Survival, Disease Manifestations, and Early Predictors of Disease Progression Among Children With Perinatal Human Immunodeficiency Virus Infection in Thailand
Objective. To describe survival and signs of human immunodeficiency virus (HIV) infection in perinatally infected children in Thailand.
Methods. At 2 large Bangkok hospitals, 295 infants born to HIV-infected mothers were enrolled at birth from November 1992 through September 1994 and followed up with clinical and laboratory evaluations every 1 to 3 months for 18 months. Infected children remained in follow-up thereafter. For the infected children, we used data collected through October 2000 to estimate survival times and compare characteristics among those whose disease progressed at rapid (died within 1 year), intermediate (died at 1–5 years), and slow (survived at least 5 years) rates.
Results. None of the 213 uninfected children died during the follow-up period. Of the 68 infected children, 31 (46%) died; median survival was 60 months (95% confidence interval: 31–89 months). The most common cause of death was pneumonia (52% of deaths). Thirty-two children (47%) started antiretroviral therapy. Six children died in their first year before developing specific signs of HIV infection; all others developed signs of HIV infection between 1 and 42 months old (median: 4 months). Severe clinical (Centers for Disease Control and Prevention Class C) conditions were diagnosed in 23 children at a median age of 12 months, 15 (65%) of whom died a median of 3 months later. Compared with children whose disease progressed slowly, those whose disease progressed rapidly gained less weight by 4 months old (median 1.7 vs 2.6 kg), and their mothers had higher viral loads (median 5.1 vs 4.5 log10 copies/mL) and lower CD4+ counts (median 350 vs 470 cells/μL) at delivery.
Conclusions. Among HIV-infected Thai children, survival times are longer than among children in many African countries, but shorter than among children in the United States and Europe. Signs of HIV develop early in most children. Growth failure and advanced maternal disease can predict rapid HIV disease progression and may be useful markers for treatment decisions.
- vertical human immunodeficiency virus transmission
- human immunodeficiency virus infection
The epidemic of perinatal human immunodeficiency virus (HIV) infection in Thailand began in the early 1990s, when the median HIV seroprevalence, by province, among women in antenatal care jumped from 0% in 1990 to > 2% by 1995.1 In Thailand, >30 000 children have been perinatally infected with HIV, and >7000 cases of AIDS or symptomatic HIV infection in children have been reported.2,3 Although Thailand’s national perinatal HIV prevention program, which recommends short-course zidovudine and formula replacement feeding, should reduce the incidence of HIV infection in children, ∼15 000 HIV-infected children are living in Thailand and more will continue to be identified through systematic HIV testing now provided for children born to HIV-infected mothers.4
Despite increasing availability of antiretroviral treatment for HIV-infected children, illness and death caused by the progression of perinatal HIV infection will place a substantial burden on the health care system in Thailand. Caring for these HIV-infected children requires an understanding of the manifestations of perinatal HIV infection in Thailand, where most HIV infections in children are with subtype E.5 Knowing the expected duration of survival after perinatal infection helps with counseling families of HIV-infected children. Identifying predictors of disease progression can help indicate prognosis and guide treatment.
Despite the magnitude and duration of the HIV epidemic in Thailand, there has been little systematic description of the manifestations of HIV in perinatally infected Thai children. In this study, we describe manifestations of HIV infection, estimate survival times, and identify predictors of disease progression among a cohort of 68 Thai children with perinatal HIV infection.
From November 1992 through March 1994, we enrolled 342 HIV-infected pregnant women into a prospective cohort study of perinatal HIV transmission at Rajavithi and Siriraj Hospitals in Bangkok. Children born at Rajavithi Hospital were followed up at the adjacent Queen Sirikit National Institute of Child Health (formerly known as Children’s Hospital). Women who provided their informed consent were enrolled during antenatal care, and their children were followed up from birth. The enrollment procedures for this study and analysis of perinatal transmission have been published.5,6 The study was conducted before the AIDS Clinical Trials group 076 study ended7; antiretroviral prophylaxis was not yet used to prevent mother-infant transmission, although HIV-infected mothers were advised not to breastfeed and were provided infant formula and ongoing infant feeding counseling. During the course of this study, primary prophylaxis against Pneumocystis carinii pneumonia (PCP) was not used routinely for HIV-exposed infants, but was used for children with signs of HIV infection. Antiretroviral treatment was starting to be used for symptomatic HIV-infected children, at first monotherapy and later dual nucleoside therapy. Because treatment usually was started once children developed HIV-related symptoms, we could not analyze the impact of treatment on disease progression.
Women had study visits during each remaining trimester of pregnancy, at delivery, and at 6 and 12 months’ postpartum. All children were evaluated at birth and scheduled for study visits at 1, 2, 4, 6, 9, 12, 15, and 18 months of age. Infected children have been followed up indefinitely every 3 months. At each visit the child was examined, and the child’s interim medical history was recorded along with information on hospitalizations, outpatient visits, medication use, and the child’s diagnoses. Venous blood specimens were taken within 72 hours of birth, at 2, 6, 12, and 15 months of age, and every 6 months thereafter for lymphocyte phenotyping according to published methods.5 Plasma HIV RNA concentration (viral load) was measured retrospectively for selected specimens.
The child’s infection status was determined from the results of HIV DNA polymerase chain reaction (PCR) tests done at birth and at 2 and 6 months of age.5 Infants were considered HIV-infected if they had positive test results from 2 separate samples or one positive test result and a Centers for Disease Control and Prevention (CDC) Class C HIV condition.8 Children were considered not infected if 2 separate samples, including 1 obtained at >6 months of age, had a negative PCR result or if 1 specimen was negative for HIV antibody by enzyme immunoassay testing. Children were considered to have indeterminate infection status if none of these conditions was met.
We analyzed data collected through October 2000. We estimated time of survival to death and to death or CDC Class C disease8 using the Kaplan-Meier method.9 Survival times of children lost to follow-up were censored at the time of last available information on clinical and vital status. To determine whether loss to follow-up affected the results, we compared factors found to be significant for early disease progression and time to Class C disease between children lost to follow-up and those who died or remained in follow-up. We used the χ2 or Fisher exact test to compare categorical variables and the Student t test to compare continuous variables among groups. We defined rapid progression as death within 1 year of birth, intermediate progression as death between 1 and 5 years of birth, and slow progression as survival for at least 5 years.
Population Characteristics and Follow-up
Of the 295 children born in the cohort from November 1992 through September 1994, 68 were infected with HIV, 213 were not infected, and 14 had indeterminate HIV infection status. The estimated mother-infant HIV transmission rate in this cohort was 24.2%; 66 (97%) of the infected children were infected with HIV subtype E.5 Only 1 mother breastfed her child. Among the children not infected, 193 (91%) completed 18 months of follow-up, and none was known to have died during follow-up. Three of the children with indeterminate HIV status died, all in the first month of life (sepsis, pneumonia, gastroschisis).
As of October 2000, 31 (46%) HIV-infected children had died, 21 (31%) were lost to follow-up, and 16 (24%) were still in follow-up. The median time of follow-up was 28 months (range: 2–84 months). Of the 21 children who were lost to follow-up, the age at last visit ranged from 2 to 57 months; the median age was 24 months. The main reason for loss to follow-up was family relocation outside of Bangkok. Children lost to follow-up were similar to those remaining in follow-up on most characteristics; however, children lost to follow-up had a higher median weight at 4 months old (5.6 kg vs 5.2 kg, P = .008, Table 1).
Morbidity and Mortality Among HIV-Infected Children
Of the 31 children who died, the age at death ranged from 2 to 69 months, the median age was 22 months, and 22 (71%) had started antiretroviral therapy.
The estimated median survival time after infection (ie, birth) was 60 months (95% confidence interval: 31–89 months), with an estimated 82%, 74%, 61%, 56%, 49%, and 43% surviving for 1, 2, 3, 4, 5, and 6 years, respectively. The estimated median survival time free from Class C disease or death was 33 months (95% confidence interval: 16–50 months), with 73%, 59%, 49%, 45%, 40%, and 34% surviving without Class C disease for 1, 2, 3, 4, 5, and 6 years, respectively. Survival free of Class C disease was similar for children who were lost to follow-up and those who were not. Infectious causes of death were most common: pneumonia (52%), diarrhea (19%), and other infectious causes (sepsis and/or fungal esophagitis, 10%). Causes of death, stratified by age, are shown in Fig 1.
Of the 68 infected children, 6 died at 4 to 8 months of age without having had any clinical signs of HIV infection recognized by the time of their last study visit, which was a median of 1.5 months before death. Four of these children had CD4+ cell counts <1500 cells/μL at 2 months old; 3 died from pneumonia.
The other 62 infected children developed an HIV-related sign8 at a median age of 4 months (range: 1–42 months). A Class C condition (recurrent serious bacterial infections)developed as the first sign of HIV infection in 2 children aged 4 and 11 months; 23 children had >1 sign at presentation (Table 2). The most common first signs and median ages when first noted were lymphadenopathy (n = 25 [40%], median: 7 months old), splenomegaly (n = 19 [31%], median: 3 months old), and hepatomegaly (n = 18 [29%], median: 3 months old). A Class C condition was diagnosed in 21 children at a median age of 12 months; 14 (67%) died at a median of 5 months after diagnosis (Table 3). Two children were diagnosed with lymphoid interstitial pneumonitis at 63 and 69 months old. After signs of HIV infection appeared, 32 (52%) of the 62 children with symptomatic HIV infection were prescribed antiretroviral medications, and 36 (58%) were prescribed prophylaxis against PCP. Antiretroviral therapy had been prescribed for only 12 (22%) of the 55 children in follow-up at 1 year of age, but 19 (63%) of the 30 in follow-up at 3 years of age.
Predictors of Rapid Progression
Twelve infected children died within 1 year of birth (“rapid progression”), 17 children died between 1 and 5 years of age (“ intermediate progression”), and 18 children survived for >5 years (“slow progression;” Table 1). Children who gained fewer than 2 kg by 4 months old were 7 times more likely, and children whose CD4 count was <1500 cells/μL at 2 months old were 4 times more likely to have rapid progression (Table 4). Other significant predictors were low CD4 count or high viral load in the mother at delivery and high viral load in the infant at 2 months of age.
Status of Children Still in Follow-up
Of the 16 children still in follow-up in October 2000, the median age at last visit was 73.5 months and the age range was 62 to 84 months. At the last visit, 14 children (88%) were taking antiretroviral medications, and 15 (94%) were taking trimethoprim-sulfamethoxazole as prophylaxis against PCP. The median CD4+ cell count was 452 (range: 15–1131) cells/μL. The mothers of 6 (38%) children had died, and 7 (44%) children were still living with their biological mothers.
In this large systematic cohort study of perinatally HIV-infected children in Asia, the HIV subtype, demographics, disease prevalence, medical infrastructure, and racial composition differ from those in similar studies in America, Europe, and Africa. Thus, the results of this study may be more helpful than studies from elsewhere in guiding clinicians and public health officials in this region, especially Thailand.
By 5 years after birth, an estimated 49% of the HIV-infected Thai children in our study had died, a proportion higher than that reported from Europe (26% by 6 years)10 or the United States (25%–35% by 5 years)11,12 but lower than that reported from Rwanda (62% by 5 years)13 or Uganda (54% by 2 years).14 The 1-year survival rate in this cohort (82%) is similar to that reported from a more recent study (short-course zidovudine perinatal trial) in the same 2 Bangkok hospitals during 1996–1998 (85%).15 Similar to that of other populations, the risk of dying was greater in the first 3 years of life than afterward, and pneumonia was a common cause of death, especially in the first year.12 Although specific diagnosis was unavailable in most cases, it is likely that some of these early cases of fatal pneumonia were caused by P carinii, which has its peak incidence in the first year of life among HIV-infected children.16,17 Median survival after the diagnosis of Class C disease was short (3 months), even shorter than the range of survival times reported from other studies (9–23 months).10,13 With earlier diagnosis and more effective treatment for both the Class C conditions and the underlying HIV infection becoming available, survival times in settings such as this are likely to become longer.
Except for 6 children who died in early infancy, all children developed HIV-associated signs, most appearing in the first 6 months of life, as reported in other cohorts.10,13,18 The most common signs in our cohort were lymphadenopathy, splenomegaly, and hepatomegaly. Although nonspecific, such signs can be useful for making clinical decisions regarding the care of HIV-exposed children (eg, whether to start or continue treatments). Four of the 6 infants who died before signs of HIV were recognized were severely immunosuppressed, and 3 died of pneumonia. These findings of early, fatal pneumonia should be considered when deciding on strategies for preventing PCP in infants. In a separate cohort, we evaluated a PCP prophylaxis strategy in which all infants born to HIV-infected mothers were prescribed trimethoprim-sulfamethoxazole and clinical indicators were used at 6 months old to decide whether prophylaxis should be continued. This evaluation suggested that such a PCP prophylaxis strategy can prevent death from pneumonia and that clinical findings can be used for deciding about discontinuing prophylaxis.17,19
Because understanding predictors of rapid disease progression provides useful information for counseling families about a child’s prognosis and for making decisions about clinical management, we examined the relationships between early death and several maternal, birth, and early infant factors. Several factors were observed more frequently with regard to children who died within 12 months of birth: high viral load and low CD4+ cell count in their mothers at delivery, low CD4+ cell count and high viral load in the infants at 2 months old, and poor infant weight gain by 4 months old. These factors have been associated with HIV disease progression in other cohort studies in the United States, Europe, and Africa.10,13,14,18, 20–29 Because CD4 count measurement is increasingly available in Thailand, maternal test results may be useful to guide interventions for both HIV-infected women and their children. For countries with more limited resources, an inexpensive and widely available prognostic marker such as early weight gain may be useful. Although it is unknown whether nutritional interventions would improve survival among poorly thriving children with HIV infection or whether poor weight gain itself is a manifestation of advanced HIV infection, children with poor weight gain would likely benefit from more intensive nutritional and medical interventions. A positive PCR test at birth, considered a proxy for in utero infection, was not a significant predictor of rapid progression in our cohort, although the number of children analyzed was small.
This study also indicates that avoidance of breastfeeding and use of infant formula from birth did not have any apparent adverse impact on mortality among the uninfected children born to HIV-infected mothers. Although there were several deaths among newborns with unknown status, there were none among the 213 infants known to be uninfected. These data support the safety of Thailand’s policy of providing replacement feeding for infants born to HIV-infected mothers.4
Our study has several limitations. We studied a relatively small number of HIV-infected children, thereby limiting our statistical power to precisely define survival risk or to conduct meaningful multivariable analyses of risk factors. Because antiretroviral therapy usually was started in response to disease progression, we could not adjust our analysis of risk factors for antiretroviral treatment and so cannot speculate as to the extent to which antiretroviral treatment (generally zidovudine monotherapy or dual nucleoside analog therapy) may have prolonged life. Moreover, it is likely that more potent antiretroviral regimens, such as those including a protease inhibitor, would have delayed clinical progression and death. Finally, whether the findings from these 2 large Bangkok tertiary hospitals can be generalized to other settings in Thailand or Asia is unknown.
Nonetheless, this cohort has several features that make these data useful for clinicians, counselors, and policymakers. Its prospective follow-up from birth, systematic collection of clinical and laboratory data, and relatively long follow-up period enhance the reliability of the data. The description of the manifestations of perinatally acquired HIV disease in Thailand can assist the growing number of clinicians caring for HIV-infected children in Thailand and Asia. Estimates of survival times can aid in appropriate counseling of families about the prognosis of HIV-infected children. Finally, estimates of survival times and incidence of severe disease can assist public health policymakers estimate the number of HIV-infected children requiring long-term health care and other support services. The finding that half of the HIV-infected children in Thailand may be surviving for 5 years or more even with limited antiretroviral therapy highlights the substantial resources that are needed to meet the long-term medical and social needs of these children, such as antiretroviral therapy, other health care, and psychosocial support.
Other members of the collaborative study group, not listed as coauthors, include the following: S. Neungton, P. Chaisilwattana, A. Roongpisuthipong, A. Chalermchokcharoenkit, K. Sirimai, P. Phopong, C. Bhadrakom, P. Chaiyakul, P. Rattananikhom, R. Prechanont (Faculty of Medicine Siriraj Hospital, Department of OB-GYN); M. Tuchinda, S. Pichitchaichan, W. Boonyavit (Faculty of Medicine Siriraj Hospital, Department of Pediatrics); C. Wasi (Faculty of Medicine Siriraj Hospital, Department of Microbiology); P. Chinayon, W. Siriwasin, S. Asavapiriyanont, B. In-neam, S. Supatosa, C. Kannasot, S. Sangkasuwan, S. Leampojara, P. Pramukhakul (Rajavithi Hospital Department of OB-GYN); S. Singhanati, G. Kaewchaiyo (Rajavithi Hospital Laboratory); J. Sawakwan, N. Montasewee (Rajavithi Hospital Department of Nursing); S. Horpaopan, V. Sangtaweesin, P. Na Chiengmai, R. Kulvisuthpravit, B. Phasukdee, P. Sojirat (Queen Sirikit National Institute of Child Health); K. Limpakarnjanarat, A. Bennetts, N. Chantharojwong, T. Naiwatanakul, J. Laosakkitiboran, P. Yuentrakul, C. Manopaiboon, N. Skunodom, N. Young, T. Chaowanachan (The HIV/AIDS Collaboration).
We thank the dedicated field work of the following project study nurses and social workers: S. Bhengsri, S. Henchaichon, S. Jalanchavanapate, K. Klumthanom, R. Buakor, C. Prasert, W. Sanyanusin, W. Suwannapha, S. Sorapipatana, S. Suwanmaitre, W. Triphanitchkul, C. Yuvasevee, and P. Kajohnsee. We also thank the HIV/AIDS Collaboration laboratory staff for laboratory specimen processing and testing and the study participants and their families for volunteering for this study.
- Received November 6, 2001.
- Accepted May 8, 2002.
- Reprint requests to (S.C.) HIV/AIDS Collaboration, DMS 6 Building, Ministry of Public Health, Tivanon Road, Nonthaburi, 11000, Thailand. E-mail:
- ↵Ungchusak K, Saengwonloey O, Jaemkrajang P, et al. HIV serosurveillance in Thailand: result of the 16th round, June 1998 [in Thai]. Monthly Epidemiol Surveill Rep.1999;30 :1– 7
- ↵Simonds RJ, Teeraratkul A, Kanshana S, et al. Estimating the burden of perinatal HIV infection and the effect of interventions in Thailand. Second Conference on Global Strategies for the Prevention of HIV Transmission from Mothers to Infants; Montreal, Canada; September 1999. [Abstract 375]
- ↵Guharat S, Sangwanloy O, Simonds RJ, Limpakarnjanarat K, Mastro TD. A decade of perinatal AIDS in Thailand, 1988–1998. Fifth International Congress on AIDS in Asia and the Pacific; Kuala Lumpur, October 1999. [Abstract 399]
- ↵Shaffer N, Roongpisuthipong A, Siriwasin W, et al. Maternal viral load and perinatal HIV-1 subtype E transmission, Bangkok, Thailand. J Infect Dis.1999;179 :590– 599
- ↵Centers for Disease Control and Prevention. 1994 revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Morb Mortal Wkly Rep.1994;43(No. RR-12) :1– 10
- ↵Barnhart HX, Caldwell MB, Thomas P, et al. Natural history of human immunodeficiency virus disease in perinatally infected children: an analysis from the Pediatric Spectrum of Disease Project. Pediatrics.1996;97 :710– 716
- ↵Langston C, Cooper ER, Goldfarb J, et al. Human immunodeficiency virus-related mortality in infants and children: data from the Pediatric Pulmonary and Cardiovascular Complications of Vertically Transmitted HIV (P2C2) Study. Pediatrics.2001;107 :328– 338
- ↵Spira R, Lepage P, Msellati P, et al. Natural history of human immunodeficiency virus type 1 infection in children: a five-year prospective study in Rwanda. Pediatrics.1999;104(5) . Available at: http://www.pediatrics.org/cgi/content/full/104/5/e56
- ↵Berhane R, Bagenda D, Marum L, et al. Growth failure as a prognostic indicator of mortality in pediatric HIV infection. Pediatrics.1997;100(1) . Available at: http://www.pediatrics.org/cgi/content/full/100/1/e7
- ↵Chotpitayasunondh T, Vanprapar N, Simonds RJ, et al. Safety of late in utero exposure to zidovudine in infants born to HIV-infected mothers, Bangkok. Pediatrics.2001;107(1) . Available at: http://www.pediatrics.org/cgi/content/full/107/1/e5
- ↵Abrams EJ, Matheson PB, Thomas PA, et al. Neonatal predictors of infection status and early death among 332 infants at risk of HIV-1 infection monitored prospectively from birth. Pediatrics.1995;96 :451– 458
- Abrams EJ, Weedon J, Steketee RW, et al. Association of human immunodeficiency virus (HIV) load early in life with disease progression among HIV-infected infants. J Infect Dis.1998;178 :101– 108
- Dickover RE, Dillon M, Leung KM, et al. Early prognostic indicators in primary perinatal human immunodeficiency virus type 1 infection: importance of viral RNA and the timing of transmission on long-term outcome. J Infect Dis.1998;178 :375– 387
- Mofenson LM, Korelitz J, Meyer WA, et al. The relationship between serum human immunodeficiency virus type 1 (HIV-1) RNA level, CD4 lymphocyte percent, and long-term mortality risk in HIV-1-infected children. J Infect Dis.1997;175 :1029– 1038
- Rich KC, Fowler MG, Mofenson LM, et al. Maternal and infant factors predicting disease progression in human immunodeficiency virus type 1-infected infants. Pediatrics.2000;105(1) . Available at: http//www.pediatrics.org/cgi/content/full/105/1/e8
- ↵Kuhn L, Steketee RW, Weedon J, et al. Distinct risk factors for intrauterine and intrapartum HIV transmission and consequences for disease progression in infected children. J Infect Dis.1999;179 :52– 58
- Copyright © 2002 by the American Academy of Pediatrics