MANEJO DEL SÍNDROME DE LENNOX-GASTAUT
Edwin Trevathan, M.D., M.P.H.
Associate Professor of Neurology & Pediatrics
Director, Pediatric Epilepsy Center
Washington University School of Medicine
St. Louis Childrens Hospital
One Childrens Place
St. Louis, MO 63119
Email: trevathan_e@kids.wustl.edu
Running title: Lennox-Gastaut Syndrome
Key words: Lennox-Gastaut Syndrome, epilepsy, epidemiology, clinical trials, prognosis, valproate, felbamate, lamotrigine, topiramate, ketogenic diet.
Lennox-Gastaut Syndrome (LGS) was among the first epilepsy syndromes described, and represents the most refractory of childhood epilepsies that challenge even experienced clinicians.1 Clinical research in the1990s has increased the number of available therapies for LGS. Yet reductions in seizure frequency and improvements in the quality of life have been minimal for most patients. However, recent progress offers hope for significant advances in the future.
Comparative trials of the newer therapies are few. Neurologists make therapeutic decisions under relative uncertainty as they evaluate these children with multiple daily seizures that are associated with significant neurological handicap. I present a review of the current literature and offer my approach regarding the management of children with LGS.
History and Diagnostic Criteria
In the early 1930s Lennox described the clinical features of epileptic encephalopathy an entity that included those with multiple seizure types and mental deficiency. In late 1930s Lennox and Gibbs described slow spike and wave that was thought to be a variant of the spike and wave they had previously described in petit mal epilepsy.2 Lennox in 19453 and Lennox and Davis in 1950 4 published the symptomatic triad of; (1) slow spike and wave on EEG, (2) mental deficiency, and (3) three seizure types (that we now refer to as atypical absence seizures, myoclonic seizures, and head drops evolving to axial spasms and falls). In 1966 Gastaut published observations largely derived from the thesis of Charlotte Dravet that expanded upon the observations of Lennox and Davis.5 Based upon the contributions of Lennox and others at Boston Childrens Hospital and Gastaut and colleagues of the Marsielles school, the term Lennox-Gastaut Syndrome was adopted.6
Clinical investigators have refined the diagnostic criteria of LGS over the last thirty years, but the essential elements of the syndrome have remained unchanged since Lennox and Davis publication in 1950.6-17 The syndrome is currently defined by several criteria: (1) multiple seizure types including atypical absence and seizures resulting in falls (axial tonic, massive myoclonic, and atonic seizures); (2) EEG demonstrating slow spike and wave (< 2.5 hertz) and bursts of fast rhythms at 10-12 hertz during sleep; and (3) static encephalopathy and learning disabilities, most often associated with profound mental retardation. Other seizure types usually are present including generalized tonic-clonic seizures and partial seizures.18
Epidemiology and Natural History
LGS represents about 5% of childhood epilepsy, and that the presence of slow spike and wave on EEG among children with multiple seizure types predicts the co-existence of profound mental retardation.19 Children with LGS are more likely to have cerebral palsy than children without slow spike and wave on EEG and multiple seizure types.19 As children with LGS grow into adulthood, many require institutionalization in order to receive appropriate care. Oguni and colleagues documented the progressive decline in IQ and progressive gait disturbances with age that were associated with worsening of the epileptic encephalopathy.20 The risk of serious injuries from falls associated with seizures is high, and up to 10% of children with LGS die prior to age 11 years.21-23
Therapies before 1980
Before the introduction of valproic acid in the 1970s a variety of AEDs were used in the treatment of seizures associated with LGS. Phenytoin was thought to be helpful for the axial spasms, but may exacerbate the frequency of petit mal seizures. Barbiturates were used with some apparent efficacy, but with associated sedation. Benzodiazepines were recommended, with some apparent increased risk for development of tonic status epilepticus.
Succinimides, especially methsuximide (Celontin), may have an adjunctive role in the treatment of atypical absence, tonic and myoclonic seizures associated with LGS.24 However, there are no controlled studies of methsuximide in LGS. Trimethadione is a rarely used drug with considerable efficacy against absence seizures. Lennox emphasized that trimethadione was effective against the atypical absence, myoclonic, and atonic seizures of LGS,3 but controlled trials have not been performed. Bromides have not been shown to be effective in LGS.25
Acetazolamide (Diamox), a carbonic anhydrase inhibitor, has been shown to have efficacy against multiple types of seizures and is usually well-tolerated.26-28 The role of acetazolamide in patients may deserve further study, especially since another carbonic anhydrase inhibitor (topiramate) has recently been proven to reduce atonic seizures in children with LGS.29
ACTH has been reported to benefit patients with Lennox-Gastaut Syndrome,30 but side effects and lack of objective data regarding benefit have limited interest in ACTH. Roger and colleagues have suggested that if corticosteroids are given early in the course of LGS that long-term benefit may be achieved, 31but controlled data are not available. The ketogenic diet has been used for children with refractory epilepsy since the late 1920s, but has experienced a renewal of interest in the 1990s and is discussed below in detail.
Valproic acid was approved for use in the 1970s. Although randomized clinical trials have never been conducted to prove efficacy of valproate in Lennox-Gastaut Syndrome, most epileptologists have viewed valproate as a first-line drug for LGS since the early 1980s because of: (1) VPAs efficacy against partial seizures and generalized seizures (including absence); (2) lack of exacerbation of any of the seizure types associated with LGS; (3) relative lack of sedative side-effects compared to barbiturates, and (4) relative ease of use compared to the ketogenic diet. The major risks of VPA are increased risk of neural tube defects in the offspring mothers taking VPA during the first trimester of pregnancy, idiosyncratic hepatic failure, and pancreatitis. However, the risk of injury and status epilepticus with associated morbidity and mortality is higher among patients with LGS than the risk of hepatic failure or pancreatitis from VPA. All females of child-bearing age with LGS should receive adequate birth control. Sterilization of women of child-bearing age with LGS who are unable intellectually to raise children and for whom pregnancy is associated with significant medical risks should be considered.
Treatments introduced in the 1980s & 1990s
By 1990 VPA was the obvious first choice for the treatment of seizures associated with LGS. However, most children and adults with LGS tended to receive multiple AEDs with continued multiple daily seizures. A series of clinical trials of newer AEDs in children with LGS were conducted in the 1980s and the 1990s.
Cinromide
In the early 1980s anecdotal reports of the possible efficacy of cinromide in the treatment of LGS32 led to the first major multi-center clinical trial of treatment for LGS.33 Seventy-three patients entered the double-blind, placebo-controlled, parallel study in which patients who completed a 6-week baseline were randomly assigned cinromide or placebo added to their baseline AEDs. The original goal was 40 completed patients in each study arm. The study was terminated prematurely when it was clear that cinromide was not effective. The cinromide study was important for two reasons. First, the cinromide study documented the "enormous commitment by investigators and their staff, by consultants, by the sponsor, and most notably by the patients parents" in terms of time, documentation, and patient recruitment and retention in controlled trials of this complex patient population.33 Defining and recruiting a homogenous population of patients with LGS was difficult, requiring multiple centers. Identifying and quantifying seizure activity in LGS is very difficult; many of these patients have seizures that are difficult to classify according to the International Classification of Epileptic Seizures.34 Their seizures are very frequent and often difficult to count. The seizures of LGS are often subtle, and inter-observer differences in number and type of seizures may be significant.
Second, the cinromide study highlights a large placebo response. The investigators noted that "..many patients were obviously significantly improved during the treatment period, and others were unchanged or worse. This observation had generated cautious optimism, and the results of the interim analysis were both surprising and disappointing." 33 The significant placebo response was documented in the subsequent clinical trials of LGS, .35,36 emphasizing the importance of viewing open-label studies with a degree of skepticism and making clinical decisions when possible based upon the results of well-designed controlled clinical trials.
Felbamate
In 1993 the efficacy of felbamate (FBM) was documented in a double-blind, placebo-controlled trial.36 Seizure frequency was assessed by guardian report and by serial 4-hour EEG-video monitoring sessions during the course of the study. FBM significantly reduced the number of atonic seizures compared to placebo in both the treatment (p = 0.01) and the maintenance (p = 0.002) phases of the study. Of the 37 patients who received FBM, 3 were without atonic seizures during the treatment phase, and 5 had no atonic seizures during the maintenance phase. A dose-response relationship was demonstrated for reduction of atonic seizures, with a linear reduction in the number of atonic seizures per day with increasing plasma FBM levels. There was a statistically significant reduction in seizure frequency for all seizure types among the FBM-treated patients compared to those who received placebo (p = 0.002).
Global evaluations of patient functioning and neuropsychological tests were performed on all study subjects in order to assess non-seizure-related outcomes. The global evaluation scores during the maintenance period were significantly higher (p < 0.001) among the FBM group than the placebo group.
Dodson reported the 12-month open-label study of FBM that followed the clinical trial.37 Those who converted from placebo to FBM had the same degree of improvement on FBM as those who received FBM during the trial. At the end of the double-blind trial only 2 of the 22 subjects randomized to placebo had experienced a > 50% reduction in atonic seizure. However, during the first month that these patients from the placebo group were treated with FBM 12 of the 22 subjects (55%) had a > 50% reduction in atonic seizures. Combining both the groups of patients from the randomized double-blind trial in the follow-on study, 33 of 50 or approximately two-thirds had a > 50% reduction of atonic seizures 12 months after beginning the open-label study.
No pattern of serious adverse events due to FBM was apparent at the time of FDA approval in 1993. By the summer of 1994 120,000 patients had been exposed to FBM and reports of both aplastic anemia and hepatic failure had been reported to Wallace Laboratories and the FDA. After letters had been sent to over 200,000 physicians in the USA informing them of these new risks, most patients were withdrawn from FBM and the use of FBM in LGS declined.
Over the last four years analysis of the available data have led to a better estimate of the risk of FBM. Kaufman, et.al. reviewed all case reports of aplastic anemia among patients treated with FBM. The incidence of aplastic anemia among those treated with FBM was estimated to have a lower limit of 1 per 37,037 patients and an upper limit of 1 per 4784 patients, with a "most probable" incidence of 1 per 7874 patients treated.38
Pellock and Brodie estimated that the incidence of hepatotoxicity to be about 1 per 26,00034,000 patients treated with FBM39 similar to the recently reported risk of hepatotoxicity for valproate (VPA).40 No children under the age of 13 years have been reported to have FBM-related aplastic anemia. Female sex, history of immune disorders (e.g., lupus), a history of prior blood dyscrasias, and allergic reactions to medications are probably associated with increased risk for FBM-associated aplastic anemia. These factors may later prove helpful in selecting patients for FBM treatment.
Lamotrigine
Lamotrigine, introduced initially for add-on treatment of partial seizures in adults, has been shown to have efficacy against multiple types of seizures. In September, 1998 the FDA approved lamotrigine for use in children and adults with LGS. Current data demonstrate efficacy against partial seizures in children and adults (add-on therapy and in monotherapy), generalized tonic-clonic seizures, tonic seizures, absence seizures, and atonic seizures. Apparent efficacy in open-label studies led to clinical trials of LTG among children with LGS.41,42
In 1997 a large double-blind, placebo-controlled trial of LTG and LGS with collaborators from 40 different epilepsy centers in the U.S.A. and Europe was published.35 Following a single-blind baseline in which all study patients received placebo, 169 patients (ages 3 to 25 years) were randomized to placebo or LTG added to their baseline AEDs for 16 weeks. Thirty-three percent of the LTG group and 16% of the placebo group experienced a > 50% reduction in seizure frequency of all major seizures (generalized tonic-clonic, tonic, atonic, and major myoclonic seizures; p = 0.01). Thirty-seven percent of those treated with LTG and 22% of those who received placebo had a > 50% reduction in frequency of drop attacks (atonic, tonic, and/or major myoclonic seizures that resulted in falls; p = 0.04). Forty-three percent of LTG-treated patients and 20% of placebo-treated patients had a 50% or greater reduction in frequency of generalized tonic-clonic seizures (p = 0.007). The only clinically significant adverse event was serious rash in two patients, both of whom were also receiving VPA and also had LTG dose-escalation rates that exceeded the current recommendations. Global evaluations of patients functioning in terms of speech, language, and attention were significantly improved in the LTG group.43
Eriksson and colleagues published a randomized, double-blind, crossover study of LTG as add-on therapy in 30 children with severe generalized epilepsy, 20 of whom had LGS.44 LTG was more effective than placebo during the double-blind crossover phase in reducing the frequency of tonic, tonic-clonic, and atonic seizures (p<0.0001). Thirteen of the 20 children with LGS improved in the open phase of the study and entered the double-blind phase; seven of the 20 children (35%) responded to LTG treatment with a > 50% seizure frequency reduction. Two children with LGS became seizure free on LTG. As with the large multi-center trial35, Eriksson, et.al. could not assess LTGs efficacy against atypical absence seizures. No apparent relationship between LTG blood level and response was noted, none of the children improved on placebo, and none developed a rash.
Proper dosing of LTG in children requires attention to whether they are on LTG monotherapy, LTG plus an enzyme-inducing AED (which reduces the half-life of LTG), or LTG plus VPA (which dramatically increases the half-life of LTG). Children who are taking an enzyme-inducing AED and VPA with their LTG should be dosed according to the guidelines for VPA plus LTG (see Table).
Table. Dosing recommendations for
lamotrigine in children
(adapted from Garnett, J Child Neurol
1997;12(suppl 1):S13.)
| Concurrent AED | Wks 1-2 | Wks 3-4 | Usual Main. Dose | |||
| EIAEDd | 2.0 mg/kg/d | 5.0 mg/kg/d | 5-15 mg/kg/d | |||
| mono-therapy | 0.5 mg/kg/d | 1.0 mg/kg/d | 2-8 mg/kg/d | |||
| valproate | 0.2 mg/kg/d | 0.5 mg/kg/d | 1-5 mg/kg/d |
d EIAED = enzyme-inducing anti-epileptic drugs such as phenytoin and carbamazepine. Patients who are taking both valproate and an EIAED should be dosed according to the co-administration with VPA guidelines.
Topiramate
Topiramate is a new AED that has been approved in the U.S.A. for adjunctive therapy in partial seizures in adults. Efficacy has been reported in abstract form for partial seizures in children45,46 and in the atonic seizures associated with LGS.29,47 These data, although not yet fully published at the time of this writing, have been presented at international meetings and reviewed recently by Glauser.48 Topiramate appears to be effective in reducing the frequency of atonic seizures associated with LGS. How topiramate will compare to FBM and LTG is not yet known, but the overall safety of TPM makes this option attractive in spite of the slow dose-titration schedule required to avoid cognitive and language problems associated with rapid dose escalation.
Ketogenic Diet
The ketogenic diet has been used in children with refractory seizures of multiple types since the late 1920s.49 Over the last decade there has been a resurgence of interest in the ketogenic diet, as more open-label data have become available on the classical diet that appears to offer better results with fewer side effects than the MCT oil diet introduced in the 1970s.50,51 Recently, a multi-center open-label study of the ketogenic diet has reproduced the good results previously reported by the Johns Hopkins group.52 NIH is sponsoring a placebo-controlled trial of the ketogenic diet at Johns Hopkins that will hopefully answer the question of efficacy for this old but novel treatment. If the placebo-controlled trial of the ketogenic diet confirms efficacy demonstrated in the open-label studies, the place of the diet in the armamentarium of the child neurologist may depend upon the frequency of adverse events.53
Vagus Nerve Stimulation
Vagus nerve stimulation (VNS) has been approved for treatment of intractable partial seizures in the United States.54 A Swedish group has recently reported a possible reduction in seizure frequency among children with LGS treated with VNS.55 As expected, sedative side effects are less severe with VNS than with AEDs in children, but other significant adverse events have been reported.56 Wide spread use of VNS in children and adults with LGS should await the results of the clinical trial currently in progress.
Corpus callosotomy
Patients with intractable atonic or tonic seizures that result in falls may benefit from a corpus callosotomy. Various epilepsy centers have used slightly different approaches to selecting patients for this procedure. However as long as seizures resulting in severe falls with associated injuries are targeted, a significant number of patients can achieve palliative improvement.57
Goals and Treatment Objectives
Children and adults with LGS are commonly referred to epileptologists with frequent daily seizures and significant dose-related side effects from multiple anti-epileptic drugs (AEDs). Parents and other caregivers often project unrealistic treatment objectives to their treating physician. These unrealistic therapeutic goals usually encourage polypharmacy and increased dose-related side effects of AEDs that can be as debilitating as the seizures. Once the physician makes the diagnosis of LGS, the time taken to educate the family and outline realistic treatment objectives usually pays dividends in the future by simplifying treatments, reducing dose-related side effects of AEDs, and improving the quality of life of the patient and the family.
The following concepts often help guide therapeutic decisions. First, patients with LGS almost never become seizure free and the goal is to limit the number of seizures while maintaining alertness. Second, the seizures that result in falls and injuries and the atypical absence seizures (that exacerbate the epileptic encephalopathy) are the primary therapeutic target. Third, limiting the dose-related sedative and cognitive side effects may improve quality of life for the patient. Fourth, the seizures tend to cluster and the overall course is remarkable for relative remissions and exacerbations of seizure frequency that are often unrelated to any obvious external factors; quick decisions that lead to discontinuing a drug or adding an additional drug may be regreted later. Drugs proven effective in LGS are used first. I try to limit the simultaneous major AEDs to three at any one time, and I use as-needed benzodiazepines (e.g., rectal diazepam or sublingual/oral lorazepam) for clusters of seizures that might lead to emergency room visits and admissions. If in spite of optimal use of AEDs injuries occur as a result of drop attacks, EEG-video monitoring and high resolution MRI scanning is performed in order to determine whether the patient might be a candidate for corpus callosum section. I often find it helpful to commit a plan to writing and provide the family a copy.
Therapeutic Approach
Neurologists rarely evaluate a newly diagnosed child with LGS who is on no AEDs, as most patients with LGS evolve into the clinical syndrome after months to years of seizures. However, the following approach seems reasonable based upon the available data. Broad-spectrum AEDs (effective against all seizure types) should be used primary drugs for LGS (VPA, LTG, TPM, FBM). AEDs that exacerbate primary generalized seizures are avoided if possible (e.g., CBZ, GBP). Seizures that result in falls subject the patient to more risk than the risk of aplastic anemia or hepatic failure from FBM or any of the currently available AEDs. I try to use fewer than four major AEDs at any one time.
The significant interactions between LTG and VPA enter into my decisions early in the course of the treatment. For example, if I wish to use LTG and do not anticipate long-term use of VPA, I try to discontinue the VPA prior to starting the LTG. Because I have had good experience with the combination of TPM and LTG, I will often add TPM to VPA and then wean the VPA before adding LTG to the regimen. Only if the patient fails treatment with VPA, LTG, and TPM will I use FBM. The ketogenic diet is used in children under the age of 10 years, but I try to avoid the combination of the ketogenic diet and VPA when possible.53
Once the best primary regimen for LGS is established in an individual patient I consider adding another drug to address a specific concern with a single type of seizure. Often a benzodiazepine at bedtime will reduce the frequency of tonic seizures upon awakening the next morning and/or compensate for the difficulty sleeping associated with FBM. For patients who experience clusters of seizures during minor infectious illnesses, rectal diazepam or sublingual lorazepam may reduce seizures during the acute illness and help maintain the total number of maintain drugs under three or four.
The long-term management of the patient with LGS should evolve into a partnership-like relationship between the attendees at the chronic care facility and/or the patients family and the physician. Early education of the family and caregivers helps foster better decision making later in the course of this most severe of epilepsy syndromes.
1. Wheless JW, Constantinou JE. Lennox-Gastaut syndrome. Pediatr Neurol 1997; 17:203-211.
3. Lennox WG. The petit mal epilepsies; their treatment with tridione. JAMA 1945; 129:1069-1074.
6. Erba G, Lombroso CT. La sindrome di Lennox-Gastaut. Prosp Pediatr 1973; 10:145-165.
7. Aicardi J. The problem of the Lennox Syndrome. Devel Med Child Neurol 1973; 15:77-81.
12. Dulac O, N'Guyen T. The Lennox-Gastaut Syndrome. Epilepsia 1993; 34:S7-S17
15. Yaqub BA. Electroclinical seizures in Lennox-Gastaut Syndrome. Epilepsia 1993; 34:120-127.
31. Roger J, Dravet C, Bureau M. The Lennox-Gastaut syndrome. Cleve Clin J Med 1989; 56:S172-S180
39. Pellock JM, Brodie MJ. Felbamate: 1997 Update. Epilepsia 1997; 38:1261-1264.
48. Glauser TA. Topiramate use in pediatric patients. Can J Neurol Sci 1998; 25:S8-S12
49. Swink TD, Vining EP, Freeman JM. The ketogenic diet: 1997. Adv Pediatr 1997; 44:297-329.
Para enviar comentarios sobre esta conferencia pulse en este enlace - FORO DE EPILEPSIA - .Si se le solicita un nombre de usuario y una clave de entrada, introduzca las mismas que usa para entrar en el congreso. Para enviar comentarios directamente a los autores utilice la dirección de E-Mail de correspondencia si se indica al inicio de esta conferencia.
PAGINA PRINCIPAL | CURSOS DE FORMACION | POSTERS ELECTRONICOS
CONFERENCIAS INVITADAS | COMUNICACIONES | CHARLA INTERACTIVA