Brain Stimulation (2009) 2, 234–7
Exploring the optimal site for the localization of dorsolateral prefrontal cortex in brain stimulation experiments Paul B. Fitzgerald, MBBS, MPM, PhD, FRANZCPa, Jerome J. Maller, BSc, GradDipPsych, MSc, PhDa, Kate E. Hoy, BBNSc (Hons), DPsych (Clin Neuro)a, Richard Thomson, BEng (Hons), PhDa, Zafiris J. Daskalakis, MD, FRCP(C)b a
Alfred Psychiatry Research Centre, The Alfred and Monash University School of Psychology, Psychiatry and Psychological Medicine, Melbourne, Victoria, Australia b Centre for Addiction and Mental Health, University of Toronto, Toronto, Ontario, Canada
Background Dorsolateral prefrontal cortex (DLPFC) is a common target for repetitive transcranial magnetic stimulation (rTMS) experiments and therapeutic protocols. Objective The aim of this study was to investigate the optimal method for the localization of DLPFC for use in these studies. Methods Twelve healthy subjects underwent a structural magnetic resonance imaging (MRI) scan, a TMS procedure to establish the location of the motor cortex and a neuronavigational procedure to assess the relative position of the DLPFC. Several electroencephalographic (EEG) points and a position 5 cm anterior to motor cortex were established. Results The DLPFC site used was identified as being approximately halfway between the EEG points F3 and AF3. This point is considerably more anterior than the point identified by measuring 5 cm anterior to motor cortex. The study was supported in part by a National Health and Medical Research Council (NHMRC) project grant (436710) and the Neurosciences Australia Clinical Neurobiology of Psychiatry Platform. PF and ZD have received support for research conducted with Neuronetics Inc, a TMS equipment manufacturer. PF was supported by a Practitioner Fellowship grant from the NHMRC and ZD by a NARSAD Young Investigator award. Correspondence: Professor Paul B. Fitzgerald, Alfred Psychiatry Research Centre, First Floor, Old Baker Building,The Alfred, Commercial Rd Melbourne, Victoria, Australia, 3004. E-mail address: [email protected]
Submitted January 8, 2009. Accepted for publication March 4, 2009.
1935-861X/09/$ -see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.brs.2009.03.002
Prefrontal Cortex and rTMS
Conclusions EEG points provide a useful way to optimally identify DLPFC. Ó 2009 Elsevier Inc. All rights reserved. Keywords repetitive transcranial magnetic stimulation; depression; prefrontal cortex; response; remission; antidepressant
A large number of studies have investigated the use of repetitive transcranial magnetic stimulation (rTMS) in the treatment of major depressive disorder (MDD).1-4 Most of the published trials have indicated efficacy, but many studies have reported limited effect sizes or response rates.5 One potential reason for the moderate clinical effects seen with rTMS treatment is that standard methods may not ensure that stimulation is provided to the target region in many patients. The target site in the majority of the rTMS trials conducted is the dorsolateral prefrontal cortex (DLPFC). In the original reports that used rTMS treatment in depression, a standard procedure was proposed for the localization of DLPFC that involves the localization of the motor cortex by providing TMS pulses to the relevant motor cortex inducing muscle twitches in the contralateral hand muscle, usually the abductor pollicis brevis (APB). The coil is then placed for treatment 5 cm forward from this spot in a sagittal plane. However, when the validity of this was investigated by Herwig et al.,6 it resulted in targeting of Brodman area (BA) 9 in only seven of 22 subjects. Because of this concern, we recently conducted a trial comparing rTMS treatment localized using the standard method with rTMS treatment localized to DLPFC using an magnetic resonance imaging (MRI)-based neuronavigational procedure. The latter approach resulted in a greater treatment effect than the former.7 For the purpose of this clinical trial we selected a target in DLPFC based on a recent redefinition of the outline of this region.8 We selected a site at the junction of areas 9 and 46 identified by the coordinates in the Talairach and Tournoux system of –45, 45, 35.9 As the hardware and expertise to conduct neuronavigation procedures would not be available in all clinical settings, in this study we chose to investigate whether there was a relatively reliable and simple way of accurately targeting this DLPFC site (–45, 45, 35). We were also interested in investigating the relationship between this site and that identified by the standard 5-cm method.
Methods and materials The procedure was conducted on 12 healthy subjects (men 5 7, women 5 5) with an age range of 26-57 years (mean age 5 36.4 years [SD 5 12.5]; males mean 5 39.0 [SD 5 13.4], females mean 5 32.8 [SD 5 11.3]). All subjects gave informed consent in a manner approved by the Alfred Hospital Ethics Committee.
MRI scan and processing All subjects underwent a structural T1 weighted MRI scan with six fiducial markers (vitamin E capsules) in place at a variety of anatomic locations. The scans were threedimensional (3D) sagittally orientated T1-weighted structural MR scans conducted on a 1.5T, GE Signa Imaging System (General Electric Medical Systems, Milwaukee, Wisconsin). The sequence was a contiguous AC-PC aligned IR-prepared SPGR (TR 5 8.628, TE 5 1.924, IT 5 300, matrix size 5 256 3 256, 0.86 3 0.86 mm, NEX 5 1, slice thickness 5 1.5 mm, number of slices 5 128). Slices were then resliced, voxels converted to be isotropic (1.00 mm3), and the anterior commissure reset to 0,0,0. These scans were used in two ways. First, the raw scans were rendered to produce 3D images, and we then produced a brain-only image using the brain extraction tool (BET).10 Second, the raw scans were processed through SPM99 to match Talairach space. The T1-weighted scan of colin27 was used as the template. A skull stripped brain-only image and intact whole head 3D rendered versions were produced.
TMS After the MRI, single-pulse TMS (using a Magstim 200 stimulator and 70 mm figure-of-eight coil [Magstim Company Ltd, Sheffield, United Kingdom]) was used to establish the optimal scalp location for the stimulation of the right APB muscle (M1). A distance of 5 cm was measured anterior to this point in a sagittal plane and this location was marked on the scalp. The scalp locations, C3, AF3, and F3 from the 10-20 electroencephalographic (EEG) system were also measured and marked on the subjects scalp.
Neuronavigational procedure After this, each subject underwent a neuronavigational procedure using standard techniques for the Ascension miniBIRD (Ascension Technology Corporation, Milton, Vermont) and MRIcro/MRIreg package to enable each participant’s MR brain scan to be coregistered with their real head geometry.11 The neuronavigational protocol was conducted using the raw scan yoked to the normalized images by following a number of steps. First, the participant’s head was placed in a frame that kept the head still. Each fiducial location on the MRI scan was registered to
P.B. Fitzgerald et al
the matching point on the person’s head, by registering the location of those fiducials in 3D Talairach space with the miniBIRD magnetic wand position on the head, using MRIreg. MRIreg computes a multiple linear regression to map the marker coordinates onto MRI coordinates (least squares linear estimation). Second, as a quality control, the reliability of the localization technique was confirmed by navigating the sensor/wand to known anatomic points such as nose or ears on the nonskull stripped scan and confirming that these points were accurately identified on the MRI scan. Third, the MRIreg ‘‘tracker’’ mode was used. MRIreg was used to identify the scalp surface position corresponding to the cortical brain site identified by the normalized Talairach coordinates –45, 45, 35, which was then marked on the scalp. The wand/probe was then successively placed on each of the previously marked points on the scalp (5 cm anterior to M1, AF3, F3, C3) and the corresponding Talairach coordinates were recorded from the MRIcro software.
left DLPFC was near the midpoint between the positions of F3 and AF3 (Figure 1).
Calculation of distance between AF3, F3, M1 and M1 1 5 cm anterior, and left DLPFC
The distance between the –45, 45, 35 site and the previously marked points (5 cm anterior to M1, AF3, F3) was then calculated by using these recorded coordinates. The curvature of the head between coordinates was accounted for by using the T1 image to calculate the path length, using a customized algorithm we wrote in Matlab 7.0 that included functions from SPM2. First, the oblique slice defined by the two X,Y coordinates was extracted and normalized for voxel intensity. The air/scalp interface was then defined by a 15% (of maximum) voxel intensity threshold. The arc between the two coordinates across the scalp was smoothed using a 5-point moving average filter applied in both directions. This corrected for voxel quantization error and noise artefact across the interface that would otherwise artificially increase the path distance.
Results All the subjects participated in all of the procedures. As shown in Table 1, the mean raw distance from M1 to left DLPFC was 91.81 6 7.83 mm. The DLPFC site was closest to AF3 (19.74 6 7.37 mm) and F3 (22.74 6 12.7 mm). When visualized using normalized coordinates, the Table 1 Mean Min Max SD
Figure 1 Results of each subject’s electrode positions relative to their left dorsolateral prefrontal cortex (DLPFC).
The main conclusion from this analysis is that the DLPFC site we have tried to target is substantially more anterior to the site identified by the M1 1 5 cm method and is approximately at the midpoint of a line drawn between AF3 and F3. This would allow the use of the relatively simple system for measuring 10-20 EEG coordinates as an alternative method for the localization of DLPFC in therapeutic and investigative studies. The left DLPFC, as defined by the Talairach coordinates –45, 45, 35, was actually approximately 9 cm anterior to the M1 site at which the APB muscle is optimally stimulated. This is substantially beyond the distance usually used to locate DLPFC for the purpose of rTMS studies. As the range among our sample of 12 healthy individuals was 8-10 cm and the focus of a TMS pulse is approximately 1.5-2 cm below the scalp,12 using the standard 5-cm rule would not allow the TMS pulse to actually reach this version of DLPFC in most people. An important implication of these results is that it confirms the findings of Herwig et al.6 that the standard 5cm method for localizing DLPFC seems to generally identify a spot relatively posterior in the DLPFC. In this context, it seems plausible to propose that this could be one of the major reasons for the somewhat moderate
Distances (mm) from AF3, F3, M1 ‘‘hot spot’’ and M1 1 5 cm anterior, to left DLPFC M1
M1 1 5 cm
91.81 79.51 101.15 7.83
48.16 34.79 57.18 7.98
22.71 3.15 48.46 12.70
19.74 2.45 29.48 7.37
DLPFC 5 dorsolateral prefrontal cortex; Min 5 minimum; Max 5 maximum; SD 5 standard deviation.
Prefrontal Cortex and rTMS efficacy reported in the majority of rTMS trials.5 This is supported by our recent trial results,7 but confirmation of this proposition will require replication and direct analysis. It is notable that we found that the left DLPFC is approximately at the midpoint between the F3 and AF3 electrode positions, slightly closer to F3. This is consistent with studies that have used an F3 brain stimulation site.13 As the 10-20 system is a straight-forward technique to measure electrode positions, our findings support the use of this simple system for DLPFC targeting. The use of the 10-20based system has a considerable degree of face validity as the system accounts for individual variation in head size that is not accounted for in the M1 1 5 cm approach. However, before we can propose the clinical adoption of any of the techniques for localization of the DLPFC, it is pertinent to question whether we really understand the best location for treatment targeting. In a previous report, we attempted to address this question by systematically analyzing the results of the imaging studies of depression with a meta-analysis technique.14,15 Changes in BA9 were relatively consistently identified across imaging studies. Classically defined BA9 extends relatively posterior in the prefrontal cortex. rTMS localized with the M1 1 5 cm method could well produce significant stimulation of the posterior areas of BA9, even though this is not necessarily close to any form of midpoint of DLPFC. However, there was little consistency in the direction of these changes or the hemisphere in which they were identified. This places substantial limitations on the use of existing neuroimaging studies to guide the therapeutic target for rTMS. If we were to presume that the site used in the current study (–45, 45, 35) would be an ideal site to target in rTMS treatment, our results would imply that the localization of treatment stimulation should be reconsidered. However, it is difficult to justify this assumption and the body of studies demonstrating the therapeutic effects of rTMS argues there is value in treatment at a more posterior site. Until further research can elucidate the region within DLPFC that should be optimally targeted, a conservative conclusion may be to target a site somewhat intermediary between the traditionally used location and that used here. F3 would seem to be a logical and relatively practical choice in this regard.
References 1. Fitzgerald PB, Benitez J, de Castella A, Daskalakis ZJ, Brown TL, Kulkarni J. A randomized, controlled trial of sequential bilateral repetitive transcranial magnetic stimulation for treatment-resistant depression. Am J Psychiatry 2006;163:88-94. 2. Fitzgerald PB, Brown TL, Marston NA, Daskalakis ZJ, De Castella A, Kulkarni J. Transcranial magnetic stimulation in the treatment of depression: a double-blind, placebo-controlled trial. Arch Gen Psychiatry 2003;60:1002-1008. 3. George MS, Nahas Z, Molloy M, et al. A controlled trial of daily left prefrontal cortex TMS for treating depression. Biol Psychiatry 2000; 48:962-970. 4. O’Reardon JP, Solvason HB, Janicak PG, et al. Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: a multisite randomized controlled trial. Biol Psychiatry 2007;62:1208-1216. 5. Loo CK, Mitchell PB. A review of the efficacy of transcranial magnetic stimulation (TMS) treatment for depression, and current and future strategies to optimize efficacy. J Affect Disord 2005;88: 255-267. 6. Herwig U, Padberg F, Unger J, Spitzer M, Schonfeldt-Lecuona C. Transcranial magnetic stimulation in therapy studies: examination of the reliability of ‘‘standard’’ coil positioning by neuronavigation. Biol Psychiatry 2001;50:58-61. 7. Fitzgerald PB, Hoy K, McQueen S, et al. A randomized trial of rTMS targeted with MRI based neuro-navigation in treatment resistant depression. Neuropsychopharmacology 2009;34:1255-1262. 8. Rajkowska G, Goldman-Rakic PS. Cytoarchitectonic definition of prefrontal areas in the normal human cortex: II, variability in locations of areas 9 and 46 and relationship to the Talairach Coordinate System. Cereb Cortex 1995;5:323-337. 9. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain. Stuttgart, Germany: Thieme Medical Publishers; 1988. 10. Smith SM. Fast robust automated brain extraction. Hum Brain Mapp 2002;17:143-155. 11. Rorden C, Brett M. Stereotaxic display of brain lesions. Behav Neurol 2000;12:191-200. 12. Rudiak D, Marg E. Finding the depth of magnetic brain stimulation: a re-evaluation. Electroencephalogr Clin Neurophysiol 1994;93:358-371. 13. Cerruti C, Schlaug G. Anodal transcranial direct current stimulation of the prefrontal cortex enhances complex verbal associative thought [published online ahead of print October 14, 2008]. J Cogn Neurosci doi:10.1162/jocn.2008.21143. 14. Fitzgerald PB, Laird AR, Maller J, Daskalakis ZJ. A meta-analytic study of changes in brain activation in depression. Hum Brain Mapp 2008;29:683-695. 15. Fitzgerald PB, Oxley TJ, Laird AR, Kulkarni J, Egan GF, Daskalakis ZJ. An analysis of functional neuroimaging studies of dorsolateral prefrontal cortical activity in depression. Psychiatry Res 2006;148:33-45.