Introduction

Depression is a leading cause of the global burden of disease and a major contributor to disability worldwide [1]. Approximately 4.7% of the world’s population experiences depression during a 12-month period [2]. Moreover, major depressive disorder (MDD) is one of the primary global causes of premature death [3]. Patients with MDD are more likely (approximately 20 times more) to die by suicide than those in the general population [4]. Although medication has demonstrated greater effectiveness in treating MDD than placebo [5], it is associated with side effects [6]. Further, only one-third of the patients with MDD achieve remission after initial antidepressant therapy [7]. Consequently, a key strategy to manage MDD is developing and testing novel therapies, such as high-frequency repetitive transcranial magnetic stimulation (HF-rTMS) [8], deep TMS (dTMS) [9], transcranial direct current stimulation (tDCS) [10], and transcranial alternating current stimulation (tACS) [11].

Among these interventions, rTMS is a non-invasive neuromodulation technique producing brain activity changes in relation to the applied frequency. In principle, high-frequency stimulation (≥ 5 Hz) increases neuronal excitability and inhibits synaptic transmission [12]. Increasing randomized controlled trials (RCTs) and meta-analyses have shown that active rTMS is more effective than sham stimulation in treating adult MDD [13,14,15] or adolescent first-episode MDD [16, 17]. In 2008, the US Food and Drug Administration (FDA) approved the first rTMS device for MDD treatment, utilizing a figure-of-eight (F8)-coil [18]. However, several studies have reported low rates of rTMS treatment response and remission in MDD [13, 14]. For example, Berlim et al. found that approximately 13 rTMS sessions yielded response and remission rates of 30% and 20%, respectively [13]. Nevertheless, rTMS has been demonstrated to be therapeutically safe [19,20,21]. Therefore, clinically valuable strategies are warranted to enhance rTMS efficacy.

In the case of coil types, the Hesed coil (H1-coil) generates a wider electric field than the F8-coil, thereby addressing concerns over target localization in clinical practice [22]. Moreover, the deep magnetic field created by dTMS has been suggested to enhance white matter recruitment and facilitate propagation to the subcortical areas, potentially improving antidepressant response to TMS [22, 23]. In 2013, the FDA approved a second TMS device, a dTMS instrument employing the H-coil, for MDD treatment [18]. The benefits and good tolerance of dTMS have been indicated in numerous RCTs among patients with depression [9, 24, 25]. Furthermore, a meta-analysis of 10 studies suggested that high-frequency dTMS was effective and acceptable for managing unipolar and treatment-resistant depression [26]. However, studies comparing dTMS and HF-rTMS for depression treatment have found inconsistent results [27, 28]. For example, Filipčić et al. determined that dTMS resulted in superior response rates compared with rTMS [27]. However, another study [28] revealed no significant differences in the response rates between dTMS and HF-rTMS treatments.

Currently, no systematic reviews have been published on the efficacy and safety of HF-rTMS and dTMS in adults with MDD. Therefore, this systematic review aimed to examine the safety and effectiveness of dTMS versus HF-rTMS in adults with MDD. Based on the findings of Filipčić et al. [27], we hypothesized that dTMS would yield a significantly higher response rate than HF-rTMS in adult patients with MDD.

Methods

Literature Review

Two independent investigators (NZ and YM) systematically searched six online databases (Chinese Journal Net, WanFang, PubMed, the Cochrane Library, PsycINFO, and EMBASE) up to July 6, 2023. The search terms used were as follows: (“repetitive transcranial magnetic stimulation” OR rTMS OR TMS OR “transcranial magnetic stimulation”) AND (“deep transcranial magnetic stimulation” OR “deep repetitive transcranial magnetic stimulation” OR deep rTMS OR deep TMS OR dTMS OR H-coil) AND (depress* OR dysphor* OR dysthymi* OR melanchol* OR antidepress* OR bipolar OR MDD). Additionally, hand-searching of the references from the included studies [27, 28] and the relevant reviews [22, 29]was conducted independently by two investigators (NZ and YM) to identify any additional studies .

Selection Criteria

According to the Preferred Reporting Items for Systematic Reviews and Meta Analyzes guidelines [30], the following PICOS criteria were considered when selecting the relevant studies. Participants: based on the respective studies, adults (≥ 18 years) diagnosed with MDD. Intervention versus Comparison: dTMS with the H-coil plus antidepressants versus HF-rTMS with the F8-coil plus antidepressants. Outcomes: remission (e.g., the Montgomery–Asberg Depression Rating Scale [MADRS] [31] scores ≤ 10 or Hamilton Depression Scale [HAMD] [32] scores ≤ 7) and antidepressant response (e.g., 50% reduction from the baseline HAMD or MADRS scores) rates as measured by the corresponding depression scales were the primary outcomes of the analysis. Secondary outcomes comprised depressive symptom changes as evidenced by the depression scales (e.g., HAMD or MADRS), discontinuation rates, and adverse events. Study: the inclusion of the studies was limited to published RCTs on the efficacy and safety of dTMS with the H-coil versus HF-rTMS with the F8-coil for patients with MDD. No review articles, retrospective studies, or case reports/series were included in this systematic review.

Data Extraction

Using a predetermined checklist, two investigators (NZ and YM) independently extracted the data. Specifically, data concerning the characteristics of each included study (e.g.,sex and age), stimulation parameters (e.g., intensity, duration, and train time per stimulation session), and treatment details (e.g., total pulses, total sessions, and total pulses per session) were collected. In the case of discrepancies, the two investigators attempted to reach a consensus, along with the assistance of a senior researcher (WZ), when necessary. Further, the first and/or corresponding authors of the respective studies were contacted when clarification for unclear or missing information was required.

Assessment of Study Quality

The Cochrane risk of bias tool [33] and the Jadad scale [34] were independently applied by the two investigators (NZ and YM) to assess study quality. Studies scoring 3 points on the Jadad scale were considered high quality.

Results

Database Search Results

Using the earlier mentioned databases, we initially identified 1,025 articles (Fig. 1). Ultimately, two RCTs [27, 28] that met the inclusion criteria were included.

Fig. 1
figure 1

Cochrane risk of bias tool items

Abbreviations: PRISMA=Preferred Reporting Items for Systematic reviews and Meta-Analyses; RCT=randomized clinical trial.

Full size image

Characteristics of Study Samples

The patient characteristics and dTMS/HF-rTMS parameters of each included RCT are summarized in Table 1. A total of 203 patients with MDD were enrolled in the two RCTs that compared dTMS (n = 100) with HF-rTMS (n = 103). Among them, 43.3% of the patients were male, with a median age of 23–51 years. All patients received treatment at the left dorsolateral prefrontal cortex (L-DLPFC) for 2 or 4 weeks. In the included RCTs, dTMS (18 Hz) was administered at an average dose of 1,980 pulses per session, whereas HF-rTMS (10 Hz) was provided at 1,400–3,000 pulses per session. Furthermore, the total dose of dTMS varied from 19,800 to 39,600 pulses, while that of HF-rTMS was 14,000–60,000 pulses.

Table 1 Patient characteristics and dTMS/HF-rTMS parameters in the included studies
Full size table

Quality Assessment

As shown in Fig. 2, the two included RCTs [27, 28] were judged to be “low risk” in terms of random sequence generation, blinding of outcome assessment, incomplete outcome data addressed, and selective reporting according to the Cochrane risk of bias tool. Additionally, the two RCTs had Jadad scores of 3 [27] and 4 [28], indicating their high quality.

Fig. 2
figure 2

Cochrane risk of bias tool items. +: Low risk of bias; -: High risk of bias; ?: Unclear risk of bias.

Full size image

Study-Defined Response and Remission

The antidepressant response and remission rates of dTMS versus HF-rTMS as adjunctive therapy for MDD are listed in Table 2. In the two RCTs [27, 28] the overall antidepressant response rate was significantly higher in the dTMS (60.0%) than in the HF-rTMS group (41.7%). However, only one RCT [27] reported the antidepressant remission rates, in which no significant difference was observed between the two TMS groups (dTMS group: 59.7% versus HF-rTMS group: 42.7%; P > 0.05).

Table 2 dTMS versus HF-rTMS for patients with MDD: study-defined response and change in depressive symptoms
Full size table

Improvement in Depressive Symptoms

The two RCTs [27, 28] revealed alleviation of depressive symptoms as measured by HAMD-17, although these findings were inconsistent (Table 2). Moreover, dTMS was found to be significantly superior to HF-rTMS in ameliorating depressive symptoms in the study by Filipčić et al. [27]; however, this observation was not corroborated by Yang et al. [28].

Adverse Events and Discontinuation Rates

Filipčić et al. [27] showed that compared to HF-rTMS (0%), dTMS (12%) caused more muscle twitching/spasms or jaw pain incidences (Table 3). Other adverse events (e.g., headaches and dizziness) and discontinuation rates were not significantly different between the two TMS groups (dTMS group: 12% versus HF-rTMS group: 5%) in both RCTs (Table 3).

Table 3 dTMS versus HF-rTMS for patients with MDD: rates of discontinuation and adverse effects
Full size table

Discussion

Our systematic review, encompassing two RCTs [27, 28] involving 203 adults with MDD, is the first to examine the efficacy and safety of dTMS versus HF-rTMS for treating MDD. The main findings of this systematic review are as follows: (1) dTMS provides a more pronounced overall antidepressant response than HF-rTMS; (2) dTMS causes more muscle twitching/spasms or jaw pain incidences than HF-rTMS; and (3) both RCTs have similar rates of other adverse events and discontinuation. Moreover, the two RCTs were published within the last 5 years, implying growing clinical interest in applying HF-rTMS and dTMS for patients with MDD.

In our systematic review, the dTMS group exhibited a significantly higher overall antidepressant response rate than the HF-rTMS group. However, these results were inconsistent between the two included RCTs, possibly due to the differences in the treatment courses. According to previous meta-analyses [35, 36], depression severity demonstrated greater reduction after 20 sessions than after 10 sessions of dTMS or HF-rTMS. In the included RCT by Filipčić et al., dTMS resulted in a significantly greater response rate than HF-rTMS after 20 sessions [27]. In contrast, Yang et al. found that the response rates were not statistically different after 10 sessions of dTMS versus HF-rTMS [28]. Therefore, at least 20 sessions may be required to obtain clinically meaningful effects in patients with acute MDD, regardless of rTMS or dTMS techniques [37]. Apart from the antidepressant effects of dTMS, dTMS using different H-coils has also been applied for treating obsessive-compulsive disorder (OCD) [38], smoking addiction [39], and schizophrenia [40]. For example, the Yale–Brown Obsessive-Compulsive Scale scores of patients with OCD who were treated with active dTMS were shown to be significantly lower than those of sham-treated patients [38].

The precise mechanism underlying the alleviation of depressive symptoms by dTMS in patients with MDD remains uncertain. Nevertheless, accumulating studies have suggested that individuals with MDD exhibit diminished activity in the L-DLPFC when experiencing negative emotions [41, 42]. Thus, the DLPFC, a component of the cognitive control network (CCN), represents a critical target for TMS therapy [43, 44]. In this strategy, dTMS can be used to stimulate the left DLPFC to directly affect the cognitive processes regulated by the CCN, subsequently modulating the cognitive and affective functions [44, 45]. Consequently, applying dTMS to specifically target the left DLPFC may serve as an effective intervention for MDD. Additionally, the H1-coil provides a greater penetration depth in specific brain structures than the F8-coil [46], suggesting a potential link between dTMS efficacy and the ability of the H1-coil to directly stimulate wider and deeper PFC structures [46]. However, the effect of distinct stimulation parameter settings and pulse counts on treatment efficacy remains unclear in both TMS modalities. Hence, further investigation on the varied treatment parameters of dTMS and HF-rTMS is essential.

Although the incidence of muscle twitches/spasms or jaw pain was greater in dTMS than in HF-rTMS, the rates of discontinuation and adverse events were similar across both techniques. Moreover, dTMS has been proven safe among patients with OCD [38], severe and enduring anorexia nervosa (SE-AN) [47], bipolar depression (BD) [48], Parkinson’s disease (PD) [49], and obesity [50]. For example, Knyahnytska et al. demonstrated that dTMS was low-risk and well-accepted in patients with SE-AN [47]. Similarly, dTMS was shown to be a well-received add-on therapy for patients with BD undergoing appropriate pharmacotherapy [48]. All these findings imply that using dTMS or HF-rTMS in clinical practice may be a generally safe and well-tolerated treatment strategy [51].

This systematic review has several limitations that should be considered. First, we were only able to extract data from two existing RCTs. Consequently, the study should be expanded with more RCTs in the future. Second, a meta-analysis could not be conducted due to significant heterogeneity among the included RCTs. Third, other unpublished studies with smaller (non-significant) effect sizes may be present because we did not incorporate unpublished data in this systematic review, leading to the possibility of publication bias. Finally, our systematic review was unregistered.

Conclusion

dTMS is associated with a better antidepressant response than HF-rTMS in adult patients with MDD, although both treatment modalities have favorable safety profiles. However, additional RCTs employing rigorous methodologies are warranted.