Vertical transmission remains the most common way that pediatric patients become infected with HIV. It may occur during gestation, during the intrapartum period, or in the immediate postpartum period. In the absence of any steps to prevent transmission, 20–40% of babies born to HIV-positive mothers will become infected.^1,2,3 It is thought that most babies who become infected do so in the peripartum period. Earlier infection does occur, as HIV has been found by in situ PCR (polymerase chain reaction) in the tissues of aborted fetuses during the second trimester.⁴ Transmission to babies after they have been born usually occurs through breastfeeding. This remains a problem in the developing world, where breast milk may be the only form of nutrition available, but it has been virtually eliminated in the United States, where clean water for formula is readily available.

The single most important factor in determining the risk of transmission is the viral load in maternal plasma. In one study of 552 maternal-infant pairs, none of 57 babies whose mother’s viral load was less than 1,000 copies/mL became infected, but 26 (41%) of 64 babies whose maternal viral loads were greater than 100,000 copies/mL became infected.² The plasma viral load has been shown to correlate with the amount of HIV in both cervical mucus and in vaginal secretions,⁵ which is a plausible explanation for the phenomenon. Transmission of HIV is also augmented by concomitant maternal syphilis.⁶

In the early 1990s, Pediatric AIDS Clinical Trials Group (PACTG) Protocol 076 demonstrated the effective prevention of maternal-infant transmission. This double-blind, prospective, placebo-controlled trial showed that the administration of zidovudine (ZDV or sometimes called azidothymidine [AZT]) to pregnant mothers during the last trimester of pregnancy, coupled with intravenous AZT in the immediate peripartum period and six weeks of oral AZT therapy for their neonates decreased maternal-infant transmission from 28% to 8%.³ Maternal-infant transmission decreased greatly after this standard was adopted. Of course, the success of this form of prophylaxis depends upon knowing which mothers are HIV infected. This, in turn, relies upon prenatal counseling and universal HIV testing of pregnant women. When testing is offered to pregnant women, the majority accepts it. In one exemplary early program, counseling and testing increased from 21% to 95% of pregnancies, and the perinatal HIV transmission rate in North Carolina was documented to decrease from 31% to 3.1% statewide.⁷ Testing pregnant women for HIV infection should be routine, similar to testing for syphilis or rubella immunity.

At any plasma viral load, adherence to the 076 AZT protocol decreases transmission rates. Factors that decrease the chances that a mother-infant pair will receive all three components of the protocol include older maternal age, CD4 counts greater than 500/μL, and cocaine or heroin use.⁸ Because following all three parts of the 076 protocol is cumbersome and expensive, it has been difficult to implement in developing countries. Trials in which the maternal portion, the neonatal portion, or both
were truncated suggest that a longer maternal component is more important than a longer neonatal component.⁹ In these trials, the intrapartum intravenous AZT was replaced with frequent oral AZT. In another attempt at simplification, mothers and their infants were each given a single dose of nevirapine.¹⁰ This regimen decreased transmission significantly only in mothers not already receiving standard antiretroviral therapy.¹¹ In one open-label trial, the combination of AZT and lamivudine (3TC) was also shown to be effective in preventing transmission. However, a common 3TC resistance mutation was demonstrated in 52 (39%) of 132 neonates, and adverse events, including severe anemia and thrombocytopenia, were common. Two children who were subsequently proven to be uninfected with HIV died before the age of 12 months from mitochondrial toxicities thought to be related to the combination therapy.¹²

Ultimately, antiretroviral therapies that best treat the mother’s disease probably result in the best possible outcomes for both mother and baby.¹³ The standard of care for pregnant women with HIV infection now includes use of multiple-drug therapy, at least one of which should be AZT. Fears that there would be long-term adverse outcomes in HIV-uninfected babies whose mothers received AZT during pregnancy have been largely assuaged. One study showed a slightly increased risk of first febrile seizure in these children,¹⁴ but another study that followed, for a mean of 4.2 years, HIV-uninfected children whose mothers were enrolled in protocol 076 showed no significant difference in growth, cognition, or development in those whose mothers received AZT compared with those whose mothers received placebo.¹⁵

Other factors shown to be independently associated with increased risk of maternal-infant transmission include maternal obesity and procedures that cause the baby to contact maternal blood or cervical secretions.¹⁶ In light of this finding, it is not surprising that elective cesarean delivery prior to rupture of membranes decreases transmission by about half.¹⁷ However, the efficacy of cesarean delivery in the era of highly active antiretroviral therapy (HAART) has not been evaluated. Enthusiasm for elective cesarean birth has been waning, due, in part, to the fact that transmission rates are already less than 2% in many populations, and that cesarean delivery is not without risk, especially because wound healing may be poor in patients with advanced HIV disease.

Postnatal transmission via breast milk occurs in about 15% of babies who are exclusively breastfed.¹⁸ Thus, HIV-positive mothers should not breastfeed their babies unless there is no possible alternative source of nutrition.

Horizontal transmission of HIV is usually from either sexual contact or through intravenous drug use. In recent years, there has been an increase in the number of teenagers being infected, usually through unprotected sex. It has been estimated that at least half of all new HIV infections in the United States occur among people below the age of 25, and that the majority of these young people are infected through sexual activity.¹⁹ The message of so-called safe sex (more appropriately termed “safer sex”) has been preached to a large number of people at risk. Unfortunately, adherence to this idea has been declining as improved therapies for HIV have become widely used.²⁰ Some of this laxity is due to the perception that persons on HAART are unlikely to transmit the infection. Remarkably, surveys reveal that the relaxed attitude toward HIV transmission is largely attributable to the perception that HIV infection is now a manageable disease. In one survey of 80 women, whereas only 5% of respondents believed that safer sex was no longer important, 15% said that they no longer insist on condom use, and 40% agreed that “AIDS is now a less serious threat.”²¹ Many teenagers, in particular, engage in high-risk behavior even though they understand the risk factors for HIV infection. This may be caused by the tendency of teenagers to feel a sense of immortality.

Injection drug use is a risk for acquisition of infection not only through needle sharing, but also because people who use injection drugs engage in riskier sexual behavior.²² Even alcohol use has been identified as a risk factor.²³

The Centers for Disease Control and Prevention (CDC) classification scheme for infants and children with HIV infection has changed several times. The scheme currently in use is quite straightforward. Patients are classified based on symptoms designated by letters—N, for no symptoms; A, for mild symptoms; B, for moderate symptoms; and C, for severe symptoms—and immune suppression designated by numbers: 1, for no suppression; 2, for moderate suppression; and 3, for severe suppression. The type of clinical conditions and degree of immune suppression that qualify a patient for the various designations is shown in Box 20-2.

An introduction to the various classes of antiretroviral agents, their uses, and their side effects and limitations follows. The actual clinical approach to using these agents is discussed briefly thereafter.

The workup of any HIV-infected child who presents with fever begins with an assessment of the child’s immune status. Causes of fever in children with category 1 disease (i.e., CD4 count 25% or greater) do not differ significantly from those seen in non-HIV-infected children. They are likely to have common community-acquired infections, as would any child with fever. The exception would be that HIV-infected children have a much higher incidence of pneumococcal infection than their uninfected peers (see text that follows). Children in category 2 (i.e., CD4 cell count 15–24%) are still more likely to acquire common diseases than opportunistic infections. Patients with low absolute lymphocyte counts, those in the lower end of CDC class 2, may have relatively compromised immunity, which might predispose to a broader array of infections. In all children with HIV and fever, if an obvious local source cannot be found, or if the source is pneumonia, a set of blood cultures should be obtained prior to administration of any antibiotic agent. Selected antibiotics should have good activity against Streptococcus pneumoniae, the most common pathogen. As with infections in other immunocompromised patients, the intravenous route may be necessary initially and duration of therapy may need to be longer than in the healthy host.

For patients with category 3 immune suppression (CD4 less than 15%), infection with a wide variety of pathogens is possible (including concomitant infection with more than one pathogen). As immunity is severely suppressed, signs and symptoms of disease may be mild. Evaluation needs to be thorough, and, when febrile, these patients often require admission to the hospital. It is prudent to involve an infectious-diseases trained pediatrician from the outset. Workup is more extensive than expected by the severity of symptoms; for example, intermittent or persistent headache with low-grade fever in this type of patient might be an indication for a computed tomography (CT) scan and a lumbar puncture. Blood cultures, including special cultures for Mycobacterium avium complex (see following text), should be obtained. If any localizing signs or symptoms are found, investigation can begin by focusing on those areas. Many of the problems children with HIV and AIDS may encounter are discussed in the remainder of this chapter.