Unsuspected HIV infection presenting in first year of life

BMJ 1995; 310 doi: (Published 13 May 1995) Cite this as: BMJ 1995;310:1235
  1. J A Evans, lecturera,
  2. S C Marriage, lecturera,
  3. M D S Walters, senior lecturera,
  4. M Levin, professora
  1. a Department of Paediatrics, St Mary's Hospital Medical School, London W2 1PG
  1. Correspondence to: Dr Evans.
  • Accepted 12 January 1995

Anonymous unlinked screening programmes in inner London have shown that up to 1 in 200 women booked at antenatal clinics is HIV positive. Most of these women, however, do not have a named HIV test during pregnancy and therefore remain unaware of their infection. Failure to identify these women not only results in the baby being denied adequate medical care but also prevents the use of measures that may reduce transmission of infection from mother to child.

Less than one fifth of pregnant women infected with HIV are being recognised. During 1992-3 anonymous testing of blood samples from newborn babies in three Thames regions identified 262 HIV seropositive infants, only 15% of whom were born to women known to be infected with HIV before delivery.1

Subjects, methods, and results

During 1992-3 five infants who were previously not known to be at risk of HIV infection were referred to our paediatric HIV unit with Pneumocystis carinii pneumonia (table). The pneumonia was the first sign of HIV infection in the family. Four of the children required ventilation, and one died of the pneumonia. Four children had had non-specific symptoms, including cough, diarrhoea, and poor weight gain, for up to six weeks before diagnosis.

Clinical features of five infants presenting with P carinii pneumonia leading to diagnosis of HIV infection

View this table:


All five patients presented with a life threatening complication of HIV infection as the first sign of the disease in the family. The families had to cope with the diagnosis of HIV in their child and other family members at a time when the child was critically ill.

A child born to a women known to be HIV positive is closely followed, and it is likely that the non-specific symptoms shown by these children would have received early appropriate medical intervention. In addition, our unit's policy is to give co-trimoxazole to all infants born to HIV positive women from the age of 3 weeks until their infection status is known. Co-trimoxazole has been found highly effective in preventing P carinii pneumonia in children with leukaemia and is also valuable in infants with HIV infection.2 Thus the illness may have been prevented if the mothers' HIV status had been known.

Several interventions have been shown to greatly reduce the risk of transmission of HIV from mother to child, which is one of the main aims in the care of families affected by HIV. In the developed world babies born to HIV positive women should be bottle fed as breast feeding significantly increases the risk of transmission.3 A recent study in the United States showed that giving zidovudine to HIV positive pregnant women and then to the baby for six weeks after delivery significantly reduces the transmission rate of HIV,4 and evidence is accumulating that caesarean section may also be protective.5

The reduced risk of vertical transmission, better health care of the baby and mother, and families being able to discover the HIV infection before illness develops are compelling arguments for identifying HIV infected women antenatally. The absence of an effective screening programme in Britain is denying HIV positive families simple interventions that would prevent some infants being infected and reduce the risk of life threatening complications in those that do become infected. We believe that there is an urgent need to revise the current HIV testing policy in pregnant women in areas where anonymous testing shows a high prevalence.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract