Malaria and Stress in Relation to Haemoglobins S and C*

In a previous communication (Thompson, 1962) evidence was presented in favour of the existence of partial protection against Plasmodiumn falciparum malaria among Ghanaians with either the sickle-cell or haemoglobin C trait. Comparison of mean parasite densities among young children in Accra showed significantly lighter infections among heterozygous carriers of haemoglobin S or C than among those possessing normal haemoglobin alone. The protection co ferred by haemoglobin S was apparently greater than that due to haemoglobin C, but the difference demonstrated is as not significant. The statistical treatment of these results was criticized by J. P. Garlick (personal communication, 1962) on the grounds that the t test was an invalid method of comparing untransformed parasite densities with large standard deviations and positively skewed distributions. Reappraisal of previous results has been necessary on this account, and has resulted in a more accurate assessment of the relationship between malaria and haemoglobins S and C respectively. A much smaller advantage for heterozygous carriers of haemoglobin C was forecast by Allison (1961), who thought that it might be difficult to obtain unequivocal evidence of the existence of a protective effect against malaria in th6 field. Equilibrium between the haemoglobin S and C genes in Ghana could be envisaged if it was demonstrated that haemoglobin S heterozygotes possessed a greater advantage over the normal than did haemoglobin C heterozygotes, since this would help to compensate for the known lower survival rate of haemoglobin S homozygotes as compared with their haemoglobin C counterparts. However, the accepted efficiency of haemoglobin S in limiting malarial mortality is not entirely without its disadvantages. There now exist several reports illustrating the morbidity and mortality attributed to the sickle-cell trait in negroes exposed to stress not only at high altitudes (Rotter et al., 1956) but also at sea-level (Edington and Lehmann, 1955; Ende et al., 1955; Ober et al., 1960; McCormick, 1961). Some further examples are reported in this paper.