Despite the remarkable efficacy of zidovudine (ZDV) in preventing perinatal HIV-1 transmission, a small but significant proportion of mothers who are treated during pregnancy still transmit. Several observational studies and clinical trials have reported low rates of perinatal transmission among women receiving ZDV, with rates ranging from 4 to 8% [1–6]
Failure to prevent perinatal transmission among mothers receiving ZDV in some cases may be linked to maternal ZDV drug resistance. Although a few important but limited reports have identified maternal isolates with ZDV drug-resistant mutations among transmitting mothers [2,3,7] the numbers of study subjects have been too few to assess the impact of drug resistance on perinatal transmission.
In the present study, HIV-1 reverse transcriptase (RT) gene sequence data is reported for 142 women, who were enrolled in the Women and Infants Transmission Study (WITS) between September 1989 and September 1994, and who received ZDV during pregnancy. In addition, we report the association of having any ZDV-specific RT mutation in perinatal HIV-1 transmission independent of maternal CD4 cell count, viral load, and obstetric factors known to influence the risk of infection.
Materials and methods
The women selected for this study of molecular drug-resistance were a subgroup of all women enrolled from September 1989 to September 1994 in WITS, a prospective cohort study of the natural history of HIV-1 infection in pregnant women and their infants, sponsored by the National Institute of Allergy and Infectious Diseases, National Institute of Child Health and Human Development, and the National Institute of Drug Abuse. The subgroup consisted of all women who received ZDV treatment during pregnancy and who were enrolled at WITS clinical sites in New York City, Chicago, Boston/Worcester, Houston, and San Juan (Puerto Rico). Because most of the mothers were enrolled before the treatment recommendations from the AIDS Clinical Trials Group (ACTG) 076, they often received ZDV during pregnancy for their own disease and not to prevent transmission. Women could enroll in WITS at any time during pregnancy or up to 1 week postpartum. Women were seen at study enrollment, 25 weeks and 34 weeks of pregnancy (if enrolled before these times), delivery, postpartum, and every 6 months thereafter. For each mother, data from only the first WITS enrollment pregnancy resulting in a live singleton birth were used for analysis.
RNA extraction, complementary DNA generation and DNA sequencing
Supernatants from HIV cultures at or within 6 weeks before delivery were used to infect donor peripheral blood mononuclear cells (PBMC) and establish new virus cultures according to the ACTG consensus protocol  Of the 203 eligible women, 152 had isolates available for this study. The remainder had no available isolate, either because the isolate was not saved, no culture was performed at the visit of interest, or cultures were negative. RNA was isolated from tissue culture supernatants using the Tri-reagent method as per the manufacturer's directions (Molecular Research Center, Inc., Cincinatti, OH, USA). Reverse-strand cDNA was generated from this viral template using Maloney Murine Leukemia Virus (MMLV) RT under conditions recommended by the manufacturer (Boehringer, Roche Molecular Systems, Alameda, CA, USA). The polymerase chain reaction (PCR) was used to amplify a region of the cDNA corresponding to nucleotides 2550–3321 from the HIV-1 nl43 genomic map (Genbank accession no. M19921) using AmpliTaq enzyme (Perkin-Elmer/Applied Biosystems, Norwalk, CT, USA/Foster City, CA, USA) under recommended conditions. Of 152 specimens shipped for screening, 142 yielded RT sequence data. Reasons for not obtaining evaluable PCR products included no PCR product being generated at all, or that sequencing was incomplete for the total RT region. Standard dideoxy sequencing reactions were carried out using fluorescent dye-labelled primers. Products of the sequencing reaction were then resolved electrophoretically on an ABI 7300 Automated Sequencer and analysed using Sequence Analysis 3.0 software from ABI. Sequence chromatograms were inspected visually for the presence of ZDV drug-resistance mutations at codon positions 41, 67, 70, 210, 215, and 219. Each of these sites was categorized as either wild type, mutant, or mixed. Positions where the minor peak was at least 30% of the height of the major peak (25% of total under-peak area) were categorized as mixtures in this analysis. The presence of mutants or mixtures in a specimen was confirmed by sequencing of the opposite strand.
Linkage to the Women and Infants Transmission Study database
Results from the blinded RT sequencing of maternal viral isolates at delivery were linked to the WITS database maintained at the New England Research Institute (Watertown, MA, USA). Database information on study subjects included: demographics; maternal illicit drug use (cocaine/crack, heroin, methadone, or marijuana); CD4 cell count (cells/μl) and percentage; total lymphocyte number (cells/μl); HIV-1 culture positivity at enrolment and during pregnancy (at each visit, or scored as always versus not always positive during pregnancy); plasma HIV-1 RNA level (viral copies/ml) at delivery; duration of membrane rupture at delivery; infant infection status; information as to whether mothers were treated with ZDV before pregnancy (yes/no) was also included for analysis, because specific information about the mothers' cumulative dose and duration of ZDV treatment and compliance was not available. For WITS analyses, infants are determined to be HIV infected on the basis of two separate positive PBMC HIV cultures; infants were considered to be non-infected on the basis of having all (at least two) negative cultures, at greater than one month of age and the other at equal to or greater than 6 months 
Information on infant HIV-1 infection status was available for 130 out of 142 women for whom RT sequence data were obtained. ZDV drug resistance was defined in two ways: first, viral isolates having any ZDV-associated RT mutation at RT codon positions 41, 67, 70, 210, 215, or 219 were considered resistant. Mothers with isolates having any RT mutation were compared with mothers with wild-type virus. Second, mothers having viral isolates with any mutation or combination of mutations known to be associated with high-level in-vitro phenotypic resistance (multiple mutations at RT codons 41, 67, 70, 215, or 219, or single mutations at codon 215) [10–12] were compared with mothers having isolates with single mutations at RT codons 70 or 210, or who had wild-type virus because the latter two are not associated with high-level phenotypic resistance.
Pairwise comparisons of background demographic, laboratory, and obstetric characteristics between HIV-1 transmitting and HIV-1 non-transmitting mothers were performed by comparing the proportions of both patient groups with selected characteristics (for categorical data), or by comparing the distribution of characteristics (for continuous data) using Wilcoxon rank sum tests.
The association of maternal plasma HIV-1 RNA levels with having ZDV drug-resistant viral isolates was evaluated using non-parametric or parametric methods, as appropriate; median levels of maternal plasma HIV-1 are presented with the associated 25th and 75th percentile values.
Associations of genotypic ZDV drug resistance or other determinants of perinatal transmission with infant infection status were evaluated by simple and multiple unconditional logistic regression analysis. Continuous data were log-transformed before regression, and the OR were expressed per biologically relevant metric, such as twofold higher CD4 cell count, 50 cell higher total lymphocyte count, or twofold greater plasma HIV-1 level. Estimates of association are given by OR with associated 95% CI. The P values reported are for two-tailed comparison. All analyses were performed using statistical software from the SAS Corporation (SAS, Research Triangle Park, NC, USA).
Of the 130 mothers with viral isolates yielding RT sequence data, and for whom data were available on infant HIV status, 26 (20%) transmitted HIV-1 to their infants; eight out of 34 mothers (24.4%) with ZDV-resistant virus transmitted to their infants, and 18 out of 96 (18.8%) mothers with wild-type virus transmitted. In an analysis of viral isolates from 24 of these transmitting mothers and their infants, two women with pure ZDV mutations transmitted these mutants to their infants, whereas three of the five women with mutant/wild-type mixtures transmitted wild-type virus and two transmitted ZDV-resistant virus 
This group of 142 mothers selected for sequence analysis had moderately advanced disease as assessed by immunological and virological parameters (Table 1). The median CD4 cell count at delivery was 315 cells/mm3 and the median CD4 cell percentage was 22%. Seventy-eight per cent of mothers were always HIV-1 culture-positive during pregnancy, and 89% were culture-positive at the time of delivery. The median plasma HIV-1 RNA level at delivery for this group was 24 800 copies/ml. Thirty-two per cent of these mothers had been treated with ZDV before conception, with analysis suggesting that a greater proportion of transmitting mothers (46%) than non-transmitting mothers (29%) had been treated before pregnancy (P = 0.11).
Transmitting mothers had significantly longer durations of membrane rupture before delivery (median of 9.4 h) compared with non-transmitting mothers (median of 1.4 h) (P = 0.01). Apart from differences in the duration of membrane rupture, these two groups of mothers did not vary demographically, clinically, or by levels of immunological or virological factors at delivery when analysed by pairwise comparisons.
Results from HIV-1 RT sequencing are presented in Table 2. Of the 142 mothers with sequence data, 34 (24%) had ZDV-associated drug-resistant RT mutations. Most viral isolates with ZDV drug-resistant mutations had single mutations (26/34, or 76%), with 18 out of 26 or 69% of these single mutations occurring as codon 70 (K ← R substitutions). Of all viral isolate stocks with single drug-resistant mutations, 13 out of 26 or 50% were mixtures of wild-type and mutant virus.
Similar to the findings on viral isolates with single mutations, half (three out of six) of the viral stocks from isolates with double mutations occurred as mixtures of wild-type and mutant RT sequences. In addition, two drug-resistant viral isolates had triple mutations, with one of the isolates being a mixture of wild-type and mutant virus. Seventeen out of 34 (50%) of all viral isolates with RT mutations were thus a mixture of wild-type and mutant virus.
Potential confounders of the association of ZDV drug-resistance at delivery with perinatal transmission are presented in Table 3. The CD4 cell count at delivery was inversely associated with having drug-resistance mutations after adjustment for HIV-1 plasma RNA level at delivery and ZDV use before conception, with a twofold higher level of maternal CD4 cell count at or before delivery being associated with a 45% decrease in the likelihood of having any ZDV drug-resistance mutation (OR: 0.55; 95% CI: 0.36, 0.85; P = 0.006). Alternatively, maternal CD4 cell percentage at or before delivery remained significantly associated with RT mutations after adjustment, with a twofold higher level of maternal CD4 cell percentage being associated with a 50% decrease in the likelihood of having any ZDV drug-resistance mutation (OR: 0.50; 95% CI: 0.30, 0.83; P = 0.007). In addition, the maternal plasma HIV-1 RNA level and self-reported ZDV use before conception were associated with having viral isolates with ZDV drug-resistant RT mutations. Each twofold higher plasma HIV-1 RNA level at delivery was associated with a 13% increase in the risk of having any ZDV-associated drug-resistant mutation (OR: 1.13; 95% CI: 1.01, 1.27; P = 0.03). Finally, maternal ZDV use before conception was strongly associated with mothers having ZDV drug-resistant RT mutations at delivery, after adjustment for potential confounding. Mothers who received ZDV before conception were 3.77 times as likely (95% CI: 1.21, 11.63; P = 0.02) to have virus isolates with mutations than mothers who reported not receiving ZDV.
When analysis was restricted to evaluating the risk of having RT mutations known to confer high-level ZDV phenotypic resistance (multiple mutations at RT codons 41, 67, 70, 215, or 219, or single mutations at codon 215) [10–12] results were similar, with the magnitude of OR being quite similar to results when classifying drug resistance as having any RT mutation. Furthermore, the levels of statistical significance were similar or stronger for these associations (data not shown).
The maternal plasma HIV-1 RNA level at delivery varied by the absence or presence of a ZDV drug resistance mutation (Table 4). Mothers with viral isolates having any drug-resistance mutation had 2.7-fold higher (P = 0.005), whereas those with mutations associated with phenotypic resistance had 3.3-fold higher (P = 0.04) median plasma HIV-1 RNA levels when compared with mothers with virus having no ZDV-resistant mutations.
Finally, the results of logistic regression analysis to identify independent determinants of perinatal transmission are presented in Table 5. Although having viral isolates with ZDV drug-resistance mutations was not associated with transmission (OR: 1.40; 95% CI: 0.55, 3.62) in univariate analysis, adjustment for total lymphocyte number and the duration of membrane rupture revealed an association: mothers with viral isolates having any ZDV drug-resistance mutation had a greater than fivefold risk of perinatal transmission (P = 0.01). Maternal total lymphocyte count at delivery was significantly associated with transmission by itself and after covariate adjustment: each 50 cell higher count was associated with a 6% increase in the risk of transmission (adjusted OR: 1.06; P = 0.01).
Additional analysis of T-cell subsets (CD8+DR+, CD8+CD38+, CD8+CD57+, CD16+CD56+, and CD19+ cell counts) revealed an association between CD8+CD57+ cell counts and transmission: each 50 cell count increase was associated with a 13% increase (OR: 1.13; 95% CI: 1.02, 1.31; P = 0.02), after covariate adjustment. This association was not observed for other T cell subsets that were evaluated. Furthermore when CD8+CD57+ cell counts were subtracted from total lymphocyte counts, the residual lymphocyte counts were not associated with transmission (OR: 1.04; P = 0.22). Taken together, these results suggest that the association between total lymphocyte count and transmission is explained by levels of the CD8+CD57+ T cell subpopulation.
Finally, the duration of membrane rupture was significantly associated (P = 0.02) with transmission in both simple and multiple regression models: each 4 h of increased membrane rupture had an associated 13% increase in the risk of transmission (adjusted OR: 1.13; P = 0.02). In agreement with this finding, when the duration of membrane rupture was considered as a dichotomous variable (≥ 4 h versus < 4 h) [14,15] having a duration of membrane rupture of at least 4 h was associated with a 3.15-fold risk of HIV-1 transmission (P = 0.02).
The CD4 cell count or percentage, syncytium-inducing phenotype, plasma HIV-1 RNA level, or maternal hard drug use were not associated with transmission in this cohort of women. The inclusion of these variables in multiple regression models did not strengthen the magnitude of association of drug-resistance mutations with transmission, nor did they increase the overall model goodness-of-fit.
When regression analysis was restricted to ZDV drug-resistance mutations known to confer high-level phenotypic resistance (multiple mutations at RT codons 41, 67, 70, 215, or 219, or single mutations at codon 215), the magnitude of OR (and associated P values) for maternal–infant transmission and ZDV drug resistance, total lymphocyte count, and duration of membrane rupture were not changed.
This study to evaluate the role of ZDV resistance in maternal–infant transmission included many HIV-infected women who had received ZDV for their own healthcare, before the use of ZDV for the prevention of perinatal transmission. A significant proportion (34/142 or 24%) of mothers in this substudy had developed viral isolates with at least one ZDV-associated RT mutation by the time of delivery.
Low maternal CD4 cell count or percentage, or high plasma HIV-1 RNA level at delivery were associated with mothers having isolates with ZDV-associated RT mutations. These findings corroborate earlier findings that demonstrated increased phenotypic ZDV drug-resistance in patients with low CD4 cell counts or advanced disease [16,17] Because high viral load is associated with the emergence of drug-resistant HIV-1 [18,19] it follows that mothers with the highest levels of circulating virus should have the greatest likelihood of having drug-resistant strains [20,21]
In addition, this analysis demonstrated that a maternal history of ZDV treatment before conception is strongly associated with having maternal drug-resistant virus at delivery. Mothers who were treated with ZDV before conception had a 3.77-fold risk of transmitting when compared with women who did not use ZDV (P = 0.02). This result is in agreement with previously published findings that the length of ZDV treatment is associated with phenotypic  and genotypic  ZDV drug resistance.
ZDV genotypic drug resistance was demonstrated as a strong, independent determinant of perinatal transmission among this group of women with moderately advanced disease, many of whom had been treated with ZDV treatment before conception. Having viral isolates with mutations at codons known to be associated with ZDV phenotypic resistance conferred a greater than fivefold risk of perinatal transmission when compared with mothers having wild-type virus. The association of maternal HIV-1 RT mutations with transmission was strongly confounded by maternal total lymphocyte counts, an independent predictor of perinatal transmission within this cohort.
The findings on the level of maternal HIV-1 RT mutations at delivery (24%), and that maternal HIV-1 RT mutations are associated with increased perinatal transmission, are in contrast to findings from a previously published substudy of ACTG 076  In that study, only 1.6% of the evaluable substudy population (1/61) had RT mutations at study entry, and among women with evaluable specimens at delivery (n = 48), only three (6.3%) had any RT mutation. The presence of maternal HIV-1 RT mutations at delivery was not associated with transmission. The differences in these results can be explained by differences in substudy populations. Whereas the median CD4 cell count was 553 among the 85 women included in the ACTG 076 substudy, their plasma HIV-1 RNA level at entry was 8403 copies/ml, and only seven women (8.2%) had been treated with ZDV before the study, the women included in the present study had more advanced HIV-1 disease (median CD4 cell count: 315 and plasma HIV RNA: 24 800 copies/ml) and 32% had been treated with ZDV before conception. In addition, very low numbers of women with HIV-1 RT mutations at delivery (n = 3) in the ACTG 076 substudy precluded the possibility of any meaningful statistical analysis to evaluate the role of maternal HIV-1 RT mutations in transmission.
Because women with viral isolates having RT mutations were likely to have low total lymphocyte counts, and because low total lymphocyte counts were associated with a decreased risk of perinatal transmission, the association of RT mutations with transmission was negatively confounded to appear much attenuated. Our T cell subset analysis indicates that total lymphocyte count may be a proxy for immune activation and viral burden in this population: CD8+CD57+ cell counts were associated with transmission, and the magnitude of this association was similar to the effect estimated for total lymphocyte counts. Moreover, when individual CD8+CD57+ cell counts were subtracted from total lymphocyte counts, residual lymphocyte counts were no longer associated with transmission. This suggests that it is the CD8+CD57+ T cell subset population that explains the direct association observed for total lymphocyte counts with transmission. In the present study, as was true in the overall analysis for the WITS cohort  the association of CD8+CD57+ cell counts with transmission may thus indicate CD8 cell activation in response to increased viral burden.
The association of CD8/CD57 positive lymphocytes with HIV transmission is biologically plausible. Other authors have shown that CD8/CD57 cell proliferation is associated with chronic HIV  and cytomegalovirus infection  and with the downregulation of cell-mediated cytotoxic responses [27,28] An association of CD8/CD57 cells with HIV RNA copy number or with CD4 cell depletion has been sought and not found 
Although our finding of an association of total lymphocyte counts with transmission could possibly be explained as a Type I error, the level of significance for this association (P = 0.01), in conjunction with its biological plausibility, suggests that this finding is not likely to have occurred by chance.
In contrast, whereas CD4 cell counts and total lymphocyte counts were correlated, CD4 cell counts were not at all associated with transmission (OR: 1.00). Moreover, studies in other patient populations have shown that lower CD4 cell counts, and not higher counts, are associated with perinatal transmission [6,24,30–35]
In addition to total lymphocyte counts, the duration of membrane rupture was an independent determinant of perinatal infection in this cohort. This finding does agree with an earlier report by WITS  and others [15,36,37] that the duration of membrane rupture is an important factor for transmission.
Maternal plasma HIV-1 RNA level, whether considered as a categorical or continuous variable, was not associated with perinatal transmission in the present study. This finding is in contrast with previous studies [2,24,32,33,35,38–42] which have demonstrated the association of maternal viral load with transmission. The present findings suggest that, in this group of women with moderately advanced HIV-1 infection, ZDV drug resistance in the presence of ZDV treatment was the major virological determinant of transmission, and that this effect was not explained by or was independently associated with maternal viral load.
Another explanation for this finding is that the prognostic capacity of plasma HIV-1 RNA may be modified by antiretroviral treatment and the stage of disease. For this study, all women received ZDV during pregnancy, and 76% of those women had drug-sensitive virus at delivery. It would thus be expected that ZDV would suppress viral load in most of the women included in the study. This population of mothers would most often have drug-suppressed levels of plasma HIV-1 RNA, so that other virological and immunological factors would then distinguish transmitting from non-transmitting mothers. In contrast, results from the present study and previously published reports indicate that persons with ZDV drug-resistant virus have higher plasma HIV-1 RNA [21,22] In those studies, one observes higher plasma HIV-1 RNA levels, which would be associated with perinatal transmission. This subgroup variability may explain similar results in previous studies in which maternal viral burden was not found to be prognostic of perinatal transmission in mothers treated with ZDV [36,38]
A similar lack of association of CD4 cell count with vertical transmission was observed in the present study. A lack of effect or the precision to observe the effect of CD4 cell counts may result from the great degree of immune depletion of this group of women (median CD4 cell count 315/mm3) compared with other study populations from previous reports. When compared with previous study populations [23,24,33,35] the women included in the present study would fall into high-risk subgroups (< 400 CD4 cells/mm3) designated in previous reports; thus, one may not observe the increased risk of transmission associated with low CD4 cell count in this group of women because there are few women with higher CD4 cell counts for comparison.
Of additional interest is the finding that among women included in the present substudy who received ZDV treatment of their own disease during pregnancy, the rate of perinatal transmission was 20%, or more than twice the rate reported previously (9.1%) for a similar group of women enrolled in the Ariel Project  Differences in the study populations may well explain differences in the rates of transmission. For instance, almost twice the number of women in the present study reported ZDV treatment before pregnancy (32%) when compared with the group of women in the Ariel Study (17%), potentially leading to higher rates of genotypic resistance and vertical transmission.
The findings of this study suggest that in pregnant women with moderately advanced HIV-1 infection, and among whom many have been previously treated with ZDV before conception, the development of ZDV drug resistance occurs frequently during pregnancy. Furthermore, maternal ZDV drug resistance at delivery is associated with perinatal transmission, independent of other virological, immunological, and gynecological factors, in this group of women.
Nonetheless, experience has shown that ZDV used to reduce the risk of perinatal transmission is highly effective and is the only antiretroviral agent proved in clinical trials to reduce the risk of perinatal HIV-1 infection. In geographical areas with significant levels of endemic HIV-1 infection, and as more intensive combination therapy is being offered for pregnant women for their own healthcare, a consideration of background resistance patterns will need to be taken into account to offer the best therapeutic and perinatal prevention regimens for a pregnant woman and her infant.
1. Connor EM, Sperling RS, Gelber R et al
. Reduction of maternal–infant transmission of human immunodeficiency virus type 1 with zidovudine treatment.
N Engl J Med 1994, 331:1173–1180.
2. Fang G, Burger H, Grimson R et al
. Maternal plasma human immunodeficiency virus type 1 RNA level: a determinant and projected threshold for mother-to-child transmission.
Proc Natl Acad Sci U S A 1995, 92:12100–12104.
3. Frenkel LM, Wagner LE II, Demeter LM et al
. Effects of zidovudine use during pregnancy on resistance and vertical transmission of human immunodeficiency virus type 1.
Clin Infect Dis 1995, 20:1321–1326.
4. Cao Y, Krogstad P, Korber BT et al
. Maternal HIV-1 viral load and vertical transmission of infection: the Ariel Project for the prevention of HIV transmission from mother to infant.
Nat Med 1997, 3:549–552.
5. Melvin AJ, Burchett SK, Watts D et al
. Effect of pregnancy and zidovudine therapy on viral load in HIV-1-infected women.
J Acquired Immune Defic Syndr Hum Retrovirol 1997, 14:232–236.
6. Simpson BJ, Shapiro ED, Andiman WA Reduction in the risk of vertical transmission of HIV-1 associated with treatment of pregnant women with orally administered zidovudine alone.
J Acquired Immune Defic Syndr Hum Retrovirol 1997, 14:145–152.
7. Siegrist CA, Yerly S, Kaiser L, Wyler CA, Perrin L Mother to child transmission of zidovudine-resistant HIV-1 [Letter; Comment].
Lancet 1994, 344:1771–1772.
8. Hollinger FB, Bremer JW, Myers LE, Gold JW, McQuay L Standardization of sensitive human immunodeficiency virus coculture procedures and establishment of a multicenter quality assurance program for the AIDS Clinical Trials Group.
J Clin Microbiol 1992, 30:1787–1794.
9. McIntosh K, Pitt J, Brambilla D et al
. Blood culture in the first 6 months of life for the diagnosis of vertically transmitted human immunodeficiency virus infection.
: The Women and Infants Transmission Study Group.
J Infect Dis 1994, 170:996–1000.
10. Larder BA, Kemp SD Multiple mutations in HIV-1 reverse transcriptase confer high-level resistance to zidovudine (AZT).
Science 1989, 246:1155–1158.
11. Kellam P, Boucher CA, Larder BA Fifth mutation in human immunodeficiency virus type 1 reverse transcriptase contributes to the development of high-level resistance to zidovudine.
Proc Natl Acad Sci U S A 1992, 89:1934–1938.
12. Boucher CA, O'Sullivan E, Mulder JW et al
. Ordered appearance of zidovudine resistance mutations during treatment of 18 human immunodeficiency virus-positive subjects.
J Infect Dis 1992, 165:105–110.
13. Colgrove R, Pitt J, Chung P, Welles S, Japour A Selective vertical transmission of antiretroviral resistance mutations.
AIDS 1998, 12:2281–2288.
14. Minkoff H, Burns DN, Landesman S et al
. The relationship of the duration of ruptured membranes to vertical transmission of human immunodeficiency virus.
Am J Obstet Gynecol 1995, 173:585–589.
15. Landesman SH, Kalish LA, Burns DN et al
. Obstetrical factors and the transmission of human immunodeficiency virus type 1 from mother to child.
N Engl J Med 1996, 334:1617–1623.
16. Richman DD, Grimes JM, Lagakos SW Effect of stage of disease and drug dose on zidovudine susceptibilities of isolates of human immunodeficiency virus.
J Acquired Immune Defic Syndr 1990, 3:743–746.
17. D'Aquila RT, Johnson VA, Welles SL et al
. Zidovudine resistance and HIV-1 disease progression during antiretroviral therapy.
Ann Intern Med 1995, 122:401–408.
18. Coffin JM HIV population dynamics in vivo : implications for genetic variation, pathogenesis, and therapy.
Science 1995, 267:483–489.
19. Coffin JM HIV viral dynamics.
AIDS 1996, 10 (Suppl. 3):S75–S84.
20. Coombs RW, Welles SL, Hooper C et al
. Association of plasma human immunodeficiency virus type 1 RNA level with risk of clinical progression in patients with advanced infection.
J Infect Dis 1996, 174:704–712.
21. Welles SL, Jackson JB, Yen-Lieberman B et al
. Prognostic value of plasma human immunodeficiency virus type 1 (HIV-1) RNA levels in patients with advanced HIV-1 disease and with little or no prior zidovudine therapy.
J Infect Dis 1996, 174:696–703.
22. Japour AJ, Welles S, D'Aquila RT et al
. Prevalence and clinical significance of zidovudine resistance mutations in human immunodeficiency virus isolated from patients after long-term zidovudine treatment.
J Infect Dis 1995, 171:1172–1179.
23. Eastman PS, Shapiro DE, Coombs RW, et al
., for the Pediatric AIDS Clinical Trials Group Protocol 076 Study Group. Maternal viral genotypic zidovudine resistance and infrequent failure of zidovudine therapy to prevent perinatal transmission of human immunodeficiency virus type 1 in Pediatric AIDS Clinical Trials Group 076. J Infect Dis
24. Pitt J, Brambilla D, Reichelderfer P et al
. Maternal immunologic and virologic risk factors for infant human immunodeficiency virus type 1 infection: findings from the Women and Infants Transmission Study.
J Infect Dis 1997, 175:567–575.
25. Borthwick NJ, Bofill M, Gombert WM et al
. Lymphocyte activation in HIV-1 infection.
: II. Functional defects of CD28−T cells.
AIDS 1994, 8:431–441.
26. Gratama JW, Kluin-Nelemans HC, Langelaar RA et al
. Flowcytometric and morphologic studies of HNK-1+ (Leu-7+) lymphocytes in relation to cytomegalovirus carrier status.
Clin Exp Immunol 1988, 74:190–195.
27. Landay A, Larry-Gartland G, Clement LT Characterization of a phenotypically distinct subpopulation of Lue-2+ cells that suppresses T cell proliferate responses.
J Immunol 1983, 131:2757–2761.
28. Joly P, Guillon JM, Mayaud C et al
. Cell-mediated suppression of HIV cytotoxix T lymphocytes.
J Immunol 1989, 143:2193–2201.
29. Mollet L, Sadat-Sowti B, Duntze J et al
. CD8+CD57+T lymphocytes are enriched in antigen-specific T cells capable of down-modulating cytotoxic activity.
Int Immunol 1998, 10:311–323.
30. Newell ML, Dunn DT, Peckham CS, Semprini AE, Pardi G Risk factors for mother-to-child transmission of HIV-1.
: European Collaborative Study [see Comments].
Lancet 1992, 339:1007–1012.
31. European Collaborative Study Vertical transmission of HIV-1: maternal immune status and obstetric factors.
: The European Collaborative Study.
AIDS 1996, 10:1675–1681.
32. Mayaux MJ, Blanche S, Rouzioux C et al
. Maternal factors associated with perinatal HIV-1 transmission: the French Cohort Study: 7 years of follow-up observation.
: The French Pediatric HIV Infection Study Group.
J Acquired Immune Defic Syndr Hum Retrovirol 1995, 8:188–194.
33. Dickover RE, Garratty EM, Herman SA et al
. Identification of levels of maternal HIV-1 RNA associated with risk of perinatal transmission.
: Effect of maternal zidovudine treatment on viral load.
JAMA 1996, 275:599–605.
34. Mandelbrot L, Mayaux MJ, Bongain A et al
. Obstetric factors and mother-to-child transmission of human immunodeficiency virus type 1: the French perinatal cohorts.
Am J Obstet Gynecol 1996, 175:661–667.
35. Coll O, Hernandez M, Boucher CA et al
. Vertical HIV-1 transmission correlates with a high maternal viral load at delivery.
J Acquired Immune Defic Syndr Hum Retrovirol 1997, 14:26–30.
36. 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, 11:429–435.
37. Burns DN, Landesman S, Wright DJ et al
. Influence of other maternal variables on the relationship between maternal virus load and mother-to-infant transmission of human immunodeficiency virus type 1.
J Infect Dis 1997, 175:1206–1210.
38. Sperling RS, Shapiro DE, Coombs RW et al
. Maternal viral load, zidovudine treatment, and the risk of transmission of human immunodeficiency virus type 1 from mother to infant.
N Engl J Med 1996, 335:1621–1629.
39. Weiser B, Nachman S, Tropper P et al
. Quantitation of human immunodeficiency virus type 1 during pregnancy: relationship of viral titer to mother-to-child transmission and stability of viral load.
Proc Natl Acad Sci U S A 1994, 91:8037–8041.
40. Khouri YF, McIntosh K, Cavacini L et al
. Vertical transmission of HIV-1.
: Correlation with maternal viral load and plasma levels of CD4 binding site anti-gp120 antibodies.
J Clin Invest 1995, 95:732–737.
41. Mayaux MJ, Dussaix E, Isopet J et al
. Maternal virus load during pregnancy and mother-to-child transmission of human immunodeficiency virus type 1: the French perinatal cohort studies.
: SEROGEST Cohort Group.
J Infect Dis 1997, 175:172–175.
42. Thea DM, Steketee RW, Pliner V et al
. The effect of maternal viral load on the risk of perinatal transmission of HIV-1.
: New York City Perinatal HIV Transmission Collaborative Study Group.
AIDS 1997, 11:437–444.
Principal investigators, study coordinators, program officers and funding include: Clemente Diaz, Edna Pacheco-Acosta (University of Puerto Rico, San Juan, Puerto Rico; U01 AI 34858); Ruth Tuomala, Ellen Cooper, Donna Mesthene (Boston/Worchester site, Boston, MA, USA; U01 AI 34856); Jane Pitt, Alice Higgins (Columbia Presbyterian Hospital, New York, NY, USA; U01 AI 34842); Sheldon Landesman, Hermann Mendez, Gail Moroso (State University of New York, Brooklyn, NY, USA; HD-8-2913 and OR-1-HD-25714); Kenneth Rich, Delmyra Turpkin (University of Illinois at Chicago, Chicago, IL, USA; U01 AI 34841); William Shearer, Celine Hanson, Norma Cooper (Baylor College of Medicine, Houston, TX, USA: U01 AI 34840); Mary Glenn Fowler, Judy Lew, Elaine Matzen (National Institute of Allergy and Infectious Disease, Bethesda, MD, USA); Anne Willoughby, David Burns, Jack Moye, Jennifer Read, Lynne Mofenson (National Institute of Child Health and Human Development, Bethesda, MD, USA); Katherine Davenny, Vincent Smeriglio (National Institute on Drug Abuse, Rockville, MD, USA); and Sonja McKinley, Les Kalish, Susan Ellis (New England Research Institutes, Watertown, MA, USA; N01 AI 35161).