- Email: [email protected]
Efficient reduction of stimulus artefact in TMS–EEG by epithelial short-circuiting by mini-punctures
Clinical Neurophysiology 119 (2008) 475–481 www.elsevier.com/locate/clinph
Efficient reduction of stimulus artefact in TMS–EEG by epithelial short-circuiting by mini-punctures q P. Julkunen a
a,*
, A. Pa¨a¨kko¨nen a, T. Hukkanen a, M Ko¨no¨nen S. Vanhatalo c, J. Karhu a,d
a,b
, P. Tiihonen a,
Department of Clinical Neurophysiology, Kuopio University Hospital, P.O. Box 1777, FI-70211 Kuopio, Finland b Department of Radiology, Kuopio University Hospital, Kuopio, Finland c Department of Clinical Neurophysiology, University Hospital of Helsinki, Helsinki, Finland d Nexstim Ltd., Helsinki, Finland Accepted 23 September 2007
Abstract Objective: We aimed at comparing the effects of two different electrode-to-skin contact preparation techniques on the stimulus artefact induced by transcranial magnetic stimulation (TMS) in electroencephalography (EEG) signals. Methods: Six healthy subjects participated in a combined navigated brain stimulation (NBS) and EEG study. Electrode contacts were first prepared in the standard way of rubbing the skin using a wooden stick with a cotton tip. The location of hand motor area and the motor threshold (MT) was determined for each subject. Then, the TMS-induced artefact was measured at 60%, 80%, 100% and 120% of the MT. Subsequently, the epithelium under the electrode contacts was electrically short-circuited by puncturing with custom-made needles and the stimulation sequences were replicated. The artefact was compared between the preparation techniques. Results: The TMS-induced artefact was significantly reduced after puncturing. In addition, the size and duration of the artefact depended on the applied stimulation intensity. The reduction of the artefact was largest in electrodes at and close to the stimulation site. Conclusions: Mini-puncturing technique enables more accurate analysis of TMS-induced short-latency phenomena in EEG during NBS, and it may aid in the examination of the short distance neural connectivity beneath and close to the stimulation site. Significance: This study describes a practical skin preparation method that significantly improves the utility of TMS–EEG method in studying short-latency cortical connectivity. 2007 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Navigated brain stimulation; Transcranial magnetic stimulation; Electroencephalography; Brain; Skin contact
1. Introduction Stimulation of the brain by electrical currents has been used for therapy and diagnostics for decades. Nowadays, high currents conducted directly through skin are no longer necessary. Transcranial magnetic stimulation (TMS) can transfer the electrical current to the desired stimulation focus on the cortex (Barker et al., 1985). TMS has been q *
Disclosures: Jari Karhu is affiliated with Nexstim Ltd. Corresponding author. Tel.: +358 44 7174118; fax: +358 17 173187. E-mail address: petro.julkunen@kuh.fi (P. Julkunen).
used primarily for research purposes in studies of cortical excitability, neuronal connectivity and plasticity, and in the localization of functionally active areas on the cortex (Kahkonen et al., 2004; Kobayashi and Pascual-Leone, 2003; Siebner and Rothwell, 2003; Wright et al., 2006). In the past few years, navigated brain stimulation (NBS) has provided a new, more precise way of transferring the magnetic stimulation to a certain location of the brain by combining magnetic resonance (MR) imaging with advanced image processing methods and navigating systems (Boroojerdi et al., 1999; Ettinger et al., 1998; Gugino et al., 2001; Hannula et al., 2005; Sparing et al., 2007).
1388-2457/$32.00 2007 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2007.09.139
476
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
By combining the electroencephalographic (EEG) recordings of the brain activity with TMS, the reactivity and connectivity of intra- and transcortical networks of the brain can be examined (Ilmoniemi et al., 1997; Izumi et al., 1997; Komssi et al., 2002; Schurmann et al., 2001). However, the strong currents induced in the tissue by each TMS-pulse produce large artefacts in the scalp electrodes close to the TMS-coil (Fuggetta et al., 2006; Izumi et al., 1997; Morbidi et al., 2007; Thut et al., 2005). With TMScompatible EEG recording amplifiers, the recording of the EEG can be blocked for the duration of the TMS-pulse using gain control and sample-and-hold circuits (Virtanen et al., 1999). However, the decay of the stimulus-induced charge differences between the electrode-gel and skin occurs in a time-scale that is much longer than the stimulus itself. This time-scale depends on the capacitive and resistive properties of the skin (Swanson and Webster, 1974) (Fig. 1A). Reducing the skin capacitance and resistance by electrical short-circuiting of the epithelial layers should theoretically make the decay of the induced charge differences faster and reduce the TMS-induced artefact in the EEG-electrodes. The skin-borne TMS artefacts are large. Thus, studies have either neglected tens of milliseconds from the beginning of the response (Bonato et al., 2006; Fuggetta et al., 2006), or have applied different types of online and offline artefact removal paradigms (Fuggetta et al., 2006; Ilmoniemi et al., 1997; Morbidi et al., 2007; Thut et al., 2005). This study was set up to characterize the TMS-induced artefact temporally and spatially and to test if a significant reduction of TMS-induced artefact could be achieved by short-circuiting of the epithelial layers by mini-puncturing technique (Burbank and Webster, 1978; Miller et al., 2007; Tallgren, 2005), which has been used to achieve a direct current stable electrode skin interface in Full-band EEG recordings (FbEEG) (Vanhatalo et al., 2005). A reliable comparison between the traditionally used rubbing (abrasion) and the mini-puncture technique was made possible by our NBS system that allows stimulation with highly repeatable coil location and orientation. To the best of our knowledge, the minipuncturing technique has not been used earlier together with TMS–EEG. We assumed that short-circuiting (decreasing capacitance and resistance, Fig. 1A) of the skin would result in not just smaller amplitudes of the TMS-induced artefact but also faster recovery from it. This would enable the investigation of cortical short-latency (<25 ms) responses to TMS more accurately and reliably than what has been achieved after conventional skin preparation by rubbing, which typically results in severe EEG artefacts (Bonato et al., 2006; Paus et al., 2001). 2. Methods 2.1. Overview of measurement system Six voluntary healthy subjects (3 male and 3 female) aged 21–46 years participated in the study. Subjects were scanned
with a Siemens Magnetom Avanto (Erlangen, Germany) 1.5T scanner to receive T1-weighted high-resolution 3D MR-images for the NBS. The NBS setup consisted of a navigation system, a stimulator, and a 70 mm figure-of-eight biphasic TMS-coil (Nexstim Ltd., Helsinki, Finland). During NBS, muscle activity was monitored online with an electromyography (EMG) device (ME 6000, Mega Electronics Ltd., Kuopio, Finland), and EEG was recorded with a 60channel TMS-compatible EEG device (Nexstim Ltd., Helsinki, Finland). All EEG-electrodes were referenced to an electrode placed on the right mastoid. EMG was measured from pre-gelled disposable Ag/AgCl electrodes attached to the opponens pollicis muscle. External triggering pulses controlled by a foot-switch were used to synchronize the TMS, EMG and EEG systems. Triggering pulse was gated to the EEG-amplifier for blocking the EEG input channels for 3 ms in order to prevent saturation of the amplifier (Virtanen et al., 1999). 2.2. Characterization of the artefact without amplifier blocking The characterization of the artefact was performed using a TDS210 oscilloscope (Tektronix Inc., Beaverton, OR, USA) with high impedance (10 MX) probes by measuring voltage between a reference electrode and an active electrode close to the TMS-coil centre. Two coil types were used for comparison: figure-of-eight biphasic and monophasic (Nexstim Ltd., Helsinki, Finland), both 70 mm in diameter. A pre-gelled disposable Ag/AgCl electrode placed on the right forearm skin on the radialis longus muscle served as an active electrode. A similar reference electrode was placed on the right wrist. The stimuli were delivered straight at the active electrode with the coil plane parallel to the electrode and the distance was fixed to about 0.8 cm. The oscilloscope was triggered with the stimuli. Effect of stimulus intensity on the induced artefact was evaluated with biphasic coil by using 20%, 50% and 80% of the maximum stimulator output. 2.3. Electrode preparation and measurement protocol Subjects were first prepared for combined NBS and EEG recording by rubbing the scalp at the electrode–skin contacts using a wooden stick with a cotton tip (Fig. 1) and all electrode locations were digitized. Then, the primary motor cortex area was mapped to locate the optimal cortical representation area of opponens pollicis muscle by finding the highest EMG response. EMG was also used to determine the motor threshold (MT) intensities. A compound muscle action potential with amplitude greater than 50 lV was taken as a reliable response. MT intensity was determined for each subject by finding the lowest NBS stimulation intensity that induced at least 5 responses out of 10 consecutive stimuli. Fifty TMS-pulses were then delivered at the optimal location of the opponens pollicis muscle at three to four
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
Fig. 1. (A) Schematic presentation of electrode contacts with rubbing and mini-puncturing of the scalp. Electrical model for the skin was adapted from Swanson and Webster (1974). (B) A custom-made instrument for the mini-puncturing.
second intervals. The sequence was repeated with four different intensities: 60%, 80%, 100% and 120% of the determined MT. Both spontaneous and stimulus-locked EEGs were simultaneously measured. The order of the sequences was varied randomly between the subjects.
477
After the four sequences, each electrode contact was punctured through a hole in the electrode (four punctures per electrode). For puncturing the skin, a custom-made mini-puncturing instrument was used (Fig. 1B). This instrument was prepared by inserting a hypodermic needle (Microlance-3 24G 0.55 · 25 mm, BD Medical Systems, Drogheda, Ireland) concentrically through a cut 1.2 mm diameter hypodermic needle so that the sharp tip extended 0.5 mm over the end of the outer needle (blunt gel injection needle, Microlance-3 18G 1.2 · 40 mm, BD Medical Systems, Drogheda, Ireland). The needles were fixed to this position by using hot melt glue. The length of the sharp tip in the instruments and thus the puncturing depth varied between 0.4 and 0.8 mm. The preparation technique was similar to a technique used earlier in the measurements of FbEEG (or direct current – coupled EEG; Miller et al., 2007; Tallgren, 2005). After the mini-puncture preparation, the four stimulation sequences were repeated. In the case of one subject, only 10 of the electrode–skin contacts were punctured at the stimulation site on the left hemisphere, and the spread of the stimulation artefact was examined. 2.4. EEG recording and analyses The EEG was recorded with a 1450 Hz sampling frequency and 16-bit precision. A triggering signal marking
Fig. 2. The TMS-induced potential was measured from a disposable Ag–AgCl electrode placed on the arm of one of the authors. The potentials were measured using an oscilloscope. The induced potentials for a biphasic coil are presented at 80% (A) and 50% (B) of the maximum stimulator output. Similarly, induced potentials are presented for monophasic coil at 80% (C) and 40% (D) of the maximum stimulator output. Stimulation moment is indicated by a dashed red line. The peak-to-peak amplitudes of the induced skin potential are also presented. Black downward arrow represents the moment in time when the monophasic pulse has returned to baseline.
478
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
the exact stimulation moments was recorded together with the EEG. Vertical electro-oculogram was recorded from electrodes placed above and below the right eye. The recorded EEGs were processed by using a commercial software package Brain Vision Analyzer (Brain Products GmbH, Munich, Germany). The continuous EEG was segmented to 400 ms segments. The segments were averaged with a 100 ms prestimulus baseline. Matlab 7.3 (Mathworks Inc., Natick, MA, USA) was used for analysis of the processed signals. Segments with blinks affecting the induced artefact were removed manually from the analysis. Six electrodes were chosen for the global-field-power (GFP) comparison from those showing the highest TMSinduced artefact under the coil (Fig. 2) (Lehmann and Skrandies, 1980). The GFPs were statistically compared between the two preparation techniques by using non-parametric tests (SPSS 11.5, SPSS Inc., Chicago, IL, USA). To be able to compare the induced artefacts between the subjects, the artefact power was normalized with respect to the artefact power measured with the rubbing preparation. The localized differences were visualized with CURRY v. 4.0 (Compumedics, El Paso, TX, USA). The mini-puncture preparation reduced the signal artefact as compared to rubbing for a certain duration after stimuli. When estimating this duration of improved signal recovery between the preparation techniques, two assumptions were made: (1) pre-stimulus signal and (2) eventrelated brain potentials post-stimuli were not affected by the electrode-to-skin contact preparation. The difference between the preparation techniques was considered to be significant when the absolute difference was higher than two times the standard deviation (SD) of the pre-stimulus (from 100 ms to 0 ms) signal difference. 3. Results In the oscilloscope recordings without amplifier blocking, the biphasic as well as monophasic pulses induced an intensity-dependent response in the skin potential with characteristic response waveforms (Fig. 2). At 80% intensity, the biphasic pulse had an average amplitude of 27.0 V and the monophasic pulse 11.5 V. At 20% and 50% intensity, the biphasic pulses had an average peak-to-peak amplitude of 5.7 V and 18.0 V, respectively. The duration was 300 ls for the biphasic and 500–600 ls for the monophasic pulse. The amplitudes varied depending on the stimulus intensity. In the EEG recordings with amplifier blocking, the minipuncture preparation of the scalp was more effective than mere rubbing of the scalp in minimizing the TMS-induced EEG artefact (Fig. 3). The decay of the artefact was faster after mini-puncturing compared to rubbing – the difference in the size of the artefact between the preparation techniques remained larger than two times the pre-stimulus SD in average to 21 ms after the stimulus. This difference was dependent on the stimulation intensity (Fig. 4). The average pre-stimulus SD was 1.35 lV.
Fig. 3. Averaged EEG recording of a single subject is presented showing the location-dependence of the TMS-induced artefact. Six electrodes from area of the artefact were chosen for global-field-power analysis. The signal range was from 100 ms pre-stimulus to 100 ms post-stimulus. TMSinduced artefact is presented for both conditions (rubbed and minipunctured). A close-up from the FC3-electrode signal is presented with amplitudes and recovery times.
When examining the GFP of the electrodes most prone to the artefact, a significant (p < 0.05) decrease in the artefact power was observed after the mini-puncture procedure (Fig. 5). The artefact power increased significantly with the TMS intensity (p < 0.01). Altogether, the mini-puncture technique decreased the TMS-induced artefact power up to 93% as compared to the rubbing technique. When evaluating the stimulus artefact by normalizing the artefact power with respect to the artefact after rubbing, the minipuncture technique produced a significantly lower artefact. The significance of the improvement with mini-puncture technique was increased with the TMS intensity (Fig. 5). The spread of the TMS-evoked EEG artefact in a case of mixed electrode-to-skin contact preparation techniques was observed in one male subject. Mini-puncture preparation was performed to 10 electrodes on the left side on the scalp. The boundaries between different preparation tech-
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
Fig. 4. The difference in the recovery time between the preparation techniques. This difference was determined finding a post-stimulus timepoint in which the difference between the signals was lower than two times the pre-stimulus SD of the signal difference between the preparations. The results are presented for each stimulation intensity as a group average together with SDs.
Fig. 5. (A) The normalized power (with respect to power after rubbing) of the TMS-induced artefacts (at 6 ms) is presented separately for each subject as a mean of all intensities. Only 10 of the electrode–skin contacts were punctured for subject 6, and thus the artefact powers after rubbing and mini-puncture were computed from different sides. (B) Normalized artefact power (mean of subjects) after mini-puncture with respect to the artefact power after rubbing as a function of stimulation intensity. The effect of the preparation method on the artefact becomes more significant as the stimulation intensity increases.
niques are apparent in the scalp potential maps at the time of the artefact maximum (Fig. 6). The right side (prepared with soft rubbing) showed >10 times greater artefact in stimulation area GFP as compared to the left side (prepared with mini-punctures). The TMS-induced EEG artefact decreased in each subject after the mini-puncture preparation. The spatial spreading of the artefact was similar, but its strength was significantly reduced (Fig. 7).
479
Fig. 6. Spread of the TMS-induced EEG response at 6–10 ms post-stimuli in a test subject, who had 10 electrode–skin contacts on the left side on top of M1 mini-punctured and remaining electrode-to-skin contacts rubbed. The punctured preparation area is surrounded with dashed line. The TMS-pulses with 90% MT were focused ipsilaterally for both conditions (on the left side for the punctured and on the right side for the rubbed).
Fig. 7. Spread of the TMS-induced artefact after rubbing and minipuncturing are presented for one subject. The TMS-pulses with MT intensity were focused on the left hemisphere. The difference in the spread of induced artefacts of the two techniques is also presented. Potential maps were normalized symmetrically to the maximum potential after rubbing. Below, stimulus-induced electric field densities computed by the stimulation software are presented at depth of 1 mm from the scalp (left) and on the cortex (right).
4. Discussion Studies on cortico-cortical connectivity have been hampered by the drastic stimulation artefacts, which are likely caused by the capacitive properties of the electrode-gel– skin circuits (see Fig. 1). In this study we characterized the temporo-spatial properties of this artefact, and we
480
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
describe an easy and practical method to significantly reduce the TMS artefact. The TMS-induced artefact during NBS is often mainly restricted to the stimulation area allowing accurate contralateral EEG examination during the stimulation. However, examining the whole spatial spread and short-latency properties of the TMS-evoked EEG response requires minimization of stimulus artefact. As found in this study, the spread of the artefact is clearly affected by the preparation of the electrode contact (Fig. 6). Our results indicate that the electric properties of skin epithelium have a profound effect on the TMS artefact. The conventional skin preparation consists of repeated rubbing of the outer epithelial layers, which is often used in EEG or ECG recordings to reduce the tissue impedance (Burbank and Webster, 1978; Tam and Webster, 1977). However, deeper layers of the skin epithelium, including much of the epithelium and the basal lamina, will remain intact. Their capacitive properties will lead to a charging of the skin after the TMS-pulse, which then gradually ‘‘discharges’’ and hence yields the long-lasting tail of stimulation artefact. Short-circuiting of this remaining layer by the mini-puncture technique described in the present study eliminates the remaining skin-borne component. It is notable here that mini-puncture alone (i.e. without prior rubbing) will yield about similar effect on skin electric properties (Miller et al., 2007; Tallgren, 2005). Hence our results from mini-puncture do not likely reflect a combined effect of rubbing plus mini-puncture, which in the present study was done just to be able to reliably compare the two preparation techniques. The impact of TMS-induced artefact is centred at the stimulation area below the TMS-coil. Therefore, by neglecting the signals from the electrodes with highest artefact or using signal processing techniques, the harmful effect can be reduced in the analysis. However, a technique that is able to minimize the induced artefact should be used when studying the areas under or close to the TMS-coil. Reducing the stimulation intensity reduced the stimulation artefact (Fig. 6). Recent studies have shown that TMS can evoke detectable EEG responses with very low intensities, which can be used to minimize the intensity-dependent power of the stimulus artefact (Komssi and Kahkonen, 2006; Komssi et al., 2007). The optimal length of the tip (i.e. penetration depth) was not separately studied here, but prior studies have suggested that a depth of <0.5 mm is sufficient to penetrate through the basal lamina (Burbank and Webster, 1978). However, since the skin thickness may vary, and occasional hairs may get under the instrument, previous findings suggest that for reliable results, a tip length of about 0.5 mm is likely more reliable. The short-circuiting procedure can be further ensured by performing repeated strokes (1–5 times; see Miller et al., 2007). It is notable here that an adequate mini-puncture usually does not cause any bleeding, because the skin blood vessels traverse mostly deeper in the subdermal tissue.
Any skin lesion, whether due to abrasion or mini-punctures, will expose the skin at risk of infection from exogenous contaminants. While systematic studies are lacking, our own several years’ experience (incl. Miller et al., 2007), as well as our personal communications with other laboratories, suggests that the risk from mini-puncture is negligible if compared to other skin infections in the hospital environment. As far as the risk of blood-borne contamination from the subject to the measuring devices (Ferree et al., 2001; Putnam et al., 1992) is concerned, it is theoretically possible, but the absence of bleeding and the use of electrode cap without direct skin contacts both efficiently minimize the risk. It is advisable, however, that EEG caps should go through a prudent disinfection or sterilization protocol, and that the mini-puncture tools are made from disposable components. The major practical advantages of the method we describe here are that (i) the mini-puncture instrument can be self-made in every laboratory from the very low cost standard clinical needles, (ii) the needles are disposable, hence there are no sterilization concerns, and (iii) the procedure is very easy to learn, and quick and repeatable to perform. Moreover, our experience from a larger set of FbEEG recordings has shown that the mini-puncture causes usually less discomfort than an adequate rubbing, and the actual skin damage (i.e. ‘‘invasiveness’’) from a tiny epithelial penetration is practically less than after a good rubbing (Miller et al., 2007). The charging of the scalp led to bipolar spread of the artefact potential. This spatial distribution of the artefact resembles the electric field computed by the navigation software below the scalp (Fig. 7). The location of the positive and negative artefact maxima depends on the coil location and orientation (Fig. 3). The investigation of the TMS-induced artefact as a skin potential measured from forearm with an oscilloscope indicated that the EEG-amplifier’s gain control and sampleand-hold circuits are good at guarding the TMS-related phenomena from the artefact. When comparing the amplitude of the artefact in forearm experiments with biphasic coil with the amplitude of the TMS-induced artefact in EEG from one subject’s recording with the same intensity, the reduction of the highest possible artefact was 99.99% independent from the electrode contact preparation. The remaining artefact may be reduced by electrode contact preparation techniques, such as used in this study, and with offline artefact removal (Morbidi et al., 2007). There are differences between the TMS-coils and their ability to induce artefacts. As we used a biphasic figureof-eight coil in our study, we cannot make direct conclusions of the behaviour of mini-puncture technique with monophasic coils and coils with different construction or design. When comparing monophasic and biphasic coils, the measured potential between the electrodes was significantly different (Fig. 2). As the monophasic coil induces mainly a single potential rise, the galvanic effect is supposed to be emphasized since the opposite discharging
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
phase (second, as in biphasic) is small or missing. Thus, we believe that mini-puncture technique could prove to be even more significant when using a monophasic coil together with EEG. In addition, EEG may be more prone to the intensity changes of monophasic pulse, because the duration of elevated skin potential seems to be dependent on the stimulation intensity (Fig. 2). Nevertheless, the induced potentials over the skin are normally lower when using monophasic coils due to a lower power of the induced magnetic field as compared to the biphasic coils. 5. Conclusions We conclude that TMS induces severe artefacts in the recorded EEG when applying high TMS intensities above MT. These artefacts are likely to originate from the skin. However, the artefacts are reduced when stimulation intensities below MT are used. In addition, TMS-induced artefact contaminated a significantly smaller area in the EEG when using mini-puncture preparation. Especially, the mini-puncture method is useful in the analysis of TMSevoked EEG responses with latencies shorter than 25 ms. Acknowledgement This study was supported by the Finnish Funding Agency for Technology and Innovation (TEKES) Helsinki, Finland. References Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic stimulation of human motor cortex. Lancet 1985;1:1106–7. Bonato C, Miniussi C, Rossini PM. Transcranial magnetic stimulation and cortical evoked potentials: a TMS/EEG co-registration study. Clin Neurophysiol 2006;117:1699–707. Boroojerdi B, Foltys H, Krings T, Spetzger U, Thron A, Topper R. Localization of the motor hand area using transcranial magnetic stimulation and functional magnetic resonance imaging. Clin Neurophysiol 1999;110:699–704. Burbank DP, Webster JG. Reducing skin potential motion artefact by skin abrasion. Med Biol Eng Comput 1978;16:31–8. Ettinger GJ, Leventon ME, Grimson WE, Kikinis R, Gugino L, Cote W, et al. Experimentation with a transcranial magnetic stimulation system for functional brain mapping. Med Image Anal 1998;2:133–42. Ferree TC, Luu P, Russell GS, Tucker DM. Scalp electrode impedance, infection risk, and EEG data quality. Clin Neurophysiol 2001;112:536–44. Fuggetta G, Pavone EF, Walsh V, Kiss M, Eimer M. Cortico-cortical interactions in spatial attention: A combined ERP/TMS study. J Neurophysiol 2006;95:3277–80. Gugino LD, Romero JR, Aglio L, Titone D, Ramirez M, Pascual-Leone A, et al. Transcranial magnetic stimulation coregistered with MRI: a comparison of a guided versus blind stimulation technique and its effect on evoked compound muscle action potentials. Clin Neurophysiol 2001;112:1781–92. Hannula H, Ylioja S, Pertovaara A, Korvenoja A, Ruohonen J, Ilmoniemi RJ, et al. Somatotopic blocking of sensation with navigated transcranial magnetic stimulation of the primary somatosensory cortex. Hum Brain Mapp 2005;26:100–9.
481
Ilmoniemi RJ, Virtanen J, Ruohonen J, Karhu J, Aronen HJ, Na¨a¨ta¨nen R, et al. Neuronal responses to magnetic stimulation reveal cortical reactivity and connectivity. Neuroreport 1997;10:3537–40. Izumi S, Takase M, Arita M, Masakado Y, Kimura A, Chino N. Transcranial magnetic stimulation-induced changes in EEG and responses recorded from the scalp of healthy humans. Electroencephalogr Clin Neurophysiol 1997;103:319–22. Kahkonen S, Wilenius J, Komssi S, Ilmoniemi RJ. Distinct differences in cortical reactivity of motor and prefrontal cortices to magnetic stimulation. Clin Neurophysiol 2004;115:583–8. Kobayashi M, Pascual-Leone A. Transcranial magnetic stimulation in neurology. Lancet Neurol 2003;2:145–56. Komssi S, Aronen HJ, Huttunen J, Kesaniemi M, Soinne L, Nikouline VV, et al. Ipsi- and contralateral EEG reactions to transcranial magnetic stimulation. Clin Neurophysiol 2002;113:175–84. Komssi S, Kahkonen S. The novelty value of the combined use of electroencephalography and transcranial magnetic stimulation for neuroscience research. Brain Res Rev 2006;52:183–92. Komssi S, Savolainen P, Heiskala J, Kahkonen S. Excitation threshold of the motor cortex estimated with transcranial magnetic stimulation electroencephalography. Neuroreport 2007;18:13–6. Lehmann D, Skrandies W. Reference-free identification of components of checkerboard-evoked multichannel potential fields. Electroencephalogr Clin Neurophysiol 1980;48:609–21. Miller JW, Kim W, Holmes MD, Vanhatalo S. Ictal localization by source analysis of infraslow activity in DC-coupled scalp EEG recordings. Neuroimage 2007;35:583–97. Morbidi F, Garulli A, Prattichizzo D, Rizzo C, Manganotti P, Rossi S. Off-line removal of TMS-induced artifacts on human electroencephalography by Kalman filter. J Neurosci Methods 2007;162:293–302. Paus T, Sipila PK, Strafella AP. Synchronization of neuronal activity in the human primary motor cortex by transcranial magnetic stimulation: an EEG study. J Neurophysiol 2001;86:1983–90. Putnam LE, Johnson Jr R, Roth WT. Guidelines for reducing the risk of disease transmission in the psychophysiology laboratory. SPR Ad Hoc Committee on the Prevention of Disease Transmission. Psychophysiology 1992;29:127–41. Schurmann M, Nikouline VV, Soljanlahti S, Ollikainen M, Basar E, Ilmoniemi RJ. EEG responses to combined somatosensory and transcranial magnetic stimulation. Clin Neurophysiol 2001;112:19–24. Siebner HR, Rothwell J. Transcranial magnetic stimulation: new insights into representational cortical plasticity. Exp Brain Res 2003;148:1–16. Sparing R, Buelte D, Meister IG, Paus T, Fink GR. Transcranial magnetic stimulation and the challenge of coil placement: A comparison of conventional and stereotaxic neuronavigational strategies. Hum Brain Mapp 2007. Swanson DK, Webster JG. A model for skin-electrode impedance. In: Miller HA, Harrison DC, editors. Biomedical electrode technology – Theory and practice. New York: Academic Press; 1974. p. 117–28. Tallgren P. DC-stable electrode–skin interface for human EEG recordings. Series E: Electronic Publications E 5 (http://www.hut.fi/Yksikot/ Elektroniikka/reports), Helsinki, 2005. Tam HW, Webster JG. Minimizing electrode motion artifact by skin abrasion. IEEE Trans Biomed Eng 1977;24:134–9. Thut G, Ives JR, Kampmann F, Pastor MA, Pascual-Leone A. A new device and protocol for combining TMS and online recordings of EEG and evoked potentials. J Neurosci Methods 2005;141:207–17. Vanhatalo S, Voipio J, Kaila K. Full-band EEG (FbEEG): an emerging standard in electroencephalography. Clin Neurophysiol 2005;116:1–8. Virtanen J, Ruohonen J, Naatanen R, Ilmoniemi RJ. Instrumentation for the measurement of electric brain responses to transcranial magnetic stimulation. Med Biol Eng Comput 1999;37:322–6. Wright MA, Orth M, Patsalos PN, Smith SJ, Richardson MP. Cortical excitability predicts seizures in acutely drug-reduced temporal lobe epilepsy patients. Neurology 2006;67:1646–51.
Efficient reduction of stimulus artefact in TMS–EEG by epithelial short-circuiting by mini-punctures q P. Julkunen a
a,*
, A. Pa¨a¨kko¨nen a, T. Hukkanen a, M Ko¨no¨nen S. Vanhatalo c, J. Karhu a,d
a,b
, P. Tiihonen a,
Department of Clinical Neurophysiology, Kuopio University Hospital, P.O. Box 1777, FI-70211 Kuopio, Finland b Department of Radiology, Kuopio University Hospital, Kuopio, Finland c Department of Clinical Neurophysiology, University Hospital of Helsinki, Helsinki, Finland d Nexstim Ltd., Helsinki, Finland Accepted 23 September 2007
Abstract Objective: We aimed at comparing the effects of two different electrode-to-skin contact preparation techniques on the stimulus artefact induced by transcranial magnetic stimulation (TMS) in electroencephalography (EEG) signals. Methods: Six healthy subjects participated in a combined navigated brain stimulation (NBS) and EEG study. Electrode contacts were first prepared in the standard way of rubbing the skin using a wooden stick with a cotton tip. The location of hand motor area and the motor threshold (MT) was determined for each subject. Then, the TMS-induced artefact was measured at 60%, 80%, 100% and 120% of the MT. Subsequently, the epithelium under the electrode contacts was electrically short-circuited by puncturing with custom-made needles and the stimulation sequences were replicated. The artefact was compared between the preparation techniques. Results: The TMS-induced artefact was significantly reduced after puncturing. In addition, the size and duration of the artefact depended on the applied stimulation intensity. The reduction of the artefact was largest in electrodes at and close to the stimulation site. Conclusions: Mini-puncturing technique enables more accurate analysis of TMS-induced short-latency phenomena in EEG during NBS, and it may aid in the examination of the short distance neural connectivity beneath and close to the stimulation site. Significance: This study describes a practical skin preparation method that significantly improves the utility of TMS–EEG method in studying short-latency cortical connectivity. 2007 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Navigated brain stimulation; Transcranial magnetic stimulation; Electroencephalography; Brain; Skin contact
1. Introduction Stimulation of the brain by electrical currents has been used for therapy and diagnostics for decades. Nowadays, high currents conducted directly through skin are no longer necessary. Transcranial magnetic stimulation (TMS) can transfer the electrical current to the desired stimulation focus on the cortex (Barker et al., 1985). TMS has been q *
Disclosures: Jari Karhu is affiliated with Nexstim Ltd. Corresponding author. Tel.: +358 44 7174118; fax: +358 17 173187. E-mail address: petro.julkunen@kuh.fi (P. Julkunen).
used primarily for research purposes in studies of cortical excitability, neuronal connectivity and plasticity, and in the localization of functionally active areas on the cortex (Kahkonen et al., 2004; Kobayashi and Pascual-Leone, 2003; Siebner and Rothwell, 2003; Wright et al., 2006). In the past few years, navigated brain stimulation (NBS) has provided a new, more precise way of transferring the magnetic stimulation to a certain location of the brain by combining magnetic resonance (MR) imaging with advanced image processing methods and navigating systems (Boroojerdi et al., 1999; Ettinger et al., 1998; Gugino et al., 2001; Hannula et al., 2005; Sparing et al., 2007).
1388-2457/$32.00 2007 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2007.09.139
476
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
By combining the electroencephalographic (EEG) recordings of the brain activity with TMS, the reactivity and connectivity of intra- and transcortical networks of the brain can be examined (Ilmoniemi et al., 1997; Izumi et al., 1997; Komssi et al., 2002; Schurmann et al., 2001). However, the strong currents induced in the tissue by each TMS-pulse produce large artefacts in the scalp electrodes close to the TMS-coil (Fuggetta et al., 2006; Izumi et al., 1997; Morbidi et al., 2007; Thut et al., 2005). With TMScompatible EEG recording amplifiers, the recording of the EEG can be blocked for the duration of the TMS-pulse using gain control and sample-and-hold circuits (Virtanen et al., 1999). However, the decay of the stimulus-induced charge differences between the electrode-gel and skin occurs in a time-scale that is much longer than the stimulus itself. This time-scale depends on the capacitive and resistive properties of the skin (Swanson and Webster, 1974) (Fig. 1A). Reducing the skin capacitance and resistance by electrical short-circuiting of the epithelial layers should theoretically make the decay of the induced charge differences faster and reduce the TMS-induced artefact in the EEG-electrodes. The skin-borne TMS artefacts are large. Thus, studies have either neglected tens of milliseconds from the beginning of the response (Bonato et al., 2006; Fuggetta et al., 2006), or have applied different types of online and offline artefact removal paradigms (Fuggetta et al., 2006; Ilmoniemi et al., 1997; Morbidi et al., 2007; Thut et al., 2005). This study was set up to characterize the TMS-induced artefact temporally and spatially and to test if a significant reduction of TMS-induced artefact could be achieved by short-circuiting of the epithelial layers by mini-puncturing technique (Burbank and Webster, 1978; Miller et al., 2007; Tallgren, 2005), which has been used to achieve a direct current stable electrode skin interface in Full-band EEG recordings (FbEEG) (Vanhatalo et al., 2005). A reliable comparison between the traditionally used rubbing (abrasion) and the mini-puncture technique was made possible by our NBS system that allows stimulation with highly repeatable coil location and orientation. To the best of our knowledge, the minipuncturing technique has not been used earlier together with TMS–EEG. We assumed that short-circuiting (decreasing capacitance and resistance, Fig. 1A) of the skin would result in not just smaller amplitudes of the TMS-induced artefact but also faster recovery from it. This would enable the investigation of cortical short-latency (<25 ms) responses to TMS more accurately and reliably than what has been achieved after conventional skin preparation by rubbing, which typically results in severe EEG artefacts (Bonato et al., 2006; Paus et al., 2001). 2. Methods 2.1. Overview of measurement system Six voluntary healthy subjects (3 male and 3 female) aged 21–46 years participated in the study. Subjects were scanned
with a Siemens Magnetom Avanto (Erlangen, Germany) 1.5T scanner to receive T1-weighted high-resolution 3D MR-images for the NBS. The NBS setup consisted of a navigation system, a stimulator, and a 70 mm figure-of-eight biphasic TMS-coil (Nexstim Ltd., Helsinki, Finland). During NBS, muscle activity was monitored online with an electromyography (EMG) device (ME 6000, Mega Electronics Ltd., Kuopio, Finland), and EEG was recorded with a 60channel TMS-compatible EEG device (Nexstim Ltd., Helsinki, Finland). All EEG-electrodes were referenced to an electrode placed on the right mastoid. EMG was measured from pre-gelled disposable Ag/AgCl electrodes attached to the opponens pollicis muscle. External triggering pulses controlled by a foot-switch were used to synchronize the TMS, EMG and EEG systems. Triggering pulse was gated to the EEG-amplifier for blocking the EEG input channels for 3 ms in order to prevent saturation of the amplifier (Virtanen et al., 1999). 2.2. Characterization of the artefact without amplifier blocking The characterization of the artefact was performed using a TDS210 oscilloscope (Tektronix Inc., Beaverton, OR, USA) with high impedance (10 MX) probes by measuring voltage between a reference electrode and an active electrode close to the TMS-coil centre. Two coil types were used for comparison: figure-of-eight biphasic and monophasic (Nexstim Ltd., Helsinki, Finland), both 70 mm in diameter. A pre-gelled disposable Ag/AgCl electrode placed on the right forearm skin on the radialis longus muscle served as an active electrode. A similar reference electrode was placed on the right wrist. The stimuli were delivered straight at the active electrode with the coil plane parallel to the electrode and the distance was fixed to about 0.8 cm. The oscilloscope was triggered with the stimuli. Effect of stimulus intensity on the induced artefact was evaluated with biphasic coil by using 20%, 50% and 80% of the maximum stimulator output. 2.3. Electrode preparation and measurement protocol Subjects were first prepared for combined NBS and EEG recording by rubbing the scalp at the electrode–skin contacts using a wooden stick with a cotton tip (Fig. 1) and all electrode locations were digitized. Then, the primary motor cortex area was mapped to locate the optimal cortical representation area of opponens pollicis muscle by finding the highest EMG response. EMG was also used to determine the motor threshold (MT) intensities. A compound muscle action potential with amplitude greater than 50 lV was taken as a reliable response. MT intensity was determined for each subject by finding the lowest NBS stimulation intensity that induced at least 5 responses out of 10 consecutive stimuli. Fifty TMS-pulses were then delivered at the optimal location of the opponens pollicis muscle at three to four
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
Fig. 1. (A) Schematic presentation of electrode contacts with rubbing and mini-puncturing of the scalp. Electrical model for the skin was adapted from Swanson and Webster (1974). (B) A custom-made instrument for the mini-puncturing.
second intervals. The sequence was repeated with four different intensities: 60%, 80%, 100% and 120% of the determined MT. Both spontaneous and stimulus-locked EEGs were simultaneously measured. The order of the sequences was varied randomly between the subjects.
477
After the four sequences, each electrode contact was punctured through a hole in the electrode (four punctures per electrode). For puncturing the skin, a custom-made mini-puncturing instrument was used (Fig. 1B). This instrument was prepared by inserting a hypodermic needle (Microlance-3 24G 0.55 · 25 mm, BD Medical Systems, Drogheda, Ireland) concentrically through a cut 1.2 mm diameter hypodermic needle so that the sharp tip extended 0.5 mm over the end of the outer needle (blunt gel injection needle, Microlance-3 18G 1.2 · 40 mm, BD Medical Systems, Drogheda, Ireland). The needles were fixed to this position by using hot melt glue. The length of the sharp tip in the instruments and thus the puncturing depth varied between 0.4 and 0.8 mm. The preparation technique was similar to a technique used earlier in the measurements of FbEEG (or direct current – coupled EEG; Miller et al., 2007; Tallgren, 2005). After the mini-puncture preparation, the four stimulation sequences were repeated. In the case of one subject, only 10 of the electrode–skin contacts were punctured at the stimulation site on the left hemisphere, and the spread of the stimulation artefact was examined. 2.4. EEG recording and analyses The EEG was recorded with a 1450 Hz sampling frequency and 16-bit precision. A triggering signal marking
Fig. 2. The TMS-induced potential was measured from a disposable Ag–AgCl electrode placed on the arm of one of the authors. The potentials were measured using an oscilloscope. The induced potentials for a biphasic coil are presented at 80% (A) and 50% (B) of the maximum stimulator output. Similarly, induced potentials are presented for monophasic coil at 80% (C) and 40% (D) of the maximum stimulator output. Stimulation moment is indicated by a dashed red line. The peak-to-peak amplitudes of the induced skin potential are also presented. Black downward arrow represents the moment in time when the monophasic pulse has returned to baseline.
478
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
the exact stimulation moments was recorded together with the EEG. Vertical electro-oculogram was recorded from electrodes placed above and below the right eye. The recorded EEGs were processed by using a commercial software package Brain Vision Analyzer (Brain Products GmbH, Munich, Germany). The continuous EEG was segmented to 400 ms segments. The segments were averaged with a 100 ms prestimulus baseline. Matlab 7.3 (Mathworks Inc., Natick, MA, USA) was used for analysis of the processed signals. Segments with blinks affecting the induced artefact were removed manually from the analysis. Six electrodes were chosen for the global-field-power (GFP) comparison from those showing the highest TMSinduced artefact under the coil (Fig. 2) (Lehmann and Skrandies, 1980). The GFPs were statistically compared between the two preparation techniques by using non-parametric tests (SPSS 11.5, SPSS Inc., Chicago, IL, USA). To be able to compare the induced artefacts between the subjects, the artefact power was normalized with respect to the artefact power measured with the rubbing preparation. The localized differences were visualized with CURRY v. 4.0 (Compumedics, El Paso, TX, USA). The mini-puncture preparation reduced the signal artefact as compared to rubbing for a certain duration after stimuli. When estimating this duration of improved signal recovery between the preparation techniques, two assumptions were made: (1) pre-stimulus signal and (2) eventrelated brain potentials post-stimuli were not affected by the electrode-to-skin contact preparation. The difference between the preparation techniques was considered to be significant when the absolute difference was higher than two times the standard deviation (SD) of the pre-stimulus (from 100 ms to 0 ms) signal difference. 3. Results In the oscilloscope recordings without amplifier blocking, the biphasic as well as monophasic pulses induced an intensity-dependent response in the skin potential with characteristic response waveforms (Fig. 2). At 80% intensity, the biphasic pulse had an average amplitude of 27.0 V and the monophasic pulse 11.5 V. At 20% and 50% intensity, the biphasic pulses had an average peak-to-peak amplitude of 5.7 V and 18.0 V, respectively. The duration was 300 ls for the biphasic and 500–600 ls for the monophasic pulse. The amplitudes varied depending on the stimulus intensity. In the EEG recordings with amplifier blocking, the minipuncture preparation of the scalp was more effective than mere rubbing of the scalp in minimizing the TMS-induced EEG artefact (Fig. 3). The decay of the artefact was faster after mini-puncturing compared to rubbing – the difference in the size of the artefact between the preparation techniques remained larger than two times the pre-stimulus SD in average to 21 ms after the stimulus. This difference was dependent on the stimulation intensity (Fig. 4). The average pre-stimulus SD was 1.35 lV.
Fig. 3. Averaged EEG recording of a single subject is presented showing the location-dependence of the TMS-induced artefact. Six electrodes from area of the artefact were chosen for global-field-power analysis. The signal range was from 100 ms pre-stimulus to 100 ms post-stimulus. TMSinduced artefact is presented for both conditions (rubbed and minipunctured). A close-up from the FC3-electrode signal is presented with amplitudes and recovery times.
When examining the GFP of the electrodes most prone to the artefact, a significant (p < 0.05) decrease in the artefact power was observed after the mini-puncture procedure (Fig. 5). The artefact power increased significantly with the TMS intensity (p < 0.01). Altogether, the mini-puncture technique decreased the TMS-induced artefact power up to 93% as compared to the rubbing technique. When evaluating the stimulus artefact by normalizing the artefact power with respect to the artefact after rubbing, the minipuncture technique produced a significantly lower artefact. The significance of the improvement with mini-puncture technique was increased with the TMS intensity (Fig. 5). The spread of the TMS-evoked EEG artefact in a case of mixed electrode-to-skin contact preparation techniques was observed in one male subject. Mini-puncture preparation was performed to 10 electrodes on the left side on the scalp. The boundaries between different preparation tech-
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
Fig. 4. The difference in the recovery time between the preparation techniques. This difference was determined finding a post-stimulus timepoint in which the difference between the signals was lower than two times the pre-stimulus SD of the signal difference between the preparations. The results are presented for each stimulation intensity as a group average together with SDs.
Fig. 5. (A) The normalized power (with respect to power after rubbing) of the TMS-induced artefacts (at 6 ms) is presented separately for each subject as a mean of all intensities. Only 10 of the electrode–skin contacts were punctured for subject 6, and thus the artefact powers after rubbing and mini-puncture were computed from different sides. (B) Normalized artefact power (mean of subjects) after mini-puncture with respect to the artefact power after rubbing as a function of stimulation intensity. The effect of the preparation method on the artefact becomes more significant as the stimulation intensity increases.
niques are apparent in the scalp potential maps at the time of the artefact maximum (Fig. 6). The right side (prepared with soft rubbing) showed >10 times greater artefact in stimulation area GFP as compared to the left side (prepared with mini-punctures). The TMS-induced EEG artefact decreased in each subject after the mini-puncture preparation. The spatial spreading of the artefact was similar, but its strength was significantly reduced (Fig. 7).
479
Fig. 6. Spread of the TMS-induced EEG response at 6–10 ms post-stimuli in a test subject, who had 10 electrode–skin contacts on the left side on top of M1 mini-punctured and remaining electrode-to-skin contacts rubbed. The punctured preparation area is surrounded with dashed line. The TMS-pulses with 90% MT were focused ipsilaterally for both conditions (on the left side for the punctured and on the right side for the rubbed).
Fig. 7. Spread of the TMS-induced artefact after rubbing and minipuncturing are presented for one subject. The TMS-pulses with MT intensity were focused on the left hemisphere. The difference in the spread of induced artefacts of the two techniques is also presented. Potential maps were normalized symmetrically to the maximum potential after rubbing. Below, stimulus-induced electric field densities computed by the stimulation software are presented at depth of 1 mm from the scalp (left) and on the cortex (right).
4. Discussion Studies on cortico-cortical connectivity have been hampered by the drastic stimulation artefacts, which are likely caused by the capacitive properties of the electrode-gel– skin circuits (see Fig. 1). In this study we characterized the temporo-spatial properties of this artefact, and we
480
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
describe an easy and practical method to significantly reduce the TMS artefact. The TMS-induced artefact during NBS is often mainly restricted to the stimulation area allowing accurate contralateral EEG examination during the stimulation. However, examining the whole spatial spread and short-latency properties of the TMS-evoked EEG response requires minimization of stimulus artefact. As found in this study, the spread of the artefact is clearly affected by the preparation of the electrode contact (Fig. 6). Our results indicate that the electric properties of skin epithelium have a profound effect on the TMS artefact. The conventional skin preparation consists of repeated rubbing of the outer epithelial layers, which is often used in EEG or ECG recordings to reduce the tissue impedance (Burbank and Webster, 1978; Tam and Webster, 1977). However, deeper layers of the skin epithelium, including much of the epithelium and the basal lamina, will remain intact. Their capacitive properties will lead to a charging of the skin after the TMS-pulse, which then gradually ‘‘discharges’’ and hence yields the long-lasting tail of stimulation artefact. Short-circuiting of this remaining layer by the mini-puncture technique described in the present study eliminates the remaining skin-borne component. It is notable here that mini-puncture alone (i.e. without prior rubbing) will yield about similar effect on skin electric properties (Miller et al., 2007; Tallgren, 2005). Hence our results from mini-puncture do not likely reflect a combined effect of rubbing plus mini-puncture, which in the present study was done just to be able to reliably compare the two preparation techniques. The impact of TMS-induced artefact is centred at the stimulation area below the TMS-coil. Therefore, by neglecting the signals from the electrodes with highest artefact or using signal processing techniques, the harmful effect can be reduced in the analysis. However, a technique that is able to minimize the induced artefact should be used when studying the areas under or close to the TMS-coil. Reducing the stimulation intensity reduced the stimulation artefact (Fig. 6). Recent studies have shown that TMS can evoke detectable EEG responses with very low intensities, which can be used to minimize the intensity-dependent power of the stimulus artefact (Komssi and Kahkonen, 2006; Komssi et al., 2007). The optimal length of the tip (i.e. penetration depth) was not separately studied here, but prior studies have suggested that a depth of <0.5 mm is sufficient to penetrate through the basal lamina (Burbank and Webster, 1978). However, since the skin thickness may vary, and occasional hairs may get under the instrument, previous findings suggest that for reliable results, a tip length of about 0.5 mm is likely more reliable. The short-circuiting procedure can be further ensured by performing repeated strokes (1–5 times; see Miller et al., 2007). It is notable here that an adequate mini-puncture usually does not cause any bleeding, because the skin blood vessels traverse mostly deeper in the subdermal tissue.
Any skin lesion, whether due to abrasion or mini-punctures, will expose the skin at risk of infection from exogenous contaminants. While systematic studies are lacking, our own several years’ experience (incl. Miller et al., 2007), as well as our personal communications with other laboratories, suggests that the risk from mini-puncture is negligible if compared to other skin infections in the hospital environment. As far as the risk of blood-borne contamination from the subject to the measuring devices (Ferree et al., 2001; Putnam et al., 1992) is concerned, it is theoretically possible, but the absence of bleeding and the use of electrode cap without direct skin contacts both efficiently minimize the risk. It is advisable, however, that EEG caps should go through a prudent disinfection or sterilization protocol, and that the mini-puncture tools are made from disposable components. The major practical advantages of the method we describe here are that (i) the mini-puncture instrument can be self-made in every laboratory from the very low cost standard clinical needles, (ii) the needles are disposable, hence there are no sterilization concerns, and (iii) the procedure is very easy to learn, and quick and repeatable to perform. Moreover, our experience from a larger set of FbEEG recordings has shown that the mini-puncture causes usually less discomfort than an adequate rubbing, and the actual skin damage (i.e. ‘‘invasiveness’’) from a tiny epithelial penetration is practically less than after a good rubbing (Miller et al., 2007). The charging of the scalp led to bipolar spread of the artefact potential. This spatial distribution of the artefact resembles the electric field computed by the navigation software below the scalp (Fig. 7). The location of the positive and negative artefact maxima depends on the coil location and orientation (Fig. 3). The investigation of the TMS-induced artefact as a skin potential measured from forearm with an oscilloscope indicated that the EEG-amplifier’s gain control and sampleand-hold circuits are good at guarding the TMS-related phenomena from the artefact. When comparing the amplitude of the artefact in forearm experiments with biphasic coil with the amplitude of the TMS-induced artefact in EEG from one subject’s recording with the same intensity, the reduction of the highest possible artefact was 99.99% independent from the electrode contact preparation. The remaining artefact may be reduced by electrode contact preparation techniques, such as used in this study, and with offline artefact removal (Morbidi et al., 2007). There are differences between the TMS-coils and their ability to induce artefacts. As we used a biphasic figureof-eight coil in our study, we cannot make direct conclusions of the behaviour of mini-puncture technique with monophasic coils and coils with different construction or design. When comparing monophasic and biphasic coils, the measured potential between the electrodes was significantly different (Fig. 2). As the monophasic coil induces mainly a single potential rise, the galvanic effect is supposed to be emphasized since the opposite discharging
P. Julkunen et al. / Clinical Neurophysiology 119 (2008) 475–481
phase (second, as in biphasic) is small or missing. Thus, we believe that mini-puncture technique could prove to be even more significant when using a monophasic coil together with EEG. In addition, EEG may be more prone to the intensity changes of monophasic pulse, because the duration of elevated skin potential seems to be dependent on the stimulation intensity (Fig. 2). Nevertheless, the induced potentials over the skin are normally lower when using monophasic coils due to a lower power of the induced magnetic field as compared to the biphasic coils. 5. Conclusions We conclude that TMS induces severe artefacts in the recorded EEG when applying high TMS intensities above MT. These artefacts are likely to originate from the skin. However, the artefacts are reduced when stimulation intensities below MT are used. In addition, TMS-induced artefact contaminated a significantly smaller area in the EEG when using mini-puncture preparation. Especially, the mini-puncture method is useful in the analysis of TMSevoked EEG responses with latencies shorter than 25 ms. Acknowledgement This study was supported by the Finnish Funding Agency for Technology and Innovation (TEKES) Helsinki, Finland. References Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic stimulation of human motor cortex. Lancet 1985;1:1106–7. Bonato C, Miniussi C, Rossini PM. Transcranial magnetic stimulation and cortical evoked potentials: a TMS/EEG co-registration study. Clin Neurophysiol 2006;117:1699–707. Boroojerdi B, Foltys H, Krings T, Spetzger U, Thron A, Topper R. Localization of the motor hand area using transcranial magnetic stimulation and functional magnetic resonance imaging. Clin Neurophysiol 1999;110:699–704. Burbank DP, Webster JG. Reducing skin potential motion artefact by skin abrasion. Med Biol Eng Comput 1978;16:31–8. Ettinger GJ, Leventon ME, Grimson WE, Kikinis R, Gugino L, Cote W, et al. Experimentation with a transcranial magnetic stimulation system for functional brain mapping. Med Image Anal 1998;2:133–42. Ferree TC, Luu P, Russell GS, Tucker DM. Scalp electrode impedance, infection risk, and EEG data quality. Clin Neurophysiol 2001;112:536–44. Fuggetta G, Pavone EF, Walsh V, Kiss M, Eimer M. Cortico-cortical interactions in spatial attention: A combined ERP/TMS study. J Neurophysiol 2006;95:3277–80. Gugino LD, Romero JR, Aglio L, Titone D, Ramirez M, Pascual-Leone A, et al. Transcranial magnetic stimulation coregistered with MRI: a comparison of a guided versus blind stimulation technique and its effect on evoked compound muscle action potentials. Clin Neurophysiol 2001;112:1781–92. Hannula H, Ylioja S, Pertovaara A, Korvenoja A, Ruohonen J, Ilmoniemi RJ, et al. Somatotopic blocking of sensation with navigated transcranial magnetic stimulation of the primary somatosensory cortex. Hum Brain Mapp 2005;26:100–9.
481
Ilmoniemi RJ, Virtanen J, Ruohonen J, Karhu J, Aronen HJ, Na¨a¨ta¨nen R, et al. Neuronal responses to magnetic stimulation reveal cortical reactivity and connectivity. Neuroreport 1997;10:3537–40. Izumi S, Takase M, Arita M, Masakado Y, Kimura A, Chino N. Transcranial magnetic stimulation-induced changes in EEG and responses recorded from the scalp of healthy humans. Electroencephalogr Clin Neurophysiol 1997;103:319–22. Kahkonen S, Wilenius J, Komssi S, Ilmoniemi RJ. Distinct differences in cortical reactivity of motor and prefrontal cortices to magnetic stimulation. Clin Neurophysiol 2004;115:583–8. Kobayashi M, Pascual-Leone A. Transcranial magnetic stimulation in neurology. Lancet Neurol 2003;2:145–56. Komssi S, Aronen HJ, Huttunen J, Kesaniemi M, Soinne L, Nikouline VV, et al. Ipsi- and contralateral EEG reactions to transcranial magnetic stimulation. Clin Neurophysiol 2002;113:175–84. Komssi S, Kahkonen S. The novelty value of the combined use of electroencephalography and transcranial magnetic stimulation for neuroscience research. Brain Res Rev 2006;52:183–92. Komssi S, Savolainen P, Heiskala J, Kahkonen S. Excitation threshold of the motor cortex estimated with transcranial magnetic stimulation electroencephalography. Neuroreport 2007;18:13–6. Lehmann D, Skrandies W. Reference-free identification of components of checkerboard-evoked multichannel potential fields. Electroencephalogr Clin Neurophysiol 1980;48:609–21. Miller JW, Kim W, Holmes MD, Vanhatalo S. Ictal localization by source analysis of infraslow activity in DC-coupled scalp EEG recordings. Neuroimage 2007;35:583–97. Morbidi F, Garulli A, Prattichizzo D, Rizzo C, Manganotti P, Rossi S. Off-line removal of TMS-induced artifacts on human electroencephalography by Kalman filter. J Neurosci Methods 2007;162:293–302. Paus T, Sipila PK, Strafella AP. Synchronization of neuronal activity in the human primary motor cortex by transcranial magnetic stimulation: an EEG study. J Neurophysiol 2001;86:1983–90. Putnam LE, Johnson Jr R, Roth WT. Guidelines for reducing the risk of disease transmission in the psychophysiology laboratory. SPR Ad Hoc Committee on the Prevention of Disease Transmission. Psychophysiology 1992;29:127–41. Schurmann M, Nikouline VV, Soljanlahti S, Ollikainen M, Basar E, Ilmoniemi RJ. EEG responses to combined somatosensory and transcranial magnetic stimulation. Clin Neurophysiol 2001;112:19–24. Siebner HR, Rothwell J. Transcranial magnetic stimulation: new insights into representational cortical plasticity. Exp Brain Res 2003;148:1–16. Sparing R, Buelte D, Meister IG, Paus T, Fink GR. Transcranial magnetic stimulation and the challenge of coil placement: A comparison of conventional and stereotaxic neuronavigational strategies. Hum Brain Mapp 2007. Swanson DK, Webster JG. A model for skin-electrode impedance. In: Miller HA, Harrison DC, editors. Biomedical electrode technology – Theory and practice. New York: Academic Press; 1974. p. 117–28. Tallgren P. DC-stable electrode–skin interface for human EEG recordings. Series E: Electronic Publications E 5 (http://www.hut.fi/Yksikot/ Elektroniikka/reports), Helsinki, 2005. Tam HW, Webster JG. Minimizing electrode motion artifact by skin abrasion. IEEE Trans Biomed Eng 1977;24:134–9. Thut G, Ives JR, Kampmann F, Pastor MA, Pascual-Leone A. A new device and protocol for combining TMS and online recordings of EEG and evoked potentials. J Neurosci Methods 2005;141:207–17. Vanhatalo S, Voipio J, Kaila K. Full-band EEG (FbEEG): an emerging standard in electroencephalography. Clin Neurophysiol 2005;116:1–8. Virtanen J, Ruohonen J, Naatanen R, Ilmoniemi RJ. Instrumentation for the measurement of electric brain responses to transcranial magnetic stimulation. Med Biol Eng Comput 1999;37:322–6. Wright MA, Orth M, Patsalos PN, Smith SJ, Richardson MP. Cortical excitability predicts seizures in acutely drug-reduced temporal lobe epilepsy patients. Neurology 2006;67:1646–51.