37 Postoperative Seizure Control

10.1055/b-0034-84149

37 Postoperative Seizure Control

Adhami, Seema, Harini, Chellamani

Epilepsy is a common problem in pediatric neurology. In 1975, Davidson and Falconer first showed that surgery could alter the outcome of children with temporal lobe epilepsy (TLE).1 The last two decades have seen the emergence of surgery as a valuable and viable option in the treatment of refractory focal epilepsy in children. Ten to 20 percent of cases of pediatric epilepsy fall into this category.2,3 Although much of the early literature on postoperative seizure control refers to adults with epilepsy, the importance of considering surgical treatment in children with intractable epilepsy is well recognized. Issues in pediatric epilepsy surgery differ from those in adults in several respects, including factors that may potentially affect seizure outcome. Children have a higher proportion of neoplastic lesions and malformation of cortical development in the epileptogenic substrate and more frequent extratemporal epileptogenic foci.2,4–8 Surgery on the developing brain may in itself affect postoperative seizure control in children.

Why is seizure control important as an outcome measure after epilepsy surgery? An obvious answer is that it enables physicians to consider the option of surgical treatment for an individual patient, in addition to providing a prognosis to families presented with the option of surgery for their child with refractory epilepsy. The ultimate aim of any treatment for epilepsy is to improve quality of life. Adult studies have shown that postoperative seizure control is an important predictor of quality of life.9–11 Seizure frequency has also been related to an increased risk of sudden unexpected death in adults.12 In children, as opposed to adults, an additional factor is the potentially deleterious effect of recurrent seizures on the developing brain and long-term side effects of antiepileptic drugs on achieving developmental potential.

In this chapter, we address postoperative seizure control in terms of what is known regarding the outcome of epilepsy surgery with respect to subsequent seizure control and factors that may affect postoperative seizure control. This is followed by a discussion of the medical management of epilepsy in children after epilepsy surgery.

Outcome with Respect to Seizure Control after Surgery

The most widely accepted classification of postoperative seizure outcome is the scheme proposed by Engel.13 This scheme stratifies postoperative outcome into four classes, Class I to Class IV. Patients in Class I are free of disabling seizures (excludes early postoperative seizures) and are further subdivided into four categories: (a) completely seizure free since surgery; (b) nondisabling, simple partial seizures or auras only since surgery; (c) some disabling seizures after surgery, but free of disabling seizures for at least 2 years; (d) generalized convulsion with antiepileptic drug withdrawal only. Patients in Class II have rare disabling seizures. This group is divided into four subgroups: (a) initially free of disabling seizures but have rare seizures now; (b) rare disabling seizures since surgery; (c) more than rare disabling seizures after surgery, but rare seizures for the last 2 years; (d) nocturnal seizures only. Class III describes patients with worthwhile improvement and includes patients with worthwhile seizure reduction, or prolonged seizure-free intervals amounting to greater than half the follow-up period, but not less than 2 years. Patients in Class IV have no worthwhile improvement in seizures. Favorable postoperative seizure control is generally considered as an Engel Class I or Class II outcome.

Systematic evaluation of surgical outcome data in children is limited by the heterogeneity of the surgical groups, both in terms of the surgical procedure (e.g., temporal resection, extratemporal resection, hemispherectomy, multiple subpial transection, corpus callosotomy) and the underlying cause of the seizures or epilepsy syndrome. The numbers that fall into a similar category are, therefore, not large. Potential confounders in trying to assess postoperative seizure control may be the different proportion of children in various age groups (between birth and 18 years) in each series, and, more importantly, the duration of postsurgical follow-up. Although a significant body of literature on postoperative seizure control has emerged in the last three decades, there is a lack of prospective, randomized, and controlled studies in children treated surgically for epilepsy.

Postoperative seizure control reported in seven pediatric series14–20 with reasonable sample size and duration of follow-up is summarized in Table 37.1 . The average postoperative follow-up in these children ranged from 2.7 years to almost 11 years; 67 to 81% were seizure free after surgery, and a higher proportion, 74 to 86%, had a favorable seizure outcome (Engel Class I or II).

Seizure Control after Surgery for a Structural Lesion

Refractory seizures in children caused by lesions identified on imaging are unlikely to remit with time.21 The location of the lesion, particularly temporal versus extratemporal, has also been implicated as a factor with potential influence on seizure control after epilepsy surgery.14,18–20

Seizure Control after Temporal Lobe Surgery

Evidence from retrospective studies in children treated surgically for intractable epilepsy over the last two and a half decades shows that 58 to 74% of children become seizure free. A larger percentage, 67 to 82%, show favorable seizure control.3,7,8,20,22,23 Within the last few years, randomized controlled studies have shown the efficacy of surgery compared with medical therapy in TLE in adults.24,25 Although there are no such randomized controlled trials in children, several pediatric series have documented the effectiveness of temporal resection for seizure control.16,17,26–28

In a large series reported by Benifla and colleagues ( Table 37.1 ), of the 106 children treated surgically for TLE, 74% were seizure free (Engel Class I) at a mean follow-up period of 67 months. This included 7 who became seizure free after a second procedure. Two additional patients were in Engel Class II, giving a favorable seizure outcome in 75% of patients in this series. Twenty-six patients (25%) had a less favorable outcome (Engel Class III and Class IV). In those who relapsed, the mean time to seizure recurrence was 15 months after surgery.17 In another cohort of 109 patients who underwent temporal resection and for whom follow-up data for a minimum of 5 years were available, 81.7% became seizure free (Engel Class I), and an additional 4.6% satisfied criteria for Engel Class II, resulting in favorable seizure control in 86% of the 109 patients.16

Features associated with an unfavorable outcome after temporal resection are the lack of a structural abnormality on magnetic resonance imaging (MRI), development-associated with disease, widespread disease documented by postoperative electroencephalography (EEG; occult dual disease involving the same temporal lobe, adjacent frontal lobe, or bitemporal disease), residual tumor or malignant transformation, and the need for emergency temporal lobectomy secondary to complication of invasive presurgical monitoring.20

Seizure Control after Extratemporal Resection

Seizure outcome is usually better after temporal resection than after extratemporal resection ( Table 37.2 ) with 72 to 88% seizure-free rates after temporal resection versus 54 to 60% patients becoming seizure free after extratemporal resection.14,18–20 Nevertheless, in most well-selected patients, surgery for extratemporal lesions provides an improvement in seizure control, with more than 50% of patients achieving a seizure-free status ( Table 37.2 ).

Kim and coworkers ( Table 37.2 ) found that 88% of the patients who had temporal lobe resection were seizure free, and 90% had favorable seizure outcomes. By comparison, 55% who had extratemporal resection became seizure free, and 64% had a favorable seizure outcome. This finding was despite more invasive presurgical studies in the extratemporal resection group.20 The mean duration of follow-up was more than 5 years in this study. A smaller series reported by Gilliam and colleagues, with mean follow-up of 2.7 years, showed a 72% seizure-free rate after temporal resection versus 60% seizure free after frontal lobe resection.14 Another cohort of 113 children from Italy18 showed that 54% of the patients became seizure free after extratemporal resection. After temporal resection, 91% of the patients were seizure free. The mean duration of follow-up was 55 months.18

Factors associated with an unfavorable outcome in those who had extratemporal resection were similar to the factors identified in children who had temporal resection, namely, the lack of a structural abnormality on MRI, development associated disease, more widespread disease than suspected preoperatively, residual tumor, or malignant transformation. Additional factors associated with an unfavorable outcome after extratemporal resection were multilobar resection, limited resection because of the presence of functional cortex, incomplete lesionectomy, or tumor progression.20

Seizure Control after Hemispherectomy

Successful postoperative seizure control depends primarily on removing the cortical substrate likely responsible for the seizures. Hemispherectomy has been referred to as the ultimate focal resection and is usually performed in children with diffuse hemispheric pathologies. This includes development related disorders (hemimegalencephaly, diffuse and focal cortical dysplasia [CD], polymicrogyria, microdysgenesis), acquired conditions that may be perinatal or postnatal (infarction/ischemia, infection, trauma, hemiconvulsive–hemiplegic epilepsy), or progressive disorders such as Rasmussen encephalitis and Sturge-Weber syndrome.29,30 Seizure freedom after hemispherectomy is 70 to 80% in well-selected patients. As most of the hemisphere is removed, pathology may not matter.20,31,32 In the UCLA series, which included a large percentage of cortical dysplasias and other non-CD destructive lesions, 64% of the 62 patients who had hemispherectomy showed an Engel Class I outcome at 2 years.33 Seizure-free rates of 54 to 82% have been seen after hemispherectomy for infantile hemiplegia, Rasmussen encephalitis, and Sturge-Weber syndrome.34–37 Table 37.3 gives a summary of seizure outcome after hemispherectomy for the treatment of intractable epilepsy in three studies.29,30,38 In the series reported by Devlin and colleagues,30 seizure freedom was 82% in those with acquired pathology (11 of 33 cases), 50% in those with progressive disorders (6 of 33 children), and 31% in those with developmental pathology (16 of 33). A worthwhile reduction in seizures was seen across all groups.30 By contrast, in the series from the Cleveland Clinic Foundation, the etiology of catastrophic epilepsy in 18 children younger than 2 years old who underwent hemispherectomy was not associated with the outcome of seizures after surgery. Incomplete disconnection was the only factor related to seizure outcome in this cohort.38

**Table 37.1** Postoperative Seizure Outcome in Some Pediatric Epilepsy Surgery Studies
Authors	Number of Patients	Age at Surgery, Mean (Range)	Age at Seizure Onset Mean (Range)	Duration of Epilepsy before Surgery	Duration of Follow-up Mean (Range)	Postoperative Seizure Control According to Engel Classification, % (number)
						Engel I % (No.)	Engel II % (No.)	Engel III % (No.)	Engel IV % (No.)
Gilliam et al14	33	7.75 years (8 months–12 years)	2.2 years (6 days–5 years)	—	2.7 years (7 months–6 years)	67% (22)	9% (3)	12% (4)	12% (4)
Bourgeois et al15	200	8.3 years (10 months–15years)	— (6 weeks–12 years)	2.5 years (1 months–12 years)	5.2 years (18 months–17 years)	71.3%	7.2%	2.9%	4.7%
Cossu et al16	109	13.2 years (2 months–8.9 years)	5.5 years (birth–17years)	7.7 years (01.–17.6 years)	10.9 years (5-10.2 years)	81.7%	4.6%	13.8% (15/109)
Benifla et al17	106	13.5 years	5.9 years (1 month–17 years)	5.6 years	67 months (2 years–13 years)	74% (78)	1.8% (2)	5.6% (6)	19% (20)
Massimo et al18	113	8.8 years (1–15 years)	3.1 years (0–15 years)	5.7 years (0–14 years)	55.1 months (24-115 months)	68% (77)	9% (10)	10% (11)	13% (15)
Kan et al19	58	11.2 years (1.8–21 years)	—	4.8 years (6 months–15years)	4.7 years (1–8 years)	74% seizure free
Kim et al20	134	8.5 years (8 months–18years)	4.3 years(birth–14years)	—	62.3 months (12-168 months)	69% (93)	5% (7)	7% (9)	19% (25)
Source: Gilliam F, Wyllie E, Kashden J, et al. Epilepsy surgery outcome: comprehensive assessment in children. Neurology 1997;48(5):1368–1374; Bourgeois M, Sainte-Rose C, Lellouch-Tubiana A, et al. Surgery of epilepsy associated with focal lesions in childhood. J Neurosurg 1999;90(5):833–842; Mittal S, Montes JL, Farmer JP, et al. Long-term outcome after surgical treatment of temporal lobe epilepsy in children. J Neurosurg 2005; 103(5, suppl):401–412; Benifla M, Otsubo H, Ochi A, et al. Temporal lobe surgery for intractable epilepsy in children: an analysis of outcomes in 126 children. Neurosurgery 2006;59(6):1203–1213, discussion 1213–1214; Cossu M, Lo Russo C, Francione S, et al. Epilepsy surgery in children: results and predictors of outcome on seizures. Epilepsia 2008;49(1 ):65–72; Kan P, Van Orman C, Kestle JRW. Outcomes after surgery for focal epilepsy in children. Childs Nerv Syst 2008;24(5):587–591; Kim SK, Wang KC, Hwang YS, et al. Epilepsy surgery in children: outcomes and complications. J Neurosurg Pediatr 2008;1(4):277–283

**Table 37.2** Seizure Control after Temporal and Extratemporal Resection in Children Treated Surgically for Epilepsy
Author (Number of Patients)	Temporal Resection		Extratemporal Resection
	n	% Seizure Free	n	% Seizure Free
Gilliam et al14 (n = 33)	18	72%	15	60%
Massimo et al18 (n = 113)	43	91%	70	54%
Kan et al19 (n = 58)	33	85%	25	60%
Kim et al20 (n = 134^a)	59	88%	56	55%
^aNineteen patients who had functional hemispherectomy, corpus callosotomy, or multiple subpial transaction, are not included in table.

Seizure Outcome in Relation to Pathology of the Epileptogenic Substrate

Low-grade tumors, malformations of cortical development (MCD), mesial temporal sclerosis (MTS), vascular malformations, hypoxic ischemic insults (including in utero ischemic lesions) are all causes of focal epilepsy in children.14,16–20 Table 37.4 shows the relative distribution of postoperative pathology reported in six pediatric series and whether pathology is correlated with postoperative seizure control in these patients. Low-grade tumors and MCD, together as well as independently, constitute the most common pathologies identified postoperatively. MTS as an isolated lesion is seen much less often in children (refer to Table 37.4 ). An exception to this is the series reported by Mittal and colleagues16; 45% (49/109) of patients in this series showed MTS and formed the largest single histopathological diagnosis.16 Among tumors, ganglioglioma, dysembryoplastic neuroepithelial tumor, low-grade astrocytoma, and mixed gliomas are the most frequent, and a smaller number of other tumors, such as pilocytic xanthoastrocytoma, oligodendroglioma, and choroid plexus papilloma, were identified.14,16–20 In the larger cohorts of children treated for epilepsy by surgical resection, up to 25% of children show more than one pathological finding, that is, dual or multiple pathology.16,18,20

**Table 37.3** Postoperative Seizure Control after Hemispherectomy
Author (ref)	N	Age at Surgery Mean/Median (Range)	Follow-up Mean/Median (Range)	Seizure Outcome, %
				Seizure Free	Rare Seizures	Worthwhile Improvement	No Improvement
Vining et al29	54	7.1 years	6.2 years	54%	24%	17%^a	6%^b
Devlin et al30	33	4.25 years (0.33–17 years)	3.4 years (1–8 years)	52%	9%	30% > 75% seizure reduction	9% < 75% or no seizure reduction
Gonzalez-Martinez et al38	18	11.7 months (3–22 months)	34.8 months (12–74 months)	66%		22% > 90% seizure reduction
^aModerate frequency and severity of seizures that interfered to some extent with function.

**Table 37.4** Pathology and Postoperative Seizure Control
Author (Number of Patients)	Neoplasm % (Number of Patients)	MCD % (Number of Patients)	MTS % (Number of Patients)	Multiple Findings % (Number of Patients)	Gliosis % (Number of Patients)	Miscellaneous % (Number of Patients)	No Abnormality % (Number of Patients)	Relation of Pathology to Seizure Control after Surgery
Gilliam et al14 (n = 33)	36.4% (12)	33.3% (11)	3% (1)	3% (1)	6% (2)	12% (4)	6% (2)	Pathology not related to seizure outcome
Mittal et al16 (n = 109)	35% (38)	35% (38)	45% (49)	25%	—	4.6% (5)	5.5% (6)	Pathology as an independent variable related to seizure outcome
Benifla et al17 (n = 126)	51% (64)	10% (13)	13% (16)	8% (10)	12% (15)	2% (3)	3% (5)	Better seizure outcome with neoplasm
Massimo et al18 (n = 113)	38% (43)	83% (94)	10% (11)	24% (27)	—	4% (5)	2% (2)	Better seizure outcome with glial neuronal tumor versus other path
Kan et al19 (n = 58)	28% (16)	22% (13)	28% (16)	5% (3)	—	17% (10)	—	Seizure outcome best with neoplasm and MTS
Kim et al20 (n = 124; n for study = 134, no path specimen in 10)	36% (45)	57% (71)	15% (19)	21% (26)	—	—	—	Tumor better outcome than MCD; worse outcome with dual pathology.
Abbreviations: malformations of cortical development, MCD, mesial temporal sclerosis, MTS.

As noted in the last column of Table 37.4 , better postoperative seizure control is achieved when the causative lesion is a low-grade tumor or isolated MTS.16,17,19 Seizure control is better after resection of glial neuronal tumors compared with other pathologies.18 Tumors, in general are associated with better postoperative seizure control than MCD. The presence of more than one pathology correlates with a worse seizure outcome.19,20

Among children with epilepsy, the most intractable are the youngest patients with cortical dysplasia, tuberous sclerosis, or other structural lesions.39,40 The structural lesion in these cases is often more extensive than in TLE, which has become the prototypical focal epilepsy syndrome that shows good seizure control after surgery. A study by Mathern and coworkers from UCLA compared seizure outcomes in children with symptomatic seizures from large unilateral CD and non-CD pathologies (e.g., infantile hemiplegia, perinatal stroke, infection, Rasmussen encephalitis, Sturge-Weber syndrome and miscellaneous) with TLE patients who had more focal pathology. Seizure outcomes at 6, 12, and 24 months and 5 years and 10 years after surgery were assessed.33

The mean presurgical seizure frequency was significantly greater in the CD and non-CD groups than in the group with TLE. There was no difference in postoperative seizure frequency among CD, non-CD, and TLE patients at 6, 12, and 24 months and at 10 years. The percentage of seizure-free patients at 2 years after surgery was similar in CD (65% of 64 children), non-CD (56% of 71 children) and TLE (65% of 31 children). At 5 years after surgery, seizure frequency was significantly greater in CD cases compared with TLE. The cause was unclear. The authors believed that most of the recurrent seizures were complex partial seizures and were less debilitating than the original seizures. The number of patients who had follow-up at 5 years and 10 years, however, was small.

There is some evidence to suggest that early and complete surgical resection of CD lesions may contribute to better postoperative seizure control in children with CD than in adults with CD.41–46

Only gold members can continue reading. Log In or Register to continue