Antiretroviral Therapy for Children: Title and subTitle BreakSubstantial Benefit But Limited Access

According to the United Nations Programme on HIV/AIDS (UNAIDS) global report on human immunodeficiency virus (HIV), in 2005 more than 2.3 million children were living with HIV infection, and 380 000 children with HIV had died.¹ Most of the children who die from HIV have severe manifestations of common childhood illnesses, such as diarrhea, malnutrition, acute respiratory tract infection, and tuberculosis.^{2 - 4} These children often die before the diagnosis of HIV infection can be made, and mortality is often attributed to a background high rate of early childhood mortality.

The United States is home for a diminishing proportion of HIV-infected children, with an estimated 11 000 cases in 2005.¹ There are 2 reasons for this decreasing population of HIV-infected children in the developed world. First, pregnant women are increasingly being offered HIV testing, and HIV-infected women are receiving antiretroviral medications sufficient to reduce transmission to a rate of approximately 1% to 2%.⁵ Second, the population of HIV-infected children is diminishing because those children who had been infected are now growing up and becoming adults, following the introduction of potent new drugs in 1996. According to a longitudinal study by the Pediatric AIDS Clinical Trials Group (PACTG 219), the mortality rate for HIV-infected children and adolescents decreased from more than 5% in 1996 to less than 1% in 1998.⁶ Similar survival benefits have been observed among adults.⁷

The impact of highly active antiretroviral therapy (HAART) on survival and quality of life among patients affected by HIV is hardly disputable, but concise objective evidence of the benefit of treatment on infectious complications in children has been lacking in the medical literature. Before the availability of HAART, opportunistic infections such as Pneumocystis jiroveci pneumonia, bacteremia, bacterial pneumonia, and fungal sepsis were significant fatal diseases in patients infected with HIV and remain so among patients not receiving therapy. In this issue of JAMA, Gona and colleagues⁸ provide objective evidence to substantiate the benefit of HAART or, more correctly, the association of HAART and opportunistic infection prophylaxis with incidence of opportunistic infections before and after the advent of HAART.

In a retrospective cohort study, Gona et al compared the incidence of 29 opportunistic infections in HIV-infected children who were treated in the HAART era with the incidence among those who were treated before the availability of HAART. They found that the incidence of P jiroveci pneumonia had decreased by 14-fold, disseminated Mycobacterium avium complex by 12-fold, bacteremia by 10-fold, and bacterial pneumonia by 5-fold. These reductions in opportunistic infections occurred despite a concomitant reduction in the use of specific prophylaxis medications, suggesting that immune reconstitution rather than improved prophylaxis was primarily responsible for better outcomes.

The size and diversity of the population studied allows for generalizability of the results to the US population of children with HIV infection. There were approximately 3000 children in the cohort, which constitutes about 22% of children and adolescents living with HIV in the United States. Given the diminishing population of HIV-infected children, it is unlikely that additional data will become available in this country.

There were limitations with the retrospective cohort study design, which evaluated available data from a database. Some events and clinical practices changed over time and could have affected case ascertainment and overall disease incidence. Although the incidence of bacterial pneumonia was reported, the diagnosis of this problem in children is difficult, because no consistent clinical or radiographic features reliably distinguish different etiologies of pneumonia. Radiographic pneumonia may or may not be associated with bacteremia. When such an association occurs, the relationship is not always a causal one. Specific etiologic diagnosis of bacterial pneumonia can be determined by diagnostic techniques, such as lung aspirates or bronchoalveolar lavage. These procedures are highly sensitive but are seldom used because they are considered too invasive.

An important change in clinical practice over time that has affected disease incidence is the introduction of protein conjugate vaccines in children for the prevention of invasive Hemophilus influenzae type b and Streptococcus pneumoniae infections, both of which have very high incidence in HIV-infected children. Evaluation of the incidence of these infections in HIV-uninfected age-matched children during the study period to control for changes in diagnostic criteria and treatment would have been informative as would a demonstration of a linear trend in specific opportunistic infections, but this was not possible because the number of cases per disease category was limited.

Although significant advances have been made in the improvement of the quality of life for patients with HIV/AIDS, several important challenges remain. A cure for HIV infection remains elusive and following infection, chronic suppression of viral replication with preservation of immune function remains the goal. If in the best case scenario, a combination of HAART and specific opportunistic infection prophylaxis continues to prolong survival, patients must contend with the adverse effects of long-term treatment with these agents, most of which are new and have unknown long-term effects, particularly when administered to young children.

Clearly, antiretroviral medications work. Patients receiving appropriate treatment have decreases in mortality and in the incidence of opportunistic infections. Adults receiving these medications can return to work to support their families, and children receiving these medications grow up to be parents themselves. The effect of such extraordinary therapy has no expiration date and can continue unabated for years. As the study by Gona et al⁸ has shown, antiretroviral therapy can reduce infectious complications even for common childhood illnesses, such as bacterial pneumonia and bacteremia. The effects of this intervention can be expected to be most pronounced among very young children, for whom these illnesses are common causes of death. To have the greatest benefit on HIV-associated mortality among children, the way forward must be to implement strategies for the identification and treatment of all HIV-infected infants through comprehensive programs of antenatal testing and close follow-up. A family-centered model for care that begins with a healthy pregnancy recognizes the additional survival advantage a child has with a healthy parent.

Until now, concerns about providing antiretroviral medications in settings in which the disease is highly prevalent have focused on the costs associated with these programs. Another way to reframe the issue is to consider the costs associated with not providing effective care. In a study⁹ from Malawi, 89% of children not receiving treatment for HIV disease were dead by age 3 years. The real economic and moral costs associated with the preventable deaths of large numbers of children and young adults must be considered in the equation. In the past 5 years, the debate has begun to shift from whether these treatments can be provided in developing countries to how these treatments can be provided. Through programs such as the Global Fund for AIDS, TB, and Malaria; the President's Emergency Plan for AIDS Relief; and the Clinton Foundation HIV/AIDS Initiative, the issues of “how” to provide treatment are gradually being addressed, but these efforts need to be increased substantially. For 2.3 million children living with HIV infection worldwide, the question is not whether or how but when they will receive (and, like their counterparts in the study by Gona et al, benefit from) the therapy that will allow them to reach adulthood.