Secondary Logo

Share this article on:

Delayed Seroreversion in HIV-exposed Uninfected Infants

Chatpornvorarux, Sunsanee, MD; Maleesatharn, Alan, MBA; Rungmaitree, Supattra, MD; Wittawatmongkol, Orasri, MD; Phongsamart, Wanatpreeya, MD; Lapphra, Keswadee, MD; Kongstan, Nantaka, BN.Msc; Khumcha, Benjawan, BA; Chokephaibulkit, Kulkanya, MD

The Pediatric Infectious Disease Journal: January 2019 - Volume 38 - Issue 1 - p 65–69
doi: 10.1097/INF.0000000000002196
HIV Reports

Background: Recent studies report delayed anti-HIV antibody clearance (seroreversion) among HIV-exposed uninfected infants that may affect diagnostic practices. We evaluated the age-specific seroreversion rates in Thailand.

Methods: The medical records of HIV-exposed uninfected infants born in January 2000–December 2014 were reviewed. Anti-HIV seroreversion rates at 12, 18 and 24 months were analyzed in 3 periods according to the Thai National Guidelines of prevention of mother-to-child transmission of HIV: zidovudine with or without single dose nevirapine to all women (2000–2006), adding lamivudine plus nevirapine to zidovudine in women with CD4 count <200 cells/mm3 (2007–2009) and zidovudine plus lamivudine plus boosted lopinavir to all women (2010–2014). In 2013, the serologic test kit was changed from third- to fourth-generation (4G) assay. All the infants were formula fed.

Results: Among 736 infants, the overall seroreversion rates at 12, 18 and 24 months of age were 59.38%, 94.57% and 100%, respectively. The seroreversion rates at 12 months of age declined from 68% in 2000–2006 and 65.9% in 2007–2009, to 42.9% in 2010–2014 (P = 0.001). Seroreversion rates at 18 months of age were more than 96.5% before 2013 and decreased to 79.1% in 2013–2014 (P = 0.001) with use of 4G. Multivariate analysis identified antepartum protease inhibitors treatment and the use of 4G testing as independent factors associated with delayed seroreversion.

Conclusions: Anti-HIV seroreversion delay in HIV-exposed uninfected infants was associated with use of protease inhibitors and 4G HIV testing, complicating the interpretation to exclude perinatal HIV infection.

From the Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand.

Accepted for publication August 23, 2018.

The authors have no funding or conflicts of interest to disclose.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (

Address for correspondence: Kulkanya Chokephaibulkit, MD, Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, 2 Wanglang Road, Bangkoknoi, Bangkok 10700, Thailand. E-mail:

The diagnosis of HIV infection using anti-HIV serology in infants less than 18 months of age is complicated by transplacental transfer of maternal antibodies. Virologic assays using nucleic acid testing (NAT) have been recommended for early diagnosis and treatment, as well as for monitoring the prevention of mother-to-child transmission program. However, NAT is less commonly available than HIV serology especially in resource limited setting.1 , 2 Anti-HIV serology in infants has been used limited to confirmation of HIV exposure and to exclude HIV infection in older infants with perinatal exposure. The World Health Organization recommends HIV serologic testing, using rapid diagnostic test (RDT), at around 9 months of age (or during the last immunization visit) for HIV-exposed infants who are otherwise healthy.3 If the test is negative, and the infant had not been breast-fed for at least 3 months, they are considered uninfected. A positive serology requires NAT to inform decisions regarding antiretroviral therapy. Using HIV serology in older infants to screen for HIV infection can reduce the need for NAT.

Maternal passive immunoglobulin G (IgG) levels gradually decline with increasing age of the infant, eventually resulting in seroreversion of anti-HIV antibody in exposed but uninfected infants. Before the year 2000, studies in the United States and Europe revealed the median age at clearance of maternal anti-HIV antibody as measured by enzyme-linked immunosorbent assay (ELISA) serologic assay of 9.4–10.9 months.4–6 Less than 2% of uninfected children were found to have a positive anti-HIV serology at 18 months.4 A study in 1995 in HIV-exposed uninfected South African infants reported a seroreversion rate of 95% by 12 months of age.7 More recent studies have reported an increased age at seroreversion. A study in Vietnam in 2009 reported the median age at seroreversion of 18.8 months and a seroreversion rate of only 22% at 12 months.8 Likewise, a US study during 2000–2007 reported a seroreversion rate of only 25% at 12 months.9 A study in Malawi found delayed seroreversion in infants born after the year 1999 compared with those born before 1997.10 The reasons for these findings are unclear. Prior studies found various cofactors related to seroreversion such as type of antibody test kit, maternal HIV IgG concentrations, severity of maternal HIV status, protease inhibitors (PIs)–based antiretroviral drug regimen during pregnancy, nutritional status of infants and clearance rate of HIV IgG of infants.8 , 9 , 11–14 It is not known whether this trend of increasing age at seroreversion occurs in different settings, populations and type of test.

Delayed seroreversion can influence the diagnostic approach. The utility of earlier HIV serology is reduced if the likelihood of seroreversion is low, and positive tests may increase parental anxiety. In our setting with exclusive infant formula feeding, serology has been recommended in all HIV-exposed infants at 12–18 months of age to exclude infection before discharge from the HIV clinic. Any trend towards delayed seroreversion could impact the timing of diagnostic testing. Therefore, we aimed to evaluate the proportion of seroreversion of anti-HIV antibody in HIV-exposed but uninfected infants at 12, 18 and 24 months and to determine factors associated with seroreversion at 12 months of age.

Back to Top | Article Outline


This retrospective study was conducted at the pediatric HIV clinic of Siriraj hospital, a large tertiary public hospital in Bangkok. HIV-exposed uninfected infants who were born during January 2000 to December 2014 were included. The HIV status was based on US-Centers for Disease Control and Prevention recommendations for nonbreastfed infants who required 2 or more negative virologic tests; at age ≥1 month and at age ≥4 months.1 The child must not have other laboratory or clinical evidence of HIV infection. Breastfeeding is contraindicated for HIV-exposed infants in Thailand, and the Ministry of Public Health provides free infant formula to all HIV-exposed infants up to 18 months of age. During the study period, anti-HIV serology is routinely performed in all infants born to HIV-infected mothers at 12 months of age and, if positive, the test is repeated at 18 months and again at 24 months of age if needed, according to the Thai National Guidelines. All the infants included in this study must have evidence of seroreversion. Infants who were lost to follow-up before seroreversion was documented were excluded from the study. All twins were included in the analysis.

Seroreversion rates at 12, 18 and 24 months were analyzed for each year of birth and also stratified by the period of antiretroviral therapy (ART) recommendation by the National Guidelines for prevention of mother-to-child transmission (Fig. 1).15 From 2000 to 2003, all HIV-infected pregnant women were recommended to receive zidovudine (AZT) beginning at 34 weeks of gestation. In 2004, the National Guidelines recommended AZT beginning at 28 weeks of gestation with an addition of a single dose nevirapine (NVP) during labor. In 2007, a combined regimen of AZT plus lamivudine (3TC) plus NVP was recommended to begin at 28 weeks gestation for women with CD4+ T-lymphocytes (CD4) less than 200 cells/mm3. In 2010, the recommendation was changed to AZT plus 3TC plus lopinavir/ritonavir (LPV/r) for all HIV-infected pregnant women beginning at 14 weeks gestation regardless of CD4 levels. During 2000–2009, all HIV-exposed infants were recommended to receive AZT syrup from birth until 4 weeks. From 2010, the triple combination regimen of AZT plus 3TC for 4–6 weeks plus NVP for 2–4 weeks was recommended for infants at high risk of HIV infection [maternal ART less than 4 weeks before delivery, poor compliance with the drug regimen or a high viral load (VL) at delivery]. Therefore, the stratified periods in this analysis were 2000–2006 (AZT mono therapy with or without single dose NVP), 2007–2009 (AZT/3TC/NVP in those with low CD4) and 2010–2014 (AZT/3TC/LPV/r). In a subject with missing data at 12 months and a positive test at 18 months, the 12-month test was assumed to also be positive. On the other hand, if an earlier test result of a subject was negative, the subsequent missing results were also assumed to be negative.



From 2000 to 2012, a third generation (3G) anti-HIV ELISA test kit (AxSYM HIV 1/2 gO, Abbott Diagnostics, Wiesbaden, Germany) was used. This test had a reference sensitivity of 100% and specificity of 99.4%–99.9%. From 2013, the hospital laboratory changed anti-HIV test kit to fourth-generation (4G) VIDAS HIV DUO Ultra (Biomérieux, Marcy-l’Etoile, France) with reference sensitivity of 99.5%–100% and specificity 98%–100% to improve diagnosis among patient with acute HIV infection.

Maternal and infant data were extracted from medical records to explore factors that might be associated with seroreversion at age 12 months. Maternal variables included age at delivery, duration of ART before delivery, ART regimen during pregnancy and the postpartum period, history of opportunistic infections and the CD4 count and VL at ART initiation and within 4 weeks of delivery. The infants’ data collected were gender, birth weight, antiretroviral drug regimen from birth, anti-HIV testing date and history of hospitalization.

This study was approved by the Ethics Committee for Human Research of the Faculty of Medicine at Siriraj Hospital, Mahidol University in January 2015.

Back to Top | Article Outline

Statistical Analysis

Descriptive statistics were calculated as percentages for discrete data and medians with interquartile ranges (IQR) for continuous data. The χ2 statistic was used to analyze the proportion of seroreversion by 12, 18 and 24 months of age in each period. The Mann-Whitney U test or the Kruskal-Wallis test was applied for univariate analysis. Backward stepwise multivariate logistic regression was used to explore relationships between maternal and infant variables, seroreversion at 12 months of age and the performance of the 3G anti-HIV test kit. To avoid the collinearity of potential overlapping factors, the tolerance of more than 80% was used as the cut-off of independency. Data analysis was performed using IBM SPSS Statistics 20.0 (SPSS Inc., Chicago, IL).

Back to Top | Article Outline


Of the 1356 HIV-exposed infants born between January 2000 and December 2014, 1086 (80.1%) returned to the clinic for follow-up. Sixty-seven infants (6.2%) were HIV-infected diagnosed by 2 or more positive HIV DNA polymerase chain reaction (PCR) assays, and ART was promptly started. There were 286 (26.3%) infants lost to follow-up before anti-HIV seroreversion was documented, and they were excluded from this study. Of the 736 infants (4 twins) included in this study, the characteristics of 732 mothers and infants are shown in Table 1. In the periods of 2007–2009 and 2010–2014, 3-drug combinations ART were prescribed to 47.2% and 90.8% of the mothers, respectively. The regimens prescribed were non-nucleoside reverse transcriptase inhibitor–based ART in 36.4% and 28.2%; and LPV/r-based ART in 10.8% and 62.6% in the period of 2007–2009 and 2010–2014, respectively. HIV infection was newly diagnosed at the time of antenatal care in 540 (73.8%) mothers. Eighty-two (11.2%) mothers did not receive antenatal ART. Seventy-five (10.3%) infants were born before 37 weeks gestation, and 77 (10.5%) were hospitalized during the neonatal period due to non-HIV–related conditions.



Of the overall 736 infants tested anti-HIV serology at 12 months of age, 696 were positive and were retested at 18 months of age, and 40 infants who remained positive were retested at 24 months of age. The proportion of infants that tested anti-HIV antibody negative at 12 (median, 12.1; IQR, 11.9–12.3) months, 18 (median, 17.8; IQR, 16.7–18.2) months and 24 (median, 20.2; IQR, 19.2–23.9) months are given in Figure 2. The rate of seroreversion at 12 months was 84% in 2000 and decreased to 65% in 2009 and continuously decreased from the year 2010 to 40% in 2014 (Fig. 2). The proportion of anti-HIV seroreversion by 12 months of age declined from 68% in 2000–2006 to 65.9% in 2007–2009 and then to 42.9% in 2010–2014 (P < 0.001; Fig. 3). In the year 2000–2012 in which 3G test kit was used, the proportion of seroreversion by 18 months of age was 91% and significantly decrease to 67.3% in 2013–2014 when the test kit was changed to 4G assay (P < 0.001). All infants had seroreversion by 24 months of age.





In univariate analysis, infants born to the mothers ≥35 years of age [P = 0.021; crude odds ratio (OR), 0.45; 95% confidence interval (CI): 0.23–0.88], who received antepartum PI-based treatment (P < 0.002; crude OR, 0.42; 95% CI: 0.24–0.72), who received PI-based treatment more than 3 months (P < 0.001; crude OR, 0.40; 95% CI: 0.30–0.60) and infants born in the year 4G testing was used (P < 0.001; crude OR, 0.32; 95% CI: 0.19–0.51) were less likely to serorevert by 12 months of age. Other variables such as maternal CD4, maternal history of opportunistic infection, mode of delivery, infant gender, infant birth weight, infant ART and infant history of hospitalization were not associated with seroreversion at the 12 months of age (P > 0.05 for all). In multivariate analysis, the variables included were maternal age, maternal antepartum ART regimen and year of birth according to the HIV test kit used. Anti-HIV seroreversion at 12 months was inversely associated with maternal PI-based treatment (P = 0.0331; adjusted OR, 0.65; 95% CI: 0.43–0.97) and the years of birth that use 4G test kit (P = 0.002; adjusted OR, 0.42; 95% CI: 0.24–0.72; Table, Supplemental Digital Content 1,

Back to Top | Article Outline


Anti-HIV seroreversion has been used to exclude HIV infection in infants born to HIV-infected mothers, particularly in settings where HIV DNA PCR is not available or when virologic laboratory quality control is not well established. Early diagnosis and exclusion of HIV infection are critical to guide the treatment of the infants and family, as well as to the effectively evaluate the National prevention of mother-to-child transmission program. This is one of the few studies to determine the time of seroreversion in perinatally HIV-exposed uninfected infants in Asia. We found that the proportion of anti-HIV seroreversion at 12 months of age decreased from 84% in the year 2000 to 40% in 2014, with the overall rate of 67% before and 43% after the year 2010 at which the National prevention of mother-to-child transmission first-line regimen was changed to AZT plus 3TC plus LPV/r in all HIV-infected pregnant women in second trimester onward. The maternal exposure to PI-based regimen and the use of 4G testing were found to be the independent factors inversely associated with seroreversion at 12 months of age. The implementation of the 4G ELISA test in the year 2013 occurred in the absence of other changes in practice or guidelines. The use of 4G assay was associated with a decrease in seroreversion rate at 18 months of age, from 96% before the year of use of the assay to 79% after its introduction (P < 0.001). All infants seroreverted by 24 months of age.

In the settings where HIV DNA PCR is not performed in early infancy, World Health Organization recommends provision of RDT at 9 months to rule out HIV infection for those HIV-exposed infants.3 If the infants with negative RDT are still breastfeeding, a virologic testing is required at the end of breastfeeding. In nonbreastfed infants, a negative RDT result can rule out HIV infection while a positive finding requires further virologic testing. In Thailand and other settings in Asia, anti-HIV seroreversion has been recommended to confirm that these HIV-exposed infants are not infected before discharge from the HIV clinic. This is helpful as infants may have not been tested by PCR or were tested by an uncertified laboratory. Compared with PCR assays, anti-HIV serology is less technically demanding and can be performed at a lower cost with high reliability in most laboratories.

A recent US study in 2000–2007 found that after introduction of triple combination ART, the median age of seroreversion in HIV-exposed uninfected infants was 13.99 months while seroreversion was 86% at 18 months and 95.7% at 21 months.9 This and another study in Vietnam8 also found an association between maternal exposure to PI with later age of seroreversion in infants. We did not perform frequent serologic testing to determine the precise timing of seroreversion, but we can expect that a majority of infants seroreverted between 12 and 18 months of age. Similarly, we found that maternal PI-based regimen treatment was inversely associated with seroreversion at 12 months of age.

The causes of delayed seroreversion remain unclear. It could be from higher level of maternal antibody transferred to the infants or reduction of antibody decay rate. A systematic review of measles maternal antibody revealed that the placental transfer rate was more important affecting the infant antibody level than the relatively stable antibody decay rate. The lower placental transfer rates were found in less gestational age, HIV infection and malaria.16 Previous study suggested that maternal immunologic status may impact on the time to seroreversion.11 The infants of AIDS mothers had relatively shorter time to seroreversion, and antenatal combination ART promotes increased immunity resulting in higher levels of anti-HIV antibody transfer to the infant and delayed clearance.10 , 11 The PI-based ART causes great reduction HIV VL and may affect placental transfer maternal antibody. A study in the United States9 also described an earlier age of seroreversion among infants with higher IgG levels at 3–7 months of age and a higher rate of decrease of IgG 6 months later and proposed that this was the result of increased consumption of IgG due to infections during the first 15 months of life. However, other studies in the United States and South Africa have not shown an association between maternal antibody titers and infant antibody.6 , 17 Our study and the Malawi study10 did not find an association between delayed seroreversion and maternal HIV disease stage.

It was expected that the 4G ELISA would not change the sensitivity in detecting anti-HIV antibodies but rather improved the sensitivity of detecting HIV infection from additional detection of HIV antigen during acute infection. On the other hand, another study showed that the 3G test kit may be able to detect anti-HIV serology during the fading period better than 4G test due to the greater breadth of antigens presented in the 3G test kit.18 The results of our study were found that using 4G test (2013–2014) might increase the sensitivity of detecting low level of waning anti-HIV antibody and resulted in delayed seroreversion. Of note was that despite the wide use of the LPV/r-based regimen from the year 2010, 98% of our uninfected infants seroreverted by 18 months of age before the use of 4G test kit. With the change to 4G, these HIV-uninfected infants were found less likely to be seronegative; only 79% by 18 months of age.

Our study had identifiable limitations. First, there was a 19% drop-out rate that may have limited our ability to detect significant effects of some variables on seroreversion. We did not systematically collect the data of these infants over 14 years and therefore unable to evaluate any temporal differences and the potential selective bias. We were not able to determine the exact age of seroreversion because testing was performed every 6 months. Only a third of the mothers had available VL data. And finally, we did not quantify the infants’ and maternal antibody titers and were therefore unable to correlate antibody levels in mothers and infants at time of delivery. Despite these limitations, we are able to demonstrate the increasing age at seroreversion in infants over the time during 2000–2014.

The findings from our study have important practical implications. In our study, only 40% of the infants born in the most recent years seroreverted by 12 months of age, and about 80% reverted by 18 months of age. In the settings where serology is used to exclude HIV infection in exposed, nonbreastfed infants, the test should be performed no earlier than 18 months of age. However, it is unclear whether these data can be generalized in breastfed infants on different ART regimens. Delayed seroreversion underscores the importance of access to virologic testing particularly in seropositive infants at 18 months. The current World Health Organization recommendation to use anti-HIV serology for diagnosis in children 18 months or older is not perfect, but testing at 9 months of age, particularly using 4G test kit, is perhaps of little benefit. Local guidelines using anti-HIV antibody testing for exclusion or making diagnosis of perinatal HIV infection at 18 months of age should interpret the results with cautions.

Back to Top | Article Outline


1. Panel on Antiretroviral Therapy and Medical Management of HIV-Infected Children. Guidelines for the Use of Antiretroviral Agents in Pediatric HIV Infection. Available at: Accessed November 10, 2017.
2. Schneider E, Whitmore S, Glynn KM, et al; Centers for Disease Control and Prevention (CDC). Revised surveillance case definitions for HIV infection among adults, adolescents, and children aged <18 months and for HIV infection and AIDS among children aged 18 months to <13 years–United States, 2008. MMWR Recomm Rep. 2008;57:1–12.
3. World Health Organization Consolidated guidelines on the use of antiretroviral drugs for treating and prevention HIV infection. 2016. Geneva, Switzerland: World Health Organization; Available at: Accessed September 10, 2017.
4. Simpson BJ, Andiman WA. Difficulties in assigning human immunodeficiency virus-1 infection and seroreversion status in a cohort of HIV-exposed in children using serologic criteria established by the Centers for Disease Control and Prevention. Pediatrics. 1994;93:840–842.
5. European Collaborative Study. Mother-to-child transmission of HIV infection. Lancet. 1988;332:1039–1043.
6. Chantry CJ, Cooper ER, Pelton SI, et al. Seroreversion in human immunodeficiency virus-exposed but uninfected infants. Pediatr Infect Dis J. 1995;14:382–387.
7. Moodley D, Bobat RA, Coutsoudis A, et al. Predicting perinatal human immunodeficiency virus infection by antibody patterns. Pediatr Infect Dis J. 1995;14:850–852.
8. Sohn AH, Thanh TC, Thinh le Q, et al. Failure of human immunodeficiency virus enzyme immunoassay to rule out infection among polymerase chain reaction-negative Vietnamese infants at 12 months of age. Pediatr Infect Dis J. 2009;28:273–276.
9. Gutierrez M, Ludwig DA, Khan SS, et al. Has highly active antiretroviral therapy increased the time to seroreversion in HIV exposed but uninfected children? Clin Infect Dis. 2012;55:1255–1261.
10. Gulia J, Kumwenda N, Li Q, et al. HIV seroreversion time in HIV-1-uninfected children born to HIV-1-infected mothers in Malawi. J Acquir Immune Defic Syndr. 2007;46:332–337.
11. Alcântara KC, Pereira GA, Albuquerque M, et al. Seroreversion in children born to HIV-positive and AIDS mothers from Central West Brazil. Trans R Soc Trop Med Hyg. 2009;103:620–626.
12. The Italian Register for HIV Infection in Children. Combined antiretroviral therapy reduces hyperimmunoglobulinemia in HIV-1 infected children. AIDS. 2004;18:1423–1428.
13. Bunders M, Pembrey L, Kuijpers T, et al. Evidence of impact of maternal HIV infection on immunoglobulin levels in HIV-exposed uninfected children. AIDS Res Hum Retroviruses. 2010;26:967–975.
14. Morris L, Binley JM, Clas BA, et al. HIV-1 antigen-specific and -nonspecific B cell responses are sensitive to combination antiretroviral therapy. J Exp Med. 1998;188:233–245.
15. Lolekha R, Boonsuk S, Plipat T, et al. Elimination of mother-to-child transmission of HIV—Thailand. MMWR Morb Mortal Wkly Rep. 2016;65:562–566.
16. Cáceres VM, Strebel PM, Sutter RW. Factors determining prevalence of maternal antibody to measles virus throughout infancy: a review. Clin Infect Dis. 2000;31:110–119.
17. Sherman GG, Jones SA. Oral fluid human immunodeficiency virus tests: improved access to diagnosis for infants in poorly resourced prevention of mother to child transmission programs. Pediatr Infect Dis J. 2005;24:253–256.
18. de Souza MS, Pinyakorn S, Akapirat S, et al; RV254/SEARCH010 Study Group. Initiation of antiretroviral therapy during acute HIV-1 infection leads to a high rate of nonreactive HIV serology. Clin Infect Dis. 2016;63:555–561.

seroreversion; HIV antibody; HIV-exposed uninfected infant; diagnosis; prevention of mother-to-child transmission; Thailand

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.