Introduction

Diabetic peripheral neuropathy (DPN) is one of the most common microvascular complications of both type 1 and type 2 diabetes. Depending on the criteria used, it occurs in 5–100% of diabetic patients <sup>[1 Joslinʼs diabetes mellitus Philadelphia: Lippincott Wiliams & Wilkins (2005). 

Cliquez ici pour aller à la section Références]</sup>. The most common type of DPN is distal diabetic polyneuropathy with symmetrical involvement of sensory and/or motor nerves. Patients with this complication may report no symptoms or complain of numbness and other sensation disturbances <sup>[2Vinik A.I., Park T.S., Stansberry K.B., Pittenger G.L. Diabetic neuropathies Diabetologia 2000 ;  43 : 957-973 [cross-ref]

Cliquez ici pour aller à la section Références, 3Nordin M., Nyström B., Wallin U., Hagbarth K.E. Ectopic sensory discharges and paresthesiae in patients with disorders of peripheral nerves, dorsal roots and dorsal columns Pain 1984 ;  20 : 231-245 [cross-ref]

Cliquez ici pour aller à la section Références and 4Brownlee M. Biochemistry and molecular cell biology of diabetic complications Nature 2001 ;  414 : 813-822 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. The main clinical problem is pain that has a major effect on the quality of life and can even lead to depression and suicide attempts in extreme cases. Current therapy for painful diabetic neuropathy (PDN) is mainly the symptomatic use of typical analgesics (nonsteroidal anti-inflammatory drugs [NSAID] or opioids) or coanalgesics (such as antidepressants and antiepileptics) <sup>[5Sindrup S.H., Jensen T.S. Efficacy of pharmacological treatments of neuropathic pain: an update and effect related to mechanism of drug action Pain 1999 ;  83 : 389-400 [cross-ref]

Cliquez ici pour aller à la section Références, 6Dooley D.J., Meder W.P., Suman-Chauhan N., Duffy S., Clussman H., Gothert M. Inhibition of neuronal Ca(2+) influx by gabapentin and pregabalin in the human neocortex Neuropharmacology 2002 ;  42 : 229-236

Cliquez ici pour aller à la section Références and 7Kumar D., Alvaro M.S., Julka I.S., Marshall H.J. Diabetic peripheral neuropathy. Effectiveness of electrotherapy and amitriptyline for symptomatic relief Diabetes Care 1998 ;  21 : 1322-1325 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. As a causal treatment, alpha-lipoic acid appears to be the most effective <sup>[8Ametov A.S., Barinov A., Dyck P.J., Hermann R., Kozlova N., Litchy W.J. The sensory symptoms of diabetic polyneuropathy are improved with alpha-lipoic acid: the SYDNEY trial Diabetes Care 2003 ;  26 : 770-776 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. However, despite a wide range of pharmacological agents, effective analgesic treatment of PDN is still a challenge. Various nonpharmacological symptomatic treatments of PDN have been attempted, including homoeopathy, acupuncture, low-intensity laser therapy as well as static and pulsed magnetic fields <sup>[9Abuaisha B.B., Costanzi J.B., Boulton A.J. Acupuncture for the treatment of chronic painful peripheral diabetic neuropathy: a long-term study Diabetes Res Clin Pract 1998 ;  39 : 115-121 [cross-ref]

Cliquez ici pour aller à la section Références, 10Zinman L.H., Ngo M., Ng E.T., Nwe K.T., Gogov S., Bril V. Low-intensity laser therapy for painful symptoms of diabetic sensorimotor polyneuropathy: a controlled trial Diabetes Care 2004 ;  27 : 921-924 [cross-ref]

Cliquez ici pour aller à la section Références and 11Weintraub M.I., Wolfe G.I., Barohn R.A., Cole S.P., Parry G.J., Hayat G., et al. Magnetic Research Group. Static magnetic field therapy for symptomatic diabetic neuropathy: a randomized, double-blind, placebo-controlled trial Arch Phys Med Rehabil 2003 ;  84 : 736-746 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. Of the physical methods, low frequency pulsed magnetic fields (PMF) are currently of clinical interest <sup>[12Weintraub M.I., Cole S.P. Pulsed magnetic field therapy in refractory neuropathic pain secondary to peripheral neuropathy: electrodiagnostic parameters-pilot study Neurorehabil Neural Repair 2004 ;  18 : 42-46 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>.

The effect of low frequency magnetic fields on human tissues has been reported in many studies. PMF have analgesic, vasoactive, neurostimulatory and trophic effects, among others, in humans <sup>[13Musaev A.V., Guseinova S.G., Imamverdieva S.S. The use of pulsed electromagnetic fields with complex modulation in the treatment of patients with diabetic polyneuropathy Neurosci Behav Physiol 2003 ;  33 : 745-752 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. By inducing low frequency currents, PMF can depolarize, repolarize and hyperpolarize neurons and, in this way, modulate neuropathic pain. The influence of low frequency magnetic fields on pain intensity has not been objectively examined so far in patients with painful diabetic polyneuropathy. Indeed, for years, diabetes has been considered a contraindication for magnetic therapy mainly due to a lack of research into its effects on glucose metabolism. However, the results of recent trials show positive effects of such therapy on glucose utilization <sup>[14Banaszkiewicz W., Straburzyński G. Influence of pulsed magnetic fields on selected parameters of metabolic control and acid: alkaline balance Baln Pol 1992 ;  34 : 109-119

Cliquez ici pour aller à la section Références]</sup>.

The aim of the present randomized, placebo-controlled, double-blind study was to assess whether magnetic fields influence pain intensity, the quality of life and sleep, and glycaemic control in patients suffering from painful diabetic polyneuropathy.

	Patients and methods

	Study design and patients

From February 2004 through to October 2005, 61 patients with symptomatic diabetic polyneuropathy were recruited from the Silesia region of Poland. Enrollment criteria required that all patients have a diagnosis of diabetes (any type) and painful diabetic polyneuropathy with pain disturbing their sleep at night. The diagnosis of diabetic polyneuropathy was based on simple clinical tests, including pinprick, temperature and vibration perception (using a 128Hz tuning fork), 10g monofilament pressure sensation at the distal halluces and ankle reflexes <sup>[15Boulton A.J.M., Vinik A.I., Arezzo J.C., Bril V., Feldman E.L., Freeman R. Diabetic Neuropathies. A statement by the American Diabetes Association Diabetes Care 2005 ;  28 : 956-962 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>, and was also confirmed by electroneurography (NC). All patients had to mark at least 40mm on a 100mm visual analogue scale (VAS) of pain intensity, where zero (0) meant no pain and 100mm was the worst possible pain <sup>[16Melzack R. The Short-Form McGill pain questionnaire Pain 1987 ;  30 : 191-197 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. Patients were excluded if there were other causes of neuropathic pain (such as alcohol or drugs). As a safety precaution, pregnant women and those diagnosed with neoplasm or using cardiac pacemakers were also excluded. Analgesics or other drugs taken for the treatment of chronic neuropathic pain were continued, but no new drugs were allowed during the study period.

Patients were randomized into two groups. The study group consisted of 32 people with painful diabetic polyneuropathy who had an average pain duration of 23 months and who were exposed to low frequency magnetic fields. The control group consisted of 29 patients who had an average pain duration of 28 months and who received sham exposures to magnetic fields. Table 1 shows the general characteristics of the study participants. The study protocol was approved by the ethics committee of the Medical University of Silesia in Katowice, and written informed consent was obtained from all patients before enrollment. All enrolled patients agreed to blood tests, neurophysiological examination (at the neurology department) and filling in questionnaires periodically during the five-week study (VAS Short-Form McGill Pain Questionnaire, EuroQol EQ-5D VAS worksheet questionnaire and MOS Sleep Scale).

	Experimental treatment

Magnetic field exposure was performed with the use of a Viofor JPS device (Med & Life, Komorow, Poland), which is commercially available (shaped like a bed) and generates a low frequency magnetic field of up to 100μT <sup>[17http://www.vioforjps.pl/_index.html.

Cliquez ici pour aller à la section Références]</sup>. This level is defined as magnetostimulation in contrast to magnetotherapy, where induced field values are above 100μT. The electromagnetic waves generated by the Viofor JPS are a complex sequence of pulses at a frequency of about 180–195Hz. Electrical field intensity is about 130V/m and is similar to the earthʼs electrical field. The way in which the Viofor JPS is constructed allows its use in a double-blind manner. Depending on the type of code pre-entered, the device works in a genuine or sham exposure mode and neither the observer nor the subject knows which mode is truly active. The device was precoded by the manufacturer prior to the start of the study. Active and placebo codes were randomly divided into two equal parts (in blocks for 10 people) and were only disclosed after the study had been completed by all participants.

	Magnetostimulation scheme

Each patient was exposed to a genuine or sham magnetic field for a period of 15 days (three weeks, excluding Saturdays and Sundays). Each session lasted 20min and consisted of two 10min exposures according to the following application parameters: trunk: M1, P2; intensity 4; lower limbs: M1, P2; intensity 6. M1 is an application of constant intensity of a selected field throughout the entire exposure time, and P2 is a JPS system using ionic cyclotron resonance.

	Measures of outcome

The primary outcome measures were changes in:

Secondary outcome measures were changes in conduction parameters in the peripheral nerves of the lower limbs and in HbA_1c . The study period included three weeks of genuine (low frequency magnetic field) or sham exposures and two weeks of follow-up. Pain was measured on a 100mm linear VAS (Short-Form McGill Pain Questionnaire) <sup>[16Melzack R. The Short-Form McGill pain questionnaire Pain 1987 ;  30 : 191-197 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. Quality of life was assessed using EuroQol EQ-5D VAS worksheet questionnaire, which records the respondentʼs self-rated health status on a vertical scale of 0–100 (where 0 is the worst and 100 the best health status) <sup>[18www.euroqol.org.

Cliquez ici pour aller à la section Références]</sup>. Sleep assessment was performed using the MOS Sleep Scale questionnaire, in which answers are converted into percentages (0–100%) and the greater the score, the greater the sleep disruption <sup>[19Hays R.D., Stewart A.L. Sleep measures Measuring functioning and well-being: The medical outcomes study approach Durham, NC: Duke University Press (1992). 

Cliquez ici pour aller à la section Références and 20Spritzer KL, Hays R. cliquez ici . MOS Sleep Scale. A manual for use and scoring, version 1.0, 2003.

Cliquez ici pour aller à la section Références]</sup>. Patients’ answers to all the questionnaires were evaluated five times: at baseline; at the end of each week of exposure (Weeks 1, 2 and 3); and at the end of the study (Week 5). Electroneurography (NC) was used to measure motor and sensory nerve conduction velocity (MNCV, the peroneal and tibial nerves; SNCV, the sural nerve), according to the standard procedure (Counterpoint MK2, Dantec, Denmark). The amplitude of evoked potentials and their latency were also measured in all patients. Electroneurography was performed at baseline and after three and five weeks. In all patients, assessment of diabetic autonomic neuropathy was performed once, using an Ewing battery <sup>[20Spritzer KL, Hays R. cliquez ici . MOS Sleep Scale. A manual for use and scoring, version 1.0, 2003.

Cliquez ici pour aller à la section Références]</sup>. HbA_1c was measured by high performance liquid chromatography (HPLC; Variant Biorad) at baseline and at the end of the study (Week 5).

	Statistical analyses

Data are presented as means±standard deviation for parametric data and as medians (interquartile range) for nonparametric data. The Shapiro-Wilk test for normality was used to evaluate the distribution of data, and between-group differences were analyzed by the Mann-Whitney U -test. Wilcoxonʼs test was used to assess differences between baseline and each week of the study.

	Results

	Efficacy

The baseline characteristics of the 61 study participants were similar in both groups (Table 1). We observed a significant reduction in pain intensity after Week 1 with both genuine and sham exposures that persisted until the end of the follow-up observation period (P <0.05 or P <0.01 versus baseline at any time point). As shown in Fig. 1A, the extent of pain reduction was similar in both the study and control groups, with VAS values at baseline and after three weeks of 73mm versus 33mm and 69mm versus 41mm in the two groups, respectively. There were no statistically significant differences between the groups at any time during the study. Similar improvements were observed for quality of life (Fig. 1B) and sleep (Fig. 1C), with no significant differences between the groups throughout the study period. In addition, there were no statistically significant differences in conduction velocity, amplitude of evoked potentials and latency of peripheral nerves (data not shown).

Fig. 1
	Fig. 1.  A. Pain intensity, B. Quality of life, C. Sleep disturbances. Values are means±standard error of mean; *P <0.01; ^P <0.05 versus baseline. No significant differences between groups for all presented parameters at any time during the study. Zoom

Although no specific attempt was made to improve diabetes control during the study, HbA_1c significantly decreased (P <0.01) from baseline to the end of study, most likely as a trial effect (Table 2).

	Safety

No side effects were recorded during the study.

	Discussion

Our results show that low frequency magnetic fields have a positive impact on pain, quality of life and sleep, and HbA_1c values, but are no better than a placebo. A search through the literature revealed no study in which the effect of low frequency magnetic fields in painful diabetic polyneuropathy was assessed in a randomized, double-blind, placebo-controlled trial.

In all of our studied patients, previous analgesic pharmacological treatments were ineffectual. The key inclusion criterion, other than clinical confirmation of diabetic polyneuropathy, was a value greater than or equal to 40mm on a self-rated VAS of pain intensity. As already known from clinical practice, such a VAS range corresponds to pain that disturbs sleep at night and has a major impact on the patientʼs feeling of well-being. VAS, a standard tool in clinical trials to assess the efficacy of analgesic treatments, allows for maximum sensitivity, and comparisons of pain intensity and between-patient scores <sup>[13Musaev A.V., Guseinova S.G., Imamverdieva S.S. The use of pulsed electromagnetic fields with complex modulation in the treatment of patients with diabetic polyneuropathy Neurosci Behav Physiol 2003 ;  33 : 745-752 [cross-ref]

Cliquez ici pour aller à la section Références, 21Kolia G., Milojevic J.P., Cantineau R., Ruiz B., Coudert S., Bataille Can severe acute pain escape visual analog scale screening in the ED? Am J Emerg Med 2004 ;  22 : 238-241

Cliquez ici pour aller à la section Références and 22Freynhagen R., Strojek K., Whalen E., Balkenohl M., Griesing T., et al. Efficacy of pregabalin in neuropathic pain evaluated in a 12-week, randomised, double-blind, multicentre, placebo-controlled trial of flexible- and fixed-dose regimens Pain 2005 ;  115 : 254-263 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. It also eliminates the potential bias arising from memorization of responses given on previous questionnaires, which is often the case with descriptive scales <sup>[23Supranowicz P. Assessment of usefulness of visual analog scale (VAS) for measurements of attitude of the youth in health aspect Przegl Epidemiol 2003 ;  57 : 541-551

Cliquez ici pour aller à la section Références]</sup>.

Our results show that a low frequency magnetic field generated by a Viofor JPS device according to fixed parameters of exposure (M1, P2, and intensity 4 and 6) has no advantage over the placebo effect. Such a programme of exposure is commonly used as analgesic therapy for musculoskeletal pain. Our results do not correlate with results of other studies evaluating the impact of magnetic fields in patients with painful diabetic polyneuropathy. Two small-scale studies (N =31 and N =21) reported subjective pain reduction and improvement in vibration sensation (measured with a tuning fork) <sup>[23Supranowicz P. Assessment of usefulness of visual analog scale (VAS) for measurements of attitude of the youth in health aspect Przegl Epidemiol 2003 ;  57 : 541-551

Cliquez ici pour aller à la section Références and 24Cieślar G., Sieroń A., Radelli J. Assesment of pulsed magnetic field action in patients with painful diabetic neuropathy including vibration perception Baln Pol 1995 ; 37

Cliquez ici pour aller à la section Références]</sup>. However, these studies were open and mainly assessed the impact of magnetotherapy, not magnetostimulation. Furthermore, electroneurography (NC), an objective tool to assess neuropathy, was not performed. There was also no information regarding the use of an objective pain scale <sup>[24Cieślar G., Sieroń A., Radelli J. Assesment of pulsed magnetic field action in patients with painful diabetic neuropathy including vibration perception Baln Pol 1995 ; 37

Cliquez ici pour aller à la section Références and 25Sadurska K., Łukomska E., Szwedek R., Wierusz-Wysocka B. Evaluation of efficacy of pulsed magnetic fields in treatment of diabetic neuropathy and angiopathy Baln Pol 1992 ;  34 : 109-119

Cliquez ici pour aller à la section Références]</sup>. In a different study (N =121), the authors observed a 54% reduction in pain (by VAS), but magnetotherapy was not the only intervention: all subjects had also received therapeutic massage and exercises <sup>[13Musaev A.V., Guseinova S.G., Imamverdieva S.S. The use of pulsed electromagnetic fields with complex modulation in the treatment of patients with diabetic polyneuropathy Neurosci Behav Physiol 2003 ;  33 : 745-752 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. Positive effects of PMF, including pain reduction, were also demonstrated in a pilot study by Weintraub and Cole <sup>[12Weintraub M.I., Cole S.P. Pulsed magnetic field therapy in refractory neuropathic pain secondary to peripheral neuropathy: electrodiagnostic parameters-pilot study Neurorehabil Neural Repair 2004 ;  18 : 42-46 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. However, the analyzed group was not homogeneous (not only diabetic patients were included) and there was no control group.

In our study, the reduction of pain intensity in both groups was followed by an improvement in quality of life (EQ-5D VAS) and sleep (MOS Sleep Scale). Clearly, improvement in quality of life can be largely attributed to the pain reduction. Likewise, diminished sleep disruption most likely resulted from the reduction in night-time pain, a typical feature of symptomatic diabetic polyneuropathy. The lack of differences in our questionnaire findings between the real and sham exposure groups could have been the result of difficulties in objective assessment of pain by patients. On the other hand, pain perception is highly subjective. In the case of a chronically ill patient who suffers from neuropathic pain, simply taking an interest in his/her problem may result in subjective feelings of improvement (placebo effect). Also, it is well known that the sensation of pain differs from one person to another. Psychological studies show that many factors, such as anxiety, suggestion or attention, can influence the origin, intensity and duration of pain. Although the threshold of pain perception is relatively stable, pain tolerance depends on mental status; it decreases in depression and increases in consciousness disturbances <sup>[26Wörz M. Pharmacology of pain  Warszawa: PZWL (1992). 

Cliquez ici pour aller à la section Références]</sup>. It should also be said that a similarly strong placebo effect was observed in pharmacological trials of analgesic and coanalgesic drugs <sup>[22Freynhagen R., Strojek K., Whalen E., Balkenohl M., Griesing T., et al. Efficacy of pregabalin in neuropathic pain evaluated in a 12-week, randomised, double-blind, multicentre, placebo-controlled trial of flexible- and fixed-dose regimens Pain 2005 ;  115 : 254-263 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>.

In the present study as with that of Musaev et al. <sup>[13Musaev A.V., Guseinova S.G., Imamverdieva S.S. The use of pulsed electromagnetic fields with complex modulation in the treatment of patients with diabetic polyneuropathy Neurosci Behav Physiol 2003 ;  33 : 745-752 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>, NC revealed a slight baseline impairment in conduction velocity for all the nerves studied (peroneal: 40.07 and 40.60m/s, respectively; tibial: 39.2 and 39.4m/s, respectively) along with low amplitude evoked potentials (peroneal: 1.1 and 2.4m/V, respectively; tibial: 3.0 and 2.9m/V, respectively). These data indicate lesions in the myelin sheaths and axial cylinders of the peripheral nerves in the lower limbs in both studies. Amplitude reduction is more typical for axonal polyneuropathy, as in diabetic polyneuropathy, whereas conduction velocity may be slightly diminished or even unchanged when fast fibers are relatively preserved <sup>[27 Clinical electromyography Warszawa: PZWL (1986).  [wyd II].

Cliquez ici pour aller à la section Références]</sup>. In the Musaev et al. study, there was an improvement in the examined parameters that was more significant with a magnetic field of lower frequency (10Hz) than with the 100Hz field. In our study, we observed no changes in NC in either the study or control group. Weintraub et al. obtained findings similar to ours with pulsed <sup>[12Weintraub M.I., Cole S.P. Pulsed magnetic field therapy in refractory neuropathic pain secondary to peripheral neuropathy: electrodiagnostic parameters-pilot study Neurorehabil Neural Repair 2004 ;  18 : 42-46 [cross-ref]

Cliquez ici pour aller à la section Références]</sup> and static magnetic fields <sup>[11Weintraub M.I., Wolfe G.I., Barohn R.A., Cole S.P., Parry G.J., Hayat G., et al. Magnetic Research Group. Static magnetic field therapy for symptomatic diabetic neuropathy: a randomized, double-blind, placebo-controlled trial Arch Phys Med Rehabil 2003 ;  84 : 736-746 [cross-ref]

Cliquez ici pour aller à la section Références]</sup> in patients with neuropathic pain.

NC analysis is an objective examination of neuropathy used for DPN diagnosis as well as its progression. Many trials conducted so far confirm the high repeatability of this method <sup>[28Feldman E.L., Stevens M.J., Thomas P.K., Brown M.B., Canal N., Greene D.A. A practical two-step quantitative clinical and electrophysiological assessment for the diagnosis and staging of diabetic neuropathy Diabetes Care 1994 ;  17 : 1281-1289

Cliquez ici pour aller à la section Références and 29Claus D., Mustafa C., Vogel W., Herz M., Neundörfer B. Assessment of diabetic neuropathy: definition of norm and discrimination of abnormal nerve function Muscle Nerve 1993 ;  16 : 757 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. The lack of such changes in our study suggests no impact of PMF on definite parameters on NC; however, this does not preclude a beneficial effect of PMF on pain intensity via other mechanisms such as cellular receptor modulation.

All of the above-mentioned studies assessed the effectiveness of magnetotherapy, whereas we examined the impact of magnetostimulation in patients with painful diabetic polyneuropathy. The difference between these two interventions is the frequency of the magnetic field: magnetostimulation ranges from a few to 3000Hz and induction values are below 100μT; magnetotherapy uses frequencies up to 100Hz and induction values range from 0.1–20mT. Effects of both types of magnetic field exposure are convergent on many points. Lately, magnetostimulation is of greater interest for analgesic therapy and, unlike magnetotherapy, the data confirm that it does not affect melatonin secretion <sup>[30Karasek M., Czernicki J., Woldańska-Okońska M., Żylińska K., Świętosławski J. Chronic exposure 25–80μT, 200Hz magnetic fields does not influence serum melatonin concentrations in patients with low back pain J Pineal Res 2000 ;  29 : 81-85

Cliquez ici pour aller à la section Références]</sup>. An analgesic action of melatonin has been suggested in experimental studies of rodents <sup>[31Mantovani M., Kaster M.P., Pertile R., Calixto J.B., Rodrigues A.L., Santos A.R. Mechanisms involved in the antinociception caused by melatonin in mice J Pineal Res 2006 ;  41 : 382-389 [cross-ref]

Cliquez ici pour aller à la section Références and 32Ulugol A., Dokmeci D., Guray G., Sapolyo N., Ozyigit F., Tamer M. Antihyperalgesic, but not antiallodynic, effect of melatonin in nerve-injured neuropathic mice: Possible involvements of the l-arginine-NO pathway and opioid system Life Sci 2006 ;  78 : 1592-1597 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>.

Other studies claim a positive role for PMF in glycaemic control. In experimental studies of rodents, researchers observed lowered glucose concentrations and a reduction in the insulin to glucose ratio in exposed animals compared with controls <sup>[33Laitl-Kobierska A., Cieślar G., Sieroń A., Grzybek H. Influence of alternating extremely low frequency ELF magnetic field on structure and function of pancreas in rats Bioelectromagnetics 2002 ;  23 : 49-58 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>. According to the authors, the underlying mechanism could be stimulation of insulin secretion and peripheral tissue glucose uptake by PMF. In another study, a higher absorption of ³H glucose injected into the peritoneal cavity was observed in rats exposed to magnetic fields <sup>[34Brus H. Impact of low frequency magnetic field on 3H glucose absorption in brain and selected tissues of rats including role of NO. Praca doktorska, Śląska Akademia Medyczna 2003.

Cliquez ici pour aller à la section Références]</sup>. Such effects were attributed to the field-induced changes in cell membranes and ion channels. In our pilot study (N =21), we noted a significant reduction in HbA_1c values after five weeks (three weeks of exposure and two weeks of observation) in patients exposed to magnetic fields compared with sham exposures (P <0.05) <sup>[35Wróbel M., Szymborska-Kajanek A., Karasek D., Biniszkiewicz T., Sieroń-Słotny K., Grzeszczak W., et al. Influence of alternating low frequency magnetic fields in patients with diabetic polyneuropathy: pilot study Diabetologia Doświadczalna i Kliniczna 2005 ;  5 : 59-63

Cliquez ici pour aller à la section Références]</sup>. However, after increasing the study group to 61 patients, similar reductions in HbA_1c were observed in both analyzed groups. There was no change in hypoglycaemic treatment during the study period that could have caused these reductions in HbA_1c . It may be that simply participating in a trial may have resulted in an improvement in glucose control (trial effect) due to better patient compliance during the study period. Others have confirmed that HbA_1c reduction has a positive effect on peripheral nerve function that may result in diminished pain intensity <sup>[36The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;29:977–86.

Cliquez ici pour aller à la section Références and 37UK Prospective Diabetes Study (UKPDS) Group intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–53.

Cliquez ici pour aller à la section Références]</sup>. In addition, pain reduction decreases stress and, thus, may also contribute to better glycaemic control. However, it is unlikely that improvement of glycaemia was responsible for the observed pain reduction seen after the first week of PMF exposure.

When considering the use of low frequency magnetic fields for painful diabetic polyneuropathy or in medicine in general, there arises not only the question of does it really work, but also of how does it work. One hypothesis to explain the effect of magnetic fields on human tissues is that it affects the plasma membrane transport of calcium ions; a resonant interaction at cyclotron frequency between calcium ions of geomagnetic densities is observed <sup>[38Blackman C., Benane S., Kinney L., Joines W.T., House D.E. Effects of ELF fields on calcium-ion efflux from brain tissue in vitro Radiat Res 1982 ;  92 : 510-520 [cross-ref]

Cliquez ici pour aller à la section Références and 39Halle B. On the cyclotron resonance mechanism for magnetic field effects on transmembrane ion conductivity Bioelectromagnetics 1988 ;  9 : 381-385 [cross-ref]

Cliquez ici pour aller à la section Références]</sup>.

Comparison of our results with the above-mentioned findings of other authors is difficult because of differences between studies in terms of parameters of magnetic fields used (exposure profile), exposure duration, total exposure times and devices used to generate the magnetic fields. Moreover, only a few researchers analyzed the placebo effect <sup>[11Weintraub M.I., Wolfe G.I., Barohn R.A., Cole S.P., Parry G.J., Hayat G., et al. Magnetic Research Group. Static magnetic field therapy for symptomatic diabetic neuropathy: a randomized, double-blind, placebo-controlled trial Arch Phys Med Rehabil 2003 ;  84 : 736-746 [cross-ref]

Cliquez ici pour aller à la section Références and 24Cieślar G., Sieroń A., Radelli J. Assesment of pulsed magnetic field action in patients with painful diabetic neuropathy including vibration perception Baln Pol 1995 ; 37

Cliquez ici pour aller à la section Références]</sup>. Furthermore, it must be emphasized that, in humans, the biological response is dependent on exposure at particular magnetic field strengths. A “window” is defined as a biological response that occurs only within a specific amplitude or frequency range, being moderate or absent outside of this range. This could explain the apparently conflicting findings in the studies using different exposure profiles. Finally, it needs to be emphasized that there were no major, clinically important, side effects with the use of low frequency magnetic fields in either our study or any of the above-mentioned trials.

The results we obtained may also have been determined by the specific exposure pattern (M1, P2, intensity 4 and 6) we used. For this reason, randomized, double-blind, placebo-controlled studies using a variety of magnetic field exposure protocols (specific window) are necessary to arrive at any conclusions regarding the potential benefits of this procedure in the treatment of painful diabetic neuropathy.

	Declaration of competing interests

None to declare.