Variability of Intraoperative Electrostimulation Parameters in Conscious Individuals: Language Fasciculi.

OBJECTIVE: The authors analyzed the current-intensity thresholds for electrostimulation of language fasciculi and the possible consequences of threshold variability on brain mapping. METHODS: A prospective protocol of subcortical electrostimulation was used in 50 patients undergoing brain mapping, directly stimulating presumed language fasciculi identified by diffusion tensor imaging. RESULTS: The stimulation-intensity thresholds for identification of language fasciculi varied among patients (mean minimum current intensity of 4.4 mA, range = 1.5-10 mA, standard deviation = 1.1 mA), and 23% of fascicular interferences were detected only above 5 mA. Repeated stimulation of the same site with the same intensity led to different types of interferences in 20% of patients, and a higher current intensity led to changes in the type of response in 27%. The mean minimum stimulation intensities did not differ significantly between different fasciculi, between the different types of interference obtained, or with age, sex, or type of tumor. Positive results on cortical mapping were significantly associated with positive results on subcortical mapping (P < 0.001). Subcortical intensity thresholds were slightly lower than cortical ones (mean = 4.43 vs. 5.25 mA, P = 0.034). In 23 of 50 subcortical mappings, fascicular stimulation produced no language interference. CONCLUSIONS: Individual variability of minimum stimulation-intensity thresholds for identification of language fasciculi is frequent. Nevertheless, even when a high current intensity was used, many stimulations on language fasciculi remained negative for various hypothetic reasons. Finding the optimal current intensity for identifying language fasciculi is of paramount importance to refine the clinical results and scientific data derived from brain mapping.

Endoscopic Transfontanellar Approach of a Cyst Anterior to the Brainstem Crossing the Foramen Magnum-Part II: Success of Multiple Fenestrations: Two-Dimensional Operative Video.

Premature infants with severe germinal matrix/intraventricular hemorrhage are at high risk of posthemorrhagic ventricular dilation and hydrocephalus (Video 1). We report a 4-month-old premature infant with grade III intraventricular hemorrhage and compartmentalized hydrocephalus with an unusual craniospinal cyst. The cyst extended anteriorly from the mesencephalon to the posterior wall of C6, causing severe compression of the brainstem and spinal cord. An endoscopic procedure was performed first to achieve a unique cranial fenestration, which is detailed in "Endoscopic Transfontanellar Approach of a Cyst Anterior to the Brainstem Crossing the Foramen Magnum-Part I: Failure of Unique Fenestration." One month later the clinical picture recurred. Magnetic resonance imaging revealed hydrocephalus and cyst recurrence. A second procedure with endoscopic fenestration and shunt revision was needed. The same right transfontanellar approach was chosen, using a straight 30° endoscope. The procedure and surgical technique are explained in a step-by-step fashion. Extreme care was taken to align the head and the cervical spine. This allowed access to the extreme caudal cyst membrane posterior to C6 to create multiple transfixing fenestrations of the cyst. At 18-month follow-up, the child demonstrated almost normal neurological and psychomotor development with no cyst recurrence or hydrocephalus. Our report underlines the importance of performing multiple fenestrations of such cysts. We believe that performing a transfixing fenestration through the cyst allows cerebrospinal fluid flow and prevents recurrence, as, for instance, has been shown for suprasellar arachnoid cysts.

Endoscopic Transfontanellar Approach of a Cyst Anterior to the Brainstem Crossing the Foramen Magnum-Part I: Failure of Unique Fenestration: 2-Dimensional Operative Video.

Premature infants with severe germinal matrix intraventricular hemorrhage (IVH) are at high risk of posthemorrhagic ventricular dilation and hydrocephalus (Video 1). We report the case of a 3-month-old premature infant referred for posthemorrhagic multilocated hydrocephalus. He presented with somnolence, hypotonia, and a bulging fontanelle. His past medical history included an IVH grade III with hydrocephalus initially treated by temporary ventriculosubgaleal shunting. Magnetic resonance imaging (MRI) showed persistent hydrocephalus associated to a cyst extending anteriorly from the mesencephalon to the posterior wall of the sixth cervical vertebra, causing severe brainstem and spinal cord compression. A two-step surgery was performed, consisting of an endoscopic procedure and a ventriculo peritoneal shunt placement, to achieve hydrocephalus treatment and cyst fenestration. A right transfontanellar approach with the head flexed was chosen, using a straight 30-degree endoscope. The procedure and surgical technique are explained in a step-by-step fashion in the original conditions. A large unique fenestration was performed with no postoperative complications and improvement of the neurological status. One month later the patient presented again with symptomatic intracranial hypertension with compatible clinical and imagery findings. MRI showed cyst recurrence and hydrocephalus. We therefore revised our strategy and performed an additional procedure, which is detailed in the second part of the video (Part II: Success of Multiple Fenestrations). Our report underlines the high risk of recurrence of such cysts in post-hemorrhagic arachnoiditis. Cyst recurrence may be explained by the absence of crossing flow when a unique perforation is made.

Automatic segmentation of brain tumor resections in intraoperative ultrasound images using U-Net.

To compensate for the intraoperative brain tissue deformation, computer-assisted intervention methods have been used to register preoperative magnetic resonance images with intraoperative images. In order to model the deformation due to tissue resection, the resection cavity needs to be segmented in intraoperative images. We present an automatic method to segment the resection cavity in intraoperative ultrasound (iUS) images. We trained and evaluated two-dimensional (2-D) and three-dimensional (3-D) U-Net networks on two datasets of 37 and 13 cases that contain images acquired from different ultrasound systems. The best overall performing method was the 3-D network, which resulted in a 0.72 mean and 0.88 median Dice score over the whole dataset. The 2-D network also had good results with less computation time, with a median Dice score over 0.8. We also evaluated the sensitivity of network performance to training and testing with images from different ultrasound systems and image field of view. In this application, we found specialized networks to be more accurate for processing similar images than a general network trained with all the data. Overall, promising results were obtained for both datasets using specialized networks. This motivates further studies with additional clinical data, to enable training and validation of a clinically viable deep-learning model for automated delineation of the tumor resection cavity in iUS images.

Brain-shift compensation using intraoperative ultrasound and constraint-based biomechanical simulation.

PURPOSE: During brain tumor surgery, planning and guidance are based on preoperative images which do not account for brain-shift. However, this deformation is a major source of error in image-guided neurosurgery and affects the accuracy of the procedure. In this paper, we present a constraint-based biomechanical simulation method to compensate for craniotomy-induced brain-shift that integrates the deformations of the blood vessels and cortical surface, using a single intraoperative ultrasound acquisition. METHODS: Prior to surgery, a patient-specific biomechanical model is built from preoperative images, accounting for the vascular tree in the tumor region and brain soft tissues. Intraoperatively, a navigated ultrasound acquisition is performed directly in contact with the organ. Doppler and B-mode images are recorded simultaneously, enabling the extraction of the blood vessels and probe footprint, respectively. A constraint-based simulation is then executed to register the pre- and intraoperative vascular trees as well as the cortical surface with the probe footprint. Finally, preoperative images are updated to provide the surgeon with images corresponding to the current brain shape for navigation. RESULTS: The robustness of our method is first assessed using sparse and noisy synthetic data. In addition, quantitative results for five clinical cases are provided, first using landmarks set on blood vessels, then based on anatomical structures delineated in medical images. The average distances between paired vessels landmarks ranged from 3.51 to 7.32 (in mm) before compensation. With our method, on average 67% of the brain-shift is corrected (range [1.26; 2.33]) against 57% using one of the closest existing works (range [1.71; 2.84]). Finally, our method is proven to be fully compatible with a surgical workflow in terms of execution times and user interactions. CONCLUSION: In this paper, a new constraint-based biomechanical simulation method is proposed to compensate for craniotomy-induced brain-shift. While being efficient to correct this deformation, the method is fully integrable in a clinical process.