Since the PRISM study in 1998 (1), neurologists have reported that the likelihood of a woman experiencing a multiple sclerosis (MS) relapse was reduced during pregnancy, especially during the third trimester, and again increases in the first 3–6 months postpartum, compared with her pre-pregnancy risk status (1–3).

In this scenario, the effect of breastfeeding on postpartum relapses was unclear: some studies showed potential benefits (4, 5), and others did not (6, 7). Consequently, in clinical practice, women were often offered to restart their disease-modifying drug (DMD) soon after delivery to reduce the chances of having an MS relapse. Limited available evidence about DMD safety meant many women avoided breastfeeding, despite the World Health Organization's recommendation to exclusively breastfeed children for the first 6 months of life (8).

This mini-review discusses recent evidence about a protective role for breastfeeding and provides clinicians with a practical tool for a patient-driven approach in evaluating women with MS planning to breastfeed.

To understand the biological mechanisms behind the protective role of exclusive breastfeeding in MS, we consider the hormonal and immunological changes characterizing women's reproductive years, from menstruation to pregnancy and puerperium (Figure 1).

The protective role of pregnancy, which MS shares with other autoimmune conditions, has been known for decades (1).

During pregnancy, there is a shift from a lymphocyte T-helper (Th)-1 [interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), interleukin (IL)-2] to a Th-2 (IL-4, IL-10) immune response to protect the fetal-placental unit (10). High levels of estradiol and progesterone suppress lymphocyte Th-1 activity (11), which plays a crucial role in promoting MS inflammatory activity, while placental factors sustain a Th-2 response (10). The effects of estrogen on MS activity have been studied in experimental autoimmune encephalomyelitis (12), cuprizone mice (13), and in clinical trials using estriol (14, 15).

As drop in sex hormones after delivery is associated with increased annualized relapse rate (ARR) postpartum, researchers suggested starting protective treatments from the third trimester or immediately postpartum (16).

However, not all women with MS experience rebound activity after delivery. Other immunological factors may influence the changes in relapse risk in pregnancy and postpartum, such as changes in lymphocyte T-regulatory cells (Th-17) (17) and natural killer cells (18) that occur from late pregnancy. In these studies, however, researchers did not investigate if exclusive breastfeeding impacted the immunological changes.

Researchers started to observe reduced ARR in MS women who exclusively breastfeed compared with non-breastfeeding women (5). A small study showed that MS women having relapses postpartum had increasing levels of IFN-gamma from the third trimester, but this remained low in women who exclusively breastfed (9).

Exclusive and frequent feedings (every 3–4 h) maintain high prolactin and low luteinizing hormone (LH) levels, inducing ovarian suppression (lactational amenorrhea). Since menstrual cycles are involve inflammatory processes that probably serve to prepare the endometrium for embryonic implantation (19), their suppression is could be related to anti-inflammatory effects (20). Prolactin has a controversial role: on the one hand, this hormone has recognized inflammatory activity (21), on the other hand, prolactin may help oligodendrocyte precursors to repair damaged myelin (22).

Whether a predisposition emerges during pregnancy has yet to be defined, and studies that carefully evaluate the role of breastfeeding are still lacking. Furthermore, other factors such as parity, body mass index, and fetal sex may influence the postpartum ARR risk (23).

For instance, fetal sex may influence maternal immune activity during pregnancy. A study found that a male fetus may increase maternal levels of pro-inflammatory cytokines (e.g., IL-12p70, IL-21) (24). In fact, histological studies showed that placentae from male premature babies had greater markers of inflammation than female ones, suggesting a more robust maternal immune response to male fetuses (25). However, other studies could not confirm the association between fetal sex and increased maternal production of inflammatory proteins (26). A recent study by Ross et al. considered multiple factors (23), including fetal sex, body mass index and parity, showing different patterns of inflammatory markers in pregnant women. Their findings highlight the complexity of pregnancy and postpartum regarding immune system maternal adaptation, which is likely to be influenced by multiple factors.

The PRIMS study (1) was the first large prospective study on MS and pregnancy. The effects of exclusively breastfeeding on relapses were not a specified outcome. However, a subgroup analysis reported no associations between breastfeeding and relapse rate in the first 3 months postpartum.

Since then, there has been controversy about the protective role of breastfeeding on postpartum annualized relapse rate (ARR). Prospective and retrospective studies have followed, and, although authors agree that breastfeeding has no adverse effects, some have reported beneficial effects on ARR (27, 28), whilst others have not (16, 29). One of the main biases was that the early studies included patients who were breastfeeding in general, regardless of its exclusivity and duration. As discussed before, it is thought to be lactational amenorrhea, promoted by exclusive breastfeeding, that seems to suppress inflammatory activity.

In 2009, Langer-Gould et al. (5) were the first to differentiate between exclusive and non-exclusive breastfeeding, finding an ARR significantly higher in women who did not exclusively breastfeed, irrespective of their 2-year pre-pregnancy ARR or treatment history. This observation was in line with evidence for the protective role of lactational amenorrhea (Figure 1). A meta-analysis (30) found women with MS who breastfeed were almost half as likely to experience a postpartum relapse as women who did not. However, significant heterogeneity was present (63%), driven mainly by the variable duration of the postpartum follow-up.

Although two large observational studies nourished the debate (31, 32) on breastfeeding and MS, showing different results (4, 7), recent works have supported the benefits of breastfeeding (33, 34), as summarized by a recent meta-analysis (35). Krysko et al. (35), with an overall estimated heterogeneity of 48%, showed that those who breastfed had 37% lower odds of postpartum relapse than those who did not breastfeed or did not exclusively breastfeed postpartum.

DMDs target neuroinflammation and are approved for patients with relapsing-remitting MS (RRMS) and active secondary-progressive MS (SPMS) patients (36). Approximately 30% of mothers with MS may relapse within the first 3 months postpartum (1). Effective disease management during this challenging period is essential for the wellbeing of the mother and baby.

Nine classes of DMDs with different mechanisms of action and routes of administration are currently available in MS. For each class, there are limited data available, with some studies showing individual variability (37). However, most DMDs for RRMS are not advised during breastfeeding by manufactures, the Food and Drug Administration (FDA) and European Medicines Agency (EMA) (38, 39).

The drug's molecular weight is probably the most critical factor in determining transfer to breast milk, but oral absorption is also crucial for the child. Poorly absorbed drugs are not likely to enter the child's bloodstream and cause pharmacological effects (40).

High-dose intravenous or oral methylprednisolone is first-line treatment for acute, disabling MS relapses to hasten recovery. It is typically administered for 3 or 5 days (56).

For the mother, potential side effects of corticosteroid exposure on the baby may include suppression of growth, changes in behavior or appetite, acne, and sleep disorders (57). Particular attention should be paid to mothers suffering from postpartum depression.

A study on the concentrations of methylprednisolone in the breast milk of 16 patients with MS showed that transfer of corticosteroids into breast milk is very low (58). Methylprednisolone levels dropped significantly 4 h after intravenous infusion, and the RID was 0.71% of the weight-adjusted maternal dose, which was >10 times lower than the accepted theoretical RID of 10%. It also found that fully breastfed infants would receive a lower dose than their daily cortisol output. These findings are consistent with previous reports obtained in smaller sample sizes (58). If possible, mothers should wait 2–4 h after infusion to further limit the infant's exposure.

MRI with gadolinium to confirm a relapse is not usually necessary in clinical practice. However, if its administration is considered necessary (e.g., in a differential diagnosis with progressive multifocal leukoencephalopathy), most guidelines state that breastfeeding need not be disrupted after a nursing mother receives a gadolinium-containing contrast medium (59), as only small amounts of this molecule are detectable in the milk.

Prepartum choices influence decisions about DMDs in the postpartum period. DMD initiation/discontinuation must be carefully evaluated and agreed with the patient before family planning. Clinicians should pay particular attention to natalizumab and fingolimod, whose cessation can cause relapse exacerbation during pregnancy and postpartum.

At present, there are no studies directly comparing exclusive breastfeeding vs. DMDs in terms of efficacy on postpartum ARR.

Figure 2 offers a flowchart for possible treatment choices.

The primary biological mechanism driving the protective role of exclusive breastfeeding is ovarian suppression. This suppression is usually maintained until the infant has been weaned consistently to solids (usually around 6–9 months postpartum). Therefore, it is safe to conclude that women with MS can exclusively breastfeed until solids are needed, and this could have a protective effect for at least 1 year (35). More evidence is required for longer durations. Although it has been highlighted (35) that higher prepartum ARRs might be even related to lower postpartum ARRs in exclusively breastfeeding women, further studies must confirm this. Therefore, neurologists should adopt a careful approach for women with high prepartum ARR. Finally, MS women planning pregnancy should be offered multi-disciplinary consultations to decide on their treatment journey. Particular attention should be paid to DMDs that have a relatively high risk of rebound. Then, during pregnancy and breastfeeding, MS women should be carefully monitored for relapses and DMDs should be reinstituted if indicated.

SC has done the figures included in the manuscript. All authors have contributed in the literature review and in drafting the manuscript. All authors contributed to the article and approved the submitted version.