Metabolism of tryptophan in childhood epilepsy.

A relation between pyridoxine and convulsive phenomena has been established over several years (Snyderman, Holt, Carretero, and Jacobs, 1953; Molony and Parmelee, 1954; Adams, Bessey, Bussey, and Hansen, 1954; May, 1954; Coursin, 1954, 1955). This led to a search for evidence of pyridoxine deficiency among children with cryptogenic epilepsy. Soon patients were discovered who, though receiving a normal amount of pyridoxine, stopped convulsing when given supplements of this vitamin and relapsed when these were withdrawn (Hunt, Stokes, McCrory, and Stroud, 1954; Bessey, Adams, and Hansen, 1957; Sokoloff, Lassen, McKhann, Tower, and Albers, 1959; Marie, Hennequet, Lyon, Debris, and Le Balle, 1959; Scriver, 1960; Garty, Yonis, Braham, and Steinitz, 1962; Waldinger and Berg, 1963; Scriver and Hutchison, 1963). This condition, named pyridoxine dependency, may be an inborn error ofmetabolism and should be distinguished from a dietary deficiency of the vitamin, which also results in convulsions. Cochrane (1959) reported an apparent association of the form of childhood epilepsy known as infantile spasms ('lightning fits', 'massive myoclonic epilepsy', West's syndrome), with pyridoxine deficiency. The five patients he studied excreted more than the normal amount of xanthurenic acid following the administration of tryptophan, a finding in accord with pyridoxine deficiency, and on administration of this vitamin, 'a definite decrease in the frequency of their convulsions and an improvement in mental status was observed'. This was further investigated by Bower (1961) and Bower and Hughes (1961) who reported on the tryptophan load test in a small group of normal children; in a group of patients with infantile spasms (subdivided into those associated with some other disorder, 'symptomatic' and those without, 'cryptogenic'); and in a small group which included, without differentiation, children with other forms of epilepsy and mental deficiency. The results of this preliminary study suggested that half the children with cryptogenic infantile spasms