Main findings.
The search found 1116 articles (see Fig. 1), of these 1701 in Medline, 5 in Lilacs, 270 in Cochrane, 138 in Embase, 34 in PsycINFO, and 2 in grey literature; After deleting the duplicates 1055 abstracts were read to identify compliance with the selection criteria; 960 were excluded, leaving 95 to which the text was completely read; in this last revision 65 were excluded leaving 30, of which, by making a judicious review of the PICO, 12 were excluded which did not contemplate the approach or type of the study, or included different types of pain. In the end, 18 articles complied with the PICO proposed. Those articles are summarized on Table 1.
Table 1
Summary of interventions and results from included studies
Study
|
N
|
Intervention
|
Result
|
Quality of life
|
Khedr et al., 2015
|
n = 34 oncologic neuropathic pain
|
10 sessions, 5/week. 20 Hz, 10 sec, 80% intensity, 10 sessions
Follow-up two weeks
|
Pain reduction
|
N/A
|
Quesada et al., 2020
|
N = 42 central
neuropathic pain
|
Four sessions, 20 consecutive trains of 80 pulses at 20 Hz, at 80% of the engine threshold, the interval of 84 s, during a session of 27 minutes.
|
Pain reduction
|
Remained unchanged.
|
Abdelkader et al., 2019
|
N = 20 diabetic neuropathy
|
5 sessions of 40 minutes, 15 consecutive trains (2 s duration) of 50 stimuli at 10 Hz, at 100% of the engine threshold, intervals between trains of 30 s were delivered.
|
Significant pain reduction
|
N/A
|
Ahmed et al., 2020
|
N = 30 diabetic neuropathy
|
5 sessions/20 Hz or sham/80% RMT/2000 pulses/M1
Follow-up 5days, 1, and 2 months
|
Significant pain reduction
|
N/A
|
Attal et al., 2016
|
N = 36 lumbar neuropathic radicular pain
|
3 sessions, 3 consecutive days, 3000 pulses (30 trains of 10 seconds each, with an interval between trains of 20 seconds) delivered at a frequency of 10 Hz.
|
Pain reduction
|
N/A
|
Hosomi et al., 2020
|
N = 144 neuropathic pain from different locations
|
An active rTMS session involved 10 trains at 90% RMT (50 pulses/train at 5 Hz; the interval between trains, 50 seconds)
|
Pain reduction
|
Remained unchanged
|
Hosomi et al., 2013
|
N = 70 neuropathic pain from different locations
|
1 session/5 Hz/90% RMT/500 pulses/M1
Follow-up 17 days
|
Pain reduction
|
N/A
|
Kim et al., 2020
|
N = 30 central neuropathic pain
|
A daily rTMS session was held for 5 days. 3 pulses (60 ms) at 50 Hz and a 5 Hz TBS train of last 2 sec, intervals of 10 seconds for 200 seconds (600 stimuli in total), with an 80% intensity
|
Significant pain reduction
|
N/A
|
Cervigni et al., 2018
|
N = 13 painful bladder syndrome and interstitial cystitis
|
The sessions of 30 consecutive trains of 50 stimuli delivered at 20 Hz at 110% of the RMT
|
Significant pain reduction
|
Improvement
|
Goto et al., 2020
|
N = 11 chemotherapy-induced neuropathy
|
20 Hz, 10 s, 10 trains with 80% intensity.
Four sessions
|
Pain reduction
|
N/A
|
Lindholm et al., 2015
|
N = 16 orofacial pain
|
1 session/10Hz or simulation /90% RMT/1000 pulses/contralateral S1-M1 or S2 right. Four weeks Follow-up
|
Significant pain reduction
|
Improvement
|
Pommier et al., 2016
|
N = 40 central neuropathic pain
|
Four sessions, 20Hz, 20 consecutive trains of 80 stimuli at 80% of the engine threshold, intervals between trains of 84 s, session of 26 minutes. Follow-up average of 311.8 days.
|
Pain reduction
|
N/A
|
(Ayache et al., 2016)
|
N = 66 neuropathic pain from various causes and locations
|
A single session of 10 Hz-rTMS M1
|
Significant pain reduction
|
N/A
|
(Hodaj et al., 2020)
|
N = 57 neuropathic pain from different locations
|
10 daily sessions for 2 weeks and then 2 sessions in 1 week followed by 3 weekly sessions, 2 weekly sessions, and 5 monthly sessions / 10 Hz / 80% / 2000 pulses in 20 minutes. Follow-up 0.5, 1, 3 and 6 months
|
Significant pain reduction
|
Significant improvement
|
Lawson McLean et al., 2018)
|
N = 48 neuropathic pain from various sources
|
10 Hz- rTMS on M1
|
Significant pain reduction
|
N/A
|
Lefaucheur et al., 2012
|
N = 14 neuropathic pain from different locations
|
1 session/10 Hz/90% AMT/2000 pulses/M1 preceded or not by iTBS or cTBS/80% AMT/600 pulses/M1
One week Follow-up
|
Pain reduction
|
N/A
|
Nurmikko et al., 2016
|
N = 27 neuropathic pain of different origins
|
10 Hz, 80% RMT, 3000 pulses, 3 sessions
|
Pain reduction
|
N/A
|
Sampson et al., 2011
|
N = 9 neuropathic pain from different locations
|
15 sessions/1 Hz/110% RMT/1600 pulses/ correct DLPFC
Follow-up 1, 3 weeks and at 3 months
|
Pain reduction
|
N/A
|
RMT: Resting motor threshold. AMT: Active motor threshold. rTMS: Repetitive transcraneal magnetic stimulation. cTBS: Continuous Theta burst stimulation. iTBs: Intermittent Theta burst stimulation. DLPFC: dorsolateral prefrontal cortex. |
Participant characteristics
We included 18 articles: 9 ECAS and 9 observational studies. The total sample of the studies was 707 patients, with 312 (44.14%) women and 395 (55.86%) men. The average age was 52 years old.
Intervention
All studies used rTMT in their treatment, the number of sessions vary from a single session to 15 sessions, with an average of 3.5 sessions. Follow-up was variable ranging from 7 to 311.8 days, with an average of 45 days. Pain reductions were obtained in the 18 studies.
Excluded studies
The main reasons for exclusion were different intervention, pain nature and type of study. Appendix 3.
Primary outcomes
Pain Modulation
From the 18 articles included 9 were ECAS (38–46), with a total population of 419 patients.
The article of Khedr et al., (2015) (38) , included 34 patients with a diagnosis of chemotherapy-induced neuropathy, rTMS treatment was performed in 10 daily sessions, 5 days a week. Pain assessment was performed before and after the first, fifth, and 10th session and then 15 days and one month later. Changes in the verbal descriptor scale (VDS), and visual analog scale (VAS) were significant in the intervention group, particularly at the end of treatment, 49.1% VDS and 36.7% VAS and at 15 days of follow-up 45.6% VDS and 35.5% VAS.
The study of Quesada et al., (2020) (39) included 42 adults with central pain, divided into active and simulated phases. Each phase included 4 sessions and an 8-week washout between phases. The pain was assessed with VAS assessed after each rTMS session and at the end of the phase. There was a mean difference of 20.79% R (95% CI) between the active and sham phases. 23 patients (54%) and 15 patients (35%) achieved pain relief of >30% or >50%, respectively.
Two articles included adults with diabetic neuropathy (40,41). The results were evaluated with VAS in both studies. Abdelkader et al., (2019) (40) reported a significant mean VAS reduction for insulin-dependent patients (5.10 ± 2.6 mm; p = 0.011) and non-insulin-dependent patients (4.0 ± 1.7 mm; p = 0.005). Ahmed et al., (2020) (41) found significant decreases (P<0.05) in pain severity after rTMS and after TENS intervention.
The article of Attal et al., (2016) (42), included a population of 36 patients, 24 for active stimulation and 12 for sham group. The treatment protocol included two blocks separated by an interval of three weeks. Each block consisted of three sessions of rTMS or tDCS (transcranial direct stimulation) on three consecutive days (V1, V2, V3). Treatment effects were assessed with Numerical Rating Scale (NRS) immediately before stimulation in V1, after stimulation in V2 and V3 and five days after the last session. The study concluded that active rTMS was superior to tDCS and simulation in pain intensity (F = 2.89; p = 0.023). rTMS reduced cold pain thresholds (p = 0.04) and its effect on cold pain was correlated with analgesic efficacy (p = 0.006).
The study by Hosomi et al., (2020) (44), included 144 patients with pain from different locations. 70 patients for active treatment and 67 patients for sham rTMS. Patients received 5 daily sessions of active or sham rTMS. In addition, responders whose VAS decreased an average of 10 mm or more with daily rTMS could join a continuous open-ended trial to test active rTMS administered at least once a week for 4 weeks. The VAS was recorded before and after each intervention, the decrease in the mean VAS was 8.0 (SD, 12.1) in the active stimulation group and 9.2 (SD 13.6) in the simulated group.
The article of Hosomi et al., (2013) (45), included 70 patients with neuropathic pain from different locations, 34 in group A and 36 in group B. As for the treatment, a session of rTMS was applied daily for 10 consecutive days. Pain intensity was assessed daily in each patient using a visual analog scale (VAS; 0-100 scale). In the results, the mean rates of reduction in VAS (95% CI) averaged in real and simulated rTMS were 6.51% (3.68–9.34) vs 2.44% (0.98–5.62) just after stimulation, and 3.38% (1.13–5.62) vs 0.66% (2.74–1.43) 60 min after.
Kim et al., (2020) (46), included 30 patients who presented a central neuropathic pain, 15 patients in group A and 15 in group B. A daily rTMS session was held for 5 days. Patients were evaluated with the NRS of 11-point scale. There was a significant reduction in NRS of the real group with a difference of 15.20±7.46.
In the article of Cervigni et al., (2018)(43), 13 patients with neuropathic pain caused by painful bladder syndrome/interstitial cystitis were chosen. The first group of patients received rTMS-real treatment for five consecutive days followed, by a washout period of 6 weeks, for the simulated treatment of rTMS; the second group received the same treatments in reversed order. The scales used to assess pain were VAS, Functional Neuropathic Pelvic Pain syndrome (FPPS), Neuropathic Pain Symptom Inventory, McGill Questionnaire, and Central Sensitization Inventory LUTS Urinary. In the actual stimulation phase, it obtained a significant overall reduction for VAS (F [5.45] = 5200, P = 0.001).
Of the 18 studies, 9 quasi-experimental studies were found (47–55) with a total population of 288 patients.
Goto et al,. (2020) (47), included 11 adults with oncological diagnosis and chemotherapy-induced neuropathy, who underwent 4 sessions of rTMS. Pain was assessed at baseline, pre-intervention, and post-intervention. P-VAS scores decreased significantly in pre-stimulation (mean 55.3) and after stimulation (mean 46.8).
Pommier et al., (2016) (49) included 40 adults with neuropathic pain of central origin who underwent 4 sessions separated from each other by an interval of 3-4 weeks. The evaluation was carried out after 4 sessions. In the 31 patients who responded, the average number of sessions per patient was 11, the average follow-up was 311.8 days, the average interval between sessions was 28.4 days. The average percentage of pain relief (Global %R) in the responder's group was 41%. The mean VAS score before rTMS was 6.35 vs 6.16 after four sessions.
The study of Lindholm et al., (2015) (48), had 16 adults with orofacial pain. All participants received 2 active rTMS treatments and 1 sham treatment (placebo). The 3 treatments were separated from each other by 4 weeks. NRS was assessed at baseline and after each rTMS session. Brief Pain Inventory (BPI) was evaluated at baseline, 3-5 days after treatment and 1 month after. Pain intensity (NRS) was lowest on the third day after S2 stimulation (MS: 3.8, SE: 0.6) in comparison with S1/M1 stimulation (MS: 5.4, SE: 0.6; P 5 0.0071) or simulated (MS: 5.3, SE: 0.6; P 5 0.0187). These results remain after 1 month.
Ayache et al., (2016)(50), included 66 patients with neuropathic pain of different causes and locations, who were evaluated for the analgesic effect in a single session. This study compared analgesia induced by stimulation of pain location without hand motor hot spot (hMHS) consideration, versus primary motor cortex (M1). The analgesic effect was assessed with VAS, finding that the percentage of pain reduction during the week (PPRweek) and for the days 2 and 3 (PPRd2d3) showed a significant difference during the hMHS-targeted procedure and the M1 compared to the sham procedure (p<0.05 and p<0.001, respectively).
The study of Hodaj et al., (2020) (51) had 57 pain patients from different locations. The treatment consisted of one session per day for five days for two consecutive weeks (weeks 1 and 2), then 2 sessions in the following week (week 3) for a total of 12 sessions. Maintenance therapy was conducted, consisting of a rTMS session at week 4 and then bi-monthly sessions over the next five months, for a total of 11 sessions.
Lawson McLean et al., (2018)(52), included 48 patients with atypical facial pain, rTMS treatment was performed daily for a total of nine sessions. Pain intensity was documented according to the visual analog scale.
In the study of Lefaucheur et al., (2012) (53), 14 patients with different pain locations where included. The patients underwent 3 sessions separated by four weeks. The pain was assessed with VAS for 1 week, before the first rTMS session and after each rTMS session. The mean pain score reduction score was 41% when combined with iTBS (intermittent theta-burst stimulation) and 31% compared to cTBS (continuous theta-burst stimulation).
Sampson et al., (2011) (55) performed 15 rTMS sessions on 9 patients with pain from different locations, (5 days a week for 3 weeks). p-VAS was evaluated before and after all rTMS treatments and monthly for 3 months. In the results, three subjects had a 50% decrease in pain ratings upon completion of rTMS treatments, and 1 subject responded more slowly with a 50% improvement in pain at the end of the 3-month follow-up.
The study of Nurmikko et al., (2016) (54) included 27 patients with pain from different parts of the body. Divided into 3 groups: Pharmacological treatment, rTMS, and the control group. Treatment was designed by applying 3 cycles of 5 sessions of rTMS 3-5 times per week. Patients rated their average pain every 24 hours with the NRS, and before and after 5 sessions patients completed the Brief Pain Inventory (BPI). The results showed that rTMS provided pain relief and increased the total rate of responders by 58%.
Secondary outcomes
Quality of life
In the search, we found 5 articles that evaluate the quality of life in their results. The sample amounts to 272 patients (39,43,44,48,51).
In the article of Cervigni et al., (2018) (43), they described a significant overall effect on the SF-36 sub-score related to physical pain, in the actual stimulation phase (F [5.50] = 2636, P = 0.034) and for the emotional state-related sub-score (F [5.50] = 3096, P = 0.016).
In the study of Hosomi et al., (2020)(44), the EQ-5D-5L for quality of life was recorded after each intervention, on the 1st day, after the intervention on the 5th day, and the 4th week. None of the outcomes for quality of life were significantly different between groups.
Hodaj et al., (2020) (51) used Sf-36 and RAND 36 to assess quality of life. A significant improvement for physical and mental component summaries and scores was observed in the results.
In the study by Quesada et al., (2020) (39) quality of life was assessed with EuroQol (EQ-5D), it was applied at the beginning and end of each phase. In the results, the EQ-5D remained unchanged in both the active and simulated phases.
Lindholm et al., (2015) (48) included 16 adults with orofacial pain, quality of life was assessed by NePIQoL and SF-36 at baseline and 1 month later, finding a small reduction in total NePIQol score 1 month after S2 stimulation (S: 79.8, SE: 5.7) compared to baseline (ES: 86.6, SE: 5.7; P50.0031).
Adverse effects
Of the 18 articles included for this systematic review, three (16.66%) of them (38,41,52) do not mention in their content the adverse effects of transcranial magnetic therapy, two of them (11.11%) (42,55) mentioned that these effects are observed and measured, without describing their conclusions. Three of them (16.66%) (43,45,54) mentioned adverse effects classified as mild, among them headache, pain in the extremity, dizziness and drowsiness, being these the most noted. However, ten of the articles (55.55%) (39,40,44,46–51,53) mentioned not having found severe adverse effects in their protocols.
Risk of Bias
Nine of the eighteen studies (38–46) were evaluated with the Cochrane risk-of-bias tool for randomized trials, in summary the studies showed a low risk of selection bias but high risk of performance and detection bias due to the lack of blinding of participants, personnel and outcome assessors. (Figure 2).
The studies with other methodology (47–55) were evaluated with JBI Critical Appraisal Checklist for quasi-experimental studies from the Faculty of Health and Medical Sciences at the University of Adelaide, South Australia, it can be seen that 7 of the 9 articles fulfilled more than 50% of the checklist items. (Table 2.)
Table 2
JBI Critical appraisal checklist for quasi-experimental studies
ITEM
|
Yes
|
No
|
Unclear
|
1. Is it clear in the study what is the ‘cause’ and what is the ‘effect’ (i.e. there is no confusion about which variable comes first)?
|
8
|
0
|
1
|
2. Were the participants included in any comparisons similar?
|
7
|
2
|
0
|
3. Were the participants included in any comparisons receiving similar treatment/care, other than the exposure or intervention of interest?
|
8
|
1
|
0
|
4. Was there a control group?
|
7
|
2
|
0
|
5. Were there multiple measurements of the outcome both pre and post the intervention/exposure?
|
7
|
2
|
0
|
6. Was follow up complete and if not, were differences between groups in terms of their follow up adequately described and analysed?
|
7
|
2
|
0
|
7. Were the outcomes of participants included in any comparisons measured in the same way?
|
9
|
0
|
0
|
8. Were outcomes measured in a reliable way?
|
9
|
0
|
0
|
9. Was appropriate statistical analysis used?
|
9
|
0
|
0
|
Quality of evidence
We found moderate quality of evidence for the effects of interventions in Pain Modulation. Very low-quality evidence was found about effects on changes of Quality life (See Figure 3).