DiscussionGrowth, behaviour, and educational achievement are influenced by many factors, and, although we found that the AA and AS children were alike in some of these, we cannot be certain that they were alike in all. Every child, however, had lived since early childhood in a small community in which living standards, child rearing practices, and educational opportunities varied within narrow ranges, and there were unlikely to be major differences in the backgrounds of the two groups.The absence of differences in growth and achievement between children with the two genotypes conflicts with some of the results of the study in the USA,4 in which significantly lower weights and poorer performances in psychological-intellectual tests were found in 19 AS compared with 241 AA children aged 10-16 years. No evidence was presented in that report, however, to show that the two groups were matched for socioeconomic or other factors that might have influenced the measurements. Furthermore, the selection of patients from twin studies may complicate assessment of the results. Perinatal hypoxia is more likely to occur in twins, and, although levels of Hb S are low in the sickle-cell trait at birth, the possible role of severe hypoxia precipitating sickling cannot be entirely excluded.Both studies were limited by small numbers, and carefully controlled investigations of larger groups of children are needed. Meanwhile, the available evidence does not warrant the conclusion that che sickle-cell trait affects physical and mental development. Although PGF,, was associated with a somewhat higher frequency of minor side effects than hypertonic saline, notably vomiting and diarrhoea, these were within acceptable limits. Only 59 women (8 2%) in the prostaglandin group had more than four episodes of vomiting and 11 (151o) more than four episodes of diarrhoea. Other side effects occurred only occasionally. No difference was found between the two groups in the frequency of incomplete abortion or excessive bleeding.
IntroductionEarly studies of the intra-amniotic administration of prostaglandin (PG) F,2 showed that doses of 5-25 mg often needed to be repeated to induce therapeutic abortion.'-8 When an initial dose of 25 mg was repeated in 24 hours (if abortion was not