Traumatic brain injury (TBI) remains a significant public health problem and is a leading cause of death and disability in many countries. Durable treatments for neurological function deficits following TBI have been elusive, as there are currently no FDA-approved therapeutic modalities for mitigating the consequences of TBI. Neurostimulation strategies using various forms of electrical stimulation have recently been applied to treat functional deficits in animal models and clinical stroke trials. The results from these studies suggest that neurostimulation may augment improvements in both motor and cognitive deficits after brain injury. Several studies have taken this approach in animal models of TBI, showing both behavioral enhancement and biological evidence of recovery. There have been only a few studies using deep brain stimulation (DBS) in human TBI patients, and future studies are warranted to validate the feasibility of this technique in the clinical treatment of TBI. In this review, the authors summarize insights from studies employing neurostimulation techniques in the setting of brain injury. Moreover, they relate these findings to the future prospect of using DBS to ameliorate motor and cognitive deficits following TBI.
Samuel S. Shin, C. Edward Dixon, David O. Okonkwo and R. Mark Richardson
Satoshi Maesawa, Douglas Kondziolka, C. Edward Dixon, Jeffrey Balzer, Wendy Fellows and L. Dade Lunsford
Object. Any analysis of the potential role of stereotactic radiosurgery for epilepsy requires the experimental study of its potential antiepileptogenic, behavioral, and histological effects. The authors hypothesized that radiosurgery performed using subnecrotic tissue doses would reduce or abolish epilepsy without causing demonstrable behavioral side effects. The kainic acid model in rats was chosen to test this hypothesis.
Methods. Chronic epilepsy was successfully created by stereotactic injection of kainic acid (8 µg) into the rat hippocampus. Epileptic rats were divided into three groups: high-dose radiosurgery (60 Gy, 16 animals), low-dose (30 Gy, 15 animals), and controls. After chronic epilepsy was confirmed by observation of the seizure pattern and by using electroencephalography (EEG), radiosurgery was performed on Day 10 postinjection. Serial seizure and behavior observation was supplemented by weekly EEG sessions performed for the next 11 weeks. To detect behavioral deficits, the Morris water maze test was performed during Week 12 to study spatial learning and memory. Tasks involved a hidden platform, a visible platform, and a probe trial.
After radiosurgery, the incidence of observed and EEG-defined seizures was markedly reduced in rats from either radiosurgically treated group. A significant reduction was noted after high-dose (60 Gy) radiosurgery in Weeks 5 to 9 (p < 0.003). After low-dose (30 Gy) radiosurgery, a significant reduction was found after 7 to 9 weeks (p < 0.04). During the task involving the hidden platform, kainic acid—injected rats displayed significantly prolonged latencies compared with those of control animals (p < 0.05). Hippocampal radiosurgery did not worsen this performance. The probe trial showed that kainic acid—injected rats that did not undergo radiosurgery spent significantly less time than control rats in the target quadrant (p = 0.03). Rats that had undergone radiosurgery displayed no difference compared with control rats and demonstrated better performance than rats that received kainic acid alone (p = 0.04). Radiosurgery caused no adverse histological effects.
Conclusions. In a rat model, radiosurgery performed with subnecrotic tissue doses controlled epilepsy without causing subsequent behavioral impairment.
C. Edward Dixon, Bruce G. Lyeth, John T. Povlishock, Robert L. Findling, Robert J. Hamm, Anthony Marmarou, Harold F. Young and Ronald L. Hayes
✓ Fluid percussion models produce brain injury by rapidly injecting fluid volumes into the cranial cavity. The authors have systematically examined the effects of varying magnitudes of fluid percussion injury in the rat on neurological, systemic physiological, and histopathological changes. Acute neurological experiments showed that fluid percussion injury in 53 rats produced either irreversible apnea and death or transient apnea (lasting 54 seconds or less) and reversible suppression of postural and nonpostural function (lasting 60 minutes or less). As the magnitude if injury increased, the mortality rate and the duration of suppression of somatomotor reflexes increased. Unlike other rat models in which concussive brain injury is produced by impact, convulsions were observed in only 13% of survivors. Transient apnea was probably not associated with a significant hypoxic insult to animals that survived. Ten rats that sustained a moderate magnitude of injury (2.9 atm) exhibited chronic locomotor deficits that persisted for 4 to 8 days. Systemic physiological experiments in 20 rats demonstrated that all levels of injury studied produced acute systemic hypertension, bradycardia, and increased plasma glucose levels. Hypertension with subsequent hypotension resulted from higher magnitudes of injury. The durations of hypertension and suppression of amplitude on electroencephalography were related to the magnitudes of injury. While low levels of injury produced no significant histopathological alterations, higher magnitudes produced subarachnoid and intraparenchymal hemorrhage and, with increasing survival, necrotic change and cavitation. These data demonstrate that fluid percussion injury in the rat reproduces many of the features of head injury observed in other models and species. Thus, this animal model could represent a useful experimental approach to studies of pathological changes similar to those seen in human head injury.
Antonio A. F. DeSalles, Pauline G. Newlon, Yoichi Katayama, C. Edward Dixon, Donald P. Becker, Henry H. Stonnington and Ronald L. Hayes
✓ Studies in humans have shown that sensory stimuli, presented in the context of certain tasks, can elicit a late positive component (LPC), namely P300, in the scalp-recorded evoked potential believed to reflect neural activity related to attentional processes. A similar LPC has been reported in cats and monkeys. In this study, the LPC of the auditory evoked potential (AEP) in the cat was used to detect impairment in attention to a relevant stimulus after low levels of cerebral concussion produced by a fluid percussion device. A hollow screw (for fluid percussion) and stainless steel screws (for AEP recording) were surgically placed in the skull. After recovery from surgery, animals were trained in the paradigm to obtain an LPC. Pupillary dilation was conditioned to tones. A random sequence of two discriminable tones was presented. The lower tone had a probability of 0.1 and was followed by a tail shock (tone-shock). After 400 to 1000 tone-shock presentations, animals attended to the lower tone stimulus as inferred by selective pupillary dilation. In the AEP an early positive component at 50 to 120 msec related to an alerting response was enhanced, and an LPC at 250 to 450 msec appeared in response to the paired tone-shock. Animals were then subjected to cerebral concussion. Complete recovery of normal reflexes, motor coordination, and orienting response was seen within 2 hours after injury. The LPC was suppressed for a period of at least 3 days, suggesting that low magnitudes of brain injury can disrupt higher-order neural activities. This disruption can persist despite recovery of other neurological functions.
Amy K. Wagner, Dianxu Ren, Yvette P. Conley, Xiecheng Ma, Mary E. Kerr, Ross D. Zafonte, Ava M. Puccio, Donald W. Marion and C. Edward Dixon
Dopamine (DA) pathways have been implicated in cognitive deficits after traumatic brain injury (TBI). Both sex and the dopamine transporter (DAT) 3′ variable number of tandem repeat polymorphism have been associated with differences in DAT protein density, and DAT protein affects both presynaptic DA release, through reverse transport, and DA reuptake. Catecholamines and associated metabolites are subject to autooxidation, resulting in the formation of reactive oxygen species that may contribute to subsequent oxidative injury. The purpose of this study was to determine associations between factors that affect DAT expression and cerebrospinal fluid (CSF) DA and metabolite levels after severe TBI.
Sixty-three patients with severe TBI (Glasgow Coma Scale score ≤ 8) were evaluated. The patients' genotypes were obtained using previously banked samples of CSF, and serial CSF samples (416 samples) were used to evaluate DA and metabolite levels. High-performance liquid chromatography was used to determine CSF levels of DA, 3,4-dihydroxyphenylacetic acid (DOPAC), and homovanillic acid (HVA) during the first 5 days after injury.
Mixed-effects multivariate regression modeling revealed that patients with the DAT 10/10 genotype had higher CSF DA levels than patients with either the DAT 9/9 or DAT 9/10 genotypes (p = 0.009). Females with the DAT 10/10 genotype had higher CSF DA levels than females with the DAT 9/9 or DAT 9/10 genotypes, and sex was associated with higher DOPAC levels (p = 0.004). Inotrope administration also contributed to higher DA levels (p = 0.002).
In addition to systemic administration of DA, inherent factors such as sex and DAT genotype affect post-TBI CSF DA and DA metabolite levels, a phenomenon that may modulate susceptibility to DA-mediated oxidative injury.
Michael L. Forbes, Robert S. B. Clark, C. Edward Dixon, Steven H. Graham, Donald W. Marion, Steven T. Dekosky, Joanne K. Schiding and Patrick M. Kochanek
Minimizing secondary injury after severe traumatic brain injury (TBI) is the primary goal of cerebral resuscitation. For more than two decades, hyperventilation has been one of the most often used strategies in the management of TBI. Laboratory and clinical studies, however, have verified a post-TBI state of reduced cerebral perfusion that may increase the brain's vulnerability to secondary injury. In addition, it has been suggested in a clinical study that hyperventilation may worsen outcome after TBI.
Object. Using the controlled cortical impact model in rats, the authors tested the hypothesis that aggressive hyperventilation applied immediately after TBI would worsen functional outcome, expand the contusion, and promote neuronal death in selectively vulnerable hippocampal neurons.
Methods. Twenty-six intubated, mechanically ventilated, isoflurane-anesthetized male Sprague—Dawley rats were subjected to controlled cortical impact (4 m/second, 2.5-mm depth of deformation) and randomized after 10 minutes to either hyperventilation (PaCO2 = 20.3 ± 0.7 mm Hg) or normal ventilation groups (PaCO2 = 34.9 ± 0.3 mm Hg) containing 13 rats apiece and were treated for 5 hours. Beam balance and Morris water maze (MWM) performance latencies were measured in eight rats from each group on Days 1 to 5 and 7 to 11, respectively, after controlled cortical impact. The rats were killed at 14 days postinjury, and serial coronal sections of their brains were studied for contusion volume and hippocampal neuron counting (CA1, CA3) by an observer who was blinded to their treatment group.
Mortality rates were similar in both groups (two of 13 in the normal ventilation compared with three of 13 in the hyperventilation group, not significant [NS]). There were no differences between the groups in mean arterial blood pressure, brain temperature, and serum glucose concentration. There were no differences between groups in performance latencies for both beam balance and MWM or contusion volume (27.8 ± 5.1 mm3 compared with 27.8 ± 3.3 mm3, NS) in the normal ventilation compared with the hyperventilation groups, respectively. In brain sections cut from the center of the contusion, hippocampal neuronal survival in the CA1 region was similar in both groups; however, hyperventilation reduced the number of surviving hippocampal CA3 neurons (29.7 cells/hpf, range 24.2–31.7 in the normal ventilation group compared with 19.9 cells/hpf, range 17–23.7 in the hyperventilation group [25th–75th percentiles]; *p < 0.05, Mann—Whitney rank-sum test).
Conclusions. Aggressive hyperventilation early after TBI augments CA3 hippocampal neuronal death; however, it did not impair functional outcome or expand the contusion. These data indicate that CA3 hippocampal neurons are selectively vulnerable to the effects of hyperventilation after TBI. Further studies delineating the mechanisms underlying these effects are needed, because the injudicious application of hyperventilation early after TBI may contribute to secondary neuronal injury.