Evolution of epilepsy surgery in childhood: the neurologist’s point of view

ARTICLE

Paper presented at the World Conference on Pediatric Neurosurgery 2000: State of the Art and Perspectives for the Third Millenium, Martinique, 27 November to 4 December 1999.

Surgical therapy of epilepsy in adults has now gained full acceptance, at least in developed countries. It has the advantage, when successful, of being a definitive treatment not only preventing the harmful consequences of uncontrolled, repeated seizures but also allowing the use of smaller dosages of antiepileptic drugs or even their discontinuation. Such benefits would appear to be of even greater importance for children in whom the benefits could extend over a lifetime. In spite of these potential advantages, epilepsy surgery in children has been slow to develop and the interval between onset of the disease and surgery in patients is still often very long, 10-15 years or more [1]. Nevertheless, over the past two decades, considerable work has been dedicated to epilepsy surgery in young patients [2-5]. It has been recognized as both justified, because the outcome in childhood epilepsies became predictable, and possible, thanks to the introduction of new techniques of exploration, especially modern neuroimaging.

As a result, epilepsy surgery programmes have been launched in most major paediatric centres. Application of surgical therapy to childhood epilepsies, however, raises problems different to those in adults not only for technical reasons, but even more because the disorder occurs in a developing, constantly changing organism and can modify its ultimate functioning, and because early-onset epilepsies may differ considerably from their adult counterparts. This article analyses the causes of these differences and considers some of the unresolved issues raised by paediatric epilepsy surgery.

Required conditions for surgical treatment

Three conditions are necessary for surgical treatment of epilepsy, whether in adults or in children: 1) medical intractability of the disorder; 2) feasibility of surgery; 3) likelihood of a satisfactory result, in terms of both cure or improvement of the disorder and reasonable preservation of brain functions. Practical implementation of these conditions can be considerably different in adults and children.

Intractability, defined as "inadequate seizure control in spite of appropriate medical therapy with antiepileptic drugs or adequate seizure control but with unacceptable side-effects from the AEDs" [6, 7] is a rather vague concept. What constitutes "inadequate seizure control" or "unacceptable side-effects" depends not only on biological facts but on an appreciation by the patient that may vary according to her/his lifestyle and personal objectives. It is therefore a joint decision of the patient and doctor taking into account both medical and personal factors, and is therefore highly individual.

In children, the issue is even more complicated and many factors specific to infant and childhood epilepsies make a simple transposition of rules in adults inappropriate. Even though epilepsies after 4-5 years of age, especially partial symptomatic ones, can be approached in the same general way as in adults as they raise generally similar problems although the possibilities and techniques of presurgical exploration may differ, the situation is radically different in many other cases [8]. One major problem is the stability of the epilepsy. It has long been known that some children "grow out" of their epilepsy and are thus not candidates for epilepsy surgery. Considerable work over the past 20 years has established criteria that separate the benign epilepsy syndromes from the severe or malignant ones [9]. It is now possible to know, early in the course, whether persistence of seizures is likely and whether epilepsy will interfere with cognitive and behavioural development, and therefore to establish a firm prognosis. Syndromes with excellent prognoses such as benign rolandic epilepsy or benign occipital epilepsy [10] can now be easily distinguished from more severe cases and there is little risk of advising surgery in such cases.

The characterization, from clinical and EEG points of view, of malignant epilepsy syndromes such as West syndrome, Lennox-Gastaut syndrome, severe myoclonic epilepsy or polymorphic epilepsy of infants casts a different light on infantile and early childhood epilepsies. In this new perspective, seizures are not necessarily the only or even the main problem; cognitive/behavioural development is of paramount importance. This has to be taken into account in the therapeutic decision and, even though the role of surgery in this respect remains to be established, an early interruption of the epileptic activity is clearly desirable. Most infant and childhood epilepsies are severe, with multiple, daily seizures and specific EEG changes, so the intractable nature is rapidly determined without the necessity of waiting 2 or 3 years [11].

Feasibility. Several studies have shown that epilepsy surgery is perfectly feasible in children and infants [12-19] provided a trained team including child neurologists, electroencephalographers, neuropsychologists and specialists in imaging in young patients is available, in addition to the neurosurgeon(s) and anaesthesiologists. However, the types of surgery to be performed are often different from those in adults because the lesions responsible for epilepsy syndromes are often extensive (e.g. involving more than one lobe or a whole hemisphere), leading to large resections especially hemispherectomies and similar techniques. For several syndromes, such as Landau-Kleffner [20], or some cases of bilateral epileptogenesis, only palliative surgery such as corpus callosotomy can be considered [21]. In general, the pathology responsible for the epilepsy is different in adults and children. Hippocampal sclerosis, for example, is less common in children than in adults [15, 16], although recent work has shown that it is not rare [24]. Tumours are much less common in children and are mostly represented by developmental lesions such as ganglioglioma [22, 26, 27] or developmental neuroepithelial tumours (DNET) [24] that do not pose oncological problems as they are static or very slow growing, but pose mostly an epilepsy surgery problem, i.e. that of the extent of resection necessary for an optimal result. Cortical dysplasias are the most common cause of refractory epilepsy at this age [16, 19]. Their frequency may be responsible for the common occurrence of extratemporal lesions that pose difficult problems. In addition, the extent of cortical dysplasias is often difficult to define with current techniques. The possibility of other subtle, undetectable, abnormalities in the rest of the brain [25] may be one reason for the relatively poor surgical results in some series. These conditions remain a challenge for epileptologists and surgeons, to improve the techniques of diagnosis (e.g. by improvements in anatomical and functional imaging) as well as operative methods.

The approach of such lesional epilepsies has been revolutionized by the new techniques of neuroimaging that play a decisive role in modern epilepsy surgery [29]. MRI is the most commonly used of these techniques and is being constantly improved. Functional imaging (positron emission tomography (PET), single photon emission computed tomography (SPECT), magnetic resonance spectroscopy (MRS) and functional MRI (fMRI) are becoming increasingly important for determination of epileptogenic areas and of functionally eloquent cortex. Modern computing techniques allow the combination of several modalities in the same image and comparison with mean results from groups of normal subjects (statistical parametric analysis) [30]. Progress in imaging combine with EEG techniques to give a precise definition of the epileptogenic zone(s). Advanced scalp EEG methods and combination with video-recording are widely available and a large variety of invasive recording techniques can be used. Their feasibility in young children and in behaviourally disturbed patients, however, continues to raise problems and still represents a limitation in the investigation of some cases of epilepsy with negative MRI and therefore in the feasibility of surgery in these cases.

Likelihood of satisfactory results. Results of epilepsy surgery on control of the seizures from various centres are comparable to those of most adult series [15, 19, 26]. It is, however, important to separate results in older children in whom surgery is not fundamentally different from that in adults, from those in infants and younger children less than 3-4 years of age for whom the indications and surgical techniques are often very different.

Such information is still relatively limited [8, 16, 18, 19]. Excellent results (Engel grade I) are reported in 35% [16] to 61% [15, 19] of infants less than 3 years of age at the time of surgery. Results are better for cases of tumours or Sturge-Weber syndrome than for cortical dysplasias or hemimegelancephaly in some, but not all, series [16, 31]. As in older patients, the results of surgery in general are significantly better when a dectectable lesion can be demonstrated by neuroimaging. The results of temporal lobe surgery are better than those of extratemporal resection, and those of hemispherectomy seem superior to those of focal resection in some series. In fact, comparison between reported studies are difficult because of the heterogeneous composition of series, different methods of selection and investigation, and variable causes of seizures. Furthermore, the very aim of surgical therapy in young infants need not be the same as in older patients.

Indeed, total control of seizures has been shown in older patients to be an absolute requirement for an excellent outcome in terms of quality of life [31]. In children, it may be especially important that control is obtained early, as the psychosocial consequences of epilepsy and its treatment are of great significance in terms of ultimate development and social insertion. For infants, however, other aims of surgery are conceivable. One would be the prevention of cognitive and behavioural deterioration or arrest of development, if it can be demonstrated that even subclinical epileptic activity (especially some EEG patterns such as hypsarrhythmia or continuous spike-waves of sleep) may be responsible and respond to surgery. Incidentally, this would raise the question of whether to operate on infants, even in the absence of seizures, in cases of progressive deterioration with such severe EEG patterns. Another objective would be the palliation of certain consequences of epilepsy, for example repeated falls, that are extremely disruptive for the patient and family life, even when curative surgery cannot be contemplated. Operations such as commissurotomy and subpial transections could be considered for such aims. Such a widening of the potential scope of surgery would have to be assessed on criteria different from those in conventional resective surgery. Establishment of different categories for measuring the outcome of epilepsy surgery in children, based not only on seizure control but on cognitive/behavioral improvement (or arrest of deterioration) and on improvement in the overall quality of life would be required, especially for the unfavourable categories for which complete seizure control may be unattainable but significant alleviation of the worst consequences of the disorder might be achievable. Taylor et al. [31] have proposed such a scheme and further reflection along these lines is needed.

Future perspectives

Even though successes of surgery in the treatment of epilepsy have been gratifying, there is ample room left for further research and improvement.

One important topic for research is assessment of the results of surgery. The significance of currently published results is hard to evalulate as many studies are obviously not comparable and the results probably depend as much or more on the criteria for acceptance of surgical candidates than on the methods of investigation or operative techniques used. Centres that operate on relatively "poor" candidates such as patients with mental retardation, cerebral palsy or extensive lesions will have less favourable statistics than those that accept only rigorously selected patients of the "best" category. This has a clear impact on the extent to which surgical therapy can be used: should it be reserved for a selection of relatively unusual patients with a high probability of good control and great benefit, in terms of quality of life and social insertion, or is it acceptable to include a larger number of patients with a less favourable prognosis both in terms of seizure control and of quality of results? It is clearly difficult to give an absolute answer as the issue is not purely medical but has to be also evaluated according to the cost to society and burden imposed on patients.

A somewhat similar issue arises with respect to the indications of the multiple methods of investigation for surgery. While clinical examination is of accepted importance and the depth of investigation is very much guided by the individual problem of each patient, there is still some debate about the more or less completeness and technicity required. For example, is invasive EEG recording (SEEG or otherwise) necessary for all or a majority of cases or can investigation be simplified in a significant proportion? Again, there is no simple answer, in large part because comparisons are not available which take into account not only the crude statistics but also the comparability of operated patients. Some studies indicate that, even with a well-defined lesion, the use of operative intracranial recording improves the results [15, 23, 30, 35] but the cost of complicated techniques is great both in terms of resources, human and material, and of human factors such as duration of hospitalisation and repeated investigations. The actual benefit is difficult to measure as the degree of experience of a team may be as important for some matters as the technical tools they use. I believe that a considerable effort should be made to try to devise methods that could help answer these questions.

The place of techniques other than resection surgery in treatment also demands further research. Such methods do not directly compete with resection in terms of control of seizures, so their possible benefit requires a different assessment. Some may have a relatively high degree of effectiveness for control of seizures and can therefore be assessed like more classical techniques, e.g. amygdalohippocampectomy which may be used palliatively in some temporal lobe tumours. Others, such as commissurotomy, are basically meant to limit the damage from seizures and have a rather narrow scope. Their usefulness appears to be relatively limited and better techniques should be looked for. Still other methods, especially subpial transection, are being used in several centres for the treatment of epilepsies in which the epileptogenic area cannot be resected without major functional deficits. The underlying assumption is that recruitment necessary for generating an epileptic discharge is mediated by horizontal fibres that are interrupted by the procedure, whereas vertical fibres that mediate the functions are left untouched [20]. The effectiveness of the procedure is still being evaluated. An additional interest of this method is to suggest that some surgical procedures might have a positive effect on functional deficits not directly related to clinical seizures. Suppression of the underlying epileptic discharge would allow resumption of the normal function suppressed by the massive and unmodulated paroxysmal activity. A possible mechanism to explain the persistence of deficit after a long-lasting period of epileptic activity even without clinical seizures, is that the epileptic activity as expressed by the EEG can modify the connectivity of the cortical areas involved and interfere with their specialisation. Clearly, much more needs to be known about possible mechanisms, but the possibility of interfering surgically with an abnormal function in order to restore a more normal activity is both intriguing and exciting as cognitive and/or behavioural deterioration is one of the major issues in infant epilepsy. Whether the method is effective and whether it can be extended to other areas and pathologies remains to be determined. Finally, vagus nerve stimulation is being intensively studied for patients not amenable to any form of surgical treatment and preliminary results seem interesting [7].

CONCLUSION

The recent developments in surgical treatment of epilepsy in childhood suggest that its indications may need to be redefined in several respects.

Firstly, the good results obtained lead most investigators to propose surgery as an early, if not a first option. This would apply especially to some syndromes such as mesial temporal lobe epilepsy following febrile seizures for which surgery could even be considered as indicated, as a first treatment or at least after only a brief attempt at medical control. Clearly, there is no reason to procrastinate when a definitive treatment is available with a high likelihood of cure. However, a more complete assessment of the possible adverse effects of resection, especially on memory [17], is necessary before this approach can be recommended.

Secondly, the domain of surgery may possibly be extended to patients so far excluded. These could include patients with disorders in which localisation of the responsible lesion(s) is currently imprecise, such as tuberous sclerosis, patients with multiple disabilities for whom even partial relief of seizures can improve their quality of life, and some of those with threatening or ongoing cognitive behavioural deterioration such as infantile spasms, the Lennox-Gastaut syndrome and other epileptic encephalopathies, if adequate methods of exploration and treatment can be devised.

Thirdly, efforts should be made to reduce the cost of surgery, both in terms of personal suffering and of human and financial resources. This implies a minimum agreement on the best techniques for specific problems, thus the possibility of comparative studies among the different methods.

REFERENCES

1. Aicardi J. Paediatric epilepsy surgery: how the view has changed. In: Tuxhorn I, Holthausen H, Boenigk H, eds. Paediatric epilepsy syndromes and their surgical treatment. London: John Libbey, 1997; 3-6, 894 p.

2. Engel J Jr. Surgical. Treatment of the epilepsies, 2. New York: Raven Press, 1993.

3. Engel J Jr, Pedley TA, eds. Epilepsy: a comprehensive textbook. Chapt VII Surgical Therapy 1998: 1673-910.

4. Lüders H, ed. Epilepsy surgery. New York: Raven Press, 1992.

5. Tuxhorn I, Holthausen H, Boenigk H, eds. Paediatric epilepsy syndromes and their surgical treatment. London: John Libbey, 1997.

6. Bourgeois BFD. General concepts of medical intractability. In: Luders HO, ed. Epilepsy surgery. New York: Raven Press, 1992: 77-81.

7. Shields WD. Defining medical intractability: the differences in children compared to adults. In: Tuxhorn I, et al., eds. Pediatric epilepsy syndromes and their surgical treatment. London: John Libbey 1997: 93-8.

8. Arts WFM, Geerts AT, Brouwer OF, Peters ACB, Stroink H, van Donselaar CA. The early prognosis of epilepsy in childhood: the prediction of a poor outcome. The dutch study of epilepsy in childhood. Epilepsia 1999; 40: 726-34.

9. ???

10. Panayiotopoulos CP. Benign childhood partial seizures and related epileptic syndromes. London: John Libbey, 1999; 406 p.

11. Sugimoto T, Otsubo H, Hwang P, Hoffman HJ, Jay W, Snead OC III. Outcome of epilepsy surgery in the first three years of life. Epilepsia 1999; 40: 560-5.

12. Chugani HT, Shewmon DA, Shields WD. Surgery for intractable infantile spasms: neuroimaging perspectives. Epilepsia 1993; 34: 764-71.

13. Chugani HT, Kupsky WJ, Chugani DC. Cortical dysplasia: surgical treatment and neuropathologic findings in infants and children. In: Guerrini R, et al., eds. Dysplasias of cerebral cortex and epilepsy. Philadelphia: Lippincott-Raven, 1996: 427-33.

14. Duchowny M, Levin B, Jayakar P. Temporal lobectomy in early childhood. Epilepsia 1992; 33: 298-303.

15. Duchowny M, Jayakar P, Resnik T, Harvey AS, Alvarez L, Dean P, et al. Epilepsy surgery in the first three years of life. Epilepsia 1998; 39: 737-43.

16. Hirabayashi S, Binnie CD, Janota I, Polkey CE. Surgical treatment of epilepsy due to cortical dysplasia: clinical and EEG findings. J Neurol Neurosurg Psychiatry 1993; 56: 765-70.

17. Szabo CA, Wyllie E, Stanford D, Geckler G, Kotagal P, Comair YG, Thompson AE. Neuropsychological effect of temporal lobe resection in preadolescent children with epilepsy. Epilepsia 1998; 39: 814-9.

18. Vickery BG, Hays RD, Engel J Jr, et al. Outcome assessment for epilepsy surgery: the impact of measuring health-related quality of life. Ann Neurol 1995; 37: 158-66.

19. Wyllie E, Comair YG, Kotagal P, Raja S, Ruggieri P. Epilepsy surgery in infants. Epilepsia 1996; 37: 625-37.

20. ???

21. Blume WT. Corpus callosotomy: a critical review. In: Tuxhorn I et al., eds. Pediatric epilepsy syndromes and their surgical treatment. London: John Libbey, 1997: 829-95.

22. Duchowny M, Altman N, Bruce J. Dysplastic gangliocytoma of the cerebral hemisphere. In: Guerrini R, et al., eds. Dysplasias of the cerebral cortex and epilepsy. Philadelphia: Lippincott-Raven, 1996: 93-100.

23. Rassi-Neto A, Ferraz FP, Campos CR, Brago FM. Patients with epileptic seizures and cerebral lesions who underwent lesionectomy restricted to or associated with adjacent irrritative area. Epilepsia 1999; 40: 856-64.

24. Harvey AS, Grattan-Smith JD, Desmond PM. Febrile seizure and hippocampal sclerosis: frequent and related findings in intractable temporal lobe surgery in childhood. Pediatr Neurol 1995; 12: 201-6.

25. Raymond AA, Fish DR, Sisodiya N, Alsanjari N, Stevens J, Shorvon SD. Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy. Clinical, EEG and neuroimaging features. Brain 1995; 118: 629-60.

26. Zentner J, Wolf HK, Ostertun B. Gangliomas: clinical, radiological and histopathological findings in 51 patients. J Neurol Neurosurg Psychiatry 1994; 57: 1497-502.

27. Raymond AA, Halpin SFS, Alsaujari N, Cook MJ, Kitchen ND, Fish DR. Dysembryoplastic neuroepithelial tumour features in 16 patients. Brain 1994; 117: 461-75.

28. Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf S, eds. Epileptic syndromes in infancy, childhood and adolescence. 2nd ed. London: John Libbey, 1992.

29. Duncan JS. Imaging and epilepsy. Brain 1997; 120: 339-77.

30. Duchowny M, Jayakar P. Functional cortical mapping in children. Adv Neurol 1993; 63: 149-54.

31. Taylor DC, Cross OH, Harkness W, Neville BGR. Defining new aims and providing new categories for measuring outcome of epilepsy surgery in children. In: Tuxhorn I, et al., eds. Paediatric epilepsy syndromes and their surgical treatment. London: John Libbey, 1997: 17-25.

32. Wright MC, McGuire PK, Polino JB, et al. A voxel-based method for the statistical analysis of gray and white-matter density applied to schizophrenia. Neuroimage 1995; 2: 244-51.

33. Schachter SC, Super CB. Vagus nerve stimulation. Epilepsia 1998; 39: 677-86.

34. Morrell F, Whisler W, Smith MC. Landau-Kleffner syndrome. Treatment with subpial transection. Brain 1995; 118: 1529-46.

35. Fried I, Cascino G. Lesional surgery. In: Engel J Jr, ed. Surgical treatment of the epilepsies, 2. New York: Raven Press, 1993: 501-9.