Diminished Corticomotor Excitability in Gulf War I Veterans with Chronic Pain Symptoms

Background: Diminished motor cortical excitability, as a measurement of increased resting motor threshold with transcranial magnetic stimulation (TMS), is known to be associated with chronic pain conditions. Over 175,000 veterans who were deployed to the Persian Gulf in 1990-91 suffer from an unexplained multisymptom illness termed Gulf War Illness (GWI) with chronic headaches and diffuse body pain as the most commonly reported symptoms. This study hypothesized that veterans with Gulf War Illness related pain exhibit diminished corticomotor excitability associated with chronic pain states and aimed to assess the resting motor threshold (RMT) at the primary motor cortex (M1) in veterans with GWI. Methods: Single pulse TMS was administered over the left motor cortex, using anatomical scans of each subject to guide the TMS coil, starting at 25% of maximum stimulator output (MSO) and increasing in steps of 2% until a motor response with a 50 µV peak to peak amplitude, dened as the RMT, was evoked at the exor pollicis brevis muscle. RMT was then analyzed using Repeated Measures Analysis of Variance (RM-ANOVA). Results: Gulf War Veterans (GWV) with chronic headaches and body pain (N=20) had a signicantly (P<0.001) higher average resting motor threshold (%± SD) of 77.2%±16.7% compared to age and gender matched GWV Controls (N=20), whose average was 55.6%±8.8%. Conclusion: This study demonstrated that a higher level of stimulation was required to evoke a motor response in veterans with GWI related pain. This diminished corticomotor excitability that suggests a reduced state of supraspinal pain modulatory function. Though further studies will be needed, TMS may provide a means of effectively rectifying this modulatory decit. the GWV-HAP group were in Combat (C) roles, compared to three subjects in the GWV Control group. The remaining subjects were Non-Combat (NC). Body Mass Index (BMI±SD) in the GWV-HAP group was higher at 30.3±4.0 compared to the GWV Control group who had an average BMI of 29.8±4.0. BMI data was missing from one subject in the GWV Control group, who was excluded from the calculation of its average.


Introduction
Gulf War Illness (GWI) is a chronic multisymptom illness that uniquely affects military personnel of the 1990-91 Persian Gulf War. An estimated 25-32% of the 700,000 veterans returning from theater in the Persian Gulf have exhibited a sequela of symptoms characterized by GWI. These symptoms include neurological dysfunction, gastrointestinal, respiratory, or dermatological issues as well as pain and fatigue. (1,2) Of those, headaches and diffuse body pain have been reported as two of the most debilitating conditions of Gulf War Illness. (3,4) While the pathophysiology underlying pain symptomology has not been well de ned, assessing the underlying cortical excitability can provide a new level of understanding in any associated aberrant central pain neuromodulatory functions, and thus potentially providing guidance for therapeutic intervention. While it is widely accepted that chronic pain states can be associated with impaired corticomotor excitability represented by a diminished supraspinal modulatory state from traumatic or non-traumatic causes. (5)(6)(7)(8)(9)(10) To the authors' best knowledge, no study has been conducted to assess the corticomotor excitability in patients with GWI related chronic headache and pain problems.
Transcranial magnetic stimulation (TMS) can evoke action potentials on the cerebral cortex, and is one of the most established methods for assessing motor cortical excitability via the measurement of the resting motor threshold (RMT), de ned as the minimum percentage of the maximum intensity needed to illicit a motor response demonstrated on an electromyogram. (11) To further the understanding of the pathophysiology of pain associated with GWI, this study compared RMT in patients with GWI-associated chronic headaches, muscle, and joint pain, with gender and age matched Gulf War veterans who served in the same period, but did not exhibit these pain symptoms, hypothesizing that cortical excitability of the motor cortex is suppressed in patients with GWI.

Methods
Veterans who served in the Persian Gulf War I between August 1990 and July 1991 were screened and enrolled based on the study protocol approved by the Institutional Human Subject Protection Committee. The study consisted of three visits as follows: 1) Informed Consent and Screening; 2) MRI Scan; and 3) Resting Motor Threshold Procedure with TMS. On average, study participants completed all three visits within two weeks.

Screening and Demographics
The main study inclusion criteria consisted of male or female veterans between the ages of 18 and 65 years old who served in the Persian Gulf for at least thirty consecutive days between August 1990 and July 1991. Veterans in the Gulf War Illness-associated headache and pain group (GWV-HAP) were also required to meet the following diagnostic criteria: 1. CDC Criteria for Gulf War Illness (at least 1 symptom in 2 of the following domains) ( (14) At least 5 headache attacks, ful lling the following criteria i and ii: Headache attacks lasting 4-72 hours Occupied by nausea and/or photophobia and phonophobia In addition, GWV-HAP subjects were required to have an average headache exacerbation intensity greater than or equal to 3/10 on a numerical rating scale (NRS), occurring at least three times a week and at least one of which should have lasted more than four hours in the past three months. (15) They were also required to have daily muscle pain with an intensity of greater than or equal to 3/10 on a NRS and daily joint pain with an intensity of greater than or equal to 3/10 on a NRS. (15) Veterans who met the initial study criteria of a male or female veteran between 18 and 65 years old who served in the Persian Gulf for at least thirty consecutive days between August 1990 and July 1991, but did not meet both GWI diagnostic criteria and the pain conditions de ned above for GWV-HAP were recruited for the GWV Control group. GWV Control group veterans were gender and age-matched within one year to a veteran in the GWV-HAP group.
Veterans were excluded from both GWV-HAP and GWV Control groups of the study if they had any of the following: pregnancy; history of pacemaker implant; any ferromagnetic material in the brain and/or body; history of dementia; major psychiatric diseases; life threatening diseases; presence of any chronic neuropathic pain; history of seizures; pending litigation; low back pain with mechanical origins; lack of ability to understand or speak English; history of traumatic brain injury; chronic tension or cluster headaches; or ongoing cognitive rehabilitation or treatment for Post-Traumatic Stress Disorder (PTSD).
After subjects were enrolled into the study, demographic information pertaining to age, gender, race, body mass index (BMI), and Persian Gulf War durations and duties were collected. Those who served in the 1990-91 Persian Gulf War were indicated as serving under "Gulf War I," whereas those who served in both the 1990-91 Persian Gulf War and post-1991 Gulf War were indicated as "Both." Time served in the Persian Gulf was reported in months. Military branches were reported as either Navy, Marines, Army, or Other, with "Other" consisting of the Coast Guard and Air Force branches. Combat status was reported based on military occupational specialty, with combat roles coded as "C" and non-combat roles coded as "NC."

Neuroimaging Data Acquisition
Neuroimaging data was acquired through magnetic resonance imaging (MRI) using a General Electric Discovery MR750 3.0T scanner. During the scan, subjects were instructed to keep the head still and padding was provided between the subject's head and the scanner head coil to minimize head movement. Anatomical scans were obtained with magnetization prepared rapid gradient echo (MP RAGE) samplings (176 slices, T1 450, TE 3.172, TR (ms) 8.132, 256 × 256 and 1 mm slice thickness).

Study Procedure: Resting Motor Threshold Preparation
Subjects were asked to sit in a comfortable chair and relax as much as possible. Electromyography (EMG) recordings (Xltek, Oakville, Ontario, Canada) were collected through silver-silver chloride surface electrodes attached to the contralateral exor pollicis brevis muscle and a 3 cm diameter ground electrode placed on the back of the hand. Manufacturer preinstalled software was used to collect signal with a recording time window of 200 ms, gain of 100 uV, notch lter 60 Hz, LFF 20 Hz, and HFF 10 k Hz.
TMS was performed with a gure-of-eight Cool-B65 coil connected to MagPro R30 (Alpine BioMed, Fountain Valley, CA, USA).

TMS Neuronavigation System
All anatomical MRI images were processed using BrainVoyager TMS Neuronavigation Software (Brain Innovation, Maastricht, The Netherlands), here anatomical scans were oriented along the anterior commissure (AC) and posterior commissure (PC) plane. The transformed images were used to create 3D head and brain meshes, where head ducial points (FDP) were marked. Utilizing the BrainVoyager ultrasound-based co-registration system via sensors attached to each subject's face and the TMS coil, stereotaxic data was used to determine the spatial position of the TMS stimulation site in relation to each subject. A digitizing pen with ultrasound sensors was used to mark each subject's face, which was coregistered to corresponding FDPs on a subject's head mesh. As a result, TMS coil was visualized in real time and space, allowing the investigator to focus the magnetic ux on a speci c target region.

Resting Motor Threshold
Under the guidance of BrainVoyager Neuronavigation, a constant suprathreshold stimulus intensity was applied from the TMS coil on the left primary motor cortex (LMC) in an ascending order. Given the pain symptoms in this patient population are diffuse, the laterality of the testing side was chosen for the ease of operation and location derived from previous pain related studies. A single pulse TMS was delivered to the LMC starting at 25% of maximum stimulator output (MSO) and increasing in steps of 2% until a motor response at the exor pollicis brevis muscle of 50 µV peak to peak amplitude in ve out of ten consecutive trials was evoked. The MSO of the MagPro R30 used in this procedure was 1850V, and the minimum stimulus required for the evoked response was de ned as the resting motor threshold (RMT).
(16) The cortical location for exor pollicis brevis muscle was obtained by using the personalized brain mesh to trace the precentral gyrus region corresponding to the contralateral hand in the homunculus layout, which was con rmed using the EMG. Once the cortical location for the obtained RMT was marked, the distance between the center of the TMS coil to the target site was recorded. (17) Data Analysis The within subject design was used for this study. All GWV-HAP veterans were age and gender matched to a counterpart control group, and there was no signi cant difference between them on these measurements using Chi-Square for categorical data and Analysis of Variance for continuous data. Furthermore, these two groups were not signi cantly different on race, Gulf War period, duration, number of deployments, military branch, combat status, nor BMI. The primary outcome, resting motor threshold (RMT) data, was normally distributed and analyzed using Repeated Measures Analysis of Variance (RM-ANOVA) with one within-factor of headache and pain; GWV-HAP for those who met the headache and pain criteria, and GWV Control for those who did not meet the headache and pain criteria; using SPSS version 23.

Demographics
Forty-seven veterans were screened based on the study protocol approved by the Institutional Human Subject Protection Committee. Out of the forty-seven veterans, one failed the study screening, ve dropped prior to study procedures due to time commitment or MRI complications, and one was excluded due to information shared after the study procedures that would exclude him or her from the study. The remaining forty gender and age-matched veterans (20 GWV-HAP and 20 GWV Control) were enrolled in the study and their data analyzed. Caucasians, three Asians, two African-Americans, two Hispanics, and one Paci c Islander, whereas the GWV Control group consisted of eleven Caucasians, ve African-Americans, two Hispanics, one Paci c Islander, and one African-American/Asian. In the GWV-HAP group, nine veterans were in the Navy branches, seven in the Marines, three in the Army, and one in Other, compared to ten Navy, six Marine, and four Army subjects in the GWV Control group. Eight subjects in the GWV-HAP group were in Combat (C) roles, compared to three subjects in the GWV Control group. The remaining subjects were Non-Combat (NC). Body Mass Index (BMI±SD) in the GWV-HAP group was higher at 30.3±4.0 compared to the GWV Control group who had an average BMI of 29.8±4.0. BMI data was missing from one subject in the GWV Control group, who was excluded from the calculation of its average.  There were no signi cant differences found between groups in all aforementioned demographic categories (see Table 3). There were no signi cant differences found between groups in all demographic categories where P = 0.05.

Discussion
Supraspinal pain perception consists of mainly three functional regions: 1) sensory discriminatory regions such as the primary and secondary somatosensory cortices (SSC1 and SSC2) and inferior parietal lobe (IPL); 2) affective regions such as anterior cingulate cortex (ACC) and insula (IN); and 3) modulatory regions involving motor and various regions of prefrontal cortices (PFCs). (18,19) Additionally, the insula (IN) has been known to play a role in assessing the magnitude of pain, while the inferior parietal lobe (IPL) aids in distinguishing spatial discrimination in pain perception.(20-23) Chronic pain state is often associated with a mal-adaptation in the supraspinal pain processing, which is often accompanied with diminished modulatory functional connectivity from the prefrontal cortices with diminished motor cortex excitability as re ected by an elevated resting motor threshold. (24,25) Persian Gulf War veterans in this study demonstrated an increase in RMT level compared to their nonpain Gulf War veteran counterparts. The need for a higher level of stimulation in Gulf War veterans is consistent with other chronic pain conditions resulted from either direct or indirect neuronal traumas which resulted in impaired supraspinal pain modulation (26)(27)(28)(29)(30)(31)(32)(33)(34). Although the pathophysiology behind the headaches and diffuse body pain in this population has not been explicitly de ned, the observed functional de cit in supraspinal cortical modulatory function serves as a signi cant step in understanding the pathophysiology underlying the high prevalence of chronic pain states in this patient population. (35)(36)(37) While transcranial magnetic stimulation-evoked resting motor threshold has been demonstrated to be a reliable method for assessing cortical excitability, it may also be a viable solution for the chronic pain experienced by these veterans. TMS is approved by the United States Food and Drug Administration for treating depression and migraine headaches. (38)(39)(40) In addition, multiple meta-analyses and panel consensus review de nitively supported a high level of evidence for treating central neuropathic states, such as in the management of chronic and debilitating headaches in mild Traumatic Brain Injury patients. (17,41) High frequency (> 1 Hz) TMS works by evoking action potentials to directly excite the motor cortex and potentially regain cortical excitability, which cannot be accomplished with traditional pharmacological methods. Since Gulf War veterans with Gulf War Illness-associated headaches and pain present with a similarly increased RMT pro le, it is possible they may also nd relief with repetitive transcranial magnetic stimulation in the form of increased pain modulatory function.
Limitations to the study include those due to an all-male study population and recruitment from a single site. Although all subjects were successfully gender and age-matched to a counterpart in the other group of the study, all forty veterans were male which is not indicative of the true veteran population that was deployed to the Persian Gulf in 1990-91. While female veterans were screened for the study, they were far fewer in number compared to male veterans who met the criteria. This is most likely explained by the inherently In short, the observed result from the current study suggests that the high prevalence of headaches and diffuse body pain in 1990-91 Persian Gulf War veterans is associated with impaired cortical excitability and pain modulation. This was expressed by a higher resting motor threshold in GWV-HAP veterans compared to control. The signi cance of transcranial magnetic stimulation in this study is two-fold. TMS was utilized as the primary modality in assessing resting motor threshold. In a clinical setting, the same TMS machine may also provide veterans potential bene t long-term pain relief that is currently lacking in the management of this multisymptom illness.

Consent for publication
Written informed consent for publication of clinical details was obtained from the patients.

Availability of data and materials
The data used and/or analyzed in this study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This study received funding support from the following agency: Department of Defense Congressionally Directed Medical Research Program Grant (W81XWH-16-1-0754).
Authors' contributions KL, AK, and VMS assisted in conducting the study, data analysis, and manuscript preparation. SG assisted in the study design, data analysis, and manuscript preparation. JJ and JW assisted in patient evaluation, recruitment, and screening. RL assisted in neuroimaging assessment. MV assisted in conducting the study and patient evaluation. TR assisted in patient evaluation. AL designed and conducted the study and assisted in manuscript preparation.