Etiology

HIV-1 and HIV-2 are members of the Retroviridae family and belong to the Lentivirus genus, which includes cytopathic viruses causing diverse diseases in several animal species. The HIV-1 genome contains 2 copies of single-stranded RNA that is 9.2 kb in size. At both ends of the genome there are identical regions, called long terminal repeats, which contain the regulation and expression genes of HIV. The remainder of the genome includes 3 major sections: the GAG region, which encodes the viral core proteins (p24, p17, p9, and p6, which are derived from the precursor p55); the POL region, which encodes the viral enzymes (i.e., reverse transcriptase [p51], protease [p10], and integrase [p32]); and the ENV region, which encodes the viral envelope proteins (gp120 and gp41, which are derived from the precursor gp160). Other regulatory proteins, such as tat (p14), rev (p19), nef (p27), vpr (p15), vif (p23), vpu in HIV-1 (P16), and vpx in HIV-2 (P15) are involved in transactivation, viral messenger RNA expression, viral replication, induction of cell cycle arrest, promotion of nuclear import of viral reverse transcription complexes, downregulation of CD4 receptors and class I major histocompatibility complex, proviral DNA synthesis, and virus release and infectivity (Fig. 268-1).

Figure 268-1 The human immunodeficiency virus and associated proteins and their functions.

The major external viral protein of HIV-1 is a heavily glycosylated gp120 protein that is associated with the transmembrane glycoprotein gp41; gp41 is very immunogenic and is used to detect HIV-1 antibodies in diagnostic assays; gp120 is a complex molecule that includes the highly variable V3 loop. This region is immunodominant for neutralizing antibodies. The heterogeneity of gp120 presents major obstacles in establishing an effective HIV vaccine. The gp120 glycoprotein also carries the binding site for the CD4 molecule, the most common host cell surface receptor of T lymphocytes. This tropism for CD4⁺ T cells is beneficial to the virus because of the resulting reduction in the effectiveness of the host immune system. Other CD4-bearing cells include macrophages and microglial cells. The observations that CD4⁻ cells are also infected by HIV and that some CD4⁺ T cells are resistant to such infections suggests that other cellular attachment sites are needed for the interaction between HIV and human cells. Several chemokines serve as co-receptors for the envelope glycoproteins, permitting membrane fusion and entry into the cell. Most HIV strains have a specific tropism for 1 of the chemokines, including the fusion-inducing molecule CXCR-4, which has been shown to act as a co-receptor for HIV attachment to lymphocytes, and CCR-5, a β chemokine receptor that facilitates HIV entry into macrophages. Several other chemokine receptors (CCR-3) have also been shown in vitro to serve as virus co-receptors. Other mechanisms of attachment of HIV to cells use non-neutralizing antiviral antibodies and complement receptors. The Fab portion of these antibodies attaches to the virus surface, and the Fc portion binds to cells that express Fc receptors (macrophages, fibroblasts), thus facilitating virus transfer into the cell. Other cell surface receptors, such as mannose-binding protein on macrophages or DC-specific C-type lectin (DC-SIGN) on dendritic cells, also bind to the HIV-1 envelope glycoprotein and increase the efficiency of viral infectivity. Cell-to-cell transfer of HIV without formation of fully formed particles is a more rapid mechanism of spreading the infection to new cells than direct infection by the virus.

Following viral attachment, gp120 and the CD4 molecule undergo conformational changes, and gp41 interacts with the fusion receptor on the cell surface. Viral fusion with the cell membrane allows entry of viral RNA into the cell cytoplasm. This process involves accessory viral proteins (nef, vif) and binding of cyclophilin A (a host cellular protein) to p24. Viral DNA copies are then transcribed from the virion RNA through viral reverse transcriptase enzyme activity, and duplication of the DNA copies produces double-stranded circular DNA. The HIV-1 reverse transcriptase is error prone and lacks error-correcting mechanisms. Thus, many mutations arise, creating wide genetic variation in HIV-1 isolates even within an individual patient. The circular DNA is transported into the cell nucleus, where it is integrated into chromosomal DNA and referred to as the provirus. The provirus has the advantage of latency, as it can remain dormant for extended periods. Integration usually occurs near active genes, which allow a high level of viral production in response to various external factors such as an increase in inflammatory cytokines (by infection with other pathogens) and cellular activation. Depending on the relative expression of the viral regulatory genes (tat, rev, nef), the proviral DNA may encode production of the viral RNA genome, which in turn leads to production of viral proteins necessary for viral assembly.

HIV-1 transcription is followed by translation. A capsid polyprotein is cleaved to produce, among others, the virus-specific protease (p10). This enzyme is critical for HIV-1 assembly. Several HIV-1 antiprotease drugs have been developed, targeting the increased sensitivity of the viral protease, which differs from the cellular proteases. The RNA genome is then incorporated into the newly formed viral capsid that requires zinc finger domains (p7) and the matrix protein (p17). As new virus is formed, it buds through specialized membrane areas, known as lipid rafts, and is released.

Full length sequencing of the HIV-1 genome demonstrated 3 different groups (M [main], O [outlier], N [non-M, non-O]) probably occurring from multiple zoonotic infections from primates in different geographic regions. The same technique identified 8 groups with HIV-2 isolates. Group M diversified to 9 subtypes (or clades A to D, F to H, J and K) In each region of the world, certain clades predominate, for example, clade A in Central Africa, clade B in the USA and South America, clade C in South Africa, clade E in Thailand, and clade F in Brazil. While some subtypes have been identified for Group O, none was found with any of the HIV-2 groups. Clades are mixed in some patients due to HIV recombination, and some crossing between groups (i.e., M and O) has been reported.

HIV-2 has a similar life cycle to HIV-1 and is known to cause infection in several monkey species. Subtypes A and B are the major causes of infection in humans but rarely cause infection in children. HIV-2 differs from HIV-1 in its accessory genes (for example, it has no vpu gene but contains the vpx gene, which is not found in HIV-1). It is most prevalent in Western Africa, but increasing numbers of cases are reported from Europe and Southern Asia. The diagnosis of HIV-2 infection is more difficult because of major differences in the genetic sequences between HIV-1 and HIV-2. Thus, several of the standard confirmatory assays (immunoblot), which are HIV-1 specific, may give indeterminate results with HIV-2 infection. If HIV-2 infection is suspected, a combination screening test that detects antibody to HIV-1 and HIV-2 peptides should be used. In addition, the rapid HIV detection tests have been less reliable in patients suspected to be dually infected with HIV-1 and HIV-2, because of lower antibody concentrations against HIV-2.

Epidemiology

The World Health Organization (WHO) estimated that in 2009, 2.5 million children worldwide were living with HIV-1 infection, 90% of who were from Sub-Saharan Africa. While between 2004 and 2009 the global number of children born with HIV decreased by 24% and deaths from AIDS-related illnesses among children <15 yr of age declined by 19%, still 370,000 children (<15 yr) were newly infected with HIV in 2009 alone. These trends reflect the slow but steady expansion of services to prevent transmission of HIV to infants and an increase in access to treatment for children. Worldwide, 50% of HIV-infected individuals are women, most of who have become infected through heterosexual contact in their child-bearing years. Through 2009, an estimated 16.6 million children have been orphaned by AIDS, which is defined as having one or both parents die from AIDS.

There have been 9,600 American children <13 yr of age diagnosed with AIDS from the beginning of the epidemic through 2007. The number of U.S. children with AIDS diagnosed each year increased from 1984 to 1992 but then declined by more than 95% to <100 cases annually by 2003, largely due to the success of prenatal screening and perinatal antiretroviral treatment of HIV-infected mothers and infants. There are now ~8,500 children and adolescents living with HIV or AIDS in the USA. Virtually all HIV infections in children <13 yr of age in the USA are the result of vertical transmission from an HIV-infected mother. Children of racial and ethnic minority groups are disproportionately over-represented, particularly non-Hispanic African-Americans and Hispanics. Race and ethnicity is not a risk factor for HIV infection but more likely reflects other factors that may be predictive of increased risk for HIV infection, such as lack of educational and economic opportunities and higher rates of intravenous drug use. New York, Florida, and California account for most cases of HIV in children in the USA.

Although adolescents (13-24 yr of age) represent a minority of U.S. AIDS cases (approximately 5%), they constitute a growing population of newly infected individuals, with 15% of all new cases of HIV diagnosed in 2006 occurring between the ages of 13 and 24 yr. The highest incidence of new adolescent infections has been among African-American males who have sex with males, and more than 50% report being unaware of their diagnosis. Considering the long latency period between the time of infection and the development of clinical symptoms, reliance on AIDS case definition surveillance data significantly under-represents the impact of the disease in adolescents. Based on a median incubation period of 8-12 yr, it has been estimated that 15-20% of all AIDS cases were acquired between 13 and 19 yr of age.

Risk factors for HIV infection vary by gender in adolescents. More than 87% of males between the ages of 13 and 19 yr with HIV/AIDS acquire infection through sex with males (MSM). In contrast, 88% of adolescent females with AIDS are infected through heterosexual contact. As in the pediatric population, adolescent racial and ethnic minority populations are over-represented, especially among females. A greater proportion of female adolescents have AIDS (male : female ratio 1.5 : 1) than do female adults >25 yr of age (male : female ratio 2.9 : 1).

Pathogenesis

HIV infection affects most of the immune system and disrupts its homeostasis (Fig. 268-2). In most cases, the initial infection is caused by low amounts of a single virus. Therefore, disease may be prevented by prophylactic drug(s) or vaccine. When the mucosa serves as the portal of entry for HIV, the 1st cells to be affected are the dendritic cells. These cells collect and process antigens introduced from the periphery and transport them to the lymphoid tissue. HIV does not infect the dendritic cell but binds to its DC-SIGN surface molecule, allowing the virus to survive until it reaches the lymphatic tissue. In the lymphatic tissue (e.g., lamina propria, lymph nodes), the virus selectively binds to cells expressing CD4 molecules on their surface, primarily helper T lymphocytes (CD4⁺ T cells) and cells of the monocyte-macrophage lineage. Other cells bearing CD4, such as microglia, astrocytes, oligodendroglia, and placental tissue containing villous Hofbauer cells, may also be infected by HIV. Additional factors (co-receptors) are necessary for HIV fusion and entry into cells. These factors include the chemokines CXCR4 (fusion) and CCR5. Other chemokines (CCR1, CCR3) may be necessary for the fusion of certain HIV strains. Several host genetic determinants affect the susceptibility to HIV infection, the progression of disease, and the response to treatment. These genetic variants vary in different populations. A deletion in the CCR5 gene that is protective against HIV infection (CCR5Δ32) is relatively common in whites but is rare in blacks. Several other genes that regulate chemokine receptors, ligands, histocompatibility complex, and cytokines have also been found to influence the outcome of HIV infection. Usually, CD4⁺ lymphocytes migrate to the lymphatic tissue in response to viral antigens and then become activated and proliferate, making them highly susceptible to HIV infection. This antigen-driven migration and accumulation of CD4 cells within the lymphoid tissue may contribute to the generalized lymphadenopathy characteristic of the acute retroviral syndrome in adults and adolescents. HIV preferentially infects the very cells that respond to it (HIV-specific memory CD4 cells), accounting for the progressive loss of these cells and the subsequent loss of control of HIV replication. The continued destruction of memory CD4+ cells in the gastrointestinal tract leads to reduced integrity of the gastrointestinal epithelium followed by leakage of bacterial particles into the blood and increased inflammatory response, which cause further CD4+ cell loss. When HIV replication reaches a threshold (usually within 3-6 wk from the time of infection), a burst of plasma viremia occurs. This intense viremia causes flu or mononucleosis-like symptoms (fever, rash, lymphadenopathy, arthralgia) in 50-70% of infected adults. With establishment of a cellular and humoral immune response within 2-4 mo, the viral load in the blood declines substantially, and patients enter a phase characterized by a lack of symptoms and a return of CD4 cells to only moderately decreased levels.

Figure 268-2 HIV life cycle and antiretroviral drug targets. Present antiretroviral drugs span 6 classes that target 5 unique steps in the HIV life cycle (binding, fusion, reverse transcription, integration, and proteolytic cleavage). The most common drugs used in resource-rich regions to target each step are shown. Extracellular virions enter their target cell through a complex 3-step process, which is (1) attachment to the CD4 receptor, (2) binding to the CCR5 or CXCR4 co-receptors, or both, and (3) membrane fusion. Maraviroc blocks CCR5 binding and enfuvirtide blocks fusion. The HIV reverse transcriptase enzyme catalyses transcription of HIV RNA into double-stranded HIV DNA, a step inhibited by nucleoside analogues and non-nucleoside reverse transcriptase inhibitors (NNRTIs). The HIV integrase enzyme facilitates incorporation of HIV DNA into host chromosomes and this step is inhibited by raltegravir and other integrase inhibitors. After transcription and translation of the HIV genome, immature virions are produced and bud from the cell surface. The HIV protease enzyme cleaves polypeptide chains, allowing the virus to mature. This last step is inhibited by HIV protease inhibitors.

(From Volberding PA, Deeks SG: Antiretroviral therapy and management of HIV infection, Lancet 376:49–60, 2010, Fig 1, p 50.)

Early HIV-1 replication in children has no apparent clinical manifestations. Whether tested by virus isolation or by PCR for viral nucleic acid sequences, fewer than 40% of HIV-1–infected infants demonstrate evidence of the virus at birth. The virus load increases by 1-4 mo, and almost all HIV-infected infants have detectable HIV-1 in peripheral blood by 4 mo of age.

In adults, the long period of clinical latency (up to 8-12 yr) is not indicative of viral latency. In fact, there is a very high turnover of virus and CD4 lymphocytes (more than a billion cells per day), gradually causing deterioration of the immune system, marked by depletion of CD4 cells. Several mechanisms for the depletion of CD4 cells in adults and children have been suggested, including HIV-mediated single cell killing, formation of multinucleated giant cells of infected and uninfected CD4 cells (syncytia formation), virus-specific immune responses (natural killer cells, antibody-dependent cellular cytotoxicity), superantigen-mediated activation of T cells (rendering them more susceptible to infection with HIV), autoimmunity, and programmed cell death (apoptosis). The viral burden is greater in the lymphoid organs than in the peripheral blood during the asymptomatic period. As HIV virions and their immune complexes migrate through the lymph nodes, they are trapped in the network of dendritic follicular cells. Because the ability of HIV to replicate in T cells depends on the state of activation of the cells, the immune activation that takes place within the microenvironment of the lymph nodes in HIV disease serves to promote infection of new CD4 cells as well as subsequent viral replication within these cells. Monocytes can be productively infected by HIV yet resist killing, explaining their role as reservoirs of HIV and as effectors of tissue damage in organs such as the brain.

Cell-mediated and humoral responses occur early in the infection. CD8 T cells play an important role in containing the infection. These cells produce various ligands (MIP-1α, MIP-1β, RANTES), which suppress HIV replication by blocking the binding of the virus to the co-receptors (CCR5). HIV-specific cytotoxic T lymphocytes (CTLs) develop against both the structural (ENV, POL, GAG) and regulatory (tat) viral proteins. The CTLs appear at the end of the acute infection, as viral replication is controlled by killing HIV-infected cells before new viruses are produced and by secreting potent antiviral factors that compete with the virus for its receptors (CCR5). Neutralizing antibodies appear later in the infection and seem to help in the continued suppression of viral replication during clinical latency. There are at least 2 possible mechanisms that control the steady-state viral load level during the chronic clinical latency. One mechanism may be the limited availability of activated CD4 cells, which prevent further increase in viral load. The other mechanism is development of an active immune response, which is influenced by the amount of viral antigen and limits viral replication at a steady state. There is no general consensus about which of these 2 mechanisms is more important. The CD4 cell limitation mechanism accounts for the effect of antiretroviral therapy, whereas the immune response mechanism emphasizes the importance of immune modulation treatment (cytokines, vaccines) to increase the efficiency of immune-mediated control. A group of cytokines that includes tumor necrosis factor-α (TNF-α), TNF-β, interleukin 1 (IL-1), IL-2, IL-3, IL-6, IL-8, IL-12, IL-15, granulocyte-macrophage colony-stimulating factor, and macrophage colony-stimulating factor plays an integral role in upregulating HIV expression from a state of quiescent infection to active viral replication. Other cytokines such as interferon-γ (IFN-γ), IFN-β, and IL-13 exert a suppressive effect on HIV replication. Certain cytokines (IL-4, IL-10, IFN-γ, TGF-β) reduce or enhance viral replication depending on the infected cell type. The interactions among these cytokines influence the concentration of viral particles in the tissues. Plasma concentrations of cytokines need not be elevated for them to exert their effect, because they are produced and act locally in the tissues. Thus, even during states of apparent immunologic quiescence, the complex interaction of cytokines sustains a constant level of viral expression, particularly in the lymph nodes.

Commonly HIV isolated during the clinical latency period grows slowly in culture and produces low titers of reverse transcriptase. These isolates use CCR5 as their co-receptor. By the late stages of clinical latency, the isolated virus is phenotypically different. It grows rapidly and to high titers in culture and uses CXCR4 as its co-receptor. The switch from CCR5-receptor to CXCR4 receptor increases the capacity of the virus to replicate, to infect a broader range of target cells (CXCR4 is more widely expressed on resting and activated immune cells), and to kill T cells more rapidly and efficiently. As a result, the clinical latency phase is over and progression toward AIDS is noted. The progression of disease is related temporally to the gradual disruption of lymph node architecture and degeneration of the follicular dendritic cell network with loss of its ability to trap HIV particles. The virus is freed to recirculate, producing high levels of viremia and an increased disappearance of CD4 T cells during the later stages of disease.

Before HAART was available, 3 distinct patterns of disease were described in children. Approximately 15-25% of HIV-infected newborns in developed countries present with a rapid disease course, with onset of AIDS and symptoms during the 1st few months of life and a median survival time of 6-9 mo if untreated. In resource-poor countries, the majority of HIV-infected newborns will have this rapidly progressing disease. It has been suggested that if intrauterine infection coincides with the period of rapid expansion of CD4 cells in the fetus, the virus could effectively infect the majority of the body’s immunocompetent cells. The normal migration of these cells to the marrow, spleen, and thymus would result in efficient systemic delivery of HIV, unchecked by the immature immune system of the fetus. Thus, infection would be established before the normal ontogenic development of the immune system, causing more severe impairment of immunity. Most children in this group have a positive HIV-1 culture and/or detectable virus in the plasma (median level 11,000 copies/mL) in the 1st 48 hr of life. This early evidence of viral presence suggests that the newborn was infected in utero. The viral load rapidly increases, peaking by 2-3 mo of age (median 750,000 copies/mL) and staying high for at least the 1st 2 yr of life.

The majority of perinatally infected newborns (60-80%) in developed countries present with a much slower progression of disease, with a median survival time of 6 yr representing the 2nd pattern of disease. Many patients in this group have a negative viral culture or PCR in the 1st wk of life and are therefore considered to be infected intrapartum. In a typical patient, the viral load rapidly increases, peaking by 2-3 mo of age (median 100,000 copies/mL) and then slowly declines over a period of 24 mo. The slow decline in viral load is in sharp contrast to the rapid decline after primary infection seen in adults. This observation can be explained only partially by the immaturity of the immune system in newborns and infants.

The 3rd pattern of disease occurs in a small percentage (<5%) of perinatally infected children referred to as long-term survivors (LTS), who have minimal or no progression of disease with relatively normal CD4 counts and very low viral loads for longer than 8 yr. Mechanisms for the delay in disease progression include effective humoral immunity and/or CTL responses, host genetic factors (e.g., HLA profile), and infection with attenuated (defective gene) virus. A subgroup of the LTS called “elite survivors” has no detectable viruses in the blood and may reflect different or greater mechanisms of protection from disease progression.

HIV-infected children have changes in the immune system that are similar to those in HIV-infected adults. CD4 cell depletion may be less dramatic because infants normally have a relative lymphocytosis. A value of 1,500 CD4 cells/mm³ in children <1 yr of age is indicative of severe CD4 depletion and is comparable to <200 CD4 cells/mm³ in adults. Lymphopenia is relatively rare in perinatally infected children and is usually only seen in older children or those with end-stage disease. Although cutaneous anergy is common during HIV infection, it is also frequent in healthy children <1 yr of age, and thus its interpretation is difficult to interpret in infected infants. The depletion of CD4 cells also decreases the response to soluble antigens such as in vitro mitogens phytohemagglutinin and concanavalin A.

Polyclonal activation of B cells occurs in most children early in the infection, as evidenced by elevation of IgA, IgM, IgE, and particularly IgG (hypergammaglobulinemia), with high levels of anti–HIV-1 antibody. This response may reflect both dysregulation of T-cell suppression of B-cell antibody synthesis and active CD4 enhancement of B-lymphocyte humoral response. As a result, antibody response to routine childhood vaccinations may be abnormal. The B-cell dysregulation precedes the CD4 depletion in many children, and may serve as a surrogate marker of HIV infection in symptomatic children in whom specific diagnostic tests (PCR, culture) are not available or are too expensive. Despite the increased levels of immunoglobulins, some children lack specific antibodies or protective antibodies. Hypogammaglobulinemia is very rare (<1%).

Central nervous system (CNS) involvement is more common in pediatric patients than in adults. Macrophages and microglia play an important role in HIV neuropathogenesis, and data suggest that astrocytes may also be involved. Although the specific mechanisms for encephalopathy in children are not yet clear, the developing brain in young infants is affected by at least 2 mechanisms. The virus itself may directly infect various brain cells or cause indirect damage to the nervous system by the release of cytokines (IL-1α, IL-1β, TNF-α, IL-2) or reactive oxygen from HIV-infected lymphocytes or macrophages.

Appropriate therapy with antiretroviral agents may result in immune reconstitution inflammatory syndrome (IRIS), which is characterized by an increased inflammatory response from the recovered immune system to subclinical opportunistic infections (e.g., tuberculosis, herpes simplex virus (HSV) infection, toxoplasmosis, CMV infection, cryptococcal infection). This condition is more commonly observed in patients with progressive disease and severe CD4⁺ T-lymphocyte depletion. Patients with IRIS develop fever and worsening of the clinical manifestations of the opportunistic infection or new manifestations (e.g., enlargement of lymph nodes, pulmonary infiltrates, etc.), typically within the 1st few weeks after initiation of antiretroviral therapy. Determining whether the symptoms represent IRIS, worsening of a current infection, a new opportunistic infection, or drug toxicity is often very difficult. If the syndrome does represent IRIS, adding nonsteroidal anti-inflammatory agents or corticosteroids may alleviate the inflammatory reaction, although the use of corticosteroids is controversial. The inflammation may take weeks or months to subside. In most cases, continuation of anti-HIV treatment while treating the opportunistic infection (with or without antiinflamatory agents) is sufficient. If opportunistic infection is suspected prior to initiation of antiretroviral therapy, appropriate antimicrobial treatment should be given first.

Clinical Manifestations

The clinical manifestations of HIV infection vary widely among infants, children, and adolescents. In most infants, physical examination at birth is normal. Initial symptoms may be subtle, such as lymphadenopathy and hepatosplenomegaly, or nonspecific, such as failure to thrive, chronic or recurrent diarrhea, respiratory symptoms, or oral thrush, and may be distinguishable only by their persistence. Whereas systemic and pulmonary findings are common in the USA and Europe, chronic diarrhea, wasting, and severe malnutrition predominate in Africa. Clinical manifestations found more commonly in children than adults with HIV infection include recurrent bacterial infections, chronic parotid swelling, lymphocytic interstitial pneumonitis (LIP), and early onset of progressive neurologic deterioration.

The HIV classification system is used to categorize the stage of pediatric disease by using 2 parameters: clinical status and degree of immunologic impairment (Table 268-1). Among the clinical categories, category A (mild symptoms) includes children with at least 2 mild symptoms such as lymphadenopathy, parotitis, hepatomegaly, splenomegaly, dermatitis, and recurrent or persistent sinusitis or otitis media (Table 268-2). Category B (moderate symptoms) includes children with LIP, oropharyngeal thrush persisting for >2 mo, recurrent or chronic diarrhea, persistent fever for >1 mo, hepatitis, recurrent (HSV) stomatitis, HSV esophagitis, HSV pneumonitis, disseminated varicella (i.e., with visceral involvement), cardiomegaly, or nephropathy (see Table 268-2). Category C (severe symptoms) includes children with opportunistic infections (e.g. esophageal or lower respiratory tract candidiasis, cryptosporidiosis (>1 mo), disseminated mycobacterial or cytomegalovirus infection, Pneumocystis pneumonia, or cerebral toxoplasmosis [onset >1 mo of age]), recurrent bacterial infections (sepsis, meningitis, pneumonia), encephalopathy, malignancies, and severe weight loss.