Arch Dis Child: first published as 10.1136/adc.47.252.309 on 1 April 1972. Downloaded from

Short Reports 309 infection. Since Case 18 failed to respond ade- Summary quately to either arginine alone or arginine with Twenty-six subjects with chromosomal anomalies oestrogen priming, an insulin response test was were studied for responsiveness of plasma growth performed, and she also failed to respond to this. hormone levels to one or more stimuli (arginine, oestrogen priming followed by arginine, and insulin- Discussion induced hypoglycaemia). 4 of these subjects The majority of subjects with chromosomal anoma- required more than one test to demonstrate a lies also have a disorder of growth. Tall stature in positive response, and one subject males with multiple X chromosome disorders, and failed to respond to any of 3 tests. It is concluded short stature in subjects with the XO syndrome or that the production of growth hormone in these autosomal aneuploidies are characteristic. No patients is usually normal. adequate explanation for these disorders of growth in subjects with chromosomal anomalies has been REFERENCus advanced. Plasma growth hormone has not been Baker, H. W. G., Best, J. B., and Burger, H. G. (1970). Arginine- infusion test for growth-hormone secretion. Lancet, 2, 1193. studied in these subjects except in a few patients Best, J., Catt, K. J., and Burger, H. C. (1968). Non-specificity of with the XO and XXY syndromes (Frasier, arginine infusion as a test for growth-hormone secretion. Lancet, 2, 124. 1967; Hillman and Colle, 1969; Lundberg andWahl- Frasier, S. D. (1967). The serum growth-hormone response to strom, 1970) in whom the PGH responses were not hypoglycemia in . Journal of Pediatrics, 71, 625. Hambert, G. (1966). Males with Positive Sex Chromatin, p. 22. different from those in normal control subjects. Akademiforlaget, Goteborg. In the present study all subjects with anomalies Hillman, D. A., and Colle, E. (1969). Plasma growth hormone and ofthe sex chromosomes had positive PGH responses insulin responses in short children. American Journal of of Children, 117, 636. (Table). It is of interest that among this group Lundberg, P. O., and Wahlstrom, J. (1970). Hormone levels in the 2 subjects who required more than one test for men with extra Y chromosomes. (Letter to the Editor.) Lancet, 2, 1133. demonstration of a positive PGH response also had Merimee, T. J., Rabinowitz, D., and Fineberg, S. E. (1969). marked growth retardation (more than 2 SD below Arginine-initiated release of human growth hormone. New England Journal of Medicine, 280, 1434. the mean). Moorhead, P. S., Nowell, P. C., Mellman, W. J., Battips, D. M., Among the subjects with autosomal aneuploidy, and Hungerford, D. A. (1960). Chromosome preparations of 3 failed to respond adequately to the initial arginine leukocytes cultured from human peripheral blood. Experimen- tal Cell Research, 20, 613. stimulation test. Two of these (Cases 16 and 24) Parker, M. L., Hammond, J. M., and Daughaday, W. H. (1967). did exhibit adequate PGH responses when the The arginine provocative test: an aid in the diagnosis of hypo- somatotropism. Journal of Clinical and Meta- test was repeated with oestrogen priming. In one bolism, 27, 1129. case (Case 18), a 1696-year-old female who had a Penrose, L. S., and Smith, G. F. (1966). Down's Anomaly, p. 46. age of 3-4 years and mosaicism for a cell Churchill, London; Little Brown, Boston. http://adc.bmj.com/ height Raiti, S., Davis, W. T., and Blizzard, R. M. (1967). A comparison line with possible trisomy D, none of the 3 stimula- of the effects of insulin hypoglycaemia and arginine infusion on tion tests (arginine, arginine with oestrogen priming, release of human growth hormone. Lancet, 2, 1182. insulin) resulted in an adequate response of 5 R. H. A. RuvALCABA,* H. C. THULINE, and V. C. ng/ml above baseline. These results suggest an KELLEY anomaly of growth hormone production in this Rainier School, Buckley, Washington, and Depart- patient which probably contributed to her dwarfism. ment of Pediatrics, University of Washington, The average height of adult patients with Down's Seattle, U.S.A. on October 1, 2021 by guest. Protected copyright. syndrome is approximately 151 cm for males and Washington., 141 cm for females (Penrose and Smith, 1966), *Correspondence to Dr. R. H. A. Ruvalcaba, Rainier School, both values being 3 SD below the mean height for Buckley, Washington 98321, U.S.A. normal subjects. The mean height of 50 adult male subjects with X chromosome anomalies (Klinefelter's syndrome) reported by Hambert Folic Acid Replacement in (1966) was 180-3 cm ± 6 8 cm, which is nearly 1 SD above the mean height for normal adult Folate-deficient Children on males. Since the groups of subjects with Anticonvulsants Klinefelter's and Down's syndromes in our study had normal PGH responses, the characteristic short Reynolds (1967) found that folic acid reversed stature in subjects with Down's syndrome and the the retarding effect of anticonvulsants in 22 out of tall stature in males with multiple X chromosome 26 folate-deficient adult patients, and Neubauet anomalies are probably not related to anomalies of (1970) noted a similar improvement in 28 out of 50 growth hormone production. children. Reynolds (1967) also reported an in- Arch Dis Child: first published as 10.1136/adc.47.252.309 on 1 April 1972. Downloaded from

310 Short Reports crease in fit frequency. Neither study was con- Some children were included in the controlled trolled. Because the increased energy, drive, and trial whose serum folate levels were within the 'speed of cerebration' described by these authors normal adult range for this laboratory (3-10 ng/ml) would benefit children's education, we undertook but the red cell folate levels (Dacie and Lewis, 1968) a control trial of folic acid versus placebo in folate- confirm that they were folate-deficient (Fig.). There deficient children attending a convulsion clinic. was a marked rise in serum and red cell folate levels Simple reaction time and the numbers of hours slept within 9 days of taking folic acid (Fig.). Though were chosen as the objective measurements most there was a significant rise in Hb concentration likely to reflect the changes in behaviour noted by (P <0 03), no variation was found in mean corpus- Reynolds and Neubauer. Fit frequency was also cular volume, serum iron, or serum B12 recorded. (Matthews, 1962). Patients and Methods We first measured the levels of serum folate (Ball and Giles, 1964), Hb concentration, and mean corpuscular I volume in 39 children (25 boys and 14 girls) who had been taking anticonvulsant drugs for over 3 months. 1,200 We compared their levels with those in 25 children (15 boys and 10 girls) of similar age who were not taking 1,000 anticonvulsants; they were either outpatients or they had been admitted to hospital very recently. -g8T 800 Of the anticonvulsant group, 25 children had serum 4) folate levels below 5 ng/ml. With their parents' permis- 4.- sion, 19 of these entered a trial comparing an 8-week 6600- period of treatment with oral folic acid, 5 mg daily, with a similar period on placebo. The children were used as their own controls and the placebo was given first. o-400 The simple reaction time of those children able to co- operate with the test was estimated on 3 occasions 200- during each treatment. They pressed a button in response to a light signal; we varied the period between 0 signals by hand in a sequence derived from a table of time was measured FIG.-Red cell folate levels during placebo and folic random numbers and the reaction acid administration: shaded area indicates the normal and displayed electronically. The parents kept records adult this of the days on which fits occurred and the number of range for laboratory. hours slept. Haematological data obtained while on http://adc.bmj.com/ placebo were compared with those in the second and Up to 8 weeks after the introduction of folic eighth week on folic acid. acid there was no significant change in the reaction time of the 16 children able to co-operate with the Results test. Similarly there was no significant difference The children taking anticonvulsant treatment in the number of hours slept by the 16 children with had significantly lower serum folate levels than the completed records. Of the 18 children with children of the same age not taking these drugs completed records of days on which fits occurred, (Table), but there was no significant difference in 9 were free of fits during the trial, and the number on October 1, 2021 by guest. Protected copyright. the Hb or mean corpuscular volume (P >0-1). of days on which the other 9 had fits was not affected by folic acid. Two children complained of 'dizziness' while on folic acid but there was no TABLE such complaint while taking the placebo. Comparison of Serum Folate Levels in Children with and without Anticonvulsant Treatment Discussion

No. of Cases Our preliminary survey confirms previous obser- Group vations that anticonvulsant therapy is frequently Serum Folate Serum Folate Total associated with low serum folate levels in children median* (Dahlke and Mertens-Roesler, 1967; Neubauer, Anticonvulsant 25 14 39 1970). No anticonvulsant 7 18 25 The lack of behavioural response to folic acid replacement in this trial is in keeping with the P =*001 (Fisher-Yates 2 x 2 contingency tables). *Median serum folate level (all children) = 5 ng/ml. findings of 3 double-blind trials in adults (Grant Arch Dis Child: first published as 10.1136/adc.47.252.309 on 1 April 1972. Downloaded from

Short Reports 311 and Stores, 1970; Jensen and Olesen, 1970; Dahlke, M. B., and Mertens-Roesler, E. (1967). Malabsorption of folic acid due to diphenylhydantoin. Blood, 30, 341. Ralston, Snaith, and Hirley, 1970). It is unlikely Grant, R. H. E., and Stores, 0. P. R. (1970). Folic acid in folate- that the differences between these results and those deficient patients with epilepsy. British Medical Journal, 4, 644. of Reynolds (1967) and Neubauer (1970) can be Jensen, 0. N., and Olesen, 0. V. (1970). Subnormal serum folate explained by shorter exposure to folic acid; Neu- due to anticonvulsive therapy: a double-blind study of the bauer (1970) noticed mental improvement in effect of folic acid treatment in patients with drug-induced sub- normal serum folates. Archives of Neurology, 22, 181. younger children from 5 to 8 weeks after starting Matthews, D. M. (1962). Observations on the estimation of serum folic acid and the adults reported by Grant and vitamin B12 using Lacrobacillus leichmannii. Clinical Science, 22, 101. Stores (1970) received folic acid for 6 months with Neubauer, C. (1970). Mental deterioration in epilepsy due to no effect. We have now shown that normal . British Medical Journal, 2, 759. Ralston, A. J., Snaith, R. P., and Hirley, J. B. (1970). Effects of levels of folate are achieved very rapidly in both the folic acid on fit-frequency and behaviour in epileptics on anti- red cells and serum when folic acid is given to convulsants. Lancet, 1, 867. folate-deficient patients on anticonvulsants; these Reynolds. E. H. (1967). Effects of folic acid on the mental state and fit-frequency of drug-treated epileptic patients. Lancet, results taken together with the lack of behavioural 1, 1086. response in the controlled trials suggests that, in this context, folic acid has little direct effect on J. R. Moonu* and E. W. BALL mental function. The possibility remains that The Children's Hospital, and the Haematology the very slow deterioration in mental performance Department, Selly Oak Hospital, Birmingham. noted in some epileptics might be related to chronic *Correspondence to Dr. J. R. Moore, Good Hope Hospital, folic acid deficiency. Sutton Coldfield, Warwickshire. We also confirmed the results of the controlled trials in adults (Grant and Stores, 1970; Jensen and Olesen, 1970; Ralston et al., 1970) in not showing an increase in fits due to folic acid. Folic Trimethoprim/Sulphamethoxazole acid has undoubtedly provoked fits in some indivi- in Pertussis: Comparison with duals (Chanarin et al., 1960; Reynolds, 1967), but this response seems unusual. Tetracycline Antibiotics, if given early in the , are Summary moderately effective in reducing the frequency and Significantly lower serum folate levels were severity of cough in pertussis (MRC Report, 1953). found in 39 children taking anticonvulsant drugs Tetracycline, chloramphenicol, and erythromycin than in 25 controls (P = 0 01). Because of

have all been recommended (MRC Report, 1953; http://adc.bmj.com/ reports of improved mental function after folic Christie, 1969; Bass et al., 1969). acid replacement in folate-deficient epileptics, 19 Trimethorprim/sulphamethoxazole is a bacterici- children entered a trial of folic acid versus placebo. dal combination active in vitro against Bordetella There was no significant difference in simple pertussis (Bushby, 1969), and it was considered reaction times, in numbers of hours slept, or in fit that it might prove a suitable alternative agent in frequency. The increase in fits due to folic acid the treatment ofpertussis. It was therefore decided which has been reported seems to be an uncommon to undertake a direct comparison between this agent response to the vitamin. and tetracycline, during a recent outbreak of the on October 1, 2021 by guest. Protected copyright. disease in Nigeria. We would like to thank Drs. G. F. A. Harding and P. M. Jeavons for help with this project; Mr. R. S. Easterby for advice about the electronic equipment Materials and Methods which was kindly loaned by the University of Aston in Patients were included in the trial if Bord. pertussis Birmingham; and Macarthy's Laboratories for supplying was isolated from the naso-pharynx, or if they had a folic acid and inert tablets through the agency of Mr. typical 'whooping' cough, and a relative and absolute R. H. Leach, pharmacist, at the Children's Hospital. lymphocytosis. No account was taken of previous vaccination history. Patients accepted into the trial REFRENCES were randomly allocated to one of two treatment Ball, E. W., and Giles, C. (1964). Folic acid and vitamin B12 groups. levels in pregnancy and their relation to megaloblastic anaemia. Children under 2 old Journal of Clinical Pathology, 17, 165. (a) Tetracyline group. years Chanarin, I., Laidlaw, J., Loughridge, L. W., and Mollin, D. L. were given 62 - 5 mg tetracycline 6-hourly, and older (1960). Megaloblastic anaemia due to phenobarbitone: the children 125 mg 6-hourly, for one week. convulsant action of therapeutic doses of folic acid. British (b) Trimethoprimlsulphamethoxazole group. Children Medical Journal, 1, 1099. Dacie, J. V., and Lewis, S. M. (1968). Practical Haematology, 4th under 6 months old were given 20 mg trimetho- ed., p. 407. Churchill, London. prim with 100 mg sulphamethoxazole twice daily 11