Neurology -- Correspondence for Mottaghy et al., 60 (9) 1539-1541

HOME

HELP

FEEDBACK

SUBSCRIPTIONS

Correspondence published:

Repetitive TMS temporarily alters brain diffusion: Thomas Duning, Andreas Rogalewski, Olaf Steinstraeter, Harald Kugel, Andreas Jansen, Caterina Breitenstein and Stefan Knecht (11 December 2003)

Reply to Duning et al: Felix Mottaghy, A. Pascual-Leone, MD, PhD, G. Schlaug G, MD, PhD (11 December 2003)

Repetitive TMS temporarily alters brain diffusion 11 December 2003

Thomas Duning,
Department of Neurology, University of Muenster, Germany
Albert-Schweitzer-Str. 33, 48129 Muenster, Germany,
Andreas Rogalewski, Olaf Steinstraeter, Harald Kugel, Andreas Jansen, Caterina Breitenstein and Stefan Knecht
Send Correspondence to journal:
Re: Repetitive TMS temporarily alters brain diffusion

duningt{at}uni-muenster.de Thomas Duning, et al.

Repetitive transcranial magnetic stimulation (rTMS) can temporarily alter brain excitability and behavior. [1,2] Based on diffusion-weighted magnetic resonance measurements (DWI) of apparent diffusion coefficients (ADC), Mottaghy et al. proposed that rTMS may alter brain diffusion. [3]
This is an important finding with respect to the safety of rTMS. However, comparison of serial DWI is difficult, because results can vary. We would like to illustrate some methodological problems with DWI based on a replication attempt using a 3 Tesla MR system. Measures of ADC were not reproducible unless the following aspects were considered:
1. For calculation and comparison of relative ADCs, the regions of interests (ROI) must be precisely defined. For example, if meninges or cerebrospinal fluid are included, even smallest misalignments will markedly affect ADC values (Fig. 1). This also makes an interhemispheric comparison of ROIs and appropriate ADC values problematic because the hemispheres differ regionally in morphology.
2. The ADC values of pre- and post-TMS scans must be statistically compared. Our data show that the variance within the pre- and post scans was as extensive as the variance between conditions. Mottaghy et al only normalized ADC values to the last DWI of each condition and did not report the variance of the pre-TMS condition.
We have not yet observed a consistent change in proton diffusion after 20 minutes of rTMS at 1Hz and 100% of motor threshold over the left hand motor cortex. This is not unexpected since DWI findings are often negative in patients with transient ischemic attacks or even in acute strokes .[4,5]
We believe that Mottaghy et al have raised an important issue in examining the effects of rTMS on brain diffusion. However, given the current lack of experience with DWI, more work is needed to draw firm conclusions about alterations in brain diffusion.
Figure Legend: Influence of different regions of interest (ROI) on the time course of the apparent diffusion coefficient (ADC) - changes after repetitive transcranial magnetic stimulation (rTMS) Figures A-C illustrate three different ROIs (superimposed on T1-weighted MR-image), each centred on the left hand area of the precentral gyrus. ROI in figure A covers the complete intracranial volume (meninges and cerebrospinal fluid included), whereas ROIs in B and C are restricted to the extracted brain B: Gray and white matter, C: Limited to the gray matter. Tables below these figures display the time course of the intensity levels of the ADC images averaged over the associated ROIs. MR data sets before (black bars) and after (grey bars) rTMS are normalized to the last scan of the train (marked by the line and taken as 100%, similar to the study by Mottaghy et al.). Acquisition time of the diffusion scans was 141 seconds. DWI scans after rTMS were acquired in direct succession (Time points I-VI
The characteristics of the ADC time course highly depends on the underlying ROI and can lead to varied interpretations. Therefore, for reproducible results a detailed definition of the used ROI is needed.
References
1. Kobayashi M, Pascual-Leone A. Transcranial magnetic stimulation in neurology. Lancet Neurol. 2003 Mar;2(3):145-156.
2. Currà, N. Modugno, M. Inghilleri, M. Manfredi, M. Hallett, and A. Berardelli. Transcranial magnetic stimulation techniques in clinical investigation. Neurology. 2002 Dec;59:1851-1859.
3. Mottaghy FM, Gangitano M, Horkan C, Chen Y, Pascual-Leone A, Schlaug G. Repetitive TMS temporarily alters brain diffusion. Neurology. 2003 May 13;60(9):1539-1541.
4. Lovblad KO, Laubach HJ, Baird AE, Curtin F, Schlaug G, Edelman RR, Warach S. Clinical experience with diffusion-weighted MR in patients with acute stroke. AJNR Am J Neuroradiol. 1998 Jun-Jul;19(6):1061-1066.
5. Oppenheim C, Stanescu R, Dormont D, Crozier S, Marro B, Samson Y, Rancurel G, Marsault C. False-negative diffusion-weighted MR findings in acute ischemic stroke. AJNR Am J Neuroradiol. 2000 Sep;21(8):1434-1440.

Reply to Duning et al 11 December 2003

Felix Mottaghy,
Department of Nuclear Medicine, University Hospital Ulm
Robert-Koch-Str. 8, D-89070 Ulm, Germany,
A. Pascual-Leone, MD, PhD, G. Schlaug G, MD, PhD
Send Correspondence to journal:
Re: Reply to Duning et al

fmottaghy{at}yahoo.de Felix Mottaghy, et al.

A reduction of apparent diffusion coefficients (ADC) in diffusion- weighted imaging (DWI) after repetitive transcranial magnetic stimulation (rTMS), even if transient and minimal, is of importance regarding safety and physiological effects of rTMS. Li et al [2] failed to find changes in DWI in patients with depression when rTMS was applied prefrontally. Duning et al now report not having been able to replicate our findings in normal subjects. [1] Replication studies of new findings are essential, but it is important to make sure that experimental conditions are the same. Duning et al used a 3 Tesla scanner, which will lead to a significantly greater image distortion than the 1.5 Tesla system we used. This is particularly true for echo planar images, which we assume Duning et al used to acquire DWI. It is not surprising that Duning et al got highly variable results given the ROIs used. The first two ROIs clearly contain CSF, which typically has very high ADC values. The third ROI seems to be more appropriate for a motor region, but it would be critical to know how its extent was defined and how DWI and T1 weighted images were co-registered to prevent CSF signal from influencing the tissue signal. We took great care in defining our ROI. [1]
First, each individual, high resolution (1mm3) T1-dataset was standardized into Talairach space. This transformation matrix was then applied to each DWI-dataset. We then determined ROIs within the posterior bank of the precentral sulcus of each hemisphere separately. [3] The dorso -ventral extent was similar in each subject and was chosen based on functional imaging studies of index finger movements. The regional ADC values were calculated offline using regional signal intensities from the b0 and b1000 of each DWI dataset. This was done for each hemisphere and for each subject individually. Our ROIs did not differ significantly in size between right and left hemisphere. Comparing the variance of the data pre- and post- rTMS yielded also no significant difference validating our approach to compare the time-course of ADC changes within condition.
Regarding the significance of our findings, it is important to note that ADC reduction over the targeted M1 after rTMS was minimal and recovered over few minutes. [1] Various animal experiments and some human studies [4] have indicated that such minimal ADC reductions may have physiological implications and be associated with changes in blood flow, but they are significantly different than what is commonly seen in ischemia (typically less than 70% of normal). [5]
References
1. Mottaghy FM, Gangitano M, Horkan C, Chen Y, Pascual-Leone A, Schlaug G. Repetitive TMS temporarily alters brain diffusion. Neurology. 2003;60:1539 -1541.
2. Li X, Nahas Z, Lomarev M, Denslow S, Shastri A, Bohning DE, George MS. Prefrontal cortex transcranial magnetic stimulation does not change local diffusion: a magnetic resonance imaging study in patients with depression. Cogn Behav Neurol. 2003; 16:128-35
3. Amunts K, Schlaug G, Jäncke L, Steinmetz H, Schleicher A, Dabringhaus A, Zilles K. Motor cortex and hand motor skills: structural compliance in the human brain. Human Brain Mapping 1997;5:206-215.
4. Schlaug G, Siewert B, Benfield A, Edelman RR, Warach S. Time course of the apparent diffusion coefficient (ADC) abnormality in human stroke. Neurology 1997;49:113-119.
5. Schlaug G, Benfield A, Baird AE, Siewert B, Lovblad KO, Parker RA, Edelman RR, Warach S. The ischemic penumbra: defining brain tissue at risk for infarction using diffusion and perfusion MRI. Neurology 1999;53:1528- 1537.