Evaluation of Transcranial Magnetic Stimulation Efficiency in Major Depressive Disorder Patients: A Magnetic Resonance Spectroscopy Study.

OBJECTIVE: Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for major depressive disorder (MDD). This study evaluated the antidepressant effect of rTMS and examined how it affected N-asetyl aspartate (NAA), choline (Cho), creatine (Cr), lactate (Lac), myoinositol (mIns), glutamate (Glu), glutathione (GSH), and glutamine (Gln) metabolite levels in the left dorsolateral prefrontal cortex (DLPFC) of MDD patients who were not receiving antidepressant medication.
METHODS: In total, 18 patients (10 female, 8 male) were evaluated. Each patient underwent H magnetic resonance spectroscopy (H-MRS) before and within 3 days of completion of TMS therapy. All patients completed 20 sessions of rTMS directed at the left DLPFC over a 2-week period. The Hamilton Depression Scale (HAMD) scores of patients were calculated, and their responses to treatment were assessed within 1-3 days of completion of TMS.
RESULTS: We found statistically significant differences in HAMD scores before and after rTMS. Moreover, the peak metabolite ratios of NAA/Cr, GSH/Cr, and Gln/Cr were significantly higher after rTMS compared to those before rTMS.
CONCLUSION: Increased understanding of the mechanism of action of TMS will improve its application and may stimulate development of new-generation therapeutic agents.

INTRODUCTION

Repetitive transcranial magnetic stimulation (rTMS) is a noninvasive technique of brain stimulation that is based on the principles of electromagnetic induction to produce an electrical current on the surface of the brain cortex [1]. It is an efficient treatment option, particularly for patients with major depressive disorder (MDD) who are resistant to antidepressant medications [2,3]. Previous findings revealed that the antidepressant effect of rTMS results from its administration to a region of the prefrontal cortex [4,5]. Studies regarding the brain regions for administration have revealed that stimulating the dorsolateral Print ISSN 1738-3684 / On-line ISSN 1976-3026 OPEN ACCESS prefrontal cortex (DLPFC) is more efficacious. Rapid rTMS administration to the left side has an antidepressant effect, whereas stimulating the right side does not [6,7].

Very little is known about the neurobiological mechanisms of the antidepressant response to rTMS. Although its effect on cortical circuits is known, it is unclear whether the changes in the excitation–inhibition balance of synaptic networks affect the antidepressant response [8]. Animal trials have shown that rTMS accelerates serotonin and dopamine turnover [9]. It reduces 5-hydroxytryptamine (5HT)2 and beta adrenergic receptors in the frontal cortex while increasing 5HT1A receptors in the frontal cortex and cingulate gyrus. On the other hand, rTMS upregulates N-methyl-D-aspartate receptors in the ventromedial hypothalamus, basolateral amygdala and parietal cortex [10].

1H Magnetic resonance spectroscopy (H-MRS) is a non-invasive imaging method used to detect endogenous chemical metabolites, including N-acetylaspartate (NAA), choline (Cho), creatine (Cr), lactate (Lac), myoinositol (mIns), glutamate (Glu), glutathione (GSH), and glutamine (Gln). It is possible to yield information about the neurobiological structure of the brain using the concentrations of 17 different neurochemical substances determined by H-MRS on 3 Tesla MR systems [11]. The standard target region for rTMS is the left DLPFC. It has been emphasized in the literature that the same brain locations should be selected for 1H-MRS to assess metabolite changes in response to rTMS [12]. Cr is a reference metabolite that resonates at 3.0 ppm in 1H-MRS. As NAA is a neuronal marker, a low NAA/Cr ratio represents neuronal loss and axonal damage. Furthermore, a high Cho/Cr ratio reflects synthesis of membranes and cells because Cho exists in the cell membrane.

Several studies have investigated the efficacy of drug treatment [13], group therapy,14 and mood-regulating medications [15] by 1H-MRS, but to the best of our knowledge, the therapeutic efficacy of rTMS in the treatment of depression has not been investigated. Initial studies regarding the antidepressant mechanism of TMS revealed an association between frontal perfusion and the response of rTMS [16]. Subsequent investigations have shown that TMS leads to an augmentation in ion flow via a depolarization wave as well as to several changes in 5HT2 and noradrenergic receptors in neurons [17]. Some authors have also revealed that TMS acts on monoaminergic systems, such as serotonin, dopamine, gamma-aminobutyric acid and Glu [18,19].

In this study, we evaluated the antidepressant response of rTMS and how it affects the levels of NAA, Cr, Cho, Gln, GSH, mIns, Glu and Lac metabolites in the left DLPFC of patients with MDD who do not take antidepressant medication.

METHODS

Participants

In a power analysis [α=0.005 1–β (power): 0.80], the required sample size was calculated as at least 17 patients, so that the average change in brain metabolites in the depressed patients who undergo rTMS should be 6.5 units in 1H-MRS. A consort diagram model was used to describe the study patients.

Twenty-three patients who were diagnosed with MDD based on The Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5), criteria by a psychiatrist using the Structured Clinical Interview for DSM-5 and who did not take any antidepressant medication for at least 3 months were included. Other inclusion criteria were non-smokers, non-users of alcohol or other substances, no suicidal thoughts and/or any previous attempt, and eligible for rTMS treatment. All patients signed written informed consent forms to participate in the study. Exclusion criteria were pregnancy or lactation, diagnosis of a bipolar or psychotic disorder, a clinically significant personality disorder, history of a substance abuse disorder or substance addiction, history of epilepsy or any other neurological illness, history of closed head trauma accompanying loss of consciousness and left-handedness. Each patient underwent 1H-MRS before and within 3 days of completing TMS therapy. Three patients were excluded due to inappropriate spectra caused by motion artefacts, and two patients were excluded because they did not attend the second spectroscopic imaging after TMS. No side effects were observed during TMS treatment, and all patients completed the treatment. Eighteen patients were evaluated for the analysis. All patients underwent H-MRS and completed the HAMD at the same time within 1–3 days after completion of 2 weeks of TMS treatment to assess treatment responses. This study protocol was approved by the Ethics Committee of Inonu University School of Medicine (IRB authorization number: 2017/05) and conducted in accordance with the ethical principles of the 2013 Declaration of Helsinki and Good Clinical Applications.

rTMS protocol

All patients completed 20 sessions of rTMS directed to the left DLPFC over a 2-week period using the Neuro-MS/D TMS Therapy System (Neurosoft, Ltd., Ivanova, Russia). Individual sessions consisted of 37.5 min (3,000 pulses; 30-s duty cycle of 4 s on and 26 s off) and 10-Hz excitatory TMS twice a day for 10 days (Monday–Friday over a 2-week period). The resting motor threshold was defined as the stimulus strength over the thumb area of the motor cortex that produced visually detectable thumb motion. The standardized treatment location was the left DLPFC (middle frontal gyrus). Initial collocation was determined according to the “5-cm rule” (5 cm forward to the resting motor threshold site along a left superior oblique plane, with a rotation point about the tip of the participant’s nose) [20].

1H-MRS neuroimaging procedures

All patients underwent MRS imaging before and after the rTMS procedure. Whole MRS images were taken at the Inonu University School of Medicine, Radiology Department on a 3T MR device (Magnetom Skyra-Siemens, Erlangen, Germany) using of a 20-channel phase array head coil. The same brain regions were carefully selected to assess the effect of TMS on the metabolite ratios during MRS [12]. To determine voxel localization and exclude parenchymal lesions, T1-weighted sagittal three-dimensional magnetization-prepared rapid-acquisition gradient echo and T2-weighted fluid-attenuated inversion recovery axial–sagittal images were obtained, respectively. The imaging parameters were as follows: TR: 2,300 ms; TE: 2.98 ms; slice thickness: 1 mm; FOV 256 mm; matrix size: 240×256. After ruling out pathological lesions on the T2 sequence, thin slice images (1 mm) at three orthogonal planes (sagittal, axial and coronal) were obtained by multiplanar reconstruction. The borders of the dorsolateral prefrontal lobe were delineated on the coronal plane by tracing a line manually, combining the superior frontal sulcus, lateral fissure and outer surface of the cortex in the left cerebral hemisphere at the level of the corpus callosum (genu) (Figure 1). We placed a single 15×20×15 mm volume of interest in a cortical region outside the sulci to avoid cerebrospinal fluid contamination, and single voxel spectroscopy–short echo spectroscopic imaging was performed using a point-resolved spectroscopy sequence (TR: 2000, TE: 30) (Figure 2). The optimal proton signal was obtained by additional “manual shimming” to enhance local magnetic homogeneity in the voxel. Spectroscopic data were transferred to a work station, and peak metabolite ratios were calculated automatically using software (syngo.via-Siemens Healthineers, Erlangen, Germany) after establishing the spectra by proportionating the integral values of the metabolites to that of Cr, which was used as the reference metabolite [21].

Figure 1.

DLPFC segmentation by manually tracing. DLPFC boundaries delineated through a line combining superior frontal sulcus (short arrow), posterior lateral sulcus (long arrow) and outer surface of cortex. DLPFC: dorsolateral prefrontal cortex.

Figure 2.

VOI placing into left DLPFC at 3 planes and MR spectrum. VOI: Volume of Interest, DLPFC: Dorsolateral Prefrontal Cortex, MR: magnetic resonance

Statistical analysis

Statistical analyses were performed using the Statistical Program for Social Sciences for Windows, version 17.0, software (SPSS Inc., Chicago, IL, USA). Descriptive statistics for continuous variables are presented as means and standard deviations, and the categorical variables are presented as numbers and percentages. Normality of the data distribution was assessed using the Shapiro-Wilk test. Normally distributed data are summarized as means and standard deviations, and the t-test was used to compare dependent groups. Data not following a normal distribution are summarized as medians and ranges and were compared using the Wilcoxon test. Spearman’s rank correlation test was used to assess the strength of the relationship between the metabolite levels and HAMD scores. A p-value< 0.05 was considered significant.

RESULTS

The mean age of the patients included in this study was 43.38±11.14 years. Of the 18 patients, 10 were female (55.6%) and 8 were male (44.4%). Three patients (16.7%) were additionally diagnosed with an anxiety disorder. A significant difference was observed in the HAMD scores before versus after rTMS (p<0.001). Moreover, the peak metabolite ratios of NAA/Cr, GSH/Cr and Gln/Cr after rTMS were significantly higher compared with those prior to rTMS (p=0.016, p=0.040 and p=0.008, respectively) (Table 1).

Table 1.

Comparison of metabolite ratios prior to and after rTMS

	Prior to TMS	After TMS	p value
	N=18	N=18
NAA/Cr	2.2±0.25	2.31±0.27	0.016
Cho/Cr	0.85±0.13	0.81±0.1	0.399
mIns/Cr	1.33±0.37	1.41±0.41	0.503
Glu/Cr	1.89±0.74	2.03±0.49	0.467
GSH/Cr	1.17±0.31	1.4±0.37	0.040
Lac/Cr	0.2 (0–1.49)	0.53 (0.05–5.46)	0.112
Gln/Cr	0.17 (0–4.17)	0.62 (0.14–9.36)	0.008
HAMD	23.06±5.97	11.5±2.5	<0.001

rTMS: repetetive transcranial magnetic stimulation, N: number of samples, NAA: N-asetyl aspartate, Cho: choline, Cr: creatine, Lac: lactate, mIns: myoinositol, Glu: glutamate, GSH: glutathione, Gln: glutamine, HAMD: Hamilton Depression Scale

The increases in the NAA/Cr, GSH/Cr and Gln/Cr ratios were not correlated with the change in the HAMD score after rTMS (p=0.176, p=0.998, and p=0.649, respectively).

DISCUSSION

The main goal of our study was to assess the effect of rTMS on brain chemical metabolites in patients with MDD. Despite the increasing number of studies in the past decade that have investigated the efficacy and action mechanism of rTMS, the antidepressant effect of rTMS is poorly understood.

The most conspicuous findings of our study were the significant increases in peak NAA/Cr, GSH/Cr, and Gln/Cr metabolite ratios and the decrease in the HAMD score after rTMS. The DLPFC is the major region of the prefrontal cortex thought to play an important role in the pathogenesis of MDD and contribute to mood regulation and executive functions [22]. The standard target region for rTMS is the left DLPFC. It has been emphasized in the literature that the same brain locations should be selected for 1H-MRS to assess the metabolite changes in response to rTMS [20]. Therefore, we measured the peak metabolite ratios only in the left DLPFC by 1H-MRS. MRS studies on depression have revealed metabolite changes in several brain regions, such as the prefrontal cortex [23], anterior cingulate cortex [24], occipital cortex [25], cerebellum [24], right caudate nucleus and putamen [26] and hippocampus [27].

NAA exists in neurons but not astrocytes and is responsible for neuronal integrity [28]. Wang et al. [23] reported lower NAA/Cr ratios in the right and left DLPFC of patients with moderate and severe MDD, compared with healthy subjects, in a MRS study. Another study also reported a lower NAA/Cr ratio in the prefrontal cortex of patients with MDD than in healthy controls [29]. The decline in the NAA/Cr ratio in the prefrontal cortex has been attributed to reduced neuronal density or neuronal dysfunction. Several studies have revealed that the decrease in the NAA level might play a role in the etiology of psychiatric diseases. Furthermore, some researchers have focused on the therapeutic efficacy of NAA. A study investigating the outcomes of lithium treatment revealed that lithium administered for 4 weeks increases NAA concentrations in all brain regions, including the frontal, temporal, parietal and occipital lobes. This result was attributed to the neuroprotective effect of lithium [30]. In contrast, after pharmacological treatment, increases in hippocampal NAA and Cho metabolite levels [12], the basal ganglia Cho/Cr ratio [31], levels of NAA, Glu, Gln and mIns in the anterior cingulate cortex [32], the NAA/Cr ratio in the medial prefrontal cortex [33] and the Cho/Cr ratio in the ventral prefrontal cortex [34] were also recorded. These metabolite changes were accepted as indicators of the neuronal regenerative effect due to antidepressant medication [35].

Glu is the leading excitatory neurotransmitter in the brain, and its homeostasis is modulated by the Glu/Gln cycle between neurons and astrocytes. Evidence shows that hypoactivity and lower levels of Glu/Gln in the cortex of patients with MDD are closely associated with a lower glutamatergic signaling frequency [36]. One study revealed that direct infusion of exogenous Gln into the medial prefrontal cortex results in an improvement in depressive-like behaviors in mice [37]. Son et al. [36] suggested that depressive behaviors due to chronic stress in mice are associated with hypoactive neurons in the medial prefrontal cortex, and that exogenous Gln can be used to augment glutamatergic neurotransmission as an alternative antidepressant. Michael et al. [38] reported lower Glu/Gln levels prior to unipolar electroconvulsive therapy in the left DLPFC of patients with severe MDD compared with the control group and showed normalized levels after electroconvulsive therapy without any significant difference between patients and controls.

GSH is an antioxidant that protects cells and their contents against oxidative stress to maintain tissue functions. Therefore, it is a sensitive endogenous determining factor of oxidative stress [39]. Wide-ranging clinical evidence reveals that the inflammatory response and oxidative damage are intensified in patients with MDD [40,41]. Furthermore, some studies have reported that long-term antidepressant therapy (12–14 weeks) reverses the increased oxidative stress in patients with MDD [40]. Nevertheless, it remains unclear whether the decrease in GSH is an etiological factor in MDD or an outcome of the disease itself. However, the therapeutic usage of agents such as acetylcysteine for adult neuropsychiatric disorders is still a subject under investigation [42,43].

In our study, the increases in the peak metabolite NAA/Cr, GSH/Cr and Gln/Cr ratios after rTMS treatment suggest that the therapeutic effect of TMS result from these metabolites, which are thought to play a role in the pathogenesis of depression. However, more evidence is needed to conclude that TMS has a regenerative effect on neuronal damage via these metabolites. Additionally, it remains unclear how TMS leads to increased levels of the metabolites.

One of the limitations of our study was the small number of patients. Further studies involving a larger sample size will contribute to a better understanding of how TMS works in depression treatment. Selecting patients who have not taken medications and minimizing variables that may affect the results by using broad exclusion criteria will render our findings more valuable. Three MDD patients included in the study had an additional diagnosis of anxiety disorder, Although this partially reduced the homogeneity of patients’ diagnostic profiles, depression is often accompanied by an anxiety disorder.

Another limitation was the lack of attention to the long-term effects of TMS and the lack of a placebo control group, which would have improved our ability to distinguish the effects of TMS from those of other interventions.

In conclusion, studies investigating the antidepressant mechanism of TMS are of great importance in two respects. First, they increase evidence on the abnormal circuits that are affected, and second, the findings will help determine whether individualizing depression treatment by determining relevant indicators predicting the treatment response will lead to better clinical outcomes. Understanding the mechanism of action of TMS will improve its application and perhaps stimulate the development of a new generation of therapeutic agents in the future.