- Email: [email protected]
High frequency repetitive transcranial magnetic stimulation (rTMS) of the left dorsolateral cortex: EEG topography during waking and subsequent sleep
Psychiatry Research: Neuroimaging Section 107 Ž2001. 1᎐9
High frequency repetitive transcranial magnetic stimulation ž rTMS/ of the left dorsolateral cortex: EEG topography during waking and subsequent sleep Thomas Graf b, Judith Engeler a , Peter Achermann b, Urs P. Mosimanna , Roger Nossa , Hans-Ulrich Fischa , Thomas E. Schlaepfer a,U a
Psychiatric Neuroimaging Group, Uni¨ ersity Hospital of Bern, Murtenstrasse 21, 3010 Bern, Switzerland b Zurich, Switzerland Institute of Pharmacology and Toxicology, Uni¨ ersity of Zurich, ¨ ¨ Received 20 July 2000; received in revised form 8 March 2001; accepted 22 April 2001
Abstract Repetitive transcranial magnetic stimulation ŽrTMS. is a novel research tool in neurology and psychiatry. It is currently being evaluated as a conceivable alternative to electroconvulsive therapy for the treatment of mood disorders. Eight healthy young Žage range 21᎐25 years. right-handed men without sleep complaints participated in the study. Two sessions at a 1-week interval, each consisting of an adaptation night Žsham stimulation. and an experimental night ŽrTMS in the left dorsolateral prefrontal cortex or sham stimulation; crossover design., were scheduled. In each subject, 40 trains of 2-s duration of rTMS Žinter-train interval 28 s. were applied at a frequency of 20 Hz Ži.e. 1600 pulses per session. and at an intensity of 90% of the motor threshold. Stimulations were scheduled 80 min before lights off. The waking EEG was recorded for 10-min intervals ; 30 min prior to and after the 20-min stimulations, and polysomnographic recordings were obtained during the subsequent sleep episode Ž23.00᎐07.00 h.. The power spectra of two referential derivations, as well as of bipolar derivations along the antero-posterior axis over the left and right hemispheres, were analyzed. rTMS induced a small reduction of sleep stage 1 Žin min and percentage of total sleep time. over the whole night and a small enhancement of sleep stage 4 during the first
U
Corresponding author. Tel.: q41-31-632-8945; fax: q41-31-632-8944. E-mail address: [email protected] ŽT.E. Schlaepfer..
0925-4927r01r$ - see front matter 䊚 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 9 2 5 - 4 9 2 7 Ž 0 1 . 0 0 0 8 3 - X
2
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
non-REM sleep episode. Other sleep variables were not affected. rTMS of the left dorsolateral cortex did not alter the topography of EEG power spectra in waking following stimulation, in the all-night sleep EEG, or during the first non-REM sleep episode. Our results indicate that a single session of rTMS using parameters like those used in depression treatment protocols has no detectable side effects with respect to sleep in young healthy males. 䊚 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Repetitive transcranial magnetic stimulation; Brain stimulation; Sleep EEG; Waking EEG; Spectral analysis; Polysomnography
1. Introduction Repetitive transcranial magnetic stimulation ŽrTMS. is a novel research tool used in neurology and psychiatry. In addition to this use, it is currently under evaluation as a possible alternative modality to electroconvulsive therapy for the treatment of mood disorders ŽGeorge et al., 1999; Post et al., 1998, 1999.. rTMS allows the focal stimulation of small, limited cortical areas by the induction of electrical currents which are induced by a strong magnetic field under a stimulating coil. This stimulation results in a local hypermetabolism ŽGeorge et al., 1996.. Depending on rTMS parameters Žlow or high frequency stimulation., a lasting inhibition or facilitation of cortical excitability can be induced ŽPascual-Leone et al., 1998.. Cortical excitability is modulated beyond the duration of rTMS trains. Left prefrontal rTMS in patients with medication-resistant major depression improved mood ratings for up to 3 months post-stimulation ŽTriggs et al., 1999.. Recently Cohrs et al. Ž1998. applied high-frequency rTMS to young healthy subjects prior to sleep. They reported a prolonged REM sleep latency and no effect on mood ratings. We investigated whether local activation of a specific brain region during wakefulness by highfrequency rTMS would affect the EEG recorded above this region during waking and subsequent sleep. 2. Methods 2.1. Subjects Eight healthy young right-handed men Žmean age 22.8" 1.7 S.E.M., range 21᎐25 years. partici-
pated in the study. Handedness was evaluated by the Edinburgh Inventory ŽOldfield, 1971.. All subjects were naive regarding the rTMS procedure and a screening night served to exclude those with sleep apnea and nocturnal myoclonus. The subjects reported themselves to be in good health and free of sleep complaints. All subjects were screened for psychiatric Že.g. drug abuse, affective disorders. and medical morbidity in a comprehensive interview. One subject had to be excluded due to psychiatric symptoms. The study protocol was approved by the Institutional Review Board of the University of Bern. Informed written consent was obtained from the subjects, and they were paid for their participation. Subjects were instructed to abstain from caffeine and alcohol and to keep a regular sleep᎐wake schedule Ž23:00᎐07:00 h. for at least 3 days prior to the study. Compliance with the latter instruction was verified with activity monitors worn on the wrist of the non-dominant arm. 2.2. Study protocol and rTMS procedures Two sessions at a 1-week interval, each consisting of an adaptation night Žsham stimulation. and an experimental night ŽrTMS in the dorsolateral prefrontal cortex or sham stimulation; crossover design., were scheduled. All subjects were stimulated 80 min before lights off. Ten-minute waking EEGs were recorded approximately 30 min before and after rTMS. During these EEG recordings, subjects were lying in bed, fixating a black dot on the ceiling. Time in bed after lights off Ž; 23:00 h. was 8 h. Stimulation was performed with a Cadwell High Speed MES-10 ŽFDA Reference 噛K905059D.
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
rTMS stimulator with a figure eight-shaped water-cooled coil Žouter diameter of each coil, 7.5 cm.. The same investigator performed all stimulations. Motor threshold was determined individually before sham and real stimulations. The stimulation intensity was 90% of the motor threshold of the right abductor pollicis brevis muscle. Safety guidelines of the International Society of Transcranial Stimulation ŽISTS. ŽWassermann, 1998. were followed. Forty trains Žinter-train interval, 28 s. of 2-s duration at a frequency of 20 Hz were applied for 20 min Ži.e. 1600 pulses per session.. The following protocols were used: 䢇
䢇
䢇
Adaptation night: The subject sat for 20 min in the stimulation chair. The coil was held 25 cm to the left of the head at an angle of 90⬚ away from the subject Žadaptation to the noise of rTMS.. Real stimulation: The stimulation location was 5 cm anterior and 2 cm to the left of the cortical representation of the contralateral abductor pollicis brevis muscle with the handle of the stimulating coil pointing backwards. Sham stimulation: The subject sat for 20 min in the stimulation chair; the left edge of the coil was put at an angle of 90⬚, at the same position as during the real stimulation, thus minimizing the magnetic field reaching the cortex. All other parameters were exactly the same as in the real stimulation setting.
2.3. Visual analog scales To assess effects of rTMS on mood, anxiety, tiredness and discomfort, standard visual analog scales were used ŽCohrs et al., 1998; Dearing et al., 1997; George et al., 1996.. 2.4. EEG recording and analysis The electroencephalogram ŽEEG., electrooculogram ŽEOG ., submental electromyogram ŽEMG., and electrocardiogram ŽECG. were recorded by a polygraphic amplifier ŽPSA24, Braintronics Inc., Almere, The Netherlands., digitized, and transmitted via fiber-optic cables to a
3
personal computer Žfor details, see Endo et al., 1998.. EEG, EMG and EOG signals were conditioned by the following analog filters: a high-pass filter Žy3 dB at 0.16 Hz., a low-pass filter Žy3 dB at 102 Hz, - y40 dB at 256 Hz., and a notch filter Ž50 Hz.. Data were sampled with a frequency of 512 Hz, digitally filtered ŽEEG and EOG: low-pass FIR filter, y3 dB at 49 Hz; EMG: band-pass FIR filter,y 3 dB points at 15.6 and 54 Hz., and stored with a resolution of 128 Hz. EEG electrodes were placed at the locations F3, F4, C3, C4, P3, P4, O1, O2 and Cz Žderivations: C3A2, all other electrodes referenced against Cz.. Standard gold electrodes were used for EEG recordings and remained on the scalp during rTMS stimulation. Exposing the electrodes to 400 pulses at 20 Hz at 100% of the maximal machine stimulator output power showed that the temperature of the electrodes Ždue to heating induced by rTMS. did not exceed 40⬚C. This is a much stronger stimulation than used in the actual experiment Ž20 pulses at 20 Hz at approx. 65% of maximal stimulator output every 28 s., thus demonstrating the safety of stimulating with EEG electrodes at therapeutic parameters with regards to overheating of the electrodes. EEG recordings were carried out for 10-min epochs prior to and after stimulation as well as during the subsequent nocturnal sleep Ž23.00᎐07.00 h.. The power spectra of 4-s epochs ŽFFT routine, Hanning window. were computed ŽAchermann and Borbely, ´ 1997.. Artifacts were identified by visual inspection and on the basis of power in the 0.75᎐4.5-Hz and 20᎐40-Hz bands. Only 4-s epochs without artifacts were used for spectral analysis. For the waking EEG before and after real or sham stimulation, between 13 and 127 artifact-free 4-s epochs were averaged Žmean 87, S.D. 21.. For the sleep EEG, power spectra of five consecutive 4-s epochs were averaged and matched with the sleep scores. Sleep stages were visually scored for consecutive 20-s epochs according to the criteria of Rechtschaffen and Kales Ž1968.. Only artifact-free 20-s epochs Žfive consecutive 4-s epochs. were used for calculating the average spectra over the whole night and the first non-REM sleep episode. Differences in power spectra of referential
4
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
derivations ŽC3A2 and F3A2. were determined for both the waking EEG Žbefore and after stimulation. as well as for the sleep EEG Žnights following real and sham stimulation.. Regional differences were further investigated by comparing the power spectra of bipolar derivations along the antero-posterior axis over the left ŽF3C3, C3P3, P3O1. and the right ŽF4C4, C4P4, P4O2. hemisphere. The derivation P4O2 could not be used due to technical problems in several subjects. Sleep variables and spectral data were tested with ANOVAs for repeated measures ŽrANOVA. with the between-factor ‘order’ ŽrTMS during first or second session. and the within-factor ‘treatment’ ŽrTMS or sham stimulation.. Tests were carried out for each 0.25-Hz bin and for five traditional EEG frequency bands Ž ␦: 0.625᎐4.375 Hz; : 4.375᎐7.875 Hz; ␣: 7.875᎐11.875 Hz; : 11.875᎐14.875 Hz; : 14.875᎐24.875 Hz.. Significant effects were assessed by two-tailed paired t-tests. Sleep variables and EEG power density values were log-transformed prior to statistical tests. Spectral data of the waking EEG were tested by comparing the ratios after and before real stimulation with the ratios after and before sham stimulation. Regional differences were analyzed by comparing bipolar derivations with rANOVAs with the factors ‘treatment’ ŽrTMS or sham stimulation., ‘hemisphere’ Žleft, right., and ‘derivation’ Žfrontalo-central, centralo-parietal .. An additional factor ‘sequence’ Žbefore and after real or sham stimulation. was used for the waking EEGs. Due to technical problems, the parietooccipital derivations were not used in these rANOVAs.
3. Results No subject had to be excluded from the study, but one subject complained of moderate pain in the region of the left trigeminal nerve during real stimulation. No other side effects were observed, and no effect of rTMS on mood, anxiety or tiredness was detected with the visual analog scales. rTMS of the left dorsolateral cortex caused a minor reduction of sleep stage 1 ŽTable 1.. A significant interaction Ž‘treatment’= ‘order’. was
Table 1 Sleep variables after real rTMS and sham stimulation Žmean and S.E.M., N s 8.
U
Stage 1 Ž%. Stage 2 Ž%. SWS Ž%. REM sleep Ž%. SL Žmin. RL Žmin.q TST Žmin. Sleep efficiency Ž%.
rTMS
Sham
8.1" 1.8 48.9" 1.1 20.1" 2.5 22.9" 1.3 14.3" 1.9 72 " 8 444 " 4 92.6" 0.8
9.9" 1.6 46.8" 2.1 19.5" 2.1 23.9" 1.9 10.3" 2.4 79 " 17 445 " 6 92.9" 1.3
SWS: slow wave sleep Žstages 3 and 4.. SL: Sleep latency Žmin from lights off to first epoch of stage 2.. RL: REM sleep latency Žmin from sleep onset to first epoch of REM sleep.. TST: Total sleep time. Sleep efficiency Žtime asleep as percentage of time in bed.. Percentages are expressed with respect to TST. U Significant reduction compared with sham Ž P- 0.05; factor ‘treatment’ of two-way rANOVAs.. A significant interaction Ž‘treatment’= ‘order’. was observed for the sleep latency, together with trends Ž P- 0.1. for the factors ‘treatment’ and ‘order’. This was the result of a shorter average sleep latency of the four subjects who were sham stimulated in the first experimental session in the night following sham stimulation. qOne subject was excluded from this comparison because of a skipped first REM episode during the night following sham stimulation.
observed for the sleep latency, together with trends Ž P- 0.1. for the factors ‘treatment’ and ‘order’. This was the result of a shorter average sleep latency ŽSL. in the night following sham stimulation of the four subjects which were sham stimulated in the first experimental session. None of the other sleep variables revealed significant differences. Contrary to Cohrs et al. Ž1998., no significant increase of REM sleep latency ŽRL; Fig. 1., RL plus SL ŽSL: sleep latency., or the non-REM᎐REM sleep cycle length was observed. One subject was excluded from this comparison because of a skipped first REM episode during the night following sham stimulation. Analyzing the non-REM᎐REM sleep cycles separately revealed that in the first non-REM sleep episode, the occurrence of sleep stage 4 was slightly enhanced wreal: 18.2 Ž3.9 S.E.M.. min, sham: 13.8 Ž2.9. min; P- 0.05, N s 7, factor ‘treatment’ of two-way rANOVAx. The waking EEG Žreferential C3A2, F3A2 or
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
Fig. 1. REM sleep latencies Žh. for individual subjects Ž N s 7. in the nights following sham and real stimulation. One subject was excluded from this comparison because of a skipped first REM sleep episode during the night following sham stimulation.
bipolar. was analyzed by comparing power spectra after real and sham stimulation and by comparing
5
the ratios afterrbefore real or sham stimulation with each other ŽFig. 2.. Tests carried out for all 0.25-Hz bins revealed that the factors ‘treatment’ and ‘order’ and the interaction ‘treatment’= ‘order’ were significant in only a few bins. The factor ‘treatment’ was significant in two consecutive bins Ž10.375᎐10.875 Hz. for the derivation F3C3 ŽFig. 2. and in three consecutive bins Ž10.625᎐11.375 Hz. for the derivation C3A2. However, considering that 100 rANOVAs were carried out for each spectrum, these deviations with P-values ) 0.01 do not amount to a significant effect due to rTMS in the waking EEG. Additional tests based on traditional frequency bands yielded no significant effects for the factor ‘treatment’. The factor ‘order’ was significant in three of 40 comparisons and the interaction ‘treatment’= ‘order’ was significant in one case. The sleep EEG was analyzed by comparing power spectra for nights following sham and real stimulation. Whole night spectra, as well as spectra for the first non-REM sleep episode ŽFig. 3.,
Fig. 2. Mean EEG power density spectrum ŽF3C3, N s 8. during waking. The curves depict power in the 10-min recording after rTMS or sham stimulation Žmean " 1 S.E.M. for 0.25-Hz bins. expressed relative to the corresponding value of the recording before the stimulation. Bottom bars indicate frequency bins for which power was significantly different after rTMS than after sham stimulation Žfactor ‘treatment’ of two-way rANOVAs.. Note that P-values are plotted in the reverse direction. Factor ‘order’ was significant only in one frequency bin and a significant interaction ‘treatment’= ‘order’ was observed in two single bins.
6
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
were investigated by testing all 0.25-Hz bins and the EEG frequency bands ␦, , ␣ , , . Again, the factors ‘treatment’ and ‘order’ and the interaction ‘treatment’= ‘order’ were significant in only a few single 0.25-Hz bins. The same tests, that were applied to the broader frequency bands, revealed a significant increase Ž P- 0.05, factor ‘condition’. of ␦ power in the derivation F4C4 in the first non-REM sleep episode. Regional differences of the waking and sleep EEG were further analyzed by comparing spectral power in the frequency bands ␦, , ␣ , and  of bipolar derivations with three- and four-way rANOVAs. No significant differences were found for the sleep EEG. For the waking EEG, the factor ‘sequence’ Žbefore and after real or sham stimulation. was significant in all bands, except the ␣ band. This is the result of lower EEG power in the waking recordings after real or sham stimulation ŽFig. 2.. In addition, the factor ‘derivation’ and the interaction ‘condition’= ‘sequence’= ‘derivation’ were significant in the ␦
band ŽFig. 4.. A number of post-hoc tests were carried out in order to understand the nature of this interaction: 1. Three-way rANOVAs with the factors ‘derivation’, ‘hemisphere’ and ‘condition’ were carried out with the data of the waking recordings after real or sham stimulation and with the relative changes afterrbefore real or sham stimulation. 2. Three-way rANOVAs with the factors ‘condition’, ‘sequence’, and ‘derivation’ or ‘hemisphere’ were carried out for pairs of bipolar derivations Ž F3C3 ᎐ F4C4, F3C3 ᎐ C3P3, F4C4᎐C4P4, C3P3᎐C4P4.. 3. Two-way rANOVAs with the factors ‘condition’ and ‘sequence’ were carried out for each of the four derivations. The factor ‘condition’ or an interaction involving this factor was not significant in any of these rANOVAs.
Fig. 3. Mean EEG power density spectrum ŽC3A2, N s 87. in non-REM sleep Žstages 2, 3 and 4. of the first non-REM sleep episode. The curve depicts power in the night following real stimulation Žmean " 1 S.E.M. for 0.25-Hz bins. expressed relative to the corresponding value of the night following sham stimulation. Bottom bars indicate frequency bins for which power was significantly different after rTMS than after sham stimulation Žfactor ‘treatment’ of two-way rANOVAs.. Note that P-values are plotted in the reverse direction. A significant interaction ‘treatment’= ‘order’ was observed in three single bins.
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
7
Fig. 4. Relative ␦ power Žmean " 1 S.E.M.. of bipolar derivations in the wake recordings before and after real and sham stimulation. For each derivation, power was normalized to the average over all four recordings. A four-way rANOVA revealed that the factors ‘sequence’ and ‘derivation’ and the interaction ‘condition’=‘sequence’= ‘derivation’ were significant.
4. Discussion We investigated whether local activation of a specific brain region by rTMS with parameters used in depression treatment trials during wakefulness affects the EEG recorded from this region during waking and subsequent sleep. Unilateral rTMS of the left dorsolateral cortex applied to left prefrontal areas did not alter the waking and sleep EEG and caused only a minor reduction of sleep stage 1 and a slightly enhanced occurrence of sleep stage 4 in the first non-REM sleep episode. Contrary to the findings of Cohrs et al. Ž1998., no significant increase of REM sleep latency ŽRL; Fig. 1., RL q SL, nor the nonREM᎐REM sleep cycle length was observed. In the alpha range of the waking EEG, the factor ‘treatment’ of the two-way rANOVAs was significant for two and three consecutive bins for the derivations F3C3 ŽFig. 2. and C3A2, respectively. Because the P-values are ) 0.01, the effect is not statistically significant if corrections for multiple testing are considered. Nevertheless, alpha activity was somewhat enhanced in the waking recordings after real and sham stimulations when compared with the recordings before the stimulations ŽFig. 2.. This observation may indi-
cate that subjects became more relaxed over the course of an experimental session, in particular after sham stimulation. Power in the ␦, , and bands was lower in the waking recordings after real or sham stimulation ŽFigs. 2 and 4, factor ‘sequence’ in four-way rANOVAs.. In addition, the interaction ‘condition’= ‘sequence’= ‘derivation’ was significant in the ␦ band, but none of the post-hoc tests showed any significant effects involving the factor ‘condition’. Small variations in the non-specific effect of lower power in the wake recordings after real or sham stimulation may, therefore, be responsible for the observed interaction. During the first non-REM sleep episode, ␦ power in the derivation F4C4 was significantly enhanced Ž P- 0.05.. Considering corrections due to multiple testing, it is questionable whether this represents a real effect. However, the increased ␦ power may be related to the enhanced occurrence of sleep stage 4 in the first non-REM sleep episode. In summary, the observed effects on both the waking and sleep EEG and its topography are small, and may, in some cases, be chance effects arising from the large number of tests or may be related to non-specific effects Že.g. lower power in
8
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
the waking recording after real or sham exposure.. Different factors may be responsible for the absence of clear EEG effects due to rTMS: 1. Stimulation was too focal to result in detectable EEG changes; 2. Stimulus strength Žduration andror intensity . was insufficient to initiate long-term EEG changes. Indeed, a vibratory stimulus of 6-h duration applied to the dominant hand prior to sleep was needed to induce increased slow-wave activity over the somatosensory cortex during the first hour of sleep ŽKattler et al., 1994.. On the other hand, 15᎐30 min exposure to EMF Žcorresponding to the radiation emitted by mobile phones. resulted in increased power in the sigma range ŽBorbely ´ et al., 1999; Huber et al., 2000.. Recently, we demonstrated ŽMosimann et al., 2000. that rTMS at 100% of the motor threshold may not have a modulatory effect on mood as was reported in previous studies ŽPascual-Leone et al., 1996; Wassermann, 1998.. Effects of rTMS may possibly only be detected at stimulation intensities much higher than those used previously. 3. Placement of the electrodes did not allow us to detect changes induced by rTMS. 4. The sample size might have been too small to detect significant effects. Our results demonstrate that rTMS applied in a therapeutic dose in young healthy males had no detectable effects on sleep and the sleep EEG. This shows that the putative anti-depressant effect of the method might not be mediated by strong influences on sleep patterns, and that rTMS is a safe and non-invasive method with respect to its influence on sleep. Furthermore, it is interesting to observe that while acute rTMS stimulation induces changes in regional cerebral blood flow ŽGeorge et al., 1999., we were unable to detect EEG changes during the subsequent night. Novel therapeutic agents must undergo extensive examination of safety in animals and healthy volunteers before testing their efficacy in clinical
trials. This scenario has been, to some extent, reversed with respect to rTMS applications in psychiatric conditions. This might be partly due to the great enthusiasm created by positive clinical trials in major depression and mania ŽGeorge et al., 1995, 1997; Klein et al., 1999; Menkes et al., 1999; Pascual-Leone et al., 1996.. Despite the fact that rTMS has shown some potential to positively influence psychiatric symptoms, the possibility for short- and long-term adverse effects on cognition and behavior has to be assessed rigorously. The message conveyed by this study is a mixed one: While we were not able to show that the anti-depressant effects of rTMS might be mediated by a clinically significant influence on sleep, we demonstrated that rTMS applied under therapeutic parameters is safe in regard to its influence on sleep in healthy young subjects.
Acknowledgements We thank Cornelia Huber, Viola Meier and Sandra Puliafito for help with the data collection and Dr Irene Tobler for comments on the manuscript. This study was supported by the Swiss National Science Foundation Grants Nr. 3100053005.97, 4038-044046.96 and 3231-044523.96 and the Human Frontiers Science Program grant RG-81r96. References Achermann, P., Borbely, ´ A.A., 1997. Low-frequency Ž- 1 Hz. oscillations in the human sleep EEG. Neuroscience 81, 213᎐222. Borbely, ´ A.A., Huber, R., Graf, T., Fuchs, B., Gallmann, E., Achermann, P., 1999. Pulsed high-frequency electromagnetic field affects human sleep and sleep electroencephalogram. Neuroscience Letters 275, 207᎐210. Cohrs, S., Tergau, F., Riech, S., Kastner, S., Paulus, W., Ziemann, U., Ruther, E., Hajak, G., 1998. High-frequency repetitive transcranial magnetic stimulation delays rapid eye movement sleep. Neuroreport 9, 3439᎐3443. Dearing, J., George, M.S., Greenberg, B.D., Wassermann, E.M., Schlaepfer, T.E., Murphy, D.L., Hallet, M., Post, R.M., 1997. Mood effects of prefrontal repetitive high frequency transcranial magnetic stimulation ŽrTMS. in healthy volunteers. CNS Spectrums 2, 53᎐68. Endo, T., Roth, C., Landolt, H.P., Werth, E., Aeschbach, D., Achermann, P., Borbely, ´ A.A., 1998. Selective REM sleep
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9 deprivation in humans: effects on sleep and sleep EEG. American Journal of Physiology 274, R1186᎐R1194. George, M.S., Lisanby, S.H., Sackeim, H.A., 1999. Transcranial magnetic stimulation: applications in neuropsychiatry. Archives of General Psychiatry 56, 300᎐311. George, M.S., Wassermann, E.M., Kimbrell, T.A., Little, J.T., Williams, W.E., Danielson, A.L., Greenberg, B.D., Hallett, M., Post, R.M., 1997. Mood improvement following daily left prefrontal repetitive transcranial magnetic stimulation in patients with depression: a placebo-controlled crossover trial. American Journal of Psychiatry 154, 1752᎐1756. George, M.S., Wassermann, E.M., Williams, W.A., Callahan, A., Ketter, T.A., Basser, P., Hallett, M., Post, R.M., 1995. Daily repetitive transcranial magnetic stimulation ŽrTMS. improves mood in depression. Neuroreport 6, 1853᎐1856. George, M.S., Wassermann, E.M., Williams, W.A., Steppel, J., Pascual-Leone, A., Basser, P., Hallett, M., Post, R.M., 1996. Changes in mood and hormone levels after rapid-rate transcranial magnetic stimulation ŽrTMS. of the prefrontal cortex. Journal of Neuropsychiatry and Clinical Neuroscience 8, 172᎐180. Huber, R., Graf, T., Cote, K.A., Wittmann, L., Gallmann, E., Matter, D., Schuderer, J., Kuster, N., Borbely, ´ A.A., Achermann, P., 2000. Exposure to pulsed high-frequency electromagnetic field during waking affects human sleep EEG. Neuroreport 11, 3321᎐3325. Kattler, H., Dijk, D.J., Borbely, ´ A.A., 1994. Effect of unilateral somatosensory stimulation prior to sleep on the sleep EEG in humans. Journal of Sleep Research 3, 159᎐164. Klein, E., Kreinin, I., Chistyakov, A., Koren, D., Mecz, L., Marmur, S., Ben-Shachar, D., Feinsod, M., 1999. Therapeutic efficacy of right prefrontal slow repetitive transcranial magnetic stimulation in major depression: a double-blind controlled study. Archives of General Psychiatry 56, 315᎐320. Menkes, D.L., Bodnar, P., Ballesteros, R.A., Swenson, M.R., 1999. Right frontal lobe slow frequency repetitive transcranial magnetic stimulation ŽSF r-TMS. is an effective treatment for depression: a case-control pilot study of safety and efficacy. Journal of Neurology, Neurosurgery and Psychiatry 67, 113᎐115.
9
Mosimann, U.P., Rihs, T.A., Engeler, J., Fisch, H.U., Schlaepfer, T.E., 2000. Mood effects of repetitive transcranial stimulation ŽrTMS. of left prefrontal cortex in healthy volunteers. Psychiatry Research 94, 251᎐256. Oldfield, R.C., 1971. The assessment and analysis of handedness ᎏ the Edinburgh inventory. Neuropsychologia 9, 97᎐113. Pascual-Leone, A., Catala, M.D., Pascual-Leone Pascual, A., 1996. Lateralized effect of rapid-rate transcranial magnetic stimulation of the prefrontal cortex on mood. Neurology 46, 499᎐502. Pascual-Leone, A., Tormos, J.M., Keenan, J., Tarazona, F., Canete, C., Catala, M.D., 1998. Study and modulation of human cortical excitability with transcranial magnetic stimulation. Journal of Clinical Neurophysiology 15, 333᎐343. Post, R.M., Frye, M.A., Denicoff, K.D., Leverich, G.S., Kimbrell, T.A., Dunn, R.T., 1998. Beyond lithium in the treatment of bipolar illness. Neuropsychopharmacology 19, 206᎐219. Post, R.M., Kimbrell, T.A., McCann, U.D., Dunn, R.T., Osuch, E.A., Speer, A.M., Weiss, S.R., 1999. Repetitive transcranial magnetic stimulation as a neuropsychiatric tool: present status and future potential. Journal of Electroconvulsive Therapy ŽJECT. 15, 39᎐59. Rechtschaffen, A., Kales, A., 1968. A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. National Institutes of Health, Bethesda, Maryland. Triggs, W.J., McCoy, K.J., Greer, R., Rossi, F., Bowers, D., Kortenkamp, S., Nadeau, S.E., Heilman, K.M., Goodman, W.K., 1999. Effects of left frontal transcranial magnetic stimulation on depressed mood, cognition, and corticomotor threshold. Biological Psychiatry 45, 1440᎐1446. Wassermann, E., 1998. Risk and safety of repetitive transcranial magnetic stimulation: report and suggested guidelines from the International Workshop on the Safety of Repetitive Transcranial Magnetic Stimulation, June 5᎐7, 1996. Electroencephalograpy and Clinical Neurophysiology 108, 1᎐16.
High frequency repetitive transcranial magnetic stimulation ž rTMS/ of the left dorsolateral cortex: EEG topography during waking and subsequent sleep Thomas Graf b, Judith Engeler a , Peter Achermann b, Urs P. Mosimanna , Roger Nossa , Hans-Ulrich Fischa , Thomas E. Schlaepfer a,U a
Psychiatric Neuroimaging Group, Uni¨ ersity Hospital of Bern, Murtenstrasse 21, 3010 Bern, Switzerland b Zurich, Switzerland Institute of Pharmacology and Toxicology, Uni¨ ersity of Zurich, ¨ ¨ Received 20 July 2000; received in revised form 8 March 2001; accepted 22 April 2001
Abstract Repetitive transcranial magnetic stimulation ŽrTMS. is a novel research tool in neurology and psychiatry. It is currently being evaluated as a conceivable alternative to electroconvulsive therapy for the treatment of mood disorders. Eight healthy young Žage range 21᎐25 years. right-handed men without sleep complaints participated in the study. Two sessions at a 1-week interval, each consisting of an adaptation night Žsham stimulation. and an experimental night ŽrTMS in the left dorsolateral prefrontal cortex or sham stimulation; crossover design., were scheduled. In each subject, 40 trains of 2-s duration of rTMS Žinter-train interval 28 s. were applied at a frequency of 20 Hz Ži.e. 1600 pulses per session. and at an intensity of 90% of the motor threshold. Stimulations were scheduled 80 min before lights off. The waking EEG was recorded for 10-min intervals ; 30 min prior to and after the 20-min stimulations, and polysomnographic recordings were obtained during the subsequent sleep episode Ž23.00᎐07.00 h.. The power spectra of two referential derivations, as well as of bipolar derivations along the antero-posterior axis over the left and right hemispheres, were analyzed. rTMS induced a small reduction of sleep stage 1 Žin min and percentage of total sleep time. over the whole night and a small enhancement of sleep stage 4 during the first
U
Corresponding author. Tel.: q41-31-632-8945; fax: q41-31-632-8944. E-mail address: [email protected] ŽT.E. Schlaepfer..
0925-4927r01r$ - see front matter 䊚 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 9 2 5 - 4 9 2 7 Ž 0 1 . 0 0 0 8 3 - X
2
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
non-REM sleep episode. Other sleep variables were not affected. rTMS of the left dorsolateral cortex did not alter the topography of EEG power spectra in waking following stimulation, in the all-night sleep EEG, or during the first non-REM sleep episode. Our results indicate that a single session of rTMS using parameters like those used in depression treatment protocols has no detectable side effects with respect to sleep in young healthy males. 䊚 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Repetitive transcranial magnetic stimulation; Brain stimulation; Sleep EEG; Waking EEG; Spectral analysis; Polysomnography
1. Introduction Repetitive transcranial magnetic stimulation ŽrTMS. is a novel research tool used in neurology and psychiatry. In addition to this use, it is currently under evaluation as a possible alternative modality to electroconvulsive therapy for the treatment of mood disorders ŽGeorge et al., 1999; Post et al., 1998, 1999.. rTMS allows the focal stimulation of small, limited cortical areas by the induction of electrical currents which are induced by a strong magnetic field under a stimulating coil. This stimulation results in a local hypermetabolism ŽGeorge et al., 1996.. Depending on rTMS parameters Žlow or high frequency stimulation., a lasting inhibition or facilitation of cortical excitability can be induced ŽPascual-Leone et al., 1998.. Cortical excitability is modulated beyond the duration of rTMS trains. Left prefrontal rTMS in patients with medication-resistant major depression improved mood ratings for up to 3 months post-stimulation ŽTriggs et al., 1999.. Recently Cohrs et al. Ž1998. applied high-frequency rTMS to young healthy subjects prior to sleep. They reported a prolonged REM sleep latency and no effect on mood ratings. We investigated whether local activation of a specific brain region during wakefulness by highfrequency rTMS would affect the EEG recorded above this region during waking and subsequent sleep. 2. Methods 2.1. Subjects Eight healthy young right-handed men Žmean age 22.8" 1.7 S.E.M., range 21᎐25 years. partici-
pated in the study. Handedness was evaluated by the Edinburgh Inventory ŽOldfield, 1971.. All subjects were naive regarding the rTMS procedure and a screening night served to exclude those with sleep apnea and nocturnal myoclonus. The subjects reported themselves to be in good health and free of sleep complaints. All subjects were screened for psychiatric Že.g. drug abuse, affective disorders. and medical morbidity in a comprehensive interview. One subject had to be excluded due to psychiatric symptoms. The study protocol was approved by the Institutional Review Board of the University of Bern. Informed written consent was obtained from the subjects, and they were paid for their participation. Subjects were instructed to abstain from caffeine and alcohol and to keep a regular sleep᎐wake schedule Ž23:00᎐07:00 h. for at least 3 days prior to the study. Compliance with the latter instruction was verified with activity monitors worn on the wrist of the non-dominant arm. 2.2. Study protocol and rTMS procedures Two sessions at a 1-week interval, each consisting of an adaptation night Žsham stimulation. and an experimental night ŽrTMS in the dorsolateral prefrontal cortex or sham stimulation; crossover design., were scheduled. All subjects were stimulated 80 min before lights off. Ten-minute waking EEGs were recorded approximately 30 min before and after rTMS. During these EEG recordings, subjects were lying in bed, fixating a black dot on the ceiling. Time in bed after lights off Ž; 23:00 h. was 8 h. Stimulation was performed with a Cadwell High Speed MES-10 ŽFDA Reference 噛K905059D.
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
rTMS stimulator with a figure eight-shaped water-cooled coil Žouter diameter of each coil, 7.5 cm.. The same investigator performed all stimulations. Motor threshold was determined individually before sham and real stimulations. The stimulation intensity was 90% of the motor threshold of the right abductor pollicis brevis muscle. Safety guidelines of the International Society of Transcranial Stimulation ŽISTS. ŽWassermann, 1998. were followed. Forty trains Žinter-train interval, 28 s. of 2-s duration at a frequency of 20 Hz were applied for 20 min Ži.e. 1600 pulses per session.. The following protocols were used: 䢇
䢇
䢇
Adaptation night: The subject sat for 20 min in the stimulation chair. The coil was held 25 cm to the left of the head at an angle of 90⬚ away from the subject Žadaptation to the noise of rTMS.. Real stimulation: The stimulation location was 5 cm anterior and 2 cm to the left of the cortical representation of the contralateral abductor pollicis brevis muscle with the handle of the stimulating coil pointing backwards. Sham stimulation: The subject sat for 20 min in the stimulation chair; the left edge of the coil was put at an angle of 90⬚, at the same position as during the real stimulation, thus minimizing the magnetic field reaching the cortex. All other parameters were exactly the same as in the real stimulation setting.
2.3. Visual analog scales To assess effects of rTMS on mood, anxiety, tiredness and discomfort, standard visual analog scales were used ŽCohrs et al., 1998; Dearing et al., 1997; George et al., 1996.. 2.4. EEG recording and analysis The electroencephalogram ŽEEG., electrooculogram ŽEOG ., submental electromyogram ŽEMG., and electrocardiogram ŽECG. were recorded by a polygraphic amplifier ŽPSA24, Braintronics Inc., Almere, The Netherlands., digitized, and transmitted via fiber-optic cables to a
3
personal computer Žfor details, see Endo et al., 1998.. EEG, EMG and EOG signals were conditioned by the following analog filters: a high-pass filter Žy3 dB at 0.16 Hz., a low-pass filter Žy3 dB at 102 Hz, - y40 dB at 256 Hz., and a notch filter Ž50 Hz.. Data were sampled with a frequency of 512 Hz, digitally filtered ŽEEG and EOG: low-pass FIR filter, y3 dB at 49 Hz; EMG: band-pass FIR filter,y 3 dB points at 15.6 and 54 Hz., and stored with a resolution of 128 Hz. EEG electrodes were placed at the locations F3, F4, C3, C4, P3, P4, O1, O2 and Cz Žderivations: C3A2, all other electrodes referenced against Cz.. Standard gold electrodes were used for EEG recordings and remained on the scalp during rTMS stimulation. Exposing the electrodes to 400 pulses at 20 Hz at 100% of the maximal machine stimulator output power showed that the temperature of the electrodes Ždue to heating induced by rTMS. did not exceed 40⬚C. This is a much stronger stimulation than used in the actual experiment Ž20 pulses at 20 Hz at approx. 65% of maximal stimulator output every 28 s., thus demonstrating the safety of stimulating with EEG electrodes at therapeutic parameters with regards to overheating of the electrodes. EEG recordings were carried out for 10-min epochs prior to and after stimulation as well as during the subsequent nocturnal sleep Ž23.00᎐07.00 h.. The power spectra of 4-s epochs ŽFFT routine, Hanning window. were computed ŽAchermann and Borbely, ´ 1997.. Artifacts were identified by visual inspection and on the basis of power in the 0.75᎐4.5-Hz and 20᎐40-Hz bands. Only 4-s epochs without artifacts were used for spectral analysis. For the waking EEG before and after real or sham stimulation, between 13 and 127 artifact-free 4-s epochs were averaged Žmean 87, S.D. 21.. For the sleep EEG, power spectra of five consecutive 4-s epochs were averaged and matched with the sleep scores. Sleep stages were visually scored for consecutive 20-s epochs according to the criteria of Rechtschaffen and Kales Ž1968.. Only artifact-free 20-s epochs Žfive consecutive 4-s epochs. were used for calculating the average spectra over the whole night and the first non-REM sleep episode. Differences in power spectra of referential
4
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
derivations ŽC3A2 and F3A2. were determined for both the waking EEG Žbefore and after stimulation. as well as for the sleep EEG Žnights following real and sham stimulation.. Regional differences were further investigated by comparing the power spectra of bipolar derivations along the antero-posterior axis over the left ŽF3C3, C3P3, P3O1. and the right ŽF4C4, C4P4, P4O2. hemisphere. The derivation P4O2 could not be used due to technical problems in several subjects. Sleep variables and spectral data were tested with ANOVAs for repeated measures ŽrANOVA. with the between-factor ‘order’ ŽrTMS during first or second session. and the within-factor ‘treatment’ ŽrTMS or sham stimulation.. Tests were carried out for each 0.25-Hz bin and for five traditional EEG frequency bands Ž ␦: 0.625᎐4.375 Hz; : 4.375᎐7.875 Hz; ␣: 7.875᎐11.875 Hz; : 11.875᎐14.875 Hz; : 14.875᎐24.875 Hz.. Significant effects were assessed by two-tailed paired t-tests. Sleep variables and EEG power density values were log-transformed prior to statistical tests. Spectral data of the waking EEG were tested by comparing the ratios after and before real stimulation with the ratios after and before sham stimulation. Regional differences were analyzed by comparing bipolar derivations with rANOVAs with the factors ‘treatment’ ŽrTMS or sham stimulation., ‘hemisphere’ Žleft, right., and ‘derivation’ Žfrontalo-central, centralo-parietal .. An additional factor ‘sequence’ Žbefore and after real or sham stimulation. was used for the waking EEGs. Due to technical problems, the parietooccipital derivations were not used in these rANOVAs.
3. Results No subject had to be excluded from the study, but one subject complained of moderate pain in the region of the left trigeminal nerve during real stimulation. No other side effects were observed, and no effect of rTMS on mood, anxiety or tiredness was detected with the visual analog scales. rTMS of the left dorsolateral cortex caused a minor reduction of sleep stage 1 ŽTable 1.. A significant interaction Ž‘treatment’= ‘order’. was
Table 1 Sleep variables after real rTMS and sham stimulation Žmean and S.E.M., N s 8.
U
Stage 1 Ž%. Stage 2 Ž%. SWS Ž%. REM sleep Ž%. SL Žmin. RL Žmin.q TST Žmin. Sleep efficiency Ž%.
rTMS
Sham
8.1" 1.8 48.9" 1.1 20.1" 2.5 22.9" 1.3 14.3" 1.9 72 " 8 444 " 4 92.6" 0.8
9.9" 1.6 46.8" 2.1 19.5" 2.1 23.9" 1.9 10.3" 2.4 79 " 17 445 " 6 92.9" 1.3
SWS: slow wave sleep Žstages 3 and 4.. SL: Sleep latency Žmin from lights off to first epoch of stage 2.. RL: REM sleep latency Žmin from sleep onset to first epoch of REM sleep.. TST: Total sleep time. Sleep efficiency Žtime asleep as percentage of time in bed.. Percentages are expressed with respect to TST. U Significant reduction compared with sham Ž P- 0.05; factor ‘treatment’ of two-way rANOVAs.. A significant interaction Ž‘treatment’= ‘order’. was observed for the sleep latency, together with trends Ž P- 0.1. for the factors ‘treatment’ and ‘order’. This was the result of a shorter average sleep latency of the four subjects who were sham stimulated in the first experimental session in the night following sham stimulation. qOne subject was excluded from this comparison because of a skipped first REM episode during the night following sham stimulation.
observed for the sleep latency, together with trends Ž P- 0.1. for the factors ‘treatment’ and ‘order’. This was the result of a shorter average sleep latency ŽSL. in the night following sham stimulation of the four subjects which were sham stimulated in the first experimental session. None of the other sleep variables revealed significant differences. Contrary to Cohrs et al. Ž1998., no significant increase of REM sleep latency ŽRL; Fig. 1., RL plus SL ŽSL: sleep latency., or the non-REM᎐REM sleep cycle length was observed. One subject was excluded from this comparison because of a skipped first REM episode during the night following sham stimulation. Analyzing the non-REM᎐REM sleep cycles separately revealed that in the first non-REM sleep episode, the occurrence of sleep stage 4 was slightly enhanced wreal: 18.2 Ž3.9 S.E.M.. min, sham: 13.8 Ž2.9. min; P- 0.05, N s 7, factor ‘treatment’ of two-way rANOVAx. The waking EEG Žreferential C3A2, F3A2 or
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
Fig. 1. REM sleep latencies Žh. for individual subjects Ž N s 7. in the nights following sham and real stimulation. One subject was excluded from this comparison because of a skipped first REM sleep episode during the night following sham stimulation.
bipolar. was analyzed by comparing power spectra after real and sham stimulation and by comparing
5
the ratios afterrbefore real or sham stimulation with each other ŽFig. 2.. Tests carried out for all 0.25-Hz bins revealed that the factors ‘treatment’ and ‘order’ and the interaction ‘treatment’= ‘order’ were significant in only a few bins. The factor ‘treatment’ was significant in two consecutive bins Ž10.375᎐10.875 Hz. for the derivation F3C3 ŽFig. 2. and in three consecutive bins Ž10.625᎐11.375 Hz. for the derivation C3A2. However, considering that 100 rANOVAs were carried out for each spectrum, these deviations with P-values ) 0.01 do not amount to a significant effect due to rTMS in the waking EEG. Additional tests based on traditional frequency bands yielded no significant effects for the factor ‘treatment’. The factor ‘order’ was significant in three of 40 comparisons and the interaction ‘treatment’= ‘order’ was significant in one case. The sleep EEG was analyzed by comparing power spectra for nights following sham and real stimulation. Whole night spectra, as well as spectra for the first non-REM sleep episode ŽFig. 3.,
Fig. 2. Mean EEG power density spectrum ŽF3C3, N s 8. during waking. The curves depict power in the 10-min recording after rTMS or sham stimulation Žmean " 1 S.E.M. for 0.25-Hz bins. expressed relative to the corresponding value of the recording before the stimulation. Bottom bars indicate frequency bins for which power was significantly different after rTMS than after sham stimulation Žfactor ‘treatment’ of two-way rANOVAs.. Note that P-values are plotted in the reverse direction. Factor ‘order’ was significant only in one frequency bin and a significant interaction ‘treatment’= ‘order’ was observed in two single bins.
6
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
were investigated by testing all 0.25-Hz bins and the EEG frequency bands ␦, , ␣ , , . Again, the factors ‘treatment’ and ‘order’ and the interaction ‘treatment’= ‘order’ were significant in only a few single 0.25-Hz bins. The same tests, that were applied to the broader frequency bands, revealed a significant increase Ž P- 0.05, factor ‘condition’. of ␦ power in the derivation F4C4 in the first non-REM sleep episode. Regional differences of the waking and sleep EEG were further analyzed by comparing spectral power in the frequency bands ␦, , ␣ , and  of bipolar derivations with three- and four-way rANOVAs. No significant differences were found for the sleep EEG. For the waking EEG, the factor ‘sequence’ Žbefore and after real or sham stimulation. was significant in all bands, except the ␣ band. This is the result of lower EEG power in the waking recordings after real or sham stimulation ŽFig. 2.. In addition, the factor ‘derivation’ and the interaction ‘condition’= ‘sequence’= ‘derivation’ were significant in the ␦
band ŽFig. 4.. A number of post-hoc tests were carried out in order to understand the nature of this interaction: 1. Three-way rANOVAs with the factors ‘derivation’, ‘hemisphere’ and ‘condition’ were carried out with the data of the waking recordings after real or sham stimulation and with the relative changes afterrbefore real or sham stimulation. 2. Three-way rANOVAs with the factors ‘condition’, ‘sequence’, and ‘derivation’ or ‘hemisphere’ were carried out for pairs of bipolar derivations Ž F3C3 ᎐ F4C4, F3C3 ᎐ C3P3, F4C4᎐C4P4, C3P3᎐C4P4.. 3. Two-way rANOVAs with the factors ‘condition’ and ‘sequence’ were carried out for each of the four derivations. The factor ‘condition’ or an interaction involving this factor was not significant in any of these rANOVAs.
Fig. 3. Mean EEG power density spectrum ŽC3A2, N s 87. in non-REM sleep Žstages 2, 3 and 4. of the first non-REM sleep episode. The curve depicts power in the night following real stimulation Žmean " 1 S.E.M. for 0.25-Hz bins. expressed relative to the corresponding value of the night following sham stimulation. Bottom bars indicate frequency bins for which power was significantly different after rTMS than after sham stimulation Žfactor ‘treatment’ of two-way rANOVAs.. Note that P-values are plotted in the reverse direction. A significant interaction ‘treatment’= ‘order’ was observed in three single bins.
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
7
Fig. 4. Relative ␦ power Žmean " 1 S.E.M.. of bipolar derivations in the wake recordings before and after real and sham stimulation. For each derivation, power was normalized to the average over all four recordings. A four-way rANOVA revealed that the factors ‘sequence’ and ‘derivation’ and the interaction ‘condition’=‘sequence’= ‘derivation’ were significant.
4. Discussion We investigated whether local activation of a specific brain region by rTMS with parameters used in depression treatment trials during wakefulness affects the EEG recorded from this region during waking and subsequent sleep. Unilateral rTMS of the left dorsolateral cortex applied to left prefrontal areas did not alter the waking and sleep EEG and caused only a minor reduction of sleep stage 1 and a slightly enhanced occurrence of sleep stage 4 in the first non-REM sleep episode. Contrary to the findings of Cohrs et al. Ž1998., no significant increase of REM sleep latency ŽRL; Fig. 1., RL q SL, nor the nonREM᎐REM sleep cycle length was observed. In the alpha range of the waking EEG, the factor ‘treatment’ of the two-way rANOVAs was significant for two and three consecutive bins for the derivations F3C3 ŽFig. 2. and C3A2, respectively. Because the P-values are ) 0.01, the effect is not statistically significant if corrections for multiple testing are considered. Nevertheless, alpha activity was somewhat enhanced in the waking recordings after real and sham stimulations when compared with the recordings before the stimulations ŽFig. 2.. This observation may indi-
cate that subjects became more relaxed over the course of an experimental session, in particular after sham stimulation. Power in the ␦, , and bands was lower in the waking recordings after real or sham stimulation ŽFigs. 2 and 4, factor ‘sequence’ in four-way rANOVAs.. In addition, the interaction ‘condition’= ‘sequence’= ‘derivation’ was significant in the ␦ band, but none of the post-hoc tests showed any significant effects involving the factor ‘condition’. Small variations in the non-specific effect of lower power in the wake recordings after real or sham stimulation may, therefore, be responsible for the observed interaction. During the first non-REM sleep episode, ␦ power in the derivation F4C4 was significantly enhanced Ž P- 0.05.. Considering corrections due to multiple testing, it is questionable whether this represents a real effect. However, the increased ␦ power may be related to the enhanced occurrence of sleep stage 4 in the first non-REM sleep episode. In summary, the observed effects on both the waking and sleep EEG and its topography are small, and may, in some cases, be chance effects arising from the large number of tests or may be related to non-specific effects Že.g. lower power in
8
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9
the waking recording after real or sham exposure.. Different factors may be responsible for the absence of clear EEG effects due to rTMS: 1. Stimulation was too focal to result in detectable EEG changes; 2. Stimulus strength Žduration andror intensity . was insufficient to initiate long-term EEG changes. Indeed, a vibratory stimulus of 6-h duration applied to the dominant hand prior to sleep was needed to induce increased slow-wave activity over the somatosensory cortex during the first hour of sleep ŽKattler et al., 1994.. On the other hand, 15᎐30 min exposure to EMF Žcorresponding to the radiation emitted by mobile phones. resulted in increased power in the sigma range ŽBorbely ´ et al., 1999; Huber et al., 2000.. Recently, we demonstrated ŽMosimann et al., 2000. that rTMS at 100% of the motor threshold may not have a modulatory effect on mood as was reported in previous studies ŽPascual-Leone et al., 1996; Wassermann, 1998.. Effects of rTMS may possibly only be detected at stimulation intensities much higher than those used previously. 3. Placement of the electrodes did not allow us to detect changes induced by rTMS. 4. The sample size might have been too small to detect significant effects. Our results demonstrate that rTMS applied in a therapeutic dose in young healthy males had no detectable effects on sleep and the sleep EEG. This shows that the putative anti-depressant effect of the method might not be mediated by strong influences on sleep patterns, and that rTMS is a safe and non-invasive method with respect to its influence on sleep. Furthermore, it is interesting to observe that while acute rTMS stimulation induces changes in regional cerebral blood flow ŽGeorge et al., 1999., we were unable to detect EEG changes during the subsequent night. Novel therapeutic agents must undergo extensive examination of safety in animals and healthy volunteers before testing their efficacy in clinical
trials. This scenario has been, to some extent, reversed with respect to rTMS applications in psychiatric conditions. This might be partly due to the great enthusiasm created by positive clinical trials in major depression and mania ŽGeorge et al., 1995, 1997; Klein et al., 1999; Menkes et al., 1999; Pascual-Leone et al., 1996.. Despite the fact that rTMS has shown some potential to positively influence psychiatric symptoms, the possibility for short- and long-term adverse effects on cognition and behavior has to be assessed rigorously. The message conveyed by this study is a mixed one: While we were not able to show that the anti-depressant effects of rTMS might be mediated by a clinically significant influence on sleep, we demonstrated that rTMS applied under therapeutic parameters is safe in regard to its influence on sleep in healthy young subjects.
Acknowledgements We thank Cornelia Huber, Viola Meier and Sandra Puliafito for help with the data collection and Dr Irene Tobler for comments on the manuscript. This study was supported by the Swiss National Science Foundation Grants Nr. 3100053005.97, 4038-044046.96 and 3231-044523.96 and the Human Frontiers Science Program grant RG-81r96. References Achermann, P., Borbely, ´ A.A., 1997. Low-frequency Ž- 1 Hz. oscillations in the human sleep EEG. Neuroscience 81, 213᎐222. Borbely, ´ A.A., Huber, R., Graf, T., Fuchs, B., Gallmann, E., Achermann, P., 1999. Pulsed high-frequency electromagnetic field affects human sleep and sleep electroencephalogram. Neuroscience Letters 275, 207᎐210. Cohrs, S., Tergau, F., Riech, S., Kastner, S., Paulus, W., Ziemann, U., Ruther, E., Hajak, G., 1998. High-frequency repetitive transcranial magnetic stimulation delays rapid eye movement sleep. Neuroreport 9, 3439᎐3443. Dearing, J., George, M.S., Greenberg, B.D., Wassermann, E.M., Schlaepfer, T.E., Murphy, D.L., Hallet, M., Post, R.M., 1997. Mood effects of prefrontal repetitive high frequency transcranial magnetic stimulation ŽrTMS. in healthy volunteers. CNS Spectrums 2, 53᎐68. Endo, T., Roth, C., Landolt, H.P., Werth, E., Aeschbach, D., Achermann, P., Borbely, ´ A.A., 1998. Selective REM sleep
T. Graf et al. r Psychiatry Research: Neuroimaging 107 (2001) 1᎐9 deprivation in humans: effects on sleep and sleep EEG. American Journal of Physiology 274, R1186᎐R1194. George, M.S., Lisanby, S.H., Sackeim, H.A., 1999. Transcranial magnetic stimulation: applications in neuropsychiatry. Archives of General Psychiatry 56, 300᎐311. George, M.S., Wassermann, E.M., Kimbrell, T.A., Little, J.T., Williams, W.E., Danielson, A.L., Greenberg, B.D., Hallett, M., Post, R.M., 1997. Mood improvement following daily left prefrontal repetitive transcranial magnetic stimulation in patients with depression: a placebo-controlled crossover trial. American Journal of Psychiatry 154, 1752᎐1756. George, M.S., Wassermann, E.M., Williams, W.A., Callahan, A., Ketter, T.A., Basser, P., Hallett, M., Post, R.M., 1995. Daily repetitive transcranial magnetic stimulation ŽrTMS. improves mood in depression. Neuroreport 6, 1853᎐1856. George, M.S., Wassermann, E.M., Williams, W.A., Steppel, J., Pascual-Leone, A., Basser, P., Hallett, M., Post, R.M., 1996. Changes in mood and hormone levels after rapid-rate transcranial magnetic stimulation ŽrTMS. of the prefrontal cortex. Journal of Neuropsychiatry and Clinical Neuroscience 8, 172᎐180. Huber, R., Graf, T., Cote, K.A., Wittmann, L., Gallmann, E., Matter, D., Schuderer, J., Kuster, N., Borbely, ´ A.A., Achermann, P., 2000. Exposure to pulsed high-frequency electromagnetic field during waking affects human sleep EEG. Neuroreport 11, 3321᎐3325. Kattler, H., Dijk, D.J., Borbely, ´ A.A., 1994. Effect of unilateral somatosensory stimulation prior to sleep on the sleep EEG in humans. Journal of Sleep Research 3, 159᎐164. Klein, E., Kreinin, I., Chistyakov, A., Koren, D., Mecz, L., Marmur, S., Ben-Shachar, D., Feinsod, M., 1999. Therapeutic efficacy of right prefrontal slow repetitive transcranial magnetic stimulation in major depression: a double-blind controlled study. Archives of General Psychiatry 56, 315᎐320. Menkes, D.L., Bodnar, P., Ballesteros, R.A., Swenson, M.R., 1999. Right frontal lobe slow frequency repetitive transcranial magnetic stimulation ŽSF r-TMS. is an effective treatment for depression: a case-control pilot study of safety and efficacy. Journal of Neurology, Neurosurgery and Psychiatry 67, 113᎐115.
9
Mosimann, U.P., Rihs, T.A., Engeler, J., Fisch, H.U., Schlaepfer, T.E., 2000. Mood effects of repetitive transcranial stimulation ŽrTMS. of left prefrontal cortex in healthy volunteers. Psychiatry Research 94, 251᎐256. Oldfield, R.C., 1971. The assessment and analysis of handedness ᎏ the Edinburgh inventory. Neuropsychologia 9, 97᎐113. Pascual-Leone, A., Catala, M.D., Pascual-Leone Pascual, A., 1996. Lateralized effect of rapid-rate transcranial magnetic stimulation of the prefrontal cortex on mood. Neurology 46, 499᎐502. Pascual-Leone, A., Tormos, J.M., Keenan, J., Tarazona, F., Canete, C., Catala, M.D., 1998. Study and modulation of human cortical excitability with transcranial magnetic stimulation. Journal of Clinical Neurophysiology 15, 333᎐343. Post, R.M., Frye, M.A., Denicoff, K.D., Leverich, G.S., Kimbrell, T.A., Dunn, R.T., 1998. Beyond lithium in the treatment of bipolar illness. Neuropsychopharmacology 19, 206᎐219. Post, R.M., Kimbrell, T.A., McCann, U.D., Dunn, R.T., Osuch, E.A., Speer, A.M., Weiss, S.R., 1999. Repetitive transcranial magnetic stimulation as a neuropsychiatric tool: present status and future potential. Journal of Electroconvulsive Therapy ŽJECT. 15, 39᎐59. Rechtschaffen, A., Kales, A., 1968. A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. National Institutes of Health, Bethesda, Maryland. Triggs, W.J., McCoy, K.J., Greer, R., Rossi, F., Bowers, D., Kortenkamp, S., Nadeau, S.E., Heilman, K.M., Goodman, W.K., 1999. Effects of left frontal transcranial magnetic stimulation on depressed mood, cognition, and corticomotor threshold. Biological Psychiatry 45, 1440᎐1446. Wassermann, E., 1998. Risk and safety of repetitive transcranial magnetic stimulation: report and suggested guidelines from the International Workshop on the Safety of Repetitive Transcranial Magnetic Stimulation, June 5᎐7, 1996. Electroencephalograpy and Clinical Neurophysiology 108, 1᎐16.