A New Model of Experimental Hemispherotomy in Young Adult Rattus Norvegicus: a Neural Tract Tracing and SPECT in Vivo Study

LABORATORY INVESTIGATION

A new model of experimental hemispherotomy in young adult Rattus norvegicus: a neural tract tracing and SPECT in vivo study

Ivair Matias Jr., ScM,1,2 Daoud Hibrahim Elias-Filho,2 Camila Araújo Bernardino Garcia, ScM,1 Guilherme Henrique Silva,1 Jorge Mejia, ScD,3 Francisco Romero Cabral, ScD,3 Ana Cláudia Camargo Miranda, ScM,3 Sérgio Gomes da Silva, ScD,3,5 Luíza da Silva Lopes, MD, ScD,1 Norberto Cysne Coimbra, MD, ScM, ScD,2,4 and Hélio Rubens Machado, MD, ScM, ScD1,4

1Laboratory of Pediatric Neurosurgery and Developmental Neuropathology, Department of Surgery and Anatomy, 2Laboratory of Neuroanatomy and Neuropsychobiology, Department of Pharmacology, and 4Neuroelectrophysiology Multiuser Centre and Neurobiology of Pain and Emotions Laboratory, Department of Surgery and Anatomy, Ribeirão Preto Medical School of the University of São Paulo; 3Hospital Israelita Albert Einstein; and 5Núcleo de Pesquisas Tecnológicas (NPT), Universidade de Mogi das Cruzes, São Paulo, Brazil

OBJECTIVE The objective of this study was to describe a new experimental model of hemispherotomy performed on laboratory animals. METHODS Twenty-six male young adult Wistar rats were distributed into two groups (surgery and control). The nonfluorescent anterograde neurotracer biotinylated dextran amine (BDA; 10,000 MW) was microinjected into the motor cortex area (M1) according to The Rat Brain in Stereotaxic Coordinates atlas to identify pathways and fibers disconnected after the experimental hemispherectomy. SPECT tomographic images of 99mTc hexamethylpropyleneamine oxime were obtained to verify perfusion in functioning areas of the disconnected and intact brain. A reproducible and validated surgical procedure is described in detail, including exact measurements and anatomical relationships. An additional 30 rodents (n = 10 rats per group) were divided into naïve, sham, and hemispherotomy groups and underwent the rotarod test. RESULTS Cortico-cortical neural pathways were identified crossing the midline and contacting neuronal perikarya in the contralateral brain hemisphere in controls, but not in animals undergoing hemispherotomy. There was an absence of perfusion in the left side of the brain of the animals undergoing hemispherotomy. Motor performance was significantly affected by brain injuries, increasing the number of attempts to maintain balance on the moving cylinder in the rotarod test at 10 and 30 days after the hemispherotomy, with a tendency to minimize the motor performance deficit over time. CONCLUSIONS The present findings show that the technique reproduced neural disconnection with minimal resection of brain parenchyma in young adult rats, thereby duplicating the hemispherotomy procedures in human patients. https://thejns.org/doi/abs/10.3171/2017.12.JNS171150 KEYWORDS hemispherotomy; biotinylated dextran amine neurotracer; cerebral cortex; corpus callosum; small-animal SPECT; epilepsy

emispherectomy has been used for decades in the scribed as an effective surgical technique for hemispheric treatment of children with refractory epilepsy and disconnection.14,48 This technique allows a complete dis- unilateral hemispheric lesions. While anatomi- connection between both brain hemispheres with good calH hemispherectomy is still performed, other techniques, results regarding seizure control; it also causes minimal such as hemispherotomy and functional hemispherectomy, intraoperative bleeding and the shortest possible surgical are now also used to disconnect a cerebral hemisphere.37,48 time, with small skin incisions, craniotomies, and the least Introduced in the 1990s, hemispherotomy has been de- possible resection of the cerebral parenchyma.13 A system-

ABBREVIATIONS BDA = biotinylated dextran amine; CPu = caudatus/putamen; HMPAO = hexamethylpropyleneamine oxime. SUBMITTED May 9, 2017. ACCEPTED December 11, 2017. INCLUDE WHEN CITING Published online June 8, 2018; DOI: 10.3171/2017.12.JNS171150.

©AANS 2018, except where prohibited by US copyright law J Neurosurg June 8, 2018 1

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al. atic review of the literature related to hemispherotomy for of the Ribeirão Preto Medical School of the University of the treatment of refractory epilepsy found that this surgical São Paulo and of the Hospital Israelita Albert Einstein in procedure is highly effective in the treatment of refractory agreement with the animal research ethics adopted by the epilepsy in children, particularly considering acquired National Council for Animal Experimentation Control, and progressive etiologies, and that hemispherotomy pro- affiliated with the International Council for Laboratory cedures are associated with a more favorable complication Animal Science. All efforts were made to minimize suf- profile.36 fering, as well as the number of animals used in this study. The clinical examination after functional hemispherectomy in children often demonstrates the presence of func- Experimental Groups tional deficits in the proximal and distal muscles of the Rats were distributed into two experimental groups for contralateral side to the surgical intervention. The hand neural tract tracing and SPECT procedures (n = 26). Neu- and wrist functions have the highest deficits (94%), fol- ral tract tracing (n = 4) and SPECT (n = 1) procedures lowed by arm dysfunction (54%) and leg and foot disabili- comprised a control group (n = 5) that did not undergo ties (45%), compared with the unaffected side.10 Research operations and the surgical group (n = 21) that underwent suggests2,28 that after surgical treatment of pharmacologi- hemispherotomy (8 did not survive the procedure; 10 cally refractory epilepsy, the still immature brain tissue survivors underwent neural tract tracing, and 3 survivors has the potential to make an impressive recovery. Howev- underwent SPECT, comprising 13 rats for the surgical er, with respect to neural reorganization, the mechanisms group). Rats received an injection of the anterograde bio- of brain plasticity are still not well understood. tinylated dextran amine (BDA) neurotracer into the right In the field of experimental neurosurgery, research has cerebral hemisphere. BDA was microinjected into the been conducted to analyze the behavioral and clinical ef- primary motor cortex (M1) or into the nucleus caudatus/ fects of neurosurgical intervention on functional/morpho- putamen (CPu) according to stereotactic coordinates from logical adaptations of neural structures, resulting in neural Paxinos and Watson’s atlas (The Rat Brain in Stereotaxic reorganization and functional recovery after hemispher- Coordinates35) as follows: anteroposterior = 0.12 mm to ectomy and hemidecortication.29,45,49 These studies were M1 and 1.92 mm to the CPu, middle-lateral = 2.0 mm to based on animal models of surgical techniques using ro- M1 and 2.6 mm to the CPu, and dorsoventral = 5.8 mm to dents, felines, and primates (both young and adult labora- M1 and 5.0 mm to the CPu. The surviving animals (n = 4, tory animals). For example, several studies22,33,44 used cats control group; n = 10, surgical group) were killed 10 days and/or rats to investigate experimental models of focal after either neural tract tracing or tracing in addition to cortical seizure, performing amygdalohippocampectomy, hemispherotomy invasive procedures. callosotomy, and multiple subpial transections aimed at As noted above, 4 rats were selected from the groups epilepsy control by surgical intervention. in this study and submitted to functional imaging pro- Experimental neurological approaches investigate the cedures, distributed into two independent experimental complex neurobiological mechanisms involved in epilep- groups: 1) the control group (n = 1) and 2) the surgical togenesis and the maintenance of seizures,25 the develop- group (n = 3). Two weeks after the procedure, rats from ment of new antiepileptic drugs,9,18,20 alterations in synap- both groups (control and surgical) were anesthetized and tic plasticity,26,50 the neurophysiological and neuropharma- underwent the SPECT imaging technique (Fig. 1). cological bases of the postictal immobility syndrome,8,12 the characteristics of ictal and postictal states,31 and post- Neurosurgery: Experimental Hemispherotomy surgical recovery skills.5,40 None of these lines of research, The animals were placed individually in an induction however, are related to the resection of relevant cortical re- chamber (150 × 150 × 300 mm), and anesthesia was in- gions or disconnection of cortical tracts in animals under- duced with 5% isoflurane (Cristália Pharmaceutical and going acute and chronic models of generalized, multifocal, Chemical Products Ltd.) in 100% oxygen with a delivery or limbic epileptic induction and cortical reorganization. rate of 1 L/min until loss of righting reflex. Anesthesia The aim of the present work is to describe a new ex- was then maintained with isoflurane at 100% oxygen with perimental neurosurgical model of hemispherotomy in a flow of 1.5–2.0 L/min, administered by means of endo- young Wistar rats. The new surgical approach not only tracheal intubation and kept under controlled ventilation clarifies neuronal mechanisms underlying behavioral re- (CR-Bion). Each rat was treated with an intramuscular in- covery from early cortical injury, but also may open new jection of penicillin-G benzathine (120,000 UI; 0.1 ml), perspectives on further investigation into the plasticity of followed by subcutaneous injection of the analgesic and corticospinal pathways and motor and sensory posthemi- antiinflammatory flunixin meglumine (2.5 mg/kg), and spherotomy recoveries. was then positioned on a surgical table for small rodents in ventral decubitus. Methods The limits for the craniotomy were as follows: anterior = 2 mm in front of the coronal suture, medial = 0.5 mm Experimental Neurosurgery and Imaging Procedures laterally to the sagittal suture (with preservation of a thin Animals bone surface over the superior sagittal sinus), posterior = Twenty-six male Wistar rats (weighing 260 ± 10 g), with on the surface of the lambdoid suture, and lateral = near a mean age of 45 days, were used in the present study. All the temporal muscle insertion. A C-shaped incision was protocols were used in compliance with the recommenda- made in the dura mater, at the level of the medial line, tions of the Ethics Committee in Animal Experimentation under a surgical microscope (D.F. Vasconcellos S.A.) with

2 J Neurosurg June 8, 2018

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

FIG. 1. A: Flowchart showing the experimental design of the present work. Rats in the control group did not undergo neurosurgery. However, they were evaluated through neural tract tracing and neuroimaging procedures. Surgical intervention corresponds to the animals that underwent experimental hemispherotomy (EH). The EH group was initially composed of 21 animals: 8 died, 10 underwent neural tract tracing, and 3 underwent SPECT. FIG. 1. (continued)→

J Neurosurg June 8, 2018 3

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

FIG. 1. B: Timeline of neuroanatomo-physiological investigation: 12 hours after neural tract tracing the animals were submitted to EH. The follow-up of the clinical evaluation occurred during 10 days, and the acquisition of SPECT images was performed 2 weeks after EH. BDA staining = visualization of interhemispheric neuroconnections; Histology 1 = tissue preparation (brain): perfusion, diaphanization, and paraffin fixation; Histology 2 = brain sectioning for the histochemical procedure. C: Timeline of the rotarod test procedure: habituation in the FMRP-USP (Faculdade de Medicina de Ribeirão Preto–University of São Paulo) Department of Surgery and Anatomy animal facility for 3 days, rotarod test before surgery on the 4th day, experimental hemispherotomy or sham procedure, and rotarod test 10 and 30 days after surgery. a microsurgical hook and scissors, within the craniectomy through a parasagittal plane, moving the leucotomy in- limits. strument from rostral (lesioning the amygdaloid complex) After meticulous hemostasis, using bipolar forceps for to caudal aspects of the temporal lobe, reaching approxi- microsurgery, the hemispherotomy was then conducted mately 1 mm behind the posterior limit of the craniotomy according to the following steps. The disconnection of the (Fig. 2F). frontal lobe was performed 1 mm in front of the coronal After a careful reexamination of the hemostasis (model suture projection, using a round-tip dissector, and the fron- HF-120, WEM Electronic Equipment Ltd.), the dura ma- tal lobe was disconnected immediately in front of the an- ter flap was repositioned and the skin was sutured with terior commissure to the floor of the skull base (Fig. 2A). 4-0 nylon wire (black NP53340, Polysuture). At the end The cerebral parenchyma was resected by aspiration, of the surgical procedure, intubation was maintained in which was performed starting at the site of the mechani- the experimental animal for oxygenation in an automatic cal disconnection of the frontal lobe. The deeper limit of breathing machine. the resection was determined by the appearance of sub- Each animal was transferred to its original cage after telencephalic structures and the archeocortex. In a similar the anesthesia recovery period, and was observed daily to manner, the anterior and lateral limits of the resection were record clinical condition, food and water intake, behavior, the anterior horn of the lateral ventricle and the CPu. The and neurological conditions. posterior limit of the cortical ablation was achieved when the resection reached the surface of the dorsal hippocam- SPECT Imaging pus. Using this surgical approach, the surgeon maintained Independent groups of Wistar rats comprising 1 rat the integrity of these structures (see Fig. 2B for details). from the control group and 3 rats from the hemispheroto- After the parenchymal resection, the microscope was my group were submitted to the SPECT imaging protocol. used to view the surface of the corpus callosum (ana- Two weeks after surgery, brain perfusion was evaluated in tomical landmarks: anterior, caudate-putamen and lateral those animals by 99mTc hexamethylpropyleneamine oxime ventricle; posterior, hippocampus) to position the instru- (HMPAO) SPECT imaging. HMPAO was obtained as a ment used for the leucotomy (12 mm in length, 0.25 mm lyophilized commercially available kit and reconstituted in diameter, with a sharpened and 10° bent tip), placing with 99mTc pertechnetate according to the manufacturer’s the cutting edge parallel to the direction of the body of instructions. Each animal was anesthetized with ketamine the corpus callosum. The corpus callosum (body) was sectioned along the rostrocaudal axis, following the curvature at 90 mg/kg (0.2 ml of 10% solution) and xylazine at 9 mg/kg (0.1 ml; Dopaser, Hertape/Calier) and received 350 of this commissure, positioning the tip of the instrument 99m slightly upward. The anterior limit of the disconnection MBq of Tc HMPAO, peripherally injected though the (toward the genu of the corpus callosum) was set at 2 mm superficial dorsal vein of the penis. For image acquisition, anterior to the craniotomy, and the posterior limit (aiming we used an upgraded clinical gamma camera equipped at the splenium of the corpus callosum) was set at 1 mm with a 1.5-mm-diameter pinhole collimator and a comput- posterior to the craniotomy, positioning the leucotomy in- er-controlled rotating stage for target positioning. Projec- strument slightly downward (Fig. 2C). tions were recorded beginning shortly after radiopharma- After resection of the cerebral parenchyma, removal of ceutical injection. Animals were placed in a vertical posi- subcortical white matter, and disconnection of the corpus tion into the animal holder, with the head in front of the callosum, the dorsal hippocampus was identified and re- pinhole collimator, and 40 projections of 45 seconds each sected by aspiration. The hippocampectomy was carefully were recorded, for a total imaging time of 30 minutes. performed, up to visual identification of the dorsal thala- The energy window was set to 140 keV ± 10%, and the mus in the floor of the lateral ventricle. After visual identi- collimator-to-detector and collimator-to-rotation axis dis- fication, the fornix was also carefully sectioned (Fig. 2D). tances were set to 315 mm and 43 mm, respectively. The The external capsule disconnection was performed projections were recorded in 256 × 256 pixels and resam- with a microsurgical dissector, following the external cap- pled to 128 × 128 pixels for processing. The tomographic sule laterally to the caudate-putamen. The region of the images were obtained using an in-house implemented ver- frontal disconnection was considered the anterior limit of sion of the ordered subset expectation maximization al- this procedure. The posterior limit of this leucotomy was gorithm, with 10 iterations, 4 subsets of projections, and a considered to be the lambdoid suture (dorsally) and the 3D smoothing operation with a 1.5-voxel-width Gaussian lateral ventricle (laterally). The inferior limit of the exter- kernel between iterations.30 nal capsule leucotomy was the skull base (Fig. 2E). The dissection instrument was inserted deep into the Neural Tract Tracing brain, following the lateral wall of the lateral ventricle Animals were anesthetized with ketamine (90 mg/kg

4 J Neurosurg June 8, 2018

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

FIG. 2. Steps of the neurosurgical procedure (A–F). A: Sagittal encephalic view: disconnection of the frontal lobe and anterior commissure (arrowheads). B: Sagittal view: removal of cortex and subcortical white matter tissue block (dashed line). C: Cal- losotomy: corpus callosum and anterior commissure disconnection (arrowhead and dashed line). D: Sagittal and posterior coronal section of the brain, showing dorsal hippocampectomy (dashed line). E: Disconnection of external capsule (dashed line). F: Anteroposterior leucotomy from the amygdaloid complex to the occipital lobe. Parasagittal (G and I) and transverse (H, J, and K) sections of the rat encephalon showing representative histological views of the following surgical procedures: frontal disconnection and cortical resection (G); cortical resection with hippocampal preservation (H); calosotomy (I); cortical resection, dorsal hippocampectomy, and disconnection of the external capsule (J); and posterior leucotomy (K). Bars = 1000 mm. intraperitoneally) and xylazine (9 mg/kg intraperitone- jection to neighboring areas.6,39 Briefly, micropipettes were ally) and fixed in a stereotactic frame apparatus (David made of a fused silica capillary (external diameter 150 Kopf). The upper incisor bar was set 3.0 mm below the mm, internal diameter 75 mm; Cluzeau Info Labo C.I.L.). interaural line so that the skull was horizontal between the The capillary was fixed within a device made with needles bregma and lambda. For the control group, the scalp was of 0.60 × 25 mm and 1.00 × 25 mm (Becton-Dickinson) anesthetized with 2% lidocaine and then surgically ex- to prevent cannula breaks. The micropipette was attached posed. In the experimental surgery group, an incision was to a 5-ml Hamilton syringe by polyethylene tube and intro- made in the midline with a no. 15 blade scalpel, cutting duced through the guide cannula. The displacement of an the skin and subcutaneous tissue. The nonfluorescent an- air bubble inside the polyethylene tube that connected the terograde neurotracer BDA was microinjected in the limb syringe to the injection needle was used to monitor the mi- area according to a stereotactic atlas.35 Rats subsequently croinjections. Injection micropipettes were left in place for received an intramuscular injection of penicillin-G benza- 2 minutes after the end of each microinjection to allow for thine (600,000 IU, 0.2 ml; Fort Dodge) and a subcutane- local drug diffusion. In the group undergoing surgery, the ous injection of the antiinflammatory and analgesic drug anterograde neurotracer (BDA, 10,000 MW) was injected flunixin meglumine (2.5 mg/kg, Schering-Plough). into the motor cerebral cortex 12 hours before the experi- The BDA (10%, either 10,000 or 3000 MW; Molecu- mental hemispherotomy, and the bidirectional neurotracer lar Probes) was dissolved in 0.01 M phosphate buffer (pH (BDA, 3000 MW) was microinjected into the CPu 72 7.4). Microinjections were performed under anesthesia, hours after the hemispherotomy. during surgery. Infusions were delivered using an infusion pump (Harvard Apparatus) at a volume of 0.5 ml BDA Histological Analysis over the course of 5 minutes. The drug dose and injection Rats were deeply anesthetized with urethane (3 g/kg time were based on previous studies.7,38,47 We used a glass intraperitoneally, Sigma-Aldrich) 10 days after neural micropipette for microinjections according to a technique tract tracing (control group) or neural tract tracing in ad- described elsewhere to minimize the spreading of the in- dition to hemispherotomy procedures (surgery group),

J Neurosurg June 8, 2018 5

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al. and transcardially perfused with 0.9% buffered saline fol- test. Normally distributed data underwent a repeated mea- lowed by 4% paraformaldehyde (Sigma-Aldrich) in 0.1-M sures 2-way ANOVA followed by Bonferroni’s post hoc phosphate buffer (pH 7.4). Subsequently, the cerebrum was test to compare the different groups of rats and follow-up removed en bloc through a vertex craniectomy and refixed periods, with the objective of evaluating the significant in 70% ethanol. Brains were then sectioned to obtain an effect of the type of injury (treatment) in different time anterior and posterior division, using a coronal plane pass- windows. In all these procedures, the statistical program ing through the cortical neurotracing point as a reference. GraphPad Prism (version 5.0) was used, and p values < Each brain section was dehydrated in a graded alcohol 0.05 were considered statistically significant. series, diaphanized in xylene, and embedded in paraffin. The brain divisions were coronally cut into 10-μm-thick Results sections and mounted on histological slides. BDA labeling was visualized using a standard avidin biotinylated horse- Experimental Neurosurgical and Imaging Procedures radish peroxidase method and a nickel-intensified per- Twenty-six rats were used in the present experimen- oxidase 3,3′-diaminobenzidine dihydrochloride reaction. tal neurosurgery study: 5 nonoperated rats were used as Slides were thoroughly washed in 0.1 M phosphate buffer controls and 21 rats underwent hemispherotomy surgery. (pH 7.4) after incubation, mounted on gelatin-coated glass Among the 21 operated animals, 13 survived to the end of slides, counterstained using the Nissl cresyl violet method the experiment and were used for structural (neural tract in a robotized autostainer (CV 5030, Leica Autostainer tracing, n = 10) and functional (SPECT, n = 3) neuroanato- XL), and viewed under a photomicroscope (AxioImager mo-physiological investigation. The other 8 operated rats Z1, Zeiss). The brain areas were analyzed and the histolog- died within the first 24 hours of surgery due to profuse ically confirmed sites of intracerebral neurotracer deposits intraoperative bleeding. The mean duration of the hemi- were identified according to Paxinos and Watson’s atlas.35 spherotomy was 50 ± 10 minutes. In the immediate postoperative period (i.e., the first Behavioral Study 24 hours), the following features of the hemispherotomy Animals procedure were recorded: tremors (8%, n = 1), bleeding through the skin (31%, n = 4), suffusion at the site of sur- All behavioral experiments were performed on young gery (69%, n = 9), decreased motor activity (85%, n = 11), adult male Wistar rats. The rats (n = 30) were randomly as- palpebral ptosis (38%, n = 5), increased muscular tonus of signed to 1 of 3 groups: control (10 rats), sham (craniotomy the dorsal muscles (69%, n = 9), and discrete bleeding in performed with an electric drill, 10 rats), and experimental the eyes and nose (15%, n = 2). neurosurgical groups (hemispherotomy, 10 rats). The sham On the 2nd day of follow-up, the following were ob- and control groups did not undergo brain injury. served: a decrease in water and food intake (85%, n = 11), Craniotomy Procedures decreased motor activity (85%, n = 11), suffusion at surgery focus (69%, n = 9), aggressiveness (69%, n = 9), dys- The sham rats underwent the same sedation procedure pnea (8%, n = 1), palpebral ptosis (38%, n = 5), increased for the experimental hemispherotomy described above. muscular tonus of the dorsal muscles (85%, n = 11), and The limits of the craniotomy were the same as the neu- discrete bleeding in the eyes and nose (15%, n = 2). rosurgical intervention; the osseous flap was removed, the On the 3rd day of clinical observation, water and food dura mater was preserved, and the skin was sutured. intake decreased in 69% (n = 9), 85% showed grooming Rotarod Test neglect (n = 11), 85% showed decreased muscle tone (n = 11), 69% of operated animals demonstrated aggressiveness We used the modified procedure for the standard ro- (n = 9), 31% showed suffusion at the surgical wound (n = tarod test to emphasize motor skill learning, coordination, 4), and 46% showed a decrease in motor activity (n = 6). 17,21 and balance. We used an automated device (acrylic box, Also observed was palpebral ptosis (23%, n = 3), increased 530 × 400 × 410 mm in height, width, and length, respec- muscular tonus of the dorsal muscles (85%, n = 11), and tively; model EFF 411, Insight) and 3 experimental groups discrete bleeding in the eyes and nose (15%, n = 2). (control, sham surgical procedure, and hemispherotomy On the 4th day after the hemispherotomy, the animals groups) were submitted to the rotarod test without previ- showed reduced leg movements on the contralateral side to ous training to remain 120 seconds on the rolling rod. the surgery, difficulty in initiating movements, and unilat- Acceleration of the rotation was fixed and was set at a eral ambulation (walking to the right). relatively slow speed (8 rpm). For the test, evaluations were On the 5th day after surgery, the animals displayed ex- performed preoperatively and then 10 and 30 days after ploratory behavior in the home cage. However, there was neurosurgery. The number of falls experienced by the rat paresis contralateral to the surgery, they dragged their par- during the procedure was recorded, and the animal was alyzed legs, there was increased spinal curvature, and they placed back on the rolling rod immediately after falling. lost walk balance. The falls were recorded until the rat remained on the roll- The anterograde neurotracer microinjected into the ing rod for 120 seconds. A timeline of this procedure is cerebral cortex reached both the external and internal provided in Fig. 1C. pyramidal layers of the motor cortex area (Fig. 3). In the control group, BDA-labeled corticostriatal pathways were Statistical Analysis found ipsilateral (Fig. 4A) to the microinjection sites of Data were analyzed using the Shapiro-Wilk normality BDA, and BDA-labeled cortico-cortical neural pathways

6 J Neurosurg June 8, 2018

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

FIG. 3. Schematic view of the neuroanatomical procedure (neural tract tracing). A: Upper shows BDA anterograde neurotracer microinjections (arrowhead) in Rattus norvegicus M1 area of the neocortex (right hemisphere). Lower shows transverse section of Wistar rat brain demonstrating, in a schematic view, reciprocated neural pathways connecting both brain hemispheres through the corpus callosum. The photomicrograph on the left represents a histologically confirmed site of BDA microinjection in the telencephalon of the Wistar rat. B: Photomicrograph of a coronal section of the telencephalon of the Wistar rat showing a histologically confirmed site of intratelencephalic microinjection of the anterograde BDA neurotracer in the M1 cortical area. C: Representa- tion of histologically confirmed sites of BDA microinjections in rat telencephalon of control and hemispherotomy-treated animals depicted in modified diagrams from the rat stereotactic atlas of Paxinos and Watson.

J Neurosurg June 8, 2018 7

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

FIG. 4. Photomicrographs of coronal sections of the telencephalon of a representative control Wistar rat, showing cortico-cortical pathways. A: Cortical fugal pathways (arrow) originated from the right brain hemisphere ipsilateral to the site of microinjection of the anterograde BDA neurotracer in the M1 cortical area, also sending inputs to the ipsilateral neostriatum (arrowheads). B: Intertelencephalic cortical pathways (arrowheads) crossing the midline in the corpus callosum (CC). C: Enhanced cortical fugal pathways (arrowheads) originating from the right brain hemisphere ipsilateral to the site of microinjection of BDA in the motor cortical area crossing the midline in the corpus callosum. D: Coronal sections of the left brain hemisphere showing BDA-labeled fibers arrowheads( ) reaching the multiform cortical layer contralateral to the site of microinjections of the anterograde neurotracer, surrounding local neurons. were widely found crossing the corpus callosum (Fig. Behavioral Study 4B and C), reaching the contralateral cerebral cortex, in According to the 2-way ANOVA, there was a signifi- which terminal-connecting BDA-labeled fibers surround cant effect of treatment (F[34,24] = 2, p < 0.0001), time neuronal bodies situated in the multiform cortical layer (F[2,19] = 27, p < 0.007), and treatment versus time in- (Fig. 4D). teraction (F[34,24] = 21.26, p < 0.0001) in the behavioral However, in rodents undergoing hemispherotomy, mi- study. Although in the preoperative period (control, sham croinjections of the anterograde neurotracer into the pri- procedure, or experimental surgery groups) motor perfor- mary motor cortex showed BDA-labeled corticofugal fi- mance was significantly affected by brain injuries, there bers (Fig. 5A and B) ipsilateral to the intracortical deposits was an increase in the number of attempts to maintain bal- of BDA sending ipsilateral projections to the neostriatum ance on the moving cylinder in the rotarod test at 10 days (Fig. 5A). Although there were labeled fibers in the corpus (Bonferroni’s post hoc test, p < 0.001) and 30 days (Bon- callosum (Fig. 5C), there was a rarefaction of interhemi- ferroni’s post hoc test, p < 0.001) after the hemispherotomy spheric projections (Fig. 5B) crossing the midline through compared with the control groups (Fig. 8). However, there the corpus callosum (Fig. 5C), with consequent absence was a tendency to recover the functional deficit over time. of BDA-labeled inputs to the contralateral cortical hemisphere (Fig. 5D), suggesting a clear interruption of cortico-cortical pathways disconnected by the experimental Discussion hemispherotomy procedure. Ipsilateral cortico-neostriato- The present experimental neurosurgical approach re- nigral pathways and the nigro-tectal pathway are examples sulted in a successful hemispherotomy in all animals un- of preserved connections in the hemispherotomy group dergoing neurosurgery, with a survival rate of 70% and (Fig. 6). good postsurgical recovery. The hemispherotomy was SPECT tomographic images of 99mTc HMPAO distribu- confirmed by neural tract tracing and SPECT procedures. tion were obtained to verify perfusion in functioning areas The anterograde neurotracer, unilaterally microinjected of the brain. In control animals, brain perfusion was uni- into the motor cortex, labeled both cortico-neostriatal and form through both brain hemispheres, as shown in hori- interhemispheric pathways in the control group, but there zontal and coronal slices (Fig. 7A upper, 7B). However, in was a clear rarefaction of BDA-labeled fibers crossing the the animals that underwent hemispherotomy, hypoperfu- midline in Wistar rats with hemispherotomy, showing sion was observed in the hemisphere ipsilateral to the sur- the efficiency of the surgical brain disconnection. This gery, as shown in horizontal and coronal slices (Fig. 7A structural evidence was supported by SPECT studies that lower, 7C–E). showed a clear impairment in the perfusion of the ipsilat-

8 J Neurosurg June 8, 2018

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

FIG. 5. Coronal brain sections of a representative hemispherotomy-treated Wistar rat. A: Intrahemispheric cortico-neostriatal connections (arrowheads) and cortico-cortical pathways (arrow) originating from the right brain hemisphere ipsilateral to the microinjection site of anterograde BDA neurotracer in the M1 cortical area. B: BDA-labeled interhemispheric neural pathways (arrow) reaching the corpus callosum ipsilateral to the microinjection site performed in the M1 cortical area. C: Rarefaction of BDA-labeled fibers reaching the site of disconnection arrowheads( ) of the corpus callosum near the midline. D: Note the lack of either BDA-labeled axonal fibers or neuronal perikarya in the contralateral cerebral cortex layers. eral brain hemisphere in the hemispherotomy group. The ample, cerebral convolutions that are absent in the murine structural and functional neuroanatomical approaches brain are more prominent in primates. In terms of com- combined with detailed neurosurgical interventions high- parative anatomy, the rat brain developed in the longitudi- light the efficiency of this procedure, with a clear potential nal plane (sagittal and frontal aspects). In that context, it is for the study of pharmacologically refractory epilepsy in possible to perform the disconnections of the subcortical laboratory animals. structures, visualizing some encephalic structures from Catastrophic epileptic syndromes have a highly unfa- the midline to parasagittal planes to determine relevant vorable outcome both in the evolution of the disease and anatomical references, such as the lateral ventricle, the in the aftereffects of treatments. Since the initial proposal CPu that is located posterior to the anterior commissure, for functional hemispherectomy by Rasmussen in 1983,37 and the corpus callosum.1 surgical procedures with more restricted cortical resection The encephalic structures reached in the experimental associated with neural tract disconnections have been the surgical approach used in this work were lesioned/sec- best choice of treatment for affected patients.48,51 On the tioned to simulate the clinical approach (hemispherotomy) other hand, the processes involved in the reorganization of performed in humans to achieve complete disconnection the highly modified brain areas after surgery still require of the brain hemisphere. This approach is as follows: 1) further clarification. internal disruption of the internal capsule and corona ra- One way to study the lesions resulting from a disease diata, 2) resection of the medial temporal structures, 3) and its treatments is to use animal models of surgical in- callosotomy, and 4) disruption of the frontal horizontal terventions. Thus far, experimental models that simulate fibers.16,32,48 surgeries used in the treatment of pharmacologically re- The possible functional recovery that occurs in patients fractory epilepsy are hemispherectomy and hemidecor- with CNS injury can also be observed in animal models.32 tication performed in neonates and in young adult labo- Thus, experimental studies are also performed to better ratory animals.23,24 In the present study, we describe a understand the mechanisms of neuroplasticity involved technique that involves a vertical approach, similar to the in the functional recovery of these patients.28,29,34,45,49 We parasagittal vertical approach demonstrated by Delalande observed that animals submitted to experimental hemi- et al.14–16 as a modified procedure of functional hemispher- spherotomy displayed a motor deficit 10 days after the ectomy proposed by Rasmussen.37 We adopted the vertical surgical procedures. However, there was a significant im- approach due to the structural anatomy of the rat brain. provement in motor coordination and balance displayed by Although human and rat brains have some similari- these animals in the rotarod test 30 days after the hemi- ties, the primitive brain is more developed in rats and the spherotomy, suggesting brain neuroplasticity. The func- neocortex is more developed in humans, in which, for ex- tional deficits were mainly presented through changes in

J Neurosurg June 8, 2018 9

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

FIG. 6. Coronal encephalic sections of a representative hemispherotomy-treated Wistar rat, showing preserved ipsilateral connections. A–C: The cortico-neostriatal pathway, with neurons (black arrows) situated in the internal pyramidal layer (A and B) sending axons (arrowheads) to the CPu (C). D: The neostriato-nigral GABAergic projection, represented by axons and terminal buttons (arrowheads) situated in the substantia nigra pars reticulata (SNpr). E and F: The nigro-striatal dopaminergic pathway, represented by neurons (arrows) situated in the substantia nigra pars compacta (SNpc) connected to the CPu (black arrowheads indicate neurites). The arrow in C shows a representative site of the BDA bidirectional neurotracer (BDA, 3000 MW) microinjections in the neostriatum. The neuronal body (arrow) shown at the bottom of panel D is situated in the SNpc. Axons (black arrowheads) and puncta (open arrowheads) suggesting terminal buttons shown in panel F on the right are situated in the SNpr. Counterstaining was conducted using Nissl cresyl violet. motor coordination causing more falls from the moving axonal reorganization of the remaining neurons in corti- rod, possibly associated with difficulties in using sensory cal territories ipsilateral to the lesions and the correlation information, and postural adjustments related to decreases of these neural substrates with motor recovery based on in general physical conditions.41,52 the development of new sprouting and axonal projections. Correlating our data with data from other studies, we Comparing our findings with those already described in observed that the motor recovery of the animals undergo- other hemispherectomy and hemidecortication models, we ing the experimental hemispherotomy may be related to observed a similarity to the findings obtained by Machado cortical/subcortical reorganization3,4 and functional recov- et al.28 regarding the total functional recovery period (less ery. 28,29,41 The functional recovery shown in the rats under- than 2 weeks). Marino et al.29 determined that the quality going hemispherotomy in the present work is similar to of functional recovery is related to the age of the rat at the that reported elsewhere.19,34,53 In addition, the neurosurgical time of surgery; i.e., younger rats have a greater chance of procedure preserved some critical connections subserving complete recovery. We used young rats, similar in weight the motor function, such as the cortico-neostriatum-nigral to the animals used by Marino et al.,29 and we obtained pathways and the nigro-striatal projections. similar results in terms of functional recovery. The present study proposes an experimental model that In the present experimental neurosurgical approach, we provides an opportunity to study the mechanisms of neu- observed a mortality rate of 30%, with death always oc- roplasticity in the brain hemisphere, which is contralateral curring within the first 24 hours of the procedure. As 71% to resection (preserved structures), but also considers some of the deaths occurred in the first half of the total number

10 J Neurosurg June 8, 2018

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

FIG. 7. SPECT images of the brain blood perfusion tracer 99mTc HMPAO in 1 normal and 3 hemispherotomy-treated rats. A: Representative horizontal and coronal sections through the brain images of the normal and one of the hemispherotomy-treated animals. B–E: Horizontal sequence of coronal sections through the brain images of the normal and 3 hemispherotomy-treated animals.

FIG. 8. Effects of the hemispherotomy on motor performance of Wistar rats in the rotarod test during a 30-day follow-up. Data are presented as mean ± standard error of the mean (n = 10 per group). *p < 0.05, **p < 0.01 compared with the control group (naïve rats); #p < 0.05, ##p < 0.01 compared with the sham procedure group, according to the repeated 2-way ANOVA, followed by Bonferroni’s post hoc test. PreOp = preoperative/baseline responses.

J Neurosurg June 8, 2018 11

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al. of operated animals, these deaths may be related to the recover better? Neurobehavioural plasticity after early brain learning/training process of surgical procedures. Further- insult. Brain 134:2197–2221, 2011 more, previous studies do not provide details about the 3. Axelson HW, Winkler T, Flygt J, Djupsjö A, Hånell A, Marklund N: Plasticity of the contralateral motor cortex fol- mortality rate of their procedures or the timing of those lowing focal traumatic brain injury in the rat. Restor Neurol deaths,29 nor do they mention the occurrence of death after 29,46,55 Neurosci 31:73–85, 2013 surgery. 4. Brus-Ramer M, Carmel JB, Martin JH: Motor cortex bilat- Advantages of the experimental model of hemispherot- eral motor representation depends on subcortical and inter- omy proposed here include the opportunity to study the hemispheric interactions. J Neurosci 29:6196–6206, 2009 morphology and functional characteristics of novel neural 5. Bulteau C, Jambaqué I, Chiron C, Rodrigo S, Dorfmüller G, connections arising from adaptive neuroplasticity in the Dulac O, et al: Language plasticity after hemispherotomy of disconnected brain through the emergence of new subcor- the dominant hemisphere in 3 patients: implication of non- linguistic networks. Epilepsy Behav 69:86–94, 2017 tical projections and their inputs to the ventral horn of the 6. Carvalho MC, Santos JM, Bassi GS, Brandão ML: Participa- spinal cord. In addition, we should highlight the functional tion of NK1 receptors of the amygdala on the processing of role played by neuronal activity in the basal nuclei, brain- different types of fear. Neurobiol Learn Mem 102:20 –27, stem-descending reticular-spinal projections, and cerebel- 2013 lum in recovering animals; neuronal and myelin plastic- 7. Castellan-Baldan L, da Costa Kawasaki M, Ribeiro SJ, Calvo ity;11 electrical stimulation;27,54 and neurophysiological F, Corrêa VMA, Coimbra NC: Topographic and functional mechanisms42 previously suggested by several research- neuroanatomical study of GABAergic disinhibitory striatum- ers.34,43,46,49 nigral inputs and inhibitory nigrocollicular pathways: neural hodology recruiting the substantia nigra, pars reticulata, for The current model of hemispherotomy also enables the the modulation of the neural activity in the inferior collicu- development of new rehabilitation strategies in rodents un- lus involved with panic-like emotions. J Chem Neuroanat dergoing experimental hemispherotomy for the improve- 32:1–27, 2006 ment of postsurgical recovery after hemispheric injuries, 8. Coimbra NC, Freitas RL, Savoldi M, Castro-Souza C, Segato and it allows us to describe the neurological bases of this EN, Kishi R, et al: Opioid neurotransmission in the postsurgical therapeutic intervention in children with pharma- ictal analgesia: involvement of m1-opioid receptor. Brain Res cologically refractory epilepsy. 903:216–221, 2001 9. Cunha AO, Mortari MR, Oliveira L, Carolino RO, Coutinho- Netto J, dos Santos WF: Anticonvulsant effects of the wasp Conclusions Polybia ignobilis venom on chemically induced seizures and The new proposed experimental model of hemispherot- action on GABA and glutamate receptors. Comp Biochem Physiol C Toxicol Pharmacol 141:50–57, 2005 omy, shown to be technically feasible and comparable to 10. de Bode S, Firestine A, Mathern GW, Dobkin B: Residual procedures currently used to surgically treat patients with motor control and cortical representations of function follow- pharmacologically refractory epilepsy, is suitable for many ing hemispherectomy: effects of etiology. J Child Neurol future studies focused on experimental neurosurgical and 20:64–75, 2005 rehabilitation areas. 11. de Faria O Jr, Pama EAC, Evans K, Luzhynskaya A, Kara- dottir RT: Neuroglial interactions underpinning myelin plasticity. Dev Neurobiol 78:93–107, 2018 Acknowledgments 12. de Freitas RL, Medeiros P, da Silva JA, de Oliveira RC, This study was supported by CNPq (grant nos. 483763/2010-1 de Oliveira R, Ullah F, et al: The m1-opioid receptor and and 474853/2013-6), FAEPA (grant nos. 1291/97, 355/2000, 68/2001, 5-HT2A- and 5HT2C-serotonergic receptors of the locus 15/2003, and 423/2017), and FAPESP (grant nos. 2007/01174-1, coeruleus are critical in elaborating hypoalgesia induced by 2012/03798-0, and 2017/11855-8). I. Matias Jr. was supported tonic and tonic-clonic seizures. Neuroscience 336:133–145, by CAPES (Scientiae Doctor Fellowship). N.C. Coimbra is a 2016 researcher (Level 1A) from CNPq (grant nos. 301905/2010-0 and 13. De Ribaupierre S, Delalande O: Hemispherotomy and other 301341/2015-0). D.H. Elias-Filho was the recipient of technician disconnective techniques. Neurosurg Focus 25(3):E14, 2008 scholarships from FAPESP (TT-2, grant no. 02/01497-1) and CNPq 14. Delalande O, Bulteau C, Dellatolas G, Fohlen M, Jalin C, (grant nos. 501858/2005-9, 372654/2006-1, 372810/2008-0, and Buret V, et al: Vertical parasagittal hemispherotomy: surgical 372877/2010-9). procedures and clinical long-term outcomes in a popula- The authors are thankful for the expert technical assistance of tion of 83 children. Neurosurgery 60 (2 Suppl 1):ONS19– Vani Maria Alves Correa (Histotechnology Laboratory, FMRP- ONS32, 2007 USP Department of Cellular and Molecular Biology and Patho- 15. Delalande O, Dorfmüller G: [Parasagittal vertical hemi- genic Bioagents), Sandra L. Balero Penharvel Martins (Histology spherotomy: surgical procedure.] Neurochirurgie 54:353– and Immunohistochemistry Laboratory, FMRP-USP Department 357, 2008 (Fr) of Surgery and Anatomy), and Daniel Mazzeto (Laboratory of Sur- 16. Delalande O, Pinard JM, Basdevant C: Hemispherotomy: gical Techniques and Experimental Surgery, FMRP-USP Depart- a new procedure for central disconnection. Epilepsia 33 ment of Surgery and Anatomy). (Suppl 3):99–100, 1992 (Abstract) 17. Dunham NW, Miya TS: A note on a simple apparatus for detecting neurological deficit in rats and mice. J Am Pharm Assoc 46:208–209, 1957 References 18. Fachim HA, Cunha AO, Pereira AC, Beleboni RO, Gob- 1. Akakin A, Yılmaz B, Akakin D, Dagbasi N, Kilic T: Three bo-Neto L, Lopes NP, et al: Neurobiological activity of dimensional anatomical microdissection of rat brain using Parawixin 10, a novel anticonvulsant compound isolated fiber dissection technique. J Anat Soc India 63:117–124, from Parawixia bistriata spider venom (Araneidae: Araneae). 2014 Epilepsy Behav 22:158–164, 2011 2. Anderson V, Spencer-Smith M, Wood A: Do children really 19. Feeney DM, Boyeson MG, Linn RT, Murray HM, Dail WG:

12 J Neurosurg June 8, 2018

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al.

Responses to cortical injury: I. Methodology and local effects structural neuroanatomy of interactive intratectal/tectonigral of contusions in the rat. Brain Res 211:67–77, 1981 mesencephalic opioid inhibitory links and nigrotectal GA- 20. Gelfuso EA, Liberato JL, Cunha AO, Mortari MR, Bele- BAergic pathways: involvement of GABAA and m1-opioid boni RO, Lopes NP, et al: Parawixin2, a novel non-selective receptors in the modulation of panic-like reactions elicited by GABA uptake inhibitor from Parawixia bistriata spider ven- electrical stimulation of the dorsal midbrain. J Chem Neuro- om, inhibits pentylenetetrazole-induced chemical kindling in anat 30:184–200, 2005 rats. Neurosci Lett 543:12–16, 2013 39. Santos JM, Macedo CE, Brandão ML: GABAergic mecha- 21. Hamm RJ, Pike BR, O’Dell DM, Lyeth BG, Jenkins LW: The nisms of hypothalamic nuclei in the expression of condi- rotarod test: an evaluation of its effectiveness in assessing tioned fear. Neurobiol Learn Mem 90:560–568, 2008 motor deficits following traumatic brain injury. J Neurotrau- 40. Save-Pédebos J, Pinabiaux C, Dorfmuller G, Sorbets SF, ma 11:187–196, 1994 Delalande O, Jambaqué I, et al: The development of pragmat- 22. Hashizume K, Tanaka T: Multiple subpial transection in ic skills in children after hemispherotomy: Contribution from kainic acid-induced focal cortical seizure. Epilepsy Res left and right hemispheres. Epilepsy Behav 55:139–145, 32:389–399, 1998 2016 23. Hicks SP, D’Amato CJ: Motor-sensory and visual behavior 41. Shiotsuki H, Yoshimi K, Shimo Y, Funayama M, Takamatsu after hemispherectomy in newborn and mature rats. Exp Y, Ikeda K, et al: A rotarod test for evaluation of motor skill Neurol 29:416–438, 1970 learning. J Neurosci Methods 189:180–185, 2010 24. Hicks SP, D’Amato CJ: Motor-sensory cortex-corticospinal 42. Song M, Martinowich K, Lee FS: BDNF at the synapse: why system and developing locomotion and placing in rats. Am J location matters. Mol Psychiatry 22:1370–1375, 2017 Anat 143:1–42, 1975 43. Takahashi M, Vattanajun A, Umeda T, Isa K, Isa T: Large- 25. Leite JP, Bortolotto ZA, Cavalheiro EA: Spontaneous recur- scale reorganization of corticofugal fibers after neonatal rent seizures in rats: an experimental model of partial epi- hemidecortication for functional restoration of forelimb lepsy. Neurosci Biobehav Rev 14:511–517, 1990 movements. Eur J Neurosci 30:1878–1887, 2009 26. Lenz M, Ben Shimon M, Deller T, Vlachos A, Maggio N: 44. Tanaka T, Hashizume K, Sawamura A, Yoshida K, Tsuda H, Pilocarpine-induced status epilepticus is associated with Hodozuka A, et al: Basic science and epilepsy: experimental changes in the actin-modulating protein synaptopodin and epilepsy surgery. Stereotact Funct Neurosurg 77:239–244, alterations in long-term potentiation in the mouse hippocam- 2001 pus. Neural Plast 2017:2652560, 2017 45. Umeda T, Funakoshi K: Reorganization of motor circuits 27. López-Alonso V, Cheeran B, Fernández-del-Olmo M: Re- after neonatal hemidecortication. Neurosci Res 78:30 –37, lationship between non-invasive brain stimulation-induced 2014 plasticity and capacity for motor learning. Brain Stimul 46. Umeda T, Isa T: Differential contributions of rostral and 8:1209–1219, 2015 caudal frontal forelimb areas to compensatory process after 28. Machado AG, Shoji A, Ballester G, Marino R Jr: Mapping of neonatal hemidecortication in rats. Eur J Neurosci 34:1453– the rat’s motor area after hemispherectomy: the hemispheres 1460, 2011 as potentially independent motor brains. Epilepsia 44:500– 47. Veenman CL, Reiner A, Honig MG: Biotinylated dextran 506, 2003 amine as an anterograde tracer for single- and double-label- 29. Marino R Jr, Machado AG, Timo-Iaria C: Functional recov- ing studies. J Neurosci Methods 41:239–254, 1992 ery after combined cerebral and cerebellar hemispherectomy 48. Villemure JG, Mascott CR: Peri-insular hemispherotomy: in the rat. Stereotact Funct Neurosurg 76:83–93, 2001 surgical principles and anatomy. Neurosurgery 37:975–981, 30. Mejia J, Miranda AC, Durante AC, Oliveira LR, Barboza 1995 MR, Rosell KT, et al: Preclinical molecular imaging: develop- 49. Wanakhachornkrai O, Umeda T, Isa K, Tantisira MH, Tan- ment of instrumentation for translational research with small tisira B, Isa T: Reorganization of sensory pathways after laboratory animals. Einstein (Sao Paulo) 14:408–414, 2016 neonatal hemidecortication in rats. Neurosci Res 79:94–98, 31. Merricks EM, Smith EH, McKhann GM, Goodman RR, 2014 Bateman LM, Emerson RG, et al: Single unit action po- 50. Willemse RB, Hillebrand A, Ronner HE, Vandertop WP, tentials in humans and the effect of seizure activity. Brain Stam CJ: Magnetoencephalographic study of hand and foot 138:2891–2906, 2015 sensorimotor organization in 325 consecutive patients evalu- 32. Morino M, Shimizu H, Ohata K, Tanaka K, Hara M: Ana- ated for tumor or epilepsy surgery. Neuroimage Clin 10:46– tomical analysis of different hemispherotomy procedures 53, 2015 based on dissection of cadaveric brains. J Neurosurg 51. Winston KR, Welch K, Adler JR, Erba G: Cerebral hemicor- 97:423–431, 2002 ticectomy for epilepsy. J Neurosurg 77:889–895, 1992 33. Ono T, Fujimura K, Yoshida S, Ono K: Suppressive effect of 52. Xerri C, Benelhadj M, Harlay F: Deficits and recovery of callosotomy on epileptic seizures is due to the blockade of body stabilization during acrobatic locomotion after focal le- enhancement of cortical reactivity by transcallosal volleys. sion to the somatosensory cortex: a kinematic analysis com- Epilepsy Res 51:117–121, 2002 bined with cortical mapping. Arch Ital Biol 142:217–236, 34. Otte WM, van der Marel K, van Meer MP, van Rijen PC, 2004 Gosselaar PH, Braun KP, et al: Altered contralateral senso- 53. Xiong Y, Mahmood A, Chopp M: Animal models of trau- rimotor system organization after experimental hemispherec- matic brain injury. Nat Rev Neurosci 14:128–142, 2013 tomy: a structural and functional connectivity study. J Cereb 54. Yoon YS, Yu KP, Kim H, Kim HI, Kwak SH, Kim BO: The Blood Flow Metab 35:1358–1367, 2015 effect of electric cortical stimulation after focal traumatic 35. Paxinos G, Watson C: The Rat Brain in Stereotaxic Coor- brain injury in rats. Ann Rehabil Med 36:596–608, 2012 dinates, ed 6. London: Academic Press, 2007 55. Yoshikawa A, Atobe Y, Takeda A, Kamiya Y, Takiguchi M, 36. Ramantani G, Kadish NE, Brandt A, Strobl K, Stathi A, Wie- Funakoshi K: A retrograde tracing study of compensatory gand G, et al: Seizure control and developmental trajectories corticospinal projections in rats with neonatal hemidecortica- after hemispherotomy for refractory epilepsy in childhood tion. Dev Neurosci 33:539–547, 2011 and adolescence. Epilepsia 54:1046–1055, 2013 37. Rasmussen T: Hemispherectomy for seizures revisited. Can J Neurol Sci 10:71–78, 1983 38. Ribeiro SJ, Ciscato JG Jr, de Oliveira R, de Oliveira RC, Disclosures d’Ângelo-Dias R, Carvalho AD, et al: Functional and ultra- Dr. Matias reports a financial relationship with Conselho Nacio-

J Neurosurg June 8, 2018 13

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC I. Matias Jr. et al. nal de Desenvolvimento Cientifico e Tecnológico (CNPq; Nation- Lopes, Machado. Drafting the article: Matias, Elias-Filho. Criti- al Council for Scientific and Technological Development), Funda- cally revising the article: Coimbra, Lopes, Machado. Reviewed ção de Amparo à Pesquisa do Estado de São Paulo (FAPESP; São submitted version of manuscript: Coimbra, da Silva, Lopes, Paulo Research Foundation), and Coordenação de Aperfeiçoa- Machado. Approved the final version of the manuscript on behalf mento de Pessoal de Nível Superior (CAPES; Coordination for the of all authors: Coimbra. Statistical analysis: Lopes, Machado. Improvement of Higher Education Personnel). Administrative/technical/material support: Coimbra, Elias-Filho. Study supervision: Coimbra, Lopes, Machado. Author Contributions Conception and design: Coimbra, Matias, Lopes, Machado. Correspondence Acquisition of data: all authors. Analysis and interpretation of Norberto Cysne Coimbra: Ribeirão Preto Medical School of the data: Coimbra, Elias-Filho, Mejia, Cabral, Miranda, da Silva, University of São Paulo, Brazil. [email protected].

14 J Neurosurg June 8, 2018

Unauthenticated | Downloaded 10/03/21 10:42 PM UTC