TRANSCRANIAL MAGNETIC STIMULATION is a non-invasive method of brain stimulation in which magnetic fields are used to induce electric currents in the cerebral cortex, depolarizing neurons. Though it has been reported that repetitive transcranial magnetic stimulation (rTMS) applied to the left dorsolateral prefrontal cortex (DLPFC) is an effective treatment for depression,1,2 its precise mechanisms are poorly understood. However, the involvement of the dopaminergic system has been suggested in elucidating the therapeutic mechanisms of rTMS. Several human studies demonstrated that acute rTMS induced the release of endogenous dopamine in the striatum.3–5 Nevertheless, we recently reported that there were no changes in [11C]raclopride binding in the striatum after 10-daily rTMS treatment in the patients with depression.6 In the present study, we have examined the effects of 10-daily sessions of rTMS on dopaminergic function in depressed patients using positron emission tomography (PET) with L-[β-11C]DOPA. L-[β-11C]DOPA can be used to assess the rate of endogenous dopamine synthesis as a way to estimate presynaptic function of the central dopaminergic system.7
We have examined the effects of repetitive transcranial magnetic stimulation (rTMS) on central dopaminergic function in patients with depression using positron emission tomography with L-[β-11C]DOPA, a ligand to assess the rate of endogenous dopamine synthesis. Eight patients were treated with 10-daily sessions of rTMS over the left dorsolateral prefrontal cortex. Positron emission tomography scanning was performed in each patient twice, before the first session and 1 day after the last session. Although four out of eight patients responded to rTMS, there were no changes in the striatal dopamine synthesis rate (k) following rTMS. These results suggest that chronic rTMS had a limited effect on the dopaminergic system.
Eight patients (four women and four men aged 36.8 ± 6.4 years [mean ± SD]) with a DSM-IV-TR diagnosis of major depressive disorder participated in this study. Three patients had a recurrent type of depressive disorder (one patient had two previous episodes and two patients had one previous episode). Duration of the present episodes was 20.5 ± 13.3 months. They were all right-handed. The patients were free of somatic or neurological disorders on the basis of their medical history, blood test, electrocardiogram, chest X-rays, brain computed tomography and electroencephalogram. Their clinical pictures were melancholic without psychotic or catatonic features. Their suicide ideas were not strong. They had no comorbid psychiatric disorders, such as anxiety disorders, alcohol dependence or personality disorders. The patients were resistant to or intolerant of drug treatment. One patient did not respond to electroconvulsive therapy. Three patients were resistant to two adequate trials of antidepressants. Two patients did not respond to one adequate antidepressant trial. Three patients had taken atypical antipsychotics as augmentation agents. They had at least a 4-week washout period from their previous medication. Fluvoxamine and lorazepam were allowed during the washout period with little clinical improvement and pharmacological dosages were kept constant during the study. The average dose of fluvoxamine during the study was 115.6 ± 88.6 mg/day. Six patients had not been able to take fluvoxamine in sufficient dosage, more than 150 mg/day, because of side-effects, such as sleepiness, anxiety, and nausea. This study was approved by the human ethics committees of Tokyo Medical and Dental University, Tokyo, Japan, and the National Institute of Radiological Sciences, Chiba, Japan. Following a description of all procedures, subjects provided written informed consent.
The Magstim Rapid System (Magstim Company Limited, Spring Gardens, Whitland, UK) with an eight-shaped coil was used to administer the rTMS treatments. Each patient was treated with 10 sessions (five times per week for 2 weeks) using the following parameters per session: 10 Hz frequency, 20 trains of 5 s duration separated by 25 s, and an intensity of 100% motor threshold (MT). These parameters were within the safety criteria.8 At entry, each patient's MT was determined using the visual method with the right first dorsal interosseous (FDI) muscle as the target muscle.9 MT was defined as the stimulus intensity that produced visibly observable right FDI muscle contractions at least five times out of 10 stimuli. rTMS was performed over the left dorsolateral prefrontal cortex (DLPFC). The point of left DLPFC was determined by moving the coil 5 cm anteriorly from the point of MT determination. The point of stimulation was marked for reference with an indelible skin marker. MT and coil placement were rechecked after the fifth treatment to assess, but neither MT nor coil placement was readjusted for any patient. During rTMS the patients wore earplugs to dampen the loud noise from the discharging coil.
To assess improvement of the clinical symptoms, all patients were evaluated with the 17-item Hamilton Rating Scale for Depression (HRSD)10 twice: 1 day before and 1 day after a series of rTMS.
The first study was performed before a series of rTMS and the second study was performed 1 day after the last session of rTMS. Each PET scan began more than 3 h after the last medication. All PET studies were performed with an ECAT EXACT HR+ (CT1-Siemens, Knoxville, TN, USA) in 3-D mode, which provides 63 planes with a 15.5-cm axial field of view.
L-[β-11C]DOPA was synthesized from [11C]carbon dioxide via D,L-[3-11C]alanine as described previously.11,12 After a 10-min transmission scan with a 68Ge-68Ga source, a bolus of L-[β-11C]DOPA was rapidly injected into the antecubital vein with a 20-ml saline flush. Injected dose was 9.9 ± 1.8 mCi and 10.2 ± 0.5 mCi before and after rTMS, respectively. The specific radioactivity was 78.3 ± 42.5 GBq/µmol and 52.3 ± 16.2 GBq/µmol before and after rTMS, respectively. Dynamic PET scanning was started immediately after the injection and continued for 60 min.13 A head fixation device with thermoplastic attachments for individual fit minimized head movement during PET measurements. All emission scans were reconstructed using a Hanning filter with a cut-off frequency of 0.4 cycle/pixel (FWHM [full width at half maximum] = 7.5 mm).
Magnetic resonance (MR) images were obtained with a Philips Gyroscan NT, 1.5 tesla. Scan parameters were 1-mm thick 3-D T1-weighted images with a transverse plane. All MR images were coregistered to the PET images using statistical parametric mapping 2 (SPM2; http://www.fil.ion.ucl.ac.uk/spm/software/spm2/). Regions of interest (ROI) were manually placed on coregistered MR images and transferred to the corresponding PET images. ROI were set to cover three adjacent slices for the right and left caudate nucleus, right and left putamen and occipital cortex. Then, time-activity curves were obtained. For quantification of the striatal utilization of L-[β-11C]DOPA, the graphical analysis method using reference brain region (occipital cortex) developed by Patlak and Blasberg7,14 was used to calculate dopamine synthesis rate (k).
Response to treatment was defined as an HRSD less than 10 points or a 50% decrease in HRSD 1 day after rTMS. Paired t-test was used to statistically analyze the difference between HRSD scores 1 day before the first session and 1 day after the last session of rTMS. Values of P < 0.05 were considered significant.
Statistical analysis of the difference between k values in patients measured before rTMS and 1 day after the last session of rTMS was performed using two-way repeated anova to find interaction between time (before and after) and regions (right and left caudate nucleus, right and left putamen). When we found any interaction, post-hoc Bonferroni correction was used for multiple comparisons. Values of P < 0.05 were considered significant.
All patients tolerated rTMS without severe complications. HRSD scores decreased significantly following a series of rTMS (17.0 ± 2.4 before treatment; 10.6 ± 6.4 after treatment; P = 0.005, paired t-test). Four out of eight patients responded to rTMS.
Two-way repeated anova revealed no significant interactions between time and regions with regard to k-values for L-[β-11C]DOPA [F (3, 5) = 0.382, P = 0.771] (Table 1).
|Region||k before TMS||k after TMS|
|R Caudate||0.01435 ± 0.00327||0.01264 ± 0.00171|
|L Caudate||0.01400 ± 0.00184||0.01361 ± 0.00430|
|R Putamen||0.01495 ± 0.00145||0.01475 ± 0.00234|
|L Putamen||0.01507 ± 0.00223||0.01446 ± 0.00150|
This study has demonstrated that k-values of L-[β-11C]DOPA did not change in the striatum following rTMS. These results suggest that the rate of dopamine synthesis in the striatum did not significantly change 1 day after the last session of rTMS.
In our previous study, rTMS induced no significant changes in [11C]raclopride binding in the striatum.6 The results were not consistent with previous reports,3–5 showing that rTMS caused a reduction in [11C]raclopride or [123I]IBZM binding, indicating increased release of endogenous dopamine. This discrepancy could be explained by the following two reasons: first, we started [11C]raclopride PET scans about 24 h after the last session of rTMS, whereas the other authors started them soon after the rTMS session. Second, we applied chronic rTMS (multiple treatments), whereas the other authors used acute rTMS (single treatment). Therefore, we suggest that release of striatal dopamine induced by rTMS might only be transient, or that dopamine release may be attenuated following chronic rTMS. Tsukada et al.15 reported that dopamine synthesis rate, as measured by L-[β-11C]DOPA, could be more sensitive to evaluate the changes in dopaminergic neuronal activity than the ligand-receptor binding. Thus, present PET studies using L-[β-11C]DOPA also indicated that chronic rTMS had limited effect on the dopaminergic system.6
A major limitation of this study is the small sample size. Another shortcoming was the study design of an open trial lacking a sham-treated rTMS group. Placebo clinical effects could not be excluded. The drugs might have influenced dopamine synthesis as three patients had taken atypical antipsychotics and all of the patients had been previously treated with antidepressants. Because treatment duration and stimulation intensity in this study were insufficient by today's standards,1 further larger sample sizes and sham-controlled studies with longer duration and higher intensity on drug-free patients are needed.
In conclusion, although rTMS showed some efficacy for the treatment of depression, rTMS failed to change the striatal dopamine synthesis rate, as measured by L-[β-11C]DOPA PET. These results suggest that chronic rTMS had limited effect on the dopaminergic system.
A part of this work was supported by research grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (16591124 and 21659274) and the Ministry of Health, Labour, and Welfare of Japan (17A-5 and 20B-1). We wish to thank Dr Tomo Terada and Dr Hidenori Atsuta for their kind assistance in conducting the rTMS treatments. We also gratefully acknowledge Dr Jun Kosaka, Dr Yota Fujimura, Dr Shoko Nozaki, Dr Miho Ohta and Ms Yoshiko Fukushima for their assistance in performing the PET experiments.