A review of: Kovacs A, Schluchter M, Easley K, et al. 1999 Cytomegalovirus infection and HIV-1 disease progression in infants born to HIV-1-infected women. N Engl J Med 341:77–84.

SEVERAL STUDIES SUGGEST that cytomegalovirus (CMV) as a cofactor is involved in the pathogenesis of the AIDS (AIDS) (1–3). A direct causal relation, however, has been difficult to clearly document in adults, because adult patients usually are infected with CMV before contracting human immunodeficiency virus type 1 (HIV-1) infection. This cohort-based, prospective study was designed to examine the association of CMV infection with the progression of disease in infants who were born to HIV-1 infected women, and whose CMV status was known (4).

HIV-1 infection was diagnosed if a child had two positive HIV-1 antibody cultures, had a positive HIV-1 antibody test at 15 mo of age, died of an HIV-1-associated condition, or had AIDS. The authors did not make use of PCR testing to establish the diagnosis of HIV-1 infection or AIDS. CMV infection was diagnosed if the child had a positive culture of saliva or urine or a positive PCR of urine at any age, or had a positive serologic test for IgG or IgM at 12 or more months of age.

Researchers assessed 440 infants, born to HIV-infected women. Seventy-five of these infants were HIV-1 infected and 365 were not. At birth, the frequency of CMV infection in the HIV-1-infected and HIV-1-uninfected infants was 4.3% and 4.5%, respectively, which is higher than the rates of 0.2–2.2 in the general newborn population. However, at 6 mo of age, CMV infection was diagnosed in 39.9% of HIV-1-infected infants and only in 15.3% of uninfected patients. The cumulative rates of CMV infection over a period of 48 mo remain significantly higher among HIV-1-infected children and the rate of CMV transmission from mothers to offspring was especially high during the first 12 mo. These data clearly show that perinatal and postnatal CMV infections are more common among HIV-1-infected children. One explanation for the increase in vertical infection rate might be that the level of CMV shedding was higher in women who transmitted HIV-1 to their infants at birth. Disease progression over years in the mothers could result in shedding more CMV in the urine or saliva. However, the authors did not perform cervical or saliva CMV cultures to probe this possibility. Alternatively, HIV-1-infected children could be more susceptible to horizontal CMV infection from their environment.

This study not only proves that HIV-1-infected children have higher rates of CMV infection acquired during the first 4 y of life, but the data support earlier clinical and epidemiological findings that CMV infection is associated with an increased risk of HIV-1 disease progression. At 18 mo of age, the infants with both CMV and HIV-1 infections had significantly higher rates of HIV-1 disease progression and CNS disease than those infected with HIV-1 alone. However, the mechanism by which CMV co-infection leads to more rapid progression remains unclear. It is possible or even likely that CMV and HIV-1, two immunosuppressive viruses, may act synergistically to accelerate disease progression.

The main conclusion that can be drawn from this study is that strategies to prevent vertical and horizontal CMV infection in HIV-1-infected infants and children should be applied to decrease and prolong disease progression and CNS disease. Such preventive measures may include the use of anti-CMV drugs applicable in early infancy, passive prophylaxis by using hyperimmune CMV antibodies or active immunization by vaccination.