Delayed Lactogenesis II and potential utility of antenatal milk expression in women developing late-onset preeclampsia: a case series

Preeclampsia is a heterogeneous and multi-system hypertensive disorder of pregnancy that affects ≈3–5% of pregnancies globally, causing significant maternal and neonatal/infant morbidity and mortality [1, 2]. The development of preeclampsia stems from the interactions between a dysfunctional placenta and maternal constitutional factors, which culminate in systemic endothelial dysfunction (e.g., maternal hypertension, proteinuria, activation of coagulation cascade) [3‐5]. Preeclampsia is diagnosed when a woman presents with new onset hypertension after 20 weeks’ gestation with proteinuria and/or signs of multi-system involvement (e.g., thrombocytopenia, renal insufficiency). Preeclampsia is characterized as early (< 34 weeks’ gestation) or late (≥ 34 weeks’ gestation) and with/without severe features indicative of multi-system organ involvement [6]. Certain pre-pregnancy and pregnancy related factors, including pre-pregnancy obesity, chronic hypertension, type 1 and 2 diabetes mellitus, advanced maternal age, and nulliparity are associated with an increased risk of preeclampsia [7].

Although the acute signs/symptoms generally resolve in the days and weeks following delivery [8], preeclampsia is associated with an increased risk for cardiovascular disease remote from pregnancy. In women with a history of preeclampsia, the risk of ischemic heart disease and cerebrovascular events is doubled, the risk of hypertension is tripled, and the risk of incident heart failure is quadrupled [9‐12]. Women with hypertensive disorders of pregnancy also have a significantly higher risk of mortality related to diabetes, ischemic heart disease, and diabetes [13, 14]. Children exposed to a preeclamptic pregnancy may be at increased risk for cardiovascular dysfunction in adolescence and adulthood [15‐17].

Breastfeeding may be one way to counteract the negative cardiovascular sequelae associated with preeclampsia. Increased duration of lactation in women is associated with reduced later life risk of hypertension, type II diabetes, hyperlipidemia, metabolic syndrome, subclinical and clinical cardiovascular disease, and cardiovascular mortality [18‐22]. Breastfed children have a reduced risk of childhood obesity and diabetes [23, 24]. Paradoxically, however, women who develop preeclampsia appear to be at-risk for premature breastfeeding discontinuation and early formula use [25, 26], for reasons which have not been previously described.

Here, we present three cases of late-onset preeclampsia and one case of severe gestational hypertension that illustrate a potential association between hypertensive disorders of pregnancy and suboptimal breastfeeding outcomes, including delayed onset of lactogenesis II and in-hospital formula supplementation. Lactogenesis II is defined as the onset of copious milk production, which typically occurs between 48 and 72 h postpartum; onset after 72 h is considered delayed and is associated with unintended breastfeeding reduction and cessation [27, 28]. Furthermore, we describe the use of antenatal milk expression (AME), or milk expression and storage beginning around 37 weeks of gestation, as a potential intervention to increase breastfeeding success in women at risk for preeclampsia and other hypertensive disorders of pregnancy.

All cases were drawn from an ongoing pilot randomized controlled trial investigating the impact of antenatal milk expression versus an education control on breastfeeding outcomes. All trial participants were healthy nulliparous women with singleton pregnancies recruited at 34–36^6/7 gestational weeks from a hospital-based midwife practice in the northeastern United States. Participants were recruited via study flyer and at their prenatal appointments after review of electronic medical records for eligibility criteria. At enrollment, women had no major risk factors for insufficient milk (e.g., breast hypoplasia, polycystic ovarian syndrome, diabetes) or preterm labor (e.g., vaginal bleeding after the first trimester, congenital anomalies, polyhydramnios). The AME protocol was modeled after an existing randomized controlled trial of AME in women with diabetes [29] and involved demonstration of hand-expression of milk beginning at 37 weeks of gestation with a study lactation consultant, weekly visits with the same lactation consultant to reinforce technique, and at-home, independent milk expression and collection one to two times per day for up to 10 min, for up to 40 gestational weeks. Women were provided containers in which to collect and freeze milk. Among women who brought milk to the hospital, they were advised to transport it in an insulated bag or cooler with two icepacks, and milk was subsequently stored in the hospital refrigerator for a maximum of 24 h. Women in the education group received weekly hand-outs addressing common breastfeeding problems from gestational weeks 37 to 40. We followed-up with all study participants at 1–2 weeks and 3–4 months postpartum to assess breastfeeding status and conduct interviews about experiences with AME.

Cases were obtained from the first third of participants recruited in the study, as an early trend was observed soon after enrollment commenced regarding development of hypertensive disorders of pregnancy and suboptimal breastfeeding outcomes. Case data was compiled from study participants’ electronic medical records (e.g., details on labor and delivery, feedings in hospital), study surveys (e.g., breastfeeding plans, timing of lactogenesis II, feedings after hospital discharge), antenatal breast milk expression logs (e.g., volume and timing of milk collection), and interviews (e.g., perceived impact of AME). Data were organized in tables to facilitate comparison across participants. In the presentation of all cases, we adhered to CARE case report guidelines guidelines, which specify inclusion of particular elements to increase transparency and accuracy of data (e.g., inclusion of dose/strength/duration of intervention, patient perspective).

A summary of the cases can be found in Table 1.

Three of the cases presented here illustrate a potential relationship between late-onset preeclampsia and other hypertensive disorders of pregnancy and early breastfeeding problems, including delayed onset of lactogenesis II and in-hospital formula use. Case 3 represents an exception to this association, with normal timing of lactogenesis II and exclusive breastfeeding in-hospital. The variability in clinical presentation among the four cases supports that any potential effect of preeclampsia or other hypertensive disorders of pregnancy on breastfeeding outcomes is likely multifactorial in nature, with possible primary etiology (e.g., preeclampsia disease course itself) and secondary causal factors (e.g., underlying maternal issue contributing to both preeclampsia and lactation problems, preeclampsia treatments).

We found two studies addressing the relationship between hypertensive disorders of pregnancy and breastfeeding outcomes, but none examining timing of lactogenesis II. Cordero et al. [25] found in their retrospective medical record analysis of 281 women diagnosed with severe preeclampsia and delivering ≥ 34 weeks of gestation, 51% of the overall sample and 78% of those intending to breastfeed initiated breastfeeding; however, only 33% of the sample was exclusively breastfeeding and 22% receiving any breast milk at the time of hospital discharge, which was lower than U.S. averages at the time but perhaps comparable to the region in which the study occurred [31, 32]. In a retrospective cohort study involving mothers with hypertensive disorders of pregnancy and matched controls, breastfeeding intention and continuation at 1 month postpartum was significantly lower among women with preeclampsia than control women [26]. These studies lend credence to a potential negative effect of preeclampsia or related factors on breastfeeding, though mechanism of action is unclear.

Three of four cases presented with lactogenesis II occurring after 3 days postpartum. Onset of lactogenesis II is triggered by a complex hormonal cascade, initiated by delivery of the placenta and a concomitant precipitous drop in progesterone. The methods by which various factors facilitate or impede this process are not well understood. Delays in lactogenesis occur at a rate of 23–44% in the U.S. [28, 33]. This is clinically significant, as delayed lactogenesis, and its corollary—formula supplementation, are both are strong prognosticators of early, unintended breastfeeding cessation [28, 34]. Delayed lactogenesis is more likely to occur among primiparous, obese, and diabetic women, as well as those who have a prolonged second stage of labor (greater than 2–3 h for nulliparous women) [35], stressful labor or delivery, cesarean section, higher birthweight infant, postpartum edema, or a perception of ineffective breastfeeding during the first 24 h postpartum [33, 36‐39]. Women in the presented cases possessed a number of these risk factors, many of which are also associated with preeclampsia and/or its treatment, including primipara status, delivery by cesarean section, stressful/prolonged labor and delivery, and early breastfeeding problems, including formula use.

While postpartum edema was not tracked as a potential cause of delayed lactogenesis II, it is a common occurrence in preeclampsia as vasculature becomes “leaky,” resulting in third-spacing of fluids [40]. It is thus possible that lower intravascular volume contributed to poor diffusion of water into mammary alveoli and delayed the onset of copious milk production until fluid balance was restored. Alternately, postpartum breast edema can contribute to difficulty in early breast milk removal, a risk factor for delayed lactogenesis II [41].

Other factors related to hypertensive disorders of pregnancy may contribute to an unfavorable endocrine profile for establishment of lactation, including stress [38, 42] and obesity [43, 44], the latter potentially relevant for Case 4 with a BMI of 30. Placental dysfunction, which typifies preeclampsia, has been generically noted as a potential contributor to delayed onset of lactation [45]; it is theoretically possible that placental hormones impacting proliferation of glandular breast tissue in preparation for breastfeeding (e.g., progesterone) are compromised in preeclampsia.

Postpartum hemorrhage, as occurred for Case 2, has the potential to contribute to delayed or insufficient milk and negative breastfeeding outcomes through several mechanisms. These include anemia and necrotic damage or shock to the pituitary gland from hypovolemia. Pituitary dysfunction affects secretion of key lactation hormones, including prolactin necessary for milk production [46]. Labor complications associated with preeclampsia, including inductions, cesarean sections, and prolonged labors are also risk factors for postpartum hemorrhage.

It is possible that medications used to treat and manage preeclampsia impacted milk supply in these cases. For example, diuretics such as Case 4 received, may contribute to a fluid deficit in relation to the volume needed for adequate milk production [47]. Case 4 was also taking a selective serotonin reuptake inhibitor (SSRI) daily; SSRIs are associated with delayed lactogenesis, potentially through perturbation of the mammary gland epithelial cells [48]. Magnesium sulfate therapy is routinely utilized for seizure prophylaxis in women with preeclampsia during the intrapartum and postpartum periods. A case report from 1993 describes a woman who received ≈ 4 days of magnesium sulfate therapy (102 g total) for management of preeclampsia and experienced delayed lactogenesis to postpartum day 10 [49]. Traditional reasons for delayed lactogenesis (e.g., delayed at-breast feeding, decreased breast stimulation) were ruled out, and it was hypothesized that the disease process or the magnesium therapy itself impacted lactogenesis. Other studies have found a dose-dependent association of magnesium sulfate therapy with delayed initiation of breastfeeding [50] and factors that impact delayed breastfeeding, including decreased infant suckling behavior, lower Apgar scores, hypotonia, and NICU admissions [51, 52]. Two of the four cases presented in this case series received magnesium sulfate therapy (Case 1: 56.4 g total; Case 4: 83.8 g total) and experienced delayed lactogenesis II. Unfortunately, research investigating the impact of magnesium sulfate therapy on lactogenesis is still lacking.

Reduced opportunity for early and effective breast stimulation and emptying, as may occur with maternal infant separation and medical complications in preeclampsia management, are known risk factors for reduced breast milk production [33, 53]. Research suggests that the earlier postnatal milk removal and suckling occurs, the more rapid the conversion to lactogenesis II (copious mature milk) post-birth and the larger the postpartum milk supply [54, 55]. One explanation is the prolactin receptor theory, which posits that early breast stimulation increases the number of prolactin receptors in breast tissue; these receptors set the theoretical maximum volume of milk production and may only be amenable within a short period around birth [56]. This may be particularly relevant for Cases 2 and 4 who delivered prior to 39 weeks gestation, as preterm and early term infants generate reduced negative pressure during suckling and removal of less milk [57].

As illustrated in Cases 1 and 3, antenatal milk expression (AME) may address a number of breastfeeding complications experienced by women with late onset preeclampsia. For example, milk stored in pregnancy can be used as supplemental feeding when at-breast feeds are not possible or successful following a complicated birth. AME may also increase prenatal confidence and commitment to breastfeeding, allowing mothers to persevere despite early postpartum setbacks (however, AME may also have unintended consequences of decreasing confidence if mothers are not able to express any milk or only small volumes). Finally, existing research [58, 59] and anecdotal feedback in our study suggest that AME may contribute to increased volume of milk both prior to and following lactogenesis II and improved short-term breastfeeding outcomes. Forster et al. [58] found that AME beginning at 36 weeks of gestation among diabetic women at low-risk for birth complications did not increase risk of NICU admission or rate of preterm birth. Because nipple stimulation in pregnancy can theoretically lead to uterine contractions and perhaps placental perfusion problems, the authors advised against unsupervised AME implementation in other high-risk groups. In our study, AME was not continued after women were diagnosed with any pregnancy complications. More research is warranted to understand the risks and benefits of AME among women who develop preeclampsia and other issues associated with postnatal breastfeeding problems. In the interim, women with prenatal risk factors for these problems, including nulliparity, with an otherwise low-risk pregnancy, may benefit from AME prior to onset of disease.

In this sample of women with hypertensive disorders of pregnancy, despite early setbacks and formula use, all were breastfeeding with little to no formula use by three to 4 months postpartum. It is unclear whether this would have been the case in a more diverse sample of women with less breastfeeding education and support. Thus, limitations of this analysis include the generalizability of findings based on the case study format and the demographically homogenous sample of women. Strengths include utilization of multiple data sources enabling methodological triangulation (e.g., electronic medical record data and maternal self-report of in-hospital feedings) and the prospective nature of data collection from pregnancy through the postpartum period, minimizing the potential for recall bias. Future research should address incidence, etiology, and interventions for breastfeeding issues among larger samples of women who develop late onset preeclampsia, while controlling for confounding variables.