The electroencephalogram in veganism, vegetarianism, vitamin B12 deficiency, and in controls.

The electroencephalogram (E.E.G.) is often abnormal in patients with pernicious anaemia. This abnormality is not related to age or to haemoglobin level and is reversible with vitamin B12 therapy. In 76 cases of pernicious anaemia with abnormal E.E.G.s (Walton, Kiloh, Osselton, and Farrall, 1954) the changes in the E.E.G. were reversed in 60 (79 %) by the administration of vitamin B12, the improvement beginning in seven to 10 days and usually being complete in a month. Serum vitamin B12 levels were not reported in that study, but the suggestion was made that the E.E.G. changes were due to a specific defect in cerebral metabolism. In another study (Smith, 1962) the E.E.G. was found to be abnormal in nine of 11 vegans, the abnormality being associated vwith low serum vitamin B12 levels, the three most abnormal E.E.G.s being associated with values of 48, 80, and 90 ,u,g./ml. Vegans are extreme vegetarians who exclude not only meat but also dairy produce and eggs from their diet. Since it has become realized that vegans may develop neurological complications from dietary vitamin B12 deficiency, many have supplemented their diet with a compound yeast extract or a milk substitute of vegetable origin which contains vitamin B12. In the present paper we report an investigation of 17 vegans. Their E.E.G.s were studied to see whether abnormalities were still present after low-content vitamin B12 supplements. In a few of the vegans, E.E.G.s were repeated after intensive vitamin B12 supplementation to try to detect reversal of abnormalities. The reporting of the diffuse types of E.E.G. abnormality seen in metabolic disorders, particularly when these are mild, is apt to be highly subjective and subject to observer error, a bias likely to be increased when automatic frequency analysis is not available as was the case with our investigation. We therefore decided that the E.E.G. scoring should be done 'blind'. Furthermore, as about 10% of the population have diffuse non-specific deviations from a well-formed alpha rhythm in the E.E.G. we decided to include control subjects so that any abnormalities in the vegans' E.E.G.s would have to differ significantly from the controls before they could be considered as possibly metabolic in origin. In addition, some vegetarians and some patients with vitamin B12 and with folic acid deficiency were included in the expectation of forming a continuum of E.E.G. abnormality, ranging from the non-metabolic to the definitely metabolically abnormal types. Another variable in the E.E G. that we thought might intrude was the mild, diffuse, non-specific abnormality due to anxiety proneness (Ulett, Gleser, Winokur, and Lawler, 1953) or to general personality or constitutional differences (Hill, 1963). We therefore included a personality questionnaire to measure extroversion and neuroticism in the groups with the expectedly more normal E.E.G., namely, the controls, the vegetarians, and the vegans. Personality and constitutional differences have been related to the type of body build (Kretschmer, 1951) so that we used a simple index calculated from the height and weight (Schrade, Boehle, and Biegler, 1960). Thinking that an error might be introduced in the body build index through the vegans being possibly underweight, we calculated an obesity index by relating the observed weight for height to standard values (Kemsley, Billewicz, and Thompson, 1962).