Pediatric epilepsy networks have both similarities and differences when compared to those in adults [1,2]. This influences pediatric epilepsy surgery techniques and practices like ECoG guided surgical resections. As epileptic networks are younger and less recrudescent in children, resections beyond the MRI abnormality based on electrical aberrations risking deficits, need a strong justification [1,7].
Among the pathological substrates in surgically remediable epilepsy, LEATs has the best outcome and these are even better in children. From an oncological perspective too, LEATs though WHO grade I and II tumors are treated surgically in view of their potential to transform to higher grade tumors [4,5]. Yet, despite supra-marginal resections or extended lesionectomies beyond morphological abnormalities, a seizure outcome failure rate of 8-25% still exists [4,5]. Perilesional tissue with focal cortical dysplasia, gliosis, infiltrating tumor residues etc. are often implicated in these surgical failures. As pediatric epilepsy networks are unique we hypothesised that ECoG utility would also differ in children when compared to adults.
ECoG to compare the pediatric and adult seizure network
Neurons surrounding the LEAT lesion form the epileptogenic zone and thus lesionectomy alone may not guarantee a good seizure outcome. One may argue that if the irritating lesion is removed, the microenvironment may return to normal and the surrounding neurons may cease to discharge abnormally, thereby making ECoG an unnecessary adjunct [3,4,6].
Based on the presumption that neurons surrounding the tumor constitute the epileptogenic zone, removing the tumor alone may not guarantee a good outcome in seizure control [6,10,11]. However, it can also be argued that if the irritating lesion is removed, the microenvironment may return to normal, and the surrounding neurons may cease to discharge abnormally. In pediatric patients, lesionectomy alone yielded very good results. Better results of lesionectomy in children may be due to a shorter history with less opportunity for permanent secondary changes such as hippocampal sclerosis.
The use of ECoG in epilepsy surgery remains controversial both in children and adults [7]. Few studies favor ECoG-guided resections while others have disputed it (Table 5). Few authors advocate its use to assess the 'completeness' of resection. Englot et al in a large meta-analysis of 773 patients with tumor related seizures (100 patients undergoing ECoG) concluded that ECoG offered no significant benefit over lesionectomy alone [12]. Several single-institution retrospective studies have been reported and the long-term benefit of ECoG remains to be convincingly established [12-28]. Qui et al reported 87.5% of patients (n = 32) who underwent ECoG guided tailored epilepsy surgery achieved an Engel I score outcome compared to 63.8% of controls (n = 103). The patients’ ages ranged from 5 to 66 years with an average age of 37 years. They infer that lesionectomy alone may be sufficient for pediatric patients, whereas adults may benefit more from ECoG [17]. Our study did not find any additional utility for ECoG in children. In our adult cohort too, ECoG was not particularly beneficial to achieve better outcomes.
Bansal et al reported a cohort of pediatric patients with tumor associated epilepsy and showed no correlation between ECoG features and improved outcomes [20]. Kim et al found that gross total resection was a better predictor of seizure freedom than ECoG [7]. In 52 pediatric patients, Wray et al reported that the presence or absence of epileptiform discharges on post resection ECoG was not predictive of seizure freedom [21].
Some papers have suggested amplitude, frequency, or morphology of discharges to be predictive of seizure recurrence, although this has not been found in our study. We agree with studies that suggest that the resection of the entire epileptogenic zone defined by EEG may not be necessary to achieve seizure freedom. There isn't enough consensus on how to use ECoG to define the epileptogenic zone [7].
Pediatric versus adult LEATs-Seizure outcome and adverse events
Tailored epilepsy surgery, rather than lesionectomy, yields better seizure outcome for intracranial neoplasms as suggested by Ogiwara et al., 2010; Tripathi et al., 2010; Voorhies and Cohen-Gadol, 2013 [7,18,24]. Guided by intraoperative ECoG, tailored epilepsy surgery targets not only the lesion alone but also the extra or potential epileptogenic foci with abnormal discharges, indicating an aggressive surgical strategy. Although gliosis at the resection site, incomplete resection, haemorrhage, residual cortical dysplasia, and neuronal injury may contribute to postoperative epileptic seizures, “coexistent pathology” or “dual pathology” at the resection borders are likely causes.
In adults due to tenacious epileptic networks, an isolated lesionectomy may not guarantee a good seizure control outcome. Many studies of pediatric patients reported that lesionectomy alone yielded satisfactory results (Kim et al., 2001; Giulioni et al., 2005) [7]. In adult patients, many large series demonstrated that GTR or even extended resection improved seizure prognosis (Awad et al., 1991; Hildebrand et al., 2005; Chang et al., 2008; Englot et al., 2011; Englot et al., 2012; Garcia-Fernandez et al., 2011) [7,12]. It is possible that neoplasms, especially those with long duration, induce secondary epileptogenesis adjacent to or even distant to the primary lesion like secondary hippocampal sclerosis (van Breemen et al., 2007). Better seizure outcome in children may be due to their shorter seizure history, with less opportunity for formation of impregnable epileptic networks. In our series, both pediatric and adult cohorts had similar good outcomes.
Our study also evaluated the post resection ECoG patterns that prompted additional tailored resections and correlated it with the histology of the tailored resection specimen. A post resection 'clean ECoG' and eloquent cortex proximity were the endpoints of the resection. Intermittent non-rhythmic spikes in the resection borders may be distractors, not worth chasing and hence were not followed by extended resections in our study. No additional adverse events related to the tailored resections were noted in both pediatric and adult groups with an acceptable safety profile.
Relevance of the study for epilepsy surgery in LMIC
About 50 million people suffer from epilepsy, of whom 80% live in resource poor countries [29,30]. The prevalence of epilepsy in children ranges from 3.2-5.5/1,000 in developed countries and 3.6-44/1,000 in underdeveloped countries. A gross imbalance exists in the availability of epilepsy surgery services with its availability restricted to about 13-21 % of LMIC and nearly 66% of HIC [30-34]. Menon and Radhakrishnan report that only one in 1000 deserving patients receive epilepsy surgery in India [39]. Therefore, the routine and non-judicious use of ECoG with its additional operation time, strain on resources and lack of sufficient supportive evidence is not often justified [35-39].
Strengths of the study
The study attempts to generate level I evidence by a randomized trial to evaluate the use of ECoG in LEATs and to compare its utility in children and adults. Though several retrospective studies and meta-analysis have looked into the utility of ECoG, this study offers more robust evidence from a single centre prospective evaluation. Many studies have addressed the vast spectrum of epileptogenic lesions but the focus on the unique epileptic network in pediatric LEATs is the hallmark of this study. This study may be of value to expand the outreach of epilepsy surgery in the developing world by advocating a judicious use of ECoG thereby making LEATs resections possible outside the realms of specialized centers.
Study Limitations
This is a single-center series from a comprehensive epilepsy care program. This limits its generalizability to non-specialised centres in terms of expertise in surgical candidacy identification.A longer duration of follow up would have helped to assess late failures and secondary epileptogenesis in rescrudescent epilepstic networks. A correlation of pre-resection intraoperative ECoG patterns with seizure outcome benefits could be of more value. Impact of intraoperative ECoG recordings on long term prognostication, cost of care, resource, optimum operating room time utilization and planning withdrawal of anti-convulsants are the many aspects unexplored by this study.Despite the limitations, the study firmly asserts the limited role of ECoG in pedaitric and adult LEATs lesionectomies and therefore advocates its judicious use in resource restricted LMICs.