Does the Academy of Breastfeeding Medicine’s Clinical Protocol #36 ‘The Mastitis Spectrum’ promote antibiotic overuse and risk worsened outcomes for breastfeeding women? A discussion paper
Revision of the Academy of Breastfeeding Medicine’s Clinical Protocol concerning lactation-related breast inflammation is timely, given the exponential increase in knowledge about the human milk microbiome over the past decade. Clinical Protocol #36 introduces important concepts, updating the Academy’s recommended clinical approach to diagnoses such as engorgement, blocked ducts, mastitis and abscess. However, I propose that Clinical Protocol #36 offers certain clinical recommendations which may be of no benefit or even worsen outcomes for breastfeeding women*, and I raise concerns about the Academy of Breastfeeding Medicine’s scientific processes for developing Clinical Protocols. This discussion paper by an Academy of Breastfeeding Medicine member does not intend offence, but aims to promote respectful discussion and debate amongst clinicians and researchers within the Academy of Breastfeeding Medicine and more broadly, whose primary commitment is to improving the well-being of breastfeeding women and their infants.
The updated Clinical Protocol #36: The Mastitis Spectrum (ABM CP#36) by Mitchell et al,1 published May 2022, makes advances in our approach to lactation-related breast inflammation.
CP#36 introduces recommendations which agree with the analyses and clinical guidelines I’d published earlier in the international open access journal Women’s Health, in July 2021, February 2022, and April 2022. My analytic syntheses of the research literature concerning lactation-related breast inflammation occur through the theoretical frames of evolutionary and complexity science, and translate into clinical guidelines for the foundational breastfeeding domain of the comprehensive programs known as Neuroprotective Developmental Care.2-4
The following elements of CP#36 are consistent with my own prior publications, which propose that breast inflammation:2-5
- Is a spectrum condition
- Is not helped by and is likely to be worsened by deep lump massage
- May elicit a systemic response that is not necessarily infective
- Mostly resolves with conservative care
- Is not the same as normal lactational glandular tissue which can feel ‘lumpy’ and tender
- Does not develop into infection in a period of hours
- Is not caused by retrograde spread of bacteria from a damaged nipple
- Is not caused by mammary candidiasis
- Is not helped by
- instructions to have baby ‘drain’ the breast
- ‘dangle’ feeds
- topical applications
- Does not require investigations of c-reactive protein or white blood cell count, since these are markers of inflammation not specific for infection.
Despite these advances, I’m concerned that ABM CP#36 contains scientific flaws, which result in recommendations which are of no benefit or may even risk worsened outcomes for some breastfeeding women. Although some recommendations which lack an evidence-base or credible theoretical frame (e.g. ‘lymphatic drainage’) may appear on the surface to be benign, unnecessary interventions exacerbate anxiety and disempower breastfeeding women, increase financial burden, and limit access to breastfeeding support to affluent women within advanced economies.6-9
1. Clinical Protocol #36 does not consistently attribute sources in accordance with scientific conventions
To give one example, in Figure 4 page 362 of ABM CP #36, a photograph labelled ‘mastitis’ appears to show a bacterial biofilm and is compared to a photograph labelled ‘healthy mammary gland’. Critical details such as source, site, preparation, type of organism, and magnification are missing from both images.
The black and white image labelled ‘mastitis’ shows a Staphylococcus epidermidis biofilm. It is a fixed biopsy sample from a human mastitis viewed under electron microscopy at 5000x magnification (personal communication by Professor Juan Rodriguez). Photographic evidence of biofilm in a biopsy of late-stage mastitis, which has been placed in a fixative for microscopy, does not support the hypothesis that pathological biofilm in milk ducts causes breast inflammation.
[Note: A 2022 case-control study from Australia published subsequent to ABM CP#36 examined the breast milk of 20 women with mastitis and 16 women without mastitis, and did not find an association between Staphylococcus epidermidis and mastitis.10]
2. Clinical Protocol #36 claims that dysbiosis is one of two fundamental causes of lactation-related breast inflammation (Figure 1 page 361).
ABM CP#36 hypothesises that pathological biofilm narrows lactiferous ducts, resulting in ductal inflammation and stromal oedema, which cause clinical presentations of breast inflammation. ABM CP#36 claims that “under physiological conditions, coagulase-negative Staphylococci and viridans Streptococci (i.e. S mitis and S salivarius) form thin biofilms that line the epithelium of the mammary ducts, allowing a normal milk flow”.
The physiological, cellular, or biochemical reasons why coagulase-negative Staphylococcus and Streptococcus bacteria – and why these two species and not other micro-organisms – are hypothesised to form a normal ductal biofilm in healthy lactating women (instead of remaining planktonic) are not discussed. This pathogenic microbiota hypothesis is illustrated in Figure 2 page 361, adapted from a 2014 Fernandez et al article which was peer-reviewed and published prior to the explosion in human microbiome and human milk microbiome science.11 12 CP#36 claims that “in the setting of dysbiosis these species proliferate and function under opportunistic circumstances whereby they are able to form thick biofilms inside the ducts, inflaming the mammary epithelium”.
Eubiosis must be defined before a dysbiosis can be described. The human milk microbiome, leukocytes, epithelial cells, oligosaccharides, exosomes and metabolome are part of the complex adaptative system of the mammary immune system.2-4 The milk microbiome is extremely complex and variable, both within the one woman and between women. Taxonomic categorisation of eubiosis in human microbiomes is increasingly thought to be unachievable and less relevant; what matters to the organism and to researchers is the function of interacting elements of the microbiome.13-20
Kvist et al found no correlation between scores for erythema, breast tension, pain or total severity of symptoms and the type of bacteria in breast milk.21 The high counts of Staphylococcus aureus and decreased microbial diversity associated with breast inflammation may be explained as homeostatic and secondary responses of the mammary immune system to breast inflammation, rather than as the cause of breast inflammation.2
3. Clinical Protocol #36 proposes that ‘hyperlactation’ is the second fundamental cause of breast inflammation (Figure 1 p 361).
3A. ABM CP#36 does not offer a workable definition of ‘hyperlactation’
This second key hypothesis poses two problems. Firstly, CP#36 states that “ductal lumens can be narrowed by edema and hyperemia associated with hyperlactation”, rather by pathological biofilm formation. That is, ABM CP#36 implicitly acknowledges the mechanical effects of raised stromal pressure on the lactiferous ducts, which are then compressed, as detailed in my paper.2 But CP#36 fails to offer mechanisms by which ‘hyperlactation’ causes stromal oedema and hyperaemia.
The mechanobiological model of breast inflammation proposes that high intralumenal and intra-alveolar pressures occur when milk production substantively exceeds the infant’s physiological needs, which I refer to as production mismatch, though there are multiple other causes.2 The mechanical effects of excessive pressures trigger inflammatory cascades, resulting in hyperaemia and increased interstitial fluid, which result in increased stromal pressures and compression of lactiferous ducts, cascading into clinical presentation. This is also how the mechanobiological model explains the link found in Cullinane et al’s cohort study of 346 breastfeeding women between pumping a few times a day and mastitis.2 22
3B. ABM CP#36 does not offer a pathophysiological model which explains how ‘hyperlactation’ might cause breast inflammation
Secondly, ABM CP#36 does not provide a workable definition of ‘hyperlactation’, a term which implies comparison with a state of ‘normal’ milk production. For example, normal breast milk volumes over a 24 hour period ranges from 478 and 1356 mls in exclusively breastfeeding women.23 Also, a woman who breastfeeds twins is not in a state of ‘hyperlactation’.
The breastfeeding domain of Neuroprotective Developmental Care proposes that milk production which consistently exceeds an infant’s physiological needs (that is, production mismatch) may occur innately, or in response to patterns of breast use, or excessive pumping. This presents as a clinical problem for the infant (functional lactose overload) or for the woman (recurrent breast inflammation) in a minority of breastfeeding pairs, and does not feature as a common reason for premature weaning.24
3C. Does ABM CP#36 make recommendations concerning management of breast inflammation through a US-centric lens?
The focus on ‘hyperlactation’ may derive from a US-centric interpretative lens. Infant lactose overload or maternal recurrent breast inflammation as a result of production mismatch may occur more often in sociocultural contexts where mechanical milk removal is common, for example, in the United States.25 26 Another example of US-centrism in the development of ABM CP#36 may be evident in the recommendation to “wear an appropriately fitting supportive bra” to prevent or manage mastitis. Although this advice may have a role for a subset of lactating women, it is not, from both an evolutionary and cross-cultural perspective, relevant to most. From the perspective of the evolutionary mechanobiological model, bras are a sociocultural innovation which may predispose some women to mastitis, and require careful management.3
3D. ABM CP#36 recommends reduced milk removal if breast inflammation emerges
By foregrounding ‘hyperlactation’ and generalising this to all breastfeeding women experiencing breast inflammation, ABM CP#36 arrives at the concerning recommendation that milk removal should be reduced when breast inflammation presents.
- Under the heading Key Information: Pathophysiology of Mastitis Spectrum Conditions p 361, ABM CP #36 states: “Reducing milk removal may transiently increase pain and erythema …; however, it ultimately prevents future episodes”.
- Under the subheading Ductal narrowing (e.g., “plugging”) p 363, ABM CP #36 states: “Patients may feel relief of a “plug” with breastfeeding because this decrease alveolar distension. However, repeated feeding in an attempt to relieve the “plug” will suppress FIL, increase milk production, and ultimately exacerbate inflammation and ductal narrowing.”
- Figure 10 p 365 states: “Patient with early inflammatory mastitis … was treated with ice, ibuprofen, acetaminophen, and feeding first off the left, less congested breast first to avoid overstimulation of the affected right breast.”
- Figure 11 p 366 states: “Bacterial mastitis that progressed from early inflammation in the inner quadrant to all quadrants being affected. This patient also pumped and continually fed the infant …. in an attempt to prevent ‘milk stasis’. This approach resulted in worsened ductal inflammation and bacterial overgrowth as well as milk obstruction.” [I contend that a single case report which fails to address multiple potential confounding factors should not be included in international clinical guidelines.]
- Figure 19 p 370 states: “Ice and decreased removal of breast milk reduce ductal narrowing and breast swelling.”
These recommendations directly contradict the approach recommended by clinical translation of the mechanobiological model of breast inflammation. The breastfeeding domain of Neuroprotective Developmental Care (NDC) supports frequent pressure gradient change in the breast stroma (that is, frequent milk ejections and associated ductal dilations) when lactation-related breast inflammation presents clinically, to counter the ductal compression which results from increased stromal pressure.2 A pre-existing mismatch between the mother’s milk production and her infant’s physiological needs is identified at presentation and appropriately managed after the high risk period of clinical inflammation has resolved.3 During the risk period, the infant is offered the breast frequently and flexibly, and not necessarily for long; hand expression may also be used to stimulate ductal dilations.
I have detailed elsewhere the features of the NDC advice to feed frequently and flexibly (See Table 2 Re-thinking benign inflammation of the lactating breast: classification, prevention and management).3
3E. ABM CP#36 recommends not continuing to attempt feeding if no milk is expressible
Under the heading Spectrum-wide recommendations c. Feed the infant on demand, and do not aim to “empty” the breasts, ABM CP #6 states: “In some instances, in which the retroareolar region is so edematous and inflamed that no milk is expressible by infant breastfeeding or hand expression, the mother should not continue to attempt feeding from the affected breast during the acute phase. She can … return to feeding from the affected breast when edema and inflammation subsides.”
In the crisis of an infant who can’t transfer milk and when there is no success with hand expression, a range of gentle strategies should be tried, ongoing, as a matter of urgency. For example, from the NDC perspective, fit and hold problems are often not identified and managed, yet may be underlying causes of severe engorgement or localised breast inflammation, due to the way fit and hold problems cause difficulty coming onto the breast. Also, once on the breast and positionally stable, an infant who may not appear to be transferring much milk is nevertheless applying repetitive vacuum and this is to be encouraged.
Advising cessation of milk removal risks worsened inflammation and abscess formation.
3F. ABM CP#36 recommends that ‘overfeeding’ an infant risks breast inflammation
ABM CP#36 states that “overfeeding from the affected breast …. is a major risk factor for worsening tissue edema and inflammation” (page 367).
Many women who believe they are ‘demand breastfeeding’ don’t offer the breast in a way that secures adequate milk production and adequate infant weight gain, due to culturally prevalent but inaccurate concerns about overfeeding a breastfed infant.
From an evolutionary perspective, it is not possible to overfeed a breastfed infant and infants will cue or communicate when they are not interested in the breast. NDC addresses multiple, often medicalised, misconceptions about infant cues.27 28 The NDC concept of frequent and flexible feeds contends that an infant self-regulates at the breast, that the breast is a ‘tool’ for dialing down the infant’s sympathetic nervous system, and that the infant can be offered the breast without waiting for cues.3
Most women do not have ‘hyperlactation”; indeed, low supply and poor weight gain are the main reasons women give worldwide for supplementing with formula.24 Popular advice to limit feeds to about eight times a day from birth, driven by fear of overfeeding, threatens many women’s capacity to meet their infant’s physiological milk needs, increases the risk of clinical breast inflammation, and is likely to not be enough to downregulate a clinical presentation of breast inflammation.
Therefore, ABM CP#36’s recommendation to not increase or to reduce milk removal does not address an possible underlying causative factor for breast inflammation (that is, supply downregulating due to the infant’s inadequate access to breast), which then may also increase the risk that a woman’s milk production fails to meet her infant’s needs after the breast inflammation has resolved.
4. Clinical Protocol #36 classifies postpartum engorgement as “a distinct clinical entity related to interstitial edema and hyperemia”.
In making engorgement a distinct clinical entity, the authors (implicitly) acknowledge that engorgement cannot be explained by the theory that each breast’s extensive lactiferous duct systems experience a generalised narrowing with pathological biofilm. That is, ABM CP#36 does not consider engorgement part of the mastitis spectrum because it clearly doesn’t fit the pathogenic biofilm hypothesis. However, ABM CP#36 fails to offer pathophysiological mechanisms for the interstitial fluid and hyperaemia of engorgement.
Applying the mechanobiological model, I have argued that engorgement belongs on the spectrum of breast inflammation, subject to the same mechanobiological aetiological model as other clinical presentations of benign lactation-related breast inflammation.3
5. Clinical Protocol #36 represents various research studies in ways that I don't consider to be accurate.
ABM CP#36 represents a number of studies in ways which I contest, analysed in Table 1 below. As a result,
- Interventions are recommended which have been investigated but lack evidence of efficacy;
- Interventions are recommended which lack pathophysiological rationale; and
- Advice is given to avoid one recommendation (nipple shields) which has demonstrated positive effects for management of breastfeeding problems.
Table 1. Queries concerning representation of studies. (I have added underscoring to assist the reader.)
Quote from Clinical Protocol #36. Recommendations are rated by the SORT taxonomy.
“A Swedish study noted that most women with inflammatory mastitis had complete resolution of symptoms without need for antibiotics or other interventions. The authors attributed this finding to a focus on symptomatic control, appreciation of the physiological anti-inflammatory response, and regular communication between patient and clinician.[Kvist 2007]”
In Kvist et al 2007, midwives gave each woman who presented with breast inflammation across the three groups studied:
1. unspecified fit and hold advice;
2. advice about interval and duration of breastfeeds with the aim of decreasing intervals between feeds;
3. advice about emptying breast by manual expression; and
4. advice about pumping or warm shower to breast.29
“A systematic review concluded that although breast massage may reduce pain, it should not be recommended as standard of care because it requires extensive training to master atraumatic approach. The most successful technique approximates manual lymphatic drainage with light sweeping of the skin rather than deep tissue massage.[Anderson et al 2019; Witt et al 2016]”
The Anderson 2019 systematic review analyses studies which investigate a variety of traditional massage approaches and Therapeutic Breast Massage for Lactation for breastfeeding problems. Anderson et al conclude: “The overall effect of breast massage on reported outcomes is uncertain.” Gua Sha, from traditional Chinese medicine, is described by ABM CP#36 as “the most successful technique” reviewed in Anderson et al. However, Gua Sha scrapes lightly from the base of the breast towards the nipple with a specialised soft instrument, resulting in decreased pain at 5 and 30 minutes afterwards. This does not approximate Manual Lymphatic Drainage, which specifically sweeps in the other direction, towards the axillae. The Japanese Oketani massage is given early in lactation and weekly if breast inflammation or supply concerns arise. It stimulates the pectoral muscles combined with a rolling massage of the breast, and showed a preventative decrease in breast pain.30 Please see my article here for further discussion of ABM CP#36’s recommendation of lymphatic massage.
“Consider lymphatic drainage to alleviate interstitial edema.[Ezzo et al 2015]”
This advice cites Ezzo et al’s 2015 Cochrane review of studies which combined Manual Lymphatic Drainage (MLD) with compression bandaging for breast-cancer related lymphoedema in the upper limb after surgical axillary node dissection.31 Subsequent systematic reviews of efficacy of MLD in 2020 and 2021 show little benefit, and suggest that prolonged tissue compression alone may be the active ingredient.32-35 Recommending lymphatic drainage on the basis of Ezzo et al’s findings also misunderstands the radically different tissue environment of the lactating breast compared to limbs post-breast cancer surgery. Read an in depth analysis of ABM CP#36’s recommendation of lymphatic drainage here.
“Consider ice for symptomatic relief”
There is no evidence to support the application of compresses, either hot or cold. Ductal dilation is not influenced by warmth, unless warmth is used as part of nipple stimulation, which releases oxytocin. Warmth may increase stromal tension and duct compression by increasing blood flow. Cold application decreases ductal diameters in the nipple, risking decreased milk transfer.36
“Sunflower or soy lecithin 5-10 gm daily by mouth may be taken to reduce inflammation in ducts and emulsify milk.” (no citation)
This hypothesis lacks an explanatory mechanism. There is no reason to believe that ingestion of oral lecithin affects distribution of fat globules in human milk or reduces inflammation in the lactiferous ducts. In a 2003 study, lecithin was directly added to a test tube of milk from mothers of prematurely born infants, resulting in less loss of fat, because the milk fats were less likely to adhere to the collecting device. This study does not support the oral ingestion of lecithin to improve outcomes for breastfeeding women.37
“Therapeutic ultrasound (TUS) uses thermal energy to reduce inflammation and relieve edema. TUS may be an effective treatment for conditions arising in the mastitis spectrum.[Mogenson et al 2020]”
This recommendation cites an article by Mogenson et al which is a narrative review, not data in an evaluative study supporting the use of TUS.38 Diepeveen et al noted in 2019 that there is little empirical evidence to support the use of TUS in lactation-related breast inflammation though it is commonly applied by Australian physiotherapists.39 The most appropriate frequency to use is unknown. Research has only examined penetration depth in non-breast tissue. In 1991 McLachlan et al reported that TUS was no more effective than placebo for engorgement.40 In 2012, a restrospective study of 25 mothers found that 23 had resolution of plugged duct following TUS but this study had serious methodological weaknesses.41 The mechanisms by which TUS is proposed to ‘[use] thermal energy to reduce inflammation’ are not clarified.
“Consider probiotics”. This recommendation is rated Levels of evidence: 1-2. Strength of recommendation: B
I dispute that this recommendation is supported by evidence.42 A 2020 review by Barker et al identified five RCTs investigating probiotic consumption for treatment or prevention of mastitis concluded there is no reliable supporting evidence. Although each RCT reported a lower incidence of mastitis in the probiotic group, there were significant methodological limitations concerning baseline characteristics, study hypotheses, lack of power calculations, definitional issues, and potential conflicts of interest.43 As ABM CP#36 recognises, probiotics did not change human milk microbiome composition when 415 breastfeeding women were randomized to receive probiotics or placebo.44
“Avoid the use of nipple shields. Available evidence does not support the use of nipple shields. Neither safety nor effectiveness has been demonstrated. Nipple shields … result in inadequate breast milk extraction.[McKenchie & Eglash 2010]”. This recommendation is rated: Level of Evidence 3. Strength of recommendation C.
The 2010 McKechnie & Eglash review cited in the clinical protocol is outdated.45 Nipple shields are often used as compensation for failure to identify and address underlying problems of positional instability or conditioned dialing up at the breast, but remain an important adjunct intervention for nipple pain and damage. Both a 2015 systematic review and a 2021 review conclude that nipple shield use substantially benefits breastfeeding when problems emerge, both in measurable outcomes and in reports by mothers.36 46 In a study which randomized nipple shield use in a group of 20 mothers with nipple pain and 28 without breastfeeding problems, nipple shields improved maternal comfort and did not impact milk removal or sucking strength in the pain group.47
"It should be noted that ultrasound studies documenting a small number of orifices approaching the nipple reflect limitations of radiographic images as compared with histological anatomy.[Ramsay 2005]"
Are the authors claiming that histological studies, unreferenced, reveal more nipple duct orifices than have been demonstrated by in Ramsay et al’s ground-breaking ultrasound study? This appears to misrepresent the Ramsay et al study, which uses ultrasound to accurately identify main ducts. Ramsay et al demonstrated dense glandular and duct tissue within a 3 cm radius of the base of nipple, and on average 9 main ducts (range 4-18).48
6. Clinical Protocol #36 inaccurately extrapolates the medical condition ‘lymphedema’ to the unique tissues of the lactating breast.
Lymphoedema does not occur in the otherwise normal lactating breast. Lymphoedema is a secondary condition, predominantly of the limbs and occasionally the breast, which results after breast cancer surgery from removal of axillary nodes. Lymphoedema is fibrotic. The rare condition of primary lymphedema is not relevant to lactation-related breast inflammation. Interstitial fluid in the breast stroma resulting from breast inflammation is not the same as the medical conditions of lymphoedema or dependent oedema.
On page 363, ABM CP#36 states that the photograph of Figure 5 shows: “Day 5 postpartum breast engorgement showing edematous nipple areolar complex and dependent lymphedema with overlying erythema”. However, this woman’s swelling is not necessarily “edematous” due to large amounts of interstitial fluid. Most glandular tissue lies in a 3 cm radius of the base of nipple and the ducts will be dilated and tense when milk is not transferring out.48 Although there is likely to be some increased interstitial fluid and stromal tension, nipple areolar complex swelling is much more likely to result from high milk volumes, associated with high intraluminal pressures, in both glandular and ductal tissue.
On page 368, ABM CP#36 states: “Lactating breasts … require support to avoid dependent lymphedema”. Please see my previous comment on bra use and the risks of using sociocultural-specific perspectives as international clinical guidelines.
On page 371 in Figure 21 entitled ‘Lymphatic Drainage’, ABM CP#36 states that lymphatic drainage “reduces swelling by assisting movement of lymph fluid, decreasing edema, softening fibrosis”. But the inflamed lactating breast is not fibrotic.
7. Clinical Protocol #36 applies diagnoses which are poorly defined and not clinically meaningful, increasing risks of unnecessary antibiotic use
Introduction of the the poorly defined clinical diagnoses of ‘inflammatory mastitis’, ‘bacterial mastitis’, ‘phlegmon’, ‘subacute mastitis’, and ‘infected galactocele’ risks perpetuation of unnecessary antibiotic use, in an era when the World Health Organisation calls urgently for antimicrobial stewardship due to the potentially catastrophic consequences of antimicrobial resistances. I explore these unnecessary clinical diagnoses in my research publication Re-thinking benign inflammation of the lactating breast: classification, prevention and management),3 and summarise aspects in Table 2 below.
Table 2. Unnecessary diagnoses in Clinical Protocol #36 may increase the risk of antibiotic use
Quote from Clinical Protocol #36
“Inflammatory mastitis presents as an increasingly erythematous, edematous, and painful region of the breast with systemic signs and symptoms such as fever, chills, and tachycardia”
Mastitis means ‘inflammation of the breast’. Because all mastitis is inflammatory, this tautology does not offer a meaningful diagnosis.
“Bacterial mastitis represents a progression from ductal narrowing and inflammatory mastitis to an entity necessitating antibiotics or probiotics to resolve … Bacterial mastitis presents as cellulitis (worsening erythema and induration) in a specific region of the breast that may spread to different quadrants … An evaluation by a medical professional should be performed if there are persistent systemic symptoms (>24 hours) such as fever and tachycardia … In the absence of systemic signs and symptoms, diagnosis should be considered if the breast is not responding to conservative measures described … Reserve antibiotics for bacterial mastitis …”
The human milk microbiome is rich in bacteria. Any lactation-related breast inflammation will result in changes in bacterial composition of the milk microbiome and breast stroma, as the mammary immune system responds homeostatically. ABM CP#36 is unable to clarify signs and symptoms which indicate a progression from ‘inflammatory’ to ‘bacterial’ mastitis. Cellulitis is a bacterial skin infection but mastitis is an inflammatory condition of the breast stroma, with secondary skin inflammation. At the severe end of the spectrum of breast inflammation this may require antibiotics. Although ‘inflammatory mastitis’ is said to be associated with systemic signs, ABM CP#36 recommends a doctor’s assessment after 24 hours of systemic signs. This suggests that after 24 hours the inflammation may be ‘bacterial’, which requires antibiotics or probiotics. But Kvist showed that fevers were not linked to abscess formation, and that even with fever, most breast inflammations resolved with conservative measures, including fit and hold intervention and increased frequency of feeds.29
“Phlegmon should be suspected with a history of mastitis that worsens into a firm, mass-like area without fluctuance. ... Acute bacterial mastitis … can progress to phlegmon. Lactational phlegmon may require extended antibiotics for complete resolution”
Phegmon is a condition which is not diagnosable by patient examination, but on imaging. Phlegmon is typically identified when the clinician investigates a lump with ultrasound to exclude abscess in breast inflammation. The decision to follow up with further imaging should be made on clinical assessment of the presenting inflammatory lump post-imaging, only. No imaging followup is necessary if the clinical presentation is resolving. The diagnosis of phlegmon made on imaging should not alter interventions offered, as abscess has been excluded. The diagnosis made on imaging reflects the subjective judgement of the radiologist that there is enough fluid collecting to warrant the term, since there are no meaningful definitions. The finding of phlegmon on imaging occurs at the severe end of the inflammatory spectrum of interstitial fluid secretion, but does not indicate significant bacterial overgrowth requiring antibiotics. Antibiotic use remains a clinical decision. ABM CP#36’s use of the diagnosis of phlegmon in the lactating breast risks both unnecessary imaging and antibiotic use. The NDC classification of lactation-related breast inflammation does not use the term phlegmon for presentations of breast inflammation.
“Subacute mastitis occurs when ductal lumens become narrowed by bacterial biofilms in the setting of chronic mammary dysbiosis.”
Subacute mastitis is not a definable condition on patient presentation. Use of this term with associations of biofilm and mammary dysbiosis increases risk of unnecessary antibiotic use. Please see my research publication for analysis of studies which claim to define subacute mastitis.3
“A galactocele develops when ductal narrowing obstructs the flow of milk to the extent that a significant volume of obstructed milk collects in a cyst-like cavity … Galactoceles, which can result from unresolved hyperlactation, can become infected … An infected galactocele requires drainage as well as antibiotics.”
This pathophysiological theory of galactocoele development does not consider the alveolar rupture, associated apoptosis and resultant tissue destruction that occurs in the subclinical inflammatory development of galactocoele. An infected galactocoele is an abscess, so the diagnosis ‘infected galactocoele’ does not have clinical meaning. ABM CP#36 does not describe the pathophysiological mechanism by which galactocoeles are hypothesised to result from hyperlactation.
8. Clinical Protocol #36 is not able to reliably apply the 2004 Strength of Recommendation Taxonomy (SORT)
ABM CP#36 makes SORT findings by grouping studies which are highly heterogenous, investigating different conditions with a wide range of different measures. In many recommendations, ABM CP#36 acknowledges that recommendations are Level C, which includes based on “consensus, usual practice … and/or opinion”. Because SORT Strength of Recommendation is based “on a body of evidence (typically more than one study)” and the Level of Evidence is based on an assessment of study design or “quality of evidence from multiple studies about a specific question or the quality of evidence supporting a clinical intervention”, it is essential in the use of SORT that each SORT recommendation results from a comprehensive review of all existing evaluations of a specific intervention available in the research literature.49 I am concerned that has not consistently occurred in the preparation of ABM CP #36. Because of the high levels of subjectivity in ABM CP#36’s application of SORT, I propose that the rating of evidence in ABM CP#36 is not meaningful, and in my view risks misleading clinicians.
I give two examples to illustrate my concern.
- ABM CP#36’s recommendation to consider probiotic use is rated Levels of evidence: 1-2. Strength of recommendation: B. To reach this conclusion, ABM CP#36 cites the Barket et al scoping review and Crepinsek Cochrane review, along with a cautionary narrative analysis by Amir et al and narrative review of bovine and human milk microbiomes by Oikonomou et al.42 43 50 16 The Crepinsek et al review investigates the use of probiotics as prevention of mastitis after childbirth (finding very low certainty of evidence) - not as intervention for breast inflammation. The Barker et al review findings do not support ABM CP#36’s SORT recommendation that probiotic use should be considered in breast inflammation, as I detail above in Table 1.
- ABM CP#36’s recommendation to avoid nipple shields is rated Level of evidence: 3. Strength of recommendation: C, but considers just one 2010 study, failing to cite other more recent studies and a systematic review on this topic, as I’ve detailed in Table 1.
Clinical lactation support remains a research frontier,51 requiring best practice implementation science, which takes the steps of a) theoretical frames; b) clinical translation into education or clinical programs; c) iterative feedback through patient contact and pilot studies; d) layers of various evaluative studies, qualitative and quantitative. This is the process I’ve followed in development of the Neuroprotective Developmental Care programs for parents and health professionals, including in the foundational breastfeeding domain which interacts with and is at the heart of the other domains, and which includes the NDC approach to lactation-related breast inflammation.
If health professionals are to offer the best possible help to breastfeeding women and their babies, and short circuit the decades-long gap between evidence and practice, theoretical frameworks need to be appropriately developed, named, and debated. It is important that clinical guidelines don’t misleadingly represent theoretical models as fact, though in the absence of guiding evaluative evidence, a model needs to be translated into clinical recommendations. Implementation science also requires careful and accurate representation of and critical analysis of existing studies.52 New medical terms and diagnoses for clinical presentations in breastfeeding pairs should only be introduced cautiously, since overmedicalisation and overtreatment is increasing internationally,7 53 54 including in breastfeeding women and their babies.55-58
In my view, Clinical Protocol #36 raises important questions about the Academy of Breastfeeding Medicine’s process for developing new clinical protocols, including processes for attribution of sources, accurate representation of existing research, the usefulness of grading of evidence, and the development of a scientific consensus position concerning interpretation and clinical translation when evaluations are limited.
Dr Pamela Douglas
Medical Director, Possums & Co., and The Possums Clinic, Sherwood, Brisbane, Australia
Associate Professor Adjunct, School of Nursing and Midwifery, Griffith University, Australia
Senior Lecturer, General Practice Clinical Unit, The University of Queensland, Australia
* I acknowledge that not all people who breastfeed are women or of binary gender. I ask those reading this article to please translate my use of gender specific nouns and pronouns to suit your identity and experience, and am very grateful for your patience and generosity in doing so, in a world that still much too frequently fails to offer inclusivity for our great diversities as humans.
Conflict of interest
Dr Pamela Douglas is Medical Director of the charity Possums & Co., which sells education products in the programs Neuroprotective Developmental Care (or the Possums programs) to health professionals www.possumsonline.com and to parents milkandmoonbabies.com. All revenue raised is returned to education and research.
1. Mitchell KB, Johnson HM, Rodriguez JM, et al. Academy of Breastfeeding Medicine Clinical Protocol #36: The Mastitis Spectrum, Revised 2022. Breastfeeding Medicine 2022;17(5):360-75.
2. Douglas PS. Re-thinking benign inflammation of the lactating breast: a mechanobiological model. Women's Health 2022;18:https://doi.org/10.1177/17455065221075907.
3. Douglas PS. Re-thinking benign inflammation of the lactating breast: classification, prevention, and management. Women's Health 2022;18:doi: 10.1177/17455057221091349.
4. Douglas PS. Overdiagnosis and overtreatment of nipple and breast candidiasis: a review of the relationship between the diagnosis of mammary candidiasis and Candida albicans in breastfeeding women. Women's Health 2021;17:DOI: 10.1177/17455065211031480.
5. Douglas PS. Re-thinking lactation-related nipple pain and damage. Women's Health 2022;18:DOI: 10.1177/17455057221087865.
6. Armstrong N. Overdiagnosis and overtreatment: a sociological perspective on tackling a contemporary healthcare issue. Sociology of Health and Illness 2020;43(1):58-64.
7. Brownlee S, Chalkidou K, Doust J, et al. Evidence for overuse of medical services around the world. The Lancet 2017;390:156–68.
8. Coon ER, Quinonez RA, Moyer VA, et al. Overdiagnosis: how our compulsion for diagnosis may be harming children. Pediatrics 2014;134(5):1-11.
9. Treadwell J. Overdiagnosis and overtreatment: generalists - it's time for a grassroots revolution. Journal of General Practice 2016;66(644):116-17.
10. Cullinane M, Scofield L, Murray GL, et al. Random amplified polymorphic DNA analysis reveals no clear link bewteen Staphylococcus epidermidis and acute mastitis. Australian and New Zealand Journal of Obstetrics and Gynaecology 2022;62:605-09.
11. Rodriguez JM, Fernandez L. Infectious mastitis during lactation: a mammary dysbiosis model. In: McGuire M, Bode L, eds. Prebiotics and probiotics in human milk: Academic Press 2017:401-28.
12. Fernandez L, Arroyo R, Espinosa I, et al. Probiotics for human lactational mastitis. Beneficial Microbes 2014;5:169-83.
13. Boix-Amoros A, Collado MC, Land VtB, et al. Reviewing the evidence on breast milk composition and immunological outcomes. Nutrition Reviews 2019;77(8):541-56.
14. Fernandez L, Pannaraj PS, Rautava S, et al. The microbiota of the human mammary ecosystem. Frontiers in cellular and infection microbiology 2020;10:Article 5866667.
15. Ruiz L, Garcia-Carral C, Rodriguez JM. Unfolding the human milk microbiome landscape in the omics era. Frontiers in Microbiology 2019;10(1378):doi:10.3389/fmicb.2019.01378.
16. Oikonomou G, Addis MF, Chassard C. Milk microbiota: what are we exactly talking about? Frontiers in Microbiology 2020;11(60):doi:10.3389/fmicb.2020.00060.
17. Sakwinska O, Bosco N. Host microbe interactions in the lactating mammary gland. Frontiers in Microbiology 2019;10:doi:10.3389/fmicb.2019.01863.
18. Dominguez-Bello M, Godoy-Vitorino F, Knight R, et al. Role of the microbiome in human development. Gut 2019;68:1108-14.
19. Dinleyici M, Perez-Brocal V, Arslanoglu S, et al. Human milk mycobiota composition: relationship with gestational age, delivery mode, and birth weight. Beneficial Microbes 2020;11(2):doi:10.3910/BM2019.0158.
20. Moossavi S, Fehr K, Derakhshani H, et al. Human milk fungi: environmental determinants and inter-kingdom associations with milk bacteria in the CHILD Cohort Study. BMC Microbiology 2020;20:146.
21. Kvist L, Larsson BW, Hall-Lord ML, et al. The role of bacteria in lactational mastitis and some considerations of the use of antibiotic treatment. International Breastfeeding Journal 2008;3(6):doi:10.1186/746-4358-3-6.
22. Cullinane M, Amir LH, Donath SM, et al. Determinants of mastitis in women in the CASTLE study: a cohort study. BMC Family Practice 2015;16:181.
23. Kent JC, Mitoulas LR, Cregan MD, et al. Volume and frequency of breastfeedings and fat content of breast milk throughout the day. Pediatics 2006;117(3):e387-95.
24. Odom E, Scanlon K, Perrine C, et al. Reasons for earlier than desired cessation of breastfeeding. Pediatics 2013;131:e726-32.
25. Labiner-Wolfe J, Fein SB, Shealy KR, et al. Prevalence of breast milk expression and associated factors. Pediatrics 2008;122 Supp 2 S63-68.
26. Johns H, Forster DA, Amir LH, et al. Prevalence and outcomes of breast milk expressing in women with healthy term infants: a systematic review. BMC Pregnancy and Childbirth 2013;13:212.
27. Douglas PS, Hill PS. A neurobiological model for cry-fuss problems in the first three to four months of life. Med Hypotheses 2013;81:816-22.
28. Douglas PS, Hill PS, Brodribb W. The unsettled baby: how complexity science helps. Arch Dis Child 2011;96:793-97.
29. Kvist LJ, Halll-Lord ML, Larsson BW. A descriptive study of Swedish women with symptoms of breast inflammation during lactation and their perceptions of the quality of care given at a breastfeeding clinic. International Breastfeeding Journal 2007;2(2):doi:10.1186/746-4358-2-2.
30. Anderson L, Kynoch K, Kildea S, et al. Effectiveness of breast massage for the treatment of women with breastfeeding problems: a systematic review. JBI Database Systematic Reviews Implement Rep 2019;17(8):1668-94.
31. Ezzo J, Manheimer E, McNeely ML. Manual lymphatic drainage for lymphedema following breast cancer treatment. Cochrane Database of Systematic Reviews 2015;5:DOI:10.1002/14651858.CD003475.pub2.
32. Thompson B, Gaitatzis K, De Jonge XJ, et al. Manual lymphatic drainage treatment for lympedema: a systematic review of the literature. Journal of Cancer Survivorship 2021;15:244-58.
33. Abouelazayem M, Elkorety M, Monib S. Breast lymphedema after conservative breast surgery: an up-to-date systematic review. Clinical Breast Cancer 2021;21(3):156-61.
34. Liang M, Chen Q, Peng K. Manual lymphatic drainage for lymphedema in patients after breast cancer surgery. Medicine 2020;99(49):e23192.
35. Brandao ML, Soares HPS, A AM. Efficacy of complex decongestive therapy for lymphedema of the lower limbs: a systematic review. Jornal Vascular Brasileiro 2020;19:e20190074.
36. Geddes DT, Gridneva Z, Perrella SL, et al. 25 years of research in human lactation: from discovery to translation. Nutrients 2021;13:1307.
37. Chan MM, Nohara M, Chan BR, et al. Lecithin decreaes human milk fat loss during enteral pumping. Journal of Pediatric Gastroenterology and Nutrition 2003;36(5):613-15.
38. Mogensen N, Portman A, Mitchell K. Nonpharmacologic approaches to pain, engorgement, and plugging in lactation. Clinical Lactation 2020;11(1):http://dx.doi.org/10.1891/2158-07184.108.40.206.
39. Diepeveen LC, Fraser E, Croft AJ. Regional and facility differences in interventions for mastitis by Australian physiotherapists. Journal of Human Lactation 2019;35(4):695-705.
40. McLachlan Z, Milne EF, Lumley J, et al. Ultrasound treatment for breast engorgement: a randomised double blind trial. Australian Journal of Physiotherapy 1991;37(1):23-28.
41. Lavigne V, Glebezon BJ. Ultrasound as a treatment of mammary blocked duct among 25 postpartum lactating women: a restrospective case series. Journal of Chiropractic Medicine 2012;11(3):170-78.
42. Amir LH, Griffin L, Cullinane M, et al. Probiotics and mastitis: evidence-based marketing? International Breastfeeding Journal 2016;111(19):doi:10.1186/s13006-016-0078-5.
43. Barker M, Adelson P, Peters MDJ, et al. Probiotics and human lactational mastitis: a scoping review. Women and Birth 2020;d33:e483-e91.
44. Simpson MR, Avershina E, Sstorre O, et al. Breastfeeding-associated microbiota in human milk following supplementation with Lactobacillus rhamnosus GG, Lactobacillus acidophilus La-5, and Bifidobacterium animalis ssp. lactis Bb-12. Journal of Diary Science 2018;101(2):889-99.
45. Mckechnie AC, Eglash A. Nipple shields: a review of the literature. Breastfeeding Medicine 2010;5(309-314)
46. Chow S, Chow R, Popovic M, et al. The use of nipple shields: a review. Frontiers in Public Health 2015;3:doi:10.3389/fpubh.2015.00236.
47. Coentro VS, Perrella SL, Lai CT, et al. Nipple shield use does not impact sucking dynamics in breastfeeding infants of mothers with nipple pain. European Journal of Pediatrics 2021;180:1537-43.
48. Ramsay DT, Kent JC, Hartmann RA, et al. Anatomy of the lactating human breast redefined with ultrasound imaging. Journal of Anatomy 2005;206:525-34.
49. Ebell MH, Siwek J, Weiss BD, et al. Strength of Recommendation Taxonomy (SORT): a patient-centred approach to grading evidence in the medical literature. American Family Physician 2004;69:548-56.
50. Crepinsek MA, Taylor EA, Michener K, et al. Interventions for preventing mastitis after childbirth (Review). Cochrane Database of Systematic Reviews 2020(9):Doi:10.1002/14651858.CD007239.pub4.
51. Stuebe AM. We need patient-centred research in breastfeeding medicine. Breastfeeding Medicine 2021;16(4):349-50.
52. Wensing M, Grol R. Knowledge translation in health: how implementation science could contribute more. BMC Medicine 2019;17(88):https://doi.org/10.1186/s12916-019-1322-9.
53. Hoffman T, Del Mar C. Patients' expectations of the benefits and harms of treatments, screening and tests - a systematic review. JAMA Internal Medicine 2015;175(2):274-86.
54. Hoffman T, Del Mar C. Clinicians' expectations of the benefits and harms of treatments, screening, and tests - a systematic review. JAMA Internal Medicine 2017;177(3):407-19.
55. Kapoor V, Douglas PS, Hill PS, et al. Frenotomy for tongue-tie in Australian children (2006-2016): an increasing problem. MJA 2018;208(2):88-89.
56. Wei E, Tunkel D, Boss E, et al. Ankyloglossia: update on trends in diagnosis and management in the United States, 2012-2016. Otolaryngology - Head and Neck Surgery 2020:https://doi.org/10.1177%2F0194599820925415.
57. Lisonek M, Shiliang L, Dzakpasu S, et al. Changes in the incidence and surgical treatment of ankyloglossia in Canada. Paedaitrics and Child Health 2017;22(7):382-86.
58. Ellehauge E, Schmidt Jensen J, Gronhoj C, et al. Trends of ankyloglossia and lingual frenotomy in hospital settings among children in Denmark. Danish Medical Journal 2020;67(5):A01200051.