Control Network Neuromodulation to Enhance Cognitive Training in Complex Traumatic Brain Injury

Concussions from blast and non-blast mechanisms can lead to prolonged post concussive symptoms (PPCS) with debilitating consequences for return to service, functional independence, and quality of life. Unfortunately, algorithmic rehabilitation modalities are of limited generalizability and have small effect sizes, potentially because of limited target engagement, while medications are modestly efficacious but carry significant risks (e.g., addiction). Cognitive symptoms after mTBI are common and debilitating: Contrary to conventional wisdom, mTBI are often not "mild" in regard to disability, with up to 22% of patients reporting functional impairment at one year following injury. In particular, executive functions (such as working memory, set shifting, and response inhibition) appear to be more sensitive to TBI damage and have a greater impact on overall functioning. Despite advances in our understanding of PPCS, no single treatment yet targets putative TBI mechanisms. Medications such as methylphenidate, while modestly helpful, carry adverse consequences such as disinhibition and cardiovascular effects, and little evidence currently exists to recommend any other class of medication. Cognitive rehabilitation is only mildly helpful: Since the 2009 meeting of the Defense Centers for Excellence of Psychological Health and Traumatic Brain Injury, little consensus has been reached regarding the utility or optimum means of delivery for therapist-based rehabilitation of cognitive deficits. Importantly, several limitations in the evidence base have impeded progress toward a standardized approach to post-TBI cognitive symptoms, including: 1) varied rehabilitation interventions used (i.e., therapist-based, computer-based); 2) varied outcome measures studied; and 3) lack of generalization of benefits to global functioning. The most recent trials of cognitive rehabilitation interventions in military populations have attempted to compare these different rehabilitation approaches. While these studies demonstrate that cognitive rehabilitation interventions can improve subjective symptoms and quality of life after mTBI, it is not clear how this benefit is achieved, nor if it is actually targeting a mTBI related mechanism. Neuromodulation can accelerate cognitive recovery: Multiple studies have described use of tDCS in TBI for cognitive performance, nearly all targeting the left DLPFC node of the CCN, with variable improvements noted from a variety of different stimulation protocols. Of note, a recent meta-analysis of tDCS for working memory in neuropsychiatric populations showed that anodal tDCS produces significant improvement in online (during stimulation) working memory accuracy (standardized mean difference = 0.77). rTMS also shows promise for treating mTBI: rTMS is a powerful neuromodulation technique that induces robust neuroplasticity, effectively treats disease, and is better characterized in terms of its mechanism of action. Recently, several small naturalistic and controlled studies have been reported, indicating that both excitatory and inhibitory rTMS are safe in the mTBI population and can improve post concussive headaches, chronic pain, tinnitus, and depression. A recent negative trial of non-specific rTMS for cognition in severe TBI highlights the important need for more precise methods of patient selection, treatment selection, and target engagement. Study procedures involve a baseline testing visit, 16 treatment visits, and a post-treatment testing visit. Participants will also be contacted at 3 months and 6 months post treatment for follow-up. Baseline Testing Visit: Study assessments for collecting demographic information, history and TBI data, symptom severity information, and neuropsychological testing will be completed at the baseline testing visit. Testing will be performed by trained study personnel under direct supervision of the study Co-Investigators. Baseline Visit, MRI scanning: Subjects will undergo a 60-minute MRI using a 3T Siemens Prisma scanner at both the Albuquerque (MRN) and Minneapolis (CMRR) sites. High-resolution T1- and T2-weighted images (1 x 1 x 1 mm resolution), DTI, pCASL, resting state fMRI, and task-related fMRI will be collected. All anatomical data will be reviewed by a board-certified neuroradiologist blinded to group identification. All positive findings will be coded for presence, location, severity and pathology of each abnormality, consistent with the imaging CDEs. Participants will undergo structural and functional MRI scanning at rest and during a multisensory working memory task. The MMWM is a continuous performance test in which subjects respond to simultaneous sequences of visual (squares on a grid) and auditory (spoken numbers) stimuli by pressing a button if stimuli in either or both sensory modalities match a previous stimulus (1-back or 2-back). Neuromodulation + Training Sessions (16 total; 1 hr each): In Albuquerque, following completion of the Baseline Visit, participants will receive either active HD-tDCS, active rTMS, or sham stimulation to the left DLPFC for a total of 30 minutes, 4 days/week, for 4 consecutive weeks. During neuromodulation sessions, subjects will describe physical sensations such as tingling or itching using a 10-point anchored Likert scale. Administration of HD-tDCS will be stopped immediately if subjects report 8 or above for discomfort, or if subjects wish to stop at any time. Subjects will have their mood, energy, pain, and arousal levels assessed using visual analog 10-point scales. These checks will occur every ten minutes during the stimulation session. HD-tDCS: The Star-Stim 8 high-definition transcranial electrical stimulator will be used to administer HD-tDCS. Targeting of the left DLPFC will be done by utilizing a standard EEG cap fitted snugly to the subject's head. Several 10-20 EEG system positions will be measured and confirmed to ensure that the cap is correctly oriented on the head. Round, 1 cm2 HD-tDCS electrodes will be utilized to deliver anodal current and receive cathodal current. Two anodal electrodes will be placed on the scalp over the functionally determined DLPFC, delivering approximately 1 mA of current each, in order to reduce overall sensation. Six return electrodes will be placed in various positions around the anodes. Precise electrode placement for each subject will be determined according to current modeling software that will use each individual's MRI T1 images to construct a 3-dimensional finite element model aiming to maximize current density within the DLPFC while minimizing current density outside of it. Current for the treatment condition will be applied at 2.0 mA for 30 minutes for a total delivered charge of 60 mA-min. Current will be ramped up over 1 minute at initiation and ramped down over 1 minute with termination. Impedances are monitored in real-time for each channel to ensure that they do not exceed recommended limits (e.g., 200 kilo-ohms). rTMS: Participants will receive 16 sessions of TMS to the functional area in the DLPFC identified with fMRI while performing a working memory task. A Magventure MagPro Transcranial Magnetic Stimulator (Albuquerque, NM) and a Magstim Rapid 2 Transcranial Magnetic Stimulator (Minneapolis, MN) will be used to administer active and sham rTMS to the left dorsolateral prefrontal cortex in 54 Veterans and Warfighters (36 active, 18 sham) with mTBI and cognitive postconcussive symptoms. At the first session, dose titration will be performed. After sitting comfortably in the chair the subject's head is held with a moldable pillow, and they are given earplugs to protect against coil discharge noise. Surface electromyography leads will be applied to clean skin over the right hand over the first dorsal interosseous (FDI) muscle. The motor cortex hotspot for the muscle will be identified with single-pulse TMS delivered to the contralateral hemisphere. Resting motor threshold will be determined to be the lowest stimulation intensity necessary to elicit a motor-evoked potential meeting TMS Clinical Society conventional criteria of 50 uV peak-to-peak on 5 out of 10 trials. The subject's fMRI data is loaded into the neuronavigation tracking computer to locate the functional hotspot within the DLPFC. An infrared camera connected to the targeting computer will track the three-dimensional positions of the subject's head and the TMS coil in real time, via affixed tracking markers. The TMS coil is then positioned over the left forehead using the co-registered MRI data and identified head landmarks. In each session, up to 1800 pulses will be delivered according to conventional parameters for excitatory TMS (e.g, 60 trains of 10 triplet pulses, frequency 5 Hz, train duration 2 seconds, intertrain interval 8 seconds). Magnetic field strength will be 120% of resting motor threshold. Ramp up of magnetic field strength may be utilized in the first session for tolerability. Side effects will be monitored and coil angle adjusted to improve tolerability if necessary without compromising placement. Sham: The sham group (n=36 total) will be split: half (n=18) will receive sham HD-tDCS, and half (n=18) will receive sham rTMS. Participants receiving sham HD-tDCS will receive a current ramp up to the intensity of the real intervention in 30 seconds, then the current will ramp down to \< 0.1 mA, an amount that has been shown not to have any physiologic effect. With 1 minute left in the stimulation session, the current will ramp up to full strength in 30 seconds, then ramp down in 30 seconds. This paradigm is used as a control condition, rather than the absence of stimulation, to equate aspects of the procedure (preparation and application of electrodes), and to give the participant a degree of physical sensation that is somewhat similar to that of the real stimulation group while remaining well below the level sufficient to affect brain function and behavior. To accomplish a double blind, the HD-tDCS machine is programmed to randomize sham versus active stimulation and keeps track of the stimulation protocol for later querying. Sham rTMS is delivered with a sham coil, which delivers no physiologically active magnetic fields to the brain. It creates a similar sound as the active rTMS coil, and features electrodes that contact the skin and deliver a mild electrical current which resembles the sensations caused by typical rTMS pulses. Double blind is maintained through use of randomized program codes assigned to each subject, and which dictate the choice of coil for the participant. Cognitive Training Tasks: For all groups, participants will be administered the APT-3 training battery for 30 minutes at each session, during treatment and sham. Each session's material will be determined by the study staff in advance according to a predetermined syllabus, and the participant will proceed through the material as efficiently as they can. If rTMS is given, the training will take place after rTMS is completed. If HD-tDCS is given, the training will take place concurrently while the stimulation is occurring. Post-treatment Visit: After the 16 neuromodulation + training sessions are completed, the testing and assessments from the Baseline Visit is repeated for the Post-treatment Visit, including neuropsychological testing, symptom assessment, and MRI. Demographic information will not be repeated. Long term follow-up (30 min each): At 3 months and 6 months after receiving neuromodulation and cognitive training, subjects will be contacted via telephone and administered symptom burden and quality of life assessment tools.