Owing to the aging of the population, the number of patients with cognitive impairment is increasing year by year, and it is predicted that 81.1 million people will be affected by 2040. The trend increases the social and economic burden, which has become the public challenge of the times (1). Among the various causes of cognitive impairment, cerebrovascular disease stands out as the second, particularly in East Asia (2). Vascular cognitive impairment (VCI) refers to cognitive decline due to brain vessel damage, and it ranges from mild vascular cognitive impairment to vascular dementia (VaD) (3). Its symptoms mainly include four cognitive areas, among which executive dysfunction is its core symptom. Studies have also demonstrated that prefrontal subcortical function impairment is a prevalent neurophysiology of VCI, which significantly impairs executive function and negatively impacts the patient’s quality of life (2, 4, 5). At present, VCI is mainly treated by prevention and drugs. However, simple drug treatment is usually not effective and there is a lack of specific treatment. More and more scholars have begun to pay attention to its non-drug treatment, among which the commonest technologies for cognitive enhancement of stroke patients include computer cognitive training, virtual reality, non-invasive brain stimulation, and brain-computer interface (6).

Transcranial magnetic stimulation (TMS) is a non-invasive regulatory technique. Its mechanism entails the application of pulsed magnetic fields generated by a coil to selectively enhance or suppress the excitability of the cerebral cortex. This alters neuronal activity and modulates the functional connectivity within the brain network, thereby influencing cognitive and executive functions (7, 8). Most cognitive studies have used the dorsolateral prefrontal cortex (DLPFC) as a stimulating region, which is mainly related to executive and attention functions. In recent years, the therapeutic effects of rTMS on executive function have been investigated, and the possible mechanisms are related to increasing synaptic connections, promoting neuroplasticity, accelerating cerebral blood flow, facilitating neuronal cell growth, and improving local brain metabolism (9). Several studies have shown the potential of TMS in the recovery of executive function of diverse neuropsychiatric diseases. Moser et al. (10) found significant improvements in TMT-B tests in middle-aged and elderly patients with refractory depression after rTMS stimulation; Zheng et al. (11) found that improved executive functioning was associated with biochemical changes in cingulate neurochemistry after high-frequency rTMS in young depressed patients; Cricstancho et al. (12) obtained similarly beneficial results after stimulation of iTBS in depressed patients. Herremans et al. (13) treated alcohol-dependent patients with high-frequency rTMS and found that it stabilized cognitive performance in an executive control task; Ameis et al. (14) found improved executive functioning after rTMS in participants with autism who had more severe deficits in adaptive functioning. In particular, since Rektorova et al. (15) revealed the beneficial effects of high-frequency rTMS on executive function in a blinded crossover trial involving seven patients with post-stroke executive dysfunction, more researchers have begun to focus on the effects of rTMS on executive function with VCI. However, it is worth noting that existing these trials are generally constrained by limited sample sizes, geographical variations, and other factors, leading to non-uniform quality levels. Consequently, further research is imperative to validate these findings. According to the search, existing meta-analyses (16–20) have focused on rTMS for the treatment of cognitive impairment in stroke, with no meta-analysis of rTMS for vascular executive dysfunction now. Therefore, this paper aims to (1) demonstrate the effectiveness of rTMS for the executive function of patients with VCI and (2) further subgroup analyses to assess the effects of multiple factors on the treatment.

Executive function is a high-level cognitive function with core components of abstract thinking, working memory, stereotype transfer, and reflective inhibition (43, 44). Executive dysfunction is the core symptom of VCI, which seriously affects the prognosis of stroke and poses a significant clinical challenge. Scholars are increasingly focused on this issue, highlighting the critical need to find an effective remedy (45). Studies have found that rTMS shows great potential in the treatment of executive function in VCI. The pulsed magnetic field of rTMS influences brain tissues (46), affecting synaptic structure (47–49), neurotransmitters (50, 51), brain-derived trophic factors (50, 52, 53), regulating neuronal excitatory gene expression (52, 54, 55), and altering neuronal membrane potentials (56, 57). The neurobiological mechanisms underlying the cognitive-enhancing effects of rTMS are not fully understood. The primary mechanism involves cortical plasticity, specifically the change of long-term potentiation (LTP) and long-term depression (LTD) (58–60). Stroke recovery relies on two competing neuroplastic remodeling mechanisms: interhemispheric competition and compensation. rTMS can enhance excitability in the affected hemisphere or modulate activity in the unaffected hemisphere (61). Animal studies employing rTMS for VCl have revealed changes in hippocampal synaptic morphologies, increased LTD induction, and heightened synaptic plasticity (52, 56, 62). The second manifestation involves an impact on the cholinergic system (63, 64). Acetylcholine (Ach) serves as an excitatory neurotransmitter crucial for learning and memory functions. Zhang et al. (50) applied rTMS (0.5 Hz, 1.33 T) to rats with vascular dementia (VD) over 30 days. They noted a substantial rise in the activities of acetylcholinesterase (AchE) and choline acetyltransferase (ChAT), an augmentation in neuronal cholinergic density, and a notable amelioration of learning and memory impairments in VD rats. Furthermore, apoptosis stands as the predominant cause of neuronal demise, notably in ischemic regions. Animal studies have demonstrated that rTMS exhibits anti-apoptotic and neuroprotective properties. By stimulating the PI3K/Akt/mTOR pathway, rTMS hinders apoptosis in hippocampal neurons (53). Guo et al. (65) identified significantly elevated levels of BDNF, TrkB, and the anti-apoptotic protein Bcl-2 in the ischemic hippocampus. In addition, astrocytes possess the capacity to shield brain tissue and mitigate disability stemming from ischemic brain disorders. rTMS (10 Hz) can enhance cognitive function by mitigating white matter lesions, reducing pro-inflammatory cytokine levels, elevating anti-inflammatory cytokine levels, and inducing the transformation of microglial cells into the M2 phenotype (66). These process influences the activity and functional connectivity of specific brain regions, potentially serving as one of the mechanisms to regulate functional brain activity and enhance executive function.

At present, rTMS is still being explored for its therapeutic applicability in the executive function of VCI. While numerous RCTs have been conducted, a comprehensive meta-analysis on this topic is yet to be established. Based on this, this study aims to fill this research gap by incorporating a total of 20 RCTs that investigate the efficacy of rTMS in improving executive function following VCI. The conclusions of the meta-analysis and systematic review are as follows (1) rTMS can improve the executive function of VCI; (2) Subgroup analyses showed that higher frequency, lower intensity, iTBS, combined comprehensive therapy, and longer interventions yielded better gains. The meta-results are consistent with the meta-analysis results of Gao et al. (18) and Han et al. (20). However, unlike these studies, they only incorporated two articles as part of their cognitive outcome analysis without conducting a comprehensive analysis. This paper addresses this limitation by providing more extensive evidence with 20 RCTs and full indicators. Alongside the inclusion of the 20 RCTs, it is noteworthy to mention that Rectorova et al. (15) conducted a well-designed crossover, sham-controlled trial in 2005. This study focused on the application of rTMS for improving executive impairment following stroke (referred to as PSEI). The results indicated that stimulating the left DLPFC led to notable improvements in patients’ problem-solving ability, as assessed by the TMT. However, the efficacy in terms of memory function and verbal fluency was found to be insignificant. In 2008, Sedlackva et al. (67) expanded their study to include the MCl-V of seven patients as a before-and-after control trial, confirming that rTMS had an equivalent effect on this subgroup. Additionally, Aref (68) conducted a study involving 40 patients to investigate the efficacy of high-frequency rTMS in treating cerebral small-vessel disease. The results from their RCT indicated that the FAB scores of the true-stimulation group were significantly higher than those of the sham-stimulation group. However, the findings were not incorporated into the present meta-analysis due to the lack of post-treatment data. Furthermore, Yuanwen et al. (38) conducted another RCT, concluding that rTMS treatment for PSEI led to improved WCST scores. Nonetheless, this study was excluded from the meta-analysis due to inconsistencies in comparison to other RCTs included in this meta-analysis. With the help of functional near-infrared spectroscopy (fNIRS), a recent study by Liu et al. (69) observed individuals with post-stroke executive impairment following rTMS activation of the left DLPFC. The researchers discovered that the time and number of error scores of the Stroop test were significantly higher in these patients, which correlated with the enhanced functional connectivity between various brain regions. Notably, this activation involved the left DLPFC, the right PMC, and the right SM1 brain regions. These trials all further support the conclusions of this article.

The potential of rTMS in treating executive function after VCI is immense, and as the evidence for its effectiveness grows, the need for further optimization becomes increasingly crucial. Currently, there is no uniform clinical standard for the treatment mode. The most common stimulation site for rTMS is DLPFC, with an intensity of 80% ~ 120% motor threshold (MT), a frequency of 5 ~ 25 Hz, and time intervals of 20 ~ 30 s. The treatment duration typically extends 3 weeks or longer (70). Numerous factors, including stimulation site, combined modalities, intervention timing, and stimulation parameters, influence the therapeutic outcomes of rTMS. Hence, this paper focuses on subgroup analysis to deepen the primary findings. In terms of stimulation sites, it is known that functional changes in the DLPFC can lead to cognitive and executive deficits (71). Since the patient is right-handed, the current RCT primarily targets the left DLPFC, with only one article in this paper examining the healthy DLPFC, which reduces heterogeneity. In terms of the combination modality of rTMS, subgroup analyses showed that rTMS combined with comprehensive therapy was superior to combined with monotherapy, suggesting that rTMS can add to cognitive routine treatment. Previous studies (72, 73) have also shown that rTMS in combination with other therapies for treating executive dysfunction is a significant trend for the future. In terms of intervention time, the results for both primary indicators 1 and 2 showed better improvement in executive functioning of ≥4 weeks, probably due to the cumulative effect of stimuli. Regarding stimulation type, iTBS is a unique type of rTMS characterized by a 50 Hz pulse with 3 pulses administered every 200 ms (7). Previous studies have confirmed that iTBS has superior effects on cognitive function than conventional rTMS (60, 61), which is consistent with the results of the subgroup analyses in this paper. Concerning stimulation frequency, current clinical research results mainly focus on high-frequency rTMS (10 Hz) to treat the executive function of VCI. 5 Hz high frequency or 1 Hz low frequency may not be effective, but numerous studies have found that low-frequency rTMS can improve cognitive function and enhance neuroplasticity (45). This paper has fewer articles on lower frequencies (one at 5 Hz, one at 1 Hz). Due to the limitations of sample size, their effectiveness needs further verification through original experiments. Regarding stimulation intensity, subgroup analyses showed that lower intensities (80%MT) may be more effective. Zhang et al. (74) explored prefrontal optimal intensity with the fNLRS and found that intensity was nonlinear with prefrontal cortical blood flow and that a 70% MT effect may be better than a 100% effect. However, the interactions between the parameters are not yet clear and need to be further explored in the future.

The therapeutic efficacy of rTMS is influenced by a variety of factors, such as age, sex, handedness, infarct area and size, cardiovascular risk factors, and education level (73). Age is an independent risk factor for executive dysfunction. The results of the study by Yaning et al. (75) showed that, in comparison to younger age groups, showed that, in comparison to younger age groups, older adults had poorer executive scores. Additionally, Mally et al. (76) reported executive function declines and rTMS therapy effects wear off faster in individuals 65 and up. Gender-specific prognoses have been observed in VCI (77). Cantone et al. (78) found that men with mild VCI had worse cognitive and functional status compared with women. This study is the first to emphasize gender-specific alterations in intracortical and corticospinal excitability following rTMS. Executive function primarily stimulates left DLPFC, as confirmed by related research. However, 21.4% of right-handed individuals have their DLPFC located mainly in the right hemisphere rather than the presumed left hemisphere, and 16.7% of left-handed individuals have their DLPFC located mainly in the left hemisphere (79). This shows that a scanner for localizing the potential DLPFC for precise stimulation also significantly impacts prognosis. Brain regions associated with executive regions include the frontal-striatal loops and the cerebellum, and injuries to these regions result in more severe impairments in executive function, which can affect the therapeutic efficacy of rTMS (80). There are no original studies on these relevant factors, and further exploration is necessary. Currently, RCT stimulation sites are mostly limited to the DLPFC, but stimulating the M1 area has been demonstrated to effectively enhance cognitive capabilities. Experiments involving rTMS stimulation of the bilateral DLPFC have yielded favorable outcomes (81). Diseases are characterized by damage to multiple networks and stimulation of other regions may also cause cognitive enhancement (82). The exact mechanism remains uncertain. It is proposed that targeting different stimulation sites such as the precuneus, parietal region, or cerebellum could improve executive dysfunction in the future.

In conclusion, this meta included 20 RCTs of executive function in patients with VCI treated by rTMS. It not only provides comprehensive evidence of its efficacy but also has some limitations: (1) the RCT has a range of subjective scales and no objective indications, which increases the heterogeneity of indicators; (2) there is less literature in foreign languages (7 foreign papers of 20 papers in this meta-analysis), which is probably related to the absence of rTMS in the latest guidelines for post-stroke cognitive dysfunction (83). So there is a greater need for further studies to promote the application of the study.

rTMS is effective in the treatment of the executive function of VCI. Higher frequency, lower intensity, iTBS, combined comprehensive therapies, and longer interventions are more effective. There are some limitations in this study, and further exploration requires more objective metrics and optimal parameters to enhance its application.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

XW: Data curation, Formal analysis, Methodology, Software, Writing – original draft. QD: Conceptualization, Software, Writing – review & editing. YuL: Resources, Supervision, Writing – review & editing. TL: Resources, Supervision, Writing – review & editing. YaL: Data curation, Investigation, Writing – review & editing. JY: Investigation, Project administration, Writing – review & editing. WZ: Funding acquisition, Project administration, Writing – review & editing.