Survival in children with perinatal HIV infection and very low CD4 lymphocyte counts.

OBJECTIVES: To evaluate clinical conditions associated with mortality in HIV-infected children with CD4+ counts <100 cells/microl. METHODS: The Pediatric Spectrum of HIV Disease Project is a longitudinal medical record review study with eight study sites in the United States, which have been enrolling children since 1989. Survival time from baseline very low CD4 count (<100 cells/microl) to death was estimated using the Kaplan-Meier method. Cox proportional hazards models were used to evaluate the effect of clinical variables on mortality. RESULTS: Of 522 children (>/=1 year of age) with serial CD4+ T-lymphocyte measurements, the median age at the first very low CD4 count was 4.8 years. The estimated median survival following the first very low CD4 count was 36 months. The following factors present at the first very low CD4 count were independently associated with a higher risk of death: younger age, weight-for-age >2 standard deviations below the mean, and previously diagnosed AIDS. The subsequent development of cytomegalovirus (CMV)-associated disease, Mycobacterium avium intracellulare (MAI) infection, wasting syndrome, or esophageal candidiasis was also independently associated with a higher risk of death. CONCLUSION: Survival in HIV-infected children with very low CD4 counts before introduction of highly active antiretroviral therapy was highly variable. Poor nutritional status and the development of CMV disease or MAI infection were associated with the shortest survival times.

Identifying human immunodeficiency virus infection at birth: application of polymerase chain reaction to Guthrie cards.

Guthrie cards containing dried blood spots from 67 children now known to be infected with human immunodeficiency virus (HIV) and 63 children now classified as having had seroreversion were retrieved from the newborn infant archives from 1986 through 1991 to determine whether the polymerase chain reaction (PCR) could predict the infection at birth. The PCR assays operating at a sensitivity capable of detecting 2 to 10 copies of HIV proviral DNA per microgram were able to detect HIV proviral DNA in 52% (35/67) of the infected neonatal blood specimens. Longer storage times did not decrease PCR positivity rates, an advantage over assays for HIV antibody. Children whose clinical progression has been aggressive had high rates of PCR positivity in neonatal specimens, 50% (7/14) in those with low CD4 cell counts during the first year of life, 71% (10/14) in those with Pneumocystis pneumonia or disseminated cytomegalovirus infection by age 1 year, 62% (18/29) in those with onset of acquired immunodeficiency syndrome by 18 months, and 66% (14/21) in those who died of the disease by 36 months of age. No evidence of HIV proviral DNA was found in any of the 63 specimens from children with seroreversion. We conclude that PCR, using routinely available dried blood spots from neonates, has applications in early diagnosis and in epidemiologic projections going beyond current seroprevalence studies.