Metabolic Changes in Medial Temporal Lobe Epilepsy Compared to Healthy Controls Using Advanced [18F] FDG PET SPM Analysis Techniques

Research Aricle

Austin J Nucl Med Radiother. 2023; 8(1): 1032.

Metabolic Changes in Medial Temporal Lobe Epilepsy Compared to Healthy Controls Using Advanced [18F] FDG PET SPM Analysis Techniques

Al Sadi R¹*; Bouhali O²; Djekidel M³

1Department of Science, Texas A & M University at Qatar, Education City, Al Rayyan, Doha, Qatar

2Qatar Computing Research Institute, Hamad Bin Khalifa University, Doha, Qatar

3Senior Author, Northwell, New York, USA

*Corresponding author: Rahaf Al Sadi Department of Science, Texas A&M University at Qatar, Education City, Al Rayyan, Doha, Qatar. Email: rahaf.alsadi@qatar.tamu.edu

Received: March 13, 2023 Accepted: April 17, 2023 Published: April 24, 2023

Abstract

Objectives: [18F] FDG PET imaging has emerged as an established modality in the evaluation of medically refractory temporal lobe epilepsy. Advanced techniques improve the sensitivity and detection of abnormalities. We sought to measure changes in metabolic activity using [18F] FDG PET in Medial Temporal Lobe Epilepsy (MTLE) patients compared to healthy controls and evaluate changes in the Seizure Onset Zone (SOZ) and remote areas.

Methods: We evaluated a cohort of 14 MTLE patients. Following a standard brain [18F] FDG PET acquisition, images were co-registered to a healthy control database using the Neuro-MIM software. Z scores were generated for different temporal and extra-temporal lobe structures for significant p values <0.05. Seizure laterality was determined by experienced epileptologists (>10 years) using intracranial or surface depth electrodes.

Results: The mean age of our patient cohort was 33.9 years (range: 9-52 years). We evaluated 7 left MTLE, 1 bilateral MTLE, and 6 right MTLE patients. Significant hypometabolic changes were seen in the amygdala, the hippocampus, and overall, the medial temporal lobe region (Z score of -2.2, -2.3, and -2.1 respectively). There was a smaller decrease in metabolic activity observed in the parahippocampal gyrus, fusiform gyrus, and the lateral temporal lobe (Z score of -1.4, -1.2, and -1.1 respectively). There were no significant differences in the basal ganglia, thalamus, prefrontal cortex, or cingulate gyrus.

Conclusions: Measurable significant differences in FDG brain metabolic activity exist in medial temporal lobe epilepsy patients compared to healthy controls. Our study shows the most significant changes are hypometabolism in the amygdala, hippocampus, and overall, the medial temporal lobe region compared to the lateral temporal lobe.

Keywords: MTLE; Epilepsy; SOZ; FDG PET

Introduction

Epilepsy is a chronic debilitating condition. Its comorbid impact can be as significant as that of severe diabetes. It has a significant socioeconomic impact on the individual patient as well as a prominent health care cost burden [1-3]. In North America, epilepsy incidence is approximately 50/100,000 per year and prevalence is 5-10/1000 [4]. The CDC estimates that about 2.0 million people in the United States have epilepsy and nearly 140,000 Americans develop the condition each year with an increase in patients with symptomatic epilepsy [4]. The lives of epilepsy patients are negatively impacted at multiple levels, with decreased independent living, and limited financial independence, due to their reduced income and less-likeliness to have full-time employment; they have driving restrictions, neuropsychological impairment, and suffer from persistent stigma in developing as well as developed countries [2-6]. The best studies found an overall standardized mortality ratio of 2.3 for epilepsy relative to the general population [4]. There is a need for more research into ways to improve epilepsy management and global awareness of the disproportionate resources allocated to other diseases and lack of funding into epilepsy progress.

Medical treatment has not been shown to impact clinical outcomes or cost-effectiveness [7-9]. While useful, Antiepileptic Drugs (AEDs) exhibit a substantial long-term health care cost and are associated with debilitating side effects [10,11]. Many epilepsy patients and young females of childbearing age, in whom a pregnancy is contemplated, may be affected directly or indirectly by their epilepsy condition itself or by the adverse effects and teratogenicity related to AEDs [12]. In the case of the one third of epilepsy patients who do not respond to AEDs, respective surgical alternatives or minimally invasive novel intracranial devices may be beneficial.

Since the late 1800’s, surgery for epilepsy has been shown to be successful in seizure control [13-15]. Since then major advances have been made [16-21]. Quality-of-life outcomes are high, and morbidity/mortality post-surgery are low [11,22-37] especially in temporal lobe epilepsy, but are dependent on the preoperative localization of the Seizure-Onset Zone (SOZ) [22,38]. Seizure freedom is usually high at 2 years [39,40]. A prospective, randomized study demonstrated the superiority of surgical versus medical treatment in temporal lobe epilepsy [39]. Surgical success rates can be improved by better noninvasive techniques to define the SOZ, such as what we are proposing here.

The SOZ, while challenging to identify, is traditionally identified by expert epileptologists using several measurement/techniques, each evaluating a certain aspect of the clinical, structural, functional, or electric-magnetic activity of the brain, in order to determine the boundaries of the SOZ. These methods vary widely in their temporal and spatial resolution. Given that there is no single perfect technique that assesses the SOZ reliably with both high temporal and spatial resolution, combining different approaches is currently the mainstay for modern pre-surgical evaluation. This involves inpatient 24-hour video monitoring of seizures while the patient is undergoing additional scalp or Intracranial Electrocorticography (IEC) or both. These data are analyzed in conjunction with MRI, MEG, and functional imaging studies including PET and SPECT, and used to guide final definition of the SOZ with or without intracranial electrode mapping. This process is challenging even in expert hands. It is guided by clinically heterogeneous and imperfect data and patients.

While MRI is used with great success, it is not always informative in cases where no lesion is found, or multiple lesions are seen. Defining a SOZ in the absence of a lesion evident on an MRI scan is challenging and can adversely impact outcomes [26,41-55]. This is where functional PET imaging with [18F] FDG, [11C] Flumazenil, [11C] AMT and [18F] MPPF has proven superior to MRI in identifying the SOZ necessary to be resected [56-81]. Traditionally, PET evaluation is done through visual review and interpretation of the scan. However, semi-quantitative or quantitative brain PET evaluations have proven more useful with numerous radiotracers including FDG. Hence, we aim here to measure and present the changes in metabolic activity using [18F] FDG PET in MTLE patients compared to healthy controls and to evaluate changes in the SOZ (i.e., medial temporal lobe) and remote areas.

Materials and Methods

Fourteen patients (with an age range of 9-52 years) with Medial Temporal Lobe Epilepsy (MTLE) were evaluated. Presurgical evaluation was carried out, whereby the patients were required to have MRI, Stereo-Electro-Encephalography (SEEG), and inter-ictal FDG PET scans performed.

Each patient underwent a PET scan with [18F] FDG 0.1mCi (3.7MBq)/kg. [18F] FDG was administered 35-45 minutes prior to imaging. A 10-minute static scan was acquired using a PET/CT scanner. PET images were reconstructed using an Ordered-Subsets Expectation Maximization (OSEM) algorithm. The appropriate corrections such as detector dead time scatter and random events, and radioactive decay were applied, in addition to attenuation correction using CT scans. The images were then co-registered to the patient’s MRI and a healthy control database using the MIMneuro (MIM Software, USA).

Consequently Z-scores were generated for different temporal and extra-temporal lobe structures for significant p values <0.05. Scans were processed and reviewed by an experienced nuclear medicine physician.

Results

A total of fourteen patients were evaluated with a mean age of 33.9 Years (range: 9-52 years). There were 7 left MTLE, 1 bilateral MTLE, and 6 right MTLE patients. The Z scores from the scans were evaluated.

As shown in (Figures 1 and 2), significant hypometabolic changes were seen in the amygdala, the hippocampus, and overall the medial temporal lobe region (Z score of -2.2, -2.3 and -2.1 respectively).