If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address reprint requests to Ronald G. Garcia, MD, PhD, Martinos Center for Biomedical Imaging, Building 149, 13th Street, Suite 2301, Charlestown, MA 02129.
Neurovascular Science Group, Fundación Cardiovascular de Colombia, Floridablanca, ColombiaMASIRA Research Institute, School of Medicine, Universidad de Santander, Bucaramanga, ColombiaMartinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
Transcranial magnetic stimulation had a significant analgesic effect in phantom limb pain (PLP) patients.
•
Transcranial magnetic stimulation was well tolerated in land mine victims with PLP.
•
Noninvasive brain stimulation might be a promising option for PLP rehabilitation.
Abstract
We evaluated the effects of repetitive transcranial magnetic stimulation (rTMS) in the treatment of phantom limb pain (PLP) in land mine victims. Fifty-four patients with PLP were enrolled in a randomized, double-blinded, placebo-controlled, parallel group single-center trial. The intervention consisted of real or sham rTMS of M1 contralateral to the amputated leg. rTMS was given in series of 20 trains of 6-second duration (54-second intertrain, intensity 90% of motor threshold) at a stimulation rate of 10 Hz (1,200 pulses), 20 minutes per day, during 10 days. For the control group, a sham coil was used. The administration of active rTMS induced a significantly greater reduction in pain intensity (visual analogue scale scores) 15 days after treatment compared with sham stimulation (−53.38 ± 53.12% vs −22.93 ± 57.16%; mean between-group difference = 30.44%, 95% confidence interval, .30–60.58; P = .03). This effect was not significant 30 days after treatment. In addition, 19 subjects (70.3%) attained a clinically significant pain reduction (>30%) in the active group compared with 11 in the sham group (40.7%) 15 days after treatment (P = .03). The administration of 10 Hz rTMS on the contralateral primary motor cortex for 2 weeks in traumatic amputees with PLP induced significant clinical improvement in pain.
Perspective
High-frequency rTMS on the contralateral primary motor cortex of traumatic amputees induced a clinically significant pain reduction up to 15 days after treatment without any major secondary effect. These results indicate that rTMS is a safe and effective therapy in patients with PLP caused by land mine explosions.
Land mines are one of the world's most disabling public health hazards causing devastating injuries such as traumatic limb amputations and associated psychological disorders.
The exact number of worldwide land mine victims is currently unknown because there is no systematic collection of reliable data. However, it is widely estimated that land mines result in 15,000 to 25,000 victims each year.
PLP is a neuropathic syndrome characterized by pain felt in the patients' remaining perception of the amputated limb after partial or complete deafferentation. This pain is usually described as a stabbing, throbbing, burning, or cramping sensation.
and is considered a challenging condition because of its negative effect on quality of life and lack of treatment response, particularly in patients with traumatic-related amputations.
Maladaptive plasticity seems to play a major role in the mechanisms of PLP. Reorganization of the primary sensorimotor cortex, including changes in motor cortex excitability and peripheral factors such as nociceptive inputs from the residual limb have been implicated in the development of this condition.
The high prevalence of PLP after amputation and its lack of treatment response have resulted in major efforts to develop interventions to decrease the pain in affected patients.
In light of PLP mechanisms, repetitive transcranial magnetic stimulation (rTMS) has been tested in this condition as a tool to block the maladaptive plasticity in the sensorimotor cortex.
rTMS applied daily over the primary motor cortex (M1) has shown pain relief effects in other neuropathic pain syndromes such as poststroke pain and spinal cord injury pain.
The RCT showed that 5 consecutive sessions of rTMS induced a significant analgesic effect compared with sham rTMS, lasting up to 2 months in 39% of the subjects. However, a recent meta-analysis judged this trial as a high risk of bias study due to a deficient randomization method, which led to an unbalanced distribution between the intervention groups.
Furthermore, the conclusion of the cited meta-analysis, after including 56 trials using noninvasive brain stimulation techniques for chronic pain treatment, was that although single doses of high-frequency rTMS of the motor cortex may have small short-term effects on chronic pain, these effects do not meet the predetermined threshold of minimal clinical significance, and there is therefore a need for larger, rigorously designed studies, particularly of longer courses of stimulation.
In light of these results, we aimed to assess in a larger sample size study and properly designed RCT, the immediate and sustained effects of a larger dose of real rTMS of M1—10 sessions—on PLP compared with sham rTMS in land mine victims. We hypothesized that 10 Hz rTMS for 2 weeks over M1 contralateral to the PLP could significantly decrease the level of pain compared with sham stimulation.
Methods
Study Design
This was a single-center, double-blinded, sham-controlled, randomized, parallel-group trial that consisted of 3 main phases: 1) a baseline evaluation consisting of a week period of observation to establish baseline measurements for pain levels, depression, and anxiety symptomatology, 2) a treatment phase consisting of daily sessions with active or sham rTMS for 5 days a week during 2 consecutive weeks, and 3) a follow-up evaluation after 15 and 30 days of treatment completion. In the baseline evaluation, we recorded demographic data, medical history, medications, and other therapies used for the treatment of PLP.
Study Population
Fifty-four patients (mean age, 33.9 ± 8.41 years; 4 female patients) were included in the study. The participants were prospectively selected from the rehabilitation department of the Regional Military Hospital and local nongovernmental organizations in Bucaramanga, Colombia. Patients were included if they fulfilled the following criteria: adults aged 18 years or older, who had amputation at any level of 1 lower limb by antipersonnel land mines with symptoms compatible with PLP. PLP was defined as a sensation of shooting, stabbing, boring, squeezing, throbbing, or burning or paresthesia or any other pain sensation in a limb that did not exist anymore.
We excluded patients with a diagnosis of complex regional pain syndrome, any pathology that could alter the course of PLP (diagnosis of cancer, immunological disorders, renal insufficiency requiring dialysis treatment, etc), pregnancy, neuropsychiatric disorders that can affect the patient ability to consent to the study participation and contraindications to rTMS, such as cardiac pacemaker, medical pumps, or implanted metals in the scalp.
Risk and safety of repetitive transcranial magnetic stimulation: Report and suggested guidelines from the International Workshop on the Safety of Repetitive Transcranial Magnetic Stimulation, June 5-7, 1996.
Transcranial magnetic stimulation for pain control. Double-blind study of different frequencies against placebo, and correlation with motor cortex stimulation efficacy.
magnetic stimulator, which provides a biphasic pulse (Magstim Company Ltd, Whitland, UK). The coil was positioned tangentially to the scalp, approximately at a 45° angle from the midline. The resting motor threshold (RMT) (of the first dorsal interosseous) was defined as the minimal intensity to induce motor evoked potentials of 50 μV peak-to-peak amplitude in at least 5 of 10 trials. Twenty trains of 6 seconds each (intertrain interval 54 seconds), using an intensity of 90% of RMT and 10 Hz frequency, were applied in each patient for 10 days during a 2-week period. For the sham treatment group, stimulation parameters were the same (location and duration), but a sham coil (Magstim Company Ltd) was used. This coil has similar appearance to the active coil in shape and weight, and produces a similar sound artifact but does not induce a scalp skin sensation nor emit a magnetic pulse within the cortex.
All sessions were administered by only 1 investigator who was not blinded to the intervention and did not participate in the outcome assessments. Participants and investigators who performed the pain assessments were blinded to treatment allocation.
Randomization
A computer-generated randomization method with a permuted block size of 6 was used to allocate subjects to the sham or active rTMS interventions. The randomization code was only given to the treating investigator on the first day of treatment session by an independent investigator not involved with any other aspect of the trial.
The blinding integrity was assessed at the end of the study. Participants were asked to guess their treatment allocation. We did not assess blinding in 2 patients because of early trial withdrawal from the study. The blinded investigators were also asked to guess the patient allocation.
Study Outcomes
All evaluations were performed by an investigator blinded to treatment allocation. The primary end point of the study was the score change in the visual analogue scale (VAS) for pain. Response was defined as a reduction of ≥30% compared with baseline (at 15 and 30 days after treatment).
The other outcome measures were considered secondary.
Pain Measurement
VAS for Pain
The response to the stimulation was evaluated by measuring the pain intensity using the VAS. This self-evaluation scale ranges from 0 to 10 as visually described in centimeter units, 0 cm indicates no pain and 10 cm the worst pain possible. This scale has been widely used in studies that evaluate pain as an outcome, and validity and reproducibility have been shown.
Because we expected daily variability in pain levels, pain was self-assessed daily at baseline for 1 week before treatment, and twice during the follow-up period (at 15 and 30 days after completing the treatment scheme). The patients were asked to continue their routine medication regimen during the study. If a patient required a change in medication dose, it was recorded and considered in the analysis.
Anxiety and Depression Symptomatology
Because depression and anxiety might be confounders of pain relief, we measured these 2 domains by using the instruments in the following sections.
Zung Self-Rating Depression Scale
This is a 20-item self-report scale that measures the 4 common characteristics of depression: pervasive affect, physiological equivalents, other disturbances, and psychomotor activities. The minimum score is 20 and the maximum score is 80. Four categories ranging from “normal” to “severely depressed” are on the basis of specific ranges of the score.
This is a 20-item questionnaire on the basis of scoring in 4 groups of manifestations: cognitive, autonomic, motor, and central nervous system symptoms. The total scores range from 20 to 80, meaning normal range to extreme anxiety levels.
Depressive and anxiety symptoms were measured once at baseline, and twice during the follow-up period (at 15 and 30 days after completing the treatment scheme).
Sample Size and Statistical Analysis
A sample size of 54 patients (27 in each arm) was calculated expecting that 60% of subjects in the active group would obtain a significant pain reduction (decrease >30% in pain level) after finishing the intervention compared with 20% in the sham group. This was on the basis of the results of a previous study.
It was considered a power of 80%, type I error of .05 (double-sided), and an adjustment for a dropout rate of 5%.
The data are presented as mean and SD and also proportion of responders in each group. Continuous variables were subjected to a Shapiro–Wilk test to determine whether the data fitted normal distribution. Baseline characteristics of patients randomized to active and sham therapies were compared using Student t-test or Wilcoxon rank sum test for continuous variables and χ2 test or Fisher exact test for categorical data. We analyzed the end point of the study using the intention-to-treat method including patients who attended at least 1 of the rTMS sessions. The missing data were considered at random, thus we used a regression imputation method to handle this issue. Such technique allowed us to substitute missing VAS values at the first (15 days after treatment, 2 subjects) and second (30 days after treatment, 6 subjects) follow-up visits, for values derived from a regression model using baseline variables as well as VAS scores from all other participants.
Risk ratios were calculated to evaluate statistically significant differences between treatment groups in the proportion of subjects attaining a clinical important pain reduction (>30%) 15 and 30 days after finishing the treatment protocol. The differences between groups in the proportion of subjects attaining a substantial clinical benefit (pain reduction >50%) were also explored. We also conducted additional analyses treating pain as a continuous variable and also for the other secondary continuous outcomes (depression and anxiety scores). For these analyses, a repeated-measure analysis of variance was performed using a 2-group (active vs sham) according to a 3 time periods (baseline, and 15 and 30 days after finishing treatment) design. Post hoc comparisons were carried out using a Scheffe test for multiple comparisons. Statistical significance was defined as P < .05. All analyses were conducted using Stata statistical software, release 11.0 (Stata Corp, College Station, TX).
Results
Fifty-four patients (n = 27 in the active group and n = 27 in the sham group) were included in the study. A participant flow diagram is shown in Fig 1. There were no significant differences in demographic and clinical characteristics at baseline between the groups (Table 1). All patients tolerated the rTMS without experiencing any serious adverse effect. Some patients experienced minor adverse effects such as headache (11.1%), neck pain (5.5%), and sleepiness (18.5%) without significant differences between groups. There were also no differences in relation to the current use of nonsteroidal anti-inflammatory drugs (28.5% vs 37.0%, P = .12), and the participation in a physical rehabilitation program (88.8% vs 85.1%, P = .68) or psychological therapy (88.8% vs 77.7%, P = .27).
Figure 1Consolidated Standards for Reporting Trials 2010 patient flow diagram.
A significantly greater mean percentage reduction in pain intensity (VAS score) was found 15 days after treatment in the active group compared with the sham stimulation group (−53.38 ± 53.12% vs −22.93 ± 57.16%; mean between-group difference = 30.44%, 95% confidence interval [CI], .30–60.58; P = .03). However, no significant differences between groups were found 30 days after treatment (−37.74 ± 52.39% vs −14.97 ± 53.88%; mean between-group difference = 22.76%; 95% CI, −6.25 to 51.79; P = .12).
Nineteen subjects (70.3%) attained a significant clinical response (pain reduction > 30%) in the active group compared with 11 (40.7%) in the sham group 15 days after treatment (risk ratio [RR] = 1.72; 95% CI, 1.03–2.89). However, no statistically significant between-group difference was found 30 days after treatment (15 [55.5%] vs 9 [33.3%]; RR = 1.66; 95% CI, .88–3.13). A higher proportion of subjects obtaining a substantial clinical benefit (pain reduction >50%) was also found in the active treatment group compared with the sham stimulation group 15 days after treatment (17 [62.9%] vs 9 [33.3%]; RR = 1.88; 95% CI, 1.02–3.46). This difference also showed a statistical trend toward significance when evaluated 30 days after treatment (13 [48.1%] vs 6 [22.2%]; RR = 2.16; 95% CI, .96–4.85).
For our secondary analyses, we assessed the effects of rTMS on pain level using a repeated measure analysis of variance. We found a significant main effect of group of treatment and time (F1,104 = 7.54, P < .01; F2,104 = 19.49, P < .0001). The analysis also showed a significant interaction term (group per time; F2,104 = 3.25, P = .04). Post hoc tests revealed a significant decrease in VAS scores 15 and 30 days after finishing the intervention in the active group, whereas no significant change was noted with sham stimulation (Fig 2). No statistically significant between-group difference was found when comparing the absolute VAS scores at day 15 (mean between-group difference = 1.42; 95% CI, −.07 to 2.93; P = .06) or day 30 (mean between-group difference = .86; 95% CI, −.59 to 2.31; P = .24) after treatment (Table 2). In relation to the scores of the depression and anxiety scales a main effect of time was found (F2,104 = 4.55, P = .01; F2,104 = 7.91, P < .0001), without significant group of treatment effects (F1,104 = .11, P = .7; F1,104 = .07, P = .7) or interactions terms (F2,104 = 1.32, P = .27; F2,104 = .2, P = .81). No statistically significant between-group difference was found when comparing the absolute scores of the depression and anxiety scales at day 15 or day 30 after treatment (Table 2).
Figure 2Variations in VAS score according to treatment group. *P < .05 compared with baseline score. The error bars represent standard errors.
Subjects and investigators did not guess correctly the treatment allocation beyond chance (P = .704; P = .571).
Discussion
The present study showed that the treatment with 10 Hz rTMS of contralateral M1 during 2 weeks in traumatic amputees with PLP induced a clinically significant pain reduction up to 15 days after treatment compared with sham stimulation. In addition, no serious adverse effects were found during the study indicating that rTMS was a safe and effective therapy in patients with PLP caused by land mine explosions.
Previous studies had shown some beneficial effects of rTMS on PLP.
These reports have evaluated either the effects of low frequency rTMS (<1 Hz), which have been shown to decrease cortical network excitability, or high frequency (>1 Hz), which may induce an opposite effect.
Transcranial magnetic stimulation for pain control. Double-blind study of different frequencies against placebo, and correlation with motor cortex stimulation efficacy.
evaluated the effect of rTMS series on phantom pain-like syndrome in 2 patients with long-lasting brachial plexus avulsion, who underwent 10 and 1 Hz rTMS during 15 days, separated by 4 and 6 weeks, respectively (at 110% of RMT, 12-minute duration) over the contralateral posterior parietal cortex to the injured limb. The authors reported a maximum pain reduction of approximately 60% and 23.6%, during the rTMS treatment compared with baseline; however, the pain decrease was not maintained in the long-term. Similarly, Di Rollo and Pallanti
in 1 patient with PLP of traumatic origin, applied 15 sessions of low-frequency rTMS (thirty 20-second trains at 80% of RMT, 15 minutes) over the ipsilesional motor cortex, showing a pain reduction of 33.3% at the end of the third week of treatment and a slight decrease (16.6%) at the follow-up visit (3 weeks after the last session). In a recent clinical trial, Ahmed et al
evaluated the analgesic effect of rTMS for chronic PLP by assigning subjects to active (n = 17) or sham (n = 10) stimulation of the contralateral M1 for 5 consecutive days (200 pulses at 20 Hz, 10-second trains, at 80% of RMT). The authors reported a significant reduction on VAS in the real stimulation group immediately after the fifth session (55%) that was maintained after 2 months (39%) compared with the sham group. Although these studies showed promising effects of rTMS in PLP, there were some methodological limitations that could have affected their results, such as a low sample size, unbalanced distribution in the treatment groups, low number of sessions (ie, 5 sessions) and the lack of standardized criteria for placebo stimulation.
In addition, the population included in these studies was heterogeneous, with different amputation locations and etiologies, which is of particular importance because these factors could be related with different pathophysiological mechanisms and/or treatment responses.
Because of the challenges in recruiting a population with PLP with similar characteristics, our study included, in a 10-day stimulation protocol, a homogeneous population consisting of 100% of subjects with traumatic lower limb amputation caused by land mine explosions. Thus our findings extend beyond previous rTMS studies on PLP and provide a more reliable estimate of effect size. The number needed to treat for 30% pain reduction with rTMS compared with sham rTMS was 4. This result indicates that 1 patient in every 4 treated with rTMS will benefit from this treatment compared with sham treatment. This effect size (number needed to treat of 4; 95% CI, 1.8–23.1) is similar to tricyclic antidepressants for the treatment of central pain.
The pain relief found in the present study could be explained by the potential effect of rTMS over the central pathophysiological mechanisms related to PLP. After a traumatic amputation, the main factors associated with PLP include maladaptive reorganization of the sensorimotor cortex, which involves a reduction in intracortical inhibition mechanisms, an imbalance between inhibitory and excitatory amino acids (gamma-aminobutyric acid and glutamate) and an increase in the excitability of corticospinal neurons.
It has been hypothesized that the administration of high-frequency rTMS over the motor cortex enhances its excitability leading to indirect activation of inhibitory projections toward the thalamus, resulting in a modulation of ascending nociceptive signal pathways.
Additionally, the modulation of thalamic activity generated by the enhancement of motor cortex excitability may influence other brain pain-related networks such as the orbitofrontal, anterior cingulate gyri, and the periaqueductal gray matter, which are related with the affective–emotional components of nociception.
Besides rTMS, other noninvasive and invasive brain stimulation methods have been explored for the treatment of chronic PLP. Transcranial direct current stimulation (tDCS) is a noninvasive method that modulates spontaneous neuronal activity with anodal stimulation enhancing cortical excitability and cathodal inducing an opposite effect.
studied the effect of anodal tDCS (1.5 mA, 15 minutes for 5 days) over M1 contralateral to the amputated limb in 8 patients with unilateral lower and upper limb amputation of different etiologies. The authors reported a pain relief immediately after the 5 sessions and up to 1 week of the last stimulation session (−41%, P = .04). Among the invasive stimulation methods, epidural motor cortex stimulation (MCS) has emerged as an alternative therapy for refractory neuropathic pain.
performed a systematic review of the effects of MCS on chronic neuropathic pain. The authors reported that 4 of 10 of the patients (40%) with PLP reported pain relief (>70%). In our study we found that 63% of the patients with PLP experienced pain relief >50%, results that are comparable with those reported in MCS studies; furthermore, these effects were present 15 days after finishing the treatment, which extend beyond the findings in patients using tDCS. It is worth noting that a recent meta-analysis evaluating the use of rTMS for the treatment of chronic pain reported significant heterogeneity, reporting a short-term analgesic effect but failing to reach pre-established criteria for a minimal clinically important difference.
One strength of our study was the inclusion of a very homogeneous population and the administration of the stimulation for 10 days, which could have contributed to the lasting and clinically significant pain reduction.
We also observed a significant reduction of depressive and anxiety symptoms after 30 days after the intervention without any differences between treatment groups, indicating an effect not attributable to rTMS. A recent meta-analysis showed that high-frequency rTMS over the dorsolateral prefrontal cortex is associated with clinically relevant antidepressant effects with a safe profile, whereas no consistent effects have been found when stimulating the motor cortex,
which could explain the lack of differential treatment effects in our study.
This study has some limitations. Although our main outcome was the proportion of subjects attaining a clinically significant pain reduction, other nonpainful phantom phenomena such as phantom limb awareness, telescoping, and phantom sensations were not assessed and could have been confounding factors in the evaluation of rTMS effects. Despite this, our results support the notion that rTMS induces a clinically significant pain relief in subjects with PLP after traumatic amputation. In addition, although we found a significant difference between groups when analyzing mean percentage reduction (difference in pain scores from baseline), there was no statistically significant between-group differences when analyzing absolute VAS scores after treatment. Although randomization in theory allows for a balanced baseline measure in both treatment groups, this balance is often not seen in small randomized trials such as ours. Therefore, analyses of differences can correct for baseline imbalances, providing a more precise estimation of treatment effects in small randomized trials.
In addition, as for any small RCT, our results need to be confirmed in large randomized trials. An additional limitation is that we stimulated the motor cortex corresponding to the first dorsal interosseous muscle of the hand contralateral to pain instead of the area corresponding to the lower limb. However, previous studies using rTMS over the hand motor cortex have reported analgesic effects in patients with chronic neuropathic pain of diverse anatomical origin,
results that were confirmed by our study. Although the sham coil used in this trial might induce a slightly different scalp sensation compared with the real stimulation coil, it is noteworthy that both coils are similar in appearance (shape and weight) and auditory artifacts. In addition, although the investigator performing the stimulation was not blinded to the intervention, she did not participate in the outcome evaluation of the subjects; therefore, it is unlikely that this fact could have influenced the obtained results. Finally, this study consisted of a small sample size, which can compromise the generalization of the results. This was in part because of the challenges of recruiting such a homogeneous study population. In addition, although we found a clinically significant effect of the stimulation on pain reduction 15 days after finishing the treatment, further studies will be necessary to determine if longer rTMS stimulation protocols could derive in even more long-lasting and maintained analgesic effects in patients with PLP.
References
Ahmed M.A.
Mohamed S.A.
Sayed D.
Long-term antalgic effects of repetitive transcranial magnetic stimulation of motor cortex and serum beta-endorphin in patients with phantom pain.
Transcranial magnetic stimulation for pain control. Double-blind study of different frequencies against placebo, and correlation with motor cortex stimulation efficacy.
Risk and safety of repetitive transcranial magnetic stimulation: Report and suggested guidelines from the International Workshop on the Safety of Repetitive Transcranial Magnetic Stimulation, June 5-7, 1996.
This study was partially supported by a grant from the Colombian Science and Technology Institute (COLCIENCIAS, project code: 6566-49-326169). Felipe Fregni is the principal investigator at Spaulding Rehabilitation Hospital of a research grant funded by NIH (5R01HD082302-02).
The authors have no conflicts of interest to declare.
30064-5&id=gr1.jpg){kind=link}
30064-5&id=gr2.jpg){kind=link}