Josué M. Avecillas-Chasin

Assessment of a method to determine deep brain stimulation targets using deterministic tractography in a navigation system

Recent advances in imaging permit radiologic identification of target structures for deep brain stimulation (DBS) for movement disorders. However, these methods cannot detect the internal subdivision and thus cannot determine the appropriate DBS target located within those subdivisions. The aim of this study is to provide a straightforward method to obtain an optimized target (OT) within DBS target nuclei using a widely available navigation system. We used T1- and T2-weighted images, fluid-attenuated inversion recovery (FLAIR) sequence, and diffusion tensor imaging (DTI) of nine patients operated for DBS in our center. Using the StealthViz® software, we segmented the targeted deep structures (subcortical targets) and the anatomically identifiable areas to which these target nuclei were connected (projection areas). We generated fiber tracts from the projection areas. By identifying their intersections with the subcortical targets, we obtained an OT within the DBS target nuclei. We computed the distances from the clinically effective electrode contacts (CEEC) to the OT obtained by our method and the targets provided by the atlas. These distances were compared using a Wilcoxon signed-rank test, with p < 0.05 considered statistically significant. We were able to identify OT coincident with the motor part of the subthalamic nucleus and the ventral intermediate nucleus. We clinically tested the results and found that the CEEC were significantly more closely related to the OT than with the targets obtained by the atlas. Our present results show that this novel method permits optimization of the stimulation site within the internal subdivisions of target nuclei for DBS.

Tractographical model of the cortico-basal ganglia and corticothalamic connections: Improving Our Understanding of Deep Brain Stimulation.

The cortico‐basal ganglia and corticothalamic projections have been extensively studied in the context of neurological and psychiatric disorders. Deep brain stimulation (DBS) is known to modulate many of these pathways to produce the desired clinical effect. The aim of this work is to describe the anatomy of the main circuits of the basal ganglia using tractography in a surgical planning station. We used imaging studies of 20 patients who underwent DBS for movement and psychiatric disorders. We segmented the putamen, caudate nucleus (CN), thalamus, and subthalamic nucleus (STN), and we also segmented the cortical areas connected with these subcortical areas. We used tractography to define the subdivisions of the basal ganglia and thalamus through the generation of fibers from the cortical areas to the subcortical structures. We were able to generate the corticostriatal and corticothalamic connections involved in the motor, associative and limbic circuits. Furthermore, we were able to reconstruct the hyperdirect pathway through the corticosubthalamic connections and we found subregions in the STN. Finally, we reconstructed the cortico‐subcortical connections of the ventral intermediate nucleus, the nucleus accumbens and the CN. We identified a feasible delineation of the basal ganglia and thalamus connections using tractography. These results could be potentially useful in DBS if the parcellations are used as targets during surgery. Clin. Anat. 29:481–492, 2016.

Multiple spinal arteriovenous fistulas: A case-based review

The occurrence of multiple spinal dural arteriovenous fistulas (AVFs) is rare. The majority of cases reported are synchronous and the lesions are mainly found at different spinal levels. Metachronous AVFs have been defined as lesions that manifest in a temporal sequence after treatment of a first AVF. In this report, we present two distinct cases of multiple spinal AVFs. Also, we review the main features of the cases previously reported, with emphasis on the proposed theories for the origin of multiple AVFs. In patients with failure to improve after treatment of a spinal DAVF, a whole-spine angiographic examination is mandatory, not only to ascertain the complete closure of the treated fistula, but also to look for a possible second lesion at a different spinal level.

Delayed posterior reversible encephalopathy syndrome (PRES) after posterior fossa surgery.

Dear Editor: PRES has been reported in association with neurosurgical interventions such as posterior fossa surgery in children, skull base surgery or during vasospasm therapy, among others [6, 8–10]. This syndrome has heterogeneous clinical manifestations such as headaches, seizures, visual changes, paresis, hemianopsia, obtundation, nausea and/or vomiting [1]. We describe the unusual case of a 19-year-old man without any known risk factors who developed a delayedonset PRES 7 days after an uneventful posterior fossa surgery. A 19-year-old man with an unremarkable past medical history presented with weight loss, dysgeusia, dizziness and blurred vision for the previous 6 months. Neurological examination revealed bilateral papilledema, left abducens nerve paresis, slight left dysmetria and nystagmus. MRI scan revealed a 5.50-cm mass in the fourth ventricle occupying the left foramen of Luschka (Fig. 1). The patient underwent a left suboccipital craniotomy, and an uneventful subtotal piecemeal resection was performed with a nodular mass remaining on the left posterolateral medulla. On the 8th postoperative day (POD), the patient experienced hypertension (180/105 mmHg), cortical blindness and a generalized tonic-clonic seizure lasting for 5 min with residual paresis of the right upper arm. An emergency CT scan revealed bilateral digitiform hypodensities with parietooccipital distribution. MRI scans revealed hyperintense lesions on T2-weighted images mainly affecting the white matter and with involvement of the basal ganglia and the parietal and occipital cortex. DWI showed no restriction, and ADC values were elevated (Fig. 1). In addition, some mild gyriform enhancement was evident on postcontrast T1weighted images. These findings were interpreted as PRES, and aggressive antihypertensive and anticonvulsant therapy was immediately initiated. A brain MRI obtained 7 days later showed a significant decrease of the previously described lesions (Fig. 1). A month later, the patient had no clinical or radiographic sequelae of PRES. This case is one of the few reported of PRES as a complication of posterior fossa surgery [8, 9]. In our case, the unusual delayed onset of PRES obscured the diagnosis. The clinical findings in conjunction with typical PRES brain MR scan images proved crucial in establishing a diagnosis. Moriarity et al. reported the first perioperative case of PRES during posterior fossa surgery with the onset of symptoms taking place immediately after surgery. In general, PRES imaging findings are described in the CT scan as hypodense areas with posterior circulation distribution. On MRI, PRES lesions are generally described as symmetrical cortical and subcortical parieto-occipital hyperintensities on T2weighted and FLAIR sequences [3]. A moderate signal enhancement on T1-weighted images with gadolinium could be explained by disruption of the blood-brain barrier [1]. These alterations can also occur as focal or asymmetrical in the frontal lobes, inferior temporo-occipital junction, basal ganglia, brainstem, cerebellum and cervical spinal cord [1]. Diffusion-weighted images (DWI) usually show no restriction and high apparent diffusion coefficient (ADC) values, indicating the presence of vasogenic edema [2]. J. M. Avecillas-Chasín :G. Gómez : J. A. Barcia Department of Neurosurgery, Institute of Neurological Sciences, Hospital Clínico San Carlos, Universidad Complutense de Madrid, Madrid, Spain

Neuroendoscopic Intraventricular Biopsy in Children with Small Ventricles Using Frameless VarioGuide System

Endoscopic biopsy for intraventricular tumors in pediatric patients with small ventricles is a challenging procedure because of the risk of morbidity during the intraventricular approach. We describe the use of the VarioGuide system for intraventricular endoscopic biopsy in 9 consecutive pediatric patients with intraventricular lesions and small ventricular size. All patients had lesions in the anterior part of the third ventricle with a median frontal and occipital horn ratio of 0.33. Patients presented with growth failure (n = 4), visual disturbances (n = 4), and seizures (n = 1). The VarioGuide system consists of an ergonomic arm with 3 joints for gross adjustment. The 3 rotational joints on the distal side of the system are adjusted according to the angles of the planned trajectory. The endoscope is adjusted to the distal side of the VarioGuide and inserted through the ring, previously set for the diameter of the endoscope and for the planned trajectory. The accuracy of the trajectory and correct ventricular cannulation are confirmed under endoscopic guidance. The biopsy is carried out according to the standard technique. In all cases, the biopsy sample provided the definitive diagnosis. Diagnoses included germinomas in 4 patients, hamartoma in 1 patient, hypothalamic astrocytoma in 2 patients, and craniopharyngioma in 2 patients. The use of the VarioGuide system for intraventricular endoscopic biopsy is highly recommended for pediatric patients with small ventricle size. This technique may help minimize the risk of unnecessary brain damage during the entrance to small ventricles.

Subdural effusion in decompressive craniectomy.

Dear Editor I have read with great interest the article BAnother treatment choice for subdural effusion with ventricle dilation^ by Fang et al. The authors report their experience treating subdural effusions (SE) in two patients with decompressive craniectomy (DC) using lumboperitoneal shunt [7]. Traumatic SE is an alteration of the cerebrospinal fluid (CSF) circulation that is associated with traumatic brain injury (TBI). However, SE has also been associated with DC for traumatic and non-traumatic conditions [11, 18, 22]. The underlying pathogenesis of SE in both conditions (DC and TBI) is different; therefore, the management of this condition could be largely different [3, 10, 13]. Traumatic SE is a relatively common complication of TBI with a benign course in the majority of cases [23]. As the authors mentioned, the origin of TSE has been related with a tear in the subarachnoid layer probably due to the head trauma producing egress of the CSF to the subdural space in a valve-like manner [1, 7, 23]. Other theories have been also proposed such as the presence of a neomembrane with increased capillary permeability leading to the development of the SE [1, 8]. Furthermore, due to the head trauma or disturbances in CSF absorption, alterations in intracranial compliance could also contribute to this complication [9, 12, 13, 15]. I agree with the authors that most SE cases are self-limited, but in rare cases SE manifests with mass effect and ventricular dilation and even in those cases the surgical management is variable [4, 16, 17, 23]. Accordingly, it is not clear if shunting procedures are the Bonly cure^ for this type of CSF alterations in the setting of DC for TBI, as stated by the authors [7]. On the other hand, CSF disturbances after DC could be caused by intracranial hypotension and the effect of atmospheric pressure [2, 5, 10]. After a large bone flap is removed from the cranium, atmospheric pressure is applied onto the brain parenchyma producing a significant alteration in the intracranial dynamics [2]. CSF absorption is carried out in a pressure-dependant manner through the arachnoid granulations with a directly proportional relationship between subarachnoid pressure and CSF absorption rate [14, 18]. Therefore, pressurerelated alterations take place into the subarachnoid space with the subsequent alteration of the CSF absorption. This mechanism is aggravated by any communication between the subdural and subarachnoid space [10] producing the SE after DC. As the authors stated, the SE acts as compensatory mechanism of the disturbed CSF circulation due to the subarachnoid/subdural gradient. Furthermore, ventricular dilation may also occur by the same mechanism [6, 18], and this event may further dis turb the CSF dynamics thus increasing the subarachnoid/subdural gradient. SE is not a common finding in patients who have undergone DC for traumatic and non-traumatic conditions [1, 11, 21]. Therefore, other factors may play a role in the development of SE after DC. Despite that the authors report a good outcome of their patients, it is necessary to bear in mind that the same radiological picture could be related to different pathophysiological mechanisms. SE could also be * Josue M. Avecillas-Chasin josue.avecillas@salud.madrid.org; josueavecillas@hotmail.com

Implantation of Depth Electrodes in Children Using VarioGuide® Frameless Navigation System: Technical Note

BACKGROUND Electrode placement in epilepsy surgery seeks to locate the sites of ictal onset and early propagation. An invasive diagnostic procedure, stereoelectroencephalography (SEEG) is usually implemented with frame-based methods that can be especially problematic in young children. OBJECTIVE To evaluate the feasibility and accuracy of a new technique for frameless SEEG in children using the VarioGuide® system (Brainlab AG, München, Germany). METHODS A frameless stereotactic navigation system was used to implant depth electrodes with percutaneous drilling and bolt insertion in pediatric patients with medically refractory epilepsy. Data on general demographic information of electrode implantation, duration, number, and complications were retrospectively collected. To determine the placement accuracy of the VarioGuide® frameless system, the mean Euclidean distances were calculated by comparing the preoperatively planned trajectories with the final electrode position observed on postoperative computed tomography scans. RESULTS From May 2011 to December 2015, 15 patients (8 males, 7 females; mean age: 8 yr, range: 3-16 yr) underwent SEEG depth electrode implantation of a total of 111 electrodes. The mean error measured by the Euclidean distance from the center of the entry point to the intended entry point was 3.64 ± 1.78 mm (range: 0.58-7.59 mm) and the tip of the electrode to the intended target was 2.96 ± 1.49 mm (range: 0.58-7.82 mm). There were no significant complications. CONCLUSION Depth electrodes can be placed safely and accurately in children using the VarioGuide® frameless stereotactic navigation system.

Individualization of deep brain stimulation targets for movement disorders

Dear Editor, I have read with great interest the paper titled “The variability of atlas-based targets in relation to surrounding major fiber tracts in thalamic deep brain stimulation” by Anthofer et al. [3]. In this work, the authors investigate the variability of the position of the atlas-based targets for essential tremor (ET) in relation with the major tracts obtained by tractography such as medial lemniscus, pyramidal tract, and dentatorubrothalamic tract (DRT). The authors include five patients with ET, one patient with dystonia, one patient with obsessive–compulsive disorder, and three patients with Parkinson’s disease (PD). The authors generate the fiber tracking using regions of interest widely accepted for the tractographical reconstruction of the DRT tract [2, 4, 8, 9]. The authors report a considerable variability of the location of the ventral intermediate (VIM) nucleus (atlas-based) in relation with the above-mentioned tracts. The main concern of this study is the spatial distribution of the fiber tracts because the artifacts of the diffusion tensor imaging (DTI) acquisition (cardiac pulsations, patient motion, and coregistration issues) may affect the correlation of the tracts with the target points given by the atlas. Therefore, it is not clear if DTI can be used as a reference to measure the variability of the atlas-based targets. The main limitation of deterministic tractography (DT) is the difficulty to trace out zones with multiple fiber orientations [5, 11]. However, the anatomical features of the DRT tract have been reproduced using DT with acceptable reliability [2, 4, 8, 9]. This tract is formed by the contralateral dentate nucleus, red nucleus, and it runs through the inferolateral border of the thalamus (Fig. 1). The DRT tract innervates the ventral lateral and ventral anterior nucleus of the thalamus [2, 13]. It is not clear if the DRT generated in this paper spans the whole thalamic connections or just the connection with the VIM nucleus (ventral lateral posterior). Therefore, the significant proportion of VIM targets (derived by the atlas) posterior to the DRT tract (64.7 %) would reflect the poor sensitivity of DTI instead of a real variability of the atlas. The use of voxel-based estimates to determine the main direction of the DRT tract could lead to inaccuracies in the spatial interpretation of the connectivity, as it has been widely demonstrated that tractography is just a gross estimate of the brain connectivity [6, 7, 12, 16]. Although the most probable source of variability comes from the atlas-based target, the comparison of two extremely variables points (atlas and tractography) would give variable results by itself [6] According to the conclusion of this study, it is necessary to move forward from the atlas in order to obtain more reliable methods for deep brain stimulation (DBS) targeting. My group published a work comparing the Euclidean distances from the atlas-based target to the clinically effective electrode’s contact (CEEC) and the Euclidean distances from the tractography-based target to the CEEC in a group of patients with PD and ET (Fig. 1). * Josué M. Avecillas-Chasin josue.avecillas@salud.madrid.org; josueavecillas@hotmail.com