Phenylketonuria in South Africa. A report on the status quo.

During the 1980s a pilot newborn screening programme for the early detection (and treatment) of amino acidopathies, especially phenylketonuria (PKU), was conducted by the Department of National Health and Population Development. The motivation for this pilot programme was the high priority accorded PKU screening in Europe and North America and the presumed similarly high incidence of this condition among South Africans of European origin. From a cohort of 59,600 newborns screened in the Pretoria area over a period of 8 consecutive years (1979-1986), only 1 case of PKU (and 1 of tyrosinaemia) was found. Statistically this result is compatible (Poisson distribution, 95% confidence interval) with a 'true' incidence of not more than 3/59,600 (or about 1/20,000) newborns. It is concluded from this result and other relevant information that newborn screening for PKU and other amino acidopathies is not cost-effective and justifiable, especially against the background of prevailing demographic conditions and more pressing health priorities in South Africa. This particular screening programme was discontinued in 1986. The results and conclusions are presented here for the record.

Family history as a risk factor for coronary heart disease in South African families.

This article presents data on, and applies a procedure for the statistical quantification of, family history as a risk factor for coronary heart disease (CHD) in three sub-samples (groups) of families: I--a healthy control group; II--families with familial hypercholesterolaemia (FH); and III--families identified by an index case with CHD. With regard to the average family history of CHD (calculated as an index for each family, and as a mean index for each group), group II differs significantly from group I and marginally significantly from group III; family groups I and III do not differ from each other statistically. By means of significance tests developed for this purpose, the groups of families are shown to be significantly heterogeneous, by being composed of families highly resistant against and susceptible to CHD. This is illustrated for example in group II, where some FH families can be shown to be highly resistant to CHD, compared with other FH families with a very strong history of CHD. The exact number and proportion of such families at different levels of significance is calculated and the actual families with the highest and lowest calculated family indices, respectively, are then identified (illustrated by examples). The practical significance of the statistical procedure of quantifying and applying family history as a risk factor for CHD, is discussed in terms of epidemiological and preventive health considerations.

Preventing cystic fibrosis in the RSA.

Prenatal diagnostic testing for cystic fibrosis (CF) in South Africa has been available by microvillar enzyme (MVE) assay since 1984 and by DNA investigation since 1987. The advantages and practical uses of these two procedures are reviewed. Over the period 1984-1989, 59 MVE assays and 13 DNA investigations (1 woman had both done; total number of pregnancies tested 71) were performed for the prenatal diagnosis of CF in high-risk families. Of the 71 pregnancies tested (65 white woman, 4 mixed race and 2 Indian), 18 fetuses were found to be affected: of these, 15 couples chose to have the pregnancies terminated. In 2 fetuses tested the MVE assays were 'equivocal' and the babies were born affected. By using population genetic and demographic data it is estimated that the present rate of prenatal diagnosis and prevention satisfies about one-quarter of the projected and practically achievable annual need. It is also tentatively shown that only about 60% of the projected number of high-risk families are at present on record. It is concluded that more systematic efforts should be directed at carefully guided information and awareness campaigns, in order to draw more CF families into the mainstream of voluntary genetic services. The identification and cloning of the CF gene (in 1989) has made it possible to extend considerably the present strategy of prevention and to include lower-risk and extended CF families in due time.

Duchenne muscular dystrophy--a molecular service.

In 1987 a carrier detection and prenatal diagnostic service for Duchenne muscular dystrophy using molecular technology was instituted at the Department of Human Genetics, University of Cape Town, to serve affected families in southern Africa. DNA samples from 100 affected male subjects and 350 of their relatives from a total of 110 families have been banked. To date restriction fragment length polymorphism (RFLP) analysis and deletion screening has been performed on the DNA of 60 male patients and 116 female relatives at risk of being carriers of the faulty gene. The DNA probes used were pERT 87-1 (MspI polymorphism) pERT 87-15, pXJ1.1, pXJ2.3 (TaqI, polymorphism), pXJ1.2 (BclI polymorphism), P20 (MspI and EcoRV polymorphism) and the cDNA probes. DNA deletions have been detected in 30 of the 60 affected boys and the carrier risks of 49 women have been determined by RFLP analyses. In those families where the risks were uncertain because the affected males had died, prenatal exclusion testing was offered to potential carriers. Two pregnancies were terminated when male fetuses were shown to be affected, since they had the same deletion as that observed in the proband.

Nonrandom association of genetic markers in a sample of South African blacks.

According to classical genetic theory, allelic genes at one locus are expected to segregate and be manifested independently of allelic genes at another locus. At the population level any significant deviation from this general hypothesis resulting from specific biologic and genetic effects can be recognized in the form of nonrandom associations between genetic markers. The present data, consisting of 24 genetic polymorphisms determined from a sample of 998 unselected and unrelated South African blacks, offers an opportunity to test whether or not any such nonrandom associations exist between the genetic markers. After appropriate statistical calculations on the population data, we found that 13 pairs of genetic polymorphisms demonstrate a nonrandom association (statistically significant). Because the results cannot be explained in terms of known biologic mechanisms, we conclude that the associations observed could be due to random statistical effects (repeated application of the chi-square test) and/or to real (as yet unknown) biologic phenomena in the population studied. This tentative conclusion can serve as a guideline for more specific investigations.

On the prevention of rhesus immunisation in the RSA.

The utilisation rate of anti-Rh antiserum in South African population groups for the years 1983-1985 was investigated. The number of pregnancies at risk of fetomaternal sensitisation because of Rh incompatibility and hence the number of women requiring prophylactic treatment with anti-Rh anti-serum shortly after delivery can be calculated from the known frequency of Rh-negative subjects in each population group. By relating the number of doses of anti-Rh antiserum actually distributed and used to the number of women requiring this prophylactic treatment, the crude utilisation rate of anti-Rh antiserum is calculated as 41-44% for all population groups combined. The rate for blacks is calculated at 14-20%, for whites 89-94%, for Indians 59-64% and for Coloureds 45-51% (the range of variation covers the years 1983-1985). These figures are discussed in terms of the approximations made for the calculations (hence 'crude utilisation rate'), the experience in other countries (specifically the USA) and the need for improving this facet of primary health care.

Linkage between the variegate porphyria (VP) and the alpha-1-antitrypsin (PI) genes on human chromosome 14.

From family studies close linkage between the gene locus for variegate porphyria (VP) and the alpha-1-antitrypsin (PI) gene became evident. The maximal lod score from male meioses was 4.33 at theta = 0.04 and from both sexes combined 3.56 at theta = 0.12. Three pedigrees were triple informative regarding loci VP, PI, and IGHC (immunoglobulin heavy chain cluster, Gm polymorphism). In two of the respective meioses recombinations were observed, and in both cases the co-segregating VP and PI alleles were separated from the Gm haplotypes. These findings argue in favour of gene order either VP:PI:IGHC or PI:VP:IGHC.

Neonatal screening for congenital hypothyroidism. A decade's review, including South Africa.

Since its introduction in 1974, neonatal screening for congenital hypothyroidism has been very extensively conducted world-wide. It is the commonest disorder found in all neonatal screening programmes, occurring in 1:3,000-8,000 births. Laboratory testing for thyroid function within the first 5 days after birth identifies this disorder weeks to months before clinical symptoms become evident and the clinical diagnosis is made. A screening programme in Pretoria (February 1981- October 1986) has identified 11 hypothyroid neonates in some 45,577 infants tested. Internationally, there is now a 10-year experience (and a follow-up for a number of programmes) of neonatal hypothyroidism diagnosed by laboratory screening tests and treated within weeks of birth. There is conclusive evidence that the physical and mental development of these children is within, or approaches to a great degree, normal limits.

Linkage relationship between the genes for adenosine deaminase and S-adenosyl-homocysteine hydrolase on human chromosome 20.

The genes for adenosine deaminase (ADA) and S-adenosyl homocysteine hydrolase (AHCY or SAHH) are known to be syntenic and within measurable distance from each other, on chromosome 20 in man. In the present study an informative family is described in which the recombination fraction (theta) between the respective genes is estimated to be about 0.18. Together with the published finding of theta = 0.15 (Eiberg and Mohr 1985) in informative Danish families, the recombination fraction for the pooled data is calculated to be theta = 0.14 (in men), theta = 0.08 (in women) and theta = 0.13 (both sexes taken together).

Alpha-1-antitrypsin genetic polymorphism in South Africa.

The alpha 1-antitrypsin (AAT) or protease inhibitor (Pi) genetic polymorphism was studied in 144 white, 100 coloured, 104 Indian and 127 black (Northern Sotho) healthy individuals (controls), in the Pretoria area. Their Pi phenotype and gene frequency distributions are compared with world-wide data on other population groups. The severely deficient Pi phenotypes S, Z and SZ jointly attain frequencies of 0.3-0.4% in coloureds and whites; in blacks and Indians the corresponding frequencies are very much lower. The implication for preventive medicine and public health is that in South Africa the sequelae of Pi deficiencies such as cirrhosis of the liver and/or emphysema of the lung are of practical importance in whites and coloureds and much less so in blacks and Indians. In 176 white breast cancer patients studied, the Pi phenotype and gene frequency distributions were found to be similar to those of healthy controls (not statistically significant). Cohorts of other patients were also phenotyped because of their low alpha 1-globulin concentrations in routine serum protein electrophoresis and/or their specific disease condition (cirrhosis of the liver or emphysema of the lung) known to be associated with AAT deficiency. These results are discussed in terms of their significance for family follow-up, genetic counselling and a preventive service. The need to avoid atmospheric pollution, especially cigarette smoke, is emphasised as a major and cost-effective preventive measure.

The plasminogen polymorphism in South African Negro populations: genetics and anthropogenetics.

Genetic polymorphism of human plasminogen (PLG) was investigated in 1252 unrelated individuals from eight South African Bantu-speaking Negro tribes. PLG phenotypes were determined by isoelectric focusing (pH 3.5-9.5 and 5-8 gradients) of neuraminidase-treated samples and subsequent detection by caseinolytic overlay or immunoblotting with specific antibody. No significant difference in the distribution of PLG alleles among the eight ethnic groups was observed. The combined allele frequencies of the common alleles in South African Negroes were 0.6977 for PLG*A, 0.2736 for PLG*B. In addition, six rare alleles were seen: PLG*A3, *A1, *M2, *B1, *B2, *B3. The rare variant PLG*B2 was proven to segregate by autosomal Mendelian inheritance in a family. The combined frequency for the rare alleles was 0.0287. The distribution of phenotypes in the total population sample was found to be in Hardy-Weinberg equilibrium. A striking difference in PLG allele distribution between Negroes from South Africa and published Negroid frequencies from North America could be observed. This difference was also seen in comparison with Mongoloid populations; in contrast, PLG frequencies for South African Negroes were similar or almost identical to known Caucasoid distributions.

On the genetic interrelationships of South African Negroes.

This study addresses the comparative genetic interrelationships between South African Negro groups. For this the genetic distances between seven ethnically defined Negro subsamples (total of 998 individuals) based on 24 genetic loci/polymorphisms are calculated by applying standard distance formulae. These computations offer an opportunity to evaluate the different polymorphisms in terms of their effects on the genetic distances. The genetic interrelationships thus computed are illustrated by way of dendrograms and are discussed in terms of their comparative significance. It follows from the findings that the Ndebele, Northern Sotho (Pedi), and Tswana form a closely related subcluster and that the Zulu and Swazi as well as the Venda and Shangana-Tsonga form two additional, more distant, subclusters. These results are discussed and tentatively interpreted against the background of the reported Khoisan admixture of the populations concerned as well as their ethnological history. The data are also compared to those derived from metric and dermatoglyphic studies. It is concluded that whereas there is some agreement between these categories of variation (genetic, metric, and dermatoglyphic) as far as the comparative evaluation of South African Negro groups is concerned, there also are discrepancies. These conclusions need to be explained in terms of evolutionary mechanisms (such as historic origins, hybridization, natural selection, and genetic drift) in order to obtain a more consistent and comprehensive comparative picture of the physical anthropology of southern African populations.

Maternal serum alpha-fetoprotein screening in Natal. Results of over 12 000 tests.

Mass screening of maternal serum alpha-fetoprotein levels for the prenatal detection of fetuses with neural tube defects (NTDs) was introduced in Natal during 1979. From then until 31 October 1982, 12318 pregnancies were monitored and 16 fetuses with NTDs identified. A further 7 abnormal fetuses, 3 with defects other than of the neural tube, were encountered. It is shown that the screening process is reliable and that it does not generate much additional work for ultrasonographers or other laboratories. Few problems related to the management of the screening programme were met; they are discussed with particular reference to the role of ultrasonography.

Screening for inborn errors of metabolism among mentally retarded patients. Outcome of a surgery at the Witrand Care and Rehabilitation Centre.

The prevalence of different types of inborn errors of metabolism among the mentally retarded patients at the Witrand Care and Rehabilitation Centre, were determined by means of a biochemical screening survey. These results are compared with those of other surveys in South Africa and abroad. One important result points to substantial differences in the recorded incidences of metabolic defects between surveys. This observation could partially be due to significant differences between the different studies in terms of methodology employed and sampling procedures. The questions raised in this regard are documented and discussed.