
BACKGROUND AND OBJECTIVES: Declining operative exposure during residency has increased the need for realistic and accessible surgical simulation models for the acquisition of microsurgical skills. Human placenta has previously been used as a biological substrate for vascular microsurgical training. In this study, we present a placenta-based training model integrated with a 3-dimensional-printed craniotomy to reproduce key steps of microsurgical resection of cerebral cavernous malformations. METHODS: Human placental tissue was combined with a 3-dimensional-printed pterional craniotomy to create a microsurgical training platform. A cavernous malformation was simulated using placental subunit fragments and clots, which were encapsulated and embedded within the placental parenchyma. The model was perfused through cannulation of umbilical vessels. The simulation procedure reproduced the main surgical steps: dural opening, vascular dissection, lesion exposure, intracapsular dissection with piecemeal removal and dural closure. The model was tested by 2 experienced neurosurgeons and 2 senior residents, who evaluated the simulator using a 5-point Likert scale questionnaire assessing realism, tissue consistency, instrument handling, training value, and overall usefulness. RESULTS: The model could be assembled in a mean time of 16 minutes at negligible costs and standard instrumentation. Up to 8 simulations could be obtained from a single placenta, increasing the model's cost-effectiveness. Evaluators rated the simulator favorably across all the investigated domains. Experts completed the procedure in a mean time of approximately 16 minutes, whereas residents required approximately 35 minutes. CONCLUSION: The described placenta-based model represents a feasible and reproducible platform for training microsurgical removal of cerebral cavernous malformation and may serve as a complementary tool in neurosurgical training programs.
BACKGROUND AND OBJECTIVE: Third ventricle colloid cysts (TVCCs) are rare, benign intracranial lesions often associated with hydrocephalus and neurological symptoms. Although various surgical techniques exist, the optimal approach remains debated. The endoscopic supracerebellar-infratentorial (SCIT) approach presents a minimally invasive alternative for accessing deep-seated lesions while preserving critical neurovascular structures. The aim of this study was to evaluate the feasibility, safety, and effectiveness of the endoscopic SCIT approach for the resection of TVCC. METHODS: A retrospective analysis was conducted on 5 patients who underwent the endoscopic SCIT approach for third ventricle lesions, including 3 cases of TVCC, one pineal apoplexy, and one pineal metastasis. Surgical outcomes, complications, and postoperative recovery were assessed through MRI and neurological examinations over a mean follow-up period of 40 months. RESULTS: Gross total resection was achieved in pineal lesion cases and total cyst content resection with various capsule resections achieved in patients with colloid cyst, with no major intraoperative complications. One patient had a preexisting ventriculoperitoneal shunt, which was removed on the third postoperative day, and no additional morbidity was observed. Postoperative imaging confirmed no recurrence over the follow-up period. The endoscopic SCIT approach provided direct visualization of the third ventricle while minimizing cortical and white matter disruption. CONCLUSION: The endoscopic SCIT approach offers a viable alternative for selected cases, particularly in patients without significant hydrocephalus, small cyst volume seated in ventral portion of the third ventricle. It provides superior visualization with minimal brain retraction, reducing the risk of neurological deficits. Although the SCIT approach is not a replacement for conventional transcallosal or transcortical methods, it represents an effective option in specific cases. Further studies with larger cohorts and longer follow-up periods are needed to validate its long-term efficacy.
BACKGROUND AND OBJECTIVES: Radiofrequency (RF) rhizotomy of the Gasserian ganglion is an established treatment for trigeminal neuralgia that requires accurate localization before lesioning. Temperature-based interrogation has been proposed as a localization strategy, but its thermal safety margin has not been quantified. We quantified thermal injury risk of temperature mapping relative to definitive lesioning. METHODS: Thermocouple logs from 12 patients undergoing RF rhizotomy with a cylindrical platinum band electrode (0.8-mm diameter, 2.0-mm contact length) included 54 mapping and 19 definitive lesion episodes. Arrhenius injury dose was computed from temperature–time data. A reduced-order effective-medium model of the Gasserian ganglion, parameterized by cerebrospinal fluid fraction and distributed heat exchange, was calibrated to burn episodes and applied to mapping episodes. Sensitivity analyses evaluated 2 Arrhenius parameter sets and ±2°C thermocouple bias. RESULTS: The Arrhenius damage integral (Ω; Ω ≥ 1 indicating predicted irreversible injury) at the thermocouple was far lower during mapping than during definitive lesioning. Under neural-relevant parameters, median mapping Ω was 3.73 × 10 −4 (IQR 4.16 × 10 −5 -1.65 × 10 −3 ; maximum 9.92 × 10 −3 ) vs 1.85 × 10 3 (IQR 5.90 × 10 1 -3.36 × 10 3 ) during burns. Predicted burn lesion radii were 1.8 to 2.2 mm, while mapping radius was 0 mm across all episodes. Mapping remained subthreshold under canonical RF parameters and ±2°C bias. CONCLUSION: Temperature mapping using multicontact cylindrical electrodes produced exposures several orders of magnitude below lesioning thresholds. These findings support a substantial thermal safety margin for temperature mapping during trigeminal RF rhizotomy.
BACKGROUND AND OBJECTIVES: Stereoelectroencephalography (SEEG) is an important methodology for invasive evaluation in medically refractory epilepsy patients. Traditionally, laser-based facial scanning has been used for stereotactic robotic registration at our center. More recently, we developed and validated a noninvasive, frame-based fiducial system integrated with intraoperative computed tomography (CT) registration, as part of our SEEG workflow. This study aims to compare the stereotactic accuracy in SEEG electrode placement using these 2 approaches. METHODS: Registration error, implantation accuracy metrics, and perioperative variables were collected for 56 refractory epilepsy patients undergoing SEEG between April 2022 and March 2024. Thirty-four patients underwent intraoperative CT registration, and 22 patients underwent laser facial scanning registration. Entry-point error, target-point error, and angular error metrics were measured for every electrode trajectory, across 1059 electrodes total. Wilcoxon rank sum tests were used to compare error measures between registration modalities. RESULTS: Overall mean age was 27.7 +/- 11.2 years, with a mean of 28.3 +/- 10.2 years in the laser group and 27.3 +/- 12.0 years in the CT group. The average number of electrodes implanted was 19 +/- 3 (intraoperative CT: 20 +/- 3; laser: 19 +/- 3). The average entry-point error was 1.52 +/- 0.91 mm (intraoperative CT: 1.28 +/- 0.83 mm; laser: 1.87 +/- 0.90 mm; P < .0001). The average target-point error was 2.75 +/- 1.48 mm (intraoperative CT: 2.65 +/- 1.43 mm; laser: 2.89 +/- 1.53 mm; P = .0050). The average angular error was 0.051 +/- 0.038 radians (intraoperative CT: 0.045 +/- 0.032 radians; laser: 0.060 +/- 0.045 radians; P < .0001). CONCLUSION: The intraoperative CT registration workflow, compared with laser facial scanning, provided significantly lower stereotactic error across all 3 error metrics. Laser facial scanning still offered an adequate error profile for SEEG. Depending on clinical necessity, these results validate intraoperative CT registration as an important step in accurate robotic-guided SEEG.
BACKGROUND AND OBJECTIVES: To evaluate the effectiveness of rigid skull base reconstruction using unsintered hydroxyapatite/poly L-lactide (u-HA/PLLA) plates in cases of high-flow intraoperative cerebrospinal fluid (CSF) leaks (Esposito grade 3) during extended endoscopic trans-sphenoidal surgery (EETSS) METHODS: We retrospectively reviewed 84 patients with intra- or suprasellar tumors who underwent EETSS from January 2019 to June 2025. Among these, 33 experienced intraoperative grade 3 CSF leaks. Rigid reconstruction was performed using either autologous bone or u-HA/PLLA plates. Postoperative CSF leak rates and complications were assessed. RESULTS: Postoperative CSF leakage occurred in 4 of 33 grade 3 cases (12.1%), all reconstructed with autologous bone. No CSF leaks occurred in any of the 14 cases reconstructed with u-HA/PLLA plates. No complications related to u-HA/PLLA plate usage were observed. CONCLUSION: Rigid reconstruction with u-HA/PLLA plates is safe and effective for preventing postoperative CSF leaks in high-risk EETSS cases. This material offers reliable fixation, bioresorbability, and radiolucency, making it suitable for skull base reconstruction.