In Support of Breast-/Chestfeeding by People With HIV in High-Income Settings Free

Abstract

First, there are well-established benefits of breastfeeding for the infant and the feeding parent. In this Viewpoint article we will mostly refer to “breastfeeding” when discussing infant feeding, as most data are among women who have breastfed. However, we will also use the terminology “chestfeeding” to acknowledge that not all individuals who engage in infant feeding consider the part of the body they feed their child from as “breasts” or identify as women. Breastfed infants have lower rates of bacterial and viral infections (ear, respiratory, urinary, and gastrointestinal), necrotizing enterocolitis, sepsis, allergies, Sudden Infant Death Syndrome (SIDS), obesity, metabolic and cardiovascular comorbidities, and hematologic cancers [1–4]. Women who breastfeed experience health benefits, including lower rates of hypertension, endometriosis, diabetes mellitus and metabolic syndrome, cardiovascular disease, and ovarian and breast cancers [5–7]. A longer duration of breastfeeding is associated with decreased all-cause mortality [5–7]. Further, there are mental health benefits (decreased postpartum depression) and greater infant–parent bonding that is facilitated by breast-/chestfeeding [8, 9]. Breast-/chestfeeding may also have a myriad of other benefits, including lower cost, convenience, availability, and safety. Formula may be prohibitively expensive for some consumers who may fall outside of income guidelines that qualify them for Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) assistance [10], a social program that provides assistance to low-income pregnant persons in the United States. The WIC program is a government-funded program that aims to safeguard the health of low-income women, infants, and children up to age 5 years who are at nutrition risk through providing them nutritious foods, information on healthy eating, and healthcare referrals. Similar programs are in place in Canada (Canada Prenatal Nutrition Program [11] and the UK Healthy Start Scheme [12]). Replacement feeding is not always safe or available in all parts of the United States, as is evident from both the Flint, Michigan, water crisis [13] and recent formula recalls due to bacterial contamination and supply chain problems [14].

It is critically important to acknowledge that there are overwhelming racial disparities that may be related to breastfeeding initiation and continuation, with Black individuals having higher infant mortality (overall and related to prematurity and SIDS); maternal morbidity and mortality; and comorbidities such as diabetes, obesity, and breast cancer outcomes; and disproportionately higher rates of human immunodeficiency virus (HIV), particularly, but not exclusively, in the United States [15–17]. Specifically, Black women are disproportionately affected by new and prevalent HIV infections, representing 54% of new HIV infections among women in the United States in 2019 [17]. Historically, Black women were 20.9 times more likely than White women (2001–2004 data) and 4.1 times more likely than Hispanic women to acquire HIV [18]. While overall increasing, rates of Black non-Hispanic infant breastfeeding are significantly lower than any other racial or ethnic group, with 73.7% ever breastfeeding and 47.8% breastfeeding at 6 months in 2017 [19, 20]. Black women in the United States also have had a complicated history with breastfeeding, historically forced to feed the infants of slave masters [21, 22]. Enslaved Black women were forced to breastfeed the infants of their slave masters and, following slavery through the 1940s, many still served as wet nurses for White families. This has been associated for some with lack of choice, a cultural mindset that breast milk was only good for the master's children and not their own, low rates of breastfeeding their own children, and with cultural normalization of not breastfeeding and shaming if one chooses to breastfeed [23, 24]. Additional factors affecting successful breastfeeding in this group include systemic racism, lack of cultural normalization, lack of support from family or community members, financial pressures necessitating a quicker return to work, shorter maternity leaves, and employment that does not provide breastfeeding support [25–28]. Systemic racism refers to the policies and practices that exist throughout a society that perpetuate unfair disadvantages for a specific population. Systemic racism in healthcare, as it pertains to breastfeeding, has resulted in the erroneous belief that Black women prefer bottle feeding over breastfeeding, making them less likely to be given breastfeeding support or resources. In employment, systemic racism has led to Black women being more likely to be in low-wage, inflexible jobs where they are not afforded the space or place to be able to exercise the option to breastfeed without consequence [25–28].

Breast-/chestfeeding may have additional unique benefits for individuals with HIV and their infants. HIV-exposed infants are at higher risk of general morbidity and mortality and would benefit most from breast-milk exposure, primarily due to undernutrition, stunting, and modest impairments in early childhood development [29]. Further, with the increasing rates of non–AIDS-defining comorbidities that can potentially be impacted by breast-/chestfeeding, individuals with HIV who can and want to breast-/chestfeed should be afforded that opportunity [30–32]. Indeed, many racial and health disparities that intersect for people with HIV may be perpetuated by the absence of breast-/chestfeeding [33].

The most recent Department of Health and Human Services’ (DHHS’) Perinatal Guidelines call for shared decision making with patients, which rightfully acknowledge patient autonomy, meeting increasing patient requests, and addressing the changing realities of perinatal HIV care [34]. Providers should also not overlook the power of patient autonomy when thinking about breastfeeding. The decision to breast-/chestfeed or formula feed is a deeply personal choice informed by one's personal and psychosocial and societal context and should be respected. Individuals who are HIV-positive may choose breast-/chestfeeding for the same reasons as an HIV-negative individual; however, there may also be additional reasons such as the risk of unintentional/inadvertent disclosure of one's HIV status by not breastfeeding in some communities. The goal of this Viewpoint article is not to vilify formula/replacement feeding, particularly as not all individuals with or without HIV can or want to breast-/chestfeed, but to clearly provide support for breast-/chestfeeding as a viable option for individuals with HIV.

PREVENTION OF MOTHER-TO-CHILD TRANSMISSION SUCCESS—CAN IT APPLY TO BREASTFEEDING?

Having an HIV diagnosis does not alter desires for childbearing and particularly with the enhanced preventing mother-to-child transmission (PMTCT) toolkit, inclusive of early HIV testing before and during pregnancy, antiretroviral therapy (ART) for the parent and the child, there are increasing numbers of individuals who are choosing to bear children [35]. The PMTCT toolkit has resulted in a decrease in the risk of HIV transmission during pregnancy and the intrapartum period from 25% to less than 1% in the setting of viral suppression during pregnancy and delivery [36, 37]. In fact, the United States recently achieved “elimination” of vertical transmission [38]. With such effective PMTCT strategies and with the overwhelming data regarding treatment as prevention (TasP) and undetectable equals untransmittable (U = U), childbearing individuals with HIV and advocates in high-income settings began to increasingly ask, Why should individuals with HIV be categorically prevented from breast-/chestfeeding their infants?

It is important to review where the concern regarding HIV transmission during infant feeding arose and how the data have evolved. Overall, the risk of perinatal transmission via breastfeeding has dramatically decreased with the described interventions, although there remains some reluctance to confidently declare that “U = U” applies to “optimal” exposures in the setting of maternal ART and undetectable viral load. Observational studies in the early 1990s reported HIV transmission in the context of breastfeeding occurring in 29% of cases (95% confidence interval [CI]: 16–42%) [39] prior to the wide availability of effective maternal ART. In 1996, the Joint United Nations Programme on HIV/AIDS (UNAIDS) recommended that women with HIV-1 living in low-resource settings be encouraged to make informed choices about infant feeding [40, 41], in part due to reduced morbidity and mortality associated with breastfeeding, particularly when compared with the morbidity with formula feeding related to infectious causes due to unsafe water and limited or unreliable availability of infant formula [42]. Notable studies demonstrated decreased infant morbidity and mortality from breastfeeding even in the setting of HIV transmissions [43, 44] in Brazil and South Africa in the era prior to effective ART. An observational study of Brazilian mother–infant dyads showed that 69 of 432 (16%) infants exposed to HIV through breastfeeding were diagnosed with HIV. Advanced maternal age, longer duration of breastfeeding, and mixed feeding (consumption of food or cereals prior to 3–6 months of age) were associated with a trend towards increased risk of transmission, although not to a statistically significant degree [43]. In an observational study from Durban, South Africa, 549 women with HIV-1 and their singleton infants who were formula-fed, exclusively breastfed, or exposed to mixed feeding were followed to estimate the proportion of infants who acquired HIV-1 [44]. Infants who exclusively breastfed (n = 103) were at the lowest risk of HIV-1 infection by 3 months of age (odds ratio [OR], 8.3; 95% CI: 2.8–13.9) compared with those who were mixed fed (had other food introduced before 3 months; OR, 19.9; 95% CI: 15.0–24.9). Subsequent studies explored the addition of maternal or infant ART to reduce HIV transmission. The Mashi study (2001–2003) in Botswana [45], designed to assess the efficacy of adding single-dose nevirapine (NVP) to maternal and infant zidovudine (ZDV) to reduce vertical transmission, established that women exposed to ART had lower HIV-1 plasma and milk RNA concentrations without differences in HIV-1 DNA milk concentrations, although it should be noted that women who were offered ART had CD4 counts less than 200 cells/mm³ or an AIDS-defining illness. Follow-up of the Mashi study investigated the safety and efficacy of extended ZDV prophylaxis in breastfed (6-month duration) versus formula-fed (1-month duration) infants; perinatal transmissions occurred in 6% of formula-fed infants and 9.5% of breastfed infants, but women in the breastfeeding group had lower infant mortality [46]. A subsequent study (2004–2007) investigating extended neonatal postexposure prophylaxis (PEP) (single-dose NVP 2 mg/kg plus ZDV 4 mg/kg twice daily × 7 days [control] vs NVP only [2 mg/kg daily for 2 weeks then 4 mg/kg daily for weeks 3–14] or NVP plus ZDV [4 mg/kg twice daily from week 2–5, then 4 mg/kg 3 times daily for weeks 6–8, then 6 mg/kg 3 times daily for weeks 9–14]) determined that either regimen was more effective than control (no infant ZDV) at preventing HIV transmission through breastfeeding—HIV-1 infections at 24 months vs control: 15.6% vs 10.8% with NVP only or 11.2% with NVP + ZDV [47, 48]. The Mma Bana study (2006–2008) [49] demonstrated HIV transmissions in 1.1% of infants whose mothers received ART while breastfeeding. In the breastfeeding, antiretrovirals, and nutrition (BAN) study (2004–2010) [50], breastfeeding mothers were randomly assigned to receive a maternal ART regimen, infant NVP, or no extended postnatal antiretroviral regimen (control group). There were fewer HIV transmissions in the infant regimen group (1.7%; 95% CI: 1.0–2.9%) compared with the maternal regimen group (2.9%; 95% CI: 1.9–4.4%). The Kesho Bora study (2005–2008) demonstrated a significant reduction in HIV breastfeeding transmission in the setting of triple maternal ART during breastfeeding over infant administration of ZDV and single-dose NVP [51]. Additional studies where mothers were exposed to ART while breastfeeding include the PROMOTE [52] and IMPAACT PROMISE trials [53]. In the PROMOTE trial (2009–2013) [52], breastfeeding transmissions occurred in 1.1% of infants (assessed at 12 months); in the PROMISE trial (2011–2014), there were no significant differences in HIV transmissions via breastfeeding between the maternal ART arm (0.7%) and the infant PEP arm (0.8%) [54].

After decades of an unwavering stance that individuals with HIV not breast-/chestfeed, evolving data, the effectiveness of ART in preventing sexual transmissions (ie, U = U), and community advocacy have led to an evolution of guidelines in resource-rich settings, including the United States [55]. Specifically, the US guidelines that first mentioned breastfeeding as an option in 2018, in 2023 incorporated stronger language supporting shared decision making with the patient [56]. There have been an increasing number of reports of breast-/chestfeeding in high-income settings (United States, Germany, Canada, Switzerland) [55–58], highlighting diverse regimens and protocols. There has been one commonality: no transmissions have been reported, underscoring that the evolution toward shared decision making and support of breast-/chestfeeding desires is welcome, needed, and overall safe.

Given that more recent HIV transmission rates are less than 1%, what are the residual concerns? Despite ART exposure, even women with undetectable plasma HIV-1 RNA can experience HIV transmission via breastfeeding [53, 54, 59, 60] (2 transmissions in the PROMISE trial, 2 in Mma Bana, 1 in the DOLPHIN-2 trial, and 2 transmissions in Safe Milk for African Children). We need to fully acknowledge that there are still unknown factors when it comes to U = U, which we cannot confidently say applies to breast-/chestfeeding at this time. Cell-associated HIV-1 DNA may still be present despite undetectable maternal plasma viral loads [45], although its significance is unclear. Further research is needed to answer questions about factors that still make HIV transmissible through breast milk in the setting of maternal ART, infant PEP, or both [34]. Additionally, safe replacement feeding options are mostly available in the United States and other high-income countries. Formula use has been established as an alternative with zero risk of HIV transmission. There are concerns about the long-term side effects associated with prolonged infant PEP. While serious adverse effects are uncommon from breast-milk exposures to nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs)/nonnucleoside reverse transcriptase inhibitors (NNRTIs)/protease inhibitors (PIs)/integrase inhibitors (INSTIs), long-term data are lacking [61–63]. There is also a small, but present, risk of transmitting resistant virus. Fogel et al [64] analyzed NVP resistance in infants who acquired HIV through breastfeeding while taking NVP as PEP exposure and found that NVP resistance was detected in 92% of infants who acquired HIV by 6 weeks and in 75% of infants who were diagnosed with HIV by 6 months. Nevirapine resistance was also assessed in infants who acquired HIV in the 3 arms of the BAN study (daily infant NVP prophylaxis, triple maternal ART, or no intervention for 28 weeks of breastfeeding) [65]. Infants acquiring HIV while receiving daily NVP were significantly more likely to develop resistance: 56% in the infant-NVP arm versus 6% in the maternal ART arm versus 11% in the control arm (P = .004). Boyce et al [66] performed a case-control study of PROMISE study participants in order to investigate whether or not NNRTI drug resistance in pregnant women and breastfeeding mothers increased the risk of HIV vertical transmission. The study found that maternal HIV drug resistance and maternal viral load were independent risk factors for vertical transmission during breastfeeding and concluded that NVP alone may be insufficient infant PEP against drug-resistant variants in maternal breast milk. HIV drug resistance was identified more frequently among transmitting versus non-transmitting mothers at the time of infant HIV diagnosis (14.6% vs 6.7%). In high-resource settings with INSTI-based regimens likely to be more widely used, there may be a higher barrier to resistance but also different resistance patterns emerging. Regardless, overall HIV transmission in high-resource settings has been minimal [67–69]. Data from a study of pretreatment INSTI resistance suggest a low rate among ART-naive nonpregnant persons (2.4%; 95% CI: 1.5–3.6%) compared with ART-experienced persons (9.6%; 95% CI: 8.3–11.0%) [70], although such data among pregnant persons are lacking. Some patients and/or providers may feel apprehensive about any additional transmission risks with breastfeeding after successfully navigating pregnancy with full adherence to guidelines.

Finally, just as successful ART use can enable people to have condomless intercourse and still prevent HIV transmission, successful maternal ART that results in undetectable maternal plasma HIV-1 RNA may, in the vast majority of cases, significantly decrease the risk of HIV transmission through breast/chest milk while still allowing HIV-exposed infants and their parents to benefit from breast-/chestfeeding.

HOW TO MOVE FORWARD?

Providers caring for pregnant people with HIV have an obligation to help patients make the best decision for themselves and their infants while navigating the evolving data about the risks of HIV transmission in the setting of maternal/parental ART and viral suppression in high-income settings. We recommend assessing each case individually, establishing a multidisciplinary consultation team to provide patients with evidence-based information about infant feeding options, including risks, benefits, and alternatives (donor milk, heat-treated milk, formula), and partnering with community organizations to further empower patients to make the best informed infant feeding choice for them and their circumstances [34, 71]. Although there are many factors to consider, providers can uphold the standards of beneficence, justice, and patient autonomy in ways that do not compromise clinical care for both parents and their infants.

References

Author notes