In the early 1980s, when the first cases of acquired immunodeficiency syndrome (AIDS) were identified, the illness appeared primarily to affect adult males. Although a few cases in children were reported as early as 1982, most pediatricians did not realize the true scope of the problem until much later in the decade. Ten years after the initial reports of pediatric AIDS, increasing proportions of AIDS cases are occurring among women and children. By September 1992, a total of 4051 cases of AIDS in children less than 13 years old had been reported to the Centers for Disease Control and Prevention (CCXC).1 An estimated 10 000 to 20 000 children in the United States are currently infected with human immunodeficiency virus type 1 (HIV-I).2 This already has had an impact on childhood mortality. In 1988, AIDS was the ninth leading cause of death in the United States for children 1 to 4 years of age.* Even more dramatic has been the impact HIV has had on Hispanic and African- American children living in particular urban centers. In New York City, HIV-related illness is the leading cause of death among Hispanic children and the second leading cause of death among AfricanAmerican children in the 1 to 4 year age group.3
As this epidemic continues to grow, it is affecting a widening social and geographic population. Thus far, the largest number of reported AIDS cases in the United States has occuned in the northeastern states, but in 1991 and 1992 the most rapid rate of increase of AIDS in both women and heterosexual men occurred in the southern and midwestern states.1 Undoubtedly, pediatric HIV-I infection will escalate in these regions as well.
The magnitude of global infection with HIV-I is disturbing. The World Health Organization reported that in 1990 more than 3 million women, most of childbearing age, were infected with HIV.4 They estimated that by the end of 1992, approximately 1 million children worldwide would have acquired HIV perinatally. In certain areas of Africa, and more recently Asia, heterosexual spread and perinatal transmission of HIV have become problems that threaten essential social and economic infrastructures.
In the United States, children infected with HIV-I generally fall into three subpopulations: 1) infants born to HIV-infected women; 2) children who received infected blood products prior to the initiation of HIV screening in 1985; and 3) adolescents. It is useful to consider the three groups separately because transmission, disease progression, and public health issues differ for each group.
EPIDEMIOLOGY OF MATERNAL-INFANT HIV-I TRANSMISSION
Infants born to HIV-infected women constitute the largest pediatric group at risk of AIDS. Over 80% of pediatric AIDS cases reported to the CDC were perinatally acquired.1 This is a direct consequence of the increasingly large number of women infected with HIV-I and the number of infants delivered annually to these women. Surveys of anonymous cord blood specimens revealed that seroprevalence varies widely between and within states. The highest rates of prevalence in 1990 were in New York (5.8/1000), Washington, DC (5.5/1000), New Jersey (4.9/1000), and Florida (4.5/1000).5 Urban areas have the highest seroprevalence rates, a pattern consistent with the epidemiology of HIV in the population as a whole.
The extent of the current pediatric HIV epidemic can be determined from the number of at-risk infants born each year and the estimated rate of vertical transmission. Prospective studies have reported motherto-infant HIV transmission in 12% to 52% of births to HIV-I infected women.3,6'10 In New York State, for example, between 1987 and September 1992, 8864 or 0.64% of heel stick specimens were HIV antibody positive (New ifork State Newborn HIV Seroprevalence Study, September 1992). Assuming a 20% transmission rate for women who do not breast-feed, as has been seen at our center and others in the United States, this likely represents an estimated 1773 infected infants. Our challenge as pediatricians remains not only to decrease the incidence of HIV in child-bearing women by educating and counseling adolescents and young adults but also to develop interventions to decrease the rate of vertical transmission.
Transmission of HIV-I from the mother to her child may be affected by the mother's disease status, the mode of delivery, perinatal feeding practices, and whether other obstetrical complications such as maternal hemorrhage or infection are present.3,6,10 Our inability to accurately determine the timing of HIV transmission to the infant complicates the assessment of risk factors. Recently, experts suggested a working definition of intrauterine HIV infection as a positive polymerase chain reaction, HIV co-culture, or p24 antigen between birth and 48 hours of life, excluding cord blood. In contrast, perinatal infection is defined as positive results of the aforementioned studies occurring more than 7 days following birth.11 Data supportive of this definition include demonstration of a pattern of viral load and p24 antigen consistent with primary infection occurring several weeks postnatally in newborns whose initial tests were negative for HIV. Use of this proposed definition to distinguish intrauterine from peripartum infection will help identify risk factors for these two modes of transmission.
Curtailing perinatal transmission also may be hindered by our failure to identify many infected women. Women with HIV infection are often asymptomatic. Most have acquired the infection through sexual contact and do not have easily identifiable risk factors. Despite obstetrical initiatives encouraging voluntary HIV antibody testing during pregnancy, many infected women remain unrecognized. As risk factors for transmission are ascertained and effective interventions are developed, identification of infected pregnant women will be crucial to reduce infection in children.
Advanced maternal HIV disease status has been consistently associated with a greater risk of intrauterine and perinatal transmission. A large prospective study of 486 children from Italy found significantly higher transmission among infants born to symptomatic mothers when compared to infants born to asymptomatic mothers (38.8% versus 21.5%).10 The European collaborative group's multicentered prospective study of 600 children born to HIV-infected mothers who were mostly asymptomatic reported a 13% to 14% transmission rate.7,8 This study found that transmission was associated with maternal p24 antigenemia and maternal CD4 lymphocyte count of less than 700/µ1.7
The role of obstetrical complications in transmission is not well defined. Caesarean section has been associated with decreased transmission of HIV in one large prospective study7 but not in others.6,9,10,12 Prematurity alone has not been associated with increased risk of perinatal infection in most studies.6,10,12 The European collaborative study, however, did find a greater risk of transmission in infants bom before 34 weeks gestation.7 The risk of vertical transmission in the presence of concurrent maternal genital infection or hemorrhage has not been studied but must be assessed since increased HIV transmission has been seen among heterosexuals with genital ulcers.13
Human immunodeficiency virus transmission via breast-feeding has been conclusively documented in cases of perinatal transmission occurring after postpartum maternal acquisition of HIV.14'20 The role of breast-feeding in vertical transmission when the mother is infected with HIV prior to delivery is less clear. Two of the European prospective cohort studies have found an increased risk of perinatal transmission in breast-fed infants,6,7,9 while a third study did not find this association.10 The majority of breast-fed infants born to HIV-infected women remain uninfected, and it is thought that immunologic components in breast milk may have a protective role in respect to HIV transmission.16,21,22 In areas of the world with poor sanitation and nutrition, the decrease in mortality associated with breast-feeding may outweigh the potential risk of HIV transmission. Therefore, breastfeeding by infected mothers is discouraged in most developed countries where safe feeding alternatives are available but not in areas with high incidences of malnutrition and mortality due to diarrheal illnesses.
NATURAL HISTORY OF PERINATAL HIV INFECTION
Human immunodeficiency virus infection in infants may be distinguished by extraordinarily rapid disease progression compared with infection in older children and adults. Whether acquired through vertical transmission or transfusion, more fulminant disease seems to occur in children who are less than 2 years of age at the time of acquisition.23 Among a cohort of 172 infants with perinatal HIV infection, 57% were diagnosed with HIV-associated disease prior to 1 year of age, while 22% were diagnosed with HIV-associated disease between I and 2 years of age. A 17% mortality rate during the first year of life, 25% by 2 years, and 50% by 3 years was observed among this cohort.24 In another study, 83% of 64 children followed prospectively with perinatally acquired HlV infection had laboratory or clinical features of HIV by 6 months of age. At 12 months, 26% of this cohort had AIDS, and 17% had died of HIV related illnesses.8 The rapid disease course found in infants differs dramatically from the long asymptomatic period following HIV-I infection in adolescents and adults. A subset of perinatally infected infants, however, remain asymptomatic for many years, which is similar to adults.8,25
The first clinical symptoms of HIV infection may be a useful indicator of the child's prognosis. Opportunistic infections, such as Pneumocystis carina pneumonia (PCP), or progressive neurologic disease are associated with rapid disease progression.23 Among cases reported to the CDC in the late 1980s, the 1 -year survival rate for children under 12 months of age with PCP was only 30% compared to 55% in this same age group without PCP.3 For older children with PCP, the 1-year survival rate was 48% compared to 72% in those without PCP. Hepatomegaly, lymphadenopathy, splenomegaly, parotitis, and lymphocytic interstitial pneumonitis are symptoms associated with longer survival.
Many early studies of children with HIV-I infection are not prospective and are heavily biased toward infants with early and rapidly progressive HIV-I disease. The prognosis for HIV-I infected children overall is probably not as bleak as these early studies imply. A prospective Italian cohort composed of perinatally infected infants identified at birth found a mortality of 8.9% in the first year of life, 3.5% per year between ages 2 and 7 years, 12% at 8 years, and 8% at 9 years of age.25 Of their cohort, 49.5% was alive at 9 years of age. The median survival was 96.2 months. Early identification of infants infected with HIV-I, interventions such as prophylaxis against PCP, and antiretroviral therapy are likely to improve the survival of HIV-infected children.
EPIDEMIOLOGY OF HIV ACQUIRED DURING CHILDHOOD
Beyond the neonatal period, HIV is transmitted to the pediatric population through infusion of infected blood products, sexual contact with HIV-infected persons, or sharing intravenous needles with infected persons. The number of children infected through these modes is significantly smaller than the number infected perinatally. Since 1985, when HIV-screening of blood products was implemented in the United States, transfusion-associated infection has become rare (less than 1 in 40 000 units of blood is potentially infected).26 Most children with transfusion-associated HIV infection acquired the infection prior to 1985.27,28 As a result, the number of new AIDS cases among children with hemophilia younger than 13 years of age reported to the CDC has declined from 26 cases during the period of October 1990 through September 1991, to 23 cases during the period of October 1991 through September 1992. Forty-six adolescents with hemophilia were reported with AIDS during the October 1990 to September 1991 period, while 49 adolescents with hemophilia were reported with AIDS the following year.1
Several reports describe sexually abused children who have acquired HIV infection.2931 One pediatric AIDS center reported that 14 of their 96 patients were confirmed victims of sexual abuse.29,30 The investigators established transmission of HIV through sexual abuse in four of their pediatric patients.30 In some of the reported cases, the perpetrator was the mother's sexual partner, and the mother also was found to be HIV infected. Determining whether a sexually abused child in this setting acquired HIV perinatally or as a result of sexual abuse is difficult.29,31
Current estimates of the incidence of pediatric sexual abuse are likely to be erroneously low, and many physicians do not routinely test for HIV infection in cases of known or suspected sexual abuse.31 The CDC does not include sexual abuse as a separate risk factor in its reports of AIDS cases.1,31 Evaluating the impact of sexual abuse in the epidemiology of pediatric HIV infection is complicated by the frequent occurrence of high-risk behaviors in previously sexually abused children. Certainly, HIV transmission resulting from pediatric sexual abuse and subsequent high-risk behaviors deserves further investigation.
Children over 2 years of age at the time of infection have a mean survival rate of 7.5 years, significantly longer than perinatally infected children.23 Interestingly, a large prospective study reported that hemophiliacs infected during childhood and adolescence are asymptomatic for a longer period of time than hemophiliacs infected as adults.28 The incidence of AIDS in these children was 0 for the first 2 years following HIV seroconversion. The incidence gradually rose to .05% in the 7.5 years following seroconversion. A strong correlation between older age and higher incidence rate of AIDS was found. Although the child's age at the time of HIV infection may affect disease progression, differences in pathogenesis relative to age have not been elucidated. Immunologic immaturity and infectious cofectors are thought to be important factors.
HIV DURING ADOLESCENCE
Adolescents are at risk of HIV infection as a result of engaging in high-risk behaviors. A national survey conducted in 1988 revealed that 52% of females and 60% of males between ages 15 and 19 have had sexual intercourse. At the age of 19, 75% of females and 86% of males report being sexually active.32 Adolescents also have a high rate of other sexually transmitted diseases that compromise mucous membrane barriers, a factor associated with an increased rate of acquisition of HIV infection.13 Although only 812 cases of AIDS in 13 to 19 year olds had been reported to the CDC by September 1992, 7549 cases were reported in 20 to 24 year olds.1 Considering an average asymptomatic period of 10.5 years, most of these young adults were infected with HIV-I during adolescence.
Data shows the prevalence of HIV-I infection in adolescents is increasing. United States Army personnel are routinely tested for HIV antibody at least every 2 years. The seroconversion among African-American military personnel less than 2 1 years of age nearly doubled from 0.75/1000 person years in 1985 to 1.5/1000 person years in 1987. 33 Overall, Army personnel under 20 years of age had an incidence rate of 0.36/1000 person years in the late 1980s.33 Looking at military recruits, adolescent African Americans had a higher seroprevalence (1.0/1000) than did their Hispanic (0.29/1000) and Caucasian (0.17/1000) counterparts.34 Seroprevalence was slightly lower in females compared to males regardless of ethnic background (0.32/1000 versus 0.35/1000). Overall, HIV seroprevalence among adolescent recruits was greatest among those who are nonwhite and living in densely populated counties with a high incidence of AIDS.
The military recruiting process selects adolescents who may be at lower risk for HIV infection than other subpopulations of US youth. Between 1987 and 1990, adolescents entering the Job Corps, a federally funded training program that recruits high school dropouts or other young people who may benefit from vocational education, were tested for HlV antibody. These 16- to 21 -year-old students were typically from socioeconomically disadvantaged environments but otherwise were not selected for high-risk behaviors. Their overall HIV seroprevalence was 3.6 per 1000.33 Males 19 years of age had a higher seroprevalence (3.7/1000) than females (3.2/1000). African- American adolescents had the highest seroprevalence (5.3/1000) compared to Hispanic (2.6/1000) and Caucasian (1.2/ 1000) adolescents. Similar to military recruits, students from larger metropolitan areas were more likely to be HIV antibody positive than students from less populated urban or rural areas, with the northeastern and southern United States reporting the greatest rates of seropositivity.
Not surprisingly, young people with particularly risky behaviors have a high incidence of HIV infection. Among 15- to 20-year-old homeless and runaway youths in New York evaluated between 1987 and 1989, 5.3% were HIV antibody positive.36 Human immunodeficiency infection virus in this population increased with age (1.3% in 15 year olds, 8.6% in 20 year olds) and con-elated with a history of high-risk activity. Human immunodeficiency virus type 1 seropositivity in males was associated with intravenous drug use, homosexual or bisexual activity, history of sexually transmitted diseases, and crack cocaine use. In contrast, HIV seropositivity in females was associated solely with a history of other sexually transmitted diseases. These alarming rates have been confirmed in other high-risk populations, including sexually transmitted disease clinics, women's health clinics, and drug treatment centers, where seroprevalence in 15 to 24 year olds has ranged from 0% to 9.6%.32 Rates were highest among adolescents reporting intravenous drug use and homosexual behavior, and varied depending on geographic location with higher rates in urban areas.
Even adolescents perceived at low risk of HIV infection are not untouched by this epidemic. A survey of college students collected from a variety of universities in 1988 found a seroprevalence of 0.2% with individual institutions ranging between 0% and 0.9%. 37 Increased age and male gender was associated with seropositivity.
While the incidence of HIV infection has leveled off in older homosexual men, infection in women, children, and adolescents is rising. Unless effective intervention measures are implemented, the number of pediatric patients with HIV and related illnesses will continue to increase. Interventions aimed at curtailing perinatal transmission potentially will have the greatest impact on decreasing the incidence of pediatric AIDS. Studies evaluating strategies aimed at this are currently underway. Educational initiatives targeting adolescents also may lower the number of pediatric patients with HIV infection. As treatments improve, the majority of HIV-infected children will survive for years and perhaps decades. The number of children with HIV infection in the upcoming decades will have a profound impact on our communities and our practice of pediatric medicine.
1. Centers for Disease Control and Prevention- HIV/AIDS Surveillance Report. Atlanta, Ga: Centers for Disease Control and Prevention; October 1992:1-18.
2. Husson RN, Piuo PA. The use of nucleoside analogues in the treatment of HIV-infected children. AIDS Res Ham Retrovirus. 1992;8:1059-1064.
3. Oxtohy MJ. Perinatally acquired human immunodeficiency virus infection. Pediatr Infect Dis. J 1990;9:609-619.
4. Chin J. Current and future dimensions of the HIV/AIDS pandemic in women and children. Lancet. 1990:336:22 1-224.
5. Gwmn M. Pappaioanou M, George JR, et al. Prevalence of HIV infection in childbearing women in the United States. JAMA. 1991;265:1704-1708.
6. Gabiano C. Tovo P-A, de Martino M, et al. Mother-to-child transmission of human immunodeficiency virus type 1: risk of infection and correlates ci transmission. Pediatrics. 1992:90:369-374.
7. European Collaborative Study. Risk factors for mother-to-child transmission of HlVl. Lancet. 1992;339:1007-1012.
8. European Collaborative Study. Children bom to women with HlV-I infection: natural history and risk of transmission. Lancet. 1991;337:253-260.
9. Blanche S, Rouzioux C, Moscato MG. et al. A prospective study of infants bom to women seropositive for human immunodeficiency vims type 1. N Engl J Med. 1989:320:1643-1648.
10. Italian Multicentre Study. Epidemiology, clinical features, and prognostic factors of pediatric HIV infection. Lancer. 1988;2(8169):1043-1046.
11. Bryson YJ, Luzuriaga K, Sullivan JL, Wara DW. Proposed definitions for in utero versus intrapartum transmission of HIV-I. N Engl J Med. 1992;327:1246-1247.
12. Hurto C, Parks WP, Lai 5, et al. A hospital-based prospective study of perinatal infection with human immunodeficiency vims type I. J Pediatr. 1991;118:347-353.
13. Cameron DW, Simonsen JN. D'Cbsta LJ. et al. Female to male transmission of human immunodeficiency virus type 1: risk factors for seroconversion in men. Lancet. 1989;2(8660):403-407.
14. Zieglcr JB. Cooper DA, Johnson RO, Gold J. Postnatal transmission of AIDSassociated retrovirus from mother to infant. Lancet. l985:l(8434):896-898.
15. Lepage P, Van de Pene P, Carnei M, et al. Postnatal transmission of HIV from mother to child. Lancet. I987;2(8555):400-40I.
16. Dunn DT, Newell ML, Ades AE. Peckham CS. Risk of human immunodeficiency virus type I transmission through breastfeeding. Lancet. 1992;340:585-588.
17. Colebundcrs R, (Capita B, Nekwei W, et al. Breastfeeding and transmission of HlV. Lancet. 1 988;2 (8626-862 7): 1487.
18. Stiehm ER, Vink P. Transmission of human immunodeficiency virus infection by breast -feeding. J Pediatr. 1 99 1 ; 1 1 8:410-4 1 2 .
19. Palasanthiran P, Ziegler JB1 Stewart GJ, et al. Breast-feeding during primary maternal human immunodeficiency vims infection and risk of transmission from mother ro infant. J infect Dis. 1993:167:441-444.
20. Van de Perre P, Simonon A, Msellati P et a). Postnatal transmission of human immunodeficiency vims type I from mother to infant. N Engl J Med. 199l;325:591598.
21. Ruff AJ. Halsey NA, Coberly J, Boulos R. Breast-feeding and maternal-infant transmission of human immunodeficiency vims type i. J Pediatr. 1992;121:325-329.
22. Senturia YD Ades AE, Peckham CS. Breastfeeding and HIV infection. Lancet. 1987;2(8555):400-401.
23. Krasinski K, Borkowsky W, Hollinan RS. Prognosis of human immunodeficiency vims infection in children and adolescents. Pediatr Infect Dis J. 1989;8:216-220.
24. Scott GB. Hutto C, Makuch RW, et al. Survival in children with perinatally acquired human immunodeficiency virus type 1 infection. N Engl J Med. 1 989; 321:1 79 1 - 1796.
25. Tovo PA, de Mattino M, Gabiano C, et al. Prognostic factors and survival in children with perinatal HIV-I infection. Lancet. 1992;339:1249-1253.
26. Dodd RY. The risk of transfusion-transmitted infection. N Engl J Med. I992;327:419422.
27. Centers for Disease Control and Prevention. Acquired immunodeficiency syndrome 1991. MMWR. 1992;41:463-468.
28. Goedert JJ, Kessler CM, Aledoft LM, et al. A prospective study of human immunodeficiency virus type 1 infection and the development of AIDS in subjects with hemophilia. N Engl J Med. 1 989:32 1 : 1 1 4 1 - 1 1 48.
29. Gutman LT, St. Claite KK. Weedy C, et al. Human immunodeficiency virus transmission by child sexual abuse. AmJ Dis Child. 1991:145:137-141.
30. Gutman LT, St. Claire KK, Weedy C. Herman-Giddens M. McKinney RE. Sexual abuse of human immunodeficiency vims-positive children. Am J Dis Child. 1992;146:1185-1189.
31. Gellen GA. Durfee MJ, BerkowirrCD, Higgens KV, Tubiolog VC. Siruarional and sociodemographic characteristics of children infected with human immunnodeficiency virus from pediatric sexual abuse. Pediatrics. 1993:91:39-44.
32. Wendell DA, Onorato IM. McCray E. Allen DM, Sweeney PA. Youth at risk: sex, drugs, and human immunodeficiency vims. AmJ Dis Child. 1992;146:76-81.
33 McNeil JG, Bmndage JE Gardner LI, et al. Trends of HIV seroconversion among young adults in the US Army, 1985 to 1 989. 7AMA. 1991:265:1709-1714.
34. Burke DS, Bmndage JF, Goldenbaum M, et al- Human immunodeficiency virus infections in teenagers. Seroprevalence among applicants for US military service. JAMA. 1990;263:2074-2077.
35. St. Louis ME, Conway GA, Hayman CR, Miller C, Petersen LR. Pondero TJ. Human immunodeficiency vims infection in disadvantaged adolescents- Findings from the US Job Corps. JAMA- 1991;266:2387-2391.
36. Stricof RL. Kennedy JT. Nattell TC, Weisfuse IB. Novick LF. HlV seroprevalence in a facility for runaway and homeless adolescents. Am J Public Health. 1991;81(S):50-53.
37. Gayle HD, Keeling RP, Garcia-Tunon M, et al. Prevalence of the human immunodeficiency vims among university students. N Engl J Med. 1 990:323:1538-1541.