Pulsed Electromagnetic Fields to Reduce Diabetic Neuropathic Pain and Stimulate Neuronal Repair: A Randomized Controlled Trial

Article Outline

• Abstract
• Methods
  • Enrollment Criteria
  • Randomization
  • Device
  • Outcome Measures
    • Primary outcome
    • Secondary outcomes
  • Statistical Analyses
• Results
  • Primary and Secondary Outcomes
  • Correlations Between Pain Measures
  • Biopsy Study
  • Blinding
  • Safety
• Discussion
  • Study Limitations
• Conclusions
• Appendix 1: Investigators
• References
• Copyright

Abstract 

Weintraub MI, Herrmann DN, Smith AG, Backonja MM, Cole SP. Pulsed electromagnetic fields to reduce diabetic neuropathic pain and stimulate neuronal repair: a randomized controlled trial.

Objective

To determine whether repetitive and cumulative exposure to low-frequency pulsed electromagnetic fields (PEMF) targeting painful feet can reduce neuropathic pain (NP), influence sleep in symptomatic diabetic peripheral neuropathy (DPN), and influence nerve regeneration.

Design

Randomized, double-blind, placebo-controlled parallel study.

Setting

Sixteen academic and clinical sites in 13 states.

Participants

Subjects (N=225) with DPN stage II or III were randomly assigned to use identical devices generating PEMF or sham (placebo) 2 h/d to feet for 3 months.

Interventions

Nerve conduction testing was performed serially.

Main Outcome Measures

Pain reduction scores using a visual analog scale (VAS), the Neuropathy Pain Scale (NPS), and the Patient's Global Impression of Change (PGIC). A subset of subjects underwent serial 3-mm punch skin biopsies from 3 standard lower limb sites for epidermal nerve fiber density (ENFD) quantification.

Results

Subjects (N=225) were randomized with a dropout rate of 13.8%. There was a trend toward reductions in DPN symptoms on the PGIC, favoring the PEMF group (44% vs 31%; P=.04). There were no significant differences between PEMF and sham groups in the NP intensity on NPS or VAS. Twenty-seven subjects completed serial biopsies. Twenty-nine percent of PEMF subjects had an increase in distal leg ENFD of at least 0.5 SDs, while none did in the sham group (P=.04). Increases in distal thigh ENFD were significantly correlated with decreases in pain scores.

Conclusions

PEMF at this dosimetry was noneffective in reducing NP. However neurobiological effects on ENFD, PGIC and reduced itching scores suggest future studies are indicated with higher dosimetry (3000–5000 G), longer duration of exposure, and larger biopsy cohort.

Key Words: Electromagnetic fields, Rehabilitation

List of Abbreviations: DPN, diabetic peripheral neuropathy, ENFD, epidermal nerve fiber density, HbA1C, glycosylated hemoglobin, IGF-I, insulin-like growth factor I, IGF-II, insulin-like growth factor II, NGF, nerve growth factor, NP, neuropathic pain, NPS, Neuropathy Pain Scale, PEMF, pulsed electromagnetic fields, PGIC, Patient's Global Impression of Change, VAS, visual analog scale, VEGF, vascular endothelial growth factor

REPETITIVE TRANSCRANIAL magnetic stimulation at the prefrontal,¹ motor,² and somatosensory cortex³ is emerging as a promising alternative therapy for disabling and refractory NP. Short-term analgesic and antinociceptive effects have also been achieved with direct stimulation of the spinal cord⁴ and lumbar nerve roots.⁵ Both low-frequency and high-frequency magnetic stimulation can influence thermal and pain thresholds in both normative and symptomatic subjects for a short time, yet the specific mechanisms of action are yet to be determined.⁶, ⁷, ⁸, ⁹, ¹⁰ Despite these preliminary data with small cohorts receiving isolated treatments only at academic clinics, there has been no information regarding its efficacy in painful DPN, which is one of the most common causes of NP. It has been estimated that 40% to 50% may experience NP.¹¹ DPN begins insidiously in the feet with preferential involvement of unmyelinated C fibers and small myelinated A delta fibers.¹² From a pathophysiological standpoint, DPN symptoms are believed secondary to ectopic firing of nociceptive afferent axons that are undergoing degeneration, with dysregulated expression of sodium, calcium, and potassium channels.¹³, ¹⁴, ¹⁵ Skin biopsies reveal prominent cutaneous denervation with length-dependent reductions in ENFD.¹⁶, ¹⁷ The mechanisms of DPN and NP are considered multifactorial.¹⁸ Impaired production of neurotrophic factors (NGF, IGF-I, IGF-II, fibroblast growth factor, and so forth),¹⁹, ²⁰, ²¹ impaired Schwann cells,¹⁹, ²² macrophage dysfunction,¹⁹, ²³ microangiopathy with ischemia and reduced VEGF,¹⁹, ²⁴ impaired voltage-gated channels (sodium, potassium, calcium),¹³, ¹⁵, ²⁵ protein kinase C dysregulation,¹⁹, ²⁶ and oxidative stress¹⁹, ²⁷, ²⁸ are believed to be contributory. Data from cell culture, animal, and human studies suggest that exogenous application of weak, nonthermal electromagnetic fields upregulates NGF, IGF-I, IGF-II, fibroblast growth product, and VEGF²⁹, ³⁰, ³¹; reorients Schwann cells³²; enhances macrophage activity³³ and endoneurial blood flow³⁴; reduces nociceptive afferent signal transduction³⁵, ³⁶, ³⁷, ³⁸; reduces free radicals³⁷, ³⁹ and oxidative stress³³, ⁴⁰; and promotes neurite outgrowth.³⁵, ⁴¹ Thus, magnetic stimulation may be an appropriate noninvasive intervention that could reduce DPN symptoms and produce disease modification.³⁵, ³⁷

Back to Article Outline

Methods 

Enrollment Criteria 

The design and conduct of the randomized controlled trial is described in the accompanying consort flow diagram (fig 1). Subjects from 18 to 87 years of age with painful DPN (Dyck stage II or III)³⁸ with moderate-severe constant pain of 4 or higher on a 0 to 10 VAS, with a duration of at least 6 months, were recruited at 16 investigative sites in 13 states within the United States (appendix 1) between August 2005 and March 2007. Pregnant women and subjects with mechanical insulin pumps or cardiac pacemakers were excluded. Subjects could remain on their stable drug medications for diabetes and pain relief, but no new analgesics or dosing increases were permitted during the trial. Subjects were enrolled only if they were on a stable analgesic regimen. Before randomization, subjects were instructed on how to tabulate VAS (0–10) pain scores (3 times a day) and a sleep interference score (VAS 0–10, once daily). All participants provided written informed consent. Two university centers performed skin-punch biopsies at randomization and at conclusion of the study that were shipped to the University of Rochester for immunohistochemistry and measurement of ENFD.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  The CONSORT diagram revealing enrollment and outcomes. A total of 245 subjects were screened, and 225 were randomized and enrolled. A 13.8% dropout occurred (31/225) with no safety issues.

Randomization 

Demographic data (age, height, weight, sex, glycosylated hemoglobin [HbA1C], family history, duration of diabetes, concomitant medications) were collected for each enrolled subject. After entry and baseline quantification of pain and sleep interruption scores, eligible patients were randomized (1:1 via computer assignment) to receive an active coded magnetized or a sham device, identical in all characteristics except for the demagnetization procedure. Subjects agreed to use the device a maximum of 2 hours a day in divided sessions of 10 to 30 minutes for 3 months. Subjects recorded daily VAS pain and sleep scores; other outcome measures (see below) were evaluated at monthly study visits. All subjects agreed not to break the blinding of the devices. A consecutive subset of patients from 2 university sites volunteered to participate in an ENFD exploratory substudy. Three-millimeter punch skin biopsies were harvested from the proximal and distal lateral thigh, and the distal leg at baseline and after 3 months of PEMF or sham exposure. The skin biopsies were fixed, cryoprotected, sectioned, and immunostained with polyclonal antibodies to the panaxonal marker, protein gene product 9.5, according to previously published methods.⁴², ⁴³ A single blind observer assessed both the linear density (fibers/mm) of nerve fibers crossing the dermal-epidermal junction ENFD (crossings) and the total linear density including intraepidermal fragments ENFD (total) from three to five 50-µM thick sections selected at random from each biopsy specimen, using previously published techniques.⁴⁴, ⁴⁵

Device 

This device (fig 2) uses 6 individual (1800-G) magnetic sphere units, 3 under each foot, that are driven individually by a 6-V DC motor. A speed control circuit allows a range of 500 to 1500 revolutions per minute. The magnetic spheres turn on one axis generating magnetic lines of force (flux), and simultaneously, turn on a second axis perpendicular to the first axis (biaxial), causing the moving flux lines to cut across tissues at varying periodic angles, inducing varying intensities of force and polarity changes, resulting in static and time-varying magnetic fields. Precise placement on the foot plates with socks allows penetration of the magnetic fields up to 5 feet as measured by an ENF meter^b (model UHS). There is exponential decay of field strength with distance from magnetic source (310,000 mG). Supersaturation of the target area from every angle at 25 times a second at maximum 1500 revolutions per minute is achieved. The barium ferrite-nylon bonded spheres do not induce any discernable sensory effects on the feet to suggest device activity.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Device: Body Energizer.^a (A), PEMF and/or identical sham demagnetized device. (B), Subject with feet in place.

Outcome Measures 

Pain is a complex experience, and none of the existing pain scales appears to be ideal for all situations. Thus, we chose a priori to employ 3 of the most commonly used validated NP measures as outcomes for the trial.

The primary outcome was a VAS (ranging from 0, no pain to 10, worst possible pain),³⁸ 3 times daily at the same time to represent a mean daily pain level.

NPS assessed 10 pain descriptors collected at baseline and the end of the study.⁴³ NPS composite (NPS 10) scores range from 0 to 100.

PGIC⁴⁶ required subjects to select 1 of 7 options describing response to treatment, ranging from “very much improved” to “very much worse.”

VAS measure of sleep disruption³⁸ secondary to pain was collected once on arising each morning.

Other secondary outcomes compared baseline and 12-week values of the neurologic examination (sensory, motor, reflex functions). Standardized nerve conduction velocities and amplitudes of common peroneal nerve (recording from the extensor digitorum brevis muscle) and sural nerves were monitored at baseline and the end of the study for abnormalities consistent with distal polyneuropathy. At the end of the study, both patients and investigators were asked for their perception of device activity.

Statistical Analyses 

Based on prior pilot VAS pain data,³⁸ a sample size of 200 patients was calculated to yield a power of 80% to detect a 25% superiority of PEMF over sham placebo with alpha equal to 0.05 and beta equal to 0.20. We allowed for a dropout rate of 20% of subjects enrolled. For the NPS, 10 composite scores (range, 0–100) were used. In addition, 2 NPS items most salient to C-fiber involvement, itchy pain and burning pain (ranges, 0–10), were analyzed separately. ENFD change scores were computed by subtracting the baseline value from the 3-month value; a positive change score indicated an increase in ENFD. Change scores as continuous measures were used for correlation analyses. To assess treatment effects on ENFD, 3 categories were constructed based on a 0.5 SD of the baseline value: (1) >0.5 SD change (indicating increase in ENFD), (2) −0.5 to 0.5 (no or little change) and (3) > −0.5 SD change (decrease). The 0.5 SD criterion was chosen to be sensitive to the different levels of variability of the ENFD measures. Associations between treatment and ENFD groups were assessed with chi-square tests.

Two (PEMF, sham) × 2 (baseline to month 3) repeated-measures analyses of variance were used to assess change in pain scores and ENFD values over the course of the study. A statistically significant treatment group × time interaction indicated greater change from baseline to the end of month 3 for 1 of the treatment groups. Independent sample t tests were used to test for possible baseline differences in mean scores and for the PGIC at 3 months.

For the a priori statistical tests of the primary outcome measure, the level of significance was set at P<.05. For the 3 secondary outcome measures, a Bonferroni correction adjusted the statistical significance level to .017. For the Pearson product moment correlation analyses between ENFD values and pain measures, the researchers controlled for familywise error rate using a sequential Bonferroni approach: significance was set at P<.008. All tests were 2-sided. All analyses were conducted in an intent-to-treat manner (expectation maximization method). The Statistical Package for the Social Sciences (version 15.0^c) was used to analyze the data.

Back to Article Outline

Results 

The flow of patients through the clinical trial is depicted in figure 1 (CONSORT diagram). Of the 245 subjects enrolled in this study, 20 cases were initially excluded because of a low baseline score. Of the 225 patients randomized, there was a dropout of 31 subjects (13.8%). These included 5 because of protocol violations, 6 from diabetic complications, 16 lost to follow-up, and 4 who did not complete the study because of allodynia. Three of these 4 cases had significant premorbid burning feet syndrome with pressure allodynia. Of the 107 patients allocated to the magnet group, 90 (84.1%) completed the 3-month study, whereas 104 of the 118 allocated to the sham group (88.1%) completed the study. The dropout rate and withdrawal pattern were similar for both groups. Baseline demographics (table 1) were similar for both groups. Women represented 56.7% of the PEMF group and 55.8% of the sham group. Mean ± SD ages were 63.6±8.6 years and 63.5±9.5 years for the PEMF and sham groups, respectively. HbA1c data were similar for both groups at 3 months. There were also no changes in motor or sensory conductions or the sensory/motor neurologic examination at 3 months. Seventy-four percent of patients who completed the study took at least 1 analgesic medication for pain, and 47% took at least 2 agents. There were no group differences in number of antiepileptic drugs, narcotics, tricyclics, selective serotonin reuptake inhibitors, or nonsteroidal anti-inflammatory drug medications taken by patients.

Table 1. Baseline Demographics and Clinical Characteristics of the Enrolled Patients

Characteristic	PEMF Group (n=90)	Sham Group (n=104)
Age (y)
Mean	61.1±10.4	60.6±12.4
Range	33–87	21–83
Weight (lb)	217.9±55.6	215.1±54.6
Height (in)	66.6±4.54	67.4±4.42
Female (%)	56.7	55.8
Years since onset of diabetes	3.9±3.0	4.0±3.0
HbA1c	7.5±1.8	7.4±1.8
Subjects with abnormal nerve conduction (%)	87.7	89.9

NOTE. Values are mean ± SD unless otherwise noted.

For the biopsy cohort (CONSORT diagram) (fig 3), of the 37 subjects enrolled in the study, 10 (27.0%) were lost to follow-up (3 magnet, 7 sham). Of the remaining 27 cases, 14 had received active magnets, and 13 had received sham devices. Women represented 64.3% of the PEMF group and 38.5% of the sham group. Mean ages were 63.6 and 63.5 years for the PEMF and sham groups, respectively.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Biopsy consort.

Primary and Secondary Outcomes 

Results for study outcomes are presented in table 2. There were no statistically significant group differences in baseline pain measures. For PGIC at 3 months, 43.7% of PEMF and 30.6% of sham group subjects reported very much or much improvement (P=.04). This result was considered a nonsignificant trend. Group differences from baseline to 3 months were not significant for VAS (P=.96), NPS 10 (P=.58), sleep scores (P=.49), or electrodiagnostic studies. Analyses controlling for baseline HbA1c (PEMF mean,7.5; sham mean,7.4) and whether subjects were taking insulin (10% PEMF; 28% sham) also did not reveal significant group differences. However, for subjects with moderate to severe itchy pain, there was a 53.7% reduction in mean itchy pain scores for the PEMF group from baseline to 3 months, whereas there was a 33.8% reduction for the sham group (P=.048). Subjects who reported higher levels of itching also reported higher levels of burning at baseline (r=.32; P<.001) and at 3 months (r=.33; P<.001).

Table 2. Results of Pain and Sleep Scales

NOTE. Values are mean ± SD unless otherwise noted.

Correlations Between Pain Measures 

At baseline, the correlations between NPS 10 (only the total composite score was analyzed) and VAS was significant (r=.57; P<.001). At 3 months, there were significant correlations between NPS 10 and VAS (r=.77; P<.001). NPS 10 and PGIC (r=.50; P<.001), and VAS and PGIC (r=.53; P<.001).

Biopsy Study 

There were no statistically significant group differences in baseline ENFD measures. At the distal leg site, there was a nonsignificant trend for an increase in mean ± SD ENFD (crossings) from baseline (1.33±2.04) to 3 months (1.56±2.34) for the PEMF group, while there was a decrease in ENFD crossings from baseline (1.05±1.64) to 3 months (0.83±1.54) for the sham group (P=.10). Similarly, there was a nonsignificant trend for an increase in ENFD total from baseline (1.83±2.93) to 3 months (2.21±3.43) for the PEMF group, while there was a decrease in ENFD total from baseline (1.28±2.10) to 3 months (1.03±1.99) for the sham group (P=.08). At the distal leg site, 4 (28.6%) of the magnet group and none of the sham group had greater than 0.5 SD increase in ENFD crossings (χ² P value=.04; Fisher exact test=.07) (fig 4). No significant group differences were noted between baseline and 3-month values for ENFD (crossings) and ENFD (total) at the distal and proximal thigh biopsy sites. At the distal thigh, Pearson correlation coefficients for all 27 cases revealed moderate associations between 3-month PGIC scores and changes in ENFD crossings (r=−.40; P=.04) and changes in ENFD total (r=−.41; P=.04); higher nerve density was related to global improvement. Over the 3 months, an increase in distal thigh ENFD crossings was moderately associated with a decrease in NPS 10 scores (r=−.49; P=.010); an increase in distal thigh ENFD total was significantly associated with a decrease in NPS 10 scores (r=−.53, P=.006) (fig 5). There were no significant correlations between changes in distal leg or proximal thigh ENFD and VAS scores.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  Number of ENFD distal leg crossings in (A) PEMF and (B) sham groups. Dark lines indicate increase in density from baseline to 3 months; light lines indicate decrease or no change.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 5. 

  Scatterplot of total distal thigh ENFD: increase in ENFD was significantly associated with a decrease in NPS 10 (difference) scores (r=−.53; P=.006).

Blinding 

At the end of the study, the perception of patients and physicians, regarding device activity was erroneous in 20% of the PEMF group and 26% of the sham group. In the absence of objective changes in neurologic examination and conduction studies, physician investigators tended to agree with the responses of their patients.

Safety 

There were no safety issues or complications except that 4 cases experienced allodynia leading to dropout (sham=PEMF).

Back to Article Outline

Discussion 

To our knowledge, this is the first multicentered, randomized, double-blind, placebo-controlled trial of cumulative exposure of PEMF targeting painful feet in subjects with NP from DPN. The results indicate that the key outcomes related to change in pain or sleep disruption were not improved by PEMF. However, there are some provocative data suggesting that neurobiological changes occurred in the epidermal innervation exploratory substudy. First, PEMF appeared to affect DPN symptoms, despite the enrollment of patients with relatively advanced DPN (Dyck stage II or III), of whom many subjects were markedly deafferented, particularly at the distal leg site (table 3). Mean ± SD ENFD total for the PEMF cohort at baseline was 1.83±2.93 (normative, 16.6±5.3).⁴⁷ This suggests that cutaneous deafferentation does not preclude a beneficial effect of PEMF on NP. Second, we observed no deleterious effect of 12 weeks of PEMF on ENFD, indicating that any effects of PEMF on DPN symptoms are not mediated via injury to nociceptive afferents. Third, we found that 29% of those receiving active PEMF showed at least a 0.5 SD increase in ENFD between the pretreatment and posttreatment time points at the distal leg skin site, while none of the sham group demonstrated such an increase. The exact significance of these changes in ENFD is uncertain and should be cautiously interpreted because of the small cohort size, but it suggests the possibility of a regenerative effect. It was unfortunate that almost one third of subjects failed to return for second biopsy. The significance of reduced itchy pain scores is also unclear but was felt to represent a C-fiber function.

Table 3. Epidermal Nerve Fiber Density at Baseline and 3 Months

	PEMF		Sham
Measure	Baseline	Month 3	Baseline	Month 3
Distal leg
Crossings	1.33±2.04	1.56±2.34	1.05±1.64	0.83±1.54
Total	1.83±2.94	2.21±3.43	1.28±2.10	1.03±1.99
Distal thigh
Crossings	5.00±1.68	4.76±2.21	4.49±1.40	4.44±2.30
Total	6.51±2.44	6.26±2.91	5.95±2.08	6.27±3.14
Proximal thigh
Crossings	7.32±3.11	7.18±2.06	6.58±1.83	7.03±2.41
Total	9.12±3.81	10.28±3.06	8.69±2.83	9.43±3.19

NOTE. Values are mean ± SD.

There are several strengths of this study. These include a large homogeneous cohort with stage II and stage III DPN. Additional strengths include the use of 3 validated pain scoring methods representing a composite of the pain experience. The high rate of study completion supports device tolerability. The inclusion of a biologic endpoint (ENFD) in a subset as another measure of neurologic safety is a strength.

Study Limitations 

It is difficult to blind subjects reliably given the ease of detecting the presence of magnetism. We believe the placebo effect was as fully controlled as possible using an inert, nonactive demagnetized sham device rather than a weak magnet because biological responses have been reported. At completion of the study, the PEMF subjects (48%) reported not knowing whether they had an active or sham device; 32% believed they had an active while 20% believed they had a sham device. For the sham subjects, 56% reported not knowing whether they had an active or sham device; 26% believed they had an active device while 18% believed they had a sham device. Another limitation is that the pain reduction was reflected only in PGIC pain scales and was not significantly different in 3 of the 4 other outcome measures. This could suggest that PEMF may be influencing other aspects of neuropathic dysfunction such as paresthesiae, dysesthesiae, itching, burning, and so forth. Andre-Obadia et al², ⁴⁸ believe that pain scores after stimulation are variable and inconsistent, with their reliability increasing in the subsequent 3 to 4 days. Thus the PGIC data reflecting a cumulative response may be more meaningful than VAS and NPS.⁴⁹, ⁵⁰ Last, the specific structures potentially influenced in the microenvironment and specific tissue dosimetry at target areas also remain unknown.

Back to Article Outline

Conclusions 

This randomized controlled trial failed to demonstrate a positive effect on pain modulation at this current dosimetry and duration of exposure. However, the potential neurobiologic effects noted from PGIC and skin biopsy data (ENFD) suggest that future studies using a higher dosimetry (3000–5000 G) with a longer duration of exposure and a larger biopsy cohort is warranted to determine whether NP can be modulated by PEMF⁵¹ and influence nerve regeneration.

Suppliers

Back to Article Outline

Appendix 1: Investigators 

The site investigators are listed alphabetically with the principal investigator listed first.

Back to Article Outline

References 

1. Borckardt JJ, Smith AR, Reeves ST, et al. Fifteen minutes of left prefrontal repetitive transcranial magnetic stimulation acutely increases thermal pain thresholds in healthy adults. Pain Res Manag. 2007;12:287–290
   • View In Article
2. Andre-Obadia N, Mertens P, Gueguen A, et al. Pain relief by rTMS (Differential effect of current flow but no specific action on pain subtypes). Neurology. 2008;71:833–840
   • View In Article
   • CrossRef
3. Topper R, Foltys H, Meister IG, et al. Repetitive transcranial magnetic stimulation of the parietal cortex transiently ameliorates phantom limb pain-like syndrome. Clin Neurophysiol. 2003;114:1521–1530
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (381 KB)
   • CrossRef
4. Schlaier JR, Eichhammer P, Langguth B, et al. Effects of spinal cord stimulation on cortical excitability in patients with chronic neuropathic pain: a pilot study. Eur J Pain. 2007;11:863–868
   • View In Article
   • CrossRef
5. Lin VW, Hsiao I, Kingery WS. High intensity magnetic stimulation over the lumbosacral spine evokes antinociception in rats. Clin Neurophysiol. 2002;113:1006–1012
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (159 KB)
   • CrossRef
6. Summers J, Johnson S, Pridmore S, Oberoi G. Changes to cold detection and pain thresholds following low and high frequency transcranial magnetic stimulation of the motor cortex. Neurosci Lett. 2004;368:197–200
   • View In Article
   • MEDLINE
   • CrossRef
7. Leo RJ, Latif T. Repetitive transcranial magnetic stimulation (rTMS) in experimentally induced and chronic neuropathic pain: a review. J Pain. 2007;8:453–459
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (134 KB)
   • CrossRef
8. Khedr EM, Korb H, Kamel NF, et al. Longlasting antalgic effects of daily sessions of repetitive transcranial magnetic stimulation in central and peripheral neuropathic pain. J Neurol Neurosurg Psychiatry. 2005;76:833–838
   • View In Article
   • MEDLINE
   • CrossRef
9. Lefaucheur JP. New insights into the therapeutic potential of non-invasive transcranial cortical stimulation in chronic neuropathic pain. Pain. 2006;122:11–13
   • View In Article
   • Full Text
   • Full-Text PDF (58 KB)
   • CrossRef
10. Lefaucheur JP, Drouot X, Menard-Lefaucheur I, et al. Neurogenic pain relief by repetitive magnetic cortical stimulation depends on the origin and the site of pain. J Neurol Neurosurg Psychiatry. 2004;75:612–616
    • View In Article
    • MEDLINE
    • CrossRef
11. Veves A, Backonja M, Malik RA. Painful diabetic neuropathy: epidemiology, natural history, early diagnosis and treatment options. Pain Med. 2008;9:660–674
    • View In Article
    • CrossRef
12. Ørstavik J, Namer B, Schmidt R, et al. Abnormal function of C-fibers in patients with diabetic neuropathy. J Neurosci. 2006;26:11287–11294
    • View In Article
13. Waxman SG. The molecular pathophysiology of pain: abnormal expression of sodium channel genes and its contributions to hyperexcitability of primary sensory neurons. Pain. 1999;(Suppl 6):S133–S140
    • View In Article
14. Cummins TR, Sheets PL, Waxman SG. The role of sodium channels in nociception: implications for mechanisms of pain. Pain. 2007;131:243–257
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (415 KB)
    • CrossRef
15. Quasthoff S. The role of axonal ion conductances in diabetic neuropathy: a review. Muscle Nerve. 1998;21:1246–1255
    • View In Article
    • MEDLINE
    • CrossRef
16. Smith AG, Ramachandran P, Tripp S, Singleton JR. Epidermal nerve innervation in impaired glucose tolerance and diabetic-associated neuropathy. Neurology. 2001;57:1701–1704
    • View In Article
    • MEDLINE
17. Polydefkis M, Griffin JW, McArthur J. New insights into diabetic polyneuropathy. JAMA. 2003;290:1371–1376
    • View In Article
    • CrossRef
18. Obrosova IG. Update on the pathogenesis of diabetic neuropathy. Curr Diab Rep. 2003;3:439–445
    • View In Article
    • MEDLINE
    • CrossRef
19. Kennedy JM, Zochodne DW. Impaired peripheral nerve regeneration in diabetes mellitus. J Peripher Nerv Syst. 2005;10:144–157
    • View In Article
    • MEDLINE
    • CrossRef
20. Ishii DN. Implication of insulin-like growth factors in the pathogenesis of diabetic neuropathy. Brain Res Brain Res Rev. 1995;20:47–67
    • View In Article
21. Jungnickel J, Gransalke K, Timmer M, Growthe C. Fibroblast growth factor receptor 3 signaling regulates injury-related effects in the peripheral nervous system. Mol Cell Neurosci. 2004;25:21–29
    • View In Article
    • MEDLINE
    • CrossRef
22. Ekersley L, Ansselin AD, Tomlinson DR. Effects of experimental diabetes on axonal and Schwann cell changes in sciatic nerve isografts. Brain Res Mol Brain Res. 2001;92:128–137
    • View In Article
    • MEDLINE
    • CrossRef
23. Miyauchi A, Kanje M, Danielsen N, Dahlin LV. Role of macrophages in the stimulation and regeneration of sensory nerves by transposed granulation tissue and temporal aspects of the response. Scan J Plast Reconstr Surg. 1997;31:17–23
    • View In Article
24. Veves A, King GL. Can VEGF reverse diabetic neuropathy in human subjects?. J Clin Invest. 2001;107:1215–1218
    • View In Article
    • MEDLINE
    • CrossRef
25. Devor M. Sodium channels and mechanisms of neuropathic pain. J Pain. 2006;7(Suppl 1):S3–S12
    • View In Article
    • Full-Text PDF (43 KB)
    • CrossRef
26. Eichberg J. Protein kinase C changes in diabetes: is the concept relevant to neuropathy?. Int Rev Neurobiol. 2002;50:61–82
    • View In Article
    • MEDLINE
    • CrossRef
27. Vincent AM, Russell JW, Low P, Feldman EL. Oxidative stress in the pathogenesis of diabetic neuropathy. Endocr Rev. 2004;25:612–628
    • View In Article
    • MEDLINE
    • CrossRef
28. Pilla AA, Muehsam DJ, Markov MS, Sisken BF. EMF signals and ion/ligand binding kinetics: predictions of bioeffective waveform parameters. Bioelectrochem Bioenerg. 1999;48:27–34
    • View In Article
    • MEDLINE
    • CrossRef
29. Longo FM, Yang T, Hamilton S, et al. Electromagnetic fields influence NGF activity and levels following sciatic nerve transection. J Neurosci Res. 1999;55:230–237
    • View In Article
    • MEDLINE
    • CrossRef
30. Fitzsimmons RJ, Ryaby JT, Mohan S, Magee FP, Baylink DJ. Combined magnetic fields increase insulin-like growth factors-II in TE-85 human osteosarcoma dome cell cultures. Endocrinology. 1995;136:3100–3106
    • View In Article
    • MEDLINE
    • CrossRef
31. Rodeman HP, Bayrenter K, Pfleiderer G. The differentiation of normal and transformed human fibroblast in vitro is influenced by electromagnetic fields. Exp Cell Res. 1989;189:610–621
    • View In Article
32. Eguchi Y, Ogiue-Ikeda M, Ueno S. Control of orientation of rat Schwann cells using an 8-T static magnetic field. Neurosci Lett. 2003;351:130–132
    • View In Article
    • MEDLINE
    • CrossRef
33. Frahn J, Lantow M, Lupke M, et al. Alteration in cellular functions in mouse macrophages after exposure to 50 Hz magnetic fields. J Cell Biochem. 2006;1:168–177
    • View In Article
34. Zhao M, Bai H, Wang E, Forrester JV, McCaig CD. Electrical stimulation directly induces pre-angiogenic responses in vascular endothelial cells by signaling through VEGF receptors. J Cell Sci. 2004;117:397–405
    • View In Article
    • MEDLINE
    • CrossRef
35. Markov MS. Magnetic field therapy: a review. Electromagn Biol Med. 2007;26:1–23
    • View In Article
    • MEDLINE
    • CrossRef
36. McLean MJ, Holcomb RR, Wamil AW, PIckett JD, Cavapol AV. Blockade of sensory action potentials by a static magnetic field in the 10 mT range. Bioelectromagnetics. 1995;16:20–32
    • View In Article
    • MEDLINE
    • CrossRef
37. Pilla AA. Mechanisms and therapeutic applications of time-varying and static magnetic fields. In:  Barnes F,  Greenebaum B editor. Handbook of biological effects of electromagnetic fields. 3rd ed.. CRC Pr; 2006;
    • View In Article
38. Weintraub MI, Wolfe GI, Barohn RA, et al. Static magnetic field therapy for symptomatic diabetic neuropathy: a randomized, double-blind, placebo-controlled trial. Arch Phys Med Rehabil. 2003;84:736–746
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (268 KB)
    • CrossRef
39. Timmel CR, Till U, Brocklehurst B. Effects of weak magnetic fields on free radical recombination reactions. Mol Phys. 1998;95:71–89
    • View In Article
40. Cantoni O, Sestili P, Fiorani M, Dacha M. The effect of 50 Hz sinusoidal electric and/or magnetic fields on the rate of repair of DNA single/double strand breaks in oxidatively injured cells. Biochem Mol Biol Int. 1995;37:681–689
    • View In Article
    • MEDLINE
41. Sisken BF. Effects of pulsed magnetic fields on neurite outgrowth from chick embryos. Bioelectromagnetics. 1996;17:293–302
    • View In Article
    • MEDLINE
    • CrossRef
42. Kennedy WR, Wendelschafer-Crabb G, Johnson T. Quantitation of epidermal nerves in diabetic neuropathy. Neurology. 1996;47:1042–1048
    • View In Article
    • MEDLINE
43. McArthur JC, Stocks EA, Hauer P, Cornblath DR, Griffin JW. Epidermal nerve fiber density: normative reference range and diagnostic efficiency. Arc Neurol. 1998;55:1513–1520
    • View In Article
44. Herrmann DN, Griffin JW, Hauer P, Cornblath DR, McArthur JC. Epidermal nerve fiber density and sural nerve morphometry in peripheral neuropathies. Neurology. 1999;53:1634–1640
    • View In Article
    • MEDLINE
45. Kennedy WR, Wendelschafer-Crabb G, Polydefkis M, McArthur JC. Pathology and quantitation of cutaneous innervation. In: 4th ed..  Dyck PJ,  Thomas PK editor. Peripheral neuropathy. 1:Elsevier: Saunders; 2005;p. 869–895
    • View In Article
46. Jensen MP, Friedman M, Bonzo D, Richards P. The validity of the Neuropathic Pain Scale for assessing diabetic neuropathy pain in a clinical trial. Clin J Pain. 2006;22:97–103
    • View In Article
    • MEDLINE
    • CrossRef
47. Herrmann DN, McDermott MP, Henderson D, et al. Epidermal nerve fiber density, axonal swellings and QST as predictors of HIV distal sensory neuropathy. Muscle Nerve. 2004;29:420–427
    • View In Article
    • MEDLINE
    • CrossRef
48. Andre-Obadia N, Peyron R, Mertens P, et al. Transcranial magnetic stimulation for pain control: double-blind study of different frequencies against placebo, and correlation with motor cortex stimulation efficacy. Clin Neurophysiol. 2006;117:1536–1544
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (199 KB)
    • CrossRef
49. Gordh TE, Stubhaug A, Jensen TS, et al. Gabapentin in traumatic nerve injury pain: a randomized, double-blind, placebo-controlled, cross-over, multi-center study. Pain. 2008;138:255–266
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (192 KB)
    • CrossRef
50. Dworkin RH, Turk DC, Farrar JT, et al. Core outcome measures for chronic pain clinical trial: IMMPACT recommendations. Pain. 2005;113:9–19
    • View In Article
    • Full Text
    • Full-Text PDF (156 KB)
    • CrossRef
51. Fregni F, Freedman S, Pasqual-Leone A. Recent advances in the treatment of chronic pain with non-invasive brain stimulation techniques. Lancet. 2007;6:188–191
    • View In Article
    • CrossRef

• a Body Energizer; Nu-Magnetics, Inc, Box 572, Port Jefferson, NY 11777-1025.
• b ENF meter, Model UHS; Alpha Lab, Inc, 3005 South 300 West, Salt Lake City, UT 84115.
• c SPSS version 15.0; SPSS, Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.