Postnatal Infant HIV Prophylaxis: A Survey of U.S. Practice

Introduction

A ccurate assessment of the risk of perinatal transmission of HIV and appropriate risk reduction for the HIV-exposed infant remain pressing concerns. A recent estimate suggests that the number of births to HIV-infected women is increasing in the United States. ¹ The perinatal HIV transmission rate varies based on receipt of prophylactic medications. Without any intervention, roughly 25% of nonbreastfed HIV-exposed infants acquire the virus. But if the standard of care (prenatal maternal highly active antiretroviral therapy [HAART] plus intrapartum and postnatal zidovudine) is completed, less than 2% are infected. ² Unfortunately, in a significant proportion of HIV-affected pregnancies not all components of perinatal prophylaxis are obtained, increasing the chance that infants acquire HIV. ^3,4 Current U.S. guidelines allow for consideration of combination postnatal prophylaxis (e.g., zidovudine plus additional antiretrovirals) in high-risk situations, although the effectiveness of this approach has not been defined. ⁵ Little is known about the actual use of combination postnatal prophylaxis in the United States.

Any use of combination postnatal prophylaxis must balance uncertain benefit with the potential risks of additional medications. Nevirapine, which has demonstrated an acceptable toxicity profile, ^6,7 is the only drug other than zidovudine cautiously suggested by U.S. guidelines for neonatal use, as an adjunct following delivery by a woman with a high viral load. ⁵ They note, however, that if a neonate is infected with HIV despite prophylaxis, exposure to a single dose of nevirapine can lead to resistance mutations. Furthermore, published case series have raised questions regarding the use of drugs that lack safety and dosing data, particularly in preterm infants. ^8,9

Methods

We conducted a cross-sectional, Web-based survey between December 2009 and January 2010 of U.S. perinatal HIV providers to determine the use of combination postnatal prophylaxis. We used multiple sources to locate providers. We included all Ryan White Title IV funding recipients (n = 83) and all members with contact information from the International Maternal Pediatric Adolescent AIDS Clinical Trials Group (IMPAACT), the Pediatric HIV/AIDS Cohort Study (PHACS), and the Perinatal HIV Guidelines Working Group (n = 86). We screened the website of each Children's Hospital in the online directory kept by the National Association of Children's Hospitals and Related Institutions (NACHRI, Alexandria, VA) for possible HIV providers (n = 326). To include as many providers of HIV-related care to exposed infants as possible, for children's hospitals that did not list HIV-specialists, we included all infectious disease specialists at that institution.

The survey focused on situations for which the U.S. guidelines recommend consideration of combination prophylaxis and other high-risk scenarios for which some practitioners might recommend combination postnatal prophylaxis (Table 1). Seven short vignettes described a hypothetical neonate with or without additional risk factors for HIV transmission (i.e., “An HIV-infected woman becomes pregnant, and receives HAART during this pregnancy. She has a viral load of >1000 copies per milliliter near term. She receives IV zidovudine during a scheduled cesarean delivery.”). For each patient-based scenario, providers were asked to recommend for the neonate either zidovudine monotherapy or “a combination of 2 or more antiretroviral drugs.” There were also 4 items related to the preferred drugs for combination postnatal prophylaxis as well as 12 Likert items assessing preference for combination prophylaxis. The survey was designed and delivered using Survey Monkey (Survey Monkey, Portland OR). The survey was pretested at a large referral practice by all practitioners who would otherwise have been eligible and by 3 clinical epidemiologists with survey experience. The only eligibility criterion was that the respondent self-identified as a provider for HIV-exposed neonates.

Results

Discussion

This study describes combination prophylaxis use in the United States and the variation between providers and between situations. More than half of surveyed practitioners reported using combination postnatal prophylaxis at least once within the last year. In standard-risk scenarios the selection of combination postnatal prophylaxis was uncommon. However, the proportion of providers who chose combination postnatal prophylaxis in higher risk scenarios ranged from 30% to as high as 80%.

Variability related to the use of combination postnatal prophylaxis exists at both individual provider and national levels. The British HIV Association, for example, strongly recommends combination prophylaxis in certain cases ¹⁰ for which the U.S. guidelines either recommend consideration of combination prophylaxis or remain silent. Because the data necessary for an accurate risk-benefit analysis of providing or not providing combination prophylaxis do not exist, it is impossible at this time to say which approach is most prudent.

It is notable that combination prophylaxis is routinely used for situations in which HIV infection risk is lower than the risk of transmission from mother to child around the time of birth. Health care workers with percutaneous exposure to infected blood have an estimated 0.3% risk of acquiring HIV. ¹¹ There have been no clinical trials of combination postexposure prophylaxis (PEP), but there is a consensus that the benefit of HIV prevention outweighs the risks of multiple antiretrovirals. ^12,13 Since the risk of mother-to-child transmission (MTCT) in the perinatally HIV-exposed neonate is at least as high as the risk of HIV transmission to a health care worker via occupational exposure, the relative reluctance to use combination postnatal prophylaxis implies a belief that the potential harm to the neonate of using additional drugs is greater than the potential for harm in adults requiring PEP.

A few studies and case reports have raised toxicity concerns regarding the use of the most commonly recommended antiretrovirals in this survey for neonatal prophylaxis, but the evidence thus far has not been compelling. A slightly higher degree of hematologic suppression has been observed with the postnatal use of zidovudine in combination with lamivudine versus zidovudine alone, ¹⁴ but the clinical significance of this finding remains unclear. ¹⁵ The safety of single-dose nevirapine has been described ⁶ ; no serious adverse effects were noted in a multinational study of a 6-week postnatal nevirapine regimen. ⁷ There have been case reports of cardiomyopathy in premature infants associated with postnatal prophylaxis regimens containing LPV/r. ^8,9 There has been only one pharmacokinetics study of LPV/r in infants between 2 and 6 weeks of age, ¹⁶ and none in younger infants.

Our study establishes that the use of combination postnatal prophylaxis for prevention of MTCT is not uncommon in the United States. Because there is no national list of all perinatal HIV providers, we sent surveys to both known HIV care providers and individuals who were considered to be likely to provide some care to HIV-exposed neonates due to practicing infectious disease medicine in children's hospitals which lack dedicated HIV care providers. We believe that our response rate reflects the inclusion of a large number of individuals in our contact list who do not actually provide HIV-related care and therefore did not respond to the survey.

Our survey respondents included individuals from 37 states and many major metropolitan areas. Although we are unable to report what percent of HIV-exposed neonates in the United States receive care from the providers who responded to this survey, the distribution of our respondents is similar to the distribution of pediatric HIV throughout the country. The most represented states in this survey (Texas, California, New York, North Carolina, Florida) were among those with the largest burden of pediatric HIV. Currently, the states with the most children living with AIDS (cumulative through 2008) were New York, Florida, New Jersey, California, and Texas, ¹⁷ while those with the greatest number of new pediatric HIV diagnoses in 2008 (among the 37 states with name-based reporting) were Florida, Texas, New York, North Carolina, and Georgia. ¹⁸

The data from our study draw attention to current variations in the provision of postnatal prophylaxis in the United States. Additional studies to assess how closely actual practice corresponds with self-reported practice patterns among providers described in our study would be useful. Further research into the safety of combination postnatal prophylaxis and its effectiveness in a variety of high-risk scenarios is necessary to allow for more evidence-based standardization of practice.