Epi/q.ria, 33(5):917-922, 1992 Raven Press, Ltd., New York 0 International League Against Epilepsy
?-Vinyl GABA (Vigabatrin) in Epilepsy: Clinical, Neurochemical, and Neurophysiologic Monitoring in Epileptic Patients A. Ylinen, J. Sivenius, A. Pitkanen, T. Halonen, J. Partanen, E. Mervaala, *J. P. Mumford, and P. J. Riekkinen Depurtment oj' Neurology and Neurophysiology, University of Kuopio, Kuopio, Finland; and "Marion Merrell Dow Research Centre, Winnersh, England
Summary: We report long-term clinical, neurochemical, and electrophysiologic data of y-vinyl GABA (GVG, vigabatrin) in three groups of patients. GVG was started as add-on therapy for 75 patients with refractory complex partial seizures (group A) and for 36 mentally handicapped patients with severe epilepsy (group B). The third group (C) consisted of 20 patients with carbamazepine (CBZ) monotherapy, in half of whom GVG monotherapy was substituted. After 3 months, 55% of patients in group A and 42% in group B were responders (reduction in seizure frequency >50%). After 6 (group A) and 3 years
(group B) of follow-up, 27 and 33% of the patients, respectively, still had good response to GVG. Neurochemical measurements showed a twofold increase in CSF GABA concentrations and minimal or no changes in other neurotransmitter-related parameters. In group C, substitution of GVG as medication tended to normalize the lengthened latencies in somatosensory evoked potentials (SEPs) observed during CBZ treatment. Key Words: Anticonvulsants-y-Vinyl-GAB A-Vigabatrin-Drug toxicity-Electroencephalography-Evoked potentials.
y-Vinyl GABA (GVG, vigabatrin) irreversibly inhibits metabolism of y-aminobutyric acid (GABA) by enzyme GABA transaminase (GABA-T). Administration of GVG in anticonvulsant doses increased GABA level two- to threefold in animal brain and in cerebrospinal fluid (CSF) of animals and humans (Jung et al., 1977; Schechter et al., 1984; Halonen et al., 1988). GVG has been shown to reduce seizure frequency >50% in -40-60% of patients (Rimmer and Richens, 1984; Ben-Menachem et al., 1986; Loiseau et al., 1986; Sivenius et al., 1987; Matilainen et al., 1988) and is particularly effective against complex partial seizures. Few long-term clinical and electrophysiological studies have been performed (Matilainen et al., 1988; Cocito et al., 1989), and data regarding the specificity of the mechanism of action in human epilepsy are not readily available. Non-primate studies have shown microvacuolization in the white matter of the brain at high GVG
doses, with prolongation of the latencies of somdtosensory evoked potentials (SEPs). This has hindered adoption of GVG into wider clinical use, although no such changes have been observed in humans or other primates (Pedersen et al., 1987; Arezzo et al., 1989; Graham, 1989). To monitor the long-term clinical and neurophysiologic effects of GVG and to evaluate the specificity of action by following changes in neurotransmitter-related parameters in CSF, we administered GVG to three different patient groups. We report a 6-year clinical follow-up of 75 patients with complex partial epilepsy and a 3-year clinical and neurophysiologic follow-up of 36 mentally handicapped epileptic patients treated with GVG. CSF samples were analyzed neurochemically for 6-7 months. For neurophysiologic analysis (EEG and evoked potentials) we randomly converted 10 of a group of 20 patients receiving carbamazepine (CBZ) monotherapy to GVG monotherapy . PATIENTS AND METHODS
Received Julv 1991 : revision accepted February 1992. Address coriespondence and reprint requests to Professor P. 5. Riekkinen at Department of Neurology, University of Kuopio, P.O. Box 1627, SF-70211 Kuopio, Finland.
The three groups were designated A, B, and C (Table 1). Group A consisted of 75 patients with 917
A . YLINEN ET A L .
918
refractory complex partial epilepsy and with two or more monthly seizures in the 2-month baseline period (BP). Patients were started on GVG 3 g/day, 1.5 g twice daily, as add-on therapy. After the first 3-month treatment period (TPl), patients showing reduction in seizure frequency 250% (responders) or those showing global improvement in their condition continued to the second 3-month treatment period (TP2). This was a dose-reduction phase in which patients received either 3 or 1.5 giday GVG under double-blind conditions. Patients who were still responsive to GVG or who showed a global improvement then continued to the long-term treatment phase (LTT). Previous antiepileptic drugs (AEDs) were unchanged from BP to the end of TP2, but during the long-term phase, a reduction in some drugs was attempted in a few patients. Lumbar CSF samples were collected at the end of BP, TP1, and TP2, after oral consent was obtained from the patients and approval was given by the ethics committee. Group B consisted of 36 mentally handicapped patients with severe epilepsy; 25 (69%) were severely mentally handicapped (IQ < 50), 9 (25%) were mildly mentally handicapped (50 < IQ < 70), 1 (3%) was borderline (70 < IQ < 90), and 1 (3%) had normal intelligence. Computed tomography (CT) scan was abnormal in 29 (81%) patients, and EEG was abnormal in all but 1. During a 3-month BP, clinical seizures were counted, baseline EEG was recorded and after oral consent was obtained from patients and parents, lumbar CSF samples were collected. GVG was then started as add-on therapy, and appropriate doses for each patient (mean 2.6 i- 0.6 giday) were evaluated for the next 3 months (dose modification phase) while previous AEDs were kept constant. Responders and patients demonstrating global improvement according to the same criteria used for group A continued to longTABLE 1. Patient clinical data Parameter
Group A
Group B
Group C
Sex (MIF) Mean age (yr) Mean age at onset of seizures Seizure type Partial onset Primarily generalized
3414 1 31
23113 29
911 I 37
10
5
75
30
20
0
6
0
Group A (75 patients with complex partial epilepsy), group B (mentally handicapped epileptic patients), group C (20 epileptic patients with carbamazepine (CBZ) monotherapy). Vigabatrin (GVG) was started as add-on therapy in groups A and B, 10 patients in group C were changed from CBZ to GVG monotheraPY. Epilepsia, V d . 33, N o . 5 , 1992
TABLE 2. Number of responders during Jollow-up in three patient groups treated with GVG Duration of GVG treatment
Group A (n = 75)
Group B (n = 36)
Group C (n = 10)
3 mo 1 Yr 2 Yr 3 Yr 6 Yr
41 (55%) 31 (41%) 20 (27%) 20 (27%) 20 (27%)
15 (42%) 12 (33%) 9 (25%) 12 (33%)
7 (70%)
Abbreviations as in Table I . Groups A, B, and C as in Table 1. n, number of patients starting therapy with GVG. Responders in groups A and B: patients who had >50% in seizure frequency. Responders in group C: patients who responded to GVG monotherapy as well as (or better than) to CBZ monotherapy.
term therapy. Lumbar CSF samples were collected, and EEGs were recorded after 7 months, 1 year, and 2 years. Group C consisted of 20 epileptic patients receiving CBZ monotherapy. Ten were randomly changed to GVG monotherapy in a 1-week period. The quantitative EEG (QEEG), SEPs, and visual evoked potentials (VEPs) were recorded in both patient groups at baseline and after a 3-month follow-up period. Controls for QEEG, SEP, and VEP studies were 120, 34, and 21 healthy subjects, respectively. RESULTS Clinical follow-up The effect of GVG on seizure frequency in the three groups is shown in Table 2 (except for the dose modification phase for group A). Results of the shorter follow-up in groups A and B were reported previously (Sivenius et al., 1987, 1991; Matilainen et al., 1989). In group A, there were 41 (55%) responders after 3 months of therapy, 31 (41%) after 1 year, 20 (27%) after 2 years, 20 (27%) after 3 years and 20 (27%) after 6 years of therapy. In group B, 15 (42%) patients were classified as responders at the follow-up point of 3 months and 12 (33%), 9 (25%) and 12 (33%) were so classified at 1 , 2, and 3 years, respectively. Side effects were generally mild and transient in both populations. Drowsiness was most commonly reported (75% in group A and 28% in group B) during the first 3 months, and usually disappeared during long-term treatment. During the dose-modification phase in group A, the median monthly seizure frequency increased in patients receiving 1.5 g/day GVG from 1.7 during TPI (3 giday GVG) to 5.7 during TP2, although this was still lower than the baseline value (1 1 seizures a month, p < 0.01, Wilcoxon’s test). In group C, patients who continued receiving CBZ therapy showed the same seizure frequency
GVG IN EPILEPSY TREATMENT
they had exhibited during the 3-month observation period. Four were seizure-free. Eight of 10 patients who were changed to GVG therapy had previously been seizure-free. Two of these patients had a seizure after changing to GVG monotherapy, 4 days and 1 day, respectively, after stopping CBZ treatment and were changed back to CBZ therapy. One patient with GVG monotherapy was discontinued because of increased gastric irritation and psychic problems. These conditions had existed during CBZ therapy but to a lesser extent. Six patients continued to be seizure-free, and seizure frequency in 1 patient was unchanged with GVG monotherapy. Three of the 6 seizure-free patients expressed a wish to continue with GVG monotherapy after the initial 3-month observation period and have now been seizure-free for 3 years. Neurochemical studies Analysis of CSF showed that levels of total GABA (T-GABA), free GABA (F-GABA), and homocarnosine (HC) were increased two- to threefold in groups A and B during 3-month GVG therapy as compared with baseline. In group A, the increase in T-GABA was greater in responders than in nonresponders, but this difference was not observed in group B. Halving the dose from 3 to 1.5 g/day in group A resulted in an almost 50% reduction in GVG level in lumbar CSF and to a significant decrease in T-GABA, F-GABA, and HC levels. In both groups A and B, there was a slight but consistent increase in glycine levels after GVG treatment, but no treatment-related changes were observed in levels of glutamine, aspartate, asparagine, and tau-
919
rine (Table 3). Levels of acetylcholinesterase (AChE), homovanillic acid (HVA), 5-hydroxyindoleacetic acid (SHIAA), somatostatin, prolactin, and P-endorphin remained constant during GVG therapy of 3, 6, and 7 months’ duration. More detailed data on the levels of the various neurotransmitter-related parameters in the CSF of these patients were reported previously (Halonen et al., 1988; Pitkanen et al., 1988; Riekkinen et al., 1989a,b). Neurophysiologic studies EEG follow-up of patients in group B showed a slight decrease in the amount of slow wave and epileptiform activity during GVG treatment, especially at the 2-year follow-up point and has been reported in detail previously (Pitkanen et al., in press). In group C, the neurophysiologic parameters of patients changed to GVG monotherapy were compared with those of patients who continued to receive CBZ monotherapy . CBZ monotherapy produced statistically significant prolongation of cortical SEP frequencies as compared with those of normal controls. After 3-month treatment with GVG, the cortical SEP peak latencies became significantly shorter (Table 4). QEEG showed no significant changes during GVG therapy as compared with CBZ therapy. All the major pattern VEP peak latencies were prolonged in epileptic patients receiving CBZ as compared with healthy controls. Change to GVG monotherapy produced no significant changes (Mervaala et al., 1989).
TABLE 3. CSF amino acid levels of patients with different responses t o GVG in two different patient populations Amino acid T-GABA BP GVG F-GABA BP GVG HC BP GVG GlY BP GVG Glu BP GVG
Group A Responders (n) 4.87 12.98
f ?
1.25 3.69
(39) (39)”
0.129 0.260
-t f
0.043 (39) 0.072” (39)
Nonresponders (n)
(28) (28)’
0.129 t 0.102 (11) 0.289 f 0.121 (11)’
0.199 f 0.085 (10) 0.359 +- 0.078 (10)’
4.176 t 1.050 (13) 12.906 f 5.799 (11)’
3.389 f 0.955 (15) 10.212 t 1.680
(28) (28)
5.27 i 1.42 6.00 f 1.72’
(32) (32)
5.16 -+ 1.81 6.12 f 1.93”
(18) (8)
(38) (38)
~
0.119 f 0.035 0.247 2 0.073 2.82 2 1.00” 7.72 -t 3.51’,‘
0.497 0.522
~
0.566 0.572
f
0.472
f 0.396
(28) (28)
5.80 f 2.49 7.06 2 1.72
(13) (ll)h
~
Nonresponders (n)
6.29 2 1.23 16.79 i 5.64
(39) (39)
f f
~
~~~~
Responders (n)
4.59 i 10.62 (28) 10.62 f 2.92”,‘ (28)
3.46 i 1.45 10.63 f 4.13’
0.592 0.547
Group B
-~
_.
5.31 15.75
f +-
1.03” (15) 5.10” (10)
(13) (11)
4.92 t 2.19 (15) 5.40 f 1.88
0.122 2 0.067 (13) 0.137 2 0.072 (11)
0.148 i 0.076 (1.5) 0.150 f 0.106 (10)
CSF, cerebrospinal fluid; BP, baseline period; F-GABA, free GABA; Glu, glutamate; Gly, glycine; HC, homocarnosine; T-GABA, total GABA; other abbreviations as in Table 1. Values (mean f SD) are expressed in p M . Statislical significance: ”p < 0.05 and ‘p < 0.01 as compared with values of responders at the same phase (Student’s t test), and ’p < 0.01 as compared with baseline values (paired Student’s t test). Epilepsia, Vol. 33, N O , 5 , 1992
_
_
_
_
A . YLINEN ET AL.
920
TABLE 4. Cortical somatosensory evoked potential peaks during CBZ and GVG monotherupy
Peaks
CBZ (n = 17) latency in ms, mean (SD)
GVG (n = 7) latency in ms, mean (SD)
Erb's point (N9) Cortical (N19)
10.7 (1 .O) 20.1 (1.7)"
10.5 (1.0) 19.6 (1.5)"
Abbreviations as in Table 1. p < 0.01, Student's two-tailed t test, controls versus CBZ group. p < 0.01, Student's two-tailed t test, CBZ versus GVG period in same patients.
'
DISCUSSION
Clinical study Our results show that about one third of patients with severe and previously drug-resistant epilepsy had reduction of at least 50% in seizure frequency during long-term follow-up of 3-6 years. GVG monotherapy appeared to be at least as effective as CBZ monotherapy in 7 of 10 patients in our preliminary 3-month trial. The proportion of patients with a reduction >50% in seizure frequency during GVG treatment during short-term follow-up is in agreement with the results of most published studies (Rimmer and Richens, 1984; Loiseau et al., 1986; Browne et al., 1987; Rimmer et al., 1987; Sivenius et al., 1987). In longterm follow-up, the number of responders is lower. Some responders also dropped out, however, and the values we report are compared with those of the entire patient population at baseline in both trials, which tends to give an artificial negative bias to the results. The number of responders after 1 year of follow-up remained relatively constant in both groups, and there was no evidence of development of tolerance, in contrast to treatment with benzodiazepines, although both types of drug act through the GABAergic system. Neurochemical study In CSF analysis, the GABAergic parameters T-GABA, F-GABA, and H C showed a two- to threefold increase after GVG treatment at 3 and 6 months as compared with baseline values. In group A (75 patients with complex partial epilepsy), the increase in T-GABA level was higher in responders than in nonresponders, but this difference was not observed in group B (36 mentally handicapped patients with severe epilepsy). Both groups showed a slight increase in glycine level after GVG treatment, but GVG apparently did not affect the levels of other amino acids such as glutamate, aspartate, asparagine, and taurine or parameters of the cholinergic (AChE), dopaminergic (HVA), serotonergic Epilepsia, Val. 33, N o . 5 , 1992
(SHIAA), or peptidergic (somatostatin, prolactin, and P-endorphin) systems. The increase in T-GABA, F-GABA, and HC levels after GVG administration is in agreement with the findings of Schechter et al. (1984) and BenMenachem et al. (1986). However, these observations are not necessarily related to good seizure control during GVG treatment. The slight increase in glycine levels in lumbar CSF after GVG could result at least partly from increased release of glycine in the spinal cord, because glycinergic innervation is most concentrated at that CNS level. Despite widespread enhancement of GABAergic transmission in the brain, the C S F markers of cholinergic, dopaminergic, serotonergic, and peptidergic systems appeared not to be significantly affected, emphasizing the specificity of the mechanism of action of GVG. Neurophysiologic study EEG studies showed a slight tendency toward a decrease in the amount of epileptiform activity and slow wave activity during GVG therapy in group B patients, especially after 2 years. Isolating the effects of a single drug in patients receiving polytherapy is difficult, however. For this reason, neurophysiologic parameters were compared in group C patients with GVG or CBZ monotherdpy. CBZ therapy was associated with slowing in occipital rhythm in QEEG and prolonged cortical latencies in SEPs. Change to GVG tended to normalize S E P values. EEG follow-up patients in group B was in agreement with the findings of previous studies (Hammond and Wilder, 1985; Tartara et al., 1986; Mervaala et al., 1989) in which no significant changes were observed in epileptiform activity o r background activity. However, a slight tendency toward a decrease in epileptiform and particularly slow wave activity at 2-year follow-up may be an indication of the effects of GVG therapy. Our findings in group C patients show that GVG therapy does not exacerbate nerve conduction parameters as compared with CBZ monotherapy, as measured by QEEG, VEP, or SEP. On the contrary, change to GVG from CBZ tended to normalize cortical latencies in SEPs. This finding is in agreement with our recent observation that, in newly diagnosed patients with epilepsy, GVG treatment tends to show beneficial effects on EEG and cognition, as compared with CBZ treatment (Kalviainen et al., 1991). Arezzo et al. (1989) showed that in dogs GVGrelated microvacuolization appears and disappears concomitantly with slowing of conduction in SEPs. The lack of functional changes in somatosensory
GVG I N EPILEPSY TREATMENT
pathways in our study is in agreement with results of previous human SEP studies in patients with polytherapy and with human neurophysiologic studies in which no microvacuolization was observed (Pedersen et al., 1987; Graham, 1989). GVG appears to be effective in long-term therapy in approximately one third of patients with previously drug-resistant epilepsy. The action of GVG appears to be specifically mediated through the GABAergic system and apparently there are no signs of neurophysiologically measurable GVGrelated dysfunction in human brain during treatment. REFERENCES Arezzo JC, Schroeder CE, Litwak MS, Steward DL. Effects of vigabatrin on evoked potentials in dogs. Br J Clin Pharmacol 1989;27:53s-60s. Ben-Menachem E, Persson LI, Hamberg A, Schechter P, Haegele KD, Mumford PJ. Changes in CSF parameters and seizure control with every-third-day, every-other-day and daily dosing of gamma-vinyl GABA. Epilepsia 1986;27:649. Browne TR, Mattson RH, Penry J K , et al. Vigabatrin for refractory complex partial seizures: multicenter single-blind study with long term follow-up. Neurology 1987;37:1 8 4 9 . Cocito L , Maffini M, Perfuuro P, Roncallo F , Loeb C. Vigabatrin in complex partial seizures: a long-term study. Epilepsy Res 1989;3: 160-6. Graham D. Neuropathology of vigabatrin. Br J Clin Pharmacol 1989;27:43S-5S. Halonen T, Lehtinen M, Pitkilnen A, Ylinen A, Riekkinen PJ. Inhibitory and excitatory amino acids in CSF of patients suffering from complex partial seizures during chronic treatment with gamma-vinyl GABA (vigabatrin). Epilepsy Res 1988;2: 246-52. Hammond E J , Wilder BJ. Effects of gamma-vinyl GABA on the human electroencephalogram. Neuropharmacology 1985;24: 975-84. Jung MJ, Lippert B, Metcalf BW, Schechter PJ. Gamma-vinyl GABA (4-amino-hex-5-enolic acid), a new selective irreversible inhibitor of GABA-T: effects on brain GABA metabolism in mice. J Neurochem 1977;29:797-802. Kalviainen R, Aikia M, Partanen J, et al. Randomized controlled pilot study of vigabatrin versus carbamazepine monotherapy in newly diagnosed patients with epilepsy: an interim report. J Child Neurol 1991;6:286&9. Loiseau P, Hardenberg JP, Pestre M, Gnyot M, Schechter PJ, Tell GP. Double-blind, placebo-controlled study of vigabatrin (gamma-vinyl GABA) in drug-resistant epilepsy. Epilepsiu 1986;27: 115-20. Matilainen R, Pitkanen A, Ruutiainen T, Mervaala E , Sarlund H , Riekkinen P. Effect of vigabatrin on epilepsy in mentally retarded patients: a 7-month follow-up study. Neurology 1988; 38:743-7. Mervaala E , Partanen J, Nousiainen M, Sivenius J , Riekkinen PJ. Electrophysiologic effects of gamma-vinyl GABA and carbamazepine. Epilepsia 1989;30:189-93. Pedersen B, Hojgaard K , Dam M. Vigabatrin: no microvacuoles in a human brain. Epilepsy Res 1987;1:74-6. Pitkanen A, Matilainen R, Ruutiainen T , Lehtinen M, Riekkinen P. Effect of vigabatrin (gamma-vinyl GABA) on amino acid levels in CSF of epileptic patients. J Neurol Neurosurg Psychiatry 1988;51:1395400. Pitkanen A, Ylinen A, Matilainen R, et al. The long-term antiepileptic efficacy of rigabatrin in drug-refactory epilepsy of mentally retarded patients: a 5-year follow-up study. Arch Neurol
92 I
Riekkinen P, Ylinen A, Halonen T, Sivenius J, Pitkanen A. Cerebrospinal fluid GABA and seizure control with vigabatrin. Br J Clin Pharmacol 1 9 8 9 ~27:873-943. ; Riekkinen P, Pitkanen A, Ylien A, Sivenius J, Halonen T. Specificity of vigabatrin for GABAergic system in human epilepsy. Epilepsia 19896;30:S 18-22. Rimmer EM, Milligan NM, Richens A. A comparison of the acute effect of single doses of vigabatrin and sodium valproate on photosensitivity in epileptic patients. Epilepsy Res 1987;1:33Y-46. Rimmer EM, Richens A. Double-blind study of gamma-vinyl GABA in patients with refractory epilepsy. Lancet 1984;l: 189-90. Schechter PJ, Hanke NFJ, Grove J, Huebert N, Sjoerdsma A. Biochemical and clinical effects of gamma-vinyl GABA in patients with epilepsy. Neurology 1984;34:182-6. Sivenius MRJ, Ylinen A, Murros K , Matilainen R, Riekkinen P. Double blind dose reduction study of vigabatrin in complex partial epilepsy. Epilepsia 1987;28:688-92. Sivenius J , Ylinen A, Murros K, Mumford J, Riekkinen PJ. Vigabatrin in drug-resistant partial epilepsy: a five-year followup study. Neurology 1991;41:562-5. Tartara A, Manni R, Galinberti CA, Hardenberg J, Orwin J , Perucca E. Vigabatrin in the treatment of epilepsy: a double blind placebo-controlled study. Epilepsia 1986;27:717-23.
RESUME Les auteurs rapportent les donnCes cliniques, neurochimiques et Clectrophysiologiques recueillies dans 3 groupes de patients trait& par gamma-vinyl-GABA (GVG, Vigabatrin). Le GVG a CtC commenct comme traitement d’adjonction chez 75 patients prCsentant des crises partielles complexes rebelles (groupe A) et chez 36 patients prksentant un handicap mental avec une Cpilepsie severe (groupe B). Le troisikme groupe (groupe C) Ctait constitut par 20 patients sous monotherapie par carbamazepine (CBZ) dont la moitie a CtC passCe a une monothirapie par GVG. Aprks 3 mois, 55% des patients du groupe A et 42% du groupe B ont t t t considtres comme rtpondeurs (rtduction du nombre d e crises de plus de 50%). Aprks 6 ans (groupe A) et 3 ans (groupe B) de suivi, 27% et 33% des patients respectivement prtsentaient encore une bonne r t p o n s e au GVG. L e s dosages neurochimiques ont mis en Cvidence une multiplication par 2 des concentrations de GABA dans le LCR, sans modifications importantes des parametres lits aux autres neuro-transmetteurs. Dans le groupe C, le passage des patients au GVG a entrain6 une normalisation des latences des potentiels CvoquCs somatosensitifs, anterieurement augmentees sous CBZ. (P. Genton, Marseille)
RESUMEN En tres grupos de enfermos 10s autores presentan informaci6n a largo plazo desde el punto de vista clinico, neuroquimico y electrofisiologico, sobre 10s efectos del gamma vinyl GABA (GVG, Vigabatrin). El GVG se inici6 como terapeutica complementaria en 75 enfermos con ataques parciales complejos refractarios a1 tratamiento (Grupo A) y en 36 enferrnos con retraso mental y epilepsia severa (Grupo B). El tercer grupo (C) consistia en 20 pacientes con monoterapia con carbamazepina (CBZ), la mitad de 10s cuales se habian cambiado a monoterapia con GVG. A 10s tres meses, el 55% de 10s enfermos del grupo A y el 42% del grupo B se clasificaron como “respuestas positivas” (reducci6n del numero de ataques en mas del 50%). DespuCs de 6 (grupo A) y 3 atios (grupo B) de seguimiento, el 27% y el 33% de 10s enfermos, respectivamente, tenian respuestas buenas a1 GVG. Las determinaciones neuroquimicas revelaron un incremento, a1 doble, en las concentraciones de GABA en el liquid0 cefalorraquideo y minimos o ningun cambio en otros parametros relacionados con neurotransmisores. En el grupo C, 10s cambios a1 tratamiento con GVG mostraron una tendencia a la normaEpilepsia, Vol. 33, No. 5 , 1992
922
A . YLINEN ET A L .
lizacion de las prolongadas latencias en 10s potenciales evocados somatosensoriales observados durante el tratamiento con CBZ. (A. Portera-Sanchez, Mudrid)
ZUSAMMENFASSUNG Wir berichten uber klinische, neurochemische und elektrophysiologische Langzeit-befunde unter Vigabatrin (GVG) bei 3 Patientengruppen. GVG wurde bei 75% mit therapieresistenten komplexen Partialanfallen als Zusatztherapie gegeben (Gruppe A) und bei 36 geistig behinderten Patienten mit schwerer Epilepsie (Gruppe B). Die dritte Gruppe umfaRte 20 Patienten mit Car-
Epilepsia, Vol. 33, N o . 5 , 1992
bamazepin-Monotherapie von denen die Halfte auf GVG umgesetzt wurden. Nach 3 Monaten waren 55% der Gruppe A und 42% der Gruppe Responder (Anfallsreduktion von iiber 50%). Nach 6 (A) und 3 Jahren (B) Nachbeobachtung zeigten noch 27% bzw. 33% der Patienten ein gutes Ansprechen. Neurochemische Untersuchungen zeigten einen Lweifachen Anstieg der Liquor GABA-Konzentration und minimale oder fehlende Veranderungen in Neurotrasmitter-abhangigen-Parametern.In Gruppe C zeigte sich eine Normalisierungstendenz der unter CBZ verlangerten Somatosen-sorischen evozierten Potentiale. (C. G. Lipinski, HeidelhergiNeckurjiemun~
?-Vinyl GABA (Vigabatrin) in Epilepsy: Clinical, Neurochemical, and Neurophysiologic Monitoring in Epileptic Patients A. Ylinen, J. Sivenius, A. Pitkanen, T. Halonen, J. Partanen, E. Mervaala, *J. P. Mumford, and P. J. Riekkinen Depurtment oj' Neurology and Neurophysiology, University of Kuopio, Kuopio, Finland; and "Marion Merrell Dow Research Centre, Winnersh, England
Summary: We report long-term clinical, neurochemical, and electrophysiologic data of y-vinyl GABA (GVG, vigabatrin) in three groups of patients. GVG was started as add-on therapy for 75 patients with refractory complex partial seizures (group A) and for 36 mentally handicapped patients with severe epilepsy (group B). The third group (C) consisted of 20 patients with carbamazepine (CBZ) monotherapy, in half of whom GVG monotherapy was substituted. After 3 months, 55% of patients in group A and 42% in group B were responders (reduction in seizure frequency >50%). After 6 (group A) and 3 years
(group B) of follow-up, 27 and 33% of the patients, respectively, still had good response to GVG. Neurochemical measurements showed a twofold increase in CSF GABA concentrations and minimal or no changes in other neurotransmitter-related parameters. In group C, substitution of GVG as medication tended to normalize the lengthened latencies in somatosensory evoked potentials (SEPs) observed during CBZ treatment. Key Words: Anticonvulsants-y-Vinyl-GAB A-Vigabatrin-Drug toxicity-Electroencephalography-Evoked potentials.
y-Vinyl GABA (GVG, vigabatrin) irreversibly inhibits metabolism of y-aminobutyric acid (GABA) by enzyme GABA transaminase (GABA-T). Administration of GVG in anticonvulsant doses increased GABA level two- to threefold in animal brain and in cerebrospinal fluid (CSF) of animals and humans (Jung et al., 1977; Schechter et al., 1984; Halonen et al., 1988). GVG has been shown to reduce seizure frequency >50% in -40-60% of patients (Rimmer and Richens, 1984; Ben-Menachem et al., 1986; Loiseau et al., 1986; Sivenius et al., 1987; Matilainen et al., 1988) and is particularly effective against complex partial seizures. Few long-term clinical and electrophysiological studies have been performed (Matilainen et al., 1988; Cocito et al., 1989), and data regarding the specificity of the mechanism of action in human epilepsy are not readily available. Non-primate studies have shown microvacuolization in the white matter of the brain at high GVG
doses, with prolongation of the latencies of somdtosensory evoked potentials (SEPs). This has hindered adoption of GVG into wider clinical use, although no such changes have been observed in humans or other primates (Pedersen et al., 1987; Arezzo et al., 1989; Graham, 1989). To monitor the long-term clinical and neurophysiologic effects of GVG and to evaluate the specificity of action by following changes in neurotransmitter-related parameters in CSF, we administered GVG to three different patient groups. We report a 6-year clinical follow-up of 75 patients with complex partial epilepsy and a 3-year clinical and neurophysiologic follow-up of 36 mentally handicapped epileptic patients treated with GVG. CSF samples were analyzed neurochemically for 6-7 months. For neurophysiologic analysis (EEG and evoked potentials) we randomly converted 10 of a group of 20 patients receiving carbamazepine (CBZ) monotherapy to GVG monotherapy . PATIENTS AND METHODS
Received Julv 1991 : revision accepted February 1992. Address coriespondence and reprint requests to Professor P. 5. Riekkinen at Department of Neurology, University of Kuopio, P.O. Box 1627, SF-70211 Kuopio, Finland.
The three groups were designated A, B, and C (Table 1). Group A consisted of 75 patients with 917
A . YLINEN ET A L .
918
refractory complex partial epilepsy and with two or more monthly seizures in the 2-month baseline period (BP). Patients were started on GVG 3 g/day, 1.5 g twice daily, as add-on therapy. After the first 3-month treatment period (TPl), patients showing reduction in seizure frequency 250% (responders) or those showing global improvement in their condition continued to the second 3-month treatment period (TP2). This was a dose-reduction phase in which patients received either 3 or 1.5 giday GVG under double-blind conditions. Patients who were still responsive to GVG or who showed a global improvement then continued to the long-term treatment phase (LTT). Previous antiepileptic drugs (AEDs) were unchanged from BP to the end of TP2, but during the long-term phase, a reduction in some drugs was attempted in a few patients. Lumbar CSF samples were collected at the end of BP, TP1, and TP2, after oral consent was obtained from the patients and approval was given by the ethics committee. Group B consisted of 36 mentally handicapped patients with severe epilepsy; 25 (69%) were severely mentally handicapped (IQ < 50), 9 (25%) were mildly mentally handicapped (50 < IQ < 70), 1 (3%) was borderline (70 < IQ < 90), and 1 (3%) had normal intelligence. Computed tomography (CT) scan was abnormal in 29 (81%) patients, and EEG was abnormal in all but 1. During a 3-month BP, clinical seizures were counted, baseline EEG was recorded and after oral consent was obtained from patients and parents, lumbar CSF samples were collected. GVG was then started as add-on therapy, and appropriate doses for each patient (mean 2.6 i- 0.6 giday) were evaluated for the next 3 months (dose modification phase) while previous AEDs were kept constant. Responders and patients demonstrating global improvement according to the same criteria used for group A continued to longTABLE 1. Patient clinical data Parameter
Group A
Group B
Group C
Sex (MIF) Mean age (yr) Mean age at onset of seizures Seizure type Partial onset Primarily generalized
3414 1 31
23113 29
911 I 37
10
5
75
30
20
0
6
0
Group A (75 patients with complex partial epilepsy), group B (mentally handicapped epileptic patients), group C (20 epileptic patients with carbamazepine (CBZ) monotherapy). Vigabatrin (GVG) was started as add-on therapy in groups A and B, 10 patients in group C were changed from CBZ to GVG monotheraPY. Epilepsia, V d . 33, N o . 5 , 1992
TABLE 2. Number of responders during Jollow-up in three patient groups treated with GVG Duration of GVG treatment
Group A (n = 75)
Group B (n = 36)
Group C (n = 10)
3 mo 1 Yr 2 Yr 3 Yr 6 Yr
41 (55%) 31 (41%) 20 (27%) 20 (27%) 20 (27%)
15 (42%) 12 (33%) 9 (25%) 12 (33%)
7 (70%)
Abbreviations as in Table I . Groups A, B, and C as in Table 1. n, number of patients starting therapy with GVG. Responders in groups A and B: patients who had >50% in seizure frequency. Responders in group C: patients who responded to GVG monotherapy as well as (or better than) to CBZ monotherapy.
term therapy. Lumbar CSF samples were collected, and EEGs were recorded after 7 months, 1 year, and 2 years. Group C consisted of 20 epileptic patients receiving CBZ monotherapy. Ten were randomly changed to GVG monotherapy in a 1-week period. The quantitative EEG (QEEG), SEPs, and visual evoked potentials (VEPs) were recorded in both patient groups at baseline and after a 3-month follow-up period. Controls for QEEG, SEP, and VEP studies were 120, 34, and 21 healthy subjects, respectively. RESULTS Clinical follow-up The effect of GVG on seizure frequency in the three groups is shown in Table 2 (except for the dose modification phase for group A). Results of the shorter follow-up in groups A and B were reported previously (Sivenius et al., 1987, 1991; Matilainen et al., 1989). In group A, there were 41 (55%) responders after 3 months of therapy, 31 (41%) after 1 year, 20 (27%) after 2 years, 20 (27%) after 3 years and 20 (27%) after 6 years of therapy. In group B, 15 (42%) patients were classified as responders at the follow-up point of 3 months and 12 (33%), 9 (25%) and 12 (33%) were so classified at 1 , 2, and 3 years, respectively. Side effects were generally mild and transient in both populations. Drowsiness was most commonly reported (75% in group A and 28% in group B) during the first 3 months, and usually disappeared during long-term treatment. During the dose-modification phase in group A, the median monthly seizure frequency increased in patients receiving 1.5 g/day GVG from 1.7 during TPI (3 giday GVG) to 5.7 during TP2, although this was still lower than the baseline value (1 1 seizures a month, p < 0.01, Wilcoxon’s test). In group C, patients who continued receiving CBZ therapy showed the same seizure frequency
GVG IN EPILEPSY TREATMENT
they had exhibited during the 3-month observation period. Four were seizure-free. Eight of 10 patients who were changed to GVG therapy had previously been seizure-free. Two of these patients had a seizure after changing to GVG monotherapy, 4 days and 1 day, respectively, after stopping CBZ treatment and were changed back to CBZ therapy. One patient with GVG monotherapy was discontinued because of increased gastric irritation and psychic problems. These conditions had existed during CBZ therapy but to a lesser extent. Six patients continued to be seizure-free, and seizure frequency in 1 patient was unchanged with GVG monotherapy. Three of the 6 seizure-free patients expressed a wish to continue with GVG monotherapy after the initial 3-month observation period and have now been seizure-free for 3 years. Neurochemical studies Analysis of CSF showed that levels of total GABA (T-GABA), free GABA (F-GABA), and homocarnosine (HC) were increased two- to threefold in groups A and B during 3-month GVG therapy as compared with baseline. In group A, the increase in T-GABA was greater in responders than in nonresponders, but this difference was not observed in group B. Halving the dose from 3 to 1.5 g/day in group A resulted in an almost 50% reduction in GVG level in lumbar CSF and to a significant decrease in T-GABA, F-GABA, and HC levels. In both groups A and B, there was a slight but consistent increase in glycine levels after GVG treatment, but no treatment-related changes were observed in levels of glutamine, aspartate, asparagine, and tau-
919
rine (Table 3). Levels of acetylcholinesterase (AChE), homovanillic acid (HVA), 5-hydroxyindoleacetic acid (SHIAA), somatostatin, prolactin, and P-endorphin remained constant during GVG therapy of 3, 6, and 7 months’ duration. More detailed data on the levels of the various neurotransmitter-related parameters in the CSF of these patients were reported previously (Halonen et al., 1988; Pitkanen et al., 1988; Riekkinen et al., 1989a,b). Neurophysiologic studies EEG follow-up of patients in group B showed a slight decrease in the amount of slow wave and epileptiform activity during GVG treatment, especially at the 2-year follow-up point and has been reported in detail previously (Pitkanen et al., in press). In group C, the neurophysiologic parameters of patients changed to GVG monotherapy were compared with those of patients who continued to receive CBZ monotherapy . CBZ monotherapy produced statistically significant prolongation of cortical SEP frequencies as compared with those of normal controls. After 3-month treatment with GVG, the cortical SEP peak latencies became significantly shorter (Table 4). QEEG showed no significant changes during GVG therapy as compared with CBZ therapy. All the major pattern VEP peak latencies were prolonged in epileptic patients receiving CBZ as compared with healthy controls. Change to GVG monotherapy produced no significant changes (Mervaala et al., 1989).
TABLE 3. CSF amino acid levels of patients with different responses t o GVG in two different patient populations Amino acid T-GABA BP GVG F-GABA BP GVG HC BP GVG GlY BP GVG Glu BP GVG
Group A Responders (n) 4.87 12.98
f ?
1.25 3.69
(39) (39)”
0.129 0.260
-t f
0.043 (39) 0.072” (39)
Nonresponders (n)
(28) (28)’
0.129 t 0.102 (11) 0.289 f 0.121 (11)’
0.199 f 0.085 (10) 0.359 +- 0.078 (10)’
4.176 t 1.050 (13) 12.906 f 5.799 (11)’
3.389 f 0.955 (15) 10.212 t 1.680
(28) (28)
5.27 i 1.42 6.00 f 1.72’
(32) (32)
5.16 -+ 1.81 6.12 f 1.93”
(18) (8)
(38) (38)
~
0.119 f 0.035 0.247 2 0.073 2.82 2 1.00” 7.72 -t 3.51’,‘
0.497 0.522
~
0.566 0.572
f
0.472
f 0.396
(28) (28)
5.80 f 2.49 7.06 2 1.72
(13) (ll)h
~
Nonresponders (n)
6.29 2 1.23 16.79 i 5.64
(39) (39)
f f
~
~~~~
Responders (n)
4.59 i 10.62 (28) 10.62 f 2.92”,‘ (28)
3.46 i 1.45 10.63 f 4.13’
0.592 0.547
Group B
-~
_.
5.31 15.75
f +-
1.03” (15) 5.10” (10)
(13) (11)
4.92 t 2.19 (15) 5.40 f 1.88
0.122 2 0.067 (13) 0.137 2 0.072 (11)
0.148 i 0.076 (1.5) 0.150 f 0.106 (10)
CSF, cerebrospinal fluid; BP, baseline period; F-GABA, free GABA; Glu, glutamate; Gly, glycine; HC, homocarnosine; T-GABA, total GABA; other abbreviations as in Table 1. Values (mean f SD) are expressed in p M . Statislical significance: ”p < 0.05 and ‘p < 0.01 as compared with values of responders at the same phase (Student’s t test), and ’p < 0.01 as compared with baseline values (paired Student’s t test). Epilepsia, Vol. 33, N O , 5 , 1992
_
_
_
_
A . YLINEN ET AL.
920
TABLE 4. Cortical somatosensory evoked potential peaks during CBZ and GVG monotherupy
Peaks
CBZ (n = 17) latency in ms, mean (SD)
GVG (n = 7) latency in ms, mean (SD)
Erb's point (N9) Cortical (N19)
10.7 (1 .O) 20.1 (1.7)"
10.5 (1.0) 19.6 (1.5)"
Abbreviations as in Table 1. p < 0.01, Student's two-tailed t test, controls versus CBZ group. p < 0.01, Student's two-tailed t test, CBZ versus GVG period in same patients.
'
DISCUSSION
Clinical study Our results show that about one third of patients with severe and previously drug-resistant epilepsy had reduction of at least 50% in seizure frequency during long-term follow-up of 3-6 years. GVG monotherapy appeared to be at least as effective as CBZ monotherapy in 7 of 10 patients in our preliminary 3-month trial. The proportion of patients with a reduction >50% in seizure frequency during GVG treatment during short-term follow-up is in agreement with the results of most published studies (Rimmer and Richens, 1984; Loiseau et al., 1986; Browne et al., 1987; Rimmer et al., 1987; Sivenius et al., 1987). In longterm follow-up, the number of responders is lower. Some responders also dropped out, however, and the values we report are compared with those of the entire patient population at baseline in both trials, which tends to give an artificial negative bias to the results. The number of responders after 1 year of follow-up remained relatively constant in both groups, and there was no evidence of development of tolerance, in contrast to treatment with benzodiazepines, although both types of drug act through the GABAergic system. Neurochemical study In CSF analysis, the GABAergic parameters T-GABA, F-GABA, and H C showed a two- to threefold increase after GVG treatment at 3 and 6 months as compared with baseline values. In group A (75 patients with complex partial epilepsy), the increase in T-GABA level was higher in responders than in nonresponders, but this difference was not observed in group B (36 mentally handicapped patients with severe epilepsy). Both groups showed a slight increase in glycine level after GVG treatment, but GVG apparently did not affect the levels of other amino acids such as glutamate, aspartate, asparagine, and taurine or parameters of the cholinergic (AChE), dopaminergic (HVA), serotonergic Epilepsia, Val. 33, N o . 5 , 1992
(SHIAA), or peptidergic (somatostatin, prolactin, and P-endorphin) systems. The increase in T-GABA, F-GABA, and HC levels after GVG administration is in agreement with the findings of Schechter et al. (1984) and BenMenachem et al. (1986). However, these observations are not necessarily related to good seizure control during GVG treatment. The slight increase in glycine levels in lumbar CSF after GVG could result at least partly from increased release of glycine in the spinal cord, because glycinergic innervation is most concentrated at that CNS level. Despite widespread enhancement of GABAergic transmission in the brain, the C S F markers of cholinergic, dopaminergic, serotonergic, and peptidergic systems appeared not to be significantly affected, emphasizing the specificity of the mechanism of action of GVG. Neurophysiologic study EEG studies showed a slight tendency toward a decrease in the amount of epileptiform activity and slow wave activity during GVG therapy in group B patients, especially after 2 years. Isolating the effects of a single drug in patients receiving polytherapy is difficult, however. For this reason, neurophysiologic parameters were compared in group C patients with GVG or CBZ monotherdpy. CBZ therapy was associated with slowing in occipital rhythm in QEEG and prolonged cortical latencies in SEPs. Change to GVG tended to normalize S E P values. EEG follow-up patients in group B was in agreement with the findings of previous studies (Hammond and Wilder, 1985; Tartara et al., 1986; Mervaala et al., 1989) in which no significant changes were observed in epileptiform activity o r background activity. However, a slight tendency toward a decrease in epileptiform and particularly slow wave activity at 2-year follow-up may be an indication of the effects of GVG therapy. Our findings in group C patients show that GVG therapy does not exacerbate nerve conduction parameters as compared with CBZ monotherapy, as measured by QEEG, VEP, or SEP. On the contrary, change to GVG from CBZ tended to normalize cortical latencies in SEPs. This finding is in agreement with our recent observation that, in newly diagnosed patients with epilepsy, GVG treatment tends to show beneficial effects on EEG and cognition, as compared with CBZ treatment (Kalviainen et al., 1991). Arezzo et al. (1989) showed that in dogs GVGrelated microvacuolization appears and disappears concomitantly with slowing of conduction in SEPs. The lack of functional changes in somatosensory
GVG I N EPILEPSY TREATMENT
pathways in our study is in agreement with results of previous human SEP studies in patients with polytherapy and with human neurophysiologic studies in which no microvacuolization was observed (Pedersen et al., 1987; Graham, 1989). GVG appears to be effective in long-term therapy in approximately one third of patients with previously drug-resistant epilepsy. The action of GVG appears to be specifically mediated through the GABAergic system and apparently there are no signs of neurophysiologically measurable GVGrelated dysfunction in human brain during treatment. REFERENCES Arezzo JC, Schroeder CE, Litwak MS, Steward DL. Effects of vigabatrin on evoked potentials in dogs. Br J Clin Pharmacol 1989;27:53s-60s. Ben-Menachem E, Persson LI, Hamberg A, Schechter P, Haegele KD, Mumford PJ. Changes in CSF parameters and seizure control with every-third-day, every-other-day and daily dosing of gamma-vinyl GABA. Epilepsia 1986;27:649. Browne TR, Mattson RH, Penry J K , et al. Vigabatrin for refractory complex partial seizures: multicenter single-blind study with long term follow-up. Neurology 1987;37:1 8 4 9 . Cocito L , Maffini M, Perfuuro P, Roncallo F , Loeb C. Vigabatrin in complex partial seizures: a long-term study. Epilepsy Res 1989;3: 160-6. Graham D. Neuropathology of vigabatrin. Br J Clin Pharmacol 1989;27:43S-5S. Halonen T, Lehtinen M, Pitkilnen A, Ylinen A, Riekkinen PJ. Inhibitory and excitatory amino acids in CSF of patients suffering from complex partial seizures during chronic treatment with gamma-vinyl GABA (vigabatrin). Epilepsy Res 1988;2: 246-52. Hammond E J , Wilder BJ. Effects of gamma-vinyl GABA on the human electroencephalogram. Neuropharmacology 1985;24: 975-84. Jung MJ, Lippert B, Metcalf BW, Schechter PJ. Gamma-vinyl GABA (4-amino-hex-5-enolic acid), a new selective irreversible inhibitor of GABA-T: effects on brain GABA metabolism in mice. J Neurochem 1977;29:797-802. Kalviainen R, Aikia M, Partanen J, et al. Randomized controlled pilot study of vigabatrin versus carbamazepine monotherapy in newly diagnosed patients with epilepsy: an interim report. J Child Neurol 1991;6:286&9. Loiseau P, Hardenberg JP, Pestre M, Gnyot M, Schechter PJ, Tell GP. Double-blind, placebo-controlled study of vigabatrin (gamma-vinyl GABA) in drug-resistant epilepsy. Epilepsiu 1986;27: 115-20. Matilainen R, Pitkanen A, Ruutiainen T, Mervaala E , Sarlund H , Riekkinen P. Effect of vigabatrin on epilepsy in mentally retarded patients: a 7-month follow-up study. Neurology 1988; 38:743-7. Mervaala E , Partanen J, Nousiainen M, Sivenius J , Riekkinen PJ. Electrophysiologic effects of gamma-vinyl GABA and carbamazepine. Epilepsia 1989;30:189-93. Pedersen B, Hojgaard K , Dam M. Vigabatrin: no microvacuoles in a human brain. Epilepsy Res 1987;1:74-6. Pitkanen A, Matilainen R, Ruutiainen T , Lehtinen M, Riekkinen P. Effect of vigabatrin (gamma-vinyl GABA) on amino acid levels in CSF of epileptic patients. J Neurol Neurosurg Psychiatry 1988;51:1395400. Pitkanen A, Ylinen A, Matilainen R, et al. The long-term antiepileptic efficacy of rigabatrin in drug-refactory epilepsy of mentally retarded patients: a 5-year follow-up study. Arch Neurol
92 I
Riekkinen P, Ylinen A, Halonen T, Sivenius J, Pitkanen A. Cerebrospinal fluid GABA and seizure control with vigabatrin. Br J Clin Pharmacol 1 9 8 9 ~27:873-943. ; Riekkinen P, Pitkanen A, Ylien A, Sivenius J, Halonen T. Specificity of vigabatrin for GABAergic system in human epilepsy. Epilepsia 19896;30:S 18-22. Rimmer EM, Milligan NM, Richens A. A comparison of the acute effect of single doses of vigabatrin and sodium valproate on photosensitivity in epileptic patients. Epilepsy Res 1987;1:33Y-46. Rimmer EM, Richens A. Double-blind study of gamma-vinyl GABA in patients with refractory epilepsy. Lancet 1984;l: 189-90. Schechter PJ, Hanke NFJ, Grove J, Huebert N, Sjoerdsma A. Biochemical and clinical effects of gamma-vinyl GABA in patients with epilepsy. Neurology 1984;34:182-6. Sivenius MRJ, Ylinen A, Murros K , Matilainen R, Riekkinen P. Double blind dose reduction study of vigabatrin in complex partial epilepsy. Epilepsia 1987;28:688-92. Sivenius J , Ylinen A, Murros K, Mumford J, Riekkinen PJ. Vigabatrin in drug-resistant partial epilepsy: a five-year followup study. Neurology 1991;41:562-5. Tartara A, Manni R, Galinberti CA, Hardenberg J, Orwin J , Perucca E. Vigabatrin in the treatment of epilepsy: a double blind placebo-controlled study. Epilepsia 1986;27:717-23.
RESUME Les auteurs rapportent les donnCes cliniques, neurochimiques et Clectrophysiologiques recueillies dans 3 groupes de patients trait& par gamma-vinyl-GABA (GVG, Vigabatrin). Le GVG a CtC commenct comme traitement d’adjonction chez 75 patients prCsentant des crises partielles complexes rebelles (groupe A) et chez 36 patients prksentant un handicap mental avec une Cpilepsie severe (groupe B). Le troisikme groupe (groupe C) Ctait constitut par 20 patients sous monotherapie par carbamazepine (CBZ) dont la moitie a CtC passCe a une monothirapie par GVG. Aprks 3 mois, 55% des patients du groupe A et 42% du groupe B ont t t t considtres comme rtpondeurs (rtduction du nombre d e crises de plus de 50%). Aprks 6 ans (groupe A) et 3 ans (groupe B) de suivi, 27% et 33% des patients respectivement prtsentaient encore une bonne r t p o n s e au GVG. L e s dosages neurochimiques ont mis en Cvidence une multiplication par 2 des concentrations de GABA dans le LCR, sans modifications importantes des parametres lits aux autres neuro-transmetteurs. Dans le groupe C, le passage des patients au GVG a entrain6 une normalisation des latences des potentiels CvoquCs somatosensitifs, anterieurement augmentees sous CBZ. (P. Genton, Marseille)
RESUMEN En tres grupos de enfermos 10s autores presentan informaci6n a largo plazo desde el punto de vista clinico, neuroquimico y electrofisiologico, sobre 10s efectos del gamma vinyl GABA (GVG, Vigabatrin). El GVG se inici6 como terapeutica complementaria en 75 enfermos con ataques parciales complejos refractarios a1 tratamiento (Grupo A) y en 36 enferrnos con retraso mental y epilepsia severa (Grupo B). El tercer grupo (C) consistia en 20 pacientes con monoterapia con carbamazepina (CBZ), la mitad de 10s cuales se habian cambiado a monoterapia con GVG. A 10s tres meses, el 55% de 10s enfermos del grupo A y el 42% del grupo B se clasificaron como “respuestas positivas” (reducci6n del numero de ataques en mas del 50%). DespuCs de 6 (grupo A) y 3 atios (grupo B) de seguimiento, el 27% y el 33% de 10s enfermos, respectivamente, tenian respuestas buenas a1 GVG. Las determinaciones neuroquimicas revelaron un incremento, a1 doble, en las concentraciones de GABA en el liquid0 cefalorraquideo y minimos o ningun cambio en otros parametros relacionados con neurotransmisores. En el grupo C, 10s cambios a1 tratamiento con GVG mostraron una tendencia a la normaEpilepsia, Vol. 33, No. 5 , 1992
922
A . YLINEN ET A L .
lizacion de las prolongadas latencias en 10s potenciales evocados somatosensoriales observados durante el tratamiento con CBZ. (A. Portera-Sanchez, Mudrid)
ZUSAMMENFASSUNG Wir berichten uber klinische, neurochemische und elektrophysiologische Langzeit-befunde unter Vigabatrin (GVG) bei 3 Patientengruppen. GVG wurde bei 75% mit therapieresistenten komplexen Partialanfallen als Zusatztherapie gegeben (Gruppe A) und bei 36 geistig behinderten Patienten mit schwerer Epilepsie (Gruppe B). Die dritte Gruppe umfaRte 20 Patienten mit Car-
Epilepsia, Vol. 33, N o . 5 , 1992
bamazepin-Monotherapie von denen die Halfte auf GVG umgesetzt wurden. Nach 3 Monaten waren 55% der Gruppe A und 42% der Gruppe Responder (Anfallsreduktion von iiber 50%). Nach 6 (A) und 3 Jahren (B) Nachbeobachtung zeigten noch 27% bzw. 33% der Patienten ein gutes Ansprechen. Neurochemische Untersuchungen zeigten einen Lweifachen Anstieg der Liquor GABA-Konzentration und minimale oder fehlende Veranderungen in Neurotrasmitter-abhangigen-Parametern.In Gruppe C zeigte sich eine Normalisierungstendenz der unter CBZ verlangerten Somatosen-sorischen evozierten Potentiale. (C. G. Lipinski, HeidelhergiNeckurjiemun~
