Individual subject sensitivity to extremely low frequency magnetic field

NeuroToxicology 27 (2006) 534–546

Individual subject sensitivity to extremely low frequency magnetic field Alexandre Legros a,*, Anne Beuter b a

Lawson Health Research Institute, Department of Imaging, St. Joseph’s Health Care, 268 Grosvenor Street, London, Ont., Canada N6A 4V2 b Institut de Cognitique, Universite´ Victor Segalen Bordeaux 2, 146 Rue Le´o Saignat, 33076 Bordeaux Cedex, France Received 12 October 2004; accepted 16 February 2006 Available online 18 April 2006

Abstract It is becoming important to specify the smallest effects of extremely low frequency (ELF) magnetic fields (MF) on human physiology. One difficulty is that some people seem more sensitive and more responsive than others to MF exposure. Consequently, within- and between-subject differences have to be taken into account when evaluating these effects. As shown in previous work, human postural tremor is sensitive to MF exposure. But data about individual responses have not been examined in detail. Thus, postural tremor of 24 subjects was evaluated under ELF MF ‘‘on’’ and ‘‘off’’ conditions in a double-blind real/sham exposure protocol. The direction of the tremor changes was analyzed individually for three tremor characteristics. Results showed that subjects with high amplitude tremor seem to be more responsive to MF exposure. MF had an instantaneous effect (between ‘‘on’’ and ‘‘off’’ conditions) and also a more delayed and persistent one (between real and sham conditions), but differences were small. Moreover, due to the within- and between-subject variability, no statistical analysis could be done. However, these results do not show any potentially harmful effect of domestic or industrial 50 Hz MF on humans. They provide a starting point to orient future studies and should be taken into account in the establishment of new exposure limits. # 2006 Elsevier Inc. All rights reserved. Keywords: Postural tremor; ELF; Magnetic field; Individual differences

1. Introduction In 1997, an international seminar was held in Bologna (Italy) on the biological effects and related health hazards of ambient or environmental static and extremely low frequency (ELF) magnetic fields (MF). Following this seminar, sponsored among others by the World Health Organization (WHO) and the International Commission on Non-Ionizing Radiation Protection (ICNIRP), Repacholi and Greenebaum (1999) recapitulated the research that is still needed to better understand MF effects on humans. They underlined the need to explore whether electrophysiological indices of central nervous system activity and function are affected by ELF MF. They also specified that published reports should include information on the smallest MF effect that could be detected in humans. Several studies have been done on electrophysiological parameters such as electroencephalogram (EEG) or evoked

* Corresponding author. Tel.: +1 519 646 6100x65959; fax: +1 519 646 6399. E-mail address: [email protected] (A. Legros). 0161-813X/$ – see front matter # 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.neuro.2006.02.007

potentials (Bell et al., 1992, 1994a,b; Cook et al., 2004; Heusser et al., 1997; Lyskov et al., 1993a,b; Lyskov and Sandstrom, 2001; Marino et al., 2004). However no consensus exists on the direction of the effects due to the wide variety of exposure procedures used (length of the exposure, intensity and frequency of the MF, recordings made sometimes during and sometimes after the exposure). Moreover, MF induce artifacts in electrophysiological data which often make recording during exposure not possible (see for example Cook et al., 2004). If such neurophysiological effects exist, they might have behavioral manifestations. Many studies have been a MF effect on cognitive performance, on reaction time (Cook et al., 1992; Kazantzis et al., 1998; Podd et al., 2002, 1995; Preece et al., 1998; Whittington et al., 1996) and on human motor control. Indeed, Thomas et al. (2001) showed that a 200 mT pulsed MF improved human standing balance. Physiological tremor is a highly sensitive indicator of neuromotor pathway integrity and may be influenced by more than 37 factors (Wachs and Boshes, 1966). Briefly, three mechanisms contributing to its generation have been described: (1) short-latency spinal feedback from stretch receptors and long-loop transcortical or transcerebellar reflex pathways from these receptors, (2) mechanical properties

A. Legros, A. Beuter / NeuroToxicology 27 (2006) 534–546

of the extremities and (3) central oscillations modulating motor neuron pool activity. Some studies analyzed the effect of transcranial magnetic stimulation (TMS), which is a high intensity MF stimulation of the cerebral cortex (up to 2.5 T), on human tremor. Results showed that TMS can reduce tremor severity in patients with essential tremor or Parkinson’s disease (Britton et al., 1993; Gironell et al., 2002; Pascual-Leone et al., 1994). Recently, we explored how a low intensity MF (1000 mT, 50 Hz) could have an effect on human distal motor control by studying postural tremor and motor control during a fingertracking task (Legros and Beuter, 2005; Legros et al., 2006). Results showed a possible MF effect on postural tremor increasing the proportion of the low frequency component

535

(between 2 and 4 Hz) and facilitating the decrease of tremor intensity over time. These results were obtained by analyzing three quantitative characteristics computed on postural tremor time series: amplitude, peakedness and proportional power in the 2–4 Hz range (Beuter and Edwards, 1999; Beuter et al., 2003; Edwards and Beuter, 2000). It has been shown that sensitivity and responsiveness to ELF MF may vary across subjects (Levallois, 2002; Bell et al., 1991). For example, Lyskov and Sandstrom (2001) indicated that patients with ‘‘electrical hypersensitivity’’ tend to be hyper-sympathotone, hyper-responsive to sensory stimulation and to have heightened arousal. Therefore, the main aim of this study was to examine within- and between-subject differences in postural tremor characteristics in relation to ELF MF exposure. Indeed if

Fig. 1. (a) Illustration of the four sequences of an experimental session. The order of presentation of these sequences was counterbalanced. X-coordinates express time, vertical grey bands represent the 62 s recording periods and solid lines show MF status (‘‘off’’ when the line is down and ‘‘on’’ when the line is up). The dotted line represents the status of sham MF (i.e., its position if MF was present). (b) Only 58 s centered on a MF transition were kept in each recording and were composed of two parts of 29 s (before and after the MF transition). (c) Recordings 1–4 and 5–8 of Fig. 5a are, respectively, equivalent and were averaged (1 with 5, 2 with 6, . . .).

536

A. Legros, A. Beuter / NeuroToxicology 27 (2006) 534–546

differences between subject responses to MF exposure would exist (subjects more or less responsive, with heterogeneous reactions), they would not be detected in group results but only through an individualized examination.

refrain from smoking or drinking coffee the morning of the experimentation. The study’s protocol was reviewed and approved by the Operational Committee for Ethics in life sciences section of the Centre National de la Recherche Scientifique (CNRS, France).

2. Materials and methods 2.2. Procedures 2.1. Subjects Thirty-six men between the ages of 20 and 50 years (37.8
8) occasionally subjected to 50 Hz MF during their work were recruited among the personnel of the French electricity company ‘‘Electricite´ de France’’ (EDF) and completed the experiment. They were volunteers and gave informed consent before their participation. None of them had previously taken part in studies involving MF exposure. Before testing they were required to complete a screening questionnaire to ensure that: they did not use drugs or medications regularly; they had never experienced an epileptic seizure; they had no limitation of hand or finger movements; they did not suffer from chronic illness (e.g., diabetes, psychiatric, cardiovascular or neurological diseases); they had no cardiac or cerebral pacemaker; and they had no metallic implant in the head or in the thorax. This information was verified by EDF’s occupational medicine service. All subjects were asked to

Subjects were all tested at the same time of day (9.00 a.m.), during a single session (Crasson et al., 1999; Tyrer and Bond, 1974), under natural lighting and the room temperature was controlled at 23 8C (Lakie et al., 1994). Their handedness was determined using the Oldfield questionnaire (Oldfield, 1971). After completing these requirements, they sat on a plastic chair placed in the middle of the MF generating device. Their dominant forearm was placed in a prone position on an armrest and the dominant tested hand was placed with the palm facing towards the ground on a molded clay support. The armrest was adjustable according to each subject’s morphology. A piece of white cardboard ( 0.6 was found between the three analyzed tremor characteristics. Individual graphical screening of subjects’ tremor shows a wide range of between-subject differences, be they in terms of amplitude or frequency. By the means of the visual examination of graphical data, four specific behaviors can be identified in subjects’ tremor time series: (1) high or low averaged amplitude, (2) high velocity segments (HVS, see Fig. 3), (3) bursts (see Fig. 4) and (4) unusual frequency content (Fig. 5). HVS are defined as short, fast movements of the index finger having at least 1 mm in amplitude and shorter than 0.2 s

(see examples in Figs. 3a and 6). Bursts are defined as a local increase of the peak-to-peak amplitude observed in the filtered posture time series (of at least twice the peak-to-peak background amplitude and lasting at least 5 s). 3.1. Graphical evaluation Visual inspection of tremor amplitude shows in most of the subjects lower values during the last 28 s than during the first 28 s of recording. This makes the detection of a MF effect difficult. Overall mean of tremor amplitude for all subjects is 0.043 mm with a between-subject range of 0.080 mm (Table 1). Several subjects have HVS which affect the computation of the characteristics. For example, subject 11 has HVS in three recordings (Fig. 6). Despite the fact that once it occurs at the time of MF transition, there is no evidence that it is linked with MF exposure (see Fig. 6d). HVS are noted in four other subjects (subjects 16, 27, 22 and 31) and they are also not linked with the presence, the absence or the transition of the MF. Bursts are present in two subjects (subjects 18 and 30) and correspond to a local increase of tremor amplitude. These events were visually

A. Legros, A. Beuter / NeuroToxicology 27 (2006) 534–546

541

Table 4 Directions of the individual changes in peakedness linked with MF exposure Mean over 8 trials

Difference on  off (real)

Off ! on

On ! off

Real

Sham

Real

Sham

Low amplitude tremor Subject 36 Subject 29 Subject 12 Subject 10 Subject 13 Subject 11 a Subject 32 Subject 16 a Subject 27 a Subject 34 Subject 20 Subject 25 Subject 28 Subject 22 a

0.780 0.784 0.790 0.785 0.801 0.789 0.758 0.767 0.789 0.754 0.791 0.784 0.783 0.772

+ + + + + +  +  + +  + +

% % % % % & & & % % % & % %

% % & & & & & % % % % & % %

& & & & & & % & % % & % & &

& % & & & & & % % % & & & %

High amplitude tremor Subject 15 Subject 30 b Subject 14 Subject 18 b Subject 17 Subject 8 Subject 33 c Subject 23 Subject 31 a Subject 26

0.789 0.744 0.769 0.763 0.784 0.781 0.807 0.771 0.742 0.786

+ +   + + + + + +

% & & % % % & % % %

% & & % % % & & % %

& & % % % % & % % &

& & & & % % & % % &

Overall mean

0.778

+

%

%

&

%

Columns present for each subject: (1) the averaged peakedness over the eight recordings of the session; (2) the direction of the changes between ‘‘off’’ and ‘‘on’’ conditions during real exposure sequence independently of the time course (a ‘‘+’’ means that peakedness was higher in the ‘‘on’’ condition and vice versa); (3)–(4) direction of the changes for an ‘‘off/on’’ transition during real and sham exposure sequences, respectively; (5)–(6) direction of the changes for an ‘‘on/off’’ transition during real and sham exposure sequences, respectively. a HVS. b Burst. c Frequency.

detected by the experimenter during the experimental session. They occur sometimes in ‘‘off’’, sometimes in ‘‘on’’, sometimes in sham conditions and they do not seem linked with MF exposure. For subject 33, the frequency content of tremor time series is very different from other subjects: his tremor is highly organized around 10 Hz and this is consistent across recordings (see Fig. 5 for example). MF does not appear to affect his 10 Hz oscillation. Following this visual analysis of all subjects’ recordings, three remarks can be made to orient deeper investigations. First, subjects have a relatively wide range of tremor amplitude (with proportions from 1 to 5 between subjects 26 and 36) and have to be classified in two groups regarding their overall mean amplitude (above and below 0.043 mm): subjects with high amplitude tremor and subjects with low amplitude tremor. Second, the direction of the changes within- and between subjects has to be individually evaluated for each characteristic. And third, the specific behaviors of subjects (HVS, bursts and unusual frequency content) have to be taken into account to interpret the direction of the changes. Four recapitulative tables have been made to support this approach: Table 1 summarizes

the direction of the changes between real and sham exposure sequences for the three characteristics; Tables 2–4 present the direction of the changes between ‘‘off’’ and ‘‘on’’ conditions for amplitude, proportional power in the 2–4 Hz range and peakedness, respectively. Each characteristic is analyzed separately. Note that all transitions, be they ‘‘off/on’’ or ‘‘on/ off’’, also correspond to a begin/end recording (it is impossible to record a transition without this time effect). 3.2. Amplitude For 12 subjects, all corresponding averaged recordings (see caption of Fig. 1c for details) show a systematic begin/end decrease in amplitude independent of the MF status (they all have arrows downwards in Table 2). Twenty-two averaged transitions concerning 12 subjects show a begin/end increase in amplitude values and only nine in the real exposure sequence (during ‘‘off/on’’ transitions for subjects 11, 20, 28 and during ‘‘on/off’’ transitions for subjects 15, 16, 22, 28, 30 and 36, see Table 2). Each of these exceptions is analyzed regarding the corresponding graphical representation and results show that

542

A. Legros, A. Beuter / NeuroToxicology 27 (2006) 534–546

Fig. 4. Recording 6 of subject 30 (sham). A burst is clearly visible in the second part of the recording (a and b). This event is not linked with MF but it influence amplitude, peakedness (c) and frequency distribution (d and e).

they are due to the presence of HVS (for subjects 11 and 16), of bursts (for subjects 30), or more often of small amplitude fluctuations or tremor irregularities occurring in the last 29 s of time series (for subjects 10, 13, 15, 17, 20, 26, 28 and 36). Moreover the increase in amplitude over time during real exposure is never reproduced from one corresponding recording to another. Fourteen subjects out of 24 have lower amplitude during averaged ‘‘on’’ than ‘‘off’’ conditions. But these differences are very small (below 10 mm, which is near the laser resolution) except for three high amplitude tremor subjects: subjects 23, 30 and 31 who have values of 14, 23 and 11 mm, respectively. The reasons are: a smaller amplitude in the ‘‘on’’ condition of one of the two ‘‘on/off’’ transitions for subject 23; a burst in the second half of one ‘‘on/off’’ transition for subject 31; a HVS during an ‘‘off’’ condition for subject 31. Thus, overall mean amplitudes are lower during the ‘‘on’’ condition. Concerning the differences between real and sham exposure sequences, half of the subjects have higher amplitude during real exposure and the other half during sham (Table 1).

tendencies across the different MF conditions: they have all arrows in Table 3 going in the same direction independently of MF status (downwards for subjects 8, 12, 27, 29 and 34 and upwards for subject 33, see Table 3). Four subjects show a decreased proportional power in the 2–4 Hz range during ‘‘off/ on’’ transition concomitant with an increase during ‘‘on/off’’ transition (subjects 10, 15, 22 and 36). Seven subjects show the opposite behavior (subjects 14, 18, 20, 26, 30, 31 and 32). But only three out of these eleven subjects (subjects 14, 15 and 36) have a consistent behavior between the two corresponding recordings (tendencies in Table 3 result from their averaging, see Fig. 1a and b for details). On average, half of the subjects have higher values during ‘‘on’’ conditions, independently of their individual specificities (HVS, burst or frequency). Concerning differences between mean values in real and sham exposure sequences, only two subjects out of ten have lower values during real exposure (see Table 1).

3.3. Proportional power in the 2–4 Hz range

Peakedness is not significantly influenced by the beginning/ end effect (Legros and Beuter, 2005) but its general tendency is to increase at the end of recordings (except for ‘‘on/off’’ transition under real exposure, see Table 4). Six subjects show a consistent begin/end tendency independently of the different MF conditions (four arrows going in the same direction:

As shown previously (Legros and Beuter, 2005), proportional power in the 2–4 Hz range is not significantly affected by the begin/end effect. Thus, the effect of the MF should be more visible for this characteristic. Six subjects exhibit similar

3.4. Peakedness

A. Legros, A. Beuter / NeuroToxicology 27 (2006) 534–546

543

Fig. 5. Recording 5 of subject 33 (‘‘off/on’’ transition). He exhibits a powerful 10 Hz component in his postural tremor (d and e) which gives a noisy aspect to his postural tremor time series (a–c). It seems independent of the MF status.

peakedness increases for subjects 8, 17, 27, 31 and 34 and decreases for subject 33). Thus, if MF had an effect on peakedness, changes during ‘‘off/on’’ and ‘‘on/off’’ transitions should go in opposite directions. This is the case for 12 subjects of whom nine showed an increase of peakedness during ‘‘off/ on’’ and a decrease during ‘‘on/off’’ transitions (subjects 10, 12, 13, 15, 22, 26, 29, 28 and 36). Three show opposite tendencies (subjects 14, 25 and 32), but only subject 28 shows consistent tendencies across all the corresponding recordings. On average, peakedness values are higher during ‘‘on’’ than during ‘‘off’’ conditions for 19 subjects and for only two subjects among the 10 with high amplitude tremor. Eight subjects have higher peakedness values during real than during sham exposure condition, but only three among the 10 with high amplitude tremor. 4. Discussion We attempted to answer three questions, namely: can we detect MF effects in human postural tremor? Are some subjects more responsive to MF exposure than others? Do responsive subjects all react in the same manner to MF exposure? First, two types of intermittent behaviors (i.e., HVS and bursts) were detected in many subjects’ postural tremor. These events represent superimposed involuntary motor activity and make it difficult to interpret results. But they are a part of human

behavior and can in turn give information about the possible MF effects on motor control activity. However, results show that there is no evidence that MF could produce or modulate these intermittent behaviors. Nevertheless, we have to take them into account to understand the results. Secondly, tremor amplitude decreases between the beginning and the end of a 60 s recording. This phenomenon has already been shown in previous work (Legros and Beuter, 2005) and was interpreted as a consequence of the relaxation of the subject induced by the experimental procedure involving sensory auditory deprivation (Legros et al., 2006; Lundervold et al., 1999). This effect is highly consistent within- and between subjects: only one subject does not exhibit such a tendency. Moreover, many subjects seem to be more sensitive than others to the effect of time on postural tremor: four of them follow the same tendencies with proportional power in the 2– 4 Hz range and peakedness. But subject 33 is an outlier: while the three other subjects decrease their proportional power in low frequencies and increase their peakedness, he goes in the opposite direction. It might be due to his unusual postural tremor which is highly organized around 10 Hz (other frequencies are negligible). Thus, when his tremor decreases in amplitude over time, the 10 Hz oscillations decrease in power, and consequently other frequencies increase in proportion (Fig. 2). Concerning peakedness, if a sinusoidal oscillation is more regular, its peakedness increases: if power of

544

A. Legros, A. Beuter / NeuroToxicology 27 (2006) 534–546

Fig. 6. The HVS produced by subject 11 during: (a) a sham recording, (b) an ‘‘off/on’’ transition and (c) an ‘‘on/off’’ transition. (d) The zoom on the raw data (grey line) and on the lowpass filtered data (40 Hz, dotted line) simultaneously to the ‘‘on/off’’ MF transition shows that the HVS begin before the transition.

the 10 Hz oscillation decreases for subject 33, his tremor is less ‘‘regular’’ and peakedness decreases. Third, a short term MF effect is difficult to argue (differences between tremor during ‘‘off’’ and ‘‘on’’ conditions), however an effect on amplitude and peakedness is possible, especially for subjects with high amplitude tremor: among the 10 high amplitude tremor subjects, seven had a lower amplitude and eight had a higher peakedness in the ‘‘on’’ condition (as seen in the lower part of the second column of Tables 2 and 4), independently of the type of transition (‘‘off/on’’ or ‘‘on/off’’). This is consistent with previous results (Legros and Beuter, 2005) showing a significant MF effect on peakedness and a non-significant tremor amplitude decrease. MF have also been shown to accentuate the relaxation effect over time (Legros et al., 2006). Fourth, results of proportional power in the 2–4 Hz range and of peakedness suggest a long term MF effect (i.e., differences between tremor during real and sham exposure) for subjects with high amplitude tremor: the presence of MF for 8 min (two 4-min exposures spaced by 4 min) induces lower peakedness for seven out of 10 subjects and higher proportional power in the 2–4 Hz range for eight out of 10 subjects (as shown by the ‘‘’’ and the ‘‘+’’, respectively, in the lower part of the last two columns of Table 1). This also confirms previous results and shows that these characteristics are sensitive to MF exposure.

The effect on peakedness was not retained in our previous work mainly because, on average, it was higher during MF ‘‘on’’ than during MF ‘‘off’’ conditions, but it was lower during real than during sham exposure and its statistical significance was small. However, individual analyses presented here suggest that subjects with high amplitude tremor tend to be more responsive to MF exposure. Moreover, it is possible that ELF MF can modulate human motor behavior with different delays corresponding to different control mechanisms. Our first assumption was that since MF induces microcurrents in the central and peripheral nervous system, it could instantaneously modulate its activity (this is shown for example by Marino et al., 2004, on an EEG study), and therefore may act on motor behavior (as shown by Thomas et al., 2001, on postural sway). But it is also possible that MF exposure lead to longer term effects. For example, Chen et al. (1997) showed that TMS at 0.9 Hz during 15 min leads to a mean decrease of cortical excitability lasting at least 15 min following the stimulation. More precisely, they showed a decrease in motor evoked potential amplitude of 19.5%. According to these authors this result may be due to a change in the level of depolarization of the postsynaptic neuron, as shown by Artola et al. (1990) with animal studies. Even if the physiological basis of these observations is unknown, Gironell et al. (2002) speculate that it could be the same mechanism that explains the ‘‘anti-tremor effect’’ they

A. Legros, A. Beuter / NeuroToxicology 27 (2006) 534–546

observed in a study showing that repetitive TMS of the cerebellum improves tremor in patients with essential tremor. But if this kind of long term effect occurred, the preceding exposure sequences undergone by subjects would have had an effect on the following one, and this is not seen in our data (see Table 1). Cook et al. (2004) showed that the effect of a 200 mT pulsed MF on human EEG disappears after a delay between 3 and 7 min. If these delays are those involved here, it could explain the effect between real and sham exposure: in a real exposure sequence, the MF effect would persist when MF is turned ‘‘off’’ and could affect the following ‘‘off’’ conditions (before it is turned ‘‘on’’ again, but in a sham condition its effect would have already disappeared). Our findings provide a new basis to orient future research. They show that MF exposure to a 50 Hz, 1000 mT MF could be detected in human postural tremor, and that some subjects are more responsive than others. The effects could be detected immediately during exposure but also after a delay. However due to the relatively wide range of between-subject variability and the very subtle within-subject recorded differences, no clear significant statistical differences exist: a larger sample size would help to get deeper insight into individual differences. Moreover, following Cook et al. (2004), real and sham exposure sessions have to be programmed on two separate days to avoid perturbations due to the possible persisting effect of the MF. To conclude, at the behavioral level of observation explored in this study, there is no evidence of a potentially harmful effect of a short exposure to a 50 Hz domestic or industrial MF and this should be taken into account for establishment of new exposure limits. Acknowledgements We thank the participants, Hydro-Que´bec for financial support (Drs. M. Plante and D. Goulet), and Dr. D. Nguyen for designing the exposure system. We acknowledge EDF and Dr. N. Foulquie´ for subjects’ recruitment, DOCO Microsyste`mes, Inc., Montreal, for the data acquisition system and Dr. P.P. Vidal for his support. References Artola A, Brocher S, Singer W. Different voltage-dependent thresholds for inducing long-term depression and long-term potentiation in slices of rat visual cortex. Nature 1990;347:69–72. Bell GB, Marino AA, Chesson AL. Alterations in brain electrical activity caused by magnetic fields: detecting the detection process. Electroencephalogr Clin Neurophysiol 1992;83:389–97. Bell GB, Marino AA, Chesson AL. Frequency-specific blocking in the human brain caused by electromagnetic fields. Neuroreport 1994a;5:510–2. Bell GB, Marino AA, Chesson AL. Frequency-specific responses in the human brain caused by electromagnetic fields. J Neurol Sci 1994b; 123:26–32. Bell GB, Marino AA, Chesson AL, Struve FA. Human sensitivity to weak magnetic fields. Lancet 1991;338:1521–2. Beuter A, Edwards R. Using frequency domain characteristics to discriminate physiologic and parkinsonian tremors. J Clin Neurophysiol 1999;16: 484–94.

545

Beuter A, Glass L, Mackey MC, Titcombe MS. Nonlinear dynamics in physiology and medicine. New York: Springer Verlag; 2003. Britton TC, Thompson PD, Day BL, Rothwell JC, Findley LJ, Marsden CD. Modulation of postural wrist tremors by magnetic stimulation of the motor cortex in patients with Parkinson’s disease or essential tremor and in normal subjects mimicking tremor. Ann Neurol 1993;33:473–9. Chen R, Classen J, Gerloff C, Celnik P, Wassermann EM, Hallett M, et al. Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation. Neurology 1997;48:1398–403. Cook CM, Thomas AW, Prato FS. Resting EEG is affected by exposure to a pulsed ELF magnetic field. Bioelectromagnetics 2004;25:196–203. Cook MR, Graham C, Cohen HD, Gerkovich MM. A replication study of human exposure to 60-Hz fields: effects on neurobehavioral measures. Bioelectromagnetics 1992;13:261–85. Crasson M, Legros JJ, Scarpa P, Legros W. 50 Hz magnetic field exposure influence on human performance and psychophysiological parameters: two double-blind experimental studies. Bioelectromagnetics 1999;20:474–86. Edwards R, Beuter A. Using time domain characteristics to discriminate physiologic and parkinsonian tremors. J Clin Neurophysiol 2000;17: 87–100. Gironell A, Kulisevsky J, Lorenzo J, Barbanoj M, Pascual-Sedano B, Otermin P. Transcranial magnetic stimulation of the cerebellum in essential tremor: a controlled study. Arch Neurol 2002;59:413–7. Heusser K, Tellschaft D, Thoss F. Influence of an alternating 3 Hz magnetic field with an induction of 0.1 millitesla on chosen parameters of the human occipital EEG. Neurosci Lett 1997;239:57–60. Kazantzis N, Podd J, Whittington C. Acute effects of 50 Hz, 100 microT magnetic field exposure on visual duration discrimination at two different times of the day. Bioelectromagnetics 1998;19:310–7. Lakie M, Walsh EG, Arblaster LA, Villagra F, Roberts RC. Limb temperature and human tremors. J Neurol Neurosurg Psychiatry 1994;57:35–42. Legros A, Beuter A. Effect of a low intensity magnetic field on human motor behavior. Bioelectromagnetics 2005;26(8):657–69. Legros A, Gaillot P, Beuter A. Transient effect of low-intensity magnetic field on human motor control. Med Eng Phys, in press. Levallois P. Hypersensitivity of human subjects to environmental electric and magnetic field exposure: a review of the literature. Environ Health Perspect 2002;110(Suppl. 4):613–8. Lundervold DA, Belwood MF, Craney JL, Poppen R. Reduction of tremor severity and disability following behavioral relaxation training. J Behav Ther Exp Psychiatry 1999;30:119–35. Lyskov E, Juutilainen J, Jousmaki V, Hanninen O, Medvedev S, Partanen J. Influence of short-term exposure of magnetic field on the bioelectrical processes of the brain and performance. Int J Psychophysiol 1993a;14: 227–31. Lyskov E, Sandstrom M, Hansson Mild K. Neurophysiological study of patients with perceived ‘electrical hypersensitivity’. Int J Psychophysiol 2001;42:233–41. Lyskov EB, Juutilainen J, Jousmaki V, Partanen J, Medvedev S, Hanninen O. Effects of 45-Hz magnetic fields on the functional state of the human brain. Bioelectromagnetics 1993b;14:87–95. Marino AA, Nilsen E, Chesson AL Jr, Frilot C. Effect of low-frequency magnetic fields on brain electrical activity in human subjects. Clin Neurophysiol 2004;115:1195–201. Nguyen DH, Richard D, Plante M. Syste`me de ge´ne´ration du champ magne´tique et de commande des essais pour l’e´tude de l’effet des champs magne´tiques sur le syste`me nerveux central et pe´riphe´rique par l’exploration des tremblements. Montre´al: Hydro-Que´bec; 2004. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97–113. Pascual-Leone A, Valls-Sole J, Toro C, Wassermann EM, Hallett M. Resetting of essential tremor and postural tremor in Parkinson’s disease with transcranial magnetic stimulation. Muscle Nerve 1994;17:800–7. Podd J, Abbott J, Kazantzis N, Rowland A. Brief exposure to a 50 Hz, 100 microT magnetic field: effects on reaction time, accuracy, and recognition memory. Bioelectromagnetics 2002;23:189–95. Podd JV, Whittington CJ, Barnes GR, Page WH, Rapley BI. Do ELF magnetic fields affect human reaction time? Bioelectromagnetics 1995;16:317–23.

546

A. Legros, A. Beuter / NeuroToxicology 27 (2006) 534–546

Preece AW, Wesnes KA, Iwi GR. The effect of a 50 Hz magnetic field on cognitive function in humans. Int J Radiat Biol 1998;74: 463–70. Repacholi MH, Greenebaum B. Interaction of static and extremely low frequency electric and magnetic fields with living systems: health effects and research needs. Bioelectromagnetics 1999;20:133–60. Thomas AW, Drost DJ, Prato FS. Human subjects exposed to a specific pulsed (200 microT) magnetic field: effects on normal standing balance. Neurosci Lett 2001;297:121–4.

Timmer J, Gantert C, Deuschl G, Honerkamp J. Characteristics of hand tremor time series. Biol Cybern 1993;70:75–80. Tyrer PJ, Bond AJ. Diurnal variation in physiological tremor. Electroencephalogr Clin Neurophysiol 1974;37:35–40. Wachs H, Boshes B. Tremor studies in normals and parkinsonism. Arch Neurol 1966;66–82. Whittington CJ, Podd JV, Rapley BR. Acute effects of 50 Hz magnetic field exposure on human visual task and cardiovascular performance. Bioelectromagnetics 1996;17:131–7.