This commentary is on the article by Gallagher et al. on pages 10621065 of this issue.
Intractable epilepsy in tuberous sclerosis: is the tuber removal not enough?
Article first published online: 7 OCT 2010
Copyright © 2010 Mac Keith Press
Developmental Medicine & Child Neurology
Volume 52, Issue 11, page 987, November 2010
How to Cite
CURATOLO, P. (2010), Intractable epilepsy in tuberous sclerosis: is the tuber removal not enough?. Developmental Medicine & Child Neurology, 52: 987. doi: 10.1111/j.1469-8749.2010.03802.x
- Issue published online: 7 OCT 2010
- Article first published online: 7 OCT 2010
In tuberous sclerosis, a substantially increased risk of developing epilepsy is present very early in life. Cortical tubers are an extremely important aspect of the tuberous sclerosis phenotype, and may have an age-dependent tendency to become epileptogenic. They can be found in the fetal brain as early as 20 weeks’ gestation and can be detected by fetal magnetic resonance imaging (MRI) at 24 to 26 weeks’ gestation; they are typically stable in size and appearance, and remain proportional to the rest of the brain in growing children. However, recent evidence shows that tubers may have a non-static behaviour;1 they can degenerate in cystic lesions and can exhibit evolving characteristics over time on subsequent MRIs. Not infrequently they have focal areas of calcifications that can progress changing appearance with time and calcified tubers may act as an epileptogenic focus.1
Cortical tubers are dysplastic lesions composed of giant cells, maloriented dysmorphic neurons, and atypical astrocytes. These abnormal neuronal populations exhibit reduced GABAergic inhibition and have intrinsic epileptogenicity. An imbalance between excitatory and inhibitory synaptic transmission may be linked with seizure initiation in tubers. However, putative ictogenic mechanisms are still largely unknown.2 Furthermore, not all tubers are the same, and a spectrum of abnormal cell types in different tubers may present in the same individual, giving a different contribution to epileptogenesis. Immunohistochemical heterogeneity and variability of the cellular phenotype reported in different tubers could explain why in children with tuberous sclerosis only some of several tubers are potentially epileptogenic.
Although in many patients one localization is consistently present, epileptogenic foci can shift from one epileptogenic tuber to another during the course of the disease, and different regions can become epileptogenic over time due to maturational phenomena. Age-dependent hyperexcitability of different lesions may lead to age-dependent seizure semiology and there is a specific developmental window during which spasm-type seizures begin.2
Since seizures at an early age are the only independent risk factor for cognitive impairment, the termination of seizures may have developmental benefits. The surgical treatment of intractable epilepsy has yielded excellent results in tuberous sclerosis. Surgery has a role even in individuals with multiple cortical tubers, and if only one region of epileptogenicity is detected, early surgery is strongly recommended. Seizure freedom after surgery has been reported when both the tuber and the electrically active adjacent cortex were resected.3
Despite significant progress in defining the epileptogenic tuber by multimodality imaging in individuals with multiple lesions, almost 40% of the patients with tuberous sclerosis suffer from surgical failure, suggesting that gains in surgical prognosis are still needed. Activation of tubers at different ages leads to re-operation and multistage surgery. Risk factors associated with seizure recurrence after surgery are poorly understood, and the perspective of future seizure recurrence should be weighed against that of a temporary relief of severe seizures during a critical period of brain development.2
Tubers are not the only determinants for epilepsy as demonstrated by a tuber-less infant with tuberous sclerosis suffering from intractable epilepsy.4 Epilepsy secondary to tuberous sclerosis could be a consequence of disorders of long range neural networks with alterations in cortical-subcortical systems connectivity. This can explain why the removal of the lesion and the epileptogenic zone is not enough to interrupt the epileptogenic networks in a significant proportion of individuals with tuberous sclerosis.
Disease-modifying agents such as mTOR inhibitors have proved their efficacy in preventing epilepsy in mouse models and in reducing seizure frequency in a patient with tuberous sclerosis with refractory seizures.5 Although caution is appropriate, these findings suggest that mTOR inhibitors may have the potential to become a novel antiepileptogenic treatment for individuals with tuberous sclerosis.