Charles Munyon

Visual-spatial memory may be enhanced with theta burst deep brain stimulation of the fornix: a preliminary investigation with four cases

Memory loss after brain injury can be a source of considerable morbidity, but there are presently few therapeutic options for restoring memory function. We have previously demonstrated that burst stimulation of the fornix is able to significantly improve memory in a rodent model of traumatic brain injury. The present study is a preliminary investigation with a small group of cases to explore whether theta burst stimulation of the fornix might improve memory in humans. Four individuals undergoing stereo-electroencephalography evaluation for drug-resistant epilepsy were enrolled. All participants were implanted with an electrode into the proximal fornix and dorsal hippocampal commissure on the language dominant (n = 3) or language non-dominant (n = 1) side, and stimulation of this electrode reliably produced a diffuse evoked potential in the head and body of the ipsilateral hippocampus. Each participant underwent testing of verbal memory (Rey Auditory-Verbal Learning Test), visual-spatial memory (Medical College of Georgia Complex Figure Test), and visual confrontational naming (Boston Naming Test Short Form) once per day over at least two consecutive days using novel test forms each day. For 50% of the trials, the fornix electrode was continuously stimulated using a burst pattern (200 Hz in 100 ms trains, five trains per second, 100 µs, 7 mA) and was compared with sham stimulation. Participants and examiners were blinded to whether stimulation was active or not, and the order of stimulation was randomized. The small sample size precluded use of inferential statistics; therefore, data were analysed using descriptive statistics and graphic analysis. Burst stimulation of the fornix was not perceived by any of the participants but was associated with a robust reversible improvement in immediate and delayed performance on the Medical College of Georgia Complex Figure Test. There were no apparent differences on either Rey Auditory-Verbal Learning Test or Boston Naming Test. There was no apparent relationship between performance and side of stimulation (language dominant or non-dominant). There were no complications. Preliminary evidence in this small sample of patients with drug-resistant epilepsy suggests that theta burst stimulation of the fornix may be associated with improvement in visual-spatial memory.

Improved learning and memory with theta-burst stimulation of the fornix in rat model of traumatic brain injury

Objective: Learning and memory deficits are a source of considerable morbidity after traumatic brain injury (TBI). We investigated the effect of different patterns of hippocampal stimulation via a fornix electrode on cognitively demanding tasks after TBI. Methods: Male Sprague‐Dawley rats underwent fluid‐percussion injury and were compared with sham‐operated rats. Electrodes were implanted into the fornix and hippocampus, and stimulation of the fornix produced robust evoked potentials in the hippocampus. A 60‐s delayed non‐match‐to‐sample (DNMS) swim T‐maze was serially performed using four stimulation patterns: no stimulation (No Stim), low‐frequency stimulation (LFS, 5 Hz), high‐frequency stimulation (HFS, 130 Hz), and theta‐burst stimulation (TBS, 200 Hz in 50 ms trains, five trains per second; 60 µA biphasic pulses). In a separate cohort of sham and injured animals, Morris water maze (MWM) was performed with or without TBS. Results: In the DNMS swim T‐maze, LFS and HFS did not significantly improve performance after TBI. However, there was a significant difference in performance between TBI + No Stim and TBI + TBS groups (P < 0.05) with no significant difference between Sham + No Stim and TBI + TBS. In the MWM, latency in the TBI + TBS group was significantly different from TBI + No Stim starting on day 2 (P < 0.05) and was not different from Sham + No Stim. The TBI + TBS group performed significantly more platform crossings in the probe trial (P < 0.01) and exhibited improved search strategy starting on day 3 (P < 0.05) compared with TBI + No Stim. Conclusions: Deficits in learning and memory after TBI are improved with TBS of the hippocampus. HFS and LFS do not appear to produce as great an effect as TBS.

A Novel Method of Translabyrinthine Cranioplasty Using Hydroxyapatite Cement and Titanium Mesh: A Technical Report

We report a novel technique for closure using titanium mesh cranioplasty in addition to hydroxyapatite cement and abdominal fat graft for acoustic neuroma. We reviewed 15 patients who underwent translabyrinthine craniectomy for resection of acoustic neuroma. Hearing loss was documented prior to surgical procedure. Over 2 years, patients underwent titanium mesh and hydroxyapatite cranioplasty with abdominal fat graft. Participants included seven men and eight women, age range 38 to 65. Main outcome measures included cosmetic outcome and incidence of cerebrospinal fluid (CSF) leak. The lesion was right-sided in seven patients and left-sided in eight. Cosmetic outcome was excellent in all. There were no cases of CSF leak. Closure used one-third the hydroxyapatite required for traditional closure. Our technique yields cosmetic results equivalent to hydroxyapatite cement alone and a comparable incidence of CSF leakage without leaving a drain in place postoperatively. The technique is easy to adopt, is more cost-effective than hydroxyapatite cement cranioplasty alone, offers greater ease of access for reoperation, and does not preclude later implantation of bone-anchored hearing aid.

Accuracy of Frame-Based Stereotactic Depth Electrode Implantation during Craniotomy for Subdural Grid Placement

Background: Frame-based stereotaxy and open craniotomy may seem mutually exclusive, but invasive electrophysiological monitoring can require broad sampling of the cortex and precise targeting of deeper structures. Objectives: The purpose of this study is to describe simultaneous frame-based insertion of depth electrodes and craniotomy for placement of subdural grids through a single surgical field and to determine the accuracy of depth electrodes placed using this technique. Methods: A total of 6 patients with intractable epilepsy underwent placement of a stereotactic frame with the center of the planned cranial flap equidistant from the fixation posts. After volumetric imaging, craniotomy for placement of subdural grids was performed. Depth electrodes were placed using frame-based stereotaxy. Postoperative CT determined the accuracy of electrode placement. Results: A total of 31 depth electrodes were placed. Mean distance of distal electrode contact from the target was 1.0 ± 0.15 mm. Error was correlated to distance to target, with an additional 0.35 mm error for each centimeter (r = 0.635, p < 0.001); when corrected, there was no difference in accuracy based on target structure or method of placement (prior to craniotomy vs. through grid, p = 0.23). Conclusion: The described technique for craniotomy through a stereotactic frame allows placement of subdural grids and depth electrodes without sacrificing the accuracy of a frame or requiring staged procedures.

The 3-dimensional grid: a novel approach to stereoelectroencephalography.

BACKGROUND: Successful surgical treatment of epilepsy requires accurate definition of areas of ictal onset and eloquent brain. Although invasive monitoring can help, subdural grids cannot sample sulci or subcortical tissue; traditional stereoelectroencephalography depth electrodes are usually placed too far apart to provide sufficient resolution for mapping. OBJECTIVE: To report a strategy of depth electrode placement in a dense array to allow precise anatomic localization of epileptic and eloquent cortex. METHODS: Twenty patients with medically intractable epilepsy either poorly localized or found to arise adjacent to eloquent areas underwent placement of arrays of depth electrodes into and around the putative area of seizure onset with the use of framed stereotaxy. Each array consisted of a “grid” of parallel electrodes in a rectangular pattern with 1 cm between entry sites. In a subset of patients, a few electrodes were placed initially, with additional electrodes placed in a second stage. Trajectories were modified to avoid cortical vessels defined on magnetic resonance imaging. Patients were monitored for 4 to 21 days to establish the precise location of seizure onset. Stimulation was performed to map cortical and subcortical eloquent regions. Electrode locations were coregistered for frameless stereotaxy during subsequent resection of seizure focus. RESULTS: Two hundred fifty-four electrodes were implanted. Discrete regions of seizure onset and functional cortex were identified, which were used during resection to remove epileptogenic tissue while preserving eloquent areas. There were no hemorrhagic or infectious complications; no patient suffered permanent neurological deficit. CONCLUSION: The 3-dimensional intraparenchymal grid is useful for identifying the location and extent of epileptic and eloquent brain. ABBREVIATION: SEEG, stereoelectroencephalography

N-butyl 2-cyanoacrylate (n-BCA) embolization of a cerebellar hemangioblastoma

Hemangioblastomas (HBs) are highly vascular tumors whose resection can be associated with significant bleeding. Angioembolization has been used as an adjunct to surgical therapy, but particle embolization of cerebellar HBs has been associated with hemorrhage and resultant morbidity and mortality. We present a case of successful n-BCA embolization of an HB of the cerebellum.

Degree of distal trigeminal nerve atrophy predicts outcome after microvascular decompression for Type 1a trigeminal neuralgia.

OBJECT Trigeminal neuralgia is often associated with nerve atrophy, in addition to vascular compression. The authors evaluated whether cross-sectional areas of different portions of the trigeminal nerve on preoperative imaging could be used to predict outcome after microvascular decompression (MVD). METHODS A total of 26 consecutive patients with unilateral Type 1a trigeminal neuralgia underwent high-resolution fast-field echo MRI of the cerebellopontine angle followed by MVD. Preoperative images were reconstructed and reviewed by 2 examiners blinded to the side of symptoms and clinical outcome. For each nerve, a computerized automatic segmentation algorithm was used to calculate the coronal cross-sectional area at the proximal nerve near the root entry zone and the distal nerve at the exit from the porus trigeminus. Findings were correlated with outcome at 12 months. RESULTS After MVD, 17 patients were pain free and not taking medications compared with 9 with residual pain. Across all cases, the coronal cross-sectional area of the symptomatic trigeminal nerve was significantly smaller than the asymptomatic side in the proximal part of the nerve, which was correlated with degree of compression at surgery. Atrophy of the distal trigeminal nerve was more pronounced in patients who had residual pain than in those with excellent outcome. Among the 7 patients who had greater than 20% loss of nerve volume in the distal nerve, only 2 were pain free and not taking medications at long-term follow-up. CONCLUSIONS Trigeminal neuralgia is associated with atrophy of the root entry zone of the affected nerve compared with the asymptomatic side, but volume loss in different segments of the nerve has very different prognostic implications. Proximal atrophy is associated with vascular compression and correlates with improved outcome following MVD. However, distal atrophy is associated with a significantly worse outcome after MVD.

Motor cortex stimulation for chronic pain.

Motor cortex stimulation produces significant relief of symptoms in many forms of refractory chronic pain disorders.

Pharmacoresistant epilepsy in hypomelanosis of Ito: Palliative surgical treatment with modified anatomic posterior quadrantic resection

• The authors report modified anatomical posterior quadrant resection for adult epilepsy.

Multitract orthogonal microelectrode localization of the subthalamic nucleus: description of a novel technique.

BACKGROUND: Microelectrode recording helps surgeons accurately localize boundaries of the subthalamic nucleus (STN) and surrounding structures in deep brain stimulation. OBJECTIVE: To describe a novel adaptation of the Ben gun device to optimize efficient mapping. METHODS: Patients who underwent STN deep brain stimulation over a 3-year period were reviewed. For the final year, the Ben gun was rotated 45° and the target was offset 1.4 mm lateral and anterior in the plane orthogonal to the intended trajectory to allow for simultaneous parallel tracks at target, 2.8 mm anterior (localizing the front of STN), and 2.8 mm lateral (identifying the internal capsule). Before this step, the initial pass consisted of 1 to 2 tracks with the frame center targeted to STN. The primary outcome measure was the number of passes required for accurate localization of the nucleus and boundaries. RESULTS: Eighty-three electrodes were implanted in 45 patients (mean age, 62; range, 37-78 years), of which 29 electrodes were placed by the use of the new technique. One electrode (4%) required more than 1 pass using the new technique compared with 36 (67%) using the older technique (P < .01). The distance from original target to final electrode position increased from 0.67 ± 0.13 mm to 1.06 ± 0.15 mm (P < .05) with a greater tendency to move the final electrode position posteriorly. There was no statistically significant difference in benefit from neurostimulation. CONCLUSION: This technique facilitates reliable localization of the STN with fewer passes, possibly decreasing the risks associated with more passes and longer duration of surgery. ABBREVIATIONS: DBS, deep brain stimulators MER, microelectrode recording PD, Parkinson disease STN, subthalamic nucleus