Optimization of intraoperative ultrasound navigation during focal cortical dysplasia surgery: a case report

List of abbreviations

• FCD – focal cortical dysplasia
• IUS – intraoperative ultrasound
• BL – Brainlab
• MRI – magnetic resonance imaging
• EEG – electroencephalogram

Focal cortical dysplasia (FCD) is a common cause of seizures in adults and is known to be the main cause of focal intractable epilepsy in children, accounting for up to 50% of cases in paediatric epilepsy surgery. Attaining adequate seizure control with consequent chance for normal cognitive development are the goals of surgical intervention in the management of refractory epilepsy with underlying histological evidence of cortical dysplasia [1–3]. However, one the necessities required during the surgical resection of the offending lesion is the ability to distinguish clearly between the atypical tissue and normal brain. These tissues are hardly differentiated from one another. Intraoperative neuronavigation is therefore essential not only to discriminate between abnormal and normal tissue but also ascertain lesion dimensions and identify vital neurovascular structures.

Intraoperative ultrasound (IUS) is an imaging modality that has been purported to efficiently circumvent these challenges. The ability to acquire real-time images at any stage of the procedure while counteracting the element of brain shift are palpable benefits of using IUS [4–8]. It is also quicker to perform than other imaging navigation modalities. Institutions are also spared from huge budgetary constraints as it is considerably economically cost effective [9][10].

Despite the overwhelming advantages of using IUS, it is not without flaws. The two-dimensional (2D) IUS systems have poor image quality, lack of spatial resolution and dynamic range. Additionally, due to the steep learning curve, many neurosurgeons are usually unable to confidently read 2D IUS views. IUS artifacts at the bottom of the resection cavity make it difficult in visualizing lesion bounders which in turn results in insufficient visualization of lesion residue.

Here-in, we discuss the technical nuances via a case of intractable epilepsy secondary to FCD type II of enhancing 2D IUS navigation by integrating (co-registering) it with pre-operative magnetic resonance imaging (MRI) in Brainlab (BL) navigation (Brainlab Digital O.R. Brainlab AG, Germany) to archive 3D IUS views. BL a form of frameless stereotactic surgery provides intraoperative guidance in real-time. This greatly improves precision needed to archive safe but adequate resection of intra axial lesions.

CASE REPORT

A 22-year-old female presented early 2022 to our institution (Federal Сentre of Neurosurgery, Tyumen, Russia) with a long-standing history of bilateral tonic-clonic focal onset epileptic seizures twice a week. Over the course of her condition, she was placed on more than two anti-epileptic drugs: Lamotrigine (200 mg BD PO), Carbamazepine (400 mg BD PO) Topiramate (100 mg BD PO), Levetiracetam (750 mg BD PO), Oxcarbazepine (600 mg BD PO) without meaningful seizure relief. Her neonatal, developmental, past medical, and surgical histories were unremarkable.

The patient underwent a full range of pre-surgical examination by a team of epilepsy specialists: neurosurgeons, neuro-epileptologists, neuropsychiatrists, neurologists, speech therapists and neuro-radiologists according to current clinical guidelines.

Preoperative evaluation

Electroencephalogram

Long-term video electroencephalogram (EEG) monitoring (8 days) using 64 scalp electrodes (10-10) was used to record epileptiform activity. Epileptiform activity during the interictal phase, were detected in the form of periodic single and grouped series of spike-wave discharges localized in the right posterior temporal region emanating from leads T6 and TP8. During the ictal phase (left sided head tilt, a deviation of the mouth to the left, with subsequent generalized tonic-clonic convulsions) the epileptiform activity emanated from TP8 origin. Ultimately, the right posterior temporal area TP8 was identified as the seizure onset zone (I.1).

Magnetic resonance imaging

Pre-operative 1.5Tesla MRI (Epilepsy Protocol) was used. On FLAIR, an intra-axial lesion (2.9 ×1.9 cm) was depicted in the posterior sections of the right temporal lobe, on the border of the middle and inferior temporal gyri, typical of cortical dysplasia (characteristic thickening of the cortex, blurring of the contours of the gray-white mater junction, as well as with the presence of a typical trans mantle dysplasia. These findings provided correlation between the lesion defined in the MRI examination, seizure semiology and EEG monitorization to demonstrate the association between the epileptic focus and the lesion.

Diagnosis

The diagnosis was made by a panel of neurosurgeons, neuroepileptologists, neuropsychiatrists, neurologists, speech therapists, and neuroradiologists: Structural focal epilepsy. Bilateral tonic-clonic focal onset epileptic seizures. Drug-resistant form. FCD of the right frontal lobe (ICD-10 G40.2).

The decision was made to perform right temporal lobe lesionectomy using optimized IUS with BL as neuronavigaton.

Integrated 2D intraoperative ultrasound with Brainlab

We used Two-dimensional (2D) FlexFocus 800 IUS (BK Medical, Denmark) as frameless neuronavigation (Figure 2C). A craniotomy localizer (Craniotomy 8862) was attached to a Linear-type convex transducer IUS probe (29×10 mm, 3.8–10 MHz). The localizer was used to affix the ultrasonic transducers in a pre-set location (Figure 2E). The patient’s MRI (Epilepsy protocol) was then uploaded onto the BL navigation system Thereafter, the IUS was connected to the BL station and images were merged using the ultrasound BL program. Live 2D ultrasound images were then overlaid on preoperative MRI patient data to enable for precise correlation of the intraoperative status of the lesion. After which, the volumetric dataset was accessible for 3D navigation in any reconstructed view. Ultrasound data was visualized with better quality images (920×1080 pixels) using BL IUS software in 3D (Figure 2D). The reconstructed 3D images could be reviewed in real time in coronal, sagittal and axial planes.

Surgical procedure

Integrated 2D intraoperative ultrasound plus the 3D Brainlab

The lead surgeon was Professor Albert Akramovich Sufianov. The 2D plus the 3D BL ultrasound scanner was used as a navigation to determine the location of the pathological focus, the structure and echogenicity of the pathological tissue in relation to the surrounding normal brain (Figures 3A, В1, В2, С). Elements of brain shift, lesion size and residue, were looked out for during the procedure clearly identified and evaluated. Microsurgical removal of epileptogenic tissue was performed following the boundaries of the lesion (Figure 3D).

Post-Operative evaluation

Early post-operative MRI images were conducted to ascertain the extent of resection and lesion residue (Figure 3E1, 3E2). MRI findings showed gross total resection of the lesion. Histological findings were consistent with FCD type IIb (G40.2) (Figure 4).

Post-operative follow-up was conducted by neurologist, neuro-epileptogist and neurosurgeons on the 1^st, 3^rd, and 6^th month. The patient remained seizure-free at a 6-month follow-up (Engel I) while being maintained on a single antiepileptic drug: oxcarbazepine (600 mg BD PO).

DISCUSSION

An important clinical issue is the current inability to effectively treat many patients with refractory epilepsy caused by FCD. The main task during epilepsy surgery is to safely resect the lesion while preserving meaningful neurological faculties in the patient. The statistically significant predictor of post-operative seizure freedom is the completeness of resection. One commonly encountered explanation for incomplete resection is the inaccurate localization of the diseased tissue which is exacerbated by brain shifting that may occur at numerous stages during surgery, from the opening of dura mater, as tissue is resected and potentially throughout surgery. This is even more critical in patients with FCD as the extent of resection greatly influences postsurgical seizure outcome [11].

The use of IUS as a navigational tool in epilepsy surgery has greatly increased over the past decade [12][13]. This has led to the development of improved and optimized solutions to the IUS images acquired during surgery. It is our opinion that the integrated 2D IUS and ultrasound installed BL systems using pre-operative MRI gives better appreciation of the lesion dimensions in the axial, coronal and sagittal planes 3D views, which can improve the extent of resection and the post-operative outcomes. This makes image interpretation easier because one may utilize the MRI image to better understand the US image and the general layout of the operating room.

Additionally, the use of ultrasound navigation software with BL systems integration enables the scanning and reconstruction of 3D ultrasound data noticeably faster than with intraoperative MRI, and this allows that the surgery is quicker and more efficient; this means that the surgical workflow is practically uninterrupted [14][15]. In a study carried out at their institution, Sacino et al. investigated the use of intraoperative MRI and found that it increased the length of the procedure by 1.5 to 3 hours. The main elements that contributed to the need for more operating time included the patient’s preparation, transportation, performing MRI, and return as well as the re-sterilization and re-gowning of the patient, nurses, and surgeon while returning in the operating room.

Irrespective of the user’s level of IUS modality expertise, the BL automated view layouts are designed for ultrasound interpretation, making it easier to analyse ultrasound data, and thus even with minimum experience, junior surgeons can localize FCD lesions. We have been using IUS and IUS with BL systems integration at our centre for over 10 years and over 8 years, respectively. The lengthy learning curve of the initial phase can be significantly reduced with the usage of IUS with BL systems integration. Even though we observed the easy facility for the least experienced surgeons, blood products, and oedema can make residue interpretation a challenge.

The clinical postoperative outcome was not directly impacted by only using BL combined with IUS. However, the authors believe that the positive postoperative result was most likely influenced by the usage of IUS [16]. These results are equivalent to those attained when even more expensive neuronavigation tools (IMRI and ICT) are used.

CONCLUSION

IUS undoubtedly meets the criteria to be regarded as a good neurosurgical planning tool. The optimization of IUS with BL is a noninvasive imaging modality that offers better appreciation of FCD lesions and therefore better quality in the extent of resection. The application of these navigational tools greatly aids the reduction of seizure burden and meaningful postoperative clinical outcome.