High Incidence of Invasive Group B StreptococcalInfections in HIV-Exposed Uninfected Infants

WHAT’S KNOWN ON THIS SUBJECT: Several recent articles fromless affluent countries have shown an increased susceptibility toinfection in HIV-exposed uninfected infants.

WHAT THIS STUDY ADDS: This is the first report of increasingsusceptibility to infection in a population of exposed uninfectedinfants from developed countries.

abstractOBJECTIVES: The occurrence of an unusual number of group B strep-tococcal (GBS) infections in HIV-exposed uninfected (HEU) infants whowere followed in our center prompted this study. The objective of thisstudy was to describe and compare the incidence and clinical presen-tation of GBS infections in infants who were born to HIV-infected and-uninfected mothers.

METHODS: All cases of invasive GBS infections in infants who wereborn between 2001 and 2008 were identified from the database of HEUinfants and from the microbiology laboratory records. The medicalcharts of all infants with GBS infection were reviewed.

RESULTS: GBS invasive infections were described for 5 (1.55%) infantswho were born to 322 HIV-infected mothers who delivered in our cen-ter. The incidence of GBS infections during the same period was 16(0.08%) of 20 158 infantswhowere born to HIV-uninfectedmothers. OneHEU infant presented a recurrent infection 28 days after completion oftreatment for the first episode. Late-onset infection wasmore frequentin HEU infants (5 of 6 vs 2 of 16 episodes in the control population). Thediseases were also more severe in HEU infants with 5 of 6 sepsis orsepsis shock in HEU infants versus 10 of 16 in control subjects, andmost HEU infants had leukopenia at onset of infection.

CONCLUSIONS: The incidence of GBS infection was significantly higherin HEU infants than in infantswhowere born to HIV-uninfectedmothers.These episodes of GBS sepsis in HEU infants were mostly of late onsetand more severe than in the control population, suggesting an in-creased susceptibility of HEU infants to GBS infection. Pediatrics 2010;126:e631–e638

AUTHORS: Cristina Epalza, MD,a Tessa Goetghebuer, MD,a

Marc Hainaut, MD,a Fany Prayez, MD,a Patricia Barlow,MD,b Anne Dediste, MD,c Arnaud Marchant, MD, PhD,d andJack Levy, MD, PhDa

aPediatric Department, bObstetrical Department, andcMicrobiological Department, St Pierre University Hospital,Brussels, Belgium; and dInstitute for Medical Immunology, FreeUniversity of Brussels, Brussels, Belgium

KEY WORDSHIV, streptococcal infection, neonatal infection, HIV-exposeduninfected infants

ABBREVIATIONSHAART—highly active antiretroviral therapyMTCT—mother-to-child transmissionHEU—HIV-exposed uninfectedARV—antiretroviralGBS—group B streptococcalCSF—cerebrospinal fluidGA—gestational ageSGA—small for GARR—risk ratioLOD—late-onset diseaseEOD—early-onset diseaseNRTI—nucleoside reverse transcriptase inhibitor

Drs Epalza and Goetghebuer contributed equally to this work.

www.pediatrics.org/cgi/doi/10.1542/peds.2010-0183

doi:10.1542/peds.2010-0183

Accepted for publication May 28, 2010

Address correspondence to Tessa Goetghebuer, MD, PediatricDepartment, St Pierre University Hospital, 322 rue Haute, 1000Brussels, Belgium. E-mail: tessa_goetghebuer@stpierre-bru.be

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2010 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they haveno financial relationships relevant to this article to disclose.

ARTICLE

PEDIATRICS Volume 126, Number 3, September 2010 e631 by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

Highly active antiretroviral therapy(HAART) has markedly improved thehealth and the long-term prognosis ofHIV-infected patients and reduced therisk for mother-to-child transmission(MTCT) of the virus.1 This has led to anincrease in the number of HIV-exposeduninfected (HEU) infants.2

The potential adverse effects of mater-nal disease and/or exposure to the an-tiretroviral (ARV) agents that are usedfor prevention of MTCT on the outcomeof pregnancy and on the health of HEUinfants have been actively investi-gated. Although a limited number ofstudies that were conducted beforethe use of MTCT prophylaxis suggestedthat HEU infants had a similar outcometo children born to HIV-uninfectedwomen,3 more recent studies from de-veloping countries showed a highermorbidity and mortality risk in HEU in-fants compared with infants born toHIV-uninfected mothers,4–8 mostly as aresult of infectious diseases. Thesestudies, most of them conducted insub-Saharan Africa, where access toMTCT prophylaxis is not generalized,showed a correlation between mortal-ity risk in HEU infants and advancedmaternal disease.5,9

With the widespread use of ARV agentsduring pregnancy, studiesmainly frommore affluent countries have at-tempted to evaluate potential adverseeffects of these drugs on the fetus orthe infant. An association between theuse of HAART during pregnancy and anincreased rate of preterm birth hasbeen described1,10,11 but not confirmedin another cohort.12 The incidence ofcongenital malformations observed inHEU infants is similar to that of the gen-eral population,13 and a transient ef-fect of ARV agents on growth duringthe first year of life has been observedin several large cohorts.14,15 An in-creased incidence of febrile convul-sion in early life16 and of mitochondrialdisorders that affect severely the neu-

rologic system17–20 have been de-scribed in HEU infants. Although re-versible anemia is generally observedduring the postnatal zidovudine pro-phylaxis,21–23 other blood lineages arealso affected in infantswho are born toHIV-infected mothers, possibly associ-ated with ARV exposure,21–24 withoutany link to clinical manifestations.

We have observed the occurrence ofan unusual number of severe group Bstreptococcal (GBS) infections duringthe first months of life in the cohort ofHEU infants who were followed up in ourcenter. In this report, we describe thecharacteristics and incidence of theseinfections in HEU infants and comparethem with GBS infections that occurredduring the same period in the childrenwho were born in the same hospital toHIV-uninfected mothers.

METHODS

Setting of the Study

Saint-Pierre Hospital is a university-affiliated hospital and referral centerfor HIV-affected children and adults lo-cated in downtown Brussels. The hos-pital has a largematernity departmentand a tertiary neonatal center.

Local Guidelines and Procedures

Prevention of MTCT of HIV and Follow-up of Exposed Infants

Children who are born to HIV-infectedwomen are followed up during the firstyears of life to diagnose HIV transmis-sion and to detect possible adverse ef-fects of ARV prophylaxis. A child is con-sidered to be uninfected when RNApolymerase chain reaction (threshold�50 copies per mL; Amplicor HIV-1 Mon-itor test [Roche, Basel, Switzerland]) orDNA polymerase chain reaction is nega-tive in at least 2 peripheral blood sam-ples. Preventive measures to reduce therisk for MTCT have been implementedover time according to published guide-lines as previously described.25

Prevention and Definition of NeonatalGBS Infections

Interventions to prevent neonatal GBSinvasive infections are implemented inthe obstetric and neonatal depart-ments according to internationalguidelines26 endorsed by the BelgianSuperior Health Council.27

Outcome Definition

GBS invasive infection was defined asthe isolation of the microorganismfrom blood or cerebrospinal fluid(CSF) in infants who were�90 days ofage.

Study Populations

The medical charts of all HEU infantswho were born between January 1,2001, and December 31, 2008, were re-viewed to identify GBS infection. Forcomparison of the characteristicsof GBS infection in HEU infants withthose of infants who were born to HIV-uninfected mothers, the microbiologiclaboratory records were searched toidentify all invasive GBS infections thatoccurred in infants who were born inthis hospital during the same period.For the comparison of GBS infectionamong HEU infants and the control in-fants, only the children who were bornin Saint-Pierre Hospital were consid-ered. The total number of births duringthe study period was extracted fromthe birth register. This study was ap-proved by the hospital’s ethics com-mittee.

Data Collection

The medical chart of all children whopresented a GBS infection and theirmother were reviewed to collect ageand ethnicity of the mother, mode ofdelivery, GBS carriage and use of anti-biotic prophylaxis. In addition, for HIV-infected mothers, date of HIV diagno-sis, ARV prophylaxis during pregnancyand labor, HIV viral load, and CD4� cellcount around delivery were collected.

e632 EPALZA et al by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

Data that were collected on infants whopresented GBS infection included gesta-tional age (GA); birthweight; presence ofGBS in cutaneous smear taken from ax-illary region, anal region, or external au-ditory canal; admission to the NICU; typeof feeding; andneonatal ARVprophylaxis.Infants who were small for GA (SGA)were defined as having a birth weightless than the third percentile accordingto Gairdner and Pearson.28

The data concerning GBS infectionswere the type of infection, age at onset,number of GBS-positive culture sites,full blood count at onset, antibiotictreatment, and evolution. GBS infec-tions were divided into early-onset(0–6 days of life) and late-onset (7–90days of life) infection.29–31 Severity wasevaluated according to internationaldefinitions.32

Statistical Analysis

Statistical analysis was performed byusing Stata 8.0 software, StatXact-9,and EpiInfo 3.5.1. Incidence risk ratio(RR) was computed according to theHIV status of the mother. �2 and Fish-er’s exact tests were used to comparethe proportions. Statistical signifi-cance was assigned by a 2-sided �level of .05. All P values are 2-tailed.

RESULTS

Characteristics of GBS Infection inHEU Infants

From January 2001 to December 2008,403 infants who were born alive to HIV-infected women were followed up inour center; 6 of them had HIV infection(Fig 1). The study population includes397 uninfected infants (including 3pairs of twins) who were born to 394mothers. Seventy-five mothers deliv-ered in another facility, and their in-fants were referred to our center forpostnatal care. Among the 394 moth-ers, 88% initiated HAART prophylaxisduring pregnancy, 33% had elective ce-sarean delivery, and 17% had emer-

gency cesarean delivery. Eighty-eightpercent of the mothers were of sub-Saharan African origin; most of themarrived in Belgium in the previousyears and were living in poor socioeco-nomic conditions.

Seven episodes of neonatal invasiveGBS infections were documented in 6of 397 infants (1 infant had 2 episodes).All of these infants were confirmed tobe HIV-uninfected at follow-up. Asshown in Table 1, the mothers of 5 ofthese 6 infants originated from sub-Saharan Africa, and the majority re-ceived the diagnosis of HIV before thepregnancy. Two mothers delivered bycesarean delivery (1 emergency and 1elective). Five mothers received HAARTduring pregnancy: 1 during the wholepregnancy, 2 from the second trimes-ter, and 2 during the third trimester.The most prescribed drugs werezidovudine (2 of 6 mothers), lamivu-dine (3 of 6 mothers), and lopinavir/ritonavir (2 of 6 mothers); 5 of 6 moth-ers received intravenous zidovudineduring labor. Viral load during thethird trimester was undetectable(�50 copies per mL) in 1, between 50and 1000 copies per mL in 4, and un-available for 1. The CD4� cell countduring the third trimester was �200cells per �L in 1 case (98 cells/�L).

As shown in Table 1, 4 (67%) of 6 in-fants were female; the median GA was35 weeks; and median birth weightwas 2380 g, with 3 (50%) of 6 beingSGA. Four (67%) of 6 infants were ad-mitted to the NICU: 3 for preterm birthand 1 for sepsis. None of the infantswas breastfed, and all received neona-tal ARV prophylaxis for a median of 37days.

GBS was isolated from 5 of 6 mothers’vaginal swabs and from 2 of 6 infants’cutaneous swabs at birth. Among the 7episodes of invasive GBS infections, 5were late-onset diseases (LOD; oc-curring at 9, 26, 33, 64, and 72 days),2 presented as septic shock, 4 pre-sented as sepsis, and 1 presented asbacteremia.32 Two infants had ameningitis-associated infection (GBSisolated from the CSF). One infant(patient 4, Table 1) presented 2 epi-sodes of GBS infection; the second ep-isode occurred 28 days after the 10-day ampicillin treatment of the firstepisode was discontinued. In 5 of 7 ep-isodes, infants had a white blood cellcount �5000/�L (range: 1350–4300/mm) at onset of infection.

All infants were treated by intravenousantibiotics. The evolution was favor-able for 5 patients, but 1 patient died.

403 infants born to HIV-infected mothers

6 HIV-infected infants 397 HIV-uninfected infants

75 infants born in another hospital 322 infants born in St Pierre hospital

1 child with GBS infection 5 children with GBS infections

6 children with GBS infections

FIGURE 1Flow diagram for the occurrence of GBS infection in the cohort of infants who were born to HIV-infected mothers.

ARTICLE

PEDIATRICS Volume 126, Number 3, September 2010 e633 by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

None of the patients was admitted atthe same time.

Comparison of GBS Infection inHEU Infants and in Infants WhoWere Born to HIV-UninfectedMothers

The review of the St Pierre Hospital’smicrobiology laboratory records iden-tified 21 infants who were born be-tween 2001 and 2008 and had GBS iso-lated in a blood culture: 16 of themwere born to HIV-uninfected mothers(referred to as control population) and5 to HIV-infectedmothers (patients 1, 3,4, 5, and 6 in Table 1).

The characteristics of the inborn in-fants who presented GBS infectionsare summarized in Table 2. The mater-

nal origin was significantly differentbetween the 2 groups: 80% of the HIV-infected mother were of sub-SaharanAfrica origin. Preterm births and in-fants whowereSGAweremore frequentamong HEU infants than in the controlpopulation, although very low birthweightpreterm infantswereseenonly inthe latter. There were significantly morecesareandeliveries inHIV-infectedmoth-ers than in the control group. Noneof theHEU infants were breastfed, comparedwith 10 (63%) of 16 infants in the controlpopulation.

The comparison of the GBS infectionsis shown in Table 3. Infections tendedto be more severe in HEU infants thanin the control population. GBSmeningi-tis (2 cases) and recurrence (1 case)

TABLE1CharacteristicsofMaternalHIVInfection,Deliveries,andClinicalPresentationofGBSInvasiveInfectionin6HEUInfants

PatientGenderGA,wkBirth

Weight,gModeofDelivery

Ageat

Onset,d

TypeofInfection

WBC/�L

atOnsetSkinSmearMother

SmearAntibiotics

inLaborHAARTStartedin

Pregnancy

VLThirdTrimester,

copiespermL

CD4Third

Trimester

NeonatalART(d)

1F

33a2360

Vaginal

0Sepsis

37000GBS

GBS

Yes

NoNA

362

ZDV(63)

2bF

38a2400cVaginal

1Sepsis

4090GBS

Negative

NoFirstTrimester

�50

448

ZDV(10)

3M

372600

Vaginal

9Sepsis,meningitis

2710Negative

GBS

NoThirdTrimester

104

422

NVP(1)/3TC(7)ZDV,(35)

4F

382830

Electivecesarean

26Sepsis

4300Negative

GBS

Yes

SecondTrimester

458

373

ZDV(36)

delivery

64Bacteremia

11000

5F

32a1700cEmergencycesarean

delivery

33Septicshock,meningitis,

death

1350LactobacillusGBS

Yes

SecondTrimester

160

580

ZDV(38)

6M

31a1350cVaginal

72Septicshock

2350Negative

GBS

Yes

ThirdTrimester

458

98ZDV(42)

WBCindicateswhitebloodcells;ART,antiretroviraltherapy;VL,HIVviralload;NA,notavailable;ZDV,zidovudine;NVP,nevirapine;3TC,lamivudine.

aNICUadmission:patients1,5,and6becauseofprematurity;patient2becauseofasphyxia;andpatient3becauseofsuspectedsepsis.

bOutborn(thispatientwasnotincludedinthecomparisonwiththecontrolpopulation).

cSGA(weightlessthanthirdpercentilefortheage).

TABLE 2 Demographic, Obstetric, and Clinical Characteristics in Infants who were Born to HIV-Infected and -Uninfected Mothers and had a GBS Invasive Infection

Parameter Infants Born to HIV-Positive Mothers

Infants Born to HIV-Negative Mothers

Pa

Total No. of infants 5 16Female gender, n (%) 3 (60) 8 (50) .13Sub-Saharan African origin, n (%) 4 (80) 3 (19) .03Preterm (�37 wk), n (%) 3 (60) 4 (25) .28GA, range, wk 31–33 26–33SGA, n (%) 2 (40) 2 (13) .23GBS� skin smears, n (%) 1 (20) 9 (57) .31GBS� maternal vaginal smears, n (%) 5 (100) 14 (88) .99Antibiotics during labor, n (%) 4 (80) 9 (69)b .99Cesarean delivery, n (%) 2 (40) 1 (6) .13a Fisher’s exact test.b Three missing values.

TABLE 3 Comparison of GBS Infections in Infants who were Born to HIV-Infected and -UninfectedMothers

Parameter HEUa Infants Born to HIV-Negative Mothers

Pb

No. of episodes of GBS infection 6 16Type of infection, n (%)Septic shock 2 (33) 2 (13)Sepsis 3 (50) 8 (50) .587Bacteremia 1 (17) 6 (37)

ParticularsMeningitis associated 2 (33) 0 (0) .070

LOD 5 (83) 2 (13) .004Age at LOD, d 9, 26, 33, 64, 72 19, 55WBC count at onset, /�/L

�5000 4 (66) 2 (13) .0595000–15000 1 (17) 5 (31)�15000 1 (17) 9 (56)

WBC indicates white blood cells.a Only children who were born in St Pierre Hospital were considered in this analysis.b Fisher’s exact test.

e634 EPALZA et al by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

were observed only in the group of HEUinfants. There were significantly moreLODs (P � .002) and more leukopeniaat onset (white blood cell count�5000/�L; P� .059) in the HEU infantsthan in control subjects.

Although the mortality rate was simi-lar in the 2 groups, the characteristicsof the infants who died differed. TheHEU infant was born at 32 weeks’ ges-tation, was admitted to the NICU, andhad an uneventful hospital course untilthe 33rd day of life, when she devel-oped GBS septic shock and meningitisthat led to death 5 days later (patient 5,Table 1). The 2 infants from the controlgroup both were very low birth weightpreterm infants, born at 26 and 27weeks’ gestation, who died duringtheir first week of life, with severe re-spiratory and neurologic problems.

Incidence Risk for GBS Infection inHEU Infants and in Infants WhoWere Born to HIV-UninfectedMothers

As shown in Table 4, the incidence riskfor GBS infections was 1.55% in HEUinfants and 0.08% in the control infantswhowere born to HIV-uninfectedmoth-ers in the same hospital during thesame period. This significant differ-ence is mostly attributable to thehigher incidence of LOD in HEU infants:1.24% vs 0.01% in the control group(RR: 125.2 [95% confidence interval:26.3–620.2]; P� .0001).

DISCUSSION

This is the first report to describe ahigh risk for GBS infection in a cohort

of HEU infants. This report is basedon the compilation of GBS infectionsobserved during the prospectivefollow-up of infants who were born toHIV-infected mothers in 1 single cen-ter. The follow-up of these children var-ies in length, but all were followed upuntil at least the 90th day of life, exclud-ing reporting bias of GBS infections inHEU infants. For comparison of therate of GBS infections in HEU infantswith that of the population of infantswhowere born to HIV-uninfectedmoth-ers, all episodes of GBS infection wereretrieved from the hospital microbiol-ogy laboratory records, and only theinborn children were included in thisanalysis. A reporting bias in the inci-dence risk for late-onset GBS infectionin infants who were born to HIV-uninfected mothers cannot be ex-cluded, because these infectionsmighthave been diagnosed in another hospi-tal; however, the incidence rate that wasobserved in this control cohort is similarto the most recent incidences reportedin the literature in settings where intra-partum antibiotic prophylaxis is used:0.05% to 0.09%, 0.04% to 0.05%, and0.02% to 0.04% for the overall GBS neo-natal infections, early-onset disease(EOD), and LOD, respectively.30,33,34

There are several possible explana-tions for the higher susceptibility ofHEU infants to GBS invasive infection.HIV-infected mothers might have ahigher rate of colonization with GBS orcould be colonized by a higher numberor by more virulent organisms thanHIV-uninfected mothers. The risk for

EOD is clearly related to the carriage ofGBS in the mother’s genital tract, withtransmission occurring during labor,whereas LOD is caused by vertically orhorizontally acquired organisms.29–31

Some studies, although in differentpopulations, reported that black eth-nicity is associated with a higher car-riage of GBS in women35 and with ahigher risk for invasive GBS infectionin neonates than white ethnicity.34

Most HIV-infectedmothers in our studyoriginated from sub-Saharan Africa,whereas most control subjects arewhite. In our center, the GBS carriageis higher in HIV-infected than in-uninfected women: 29% vs 19%, re-spectively (P � .001; unpublisheddata). The ethnicity and/or the highercarriage rate of HIV-infected womenmight contribute to the difference ob-served, but it is unlikely to account fora RR of 19. Whether the serotypes ofthe strains carried by HIV-infectedwomenweremore virulent should alsobe questioned. Indeed, serotype distri-bution varies according to the site ofinfection (blood versus CSF) and thetime of infection (EOD versus LOD) andcan change according to geographyand time. One European study re-ported that the percentage of neonatalinfections that are caused by serotypeIII is increasing over time, mainly inLOD and meningitis.36 Unfortunately,the strains that were isolated fromHEU infants and from control infantscould not be typed retrospectively. Pre-maturity is a major risk factor forearly-onset GBS infection34; however, itis unlikely that prematurity influencedthe increased incidence risk for GBSinfection in HEU infants in this study,because the rate of prematurity is sim-ilar in HEU infants to that of the controlpopulation (10.7% births�37weeks inHEU infants versus 10.9% in the popu-lation of children who were born toHIV-uninfected mothers in St PierreHospital, a referral center for compli-cated pregnancy).

TABLE 4 Risk for GBS Infection Among Infants who were Born in St Pierre University Hospital(2001–2008)

Parameter HEU (n� 322),n (%)a

Infants Born to HIV-NegativeMothers (n� 20 158), n (%)

RR 95% CIb

All GBS infections 5 (1.55) 16 (0.08) 19.6 7.5–51.7Early-onset infections 1 (0.31) 14 (0.07) 4.5 0.3–26.7Late-onset infections 4 (1.24) 2 (0.01) 125.2 26.3–620.2

CI indicates confidence interval.a Only children who were born in St Pierre hospital were considered in this analysis.b Fisher’s exact test.

ARTICLE

PEDIATRICS Volume 126, Number 3, September 2010 e635 by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

There are to date a limited number ofreports suggesting an increased sus-ceptibility of HEU infants to infection.Studies in Africa have reported an in-verse correlation between highermor-tality in HEU infants and maternalCD4� cell count at delivery5,9,37 and in-creased mortality and morbidity com-pared with their HIV-unexposed coun-terparts.5,38 A prospective evaluationfrom Latin America and the Caribbeanregion demonstrated a high incidenceof infection (mostly skin or mucousmembrane and respiratory tract infec-tions) in HEU infants during the first 6months of life, with 26% of neonatalinfections.6 A prospective study con-ducted in Rwanda, before the use ofARV prophylaxis, reported a highernumber of severe pneumonia, general-ized dermatitis, and chronic parotidi-tis in HEU infants.3 The Women and In-fants Transmission Study (WITS) andother studies from Africa describedthe occurrence of several severe infec-tions in HEU infants, including dissem-inated tuberculosis, Pneumocystispneumonia, and severe pneumoniafrom other causes.4,5,8,15,39–41 A recentarticle from South Africa focused onthe occurrence of severe infectionssuggestive of immunodeficiency, suchas Pneumocystis jiroveci pneumonia,pseudomonas sepsis, cytomegalovi-rus colitis, and group A streptococcalmeningitis.4

Whether the increased risk for infec-tion described in HEU infants in thesestudies and in our observation islinked to environmental conditions orto a possible immune dysfunction isunknown. A number of studies have re-ported an impact ofmaternal HIV infec-tion on the infants’ passive and innate

immunity. Antibody transplacentaltransfer has been shown to be re-duced during pregnancy in HIV-infected women.42 There are sugges-tions that an immunodeficient mothermay have defective transplacentaltransfer of hematopoietic cytokines,resulting in lower thymic output ofCD4� cells in the child.4,9,43,44 A reduc-tion of interleukin 12 cytokine produc-tion (intermediate profile between HIV-infected infants and control subjects)has been observed in cord bloodmononuclear cells of HEU infants,45 in-ducing potentially delay in immune cel-lular maturation or deficit in antigen-presenting cells.46–48 The hypothesis oftransplacental diffusion of viral solu-ble products or defective virus-causing immune paresis43,49,50 is sup-ported by the demonstration ofproduction of interleukin 2 by periph-eral blood cells after stimulation by HIVantigens and the detection of HIV-specific T helper and cytotoxic T cell inHEU infants that persist for severalyears.46,48–50 The possible role of peri-natal ARV exposure can also be con-sidered. Several studies showed adepressed hematopoiesis in HEUchildren, including a decrease in plate-lets, neutrophils, total lymphocytes,CD4 and CD8� cell count,21,22,24,51,52 thatmight be attributable to mitochondrialdysfunction associated with exposureto nucleoside reverse transcriptase in-hibitors (NRTIs). Indeed, NRTIs crossthe placenta53,54 and have an affinity forthe mitochondrial � polymerase. It hasbeen shown that mitochondrial DNA ofcord bloodmononuclear cells is signif-icantly reduced in HEU infants com-paredwith infants who are born to HIV-uninfected mothers,53 and the majorityof infants who are exposed to NRTI

present transient and asymptomatichyperlactatemia55,56; however, persis-tent mitochondrial dysfunction thatleads to severe neurologic impairmenthas been reported in a limited numberof HEU infants.18,20 Additional risk fac-tors of susceptibility to infection inearly life might be the absence ofbreastfeeding and the psychosocialand socioeconomic conditions of HEUinfants.

CONCLUSIONS

Our observation of increased suscepti-bility of HEU infants to GBS infection isan additional piece of evidence, thefirst from an industrialized country,that HEU infants might be more sus-ceptible to infection than infants whoare born to HIV-uninfected mothers.With the generalization of MTCT pro-phylaxis, the number of HEU infantswill continue to increase worldwide,and it is crucial to determine what thecause of this increased susceptibilityis. Prospective investigations areneeded to confirm this susceptibilityand to demonstrate possible immunedysfunction. Identification of surro-gate markers of susceptibility mighthelp to prevent early-life infectiousmorbidity in this population.

ACKNOWLEDGMENTSThis study was supported by theSmiles Foundation, Belgium. Dr March-ant is a senior research associate atthe Fonds National de la RechercheScientifique, Belgium.

We are indebted to the infants whowere included in this study and to theirmothers and to all physicians wholooked after them. We thank EdwigeHaelterman and Claire Thorne for con-tribution to the statistical analysis.

REFERENCES

1. Mother-to-child transmission of HIV infec-tion in the era of highly active antiretroviraltherapy. Clin Infect Dis. 2005;401(3):458–465

2. Agangi A, Thorne C, Newell ML. Increasinglikelihood of further live births in HIV-infectedwomen in recent years. BJOG. 2005;112(7):881–888

3. Spira R, Lepage P, Msellati P, et al. Naturalhistory of human immunodeficiency virustype 1 infection in children: a five-year pro-spective study in Rwanda. Mother-to-Child

e636 EPALZA et al by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

HIV-1 Transmission Study Group. Pediatrics.1999;104(5). Available at: www.pediatrics.org/cgi/content/full/104/5/e56

4. Slogrove AL, Cotton MF, Esser MM. SevereInfections in HIV-exposed uninfectedinfants: clinical evidence of immunodefi-ciency. J Trop Pediatr. 2010;56(2):75–81

5. Marinda E, Humphrey JH, Iliff PJ, et al. Childmortality according to maternal and infantHIV status in Zimbabwe. Pediatr Infect Dis J.2007;26(6):519–526

6. Mussi-Pinhata MM, Freimanis L, YamamotoAY, et al. Infectious disease morbidityamong young HIV-1-exposed but uninfectedinfants in Latin American and Caribbeancountries: the National Institute of ChildHealth and HumanDevelopment InternationalSite Development Initiative Perinatal Study.Pediatrics. 2007;119(3). Available at: www.pediatrics.org/cgi/content/full/119/3/e694

7. Thea DM, St Louis ME, Atido U, et al. A pro-spective study of diarrhea and HIV-1 infec-tion among 429 Zairian infants. N Engl JMed. 1993;329(23):1696–1702

8. McNally LM, Jeena PM, Gajee K, et al. Effectof age, polymicrobial disease, andmaternalHIV status on treatment response andcause of severe pneumonia in South Africanchildren: a prospective descriptive study.Lancet. 2007;369(9571):1440–1451

9. Kuhn L, Kasonde P, Sinkala M, et al. Doesseverity of HIV disease in HIV-infected moth-ers affect mortality and morbidity amongtheir uninfected infants? Clin Infect Dis.2005;41(11):1654–1661

10. Grosch-Woerner I, Puch K, Maier RF, et al.Increased rate of prematurity associatedwith antenatal antiretroviral therapy in aGerman/Austrian cohort of HIV-1-infectedwomen. HIV Med. 2008;9(1):6–13

11. Combination antiretroviral therapy and du-ration of pregnancy. AIDS. 2000;14(18):2913–2920

12. Tuomala RE, Shapiro DE, Mofenson LM, et al.Antiretroviral therapy during pregnancyand the risk of an adverse outcome. N Engl JMed. 2002;346(24):1863–1870

13. Patel D, Thorne C, Fiore S, Newell ML. Doeshighly active antiretroviral therapy in-crease the risk of congenital abnormalitiesin HIV-infected women? J Acquir ImmuneDefic Syndr. 2005;40(1):116–118

14. Hankin C, Thorne C, Newell ML. Does expo-sure to antiretroviral therapy affect growthin the first 18 months of life in uninfectedchildren born to HIV-infected women? J Ac-quir Immune Defic Syndr. 2005;40(3):364–370

15. Paul ME, Chantry CJ, Read JS, et al. Morbid-ity and mortality during the first two years

of life among uninfected children born tohuman immunodeficiency virus type1-infected women: the women and infantstransmission study. Pediatr Infect Dis J.2005;24(1):46–56

16. Landreau-Mascaro A, Barret B, Mayaux MJ,Tardieu M, Blanche S. Risk of early febrileseizure with perinatal exposure to nucleo-side analogues. Lancet. 2002;359(9306):583–584

17. Blanche S, Tardieu M, Rustin P. Mitochon-drial dysfunction and perinatal exposure toantiretroviral nucleoside analogues [inFrench]. Arch Pediatr. 2000;7(1):7–9

18. Barret B, Tardieu M, Rustin P, et al. Persis-tent mitochondrial dysfunction in HIV-1-exposed but uninfected infants: clinicalscreening in a large prospective cohort.AIDS. 2003;17(12):1769–1785

19. Blanche S, Tardieu M, Benhammou V,Warszawski J, Rustin P. Mitochondrial dys-function following perinatal exposure to nu-cleoside analogues. AIDS. 2006;20(13):1685–1690

20. Brogly SB, Ylitalo N, Mofenson LM, et al. Inutero nucleoside reverse transcriptase in-hibitor exposure and signs of possible mi-tochondrial dysfunction in HIV-uninfectedchildren. AIDS. 2007;21(8):929–938

21. Pacheco SE, McIntosh K, Lu M, et al. Effect ofperinatal antiretroviral drug exposure onhematologic values in HIV-uninfectedchildren: an analysis of the women and in-fants transmission study. J Infect Dis. 2006;194(8):1089–1097

22. Le Chenadec J, Mayaux MJ, Guihenneuc-Jouyaux C, Blanche, S. Perinatal antiretrovi-ral treatment and hematopoiesis in HIV-uninfected infants. AIDS. 2003;17(14):2053–2061

23. Bunders MJ, Bekker V, Scherpbier HJ, BoerK, Godfried M, Kuijpers TW. Haematologicalparameters of HIV-1-uninfected infantsborn to HIV-1-infected mothers. Acta Paedi-atr. 2005;94(11):1571–1577

24. Levels and patterns of neutrophil cellcounts over the first 8 years of life in chil-dren of HIV-1-infected mothers. AIDS. 2004;18(15):2009–2017

25. Goetghebuer T, Haelterman E, Marvillet I, etal. Vertical transmission of HIV in Belgium: a1986–2002 retrospective analysis. Eur J Pe-diatr. 2009;168(1):79–85

26. From the Centers for Disease Control: revi-sion of guidelines for the prevention of peri-natal group B streptococcal disease. JAMA.2002;287(9):1106–1107

27. Prevention of neonatal GBS infection. [inFrench]. Guidelines From the Belgian Supe-rior Health Council. 2003;7721:3–39

28. Gairdner D, Pearson J. A growth chart forpremature and other infants. Arch Dis Child.1971;46(250):783–787

29. Schuchat A. Epidemiology of group B strep-tococcal disease in the United States: shift-ing paradigms. Clin Microbiol Rev. 1998;11(3):497–513

30. Fluegge K, Siedler A, Heinrich B, et al. Inci-dence and clinical presentation of invasiveneonatal group B streptococcal infectionsin Germany. Pediatrics. 2006;117(6). Avail-able at: www.pediatrics.org/cgi/content/full/117/6/e1139

31. Olver WJ, Bond DW, Boswell TC, Watkin SL.Neonatal group B streptococcal disease as-sociated with infected breast milk. Arch DisChild Fetal Neonatal Ed. 2000;83(1):F48–F49

32. Goldstein B, Giroir B, Randolph A. Interna-t ional pediatr ic sepsis consensusconference: definitions for sepsis and or-gan dysfunction in pediatrics. Pediatr CritCare Med. 2005;6(1):2–8

33. Trijbels-Smeulders M, de Jonge GA,Pasker-de Jong PC, et al. Epidemiology ofneonatal group B streptococcal disease inthe Netherlands before and after introduc-tion of guidelines for prevention. Arch DisChild Fetal Neonatal Ed. 2007;92(4):F271–F276

34. Centers for Disease Control and Prevention(CDC). Trends in perinatal group B strepto-coccal disease—United States, 2000–2006.MMWR Morb Mortal Wkly Rep. 2009;58(5):109–112

35. Stapleton RD, Kahn JM, Evans LE, CritchlowCW, Gardella CM. Risk factors for group Bstreptococcal genitourinary tract coloniza-tion in pregnant women. Obstet Gynecol.2005;106(6):1246–1252

36. Fluegge K, Supper S, Siedler A, Berner R.Serotype distribution of invasive group Bstreptococcal isolates in infants: resultsfrom a nationwide active laboratory surveil-lance study over 2 years in Germany. ClinInfect Dis. 2005;40(5):760–763

37. Shapiro RL, Lockman S, Kim S, et al. Infantmorbidity, mortality, and breast milk immu-nologic profiles among breast-feeding HIV-infected and HIV-uninfected women in Bo-tswana. J Infect Dis. 2007;196(4):562–569

38. Filteau S. The HIV-exposed, uninfected Afri-can child. Trop Med Int Health. 2009;14(3):276–287

39. Jeena PM, Bobat B, Thula SA, Adhikari M.Children with Pneumocystis jiroveci pneu-monia and acute hypoxaemic respiratoryfailure admitted to a PICU, Durban, SouthAfrica. Arch Dis Child. 2008;93(6):545

40. McNally LM, Jeena PM, Lalloo U, et al. Prob-able mother to infant transmission of Pneu-

ARTICLE

PEDIATRICS Volume 126, Number 3, September 2010 e637 by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

mocystis jiroveci from an HIV-infectedwoman to her HIV-uninfected infant. AIDS.2005;19(14):1548–1549

41. Heresi GP, Caceres E, Atkins JT, Reuben J,DoyleM. Pneumocystis carinii pneumonia ininfants who were exposed to human immu-nodeficiency virus but were not infected: anexception to the AIDS surveillance case def-inition. Clin Infect Dis. 1997;25(3):739–740

42. de Moraes-Pinto MI, Almeida AC, Kenj G, etal. Placental transfer and maternally ac-quired neonatal IgG immunity in human im-munodeficiency virus infection. J Infect Dis.1996;173(5):1077–1084

43. Clerici M, Saresella M, Colombo F, et al.T-lymphocyte maturation abnormalities inuninfected newborns and children with ver-tical exposure to HIV. Blood. 2000;96(12):3866–3871

44. Weinhold KJ, Lyerly HK, Stanley SD, AustinAA, Matthews TJ, Bolognesi DP. HIV-1 GP120-mediated immune suppression and lym-phocyte destruction in the absence of viralinfect ion . J Immunol . 1989 ;142(9) :3091–3097

45. Chougnet C, Kovacs A, Baker R, et al. Influ-ence of human immunodeficiency virus-infected maternal environment on develop-ment of infant interleukin-12 production. JInfect Dis. 2000;181(5):1590–1597

46. Rich KC, Siegel JN, Jennings C, Rydman RJ,Landay AL. Function and phenotype of imma-ture CD4� lymphocytes in healthy infantsand early lymphocyte activation in unin-fected infants of human immunodeficiencyvirus-infected mothers. Clin Diagn Lab Im-munol. 1997;4(3):358–361

47. Kuhn L, Coutsoudis A, Moodley D, et al.T-helper cell responses to HIV envelope pep-tides in cord blood: protection against in-trapartum and breast-feeding transmis-sion. AIDS. 2001;15(1):1–9

48. Kuhn L, Meddows-Taylor S, Gray G, Tiemes-sen C. Human immunodeficiency virus(HIV)-specific cellular immune responses innewborns exposed to HIV in utero. Clin In-fect Dis. 2002;34(2):267–276

49. Rowland-Jones SL, Nixon DF, Aldhous MC, etal. HIV-specific cytotoxic T-cell activity in anHIV-exposed but uninfected infant. Lancet.1993;341(8849):860–861

50. Legrand FA, Nixon DF, Loo CP, et al. StrongHIV-1-specific T cell responses in HIV-1-exposed uninfected infants and neonatesrevealed after regulatory T cell removal.PLoS One. 2006;1:e102

51. Bunders M, Thorne C, Newell ML. Maternaland infant factors and lymphocyte, CD4 andCD8 cell counts in uninfected children of

HIV-1-infected mothers. AIDS. 2005;19(10):1071–1079

52. Feiterna-Sperling C, Weizsaecker K, BuhrerC, et al. Hematologic effects of maternal an-tiretroviral therapy and transmission pro-phylaxis in HIV-1-exposed uninfected new-born infants. J Acquir Immune Defic Syndr.2007;45(1):43–51

53. Poirier MC, Divi RL, Al-Harthi L, et al. Long-term mitochondrial toxicity in HIV-uninfected infants born to HIV-infectedmothers. J Acquir Immune Defic Syndr.2003;33(2):175–183

54. Aldrovandi GM, Chu C, Shearer WT, et al. An-tiretroviral exposure and lymphocytemtDNA content among uninfected infants ofHIV-1-infected women. Pediatrics. 2009;124(6). Available at: www.pediatrics.org/cgi/content/full/124/6/e1189

55. Alimenti A, Burdge DR, Ogilvie GS, Money DM,Forbes JC. Lactic acidemia in human immu-nodeficiency virus-uninfected infants ex-posed to perinatal antiretroviral therapy.Pediatr Infect Dis J. 2003;22(9):782–789

56. Noguera A, Fortuny C, Munoz-Almagro C, etal. Hyperlactatemia in human immunodefi-ciency virus-uninfected infants who are ex-posed to antiretrovirals. Pediatrics. 2004;114(5). Available at: www.pediatrics.org/cgi/content/full/114/5/e598

e638 EPALZA et al by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

DOI: 10.1542/peds.2010-0183; originally published online August 23, 2010; 2010;126;e631Pediatrics

Anne Dediste, Arnaud Marchant and Jack LevyCristina Epalza, Tessa Goetghebuer, Marc Hainaut, Fany Prayez, Patricia Barlow,

Uninfected InfantsHigh Incidence of Invasive Group B Streptococcal Infections in HIV-Exposed

  

ServicesUpdated Information &

tmlhttp://pediatrics.aappublications.org/content/126/3/e631.full.hincluding high resolution figures, can be found at:

References

tml#ref-list-1http://pediatrics.aappublications.org/content/126/3/e631.full.hat:This article cites 51 articles, 15 of which can be accessed free

Citations

tml#related-urlshttp://pediatrics.aappublications.org/content/126/3/e631.full.hThis article has been cited by 5 HighWire-hosted articles:

Subspecialty Collections

diseases_subhttp://pediatrics.aappublications.org/cgi/collection/infectious_Infectious Diseasesthe following collection(s):This article, along with others on similar topics, appears in

Permissions & Licensing

mlhttp://pediatrics.aappublications.org/site/misc/Permissions.xhttables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2010 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from

DOI: 10.1542/peds.2010-0183; originally published online August 23, 2010; 2010;126;e631Pediatrics

Anne Dediste, Arnaud Marchant and Jack LevyCristina Epalza, Tessa Goetghebuer, Marc Hainaut, Fany Prayez, Patricia Barlow,

Uninfected InfantsHigh Incidence of Invasive Group B Streptococcal Infections in HIV-Exposed

  

  http://pediatrics.aappublications.org/content/126/3/e631.full.html

located on the World Wide Web at: The online version of this article, along with updated information and services, is

 

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2010 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

by guest on February 7, 2015pediatrics.aappublications.orgDownloaded from