Placental Abnormalities Associated with Human Immunodeficiency ...

1652

Placental Abnormalities Associated with Human Immunodeficiency Virus Type 1 Infection and Perinatal Transmission in Bangkok, Thailand David A. Schwartz,1,a Suthi Sungkarat,3 Nathan Shaffer,2,5 Jirasak Laosakkitiboran,5 Wendy Supapol,5 Pichai Charoenpanich,3 Tuenjai Chuangsuwanich,4 and Timothy D. Mastro2,5

1

Departments of Pathology and Medicine, Division of Infectious Diseases, Emory University School of Medicine, and 2Division of HIV/AIDS Prevention, Centers for Disease Control and Prevention, Atlanta, Georgia; Departments of 3Obstetrics and Gynecology and 4Pathology, Faculty of Medicine, Siriraj Hospital, Mahidol University, Bangkok, and 5HIV/AIDS Collaboration, Nonthaburi, Thailand

The effects of human immunodeficiency virus (HIV) type 1 on the placenta and the role of the placenta in mother-to-child HIV-1 transmission are not well understood. Placentas from 78 HIV-infected and 158 HIV-uninfected women were examined as part of a prospective perinatal HIV transmission study in Bangkok. HIV-infected women were more likely than HIV-uninfected women to have chorioamnionitis (odds ratio [OR], 2.1; P p .03), placental membrane inflammation (PMI; OR, 2.7; P p .02 ), and deciduitis (OR, 2.3; P p .03) and less likely to have villitis (OR, 0.3; P p .02 ). However, among HIV-infected women, fewer women who transmitted infection to their child had chorioamnionitis (relative risk [RR], 0.2; P p .03), funisitis (RR, 0.4; P p .1), or PMI (RR undefined; P p .03 ). These findings suggest that, in this population, HIV-infected women are at increased risk for placental membrane inflammatory lesions, but that placental inflammatory lesions are not associated with increased perinatal HIV transmission.

Recent studies suggest that most perinatal human immunodeficiency virus (HIV) type 1 infections occur at or near birth [1–4]. This contrasts with earlier hypotheses that most transmission occurs in utero through transplacental transmission [5, 6]. Although it is presumed that the placenta plays a major role in protecting against perinatal HIV transmission, the effects of HIV on the placenta are not well understood. In view of several reports from Africa that chorioamnionitis is a risk factor for perinatal HIV transmission [7–10], there is heightened interest in determining whether there are differences in the spectrum of placental lesions, particularly inflammatory lesions, between HIV-infected and -uninfected women and whether specific lesions are related to an increased risk of perinatal HIV transmission. Currently, there are no clear answers

Received 28 December 1998; revised 9 August 2000; electronically published 8 November 2000. Informed consent was obtained from all study participants. The study was conducted according to the guidelines for human experimentation of the US Department of Health and Human Services and of the collaborating institutions. The study protocol was reviewed by the Centers for Disease Control and Prevention (CDC) institutional review board and by the Ethical Review Committee, Ministry of Public Health, Thailand. Financial support: HIV Perinatal Transmission Study, CDC; CDC contract 770214-14-PF, NIH grant AI-32341-01, and Emory Medical Care Foundation grant 247320 (to D.A.S.). a D.A.S. served as chair of the pathology work group of the National Institutes of Health (NIH) Women and Infants Transmission Study. Reprints or correspondence: Dr. Nathan Shaffer, Division of HIV/AIDS, CDC (MS E-07), Atlanta, GA 30333 ([email protected]). The Journal of Infectious Diseases 2000; 182:1652–7 q 2000 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2000/18206-0010$02.00

Downloaded from https://academic.oup.com/jid/article-abstract/182/6/1652/913756 by guest on 15 December 2017

to these questions. Findings on the effects of HIV on the placenta differ. Studies suggest both increased [11] and decreased [12] placental weight, decreased prevalence of villitis [10], and increased prevalence of chorionitis [13], chorioamnionitis, funisitis, placental membrane inflammation (PMI) [7], and villous immaturity and allantois vasculopathy [14] in placentas from HIV-infected women. However, other studies suggest no or few placental differences associated with HIV [13, 15]. Most large studies of perinatal HIV transmission have not evaluated placental pathology [16–22]. Among the few published studies that include placental pathology, several studies in Africa [7–10], India [23], and the United States [24] report increased transmission risk associated with inflammatory lesions of the membranes and cord (chorioamnionitis and funisitis). However, these findings were not confirmed in a separate analysis from Kenya [15] or in a large unpublished US study [25]. Different findings may be due to differences in study populations, number of placentas examined, diagnostic criteria and expertise, thoroughness of placental sampling, and variability in the definition of infant infection. Findings on transmission in developing countries may also be subject to misclassification, due to difficulty in distinguishing intrapartum and early postnatal transmission (via breast-feeding). Antibody-based and molecular studies of placentas from HIV-infected women have yielded markedly different results and have not clarified the pathogenic mechanisms or timing of perinatal HIV transmission [13, 14, 26–29]. In Thailand, we addressed these questions in a cohort of HIV-infected women who were enrolled antenatally and who did not breast-feed and in a control group of

JID 2000;182 (December)

Placental Pathology and HIV in Bangkok

noninfected women. We used standardized diagnostic criteria and focused our investigation on inflammatory conditions of the placenta.

Subjects and Methods Study participants. The placental pathology study was conducted as a substudy of the Bangkok Collaborative Perinatal HIV Transmission Study, which enrolled pregnant women residing in the local catchment areas of the 2 largest antenatal clinics in Bangkok from November 1992 through March 1994. The methods and main findings regarding enrollment and the characteristics of the women, HIV transmission rate, and risk factors have been described elsewhere [30, 31]. None of the women received zidovudine. In accordance with local guidelines, the HIV-infected women did not breast-feed. The perinatal HIV transmission rate for the full cohort of 281 women with known outcomes was 24.2% (95% confidence interval [CI], 19%–29%). At one study hospital from October 1993 through March 1994, a consecutive sample of 78 of 81 HIV-infected women enrolled in the main transmission study were enrolled in the placenta substudy. For logistic reasons, placentas from 3 women were not collected. During this period, 155 HIVinfected women gave birth at the study hospital. All study women had laboratory testing (including CD4/CD8 lymphocyte counts, quantitative plasma virus load, and HIV-1 subtyping [31]), and their clinical histories were obtained. Infants born to HIV-infected mothers were followed up every 2–3 months. Venous blood samples obtained at birth and at 2 and 6 months were tested for proviral HIV DNA by polymerase chain reaction (PCR). Infants were considered to be HIV infected if they had 2 positive DNA PCR test results or 1 positive PCR test result and a 1993 CDC AIDS-defining condition. Infants were defined as uninfected if they tested PCR negative on 2 samples, including 1 sample obtained at age >6 months, or if they seroreverted to HIVnegative status on antibody testing. Infants who did not meet these criteria were considered to have unknown infection status. Mothers of infected infants were considered to be “transmitters”; mothers of uninfected infants were considered to be “nontransmitters.” For the placental pathology substudy, after an HIV-infected study woman gave birth, the next 2 available HIV-negative women who gave birth were enrolled as control subjects. During this period, ∼9000 HIV-negative women gave birth. The 158 HIV-negative women were enrolled systematically, on the basis of time of delivery. Placentas from HIV-negative women were handled identically to those of HIV-infected women. Placental gross examination and processing. Immediately after delivery, placentas were placed in buffered formalin and transferred to the Department of Obstetrics and Gynecology pathology laboratory. To ensure uniformity, one obstetrician trained in placental pathology (S.S.) processed all study placentas without knowing HIV status. Specific placental pathology submission forms were used. Most specimens were fixed for !24 h, although placentas delivered on weekends were fixed longer. We used universal precautions and systematically examined and sampled all placentas in an identical manner, according to a standardized protocol. The specimens were measured, and extraplacental membranes, umbilical cord, and adherent blood clots were


1653

removed before the placental body was weighed. The placenta was sectioned serially at 1-cm intervals, to identify gross lesions. Findings were recorded by the same investigator in Bangkok on a placental pathology work sheet adapted from work sheets designed for the placental pathology component of the National Institutes of Health Women and Infants Transmission Study [32]. At least 11 sections were examined histologically: 4 sections of extraplacental membrane roll, with sampling of all 4 quadrants, and the region of membrane rupture; 2 sections of umbilical cord, including the proximal and distal regions of the cord; and 5 sections of the placental body, including the mid-disk maternal and fetal surfaces, tissue at and adjacent to the umbilical cord insertion, and the marginal placenta. In addition, grossly visible lesions (e.g., infarcts, abscesses, or hemorrhages) were sampled. Sections were processed for light microscopy, by standard histologic methods, at the study hospital. Slides were stained with hematoxylin-eosin and were shipped with corresponding pathology work sheets to Atlanta for final microscopic diagnosis. Paraffin blocks of placental tissue were saved for reference and future studies. Placental microscopic diagnosis. All placental tissue slides were examined by one pathologist (D.A.S.) with placental and infectious disease expertise. Slides and coded placental pathology work sheets were evaluated without knowledge of HIV status or transmission outcome. The pathologist had access to the following information: maternal age, gestational age, pregnancy complications, mode of delivery, placental weight, fixation time, and gross placental features. Microscopic findings were entered on a precoded form designed for thorough classification of inflammatory conditions of the placenta [32]. Funisitis (inflammation of the umbilical cord) was graded as mild, moderate, or severe; the predominant inflammatory cell type was noted; structural involvement was identified (vein, artery, and Wharton’s jelly), and associated findings (e.g., meconium-associated myonecrosis, microabscesses, vascular necrosis, or thrombosis) were described. Chorioamnionitis was semiquantitatively classified as low grade (mild-to-moderate; average of 1 abortions, and VDRL positivity were associated with HIV infection. The key placental pathology findings among HIV-infected and -uninfected women are shown in table 2. Mean placental weight was 46 g lower for HIV-infected mothers. To determine whether the association of decreased placental weight with HIV was confounded by prematurity or birth weight, their interaction was analyzed by logistic regression. HIV status (P ! .001 ) and birth weight (P ! .001) were independently associated with lower placental weight. Significantly more HIV-infected than -uninfected women had chorioamnionitis (OR, 2.1), PMI (OR, 2.7), and plasma cellular deciduitis (OR, 2.3; table 2). In placentas with chorioamnionitis, high-grade chorioamnionitis was more prevalent in HIV-infected than in -uninfected women (OR, 7.3). Also, in placentas with chorioamnionitis, acute (polymorphonuclear) chorioamnionitis tended to be more common than chronic (lymphocytic) chorioamnionitis in HIV-infected than in -uninfected women (OR, 6.3), but this was not statistically significant. More uninfected women had villitis of all histologic types. Among HIV-infected women, the absence of villitis was not correlated with low CD4 cell count (P p .8). There were no other significant differences between placentas from HIV-infected and -uninfected mothers. Placentas from transmitting and nontransmitting HIV-infected mothers. Definitive HIV infection status was known for 75 infants (96%) born to HIV-infected mothers in the pathology substudy. Of these, 17 (22.7%) were HIV infected. For mothers who transmitted infection, the frequency of several inflammatory lesions was decreased: chorioamnionitis (RR, 0.2; P p .03); PMI (RR undefined; P p .03); either villitis or chorioamnionitis (RR, 0.2; P p .03); and either villitis or PMI (RR undefined; P p .03; table 3). A trend toward decreased prevalence of funisitis in transmitting mothers was also noted. No histopathologic findings occurred with increased frequency in the placentas of transmitting mothers. However, compared with placentas of nontransmitting mothers, placentas of transmitting mothers showed a trend toward a higher frequency of


Table 1. women.


Demographic and obstetric characteristics of delivering

Characteristic Mean age (range), years Mean gestational a age (range), weeks Mean CD4 cells/mm3 (range) Primiparous Abortions (>1) PROM 124 h Cesarean section Stillbirth VDRL/TPHA positive (for syphilis) Chlamydia Gonorrhea

HIV infected (n p 78)

HIV uninfected (n p 158)

P

24 (15–36)

25 (15–37)

.14

39.1 (28–45)

39.3 (34–44)

.74

Not done 79 (50) 25 (16) 1 (0.6) 8 (5) 1 (0.6)

.001 .01 1.0 .14 1.0

474 56 16 0 8 0

(170–1040) (72) (21) (10)

5 (6) 12 (16) 2 (3)

2 (1.3) b Not done b Not done

.04

NOTE. Data are no. (%), except where noted. HIV, human immunodeficiency virus; PROM, prolonged rupture of membranes; TPHA, Treponema pallidum hemagglutination test. a By history of last menstrual period and clinical examination. Five women (2 HIV infected) had estimated gestational ages 1 42 weeks. b Chlamydia and gonorrhea testing were performed by DNA hybridization probe. Although the HIV-negative women in this placenta study were not tested for these sexually transmitted diseases, contemporaneous testing of other HIVnegative pregnant women at this hospital yielded prevalences of 9% (chlamydia) and 1% (gonorrhea) [33].

plasma cellular deciduitis and abnormal villous maturity. Only 1 of 17 HIV-transmitting mothers had a placenta with chorioamnionitis; this was low grade and neutrophilic. Among 18 nontransmitting mothers with chorioamnionitis and a grade determination (1 was missing), 10 (56%) were high grade, and 17 (94%) were predominantly neutrophilic. No associations were noted between perinatal HIV transmission and intravillous hemorrhage, infarction, or villitis. Other covariates. We found no significant differences or trends between placentas from mothers presumed to have transmitted in utero (n p 4) and intrapartum (n p 12; 1 infected child could not be classified with regard to timing) [4]. However, there were few pathologic lesions among transmitting women, and the number of in utero and intrapartum infections in this subset was small. The 1 transmitting woman with chorioamnionitis transmitted intrapartum. Among HIV-infected women, maternal HIV plasma load was not related to chorioamnionitis (median RNA copies/mL, 11,000 vs. 18,000 for women with or without chorioamnionitis, respectively; P p .5 ) or other placental findings. Maternal HIV-1 subtype distribution (97% subtype E and 3% subtype B) was similar to that of the full perinatal cohort. Because of the high predominance of subtype E, it was not possible to determine whether there might be differences in placental pathology associated with viral subtype. None of the HIV-infected mothers had PROM (124 h), so an effect on placental pathology could not be evaluated. Duration of ruptured membranes (!24 h) was not related to placental pathology. The prevalence of sexually transmitted diseases (STDs) was relatively low (table 1) [33]. Positive antenatal syph-



ilis, chlamydia, or gonorrhea test results were not associated with chorioamnionitis.

Discussion This study addressed 2 questions: Are placental lesions associated with maternal HIV infection, and are placental lesions associated with perinatal HIV transmission? If specific placental lesions are related to perinatal transmission, this might suggest a mechanism for transmission and a possible strategy for interventions. Our results indicate that HIV-infected women are at increased risk for chorioamnionitis and PMI. However, these inflammatory lesions were not associated with an increased risk of perinatal transmission in our cohort. Surprisingly, we found an inverse relationship between chorioamnionitis and perinatal transmission. Potential differences in placental pathology between HIVinfected and -uninfected women have not been well characterized. In our study, HIV-infected women were at increased risk for chorioamnionitis, PMI, and plasma cellular deciduitis but at decreased risk for villitis. The different associations of these inflammatory lesions may be related to differences in involvement of maternal or fetal cells. Unlike villitis, inflammation of the membranes and umbilical cord is thought to be mediated by fetal leukocytes [34]. Because chorioamnionitis and funisitis (PMI) are almost always associated with ascending maternal genital tract infections, their increased prevalence among HIVinfected women likely results from HIV-associated factors (e.g., sexual exposure, vaginitis, and STDs). We did not investigate the microbiologic causes of chorioamnionitis, and the cause was not determined in the African studies. Acute chorioamnionitis

1655

is also associated with PROM, a risk factor for intrapartum HIV transmission [1, 35]. However, because no HIV-infected women in our study had PROM, we could not study this association. Our finding that villitis occurred less often in HIV-infected women may be related to HIV-associated immunosuppression and suggests a diminished capacity to develop an inflammatory response. Because the inflammatory cells associated with villitis are believed to be maternal in origin [36, 37], our data support the hypothesis that HIV-infected women may have a decreased ability to form villitis-type placental lesions. In Uganda and the United States, the frequency of villitis in HIV-exposed placentas was also decreased [10, 25]. Villitis, estimated to be 7%– 18% in well-sampled placentas from geographically diverse cohorts of pregnant women, is used by pathologists as a marker to assess adequacy of sampling and diagnosis in control populations [38–40]. The villitis observed in our study is termed villitis of unknown etiology. It is not associated with a single infectious agent but is associated with increased perinatal morbidity and recurrence in future pregnancies. On the basis of several early reports from Africa [7–10], we hypothesized that chorioamnionitis would be associated with perinatal HIV transmission, particularly intrapartum transmission. Recently, a US study found chorioamnionitis to be a risk factor for transmission in the presence of zidovudine [24], and it was suggested that chorioamnionitis, if associated with PROM and preterm birth, may be an important underlying risk factor for perinatal HIV transmission [35]. However, we found the opposite association between chorioamnionitis and perinatal HIV transmission. In our study, there were significantly fewer inflammatory lesions of the placental membranes and umbilical cord

Table 2. Selected histopathologic findings in placentas of human immunodeficiency virus (HIV)–infected and -uninfected mothers. Characteristic

HIV infected (n p 78)

Mean placental weight (range), g PMI Villitis Chorioamnionitis High grade Acute Plasmacellular deciduitis Hofbauer cell hyperplasia Funisitis PMI and villitis Villitis or chorioamnionitis Infarction PMI or villitis Villitis and chorioamnionitis Intervillositis Villous stromal edema Decidual necrosis Intravillous hemorrhage Decidual vasculopathy

459 13 3 21 8 19 16 11 20 1 23 6 15 1 0 31 6 15 0

(280–700) (17) (4) (27) (40) (95) (21) (14) (26) (1) (29) (8) (19) (1) (40) (8) (19)

HIV uninfected (n p 158) 505 11 21 24 2 18 16 16 34 1 42 13 31 3 2 58 7 25 2

(300–890) (7) (13) (15) (8) (75) (10) (10) (22) (!1) (27) (8) (20) (2) (1) (37) (4) (16) (1)

Odds ratio (95% CI)

P !.001

2.7 0.3 2.1 7.3 6.3 2.3 1.5 1.3 2.0 1.2 0.9 1.0 0.7 0.0 1.1 1.8 1.3 0.0

(1.1–6.8) (0.1–0.9) (1.0–4.2) (1.1–78) (0.6–306) (1.0–5.2) (0.6–3.6) (0.6–2.5) (0.0–161) (0.6–2.2) (0.3–0.9) (0.5–2.0) (0.0–8.5) (0.0–10.8) (0.6–2.1) (0.5–6.5) (0.6–2.7) (0.0–10.8)

.02 .02 .03 .03 .1 .03 .4 .5 .6 .6 .9 .9 1.0 1.0 .7 .4 .5 1.0

NOTE. Data are no. (%), except where noted. CI, confidence interval; PMI, placental membrane inflammation (simultaneous chorioamnionitis and funisitis).


1656

Schwartz et al.

Table 3. Selected histopathologic findings in placentas of human immunodeficiency virus–infected transmitting and nontransmitting mothers. Finding Villitis or chorioamnionitis Villitis or PMI Chorioamnionitis PMI Plasmacellular deciduitis Funisitis Abnormal villous maturity Decidual necrosis Infarction Villitis (all types) Villitis and PMI Villitis and chorioamnionitis Hofbauer cell hyperplasia Intravillous hemorrhage Villous stromal fibrosis

Transmitting Nontransmitting (n p 17) (n p 58) 1 (6) 0 1 (6) 0

20 14 19 13

6 (35) 2 (12)

9 (16) 18 (31)

4 (24) 2 (12) 0 0 0

5 3 5 2 1

0 2 (12) 3 (18) 0

(35) (24) (33) (22)

(9) (5) (9) (3) (2)

Relative risk (95% CI)

P

0.2 (0.02–1.1) — 0.2 (0.02–1.2) —

.03 .03 .03 .03

2.2 (1.0–4.9) 0.4 (0.1–1.5)

.09 .1

2.3 (0.9–5.4) 1.9 (0.6–6.0) — — —

.2 .3 .6 1.0 1.0

1 (2)

—

1.0

8 (14) 11 (19) 2 (3)

0.9 (0.2–3.2) 0.9 (0.3–2.8) —

1.0 1.0 1.0

NOTE. Data are no. (%), except where noted. CI, confidence interval; PMI, placental membrane inflammation (simultaneous chorioamnionitis and funisitis); —, undefined.

(including chorioamnionitis, funisitis, and PMI) among transmitting than among nontransmitting women. Of 20 HIV-infected women with chorioamnionitis, only 1 transmitted HIV to her child, and only 1 of 17 transmitting women had histologic evidence of chorioamnionitis. Preliminary data from a US study also show no association between chorioamnionitis or funisitis and perinatal HIV transmission [25]. The differing study results may be related to several factors, including differences in background risk and cofactors for chorioamnionitis in different settings, diagnostic criteria for chorioamnionitis, antenatal and obstetric care, nutritional status, and immunosuppression levels. These inflammatory lesions of the membranes and cord result from ascending infections caused by a wide variety of microorganisms and often are associated with obstetric complications, such as PROM, amniotic infection syndrome, and fever. However, in our cohort of HIVinfected women, none of these complications was reported. Our data suggest that, at least in our Bangkok setting, chorioamnionitis, funisitis, and PMI are not risk factors for perinatal HIV transmission and that pathologic findings may differ significantly between populations. In our study population, prophylactic treatment with antibiotics to prevent chorioamnionitis [35] would likely have little or no effect in preventing perinatal HIV transmission. If placental abnormalities are related to perinatal transmission, it might be hypothesized that PMI (or chorioamnionitis), which often occurs just before delivery, might be related more directly to intrapartum transmission risk, whereas abnormalities in the placenta itself might be related more directly to in utero transmission. However, because of the low level of pla-



cental pathology among transmitting women in our study and the small number of both in utero and intrapartum transmission events, we were limited in our ability to address this. In populations in which placental abnormalities appear to be related to transmission, the relationship of specific lesions with time of transmission should be examined. Some previous studies of HIV-related placental pathology may have been limited by study design or technique. The lack of an uninfected comparison group, although not interfering with the analysis of risk factors for perinatal transmission, prevents the examination of pathologic differences between HIVexposed and -unexposed placentas and precludes the ability to establish background rates for various lesions. In some studies, sampling of the placenta was limited and suboptimal, or the rate of villitis was so low that the adequacy of microscopic interpretation may be questioned. Our study design, which was intended to optimize the clinical, laboratory, and pathologic parameters necessary to evaluate the effect of HIV on the placenta and the role of placental lesions in perinatal transmission, represents one of the most thorough samplings of placentas reported in the perinatal HIV literature. Our findings do not diminish the importance of the placenta in providing a barrier to perinatal HIV infection. However, our data suggest that it is premature to assign placental pathologic lesions a definite role as risk factors or indicators for perinatal HIV transmission. Additional well-designed studies in other populations, with uniform pathologic diagnostic criteria, are needed to help clarify the effect of HIV on the placenta and the potential role of placental lesions in perinatal HIV transmission. Acknowledgments We thank Kanchana Neeyapun, Chariya Prasert, and Sununta Henchaichon for coordinating the field study, Philip Mock for assistance with data management and analysis, and Nancy Young and staff for laboratory studies (HIV/AIDS Collaboration, Nonthaburi, Thailand); Wei Zhang for processing the specimens at Emory University (Atlanta); and the staff of the Division of Pathology, Department of Obstetrics and Gynecology, Siriraj Hospital Faculty of Medicine (Bangkok), for processing the placentas and preparing slides. Members of the Bangkok Collaborative Perinatal HIV Transmission Study Group are listed in reference 31.

References 1. Minkoff H, Burns DN, Landesmann S, et al. The relationship of the duration of ruptured membranes to vertical transmission of human immunodeficiency virus. Am J Obstet Gynecol 1995; 173:585–9. 2. Bryson YJ. Perinatal HIV-1 transmission: recent advances and therapeutic interventions. AIDS 1996; 10(Suppl 3):S33–42. 3. Kuhn L, Abrams EJ, Matheson PB, et al. Timing of maternal-infant HIV transmission: associations between intrapartum factors and early polymerase chain reaction results. AIDS 1997; 11:429–35. 4. Mock PA, Shaffer N, Bhadrakom C, et al. Maternal viral load and timing of mother-to-child HIV transmission, Bangkok, Thailand. AIDS 1999; 13:407–14.



5. Douglas GC, King BF. Maternal-fetal transmission of human immunodeficiency virus: a review of possible routes and cellular mechanisms of infection. Clin Infect Dis 1992; 15:678–91. 6. Hoff R, Berardi VP, Weiblen BJ, Mahoney-Trout L, Mitchell ML, Grady GF. Seroprevalence of human immunodeficiency virus among childbearing women: estimation by testing samples of blood from newborns. N Engl J Med 1988; 318:525–30. 7. Ryder RW, Nsa W, Hassig SE, et al. Perinatal transmission of the human immunodeficiency virus type 1 to infants of seropositive women in Zaire. N Engl J Med 1989; 320:1637–42. 8. St Louis ME, Kamenga M, Brown C, et al. Risk for perinatal HIV-1 transmission according to maternal immunologic, virologic, and placental factors. JAMA 1993; 269:2853–9. 9. Temmerman M, Nyong’o AO, Bwayo J, Fransen K, Coppens M, Piot P. Risk factors for mother-to-child transmission of human immunodeficiency virus–1 infection. Am J Obstet Gynecol 1995; 172:700–5. 10. Wabwite-Mangen F, Abramowsky C, Mmiro FA. Epidemiologic and placental pathologic studies of materno-fetal HIV transmission (MFT) [abstract 39]. Mod Pathol 1995; 8:7P. 11. Jauniaux E, Nessmann C, Imbert MC, Meuris S, Puissant F, Hustin J. Morphological aspects of the placenta in HIV pregnancies. Placenta 1988; 9: 633–42. 12. Gichangi PB, Nyongo AO, Temmerman M. Pregnancy outcome and placental weights: their relationship to HIV-1 infection. East Afr Med J 1993; 70:85–9. 13. Chandwani S, Alba Greco M, Mittal K, Antoine C, Krasinski K, Borkowsky W. Pathology and human immunodeficiency virus expression in placentas of seropositive women. J Infect Dis 1991; 163:1134–8. 14. Backe E, Jimenez E, Unger M, et al. Vertical human immunodeficiency virus transmission: a study of placental pathology in relation to risk factors. Am J Perinatol 1994; 11:326–30. 15. Temmerman M, Chomba EN, Ndinya-Achola J, Plummer FA, Coppens M, Piot P. Maternal human immunodeficiency virus–1 infection and pregnancy outcome. Obstet Gynecol 1994; 83:495–501. 16. Halsey NA, Boulos R, Holt E, et al. Transmission of HIV-1 infections from mothers to infants in Haiti: impact on childhood mortality and malnutrition. JAMA 1990; 264:2088–92. 17. European Collaborative Study. Children born to women with HIV-1 infection: natural history and risk of transmission. Lancet 1991; 337:253–60. 18. Adjorlolo-Johnson G, DeCock K, Ekpini E, et al. Prospective comparison of mother-to-child transmission of HIV-1 and HIV-2 in Abidjan, Ivory Coast. JAMA 1994; 272:462–6. 19. Lepage P, Van de Perre P, Msellati P, et al. Mother-to-child transmission of human immunodeficiency virus type 1 (HIV-1) and its determinants: a cohort study in Kigali, Rwanda. Am J Epidemiol 1993; 137:589–99. 20. Bobat R, Coovadia H, Coutsoudis A, Moodley D. Determinants of motherto-child transmission of human immunodeficiency virus type 1 infection in a cohort from Durban, South Africa. Pediatr Infect Dis J 1996; 15:604–10. 21. Braddock MR, Kreiss JK, Embree JE, et al. Impact of maternal HIV infection on obstetrical and early neonatal outcome. AIDS 1990; 4:1001–5. 22. The European Collaborative Study. Vertical transmission of HIV-1: maternal immune status and obstetric factors. AIDS 1996; 10:1675–81. 23. Kumar RM, Uduman SA, Khurranna AK. A prospective study of mother-


24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34. 35.

36.

37. 38. 39. 40.

1657

to-infant HIV transmission in tribal women from India. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 9:238–42. Van Dyke RB, Korber BT, Popek E, et al. The Ariel Project: a prospective cohort study of maternal-child transmission of human immunodeficiency virus type 1 in the era of maternal antiretroviral therapy. J Infect Dis 1999; 179:319–28. Schwartz DA. The placenta as a barrier or facilitator of HIV transmission. In: Program book of Global Strategies for the Prevention of HIV Transmission from Mothers to Infants (Crystal City, Virginia). New York: American Foundation for AIDS Research, 1998:57–59. Peuchmaur M, Delfraissy JF, Pons JC, et al. HIV proteins absent from placentas of 75 HIV-1–positive women studied by immunohistochemistry. AIDS 1991; 5:741–5. Mattern CFT, Murray K, Jensen A, Farzadegan H, Pang J, Modlin JF. Localization of human immunodeficiency virus core antigen in term human placentas. Pediatrics 1992; 89:207–9. Martin AW, Brady K, Smith SI, et al. Immunohistochemical localization of human immunodeficiency virus p24 antigen in placental tissue. Hum Pathol 1992; 23:411–4. Tscherning-Casper C, Papadogiannakis N, Anvret M, et al. The trophoblastic epithelial barrier is not infected in full-term placentae of human immunodeficiency virus–seropositive mothers undergoing antiretroviral therapy. J Virol 1999; 73:9673–8. Siriwasin W, Shaffer N, Roongpisuthipong A, et al. HIV prevalence, risk, and partner serodiscordance among pregnant women in Bangkok. JAMA 1998; 280:49–54. Shaffer N, Roongpisuthipong A, Siriwasin W, et al. Maternal viral load and perinatal HIV-1 subtype E transmission, Thailand. J Infect Dis 1999; 179: 590–9. Schwartz DA, Roberts D, Anderson V, et al. Perinatal transmission of HIV and placental pathology: diagnostic criteria standardization. In: Proceedings of the 2d Conference of Women and Children with AIDS (Edinburgh). 1993. Chaisilwattana P, Chuachoowong R, Siriwasin W, et al. Chlamydial and gonococcal cervicitis in HIV-seropositive and HIV-seronegative pregnant women in Bangkok. Sex Transm Dis 1997; 24:495–502. Benirschke K, Kaufmann P. Pathology of the human placenta. 3d ed. New York: Springer Verlag, 1995:537–59. Goldenberg RL, Vermund SH, Goepfert AR, Andrews WW. Choriodecidual inflammation: a potentially preventable cause of perinatal HIV-1 transmission? Lancet 1998; 352:1927–30. Redline RW, Paterson P. Villitis of unknown etiology is associated with major infiltration of fetal tissue by maternal inflammatory cells. Am J Pathol 1993; 143:473–9. Benirschke K, Kaufmann P. Pathology of the human placenta. 3d ed. New York: Springer Verlag, 1995:596–601. Labarrere C, Althabe O. Chronic villitis of unknown aetiology in recurrent intrauterine fetal growth retardation. Placenta 1987; 8:167–73. Russell P, Atkinson K, Krishnan L. Recurrent reproductive failure due to severe placental villitis of unknown etiology. J Reprod Med 1980; 24:93–8. Redline RW, Abramowsky CR. Clinical and pathologic aspects of recurrent placental villitis. Hum Pathol 1985; 16:727–31.