, Sturma, & Aszmann, 2018). Considering these findings, future studies should verify the combined effects of sensory discrimination and electrical stimulation on PLP.

7 Conclusion and limitations

We investigated the effects of phantom touch sensation in deafferented limbs using simple electrical stimulation of RSAs. Electrical stimulation of RSAs might project somatotopically to the corresponding deafferented limb, causing the participant to feel phantom touch sensation. Patients reported that this experience felt pleasant, evoking a sense of ownership in the deafferented limb, consequently alleviating PLP. It has been anecdotally reported that electrical stimulation of RSAs enhances the embodiment of prosthetic limbs (Chai, Sui, Li, He, & Lan, 2015) and the dexterity of prosthetic hands (D’Anna et al., 2017). We propose that electrical stimulation of RSAs might provide a viable rehabilitation technique for alleviating PLP. However, the current results revealed that electrical stimulation of RSAs alleviated PLP only in the short term, not in the long term. These findings suggest that a larger number of sessions might be needed to repair maladaptive neural plasticity. Therefore, it may be useful for future studies to apply electrical stimulation to RSAs multiple times a day to maintain PLP alleviation in the long term. In addition, electrical stimulation of RSAs did not alleviate PLP in Case 3, whose RSAs were not consistent with his PLP area. Other rehabilitation techniques for evoking phantom touch sensations in deafferented limbs should be developed for such cases. Further studies will be required to clarify the mechanisms underlying PLP alleviation via electrical stimulation of RSAs.

The current study involved several other limitations that should be considered. First, it would be useful to redesign our procedure for determining RSAs and the setting of electrical stimulation in future studies. Because the present study aimed to minimize patients’ burden and fatigue, we used a simplified procedure to determine RSAs. A previous study used a standardized procedure involving multiple sensory modalities, such as heat and vibration (Knecht et al., 1998). Future studies should implement a standardized procedure. Additionally, the experience of tactile sensation in the phantom limb among amputees was reported to depend on the frequency of electrical stimulation (D’Anna et al., 2017). Thus, future studies should determine the setting of the stimulus for each case individually. Second, we did not assess the SF-MPQ every day because of the burden for patients. However, findings regarding changes in each pain quality would be valuable. Thus, future studies should assess the change of each pain quality every day. Third, because we did not assess SF-MPQ-2 in only one session in study 1, the effects of natural variation in pain were not elucidated. In addition, in study 2, we were not able to completely exclude the possibility that PLP is affected by a range of other factors (e.g., physical activity and temperature). Fourth, we did not conduct electrical stimulation under controlled conditions (i.e., controlling the possibility that electrical stimulation evoked referred sensation or sham stimulation) because of ethical considerations.

Footnotes

Acknowledgments

This study was supported by a grant from TERUMO LIFE SCIENCE FOUNDATION. We thank Benjamin Knight, MSc., from Edanz Group () for editing a draft of this manuscript.

References

Bekrater-Bodmann,

, Reinhard,

, Diers,

, Fuchs,

, & Flor,

(2020). Relationship of prosthesis ownership and phantom limb pain: results of a survey in 2,383 limb amputees. Pain.

Chai,

, Sui,

, Li,

, He,

, & Lan,

(2015). Characterization of evoked tactile sensation in forearm amputees with transcutaneous electrical nerve stimulation. Journal of Neural Engineering, 12(6), 066002.

D’Anna,

, Petrini,

F.M.

, Artoni,

, Popovic,

, Simanic,

, Raspopovic,

, & Micera,

(2017). A somatotopic bidirectional hand prosthesis with transcutaneous electrical nerve stimulation based sensory feedback. Scienticf Reports, 7(1), 10930.

DeSantana,

J.M.

, Da Silva,

L.F.

, De Resende,

M.A.

, & Sluka,

K.A.

(2009). Transcutaneous electrical nerve stimulation at both high and low frequencies activates ventrolateral periaqueductal grey to decrease mechanical hyperalgesia in arthritic rats. Neuroscience, 163(4), 1233–1241.

Dimou,

, Biggs,

, Tonkin,

, Hickie,

I.B.

, & Lagopoulos,

(2013). Motor cortex neuroplasticity following brachial plexus transfer. Frontiers in Human Neuroscience, 7, 500.

Dworkin,

R.H.

, Turk,

D.C.

, Revicki,

D.A.

, Harding,

, Coyne,

K.S.

, Peirce-Sandner,

,... & Melzack,

(2009). Development and initial validation of an expanded and revised version of the Short-form McGill Pain Questionnaire (SF-MPQ-2). Pain, 144(1-2), 35–42.

Eugen Lontis,

, Yoshida,

, & Jensen,

(2018). Features of Referred Sensation Areas for Artificially Generated Sensory Feedback - A Case Study. Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2018, 3533–3536.

Flor,

(2002). Phantom-limb pain: characteristics, causes, and treatment. Lancet Neurology, 1(3), 182–189.

Flor,

, Denke,

, Schaefer,

, & Grusser,

(2001). Effect of sensory discrimination training on cortical reorganisation and phantom limb pain. Lancet, 357(9270), 1763–1764.

10.

Giraux,

, & Sirigu,

(2003). Illusory movements of the paralyzed limb restore motor cortex activity. Neuroimage, 20(Suppl 1), S107–111.

11.

Giuffrida,

, Simpson,

, & Halligan,

P.W.

(2010). Contralateral stimulation, using TENS, of phantom limb pain: two confirmatory cases. Pain Medicine, 11(1), 133–141.

12.

Giummarra,

(2016). Augmented reality for treatment of phantom limb pain-are we there yet? Lancet, 388(10062), 2844–2845.

13.

Ichinose,

, Sano,

, Osumi,

, Sumitani,

, Kumagaya,

S. I.

, & Kuniyoshi,

(2017). Somatosensory Feedback to the Cheek During Virtual Visual Feedback Therapy Enhances Pain Alleviation for Phantom Arms. Neurorehabilitation and Neural Repair, 31(8), 717–725.

14.

Inui,

, Tsuji,

, & Kakigi,

(2006). Temporal analysis of cortical mechanisms for pain relief by tactile stimuli in humans. Cerebral Cortex, 16(3), 355–365.

15.

Johnson,

M.I.

(2014). Transcutaneous electrical nerve stimulation (TENS) Research to support clinical practice. Oxford University press, pp. 6.

16.

King,

E.W.

, Audette,

, Athman,

G.A.

, Nguyen,

H.O.

, Sluka,

K.A.

, & Fairbanks,

C.A.

(2005). Transcutaneous electrical nerve stimulation activates peripherally located alpha-2A adrenergic receptors. Pain, 115(3), 364–373.

17.

Knecht,

, Henningsen,

, Hohling,

, Elbert,

, Flor,

, Pantev,

, & Taub,

(1998). Plasticity of plasticity? Changes in the pattern of perceptual correlates of reorganization after amputation. Brain, 121(Pt 4), 717–724.

18.

Krahe,

, Drabek,

M.M.

, Paloyelis,

, & Fotopoulou,

(2016). Affective touch and attachment style modulate pain: a laser-evoked potentials study. Philosophical Transactions of the Royal Society B Biological Sciences, 371(1708).

19.

Leonard,

, Goffaux,

, & Marchand,

(2010). Deciphering the role of endogenous opioids in high-frequency TENS using low and high doses of naloxone. Pain, 151(1), 215–219.

20.

Lontis,

E.R.

, Yoshida,

, & Jensen,

(2019). Referred Sensation Areas in Transpelvic Amputee. Annu Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2019, 6458–6461.

21.

Mancini,

, Nash,

, Iannetti,

G.D.

, & Haggard,

(2014). Pain relief by touch: a quantitative approach. Pain, 155(3), 635–642.

22.

Mercier,

, & Sirigu,

(2009). Training with virtual visual feedback to alleviate phantom limb pain. Neurorehabilitation and Neural Repair, 23(6), 587–594.

23.

Mulvey,

M.R.

, Radford,

H.E.

, Fawkner,

H.J.

, Hirst,

, Neumann,

, & Johnson,

M.I.

(2013). Transcutaneous electrical nerve stimulation for phantom pain and stump pain in adult amputees. Pain Practice, 13(4), 289–296.

24.

Nahra,

, & Plaghki,

(2003). Modulation of perception and neurophysiological correlates of brief CO2 laser stimuli in humans using concurrent large fiber stimulation. Somatosens Mot Res, 20(2), 139–147.

25.

Osumi,

, Inomata,

, Inoue,

, Otake,

, Morioka,

, & Sumitani,

(2019). Characteristics of Phantom Limb Pain Alleviated with Virtual Reality Rehabilitation. Pain Medicine, 20(5), 1038–1046.

26.

Osumi,

, Sano,

, Ichinose,

, Wake,

, Yozu,

, Kumagaya,

S. I.

,... & Sumitani,

(2020). Direct evidence of EEG coherence in alleviating phantom limb pain by virtual referred sensation: Case report. Neurocase, 26(1), 55–59.

27.

Pazzaglia,

, Leemhuis,

, Giannini,

A.M.

, & Haggard,

(2019). The Homuncular Jigsaw: Investigations of Phantom Limb and Body Awareness Following Brachial Plexus Block or Avulsion. Journal of Clinical Medicine, 8(2).

28.

Parker,

R.I.

, Vannest,

K.J.

, Davis,

J.L.

, & Sauber,

S.B.

(2011). Combining nonoverlap and trend for single-case research: Tau-U. Behavior Therapy, 42(2), 284–99.

29.

Pourrier,

S.D.

, Nieuwstraten,

, Van Cranenburgh,

, Schreuders,

T.A.

, Stam,

H.J.

, & Selles,

R.W.

(2010). Three cases of referred sensation in traumatic nerve injury of the hand: implications for understanding central nervous system reorganization. Journal of Rehabilitation Medicine, 42(4), 357–361.

30.

Qiu,

T.M.

, Chen,

, Mao,

, Wu,

J.S.

, Tang,

W.J.

, Hu,

S.N.

,... & Gu,

Y.D.

(2014). Sensorimotor cortical changes assessed with resting-state fMRI following total brachial plexus root avulsion. Journal of Neurology, Neurosurgery, and Psychiatry, 85(1), 99–105.

31.

Ramachandran,

V.S.

, & Altschuler,

E.L.

(2009). The use of visual feedback, in particular mirror visual feedback, in restoring brain function. Brain, 132(Pt 7), 1693–1710.

32.

Rolls,

E.T.

(2010). The affective and cognitive processing of touch, oral texture, and temperature in the brain. Neuroscience & Biobehavioral Reviews, 34(2), 237–245.

33.

Schmalzl,

, Ragno,

, & Ehrsson,

H.H.

(2013). An alternative to traditional mirror therapy: illusory touch can reduce phantom pain when illusory movement does not. Clinical Journal of Pain, 29(10), e10–18.

34.

Schmalzl,

, Thomke,

, Ragno,

, Nilseryd,

, Stockselius,

, & Ehrsson,

H.H.

(2011). Pulling telescoped phantoms out of the stump”: manipulating the perceived position of phantom limbs using a full-body illusion. Frontiers in Human Neuroscience, 5, 121.

35.

Shankar,

, Hansen,

, & Thomas,

(2015). Phantom pain in a patient with brachial plexus avulsion injury. Pain Medicine, 16(4), 777–781.

36.

Sumitani,

, Miyauchi,

, McCabe,

C.S.

, Shibata,

, Maeda,

, Saitoh,

,... & Mashimo,

(2008). Mirror visual feedback alleviates deafferentation pain, depending on qualitative aspects of the pain: a preliminary report. Rheumatology (Oxford), 47(7), 1038–1043.

37.

Tilak,

, Isaac,

S.A.

, Fletcher,

, Vasanthan,

L.T.

, Subbaiah,

R.S.

, Babu,

,... & Tharion,

(2016). Mirror Therapy and Transcutaneous Electrical Nerve Stimulation for Management of Phantom Limb Pain in Amputees - A Single Blinded Randomized Controlled Trial. Physiotherapy Research International, 21(2), 109–115.

38.

Tsao,

J.W.

, Finn,

S.B.

, & Miller,

M.E.

(2016). Reversal of phantom pain and hand-to-face remapping after brachial plexus avulsion. Annals of Clinical and Translational Neurology, 3(6), 463–464.

39.

Vance,

C.G.

, Dailey,

D.L.

, Rakel,

B.A.

, & Sluka,

K.A.

(2014). Using TENS for pain control: the state of the evidence. Pain Management, 4(3), 197–209.

40.

Vannest,

, & Ninci,

(2015). Evaluating intervention effects in singlecase research designs. Journal of Counseling & Development, 93(4), 403–411.

41.

Wakolbinger,

, Diers,

, Hruby,

L.A.

, Sturma,

, & Aszmann,

O.C.

(2018). Home-Based Tactile Discrimination Training Reduces Phantom Limb Pain. Pain Practice, 18(6), 709–715.

42.

Wurm,

, Klein Strandberg,

, Lorenz,

, Tillfors,

, Buhrman,

, Holländare,

, & Boersma,

(2017). Internet delivered transdiagnostic treatment with telephone support for pain patients with emotional comorbidity: a replicated single case study. Internet Interventions, 26(10), 54–64.