Wade, Nancy A. MD, MPH*||; Zielinski, Mary A. MA, MSW*; Butsashvili, Maia MD, MS||#; McNutt, Louise-Anne PhD||; Warren, Barbara L. BSN, MPH*; Glaros, Roberta MA*; Cheku, Babu MD†; Pulver, Wendy MS†; Pass, Kenneth PhD‡; Fox, Kathleen RN, MSA¶; Novello, Antonia C. MD, MPH, DrPH§; Birkhead, Guthrie S. MD, MPH* ||
In 1994, the Pediatric AIDS Clinical Trials Group Protocol 076 (PACTG 076) demonstrated that a 3-part regimen of zidovudine (ZDV) prophylaxis administered to HIV-infected pregnant women and their newborns could reduce the risk of perinatal HIV transmission by two thirds. 1 PACTG 076 was a milestone in that it was the first study to demonstrate that perinatal HIV transmission could be prevented with antiretroviral prophylaxis. It also presented a major challenge to public health and medical communities to design effective programs for the prevention of perinatal HIV transmission. This report describes the series of public health policies implemented in New York State (NYS) in response to PACTG 076 to reduce perinatal HIV transmission. It further describes the decline in perinatal HIV transmission that occurred during the first 4 years of the implementation of a comprehensive perinatal HIV prevention program and the factors associated with residual perinatal HIV transmission during this time period (1997–2000).
NYS has the largest number of births to HIV-infected pregnant women in the United States. 2 To address this problem, the NYS Department of Health (NYSDOH) implemented a comprehensive program and policies for all pregnant women designed to reduce perinatal HIV transmission. In 1995, high-risk obstetric sites were funded to provide HIV counseling and testing services in the prenatal setting as well as education and outreach to the medical and consumer community. Comprehensive educational and training materials, including consensus practice guidelines and best practices for physicians, were developed and disseminated to all obstetric and pediatric providers in the state. Medicaid and the AIDS Drug Assistance Program ensured access to antiretroviral therapy (ART). Quality indicators were developed to monitor for adherence to clinical guidelines in perinatal settings.
Since 1996, all birth facilities and regulated prenatal sites have been monitored for compliance with statutory/regulatory requirements. A NYS statute and subsequent regulations required HIV counseling and consented testing in all regulated settings and as a standard of care in all other prenatal sites. Since 1997, universal testing of all newborns has also been required through the Newborn Screening Program (NSP). In 1999, a street level outreach and education program was targeted to high-risk women in areas with the highest HIV seroprevalence and where residual HIV transmissions were detected. 3 Expedited HIV testing (1999) was required of all pregnant women in the labor/delivery setting with consent if they were not HIV tested during the current pregnancy or of the newborn immediately after birth to provide time to initiate preventive therapy. A newborn follow-up program of all HIV-exposed infants documented access to care for more than 99% of HIV-exposed infants. We examined perinatal HIV transmission rates and risk factors for HIV transmission in the population of HIV-positive pregnant women delivering in NYS from February 1997 through December 2000, the first 4 years of NYS’s comprehensive perinatal HIV prevention program in which all HIV exposed newborns were identified through universal HIV testing of the newborn.
Identification of HIV-Exposed Infants
The NYSDOH laboratory, the Wadsworth Center, tests newborn filter paper blood spots obtained on virtually all infants born in NYS for HIV (enzyme-linked immunosorbent assay [ELISA] and confirmatory Western blot) and 7 other congenital metabolic/inherited disorders as part of the state’s NSP. 4 The test results are returned to the designated physician at the hospital of birth and then to the infant’s pediatrician. Pediatricians provide results to the mother/infant and arrange for or refer both for medical care. Because results of such testing are not available until approximately 2 weeks after birth, which is too late to begin preventive care, expedited testing is required on all women/infants whose HIV infection status/exposure has not been determined during pregnancy or in the immediate newborn period. Test results are required within 48 hours. As of November 2003, the time limit for return of a newborn test result is 12 hours. Because all women/newborns are tested, newborn screening serves as surveillance for completeness of case finding.
Medical records from prenatal, intrapartum, newborn, and pediatric care of all HIV-infected women giving birth and their infants were reviewed by the NYS designated peer review agency, IPRO (Lake Success, NY). Standard algorithms and data collection forms were used to document receipt of prenatal care, antiretroviral prophylaxis provided, mode of delivery, diagnosis of infant HIV infection status, and maternal demographic information.
Perinatal HIV Transmission
Infant HIV transmission status was determined from 2 sources. Approximately 90% of HIV-exposed newborns undergo diagnostic testing at least once through the NYSDOH Wadsworth Center HIV DNA polymerase chain reaction (PCR) testing program. 5 In addition, IPRO reviewed the medical records of all seropositive infants when the infant was 6 months of age or older and recorded HIV test results.
For this study, an infant was classified as HIV infected if he/she had at least 1 positive HIV DNA PCR at any time after birth. An infant was classified as HIV uninfected if he/she had a negative PCR test at or after 28 days of age. Approximately 80% of all infants had at least 2 PCR test results after the age of 1 month.
Twin births were treated as a single case; if HIV infection status was discordant, the case was classified as infected. If a woman had a second birth during the study, each pregnancy was treated as a separate event.
Trend analyses were conducted with a χ2 test using Epi Info 2000 software to examine the annual changes (1997–2000) in infant infection status, number of prenatal visits, type and course of prenatal/newborn ART administered, and rate of elective cesarean section delivery. 6
Bivariate analyses were conducted to examine the unadjusted risk of perinatal HIV transmission according to receipt of prenatal care, maternal age, race, geographic region of residence, type and timing of ART, mode of delivery, and birth weight. Relative risks (RRs) and 95% confidence intervals were calculated using Epi Info 2000 software. 6
Multivariate logistic regression was conducted with SPSS, version 8.0, 7 software to assess the association between risk factors while adjusting for potential confounders. Adjusted odds ratios and their confidence intervals were calculated to estimate the adjusted RRs.
From February 1, 1997 through December 31, 2000, 3775 HIV-positive women gave birth in New York. Their HIV-exposed infants were identified by the NSP and were included in the study cohort if the infant infection status was determined (883 [93.8%] mothers/infants in 1997, 944 [92.8%] in 1998, 903 [93.2%] in 1999, and 762 [90.0%] in 2000). Cases with insufficient test results to document HIV infection status were classified as indeterminate and were excluded from the analysis (n = 283 or 7.5% of births to HIV-infected women). One discordant twin pair was identified in the cohort and was included as an infected case.
Mothers/infants excluded from the sample did not significantly differ by age, race, and residence (Table 1). The characteristics of the cohort showed little variation from 1997 through 2000.
Trends in Perinatal Care and HIV Transmission
The proportion of HIV-infected pregnant women who received 3 or more prenatal care visits increased during the study period from 76.4% in 1997 to 85.4% in 2000 (P < 0.05;Fig. 1). An increasing proportion of women received ART at some time during pregnancy, delivery, or during the newborn period (63.8% in 1997, 96.9% in 2000; P < 0.05); receipt of intrapartum and newborn antiretroviral regimens increased from 48.5% in 1997 to 74.2% in 2000 (P < 0.05). Receipt of combination ART in the prenatal period increased from 16.2% in 1997 to 67.2% in 2000 (P < 0.05). Elective cesarean section deliveries increased sharply from 6.5% in 1997 to 35.2% in 2000 (P < 0.05). Perinatal HIV transmission decreased each year (11.0% in 1997, 8.6% in 1998, 7.1% in 1999, and 3.7% in 2000; P < 0.05; see Fig. 1).
Factors Associated With Perinatal HIV Transmission
Based on bivariate analysis of HIV-positive women delivering, white women had a higher risk of transmission (10.4%, RR = 1.5, 95% confidence interval: 1.0–2.1) compared with black women (7.0%;Table 2). Women over the age of 40 years were 2.2 times more likely to transmit HIV (13.9%, RR = 2.2, 95% confidence interval: 1.3–3.6) than women aged 20 to 29 years (6.3%;Table 3). Residents of New York City (NYC) had a higher risk of transmission than HIV-positive women living outside NYC (8.4%, RR = 1.86, 95% confidence interval: 1.3–2.7).
Women who received no prenatal care were significantly more likely to transmit infection (16.3%, RR = 2.7, 95% confidence interval: 2.0–3.7) compared with women with 3 or more prenatal visits (6.1%; see Table 3).
Women with infants of low birth weight (1500–2499 g) and very low birth weight (<1500 g) were significantly more likely to transmit HIV. Women with infants of very low birth weight had the highest risk of transmission (17.6%, RR = 2.8, 95% confidence interval: 1.8–4.2; see Table 3).
Mode of Delivery
Vaginal delivery was associated with a higher rate of HIV transmission (9.0%) compared with elective cesarean delivery (3.2%, RR = 0.4, 95% confidence interval: 0.2–0.6). The transmission rate following nonelective cesarean section delivery was 8.2%, comparable to vaginal delivery (see Table 3).
The transmission rate was lowest among women who received prenatal combination ART with a protease inhibitor (2.5%, RR = 0.2, 95% confidence interval: 0.1–0.2) or without a protease inhibitor (2.3%, RR = 0.1, 95% confidence interval: 0.1–0.2) compared with no antiretroviral prophylaxis (16.7%). Prophylaxis with a single nucleoside analogue was also protective (5.8%, RR = 0.4, 95% confidence interval: 0.2–0.5) compared with no prophylaxis (see Table 3).
Timing of Perinatal Antiretroviral Prophylaxis
A 3-part antiretroviral regimen administered in the pre-natal, intrapartum, and newborn periods was associated with the lowest rate of transmission (3.3%, RR = 0.2, 95% confidence interval: 0.1–0.2) compared with other regimens. Abbreviated antiretroviral regimens were also associated with reductions in transmission (intrapartum/newborn period [9.4%, RR = 0.5, 95% confidence interval: 0.3–0.8] or only in the newborn period [11.9%, RR = 0.6, 95% confidence interval: 0.4–0.9]) compared with no therapy (19.4%; see Table 3).
Among the 38 infants who were HIV infected and who received abbreviated antiretroviral regimens, 37 received only ZDV (intrapartum/newborn or newborn only). One infant received only indinavir as a neonate.
Among 302 infants who were exposed to HIV and not infected and received abbreviated antiretroviral regimens, 133 of their 135 mothers received ZDV alone during the intrapartum period and 122 of 135 infants received only ZDV in the newborn period. Additional newborn regimens included ZDV/lamivudine (3TC; n = 7), ZDV/nevirapine (n = 2), nelfinavir (n = 1), and unknown antiretroviral treatment (n = 3).
Among 177 newborns who received only newborn ART, 157 received only ZDV, with the remaining receiving ZDV/3TC (n = 6), ZDV/nevirapine (n = 10), and an unknown antiretroviral (n = 4).
Multivariate Logistic Regression
A multivariate logistic regression model was developed to assess risks of vertical transmission while adjusting for other known or suspected risk factors. Factors related to the probability of HIV transmission at the bivariate level were used as independent variables in the model and included maternal demographics (age, race, and residence), the course of ART, the mode of delivery, and the infant’s birth weight. The number of prenatal visits and type of prenatal ART were omitted from the model due to a high correlation with the course of ART. Maternal age greater than 40 years (compared with women aged 20–29 years), white and Hispanic race/ethnicity (compared with black race/ethnicity), low or very low birth weight (compared with infants with normal birth weight), and residence in NYC compared with the rest of state were significantly associated with risk of vertical transmission. Abbreviated regimens of ART were also significantly associated with reductions in transmission compared with no treatment. Elective cesarean section delivery appeared to be associated with a lower probability of HIV transmission but was barely significant at the level of P < 0.05 (see Table 2).
This population-based observational study of 3492 HIV-infected pregnant women and their infants describes a substantial decline in perinatal HIV transmission in NYS over a 4-year period. A 1998 Institute of Medicine publication, Reducing the Odds Preventing Perinatal HIV Transmission, describes the steps on the pathway to perinatal HIV transmission and emphasizes the need for identification of women in the prenatal period, providing access to prenatal care, and access to ART. 8 In NYS, public health efforts have focused on intervening at all possible steps in the pathway to transmission so as to reduce perinatal transmission maximally by providing access to care for pregnant women, ensuring that they are offered testing and that providers are trained in the appropriate prophylaxis necessary to reduce transmission. Similarly, recent guidance from the Centers for Disease Control and Prevention calls for routine HIV testing of all pregnant women and for infants whose mother’s HIV infection status is unknown so as to decrease perinatal transmission. 9 Rapid HIV testing in the labor and delivery setting as well as in the postpartum setting if the mother is not previously tested is also an option.
Since 1994, when the PACTG 076 regimen became a standard of care, significant reductions in transmission have been reported in other studies and in well-defined populations. 10–12 These NYS data demonstrate that the reductions in transmission observed in research studies can be achieved on a population basis when women receive prenatal care and have access to appropriate ART in the prenatal, intrapartum, and newborn periods. NYS public health measures were instituted to ensure access to care, including HIV counseling and testing and ART prophylaxis if HIV infection was detected.
Since Public Health Service guidelines recommending treatment of the HIV-infected pregnant woman for her own health status were promulgated in 1998, the use of prenatal combination ART has increased. 13 The rates of transmission observed in our study among women receiving combination therapy are comparable to the transmission rates described in PACTG 367 (combination with a nucleoside reverse transcriptase inhibitor [0% vs. 2.9%] and combination with a protease inhibitor [1.8% vs. 2.7%]). 14 Similar rates were reported in the Woman-Infant Transmission Study (multiple antiretroviral agents [4.3%] and highly active ART [0%]). 15
Reductions in perinatal HIV transmission have occurred even when abbreviated regimens of ZDV were administered to the HIV-infected pregnant woman and her infant. In our previous analysis of NYS data, significant reductions in risk of perinatal HIV transmission were described among women initiating antiretroviral prophylaxis during the prenatal period, intrapartum period, and within the first 24 to 48 hours of a newborn’s life. 16 These findings were confirmed in this study, with significant reductions in transmission described even when antiretroviral prophylaxis was initiated in the intrapartum (9.0%) or newborn (11.9%) periods compared with no antiretroviral prophylaxis (19.3%). In other studies, decreased perinatal transmission has been described with prophylaxis administered in the intrapartum and newborn periods. 17,18 This study also describes reductions in transmission when ART was initiated in the newborn period. This finding is similar to that from a recent study from Malawi, where efficacy of ZDV/nevirapine administered in the newborn period was demonstrated. 19
The finding of higher transmission rates among white women compared with black women was unexpected. White women were more likely to receive prenatal care and were more likely to have 3 or more prenatal visits. White women were less likely to have low or extremely low birth weight infants. White women were also more likely to receive a 3-part regimen of ART and combination ART during the prenatal period. All these relationships were statistically significant. Black and white women did not differ significantly for age, receipt of any ART, mode of delivery, or when the woman found out that she was infected with HIV. The lower rates of HIV transmission in black women remain unexplained; indeed, the factors described would have been expected to decrease the transmission rate among white women. It is possible that black versus white women may have differed by other indicators that we were not able to account for adequately, including, for example, maternal viral load.
The finding of a higher rate of transmission among residents of NYC compared with the rest of the state may be explained in part by some additional risk factors. Women in NYS were less likely to have received prenatal care or to have had 3 or more prenatal visits. Women residing in NYC were as likely as women throughout the rest of the state to have received ART at some time during their pregnancy or in the newborn period. They were less likely to have received a full 3-part regimen of ART during the prenatal, intrapartum, and newborn periods, however. In addition, NYC women were less likely to deliver by cesarean section and were more likely to have delivered a low or extremely low birth weight infant. These findings would support the differences seen in transmission.
There are several limitations to this study. We did not have 100% ascertainment of infant infection status; however, the number was small and not expected to change results significantly. The study did not include maternal risk factors, plasma HIV RNA levels, or CD4 counts. Plasma HIV RNA levels may have identified women most at risk for transmission. Adherence to treatment regimens was not measured. In addition, because this was not a controlled clinical trial, dosage of medication, dosing intervals, and start dates during pregnancy may have varied. Because this was an observational study and not a clinical trial, there were many uncontrolled factors. Nevertheless, there were some important strengths. The study cohort was large and population based. Indeed, it was close to a universal cohort with complete information over 4 years.
NYS has taken major steps toward reducing perinatal HIV transmission. One benefit has been to dramatically increase the number of women being tested for HIV during prenatal care, when the full 3-part ART regiment can be given. Expedited delivery setting testing remains a critical “safety net” for women who are not tested perinatally, mostly because they did not receive prenatal care. Statutes and regulations with policies and programs have ensured access to prenatal care and receipt of ART to reduce the risk of perinatal HIV transmission. These factors coupled with clinical advances will continue to provide an environment for the lowest possible rates of transmission.
We are working to identify factors in the small number of residual transmissions that continue to occur. Lack of prenatal care remains a key issue preventing the administration of ART during pregnancy, intrapartum, and postpartum. The NYSDOH has embarked on an ambitious street level outreach targeting the highest risk women to help them navigate the medical and social service system, including accessing prenatal care. Finally, NYS is continuing a comprehensive educational program aimed at prenatal care providers and consumers to ensure that the community is knowledgeable about the most effective regimens for reducing perinatal transmission to the lowest possible levels. This combination of ongoing efforts will be necessary to sustain and further reduce the rate of perinatal HIV transmission to the lowest possible level.
1. Connor EM, Sperling RS, Gelber R, et al. Reduction in maternal-infant transmission of human immunodeficiency virus type I with zidovudine treatment. N Engl J Med. 1994;331:1173–1180.
2. National HIV Prevalence Surveys. 1997 Summary. Atlanta: National Center for HIV, STD, and TB, Centers for Disease Control and Prevention; 1997:17.
3. Official Compilation Codes Rules and Regulations of the State of New York. Tit. 10 98.2 (c) 405.21 (c) (8) (h) 751.5 (a) (16) and 754.7 (c) (3); 1996.
4. Novick LF, Berns D, Stricof R, et al. HIV prevalence in newborn infants in New York State. JAMA. 1989;261:1745–1750.
5. Charbonneau TT, Wade NA, Weiner L, et al. Vertical transmission of HIV in New York State: a basis for statewide testing of newborns. AIDS Patient Care STDS. 1997;11:227–236.
6. Epi Info 6 [computer program]. Version 6.04b. Atlanta: Centers for Disease Control and Prevention, and Geneva: World Health Organization; January 1997.
7. Version SPSS 8 for Windows [computer program]. Chicago: SPSS; 1998.
8. National Research Council. Reducing the Odds: Preventing Perinatal Transmission of HIV in the United States. Washington, DC: National Academy Press; 1999.
9. Advancing HIC prevention. New strategies for a changing epidemic—United States. Centers for Disease Control and Prevention. MMWR Morb Mortal Wky Rep. 2003;52:329–331.
10. Cooper ER, Nugent RP, Diaz C, et al. After AIDS Clinical Trial 076: the changing pattern of zidovudine use during pregnancy, and the subsequent reduction in vertical transmission of human immunodeficiency virus in a cohort of infected women and their infants. J Infect Dis. 1996;174:1207–1211.
11. Fiscus SA, Adimora AA, Schoenbach VJ, et al. Trends in human immunodeficiency virus (HIV) counseling and testing and antiretroviral treatment of HIV-infected women and perinatal transmission in North Carolina. J Infect Dis. 1999;180:99–105.
12. Mayaux MJ, Teglas JP, Mandelbrot L, et al. Acceptability and impact of zidovudine for prevention of mother-to-child human immunodeficiency virus-1 transmission in France. J Pediatr. 1997;131:857–862.
13. Public Health Service Task Force recommendations for the use of antiretroviral drugs in pregnant women with HIV-1 for maternal health and for reducing HIV-1 transmission in the United States. Centers for Disease Control and Prevention. MMWR Morb Mortal Wky Rep. 1998;47:1–30.
14. Tuomala RE, Shapiro DE, Mofenson LM, et al. ART during pregnancy and the risk of an adverse outcome. N Engl J Med. 2002;346:1863–1870.
15. Cooper ER, Charurat M, Mofenson I, et al. Combination antiretroviral strategies for the treatment of pregnant HIV-1–infected women and prevention of perinatal HIV-1 transmission. J Acquir Immune Defic Syndr. 2002;29:484–494.
16. Wade NA, Birkhead GS, Warren BL, et al. Abbreviated regimens of zidovudine prophylaxis and perinatal transmission of the human immunodeficiency virus. N Engl J Med. 1998;339:1409–1414.
17. Guay LA, Musoke P, Fleming T, et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomized trial. Lancet. 1999;354:795–802.
18. Saba J, on behalf of the PETRA Trial Study Team. Interim analysis of early efficacy of three short ZDV/3TC combination regimens to prevent mother-to-child transmission of HIV-1: the PETRA trial [abstract S-7]. Presented at the 6th Conference on Retroviruses and Opportunistic Infections, Chicago, January 1999.
19. Taha TE, Kumwenda NI, Gibbons A, et al. Short post-exposure prophylaxis in newborn babies to reduce mother-to-child transmission of HIV-1: NZAZ randomized clinical trial. Lancet. 2003;362:1171–1177.
© 2004 Lippincott Williams & Wilkins, Inc.