5
PRENATAL SCREENING FOR HIV INFECTION

There are several possible goals of an HIV screening program for pregnant women: (1) prevention of vertical and horizontal transmission of HIV infection, (2) a more informed basis for reproductive decision making, (3) early diagnosis and treatment of HIV infection in women and their children, and (4) enhancement of epidemiological and treatment research among HIV-infected women and children. Whether any or all of these goals can be accomplished through prenatal HIV screening is a central question in the development of screening policy. To understand what can realistically be expected from such a program and to assess its appropriateness require close scrutiny of the potential objectives of prenatal HIV screening. This chapter offers the committee's judgment on the likelihood that these goals currently could be achieved through HIV screening of pregnant women and the consequent recommendations.

Effects on Vertical HIV Transmission

Preventing vertical transmission of HIV infection from mother to infant appears to be a desirable objective of prenatal HIV screening. Yet the limited studies to date offer little evidence to suggest that knowledge of HIV infection status significantly affects women's decisions regarding continuation of a pregnancy or future childbearing (Barbacci et al., 1989b; Kaplan et al., 1989; Schneck et al., 1989; Selwyn et al., 1989a; Johnstone et al., 1990; Sunderland, 1990). The relationships among knowledge of infection, perceived risk of perinatal transmission, and reproductive behavior are not simple ones. Personal, familial, and cultural values and beliefs often shape perceptions of individual risk, and these forces tend to influence reproductive decision making more than abstract notions of probabilities. Some HIV-infected women may view a roughly 25 to 35



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 33
HIV Screening of Pregnant Women and Newborns 5 PRENATAL SCREENING FOR HIV INFECTION There are several possible goals of an HIV screening program for pregnant women: (1) prevention of vertical and horizontal transmission of HIV infection, (2) a more informed basis for reproductive decision making, (3) early diagnosis and treatment of HIV infection in women and their children, and (4) enhancement of epidemiological and treatment research among HIV-infected women and children. Whether any or all of these goals can be accomplished through prenatal HIV screening is a central question in the development of screening policy. To understand what can realistically be expected from such a program and to assess its appropriateness require close scrutiny of the potential objectives of prenatal HIV screening. This chapter offers the committee's judgment on the likelihood that these goals currently could be achieved through HIV screening of pregnant women and the consequent recommendations. Effects on Vertical HIV Transmission Preventing vertical transmission of HIV infection from mother to infant appears to be a desirable objective of prenatal HIV screening. Yet the limited studies to date offer little evidence to suggest that knowledge of HIV infection status significantly affects women's decisions regarding continuation of a pregnancy or future childbearing (Barbacci et al., 1989b; Kaplan et al., 1989; Schneck et al., 1989; Selwyn et al., 1989a; Johnstone et al., 1990; Sunderland, 1990). The relationships among knowledge of infection, perceived risk of perinatal transmission, and reproductive behavior are not simple ones. Personal, familial, and cultural values and beliefs often shape perceptions of individual risk, and these forces tend to influence reproductive decision making more than abstract notions of probabilities. Some HIV-infected women may view a roughly 25 to 35

OCR for page 33
HIV Screening of Pregnant Women and Newborns percent chance of transmitting the virus to their infant as an acceptable risk and choose to become pregnant or continue an existing pregnancy. Counselors and researchers have also observed that some HIV-infected women may, in fact, deny their illness and the possible impact it may have on their pregnancy and their child (Wofsy, 1987; Sunderland, 1990). Still another possibility is that some women, confronted with the reality of progressive HIV infection and their own mortality, may consider a child to be their final legacy (Selwyn et al., 1989a; Sunderland, 1990). In light of the apparently limited influence that knowledge of HIV infection has on current and future childbearing decisions, a substantial reduction in vertical transmission of HIV is not likely to be accomplished through prenatal screening. Therefore, the argument that screening could reduce vertical transmission does not by itself constitute sufficient justification for instituting HIV screening of pregnant women at this time. If successful chemoprophylaxis or immunotherapy to prevent perinatal HIV transmission becomes available in the future, the prevention of vertical transmission might then be an achievable goal of prenatal screening. Effects on Horizontal HIV Transmission Screening pregnant women for HIV infection may play a role in decreasing horizontal transmission of HIV to their partners. Yet here again the connection between an awareness of HIV infection status and behavioral change is unclear, in part because of insufficient data from which to draw conclusions about this relation. In specific populations, there is some evidence to suggest that HIV-antibody testing and counseling may be useful in facilitating change in certain risk-associated behaviors (Coates et al., 1988; Turner et at., 1989). Most of the available studies, however, have been conducted among homosexual and bisexual men and, in a few instances, intravenous drug users. Among the latter, greater change has been observed in high-risk drug use practices than in sexual behavior (Turner et al., 1989; de la Fuente et al., 1990). Whether knowledge of infection will necessarily promote the adoption of less risky behavior (e.g., safer sexual practices) among heterosexual women and men remains uncertain.1 Prenatal HIV screening, therefore, may not substan- 1    Inconsistent condom use has been reported among female IV drug users and female sexual partners of infected or at-risk individuals, despite a dear perception of risk (Miller et al., 1990). For example, one study of female sexual partners of HIV-seropositive hemophilic men found that fewer than 50 percent of female partners reported that they always used condoms and nearly 20 percent reported no condom use, despite an awareness of their partner's infection (Price et al., 1989). Among a group of Rwandan women, however, HIV

OCR for page 33
HIV Screening of Pregnant Women and Newborns tially affect horizontal HIV transmission. In light of these uncertainties, reducing horizontal transmission of infection seems a desirable but not necessarily achievable goal of prenatal screening at this time. The committee recognizes that public policy on HIV testing has often been founded on the hope that testing and counseling will have some positive impact on behavioral change. Other benefits associated with testing and counseling may include the possibility of reaching seronegative women (who continue to be at risk of infection) with HIV education and support to help them guard against the acquisition of infection. Further studies are clearly needed to evaluate ways to enhance change in high-risk sexual or drug-using practices, particularly among heterosexual couples. Once effective behavioral interventions have been identified, limiting horizontal transmission of infection may become a more potent rationale for prenatal screening. Informed Reproductive Decision Making Although knowledge of HIV infection status alone may not fundamentally alter fertility-related behavior, this information must still be regarded as germane to reproductive counseling and planning. Hence, another goal of screening pregnant women for HIV infection is to offer them a more informed basis for making reproductive decisions. A woman's knowledge of her own HIV infection may be a potentially empowering tool in evaluating her current and future reproductive options, even if such knowledge does not change her final decision. As more information accumulates regarding the effect of pregnancy on maternal disease progression and the impact of HIV infection on pregnancy outcome, learning of one's infection may become increasingly important for women facing reproductive decisions. Early Diagnosis of HIV Infection and Therapeutic Intervention During Pregnancy Another goal of screening pregnant women for HIV infection is to afford them an opportunity for early diagnosis, subsequent medical monitoring, and therapeutic intervention. If HIV infection has been diagnosed in a pregnant woman, her prenatal caregiver (and perhaps other     education and testing appeared to be important in fostering safer sexual practices (Lindan et al., 1990).

OCR for page 33
HIV Screening of Pregnant Women and Newborns specialists) can provide careful surveillance of immunologic function and the signs or symptoms indicative of HIV-related opportunistic infection, both of which facilitate optimal management of her pregnancy and her own medical care (Minkoff et al., 1990). The committee concludes that screening pregnant women for the purpose of early diagnosis and treatment is both an achievable and compelling objective.2 This conclusion rests on the fact that available therapies for HIV disease, a life-threatening condition, have been shown to delay progression and minimize symptoms of disease in nonpregnant adults. However, treatment regimens (e.g., zidovudine therapy and PCP prophylaxis) may need slight modifications for HIV-infected pregnant women. Traditionally, caregivers use caution in prescribing drug therapy during pregnancy because of possible teratogenesis or unknown fetal toxicities associated with a particular therapeutic agent. For example, in regard to PCP prophylaxis, the Public Health Service has warned that "since neither aerosol pentamidine nor oral trimethoprim-sulfamethoxazole (TMP-SMX) prophylaxis is known to be safe in association with pregnancy, it is inadvisable to give either agent to HIV-infected pregnant women. Rather, such women should be monitored carefully for symptoms, signs, or laboratory abnormalities suggestive of PCP. Prophylaxis can then be considered for use in the postpartum period" (CDC, 1989b:6). No doubt, historical precedent served as a guide in making this recommendation, but its appropriateness, given current conditions, has generated some debate among obstetricians. Many obstetrical experts contend that the standard of care for HIV-infected nonpregnant adults should also apply to HIV-infected pregnant women, based on careful monitoring of immunologic function through serial CD4+ cell counts. Asymptomatic nonpregnant adults with CD4+ cell counts of 200 or less are at significantly greater risk of Pneumocystis pneumonia than those with higher counts; the presence independently of thrush or persistent fever enhances this risk (CDC, 1989b; Phair et al., 1990). Extrapolating from data in nonpregnant adults, it has been estimated that pregnant women with CD4+ cell counts below 200 may face roughly a 20 percent (or greater, if fever or thrush is present) chance of developing PCP during pregnancy. Primary prophylaxis against PCP is currently recommended for HIV-infected nonpregnant adults with CD4+ cell counts of less than 200 (CDC, 1989b), and most obstetrical experts argue that such therapy should also be offered to HIV-infected pregnant women with this same level of immune deficiency. Zidovudine therapy currently is also recommended for 2    Specific circumstances under which screening should take place are described in subsequent recommendations.

OCR for page 33
HIV Screening of Pregnant Women and Newborns both symptomatic and asymptomatic HIV-infected nonpregnant adults with CD4+ cell counts below 500 (NIAID, 1990). Because HIV-infected pregnant women with CD4+ cell counts below 200 are at greatest risk of clinical deterioration, most obstetrical experts concur that administering zidovudine to these women during pregnancy is probably appropriate and may prevent further disease progression. There is much less consensus on whether zidovudine should be administered to HIV-infected pregnant women with moderately diminished CD4+ cells (i.e., CD4+ cell counts between 200 and 500). In this case, it is considered prudent to postpone antiretroviral therapy until the postpartum period. These women should, nevertheless, be carefully followed during pregnancy for evidence of further CD4+ cell depletion that would warrant more aggressive therapeutic intervention. Concern for potential fetal toxicity has not been absent from discussions of appropriate management and treatment of HIV-infected pregnant women. To date, there are insufficient data on zidovudine therapy during pregnancy to draw conclusions about short-term fetal toxicity or adverse pregnancy outcomes related to such therapy. There is additional uncertainty regarding zidovudine's long-term effects and potential toxicities for the infant. Similarly, the fetal and neonatal risks associated with trimethoprim-sulfamethoxazole or aerosolized pentamidine therapy during pregnancy have not been delineated. Sulfa drug therapy in newborns has been associated with the displacement of bilirubin from its protein-binding sites, leading to increased blood levels of free bilirubin that can penetrate the brain and cause severe neurological damage. Because trimethoprim-sulfamethoxazole readily crosses the placenta, there is a theoretical risk of fetal or neonatal harm, although congenital abnormalities or adverse consequences in infants as a result of maternal trimethoprim-sulfamethoxazole therapy appear to be rare (Minkoff and Feinkind, 1989). However, TMP-SMX therapy during pregnancy and its effects on infant outcome have not been extensively studied. (Studies of another form of sulfonamide therapy during pregnancy concluded that the risk of bilirubin displacement in the newborn would not be increased as a result of such therapy [Jarnerot et al., 1981a,b; Esbjorner et al., 1987].) Aerosolized pentamidine has little systemic absorption (Montgomery et al., 1987) and therefore would be expected to have minimal effects on the fetus or infant, although its use during pregnancy has received little study. As a result, despite these uncertainties, the committee finds that the health risks inherent in deferring antiretroviral treatment. or PCP prophylaxis of severely immunocompromised HIV-infected pregnant women (i.e., those with CD4+ cell counts below 200) outweigh the potential fetal or neonatal risks at this time. Therefore, the committee recommends that

OCR for page 33
HIV Screening of Pregnant Women and Newborns HIV-infected pregnant women with severely depressed CD4+ cell counts be offered therapy for which they would otherwise be eligible if they were not pregnant. The decision to initiate treatment during pregnancy should always be made in concert with the patient, with full disclosure of the associated risks and benefits of therapy. Whether to receive treatment, however, ultimately remains the woman's choice. There is a clear need to continue surveillance and assessment of fetal toxicities and adverse pregnancy outcomes that may be associated with antiretroviral therapy or PCP prophylaxis. Obstetricians who are treating their HIV-infected patients with these therapies (e.g ., zidovudine, aerosolized pentamidine, or trimethoprim-sulfamethoxazole) should be advised to report any adverse side effects or outcomes to the pharmaceutical companies.3 In particular, infants born to mothers who received antiretroviral therapy or PCP prophylaxis (especially trimethoprim-sulfamethoxazole) during pregnancy should be carefully monitored for any evidence of untoward effects. Additional Benefits of Prenatal HIV Screening In addition to the maternal benefits that may be derived from identification of HIV infection during pregnancy, there are potential collateral benefits for the infant. For instance, infants born to mothers who are known to be HIV infected could be closely followed from birth for evidence of immunologic impairment and other signs and symptoms of HIV infection, so that appropriate treatment can be administered when it is clinically indicated. As holed earlier, however, the magnitude of benefit to the infant that is associated with such careful pediatric follow-up and comprehensive care remains uncertain and thus in itself does not constitute sufficient justification for screening newborns. 4 3    Burroughs Wellcome Company has established a registry to collect observational, nonexperimental data on exposure to zidovudine during pregnancy. Physicians who are administering zidovudine therapy to their pregnant HIV-infected patients can contact the registry at (919) 248-8465 or (800) 722-9292, extension 8465. 4    The mother can also be offered counseling regarding breast feeding. Several cases of postnatal HIV transmission to infants through breast feeding have been documented, although the absolute risk of infection by this route is unclear (Oxtoby, 1988). In addition, HIV has been isolated from breast milk. These findings (as well as the availability of safe and effective alternatives to infant nutrition, that is, human milk substitutes) prompted the current recommendation in the United States that HIV-infected women be advised against breast feeding to avoid postnatal HIV transmission to infants who may not already be infected (CDC, 1985).

OCR for page 33
HIV Screening of Pregnant Women and Newborns An ancillary benefit to prenatal screening is that prospective follow-up of identified HIV-infected women and their children could advance epidemiological research regarding the natural history of HIV infection in women, particularly pregnant women, and their offspring. Prenatal HIV screening could also identify HIV-infected women Who might be eligible to participate in research protocols to study the effects of treatment (e.g., zidovudine therapy) during pregnancy on maternal disease progression and pregnancy outcome, as well as the impact of antiretroviral therapy on perinatal transmission.5 Universal Versus Selective Prenatal HIV Screening The committee, having concluded that the advantages associated with early detection of HIV infection and medical intervention provide the most compelling argument for screening pregnant women, then considered whether screening should be "universal" or "selective"; that is, should all pregnant women within a particular geographic area be offered HIV testing (''universal'' screening), or should testing be offered only to a subset of pregnant women defined by self-acknowledged HIV risk behaviors, based on prior risk assessment ("selective" screening)? Risk factor assessment as an indicator for HIV testing in pregnant women has been an insensitive predictor of HIV infection. Many women are unaware they are at risk; others are unwilling to admit their involvement in high-risk sexual or drug-using practices and thus deny their risk (Landesman et al, 1987; Krasinksi et al., 1988; Barbacci et al., 1989a; Lindsay et al, 1989; Sperling et al., 1989). To restrict offers of HIV testing to pregnant women who disclose a prior risk history is therefore likely to miss a substantial number of HIV-infected women who could benefit from early diagnosis and medical intervention. Moreover, targeting only "at-risk" pregnant women for HIV screening might be perceived as discriminatory and stigmatizing because African-American and Hispanic women and children have been disproportionately affected by the HIV epidemic. The committee strongly opposes any HIV screening of pregnant women based on racial or ethnic background. It is also concerned about screening that is narrowly focused on small geographic units. This type of screening could single out areas of highly concentrated HIV infection in women and children, which might engender further discrimination and 5    For further information regarding studies of maternal chemoprophylaxis currently under development and other possible strategies to reduce perinatal transmission, see the discussion "Medical Management of HIV-Infected Pregnant Women" in the conference summary in Appendix A.

OCR for page 33
HIV Screening of Pregnant Women and Newborns stigmatization of already disenfranchised or disadvantaged populations. Geographic units selected for screening purposes should be sufficiently large (e.g., states or counties) to limit the opportunity for such discrimination and stigmatization (see the discussion in Chapter 6). Cost can be a factor in deciding between universal and selective screening. The most costly component of screening is not the actual HIV-antibody test but the labor-intensive counseling that should accompany the communication of test results. From an economic perspective, selective screening of pregnant women based on prior risk factor assessment (with posttest counseling of all women screened) appears to be no more costly, in terms of the cost per HIV-infected woman identified, than universal screening of pregnant women (with posttest counseling reserved for HIV-positive women only). (See the discus ion "A Cost-Effectiveness Analysis of Prenatal HIV Screening" in the conference summary in Appendix A. A comparison of the cost per HIV-infected woman identified through selective versus universal screening appears in Table A-1.) In the latter case, however, an opportunity would be missed to provide education and counseling support to seronegative women. Such counseling might help these women take measures to reduce their risk of infection and thereby maintain their HIV-negative status. Because a substantial number of HIV-infected women would necessarily be missed if a selective approach to prenatal HIV. screening were pursued, and in view of the favorable cost comparisons between selective and universal screening, a universal approach to screening pregnant women for HIV infection is preferable. Universal prenatal screening would limit further discrimination and stigmatization because HIV testing would be offered to all pregnant women, within a particular geographic area, without regard to risk status. Furthermore, it would not require that women disclose socially unaccepted or illicit behavior in advance of testing. The committee recommends voluntary HIV screening (with specific informed consent) for all pregnant women in high-prevalence areas.6 The HIV test should be discussed with and offered to every pregnant woman seeking prenatal care in these areas; written informed consent should be a prerequisite to testing. Women who receive no prenatal care or who have not had an opportunity to be tested prior to delivery should be offered HIV testing at the time of labor and delivery or during the postpartum period. Additionally, in areas where prevalence levels may not warrant prenatal screening of all pregnant women at this time, health care providers should continue to offer voluntary HIV testing to pregnant women who have identified risk factors for HIV infection, in accordance 6    High-prevalence areas are defined in Chapter 6.

OCR for page 33
HIV Screening of Pregnant Women and Newborns with current HIV testing recommendations (CDC, 1987a; American Academy of Pediatrics, Task Force on Pediatric AIDS, 1988; American College of Obstetrics and Gynecology, 1988; American Medical Association, 1989). The committee recommends that all pregnant women be informed about HIV infection, its modes of transmission, risk-associated behaviors, and ways of reducing one's personal risk of infection. The committee recognizes the profound psychological and social ramifications that a diagnosis of HIV infection can have for an individual Nevertheless, it believes that the benefits that have been shown to result from early diagnosis of infection and subsequent treatment counterbalance the potential risks. In addition, unlike other adult civilian, noninstitutionalized populations at risk of infection who could also benefit from early diagnosis and treatment, pregnant women can be reached relatively easily and offered screening services. Pregnant women generally come into predictable contact with the health care system through obstetrical care. Consequently, obstetrical settings offer a unique opportunity to provide HIV counseling, testing, and treatment to women who may be at increased risk of infection.

OCR for page 33
HIV Screening of Pregnant Women and Newborns This page in the original is blank.