Human immunodeficiency virus-1 infection and the acquired immunodeficiency syndrome in African children : natural history from birth to early childhood.

Abstract

Background: in 1987, the first child with HIV-1 infection was identified in the paediatric wards at King Edward VIII Hospital in Durban. This made paediatricians aware that the epidemic had spread to the children of KwaZulu/Natal. Although information on transmission and natural history was becoming available from developed countries, little was known about the disease in developing countries. It was important to determine transmission rates and disease patterns in the local population, in order to appropriately counsel women, and for management of infected infants. In addition, with resources for laboratory diagnoses being limited in developing countries, much emphasis had to be placed on clinical findings for identification of infected children. In 1989, a retrospective analysis was made of the HIV-infected children seen over a 2-year period, between 1987 and 1989. Nine such children were identified and their clinical and biochemical features were described. It was concluded that HIV infected children presented with an identifiable pattern of signs, fairly similar to that described for children in industrialised countries. With these findings, a prospective study was undertaken, to determine the vertical transmission rate, the factors affecting this rate, and natural history of vertically transmitted I-IIV-1 infection. ix KwaZulu/Natal, being at the epicentre of the epidemic in South Africa, was a natural site for the study. Patients and Methods: a trained research worker was placed in the antenatal clinic at King Edward VIII Hospital for the specific purpose of educating, counselling, and testing of all women attending the clinic. Women attending the clinic for the first time in the index pregnancy were offered HIV testing if informed consent was obtained. Blood for HIV serology was drawn at the same time as sampling for the obligatory syphilis serology. The acceptance rate for sampling was > 95%. The majority of the women attending the clinic were black, and first attendance was generally late, into the third trimester. The same research worker was responsible for post-test counselling which was offered to all the women, not only those who tested positive. This research worker was also responsible for obtaining maternal consent for entering the newborn infant into the study. All newborn infants were seen within 48 hours of birth. At this time they were examined, growth parameters were recorded, and initial blood samples taken. These infants were then followed-up at 1 month, 2 months, 3 months, then at 3-month intervals up to 18 months, then at 6-month intervals. At each visit, a thorough clinical examination was performed, growth measurements taken, and development assessed. Record was made of any interim illness and visits to health centres, and of hospital admissions. Method of feeding was note& and details on immunisation obtained from the child's immunisation card. The children received all the x routine childhood immunisations according to the national regimen, based on WHO recommendations. Mothers were asked to bring the child to the follow up clinic for any problem, so that episodes of illness would not be missed. The women were reimbursed for transport costs to encourage follow up visits. Calculation of transmission rate and classification of infection status were made according to the recommendations of the Ghent workshop. Children were regarded as infected if they were antibody positive at 18 months or had an HIV related death. They were classified as uninfectd if the antibody test was negative at 9 months of age. Those infants who were lost to follow up before the age of nine months whilst still antibody positive and those whose cause of death could not be determined, were classified as indeterminate. The diagnosis of AIDS was based on the WHO criteria. Blood samples were taken at birth, at age one and three months, then at three month intervals to 18 months; thereafter at six month intervals. Sera were tested for HIV1 antibodies by a commercial enzyme-linked immunosorbent assay,ELISA. Samples that tested positive were confirmed by two tests, a Roche Elisa and by an immunoflourescent assay (IFA). A sample was regarded as being positive if both the second ELISA as well as the IFA or the Western Blot tested positive. xi Results: between October 1990 and March 1993, 234 infants and their 229 mothers were entered into the study. Those who did not attend a single follow up after birth were excluded from the study. The final cohort comprised 181 infants, of whom 48 were classified as infected ( including 17 deaths); 93 not infected, and 40 as indeterminate ( including 8 deaths). Maternal Data: about 60% of the mothers were under 30 years of age and were multiparous; 18% tested positive for syphilis serology; 22.9% were anaemic during pregnancy, and 37% were delivered by caesarean section. Most women lived in urban areas, and 16% chose to bottle-feed exclusively. Vertical Transmission Rate and Factors affecting this Rate: the median vertical transmission rate was 34%, (95% confidence intervals, CI 26%-42%). This figure is similar to that found in most parts of Africa, but much higher than those for Europe and USA. The maternal factors found to be associated with an increased risk of transmission were vaginal deliveries and a low haemoglobin level during pregnancy. Breastfeeding, Transmission, and Outcome: breastfeeding was found to have an increased risk of transmission, by 15 % (CI 1.8-31.8). On assessing growth and morbidity, it was noted that breastfed infants were not protected against such common childhood infections as pneumonia and diarrhoea, and that failure to thrive occurred with equal frequency in both those breastfed as well as those receiving artificial feeds. Newborn Data: when comparing newborn data between those infants who were subsequently found to be infected with those who were uninfected, it was found that there were no major differences between these groups with regard to growth parameters and neonatal complications. However, those infants with rapidly progressive disease (those who died within 24 months), were noted to have lower mean birth weights and lengths, a higher frequency of low birth weights, and tended to have more neonatal problems. Clinical Manifestations: the first differences between the infected and the uninfected infants generally manifested from about 3 months of age. HIV infected children were identifiable by higher frequencies of thrush, lymphadenopathy, skin rash, and hepatosplenomegaly in the early stages, and later on with a higher tendency to neurological and developmental abnormalities, as well as of diarrhoea. Pneumonia was found with equal frequencies in both the infected and uninfected children. The HIV infected child could be distinguished fairly early in life by the combination of the manifestations described above. Progression to AIDS: AIDS was diagnosed in 44% of all the infected children during the study period. Ninety five percent of these children were identified by 12 months of life, showing a rapid progression of the disease Longitudinal Growth: when longitudinal growth parameters were analysed in this cohort, it was found that HIV infected children were stunted from as early as 3 months of age, and remained below the international standards into early childhood. Infected children were also found to be malnourished (i.e. weight for age below international means), from an early age, and this persisted throughout early childhood. Of note, the uninfected childrens' weights, although comparable to international means initially, dropped after the first year of life. However, both groups did not have significant wasting, when compared to international means. Mortality: there were 25 known deaths during the study period. Of these, 17 were classified as HIV-related, and 8 as indeterminate. The mean age at death was 10.1 months, with 83% of all the HIV-related deaths occurring within the first year of life. The commonest diagnoses at the ti me of death were diarrhoea, pneumonia, and failure to thrive; also, thrush was common, as were neurological abnormalities.