The prevalence of suicide among veterans who served in the US military is devastatingly high. Veterans die by suicide at almost double the rate of civilians in the United States [1]. The prevalence of suicide among veterans with traumatic brain injury (TBI) is even higher. Veterans with TBI are 1.5 times more likely to die by suicide than veterans without TBI, even when controlling for comorbid psychiatric conditions [1,2]. These statistics are particularly concerning given that there have been >400,000 new TBI diagnoses among US service members since the beginning of operations Enduring Freedom, Iraqi Freedom, and New Dawn, of which >80% are mild traumatic brain injury (mTBI) [3]. Overall, between 16% and 20% of all operations Enduring Freedom, Iraqi Freedom, and New Dawn veterans have a history of mTBI, and among them, up to 22% report suicidal ideation (SI) [4].

TBI consists of any external force applied to the head of any individual, resulting in vascular and axonal damage, edema, and neuronal death [5]. TBI is clinically further subcategorized by the severity of the injury as either mild, moderate, or severe. mTBI is defined as a loss of consciousness for <30 minutes, postconcussive altered mental status or amnesia for <24 hours, a Glasgow Coma Score, following injury of 13 to 15 (15 being the maximum score), and lack of abnormalities on standard clinical magnetic resonance imaging (MRI) or computed tomography [6]. TBI is an extremely common injury, resulting in an estimated 4.8 million independent emergency room visits throughout the United States annually [7-9]. TBI results in a significant economic burden: the cost of nonfatal TBI amounts to >US $40 billion per year in the United States [10]. Globally, TBI resulted in the loss of 246.7 million disability-adjusted life years between 1990 and 2013 [11].

Given the high suicide rate among US veterans, suicide prevention is a high-priority issue for the US Veterans Administration (VA) [12]. The US VA has assembled task forces that have created clinical practice guidelines for the assessment and management of veterans at risk for suicide [13]. These guidelines establish substantial risk factors to help providers identify which veterans are at the highest risk. The guidelines specifically cite having a history of TBI, functional deficits, past SI, and a history of impulsivity as independent risk factors for developing SI [13]. TBI predisposes individuals to functional deficits [14,15] and impulsivity [16], and those with impulsivity likely struggle with functional deficits [17]. Given these findings, we hypothesize that these conditions may be interrelated and could serve as meaningful behavioral targets in the treatment of SI.

Transcranial magnetic stimulation (TMS) therapy was first described by Barker et al [18]. Seminal studies by George et al [19] demonstrated the efficacy of TMS in the treatment of depression in large, multisite sham-controlled trials. TMS was subsequently approved by the US Food and Drug Administration (FDA) in 2008 for the treatment of major depressive disorder [20]. TMS was first hypothesized as a treatment for TBI by Pape et al [21]. It was later shown to be effective in treating several TBI sequelae, including depression [22]. TMS also has demonstrated efficacy in the treatment of posttraumatic stress disorder (PTSD), chronic pain, cognitive deficits, and headaches in individuals with mTBI [23-27]. A second-generation form of TMS, intermittent theta burst stimulation (iTBS), was first shown to treat depression in 2009 in a study by Holzer and Padberg [28]. In a large, noninferiority study comparing iTBS and TMS, patients with treatment-resistant depression show similar safety and adverse event (AE) profiles and exhibit improvements in their depressive symptoms to both interventions [29].

This study aims to create a novel iTBS intervention to reduce impulsivity and SI among veterans with mTBI while improving overall social and occupational functioning. This study will be conducted at the Edward Hines, Jr. VA Hospital in Hines, Illinois. We are interested in stimulating the ventromedial prefrontal cortex (VMPFC) with iTBS because the VMPFC appears to exhibit less functional connectivity with the limbic system in individuals with mTBI [30]. This diminished connectivity is suspected to be responsible for impulsivity and SI in individuals with mTBI [30-32]. To the best of our knowledge, no established treatment approach has used VMPFC iTBS to improve social and occupational functioning, SI, and negative urgency impulsivity among individuals with mTBI and comorbid mental illness. A list of the common terms used in the study and their definitions are provided in Textbox 1.

This pilot study aimed to (1) determine the safety, feasibility, tolerability, and efficacy of frontal pole iTBS for veterans with mTBI, negative urgency impulsivity, and SI; (2) collect advanced neuroimaging data from veterans with mTBI, impulsivity, and suicidality to further our understanding of what neural changes are occurring when treatment is effective; and (3) gather preliminary clinical outcome data on impulsivity, SI, and social and occupational functioning of veterans receiving iTBS. The results of these objectives will help to guide future studies on neuromodulation as a treatment for impulsivity and SI among patients with mTBI.

Our central hypotheses are that frontal pole iTBS will be safe, feasible, and tolerable among veterans with mTBI based on existing literature that suggests frontal pole iTBS is as safe and tolerable as iTBS performed elsewhere [23,25,62]. Similarly, iTBS, when administered to individuals with mTBI is as safe and tolerable as in individuals without mTBI [63]. Given that we are using an established iTBS protocol within the established TMS safety guidelines [84], we do not anticipate significant AEs.

We further hypothesize that social and occupational functioning, negative urgency impulsivity, and SI will be improved for individuals who receive active-iTBS compared with those who receive sham-iTBS. Even if only a single domain improves, we consider this a positive finding.

In addition, we hypothesize that resting-state functional connectivity between the VMPFC and the limbic system (especially the amygdala and anterior cingulate cortex) will be strengthened for the active-iTBS group compared with the sham-iTBS group. We also expect that veterans with increased connectivity between these regions of interest (ROIs) will likely show the most functional improvements.

Pending the results of this pilot trial, larger-scale RCTs may be warranted to establish the efficacy of frontal pole iTBS for social and occupational functional deficits, negative urgency impulsivity, and SI.

The proposed study is a pilot, prospective, randomized, double-blinded, sham-controlled study to develop a frontal pole iTBS intervention. We will examine the safety, feasibility, tolerability, and efficacy of 10 sessions of iTBS over 2 weeks. Preliminary impulsivity, SI, and social and occupational functioning data will be collected before and after receiving active or sham treatment. In developing this protocol, we referenced the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) guidelines [85]. We used the CONSORT (Consolidated Standards of Reporting Trials) checklist when writing our report [86]. Refer to Multimedia Appendix 1 for the CONSORT flow diagram.

The study protocol was reviewed and approved by the Edward Hines, Jr. VA Hospital Institutional Review Board (Hines IRB number 1716074) and the University of Illinois, Chicago Institutional Review Board. The study is registered on ClinicalTrials.gov (NCT05647044). This study is designed in accordance with the ethical standards of the US VA Office of Research and Development, based on those from the Declaration of Helsinki, the Belmont Report, US Common Rule and Regulations from the Council for International Organizations of Medical Sciences. The study will be overseen by the institutional review boards of the both Edward Hines, Jr. VA Hospital and the University of Illinois, Chicago. This research study is designed on a 5-year research plan.

A Health Insurance Portability and Accountability Act (HIPAA) authorization waiver will be obtained to enable electronic medical records review of potential participants before obtaining consent. The HIPAA waiver allows research staff to access information about the patient’s medical history and demographics, as well as history of drug or alcohol abuse to determine study eligibility. If a potential participant declines to participate in the study, no further contact will be made.

If the potential participant consents to the study, they will then begin the informed consent process. Informed consent will be obtained in English by trained research staff. Research participants will have access to research staff and the principal investigator to assist with any questions or concerns until understanding is achieved. A consent note is made in the participant’s medical record either electronically or on paper. The original consent is then placed in the medical record, a copy is given to the participant, a copy is filed with the Edward Hines, Jr. VA Hospital Institutional Review Board and the original informed consent is placed in the participant’s research binder and maintained in a locked file cabinet at Hines VA Hospital. Once the informed consent process is completed, a note will be documented in the electronic medical record system and the participant will be flagged as a research participant.

Any identifiable information about the participant, including their signed consent form, is stored in a different location than study data, on encrypted VA servers in files that are only accessible to approved study team members. The collected, deidentified research data are also stored on encrypted VA servers elsewhere, in password-protected folders, only accessible to approved study team personnel. Any participant information obtained in the study will be treated as confidential and safeguarded in accordance with the Privacy Act of 1974. Information published or presented about the results of the study will be done such that no individual identifying information is shared.

Participants will be compensated up to US $350 for their participation, prorated for their time.

We will enroll 56 veterans aged between 22 and 65 years with a history of mTBI and impulsivity. They must also report that they have experienced SI within the last 3 months. mTBI will be defined according to the VA and Department of Defense guidelines [87] and use aspects of the mTBI symptom attribution and classification algorithm [88] when determining eligibility. mTBI must have occurred at least 1 year before enrollment to ensure that the injury has stabilized. Though it is not mandatory, veterans may also have diagnoses of depression, PTSD, or bipolar spectrum illness (without a history of psychosis). Recent SI (within the past 3 months) will be defined as a score of ≥1 on the Columbia Suicide Severity Rating Scale [89]. Impulsivity will be defined as having any mentions of impulsive behavior in the medical chart and any positive results on the Urgency, Premeditation (lack of), Perseveration (lack of), Sensation Seeking, Positive Urgency Impulsive Behavior Scale–Negative Urgency subscale (UPPS-P) [41]. For safety, we will exclude persons with contraindications to TMS or MRI because of increased risk of AEs. We will also exclude participants whose SI is active, with intent and plan, as they require emergency psychiatric intervention. We will not exclude individuals based on sex, gender identity, ethnicity, or race. To enhance generalizability, we will include veterans regardless of previous combat or deployment experience. As cannabis and alcohol use is prevalent among veterans, intermittent or low amounts of cannabis or alcohol consumption will not be considered exclusionary if the use does not meet the criteria for a moderate substance use disorder. Detailed information on inclusion and exclusion criteria is listed below (Textbox 2).

To ensure we have adequate power to detect a difference in our primary outcome, the Social and Occupational Functioning Assessment Scale (SOFAS), we conducted a power analysis. Assuming a significance level of 0.05, a mean SOFAS score of 44.6, and a SD of 15, with 50 veterans, we will have 82% power to detect a significant interaction between time and treatment assuming no change over time in the sham-iTBS group and a change of 13 points in the active iTBS group [90]. On the basis of similar studies of iTBS protocols for patients with depression and PTSD, receiving treatment of similar duration, attrition rates are expected to be low—under 10% [58,90,91]. As such, we plan to enroll participants to ensure we have an adequate sample size considering 10% attrition. Thus, our total sample size will be 56 with 28 in the active iTBS group and 28 in the sham iTBS group.

We will recruit veterans receiving care from the TBI and polytrauma program, primary clinic, women’s health clinic, and the Mental Health Service Line at Edward Hines, Jr. VA Hospital in Hines, Illinois, United States. Personnel at all mentioned clinics will be educated on the study and encouraged to give institutional review board (IRB)–approved flyers containing contact information to veterans who may be a good fit for the study. We will be added as cosigners in the electronic medical records of Hines VA patients who may be deemed a good fit for this study. We will also mail informational letters to veterans who have completed past TBI studies and permit us to contact them about future studies through a TBI Data Repository (Hines IRB number14-003), which currently includes >300 veterans. Research candidates at the Hines VA will also be identified using administrative electronic health record data available through the VA Informatics and Computing Infrastructure. The VA Informatics and Computing Infrastructure database will search for TBI within the past 10 years (FY2013-FY2023) according to appropriate International Classification of Diseases (ninth and tenth editions), clinical modification, codes reflecting study inclusion criteria. All potentially eligible veterans will receive an informational letter and phone call to gauge interest and perform initial eligibility screens.

Each participant will be randomized to receive active or sham treatment throughout the protocol, with 50% (28/56) receiving active treatment and 50% (28/56) of participants receiving placebo. We will use a T1 MRI to localize the FP stimulation site. This region is correlated with electrode FP2 on a 10 to 20 electroencephalogram, a region overlapping with the right VMPFC [22,82]. To ensure precise and repeatable coil positioning over the frontal pole FP2 site, the Localite (neuronavigation) system and an MRI instrument marker will be used at every iTBS session.

Our 9-minute iTBS protocol will be delivered using the MagVenture Mag-Pro X100 with the MagOption stimulator, including active and placebo coils (C-B60 Butterfly coils). The consensus in the literature is that iTBS can safely deliver 1800 pulses at 110% of the MT to the frontal pole [82,83]. iTBS parameters include 3 pulses of stimulation at 50 Hz, repeated every 200 miliseconds. The interpulse interval is 20 miliseconds. A 2-second train of TBS is repeated every 10 seconds for a total of 570 seconds, equaling a total of 1800 pulses delivered over 9 minutes. To ensure that the participants can tolerate the treatment, stimulation intensity will begin at 60% to 70% of the MT intensity and will slowly increase as the patient tolerates to 110% of the patient’s MT intensity over the first few treatment sessions [42].

A partial HIPAA waiver and waiver of informed consent for screening purposes will provide regulatory approval to screen potential candidates to determine study eligibility. Once identified as a potential research candidate, candidates will be contacted for an initial phone screening to determine eligibility criteria. Demographics, mental illness diagnosis, current medication use, and history of impulsive behaviors (eg, history of fights, violence, anger, and irritability) will be cross-verified in the patient’s electronic medical record. A thorough chart review will be completed to cross-reference current medications with study eligibility criteria to identify antiepileptics or medications that could lower the seizure threshold. If review findings indicate possible contraindications to TMS or MRI related to a metal implant, then the model and manufacturer of the implant will be obtained to determine whether it is safe to expose the implant to a strong magnet. Manufacturer recommendations regarding safety will be followed. If potential participants meet all eligibility criteria, they will be invited for their first research visit, an in-person screening (visit 1). There will be 14 visits for this research study (Figure 2).

After meeting with the research staff and signing the informed consent, participants will complete several baseline diagnostic and screening measures (Table 1). These measures will be used to track changes in symptoms of impulsivity, suicidality, depression, anger, aggression, and social and occupational functioning throughout the study. If the participant is a woman, sexually active, and not on any form of physician-prescribed birth control, they will be asked to take a urine pregnancy test at this visit. Veterans who have completed baseline measures and have met eligibility criteria will be asked to complete a pretreatment resting-state fMRI at the University of Illinois, Chicago, an institution partnering with us for this study (visit 2). After the pretreatment MRI, participants will be invited back to the Hines VA to complete their MT and brain mapping using the TMS Neural Navigator (Localite) system (visit 3).

Participants will be randomized to either active or sham treatment. Blinding procedures for active and sham iTBS will be delivered with the MagProX100 with MagOption stimulator and Magpro Cool Coil B65 A/P (MagVenture), which can be switched to active or sham. The coil is identical visually for the sham and active conditions. Veterans and researchers will wear headphones connected to the noise generator during treatment to hear active iTBS for both conditions, thereby maintaining the blind. The sham coil and scalp electrodes will be placed in the same location for all participants and will mimic the sensation produced by the active coil but will not alter brain physiology. Thus, the sham stimulation looks, sounds, and feels like active stimulation. A single unblinded study team member will determine which participants are randomly placed in the active or sham groups.

After the MRI and MT have been completed, participants will be randomized, and the intervention will begin (visits 4-13). Intervention sessions will occur once daily for 10 business days, with each session lasting 30 minutes to 1 hour. Research staff will complete a TMS safety rating scale before and after every iTBS session to assess for changes from baseline for vital signs, sleep duration, fatigue level, recent seizures, and tinnitus. Pain and tolerability measures will be completed at the end of every session to assess daily pain and session tolerability. After iTBS session 5 (visit 8), each participant will be asked to complete the self-report rating scales and neurocognitive testing they completed at baseline.

All veterans who have completed the iTBS treatment course of 10 sessions will be asked to return to the University of Illinois, Chicago’s Center for Magnetic Resonance Research to obtain a second resting-state fMRI for their final research visit. Posttreatment MRI data collection is expected to be completed 72 hours after the iTBS course. This time spacing between stimulation and imaging has been done in similar studies [78] and is thought to be a long enough gap post-iTBS completion not to represent the immediate effects of the most recent stimulation.

A statistical significance level corresponding to α=.05 will be used to test hypotheses.

Descriptive statistics of feasibility, safety, and tolerability measures will be computed for all participants and compared between active- and sham-iTBS groups. We will calculate the categorical variables’ frequencies and compare them using chi-square analyses for the descriptive statistics [126]. For the continuous variables (ie, daily pain ratings), we will compute mean, median, range, and SDs and compare using 2-tailed t tests.

For functional outcome measures, we will estimate a mixed model ANOVA to measure differences within-participants (how the SOFAS changes over time), between-participants (how the SOFAS differs between active and sham iTBS), and whether there is a significant interaction between time and treatment [127]. We will assume equal variance between test and control groups [128]. If the variance is unequal, we may transition to a generalized linear mixed (GLM) effects model [129].

For our neuroimaging analysis, to evaluate the VMPFC to the amygdala and anterior cingulate relationships and how they change before and after treatment, we will define ROIs for the amygdala and anterior cingulate with clear anatomical boundaries identified using Freesurfer [130]. Average time-series data will be extracted, and GLM will be used to quantify the relationship between the seeds and targets. The GLM results will yield individual r values (ie, correlations), which will be normalized into Z scores using Fisher R-to-Z transformation [131]. The individual maps of Z scores will be the primary measure of connectivity between the seed and target ROIs. We will then implement an ROI-to-ROI analysis within the connectivity toolbox to evaluate resting-state functional connectivity between the amygdala, anterior cingulate, and VMPFC [132]. We will assume a normal distribution and use Fisher’s Z test to compare active-iTBS and sham-iTBS groups, both before and after treatment [133]. We will adjust for a priori covariates, including age, sex, and baseline psychiatric diagnoses. All analyses will be corrected for multiple comparisons using false-discovery rate correction. We will use linear mixed models to test how connectivity changes correspond to functional and mental health outcome changes, with functional and behavioral measures included as the dependent variables [134].

To inform future research studies, we will estimate regression models that include additional covariates such as demographics, psychotropic medication use, comorbid psychiatric diagnoses, and scores on the Inventory of Depressive Symptomatology, Self-Report and the PTSD Checklist for DSM-5, in addition to testing our hypotheses for the primary and secondary outcomes.

This novel protocol represents the first pilot RCT for frontal pole iTBS for mTBI, negative urgency impulsivity, and suicidality. We hypothesize that iTBS administered to the frontal pole in US veterans with mTBI, impulsivity, and suicidality will be safe and well tolerated as studies show that both TMS and iTBS appear safe and effective when administered to other anatomical sites in individuals with mTBI [23-26]. Furthermore, studies demonstrated frontal pole iTBS is tolerable with good safety outcomes in individuals without mTBI [63].

We hypothesize that we can improve social and occupational functioning, impulsivity, and SI among US veterans with mTBI. Other studies have specifically found meaningful change in both the UPPS-P and SOFAS over a 2-week neuromodulation intervention period among individuals with major depression [90,135]. Several repetitive TMS (the precursor to iTBS) trials among individuals with SI and major depression have shown reductions in suicidality [71-74]. We anticipate these results will generalize to individuals with mTBI when treated with iTBS [72,136].

Finally, we hypothesize that we will be able to identify changes in neural connectivity among individuals who respond to our iTBS protocol by comparing pretreatment fMRIs with posttreatment fMRIs. Several studies on both iTBS and TMS, administered to the DLPFC, of a variety of treatment durations, showed increased connectivity between the frontal cortex and areas of the limbic system [137]. We expect to see similar results with iTBS administered to the VMPFC.

We will add to the existing literature in several ways. First, there are no widely established treatments for poorer functional outcomes, suicidality, or impulsivity that occur in the context of mTBI. If we can develop an effective intervention that targets all of these likely interrelated conditions, we will be adding important findings to the literature base and we could also improve the quality of life and treat SI among people with mTBI. By further establishing the frontal pole as a tolerable and effective iTBS stimulation site, we could also encourage other researchers to explore this site for potential therapeutic benefit for a variety of conditions. Finally, by identifying fMRI connectivity changes before and after iTBS treatment, we will add to the literature base regarding how neuromodulation works mechanistically, which is still unclear. Moreover, if we can identify specific fMRI connectivity changes correlating to symptomatic changes, it may help us identify future neuromodulatory treatment sites of interest, to continue to further explore this very promising treatment modality.

The proposed study has several key strengths. First, the study uses an established FDA-approved device readily available in both VA and civilian health care systems throughout the United States. Compared with traditional pharmacological and behavioral approaches, iTBS has a reduced latency to treatment response, which could be lifesaving among individuals with SI. Our approach also differs from previous studies that primarily focused on improving SI via the treatment of depressive symptoms. We provide a novel framework for understanding and ultimately treating SI, which could be helpful for individuals who have yet to respond to past evidence-based treatments for this condition. Finally, by targeting the VMPFC and including neuroimaging from before and after treatment, this study has the potential to identify a novel target for iTBS, increase insight into the underlying neurological mechanisms of SI, and identify which individuals are more likely to respond to a neuromodulatory intervention.

There are several limitations to consider. The small sample size reduces the study power, which may limit the detection of statistically significant outcome differences. However, this sample size reflects the pilot nature of the study and funding mechanism. Variability in the baseline characteristics of the patient population (ie, co-occurring psychiatric and somatic conditions, coadministration of other medications, etc) that are being studied in detail in this project may impact our findings. Covariates will be explored; however, given the small sample size, only select covariates can be added to the final models. The relatively short follow-up time of 72 hours after the final iTBS treatment will not provide insight into the long-term effects of our intervention. Future studies would benefit from additional follow-up outcomes assessments and imaging several months after the treatment to assess the intervention’s durability. Although prior studies used a dose of 110% MT and a time frame of 10 sessions, we are still determining if these parameters are optimal. A future dosing study would help optimize the treatment parameters. This study does not select targets based on resting-state data; however, in collecting these data in tandem using neuronavigation, we can potentially target more specific sites in the future.

The proposed study integrates past findings from impulsivity, TBI, neuroanatomical, neuroimaging, neurorehabilitation, and neurostimulation research to create a novel, but highly informed treatment strategy. In this study, we hope to induce neuroplasticity in the connections between the VMPFC and limbic system, through neuromodulation at the frontal pole. In inducing neuroplasticity with iTBS, in an effort to repair damaged neural connections, we hope to mitigate the development of impulsive behaviors, improve social and community functioning, and decrease SI, as we believe all 3 of these conditions would benefit from strengthening the VMPFC-limbic system network. Data obtained by using validated scales will inform future studies.

Moreover, the collection of pretreatment neuroimaging may help us understand what neural signatures predispose individuals to respond favorably to frontal pole iTBS, so we might be able to predict who is most likely to benefit from treatment, following comparable work in predicting response to neurostimulation in individuals with major depression [138]. Thus, findings from the proposed project are expected to have the capacity to (1) help identify which individuals are more likely to respond to a neurostimulation intervention for impulsivity and suicidality related to mTBI and (2) better understand what neural changes we are eliciting. Advancing knowledge in both areas will inform future studies and enable a more rapid determination of which treatments will be most effective for a given individual and determining where connectivity changes are occurring may allow us to further optimize treatment effects.