Use of enhanced perinatal human immunodeficiency virus surveillance methods to assess antiretroviral use and perinatal human immunodeficiency virus transmission in the United States, 1999-2001

Article Outline

• Abstract
• Materials and Methods
• Results
• Comment
• References
• Copyright

Objective

Significant reductions in perinatal human immunodeficiency virus (HIV) transmission have been demonstrated in which the HIV-infected mothers and their HIV-exposed infants receive prenatal, intrapartum, and neonatal antiretroviral therapy.

Study Design

We used data that were collected through the Enhanced Perinatal Surveillance system for HIV-exposed singleton births that occurred 1999-2001 in 24 sites.

Results

The overall infant infection rate for the 3 years was 4.7%. Compared with zidovudine monotherapy, those patients who received zidovudine with other drugs that included a protease inhibitor and those who received zidovudine and other drugs with no protease inhibitor were less likely to have an infected infant (adjusted odds ratio, 0.4 [95% CI, 0.3-0.07]; adjusted odds ratio, 0.5 [95% CI, 0.3-0.8], respectively).

Conclusion

These data support the current treatment recommendations and show that infants were less likely to be infected when the mothers were given a prenatal antiretroviral therapy regimen that contained zidovudine with additional antiretroviral drugs with or without a protease inhibitor in addition to receiving antiretrovirals during delivery and neonatally.

Key words: antiretroviral, HIV/AIDS, perinatal, surveillance

In 1994, the results of a randomized trial Pediatric AIDS Clinical Trial Group (PACTG) 076 demonstrated that an intensive regimen of maternal zidovudine from the second trimester through the intrapartum period that was followed by 6 weeks of infant zidovudine could reduce perinatal human immunodeficiency virus (HIV) transmission by almost 70%.¹ In response to these striking results, the United States Public Health Service rapidly issued recommendations for the use of zidovudine for the reduction of perinatal HIV transmission in 1994.² As a result of these findings, in 1995, the Centers for Disease Control and Prevention (CDC) issued recommendations for the universal counseling and voluntary testing of all pregnant women to ensure that all HIV-infected pregnant women and their newborn infants have access to this perinatal HIV-prevention regimen.³ Revised recommendations for HIV screening of pregnant women were issued in 2001⁴ and in 2006⁵ that reemphasized the importance of universal voluntary screening of all pregnant women, the need to reduce barriers to healthcare providers offering screening by the use of an opt-out approach, the importance of retesting in the third trimester of pregnancy in some jurisdictions, and rapid testing at labor and delivery for women whose HIV status was still unknown. Since the publication of the PACTG 076 study results, other observational studies in the United States have confirmed dramatic decreases in perinatal transmission.⁶, ⁷, ⁸, ⁹ Perinatal cohort and clinical trial data from 1997 to the present time suggests that zidovudine, when combined with other antiretrovirals, and obstetric interventions such as scheduled cesarean section before labor onset and avoidance of breast feeding, can reduce perinatal HIV transmission to ≤2%.¹⁰, ¹¹

In the United States, 91% of all reported pediatric acquired immunodeficiency syndrome (AIDS; required reporting in all states) cases and 85% of all pediatric HIV cases that were reported from the 35 states and 4 territories with HIV surveillance are attributed to perinatal transmission of HIV.¹² Pediatric AIDS incidence has declined by 94% from 1992-2003¹²; most of this decline is due to the reduction in the number of cases that were attributed to perinatal HIV transmission. Despite this decline, an estimated 280-370 perinatally infected infants were born in 2000.¹³ Additionally, more than one-half of perinatal HIV infections in children were attributed to missed opportunities for perinatal HIV prevention.⁹

Data on the effect of antiretroviral use and its impact on perinatal HIV infant infection rates from a population-based national enhanced perinatal surveillance (EPS) system have not been described previously in the literature. Therefore, the objectives of this article were to describe recent surveillance trends in antiretroviral use for the prevention of perinatal HIV transmission in the United States and to assess the impact of antiretroviral prophylaxis on perinatal HIV transmission rates with the use of data from the EPS system from 19 states and territories and 5 US cities for HIV-exposed infants who were born in 1999, 2000, and 2001.

Back to Article Outline

Materials and Methods 

The EPS project constitutes an integrated surveillance system for HIV-infected mothers and their perinatally exposed infants and was designed as an extension of routine surveillance for HIV/AIDS.

State and local health departments that conducted surveillance of adult and pediatric HIV infection also conducted the EPS project, which consisted of enhanced case ascertainment of HIV-exposed infants and HIV-infected mothers. These enhanced surveillance methods consisted of (1) increased efforts to completely ascertain mother-infant pairs by matching birth registries to HIV/AIDS surveillance registries and (2) the systematic collection of supplemental data from multiple sources for both the mother and infant from all available medical charts, which included the maternal HIV clinic; prenatal and labor/delivery medical records; the newborn infant and pediatric records; data from laboratory reports; and data from birth and death certificates.

State health departments and the 6 independently funded city health departments that received federal funds for HIV/AIDS surveillance that have reported ≥60 HIV-positive women who gave birth per year (as determined from the 1994 Survey of Childbearing Women [SCBW]) were eligible for funding for the EPS project. The study sites that were funded through the cooperative agreement mechanism for 2003 were Alabama, California, Chicago, Los Angeles, Connecticut, District of Colombia, Florida, Houston, Louisiana, Maryland, Michigan, Mississippi, North Carolina, New Jersey, New York State, New York City, Ohio, Pennsylvania, Philadelphia, Puerto Rico, South Carolina, Tennessee, Texas, and Virginia. These sites represented 89% of all perinatal AIDS cases that were reported in 2003.

The infants who were identified through enhanced surveillance were followed up by the health department every 6 months until their HIV infection status was determined. Data were collected on HIV-infected mothers and their HIV-exposed infants who were born in 1999, 2000, and 2001. Women who were known to be HIV-infected in pregnancy (tested before or at delivery) and women who were not known to be HIV-infected in pregnancy but whose child was reported to surveillance because of a positive laboratory test that was indicative of HIV were included. All infants who were born in or had received care in the specific project site (ie, the state or city) were eligible for inclusion. For women who gave birth more than once during the time period, a separate confidential HIV/AIDS case report and EPS abstraction form were completed. If a woman had twins or triplets, a separate confidential HIV/AIDS case report form and EPS abstraction form were completed for each infant.

Variables that were used in the analysis included year of birth, maternal age, race/ethnicity, prenatal care, maternal HIV testing, and receipt of prophylactic antiretroviral therapy (ART). Prenatal care was defined as a dichotomous variable with “yes” being coded as the receipt of any prenatal care visits. Maternal HIV testing was defined as the relationship between maternal testing for HIV and the birth of the exposed infant. Receipt of prophylactic ART was based on the 3 time periods when ART could have been received: prenatally, during delivery, or neonatally during the first 6 weeks of life. Zero (0) arms was defined as no ART received during any of the 3 time periods; 1 arm was defined as ART received at only 1 of the 3 time periods; 2 arms was defined as ART received 2 of the 3 time periods; and 3 arms was defined as ART received during all 3 time periods. For the outcomes in which we assessed receipt of ART, we used the number of HIV-infected women for whom data on antiretroviral use were available and defined it as known receipt or known nonreceipt of antiretroviral drugs. Prenatal ART regimens were categorized in the following manner: (1) zidovudine monotherapy; (2) zidovudine and other drugs with a protease inhibitor (PI; zidovudine in combination with other drugs at least 1 of which was a PI); (3) zidovudine and other drugs with no PI (zidovudine in combination with other drugs in which none were PIs); (4) other drugs with PI, no zidovudine (an ART combination that includes a PI but does not include zidovudine); (5) other drugs with no PI, no zidovudine (an ART combination that does not include a PI or zidovudine); and (6) monotherapy, no zidovudine (an ART regimen with 1 drug that was not zidovudine). For defining HIV infection in perinatally exposed infants, we used the revised CDC surveillance HIV case definition for adults and children to classify children as infected with HIV, not infected with HIV, or indeterminate.¹⁴

All analyses were conducted with SAS software (version 9.1; SAS Institute Inc, Cary, NC). There were 8530 singleton births reported through the EPS system for children who were born 1999-2001. Of the 8530 singleton births, 1186 birth records contained incomplete data on ART use, which resulted in a sample size of 7344 births. To accommodate the logistic regression analysis, an additional 347 records were excluded because of missing or unknown data on maternal age at delivery, maternal race/ethnicity, receipt of any prenatal care, maternal timing of HIV test, and type of delivery, which resulted in a sample size of 6997. Because of the small sample sizes and to accommodate the logistic regression model, we further excluded 23 records of women among whom 10 women received prenatal ART only, 3 women received intrapartum ART only, and 10 women received prenatal and intrapartum ART, which resulted in a sample size of 6974 records.

Chi-square test for trend was used to test for trends in infant HIV infection status from 1999-2001. Stratified analysis was used to examine the univariate relationship between demographic variables and infant HIV infection. Unadjusted odds ratios (OR) and 95% CIs were used to assess the relationship between each variable and the outcome of infant HIV infection status. Unconditional logistic regression was used to produce ORs and 95% CIs for the relationship between covariates and infant HIV-infection status.

This project was deemed exempt from CDC Institutional Review Board review because it was determined that the project constitutes data collection for the purposes of disease surveillance and program evaluation and is not research. For this project, data were abstracted from existing medical and ancillary records. There was no contact with individual patients. Data were collected by trained health department personnel or their agents, and no names of individuals were reported to CDC.

Back to Article Outline

Results 

There were 8530 singleton births reported through EPS for children who were born 1999-2001. The overall infant infection rate for the 3 years was 4.7% (95% CI, 4.2,5.1). The infant infection rates for 1999, 2000, and 2001 were 6.0% (95% CI, 5.1, 6.9), 4.2% (95% CI, 3.5, 4.9), and 3.9% (95% CI, 3.2, 4.6), respectively. The infant infection rates declined by year, but the trend over time was not statistically significant (P = .0680).

Of the 8530 singleton births, 1186 records (13.9%) contained incomplete data on ART use, which resulted in a sample size of 7344 births. Receipt of arms of ART by infection status is presented in Table 1. Eighty-two percent of the mother-infant pairs (n = 6029) received 3 arms of ART. The infant infection rate among infants who received 3 arms of ART was 2.5% (95% CI, 2.1, 2.8). Of the 7344 mother-infant pairs, 638 pairs (8.7%) received 2 arms of ART; 359 pairs (4.9%) received 1 arm of ART, and 292 pairs (4.0%) received 0 arms of ART. The infant infection rates among the pairs who received intrapartum and neonatal ART and neonatal-only ART were 4.7% (95% CI, 2.4%, 7.0%) and 11.1% (95% CI, 7.9%, 14.3%), respectively. Among mother-infant pairs who received no ART at any time (0 arms of ART), the infant infection rate was 25.7% (95% CI, 20.7%, 30.7%).

TABLE 1. Arms of antiretroviral prophylaxis received by infant infection status, singleton births to HIV-infected mothers, 24 sites, United States, 1999-2001 (n = 7344)

Demographic characteristics by infant infection status are shown in Table 2. Mothers 13-19 years of age were more likely to have an infected infant (OR, 1.7; 95% CI, 1.1-2.6), but there were no differences in infant infection status by race/ethnicity. Ninety-four percent of mothers received some prenatal care; however, mothers with no prenatal care were more likely than mothers with prenatal care to have an infected infant (OR, 2.4; 95% CI, 1.7- 3.5). Most of the mothers (92%) were tested before or during the pregnancy. Women who were tested for HIV at delivery or after birth were more likely than women who were tested for HIV before and during pregnancy to have an infected infant (OR, 3.1; 95% CI-1.9, 4.9; OR, 8.1; 95% CI, 5.8, 11.3).

TABLE 2. Demographic and clinical characteristics of singleton births to HIV-infected mothers, 24 sites, United States, 1999-2001 (n = 6997)

	Infected		Total
Characteristic	N	%	N	Unadjusted OR	95% CI
Total	259	3.7	6997
Year of birth
1999	98	4.4	2240	1.3	1.0-1.8
2000	86	3.4	2540	1.0	0.7-1.4
2001	75	3.4	2217	1.0	Referent
Maternal age at delivery (y)
13-19	26	5.1	509	1.7	1.1-2.6
20-29	114	1.6	3641	1.0	Referent
30-39	110	4.2	2652	1.3	1.0-1.7
≥40	9	4.6	195	1.5	0.7-3.0
Maternal race/ethnicity
White	27	3.3	826	1.0	Referent
Black	178	3.6	4887	1.1	0.7-1.7
Hispanic	53	4.3	1224	1.3	0.8-2.1
Other	1	1.7	60	0.5	0.1-3.8
Receipt of any prenatal care
No	34	7.9	431	2.4	1.7-3.5
Yes	225	3.4	6566	1.0	Referent
Maternal timing of HIV test
Refused test	1	33.0	3	17.1	1.5-189.7
Tested HIV+ before or during pregnancy	183	2.8	6450	1.0	Referent
Tested HIV+ at delivery	22	8.3	266	3.1	1.9-4.9
Tested HIV+ after birth	53	19.1	278	8.1	5.8-11.3
Type of delivery
Cesarean section	113	3.1	3678	0.7	0.5-0.9
Vaginal	146	4.4	3319	1.0	Referent
Arms received
0 arms	59	22.3	265	11.6	8.3-16.2
1 arm: neonatal ART only	38	11.0	344	5.0	3.4-7.3
1 arm: intrapartum ART only	0	0.0	3	5.7	0.3-111.9
1 arm: prenatal ART only	1	10.0	10	4.5	0.6-35.7
2 arms: prenatal ART & intrapartum ART	0	0.0	10	1.9	0.1-32.9
2 arms: prenatal ART & neonatal ART	8	2.7	302	1.1	0.5-2.3
2 arms: intrapartum ART & neonatal ART	14	4.6	306	1.9	1.1-3.4
3 arms: prenatal ART & intrapartum ART & neonatal ART	139	2.4	5757	1.0	Referent

The logistic model that examined the relationship of arms of ART that was received and infection status of infants is presented in Table 3. When we controlled for maternal age at delivery, maternal race/ethnicity, receipt of prenatal care, timing of maternal HIV test, and delivery type, the receipt of ART by arms remained statistically significant. In addition, study site (not shown) was an independent risk factor for infants who became infected and maternal age of ≥40 was a significant protective factor for infants not becoming infected.

TABLE 3. Final logistic regression model^⁎ examining arms of treatment and infant infection status (n = 6974)

Variable	OR	95% CI
Maternal age (y)
13-19	1.1	0.7-1.8
20-29	1.0	Referent
30-39	1.1	0.5-2.3
40+	0.7	0.5-0.9
Race
White	1.0	Referent
Black	1.0	0.6-1.5
Hispanic	1.3	0.8-2.2
Other	0.4	0.1-3.2
Prenatal care
No	1.0	0.6-1.6
Yes	1.0	Referent
Timing of maternal HIV test
Refused test	3.2	0.2-41.5
Before/during pregnancy	1.0	Referent
At delivery	1.5	0.9-2.6
After birth	0.8	0.4-1.5
Delivery type
Cesarean section	1.1	0.8-1.4
Vaginal	1.0	Referent
Arms received
0 arms	10.7	6.1-18.9
1 arm: neonatal ART only	6.2	3.6-10.6
2 arms: intrapartum ART & neonatal ART	2.0	1.1-3.7
2 arms: prenatal ART & neonatal ART	1.1	0.5-2.4
3 arms: prenatal ART & intrapartum ART & neonatal ART	1.0	Referent

Compared with mother-infant pairs who received 3 arms of ART, the odds for becoming infected were 10.7 (95% CI, 6.1, 18.9), 6.2 (95% CI, 3.6, 10.6), and 2.0 (95% CI, 1.1, 3.7), respectively, for those who received 0 arms of ART, 1 arm (neonatal only), and 2 arms (only intrapartum and neonatal arms). Among mother-infant pairs who received only the neonatal arm of ART and for whom we have data on the time of the administration of ART after birth (n = 100), approximately one-third of the pairs (32%) received ART ≤12 hours after birth; 20% of the pairs received ART between 13 and 24 hours after birth; 23% of the pairs received ART between 25 and 48 hours after birth, and 25% of the pairs received ART >48 hours after birth. The transmission rates for each group relative to the timing of the administration of the ART was 6.3%, 10%, 13%, and 16%, respectively; however, the sample sizes were small, and the precision of these estimates is limited.

Among the 5522 mother-infant pairs who received 3 arms of ART, we examined the ART regimens by infant infection status. Ninety percent of the pairs who received 3 arms of ART received zidovudine each of the 3 time periods. Of the 4976 pairs who received zidovudine each of the 3 time periods and who received 3 arms of ART, 15% of the pairs received zidovudine monotherapy; 42% of the pairs received zidovudine in combination with other drugs that included a PI; and 42% of the pairs received zidovudine in combination with other drugs that did not include a PI (data not shown).

Among the pairs who received zidovudine monotherapy during the intrapartum and neonatal periods, receipt of the type of ART during the prenatal period varied. The receipt of zidovudine monotherapy during all 3 time periods occurred in 710 mother-infant pairs (12.9%). Of the mothers (62%) who received prenatal zidovudine in combination with other drugs, 31% received combinations with PIs, and 31% received combinations without PIs (data not shown).

The receipt of prenatal zidovudine in combination with other drugs that included a PI, with intrapartum zidovudine monotherapy and neonatal zidovudine monotherapy, occurred in 1696 cases of the study population (31%). We also examined prenatal ART regimens and infection status among mother-infant pairs who received 3 arms of treatment (Table 4). When we controlled for maternal age, race, prenatal care, delivery type, and timing of HIV test, the prenatal ART regimens remained an independent predictor of infant infection status. The odds of a mother having an infected infant were less for those mothers who received either a prenatal regimen that contained zidovudine in combination with other drugs that included at least 1 PI or a prenatal regimen that contained zidovudine in combination with other drugs without any PIs, compared with mothers whose prenatal regimen consisted of zidovudine monotherapy (OR, 0.40; 95% CI, 0.3-0.7; OR, 0.50; 95% CI, 0.3-0.8, respectively). Other non–zidovudine-containing prenatal regimens demonstrated a similar effect on infant status; however, none of them was statistically significant. Two hundred thirty-seven women were prescribed combination ART that contained a PI but no zidovudine; 135 women were prescribed combination ART regimen that contained no PI and no zidovudine.

TABLE 4. Final Logistic Regression Model^⁎ examining prenatal ART regimens and infant infection status among mother-child pairs who received 3 arms of treatment (n = 5602)

Variable	OR	95% CI
Year of birth
1999	1.0	Referent
2000	0.7	0.5, 1.1
2001	0.7	0.4-1.1
Maternal age (y)
13-19	1.2	0.7-2.1
20-29	1.0	Referent
30-39	0.8	0.3-2.7
40+	0.7	0.5-1.0
Race
White	1.0	Referent
Black	1.0	0.6-1.7
Hispanic	1.1	0.6-2.2
Other	0.0	0.0-999.9
Delivery type
Cesarean section	1.1	0.8-1.5
Vaginal	1.0	Referent
Prenatal regimens received
Zidovudine monotherapy	1.0	Referent
Zidovudine & other drugs with PI	0.4	0.3-0.7
Zidovudine & other drugs no PI	0.5	0.3-0.8
Other drugs with PI, no zidovudine	0.6	0.2-1.4
Other drugs no PI, no zidovudine	0.3	0.1-1.5

Back to Article Outline

Comment 

We examined the impact of ART on the perinatal infant infection rates in 24 sites across the United States from 1999-2001. These 24 sites represent 89% of the cumulative perinatal AIDS cases that were reported through 2003 in the United States.¹⁵ The proportion of infants who became infected is associated with the use and timing of the ART. Mother-infant pairs who receive all 3 arms (prenatal, intrapartum, and neonatal) have the lowest proportion of infants who are infected (2.5%; 95% CI, 2.1%, 2.8%). Those mothers who receive no treatment at all (eg, 0 arms) have 25.7% infant infection rate (95% CI, 20.7%-30.7%), which is similar to the results that were obtained by Connor et al¹ in the early 1990s. We also found that women and infants who received antiretroviral treatment during the intrapartum and neonatal periods, but not the antenatal period, received only some benefit from prophylaxis (4.7%; 95% CI, 2.4%-7.0%).

Our results among mother-infant pairs who received all 3 arms (prenatal, intrapartum, and neonatal) of ART are similar (2.5%; 95% CI, 2.1-2.8) to those reported by Wade et al¹⁶ (3.3%) and Peters et al⁹ (3%). Additionally, we found infant infection rates for mother-infant pairs who receive neonatal-only ART (11.0%; 95% CI, 7.7%-14.4%) to be comparable with that found by Wade et al (11.9%). When we further examined those pairs who received neonatal-only ART (n = 100, with dose timing data), 6.3% (95% CI, 2.1-14.6) of infants who received ART ≤12 hours were found to be infected (data not shown); this rate was approximately one-half that found by Wade et al (11.9%) and Peters et al (14%). Additionally, the EPS infant infection rate for mother-infant pairs who received intrapartum and neonatal ART (4.6%; 95%CI, 2.2%-6.9%) was approximately one-half that reported by Wade et al (9.4%) and Peters et al (8.0%).

The availability of ART became widespread by early 1997; combination ART that consists of 2 nucleoside reverse transcriptase inhibitors and either a PI or a non–nucleoside/nucleotide reverse transcriptase is the recommended standard treatment for HIV-1–infected adults who are not pregnant.¹⁷ Current antiretroviral drug recommendations for pregnant women now include offering the 3-part zidovudine regimen with additional antiretroviral drugs to women whose HIV-1 RNA levels are >1000 copies/mL, regardless of their clinical or immunologic status.¹⁸ The decision to treat pregnant women with combination antiretroviral drugs is a complex one, and clinicians are advised to discuss the benefits and potential negative consequences with pregnant women. However, in many instances, the benefits outweigh the potential negative consequences; 87% of the women in this study who gave birth from 1999-2001 received prenatal combination ART. Clinical studies have demonstrated decreasing perinatal HIV infant infection rates since the publication of the results of the PACTG 076.⁸, ¹⁹, ²⁰, ²¹ The PACTG 185 study confirmed the efficacy of the zidovudine regimen that was observed in PACTG 076 and extended the results to HIV-infected women who had advanced disease.²² More recent studies, which have also examined the relationship between perinatal infant infection rates and combination maternal ART, have found perinatal infant infection rates of approximately 2%.⁹, ¹⁰, ¹⁶, ²³, ²⁴, ²⁵, ²⁶ In a randomized trial that assessed single-dose nevirapine at delivery compared with placebo, and in the background of primarily combination antiretrovirals during pregnancy followed by infant zidovudine for 6 weeks, the infant infection rate was 1.6%.²³ In 2004 in a trial from Thailand, Lallemant et al²⁵ reported that the use of zidovudine from 28 weeks of gestation plus single-dose nevirapine to the mother at labor followed by single-dose nevirapine to the infant and 1 week of zidovudine was associated with an infant infection rate of 2%. Observational perinatal cohort studies from the United States and Europe have also documented very low infant infection rates with the use of combination antiretrovirals for perinatal HIV prevention. A nonrandomized study in France demonstrated an infant infection rate as low as 1.6% among women who received lamivudine in addition to the zidovudine regimen. In this study, the treatment regimen for the pregnant women consisted of lamivudine that was started at 32 weeks gestation in addition to zidovudine and to the infant for 6 weeks after birth in addition to zidovudine.¹¹ Another observational study by Cooper et al¹⁰ demonstrated perinatal infant infection rates to be 1.2% among women who received combination therapy with a PI and 3.8% among women who received combination therapy without a PI. A study by Wade et al¹⁶ also demonstrated low perinatal infant infection rates among women who received prenatal combination therapy with and without PIs, compared with women who received no prenatal ART (relative risk, 0.15 [95% CI, 0.10-0.24]; relative risk, 0.14 [95% CI, 0.08-0.22], respectively). In our analysis, when we examined prenatal ART regimens among those who received all 3 arms of ART, women who received combination prenatal ART that contained zidovudine and a PI or a combination prenatal ART that contained zidovudine and no PI had low infant infection rates (1.9% [95% CI, 1.3-2.4]; 2.4% [95% CI, 1.8-3.0], respectively; data not shown). Other transmission rates by prenatal ART regimen among mother-infant pairs who received 3 arms of perinatal preventive prophylactic therapy were 4.2% for women who received zidovudine monotherapy (95% CI, 2.8%-5.6%), and 2.5% for women who received combination drugs that included a PI but no zidovudine (95% CI, 0.5%, 4.5%; data not shown). When we controlled for age, race, type of delivery, year of birth, and site, the odds for having an infected infant were less for women who received either prenatal ART regimens that contained zidovudine and a PI (OR, 0.4; 95% CI, 0.3-0.7) or regimens that contained zidovudine without a PI (OR, 0.5; 95% CI, 0.3-0.8) compared with women whose prenatal ART regimen contained zidovudine alone.

These surveillance data show comparable reductions in perinatal transmission, as reported by other clinical studies, and demonstrate the effective dissemination and use of perinatal HIV treatment guidelines on a population-based level. In particular, these data provide additional support for the current treatment guidelines and show that, compared with prenatal zidovudine-only regimens, infants were less likely to be infected when mothers used a prenatal ART regimen that contained zidovudine with additional antiretroviral drugs with or without a PI. Infected women have to know their status to be effective partners in reducing perinatal HIV transmission. One of the strategies of the CDC’s Advancing HIV Prevention is the further reduction of perinatal HIV transmission by promoting the routine testing of all pregnant women and the routine screening of all infants whose mother’s HIV status is unknown at birth.²⁷

There are several limitations to the study. First, we were not able to ascertain the infection status of all HIV-exposed infants who were born between 1999 and 2001. For data reported here, the estimated completeness of ascertainment of mother-infant pairs, based on the 1994 SCBW estimates for each participating site, was 68%.²⁸ The use of the 1994 SCBW to estimate completeness assumes that the number of HIV-infected women who gave birth during 1999-2001 is the same as the number who gave birth during 1994; this assumption may not be valid. After 1994, the SCBW was conducted by 10 sites, all of which used state-specific funds to estimate the prevalence of births to HIV-infected women. At 1 site, the prevalence estimate increased 3%; however, at the other 9 sites, prevalence estimates declined substantially (range, 5-60% for 1995-2000). When the updated prevalence estimates for the 10 sites are included, completeness of EPS reporting improves to 90%. For 7 sites that piloted these enhanced surveillance methods for births during 1993-1997, completeness of ascertainment of HIV-infected mothers and HIV-exposed infants was 90%.²⁹ Second, the pediatric surveillance case definition for the classification of children <18 months of age as HIV-negative includes negative viral load or antibody tests that were conducted within a specified time period. If the negative viral load or antibody tests do not meet the required specified time periods (ie, ≥4 months of age), the child is categorized as indeterminate.¹⁴ In this analysis, the status of approximately one-quarter (25%) of the number of infants is indeterminate. It is presumed that almost all of these infants during this recent time period are not infected with HIV, but their status is indeterminate because the current CDC criteria for classification as not infected have not been met (requires at least 2 negative HIV virologic tests, 1 at >1 month of age and 1 at >4 months of age). In many instances, clinicians may be testing exposed children earlier than the 4-month time period stated in the pediatric HIV surveillance case definition and then discharging them so that their final HIV status is not known for surveillance purposes. Because we assumed that most of the indeterminate cases were uninfected, we may have underestimated perinatal HIV infant infection rates. However, that is unlikely to have occurred, because our results are comparable with other clinical studies that used a more stringent case definition of HIV infection¹⁶ in children who were born exposed to HIV infected mothers than the current surveillance pediatric HIV case definition.¹⁴ Third, sites conducted the project as population-based (n = 16) or facility-based (n = 8). In population-based sites, all HIV-exposed infants who were born to HIV-infected mothers within the geographic site defined by the project (eg, state or city) were eligible for medical chart abstraction. In facility-based sites, medical chart abstraction was conducted in selected facilities (eg, delivery hospitals, high-risk prenatal clinics, specialty pediatric clinics, or pediatric HIV clinics) within the geographic site defined by the project site. In conducting the project in these 2 ways, we could have over- or underestimated perinatal infant infection rates; however, when we examined the infant infection rates by project type (population-based compared with facility-based), we found no statistically significant difference (P = .07). Fourth, we did not assess adherence to antiretroviral medications; we were able to record only whether antiretroviral medications were prescribed. Fifth, the date and time of the onset of labor were not included in this analysis because of the high proportion of missing data (60% and 67%, respectively); consequently, we were unable to examine the impact of cesarean section on infant infection status.

The CDC, in partnerships with other Department of Health and Human Services agencies, state health departments, and national organizations, aims to further reduce and potentially eliminate perinatal HIV transmission. Current strategies include (1) supporting universal HIV screening for all pregnant women by making HIV screening part of the routine panel of tests during the prenatal period, unless the woman declines testing (the “opt out” approach); (2) supporting rapid testing at labor and delivery for women whose status is still unknown and to the newborn infant if the testing cannot be done intrapartum; and (3) social marketing to increase the awareness of women and clinicians of the importance and benefits of testing for HIV during pregnancy.

The CDC is also assisting states to monitor and evaluate their perinatal HIV prevention efforts by (1) assessing prenatal HIV testing rates in select states around the country, (2) evaluating perinatal prevention programs that use EPS data in 15 states, and (3) continuing to monitor perinatal infant infection rates by conducting EPS in states that are also conducting perinatal prevention programs.

These EPS data document the striking success in lowering perinatal infant infection rates in the United States in recent years. However, the CDC estimates that there are still >130 HIV-infected infants each year. To maximally reduce perinatal HIV transmission in the United States, we must make sure that all pregnant women are receiving prenatal care and that HIV screening is part of the routine panel of prenatal screening, unless a woman declines testing. In addition, rapid testing should be available widely and used at labor and delivery if a woman’s HIV status is still unknown; treatment should be offered to women during labor/delivery and to the infants within 12 hours of birth to reduce the risk of HIV transmission.

Given that AIDS cases among female patients increased 15% from 1999 through 2003,¹⁵ it is also imperative that public health efforts address and test innovative strategies to reduce the risk of acquiring HIV among adolescent girls and women of childbearing age.

Back to Article Outline

References 

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
   • View In Article
   • MEDLINE
   • CrossRef
2. Public Health Service Task Force recommendations for use of antiretroviral drugs in pregnant HIV-1–infected women for maternal health and interventions to reduce perinatal HIV-1 transmission in the United States (revised November 3, 2000). HIV Clin Trials. 2001;2:56–91
   • View In Article
   • MEDLINE
   • CrossRef
3. Centers for Disease Control and Prevention. Recommendations for human immunodeficiency virus counseling and voluntary testing for pregnant women. MMWR Morb Mortal Wkly Rep. 1995;44:1–14
   • View In Article
   • MEDLINE
4. Centers for Disease Control and Prevention. Revised recommendations for HIV screening of pregnant women. MMWR Morb Mortal Wkly Rep. 2001;50:63–85
   • View In Article
5. Centers for Disease Control and Prevention. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health care settings. MMWR Morb Mortal Wkly Rep. 2006;55:1–16
   • View In Article
6. Cooper E, Diaz C, Pitt J, et al. Impact of ACTG 076: use of ZDV (ZDV) during pregnancy and changes in the rate of HIV vertical transmission. Conf Retroviruses Opportunistic Infect. 1996;3:57
   • View In Article
7. Fiscus SA, Adimora AA, Schoenbach VJ, et al. Trends in human immunodeficiency virus (HIV) counseling, testing, and antiretroviral treatment of HIV-infected women and perinatal transmission in North Carolina. J Infect Dis. 1999;180:99–105
   • View In Article
   • MEDLINE
   • CrossRef
8. Harris NS, Thompson SJ, Ball RT, Hussey J, Sy F. Zidovudine and perinatal human immunodeficiency virus type 1 transmission: a population-based approach. Pediatrics. 2002;109:1–7
   • View In Article
   • CrossRef
9. Peters V, Liu K, Dominguez K, et al. Missed opportunities for perinatal HIV prevention among HIV-exposed infants born 1996-2000, pediatric spectrum of disease. Pediatrics. 2003;111:1186–1191
   • View In Article
10. Cooper ER, Charurat M, Mofenson L, et al. Combination antiretroviral strategies for the treatment of pregnant HIV-1-infected women and the prevention of perinatal HIV-1 transmission. J Acquir Immune Defic Syndr. 2004;29:484–494
    • View In Article
    • MEDLINE
    • CrossRef
11. Mandelbrot L, Landreau-Mascaro A, Rekacewicz C, et al. Lamivudine-ZDV combination for prevention of maternal-infant transmission of HIV-1. JAMA. 2001;285:2083–2093
    • View In Article
    • MEDLINE
    • CrossRef
12. Centers for Disease Control and Prevention. In: HIV/AIDS surveillance report. 2001;p. 1–4713
    • View In Article
13. Fleming PL, Lindegren ML, Byers RH, et al. Estimated number of perinatal HIV infections, United States, 2000. [abstract] In: XIV International AIDS Conference. 2002;Abstract TuPeC4773
    • View In Article
14. Centers for Disease Control and Prevention. Guidelines for national human immunodeficiency virus case surveillance, including monitoring for human immunodeficiency virus infection and acquired immunodeficiency syndrome. MMWR Morb Mortal Wkly Rep. 2004;48:1–31
    • View In Article
    • MEDLINE
15. Centers for Disease Control and Prevention. In: HIV/AIDS surveillance report. 2005;p. 1–4615
    • View In Article
16. Wade NA, Zielinski MA, Butsashvili M, et al. Decline in perinatal HIV transmission in New York State (1997-2000). J Acquir Immune Defic Syndr. 2004;36:1075–1082
    • View In Article
    • MEDLINE
    • CrossRef
17. Department of Health and Human Services. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. April 7. 2005;Last accessed: July 25, 2005. Available at: http://aidsinfo.nih.gov/guidelines/.
    • View In Article
18. Public Health Service Task Force. Recommendations for use of antiretroviral drugs in pregnant HIV-1–infected women for maternal health and interventions to reduce perinatal HIV-1 transmission in the United States. In: 2005;p. 1–16February 24
    • View In Article
19. Cooper ER, Nugent RP, Diaz C, et al. After AIDS clinical trial 076: the changing pattern of ZDV use during pregnancy, and the subsequent reduction in the vertical transmission of human immunodeficiency virus in a cohort of infected women and their infants: Women and Infants Transmission Study Group. J Infect Dis. 1996;174:1207–1211
    • View In Article
    • MEDLINE
    • CrossRef
20. Fiscus SA, Adimora AA, Schoenbach VJ, et al. Perinatal HIV infection and the effect of ZDV therapy on transmission in rural and urban counties. JAMA. 1996;275:1483–1488
    • View In Article
    • MEDLINE
21. Fiscus SA, Adimora AA, Funk ML, et al. Trends in interventions to reduce perinatal human immunodeficiency virus type 1 transmission in North Carolina. Pediatr Infect Dis J. 2002;21:664–668
    • View In Article
    • MEDLINE
    • CrossRef
22. Stiehm ER, Lambert JS, Mofenson LM, et al. Efficacy of ZDV and human immunodeficiency virus (HIV) hyperimmune immunoglobulin for reducing perinatal HIV transmission from HIV-infected women with advanced disease: results of Pediatric AIDS Clinical Trials group protocol 185. J Infect Dis. 1999;179:567–575
    • View In Article
    • MEDLINE
    • CrossRef
23. Dorenbaum A, Cunningham CK, Gelber RD, et al. Two-dose intrapartum/newborn nevirapine and standard antiretroviral therapy to reduce perinatal HIV transmission: a randomized trial. JAMA. 2002;288:189–198
    • View In Article
    • MEDLINE
    • CrossRef
24. Lallemant M, Jourdain G, Le Coeur S, et al. A trial of shortened ZDV regimens to prevent mother-to-child transmission of human immunodeficiency virus type 1: Perinatal HIV Prevention Trial (Thailand) investigators. N Engl J Med. 2000;343:982–991
    • View In Article
    • MEDLINE
    • CrossRef
25. Lallemant M, Jourdain G, Le Coeur S, et al. Single-dose perinatal nevirapine plus standard ZDV to prevent mother-to-child transmission of HIV-1 in Thailand. N Engl J Med. 2004;351:217–228
    • View In Article
    • CrossRef
26. Mandelbrot L, Le Chenadec J, Berrebi A, et al. Perinatal HIV-1 transmission: interaction between ZDV prophylaxis and mode of delivery in the French Perinatal Cohort. JAMA. 1998;280:55–60
    • View In Article
    • MEDLINE
    • CrossRef
27. Centers for Disease Control and Prevention. Advancing HIV prevention: new strategies for a changing epidemic: United States, 2003. MMWR Morb Mortal Wkly Rep. 2003;52:329–332
    • View In Article
    • MEDLINE
28. Davis SF, Byers RH, Lindegren ML, Caldwell MB, Karon JM, Gwinn M. Prevalence and incidence of vertically acquired HIV infection in the United States. JAMA. 1995;274:952–955
    • View In Article
    • MEDLINE
29. Wortley PM, Lindegren ML, Fleming PL. Successful implementation of perinatal HIV prevention guidelines: a multistate surveillance evaluation. MMWR Recomm Rep. 2001;50:17–28
    • View In Article
    • MEDLINE