[Nosology of Lennox-Gastaut syndrome].

After a short historical review, the symptomatology of the Lennox-Gastaut syndrome (LGS) as described in the past 30 years is summarized. Next, all papers published in the past 25 years and presenting the author's own patients are critically reviewed. These considerable patient data enabled to some extent supplementary statistical evaluation of the symptoms and signs of LGS. However, only three of the papers reported largely similar symptom complexes whose components were often combined. While not adequate to allow statistical evaluation, these data have been reviewed with descriptive analysis. The resulting diagnostic criteria correspond to those established by Gastaut in 1982 and are convincing because of their frequency of appearance. In addition, they confirm the 1989 description of LGS by the Commission on Classification and Terminology of the International League Against Epilepsy. These criteria and their frequency are: (1) diffuse slow spike waves in the EEG (100%). (2) tonic seizures (94%), (3) atypical absences (80%), (4) runs of rapid spikes in NREM sleep (approx. 70%), (5) status epilepticus (60%), (6) atonic seizures (43%). Resistance to therapy and persistence of epilepsy are amongst the most frequent features. Mental retardation is a leading symptom, occurring on average in 90% of cases. Reliable statistical analysis of the electroclinical data should be performed following the numerical taxonomy and should provide nosological entities and classifications based on objective, reliable and logical fundamentals. This is an indispensable prerequisite for differential diagnosis. Sections follow which discuss recent morphological and neurometabolic findings concerning the etiology as well as the genetics of LGS. The discussion of the differential diagnosis outlines the nosological delineation of LGS from epilepsy with myoclonic-astatic seizures, benign partial epilepsy of childhood with centrotemporal sharp waves, certain focal epilepsies of the frontal and temporal lobe. Lastly, the myoclonic variant of LGS is discussed. This review shows how frequently in the past LGS was investigated using deficient methodology. Additional studies should be undertaken in collaboration with experienced statisticians in order to complement the above analysis of the syndrome.