See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/286373126

Therapeutic Breast Massage in Lactation for the Management of Engorgement,

Plugged Ducts, and Mastitis

Article  in  Journal of Human Lactation · December 2015

DOI: 10.1177/0890334415619439

CITATIONS READS

12 1,399

4 authors, including:

Maya Bolman
Northeast Ohio Medical University
3 PUBLICATIONS   24 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Maya Bolman on 29 November 2017.

The user has requested enhancement of the downloaded file.

619439
research-article2015
JHLXXX10.1177/0890334415619439Journal of Human LactationWitt et al

Original Research
Journal of Human Lactation

Therapeutic Breast Massage in Lactation

1­–9
© The Author(s) 2015
Reprints and permissions:
for the Management of Engorgement, sagepub.com/journalsPermissions.nav
DOI: 10.1177/0890334415619439

Plugged Ducts, and Mastitis jhl.sagepub.com

Ann M. Witt, MD, IBCLC1,2,3, Maya Bolman, BA, BSN, IBCLC1,2,

Sheila Kredit3, and Anne Vanic, MSN, CPNP, IBCLC1,2

Abstract
Background: Many women in developed countries do not meet their breastfeeding goals and wean early because of breast
pain.
Objective: This study aimed to describe clinical response to therapeutic breast massage in lactation (TBML) in the
management of engorgement, plugged ducts, and mastitis.
Methods: Breastfeeding women presenting with engorgement, plugged ducts, or mastitis who received TBML as part of
their treatment were enrolled (n = 42). Data collected at the initial visit included demographic, history, and exam data
pre-TBML and post-TBML. Email surveys sent 2 days, 2 weeks, and 12 weeks following the initial visit assessed pain and
breastfeeding complications. A nested case control of engorged mothers (n = 73) was separately enrolled to compare
engorgement severity.
Results: Reasons for the visit included engorgement (36%), plugged ducts (67%), and mastitis (29%). Cases, compared to
controls, were significantly more likely to have severe engorgement (47% vs 7%, P < .001). Initial mean breast pain level
among those receiving TBML was 6.4 out of 10. Following TBML, there was significant improvement in both breast (6.4
vs 2.8, P < .001) and nipple pain (4.6 vs 2.8, P = .013). All women reported immediate improvement in their pain level. At
the 12-week survey, 65% found the massage treatment very helpful. The majority of the women with a new episode of
mastitis or plugged duct during the study follow-up found the techniques learned during the office visit very helpful for home
management of these episodes.
Conclusion: In office, TBML is helpful for the reduction of acute breast pain associated with milk stasis. Mothers find TBML
helpful both immediately in-office and for home management of future episodes.

Keywords
blocked ducts, breast engorgement, breastfeeding, breastfeeding experience, breastfeeding practices, breastfeeding support,
breast pain, expression, mastitis

Well Established Background

Breast pain is a major cause of weaning. The likelihood of wean- Despite being recognized as the biological norm of infant
ing increases the longer pain persists, and measures to resolve nutrition,1 many women in developed countries do not meet
pain quickly are important. Engorgement, plugged ducts, and their breastfeeding goals.2-4 US national data demonstrate
mastitis are commonly associated with acute breast pain. that women less than 1 month postpartum commonly cite
Emptying the breast is an important step in treatment algo-
rithms for these conditions, and manual expression and mas- 1
Breastfeeding Medicine of Northeast Ohio, Cleveland, OH, USA
sage techniques can help. 2
Senders Pediatrics, Cleveland, OH, USA
3
Case Western Reserve University, Cleveland, OH, USA

Date submitted: November 20, 2014; Date accepted: November 4, 2015.

Newly Expressed
Corresponding Author:
In-office, therapeutic breast massage in lactation provides an Ann M. Witt, MD, IBCLC, Breastfeeding Medicine of Northeast Ohio,
immediate, clinically significant reduction in acute breast pain 2054 South Green Road, South Euclid, OH 44121, USA.
associated with engorgement, plugged ducts, and mastitis. Email: awitt@bfmedneo.com

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
2 Journal of Human Lactation 

pain as a reason for weaning; 29% reported that “breastfeed- plugged ducts, and mastitis and was conducted from June
ing was too painful,” and 24% noted that “my breasts were 2013 through March 2014. Inclusion criteria were breast-
overfull or engorged.”2,5-7 Given that the likelihood of wean- feeding women (1) 18 years or older presenting to
ing prematurely increases the longer pain persists,6 the need BFMEDNEO with a history of acute breast pain, (2) diag-
for interventions to resolve pain quickly remains important. nosed with engorgement, plugged duct, or mastitis at the
Engorgement, plugged ducts, and mastitis are common visit, and (3) treated during the office visit with TBML.
causes of pain that may lead to the temporary or permanent Diagnosis of engorgement, plugged duct, and mastitis was
cessation of breastfeeding.8 According to the World Health made according to standard criteria,9,17 following history and
Organization, although engorgement, plugged ducts, and physical exam by the treating provider. Women were
mastitis are distinct entities, their pathogenesis involves the excluded if there was clinical evidence of abscess confirmed
common denominator of milk stasis.9 There is consensus in by follow-up ultrasound.
the literature that breast emptying has a role in the manage- Arm 2 provided a control comparison to the engorged
ment of all 3 conditions.9,10 mothers treated with TBML. It was conducted at the general
Some studies suggest that manual massage and expres- pediatric practice from December 2013 through January
sion techniques may help empty the breast, reduce pain, and 2014. Inclusion criteria were all breastfeeding mothers 18
assist in symptom resolution. Storr11 studied 25 mothers and years or older presenting at the initial newborn pediatric
found that self-massaging 1 breast prefeeding decreased later office visit. Mothers were excluded from the control group if
engorgement symptoms in the same breast. In an observa- they received TBML.
tional study, Zhao et al12 reported promising results regard- All cases were recruited by the treating practitioner. All
ing a 6-step manual technique for plugged duct treatment. participants gave written informed consent. The research
Although no studies have specifically examined the effect of study was approved by the University Hospitals of Cleveland
manual practices for mastitis, Thomsen et al13 showed that Institutional Review Board.
emptying the breast decreased the duration of mastitis and All women receiving in-office TBML completed a patient
improved treatment outcomes. These and other studies12,14-16 questionnaire at the initial visit for demographic information
support hands-on techniques for the treatment of milk stasis and clinical history. History, exam, and treatment data were
and its associated acute breast pain. However, further collected. Pain, engorgement levels, home treatments, feed-
research is needed to quantify symptom relief and the effec- ing patterns, and breastfeeding complications were collected
tiveness of particular milk removal techniques. via email questionnaires administered at 2 days, 2 weeks,
This study was undertaken to evaluate clinical symptoms and 12 weeks. All data were collected and managed using the
following in-office therapeutic breast massage in lactation REDCap electronic data capture tools hosted at University
(TBML) for the treatment of engorgement, plugged ducts, Hospitals of Cleveland.18 All women in arm 1 received treat-
and mastitis. Quantifying clinical response will provide ment with TBML from trained clinicians. Descriptions of
information for future randomized controlled trials and eval- these techniques have been outlined earlier.18,19 In brief, they
uate the role of massage and hand expression in treatment embrace the principles of
regimens. We hypothesized that TBML would provide
immediate symptom improvement, result in a clinically sig- 1. Focused gentle massage toward the axillae.
nificant decrease in acute breast pain, and that mothers would 2. Alternating gentle massage and hand expression.
find learning techniques at the visit helpful for future epi-
sodes at home. The massage is continually adjusted to the patient’s comfort
level. Details of the TBML techniques are outlined in
Appendix A (available online). The visit was not limited to
Methods TBML but was combined with a full consult by an
Study Population International Board Certified Lactation Consultant/regis-
tered nurse and/or breastfeeding medicine physician. During
This study was conducted at Breastfeeding Medicine of the visit, mothers received basic breastfeeding support (ie,
Northeast Ohio (BFMEDNEO), a referral practice for moth- latch correction, feeding patterns, milk supply assessment,
ers in Northeast Ohio needing specialized medical evalua- engorgement education) as clinically indicated. They also
tion for breastfeeding difficulties. This practice is located at received instruction on hand expression and gentle massage
a private suburban pediatric practice in Cleveland, Ohio, techniques. If clinically indicated following medical evalua-
viewed by the community as breastfeeding friendly. tion, antibiotic prescription, removal of white spot or nipple
bleb with a sterile needle,20 and breast milk cultures were
performed and recorded. For mothers diagnosed with masti-
Study Design tis, antibiotics were prescribed along with other standard
This prospective study consisted of 2 arms. Arm 1 was protocol measures including milk removal, supportive
designed to assess the effect of TBML on engorgement, measures, and analgesia as recommended.10

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
Witt et al 3

In arm 2, history, exam, and survey data were collected as Assessment of Mothers’ Perception of In-Office
in arm 1. Also like arm 1, this routine visit was integrated Therapeutic Treatment
with a lactation consultation21 that provided basic breast-
feeding support, hand expression, and massage education. At the end of the 2-day and 12-week survey, mothers were
asked, “Do you feel the massage in the office was helpful?”
If helpful, they were asked, “What was helpful about the visit
Descriptive Measures and the massage?” At the 2-week and 12-week survey, they
Descriptive variables including history of breastfeeding were asked, “Were the treatments learned during the visit
problems, prior treatment history, current breastfeeding- helpful for future episodes of mastitis or plugged duct?”
related diagnoses, breast milk bacterial species growth, and Mothers categorically responded (not helpful, somewhat
12-week data for weaning and breastfeeding complications helpful, helpful, very helpful) and were given additional
were recorded as previously described.22 Mothers presenting space to respond descriptively.
with engorgement were asked additional categorical ques-
tions on previous hand expression and massage instruction
Analyses
as well as hospital birthing practices, such as pacifier use,
formula supplementation, and rooming-in. All mothers were Once collected, the data were exported from REDCap to
asked to respond to the question, “What remedies have you SPSS software (SPSS, Inc, Chicago, Illinois, USA) and
tried at home before the initial visit?” Categorical answers analyzed under the supervision of the project investigator.
included nothing, feeding the baby more frequently, reverse Descriptive statistical analyses were performed to examine
pressure softening, hand expressing, pumping, massage, cool the distribution and normality of data. The main analyses
and warm compresses, and lecithin. tested clinical response to TBML for pain level, engorge-
ment severity, and plugged duct size. To further identify
differences in treatment responses, we analyzed the group
Outcome Measures
as a whole and then categorized 2 treatment groups:
Outcome variables included breast and nipple pain, engorge- engorgement (n = 15) and mastitis/plugged duct (MPD) (n
ment severity for the engorged mothers, and plugged duct = 27). Mastitis and plugged duct were initially evaluated as
severity for the mothers with plugged ducts. 1 subgroup since 2 patients were diagnosed with both mas-
Breast and nipple pain severity in the past 24 hours was titis and plugged duct. A further subanalysis was performed
rated on a numerical rating scale from 0 to 10, with 0 indi- separating mastitis and plugged duct treatment response.
cating pain free and 10 indicating the most severe pain.19,23 Those 2 patients diagnosed with both mastitis and plugged
Pain was rated by the patient at the initial visit premassage duct were excluded from this subanalysis to more clearly
and postmassage treatment and in 2-day and 12-week email differentiate treatment response for plugged duct and
surveys in response to the questions, “How bad has your mastitis.
breast pain been in the past 24 hours?” and “How bad has Our primary analysis was to assess pain, engorgement,
your nipple pain been in the past 24 hours?” Pain was also and plugged duct severity pre-TBML and post-TBML for the
rated during examination with manual expression of breast total sample. Categorical variables were described with fre-
milk and nipple tenderness to light touch premassage and quency and percentages and compared between engorge-
postmassage. ment and MPD groups using Pearson chi-square tests. Fisher
Maternal response to massage treatment was recorded exact test was used when the expected count was less than 5.
categorically following in-office treatment. Mothers were Continuous variables were described as means and standard
asked, “How has pain changed?” (same, more pain, less deviations (SDs) or median and range as appropriate. Paired
pain) and the percentage they felt their symptoms had t test was used to compare pain, engorgement, and plugged
improved (none, 10%, 25%, 50%, 75%, resolved). duct severity pretreatment and posttreatment, both for the
Engorgement severity was rated on Humenick et al’s24 entire group and for the subanalysis of the mastitis and
engorgement scale: (1) soft, no change; (2) slight change; (3) plugged duct groups. Power calculations were performed
firm, nontender; (4) firm, beginning tenderness; (5) firm, using Stata version 11.1 with the command SAMPSI. Power
tender; or (6) very firm and very tender. Engorgement sever- analysis revealed that in order to observe our difference in
ity was recorded for right and left breast separately, premas- mean pain levels premassage and postmassage treatment
sage and postmassage treatment, and at the 2-day survey. (β = 0.80 and α = 0.05), a sample size of 25 patients was
Plugged duct severity was rated as (0) none; (1) less than needed.
3 cm; (2) greater than 3 but less than 5 cm; (3) greater than A secondary case-control analysis compared engorged
or equal to 5 cm; or (4) multiple plugged areas. Plugged mothers treated with TBML with the control group. The
duct severity was recorded premassage and postmassage groups were compared in terms of sample characteristics,
treatment. delivery information, hospital experience, feeding patterns,

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
4 Journal of Human Lactation 

Table 1.  Sample Characteristics and Home Remedies. Initial Pain Levels, Engorgement, and Plugged
Total (N = 42), Duct Severity
Sample Characteristic No. (%)
To test our primary hypothesis that in-office TBML results in
Maternal age, median (range), y 32 (23-43) a significant decrease in acute breast pain, we assessed base-
Infant age, median (range), wk 5 (0-148) line breast and nipple pain in the 24 hours preceding the office
Private insurance 34 (85) visit, pain duration, and pain on examination (Table 2). Initial
College graduate 33 (79) mean (SD) breast pain for the entire sample was 6.43 (2.5).
Caucasian 31 (78) Median duration of pain before office visit was 24 to 48 hours
Returning to work 26 (62) for engorged mothers and 72 to 96 hours for MPD (P = .013).
Multiparous 23 (55) On examination, mothers with engorgement compared to
Exclusive breastfeeding 30 (71) MPD had higher levels of both breast tenderness (8.27 vs 6.5,
Breastfeeding goal > 12 months 27 (66) P = .041) and nipple tenderness (6.93 vs 4.3, P = .021).
Vaginal delivery 27 (64) Ninety-three percent of engorged mothers presented with
Previously taught hand expression 33 (79)
periareolar swelling. Engorgement severity was recorded on
Previously taught breast massage 22 (52)
both breasts. Each breast was considered independently for a
Remedy Tried at Home Before Initial Visit final sample of 30. On the Humenick engorgement scale,
53% were very firm, very tender. Initial evaluation of plugged
Nothing 2 (5) duct severity revealed that the swollen area measured greater
Feeding the baby more frequently 12 (29) than 5 cm for 36% of cases (Table 2).
Reverse pressure softening 3 (7)
Hand expressing 18 (43) Treatment response. We assessed clinical response to in-
Pumping 26 (62) office TBML treatment in terms of pain, engorgement, and
Massage 26 (62) plugged duct severity (Table 2).
Cool compresses 15 (36)
Warm compresses 29 (69) Pain outcome.  Breast and nipple pain significantly decreased
Lecithin 4 (10) after hands-on treatment with a mean (SD) decrease of 3.9
(2.4) for breast pain and 2.1 (3) for nipple pain (P < .01). A
subanalysis of mothers with mastitis and plugged duct
initial exam, pain, and engorgement severity at the 2-day sur-
revealed that both the mastitis and plugged duct groups inde-
vey, and breastfeeding rates and complications at the 12-week
pendently reported a significant (P < .001) decrease in mater-
survey.
nal breast pain and breast tenderness on exam following
treatment (Appendix B).
Results
Engorgement outcome.  The number of mothers with periareo-
In our primary analysis, we tested the hypothesis that in- lar swelling on exam significantly decreased (93% vs 7%,
office TBML results in a significant decrease in acute breast P < .001) following TBML. Engorgement severity was
pain. We quantitatively measured pain by asking mothers assessed on the Humenick24 6-point scale. Pretreatment mean
their impression of pain level before and after in-office mas- engorgement severity was 5.31 (between firm, tender and
sage. We also tested the hypothesis that engorgement and very firm, very tender). Posttreatment mean engorgement
plugged duct severity would decrease. severity was 3.48 (between firm, nontender and firm, begin-
During the 10-month study period, 43 mothers received ning tenderness). Mean (SD) engorgement severity signifi-
TBML for acute breast pain associated with milk stasis. cantly changed by 1.82 (1.6) levels (P < .01). When looking
One was excluded from the analysis because of abscess at specific levels of engorgement, significantly fewer mothers
diagnosed upon ultrasound evaluation. We analyzed 42 reported very firm, very tender breasts posttreatment (0%) as
mothers diagnosed with engorgement (36%), plugged compared to pretreatment (53%) (P < .01).
ducts (67%), and mastitis (29%) (Table 1). Median mater-
nal age was 32 years and median infant age was 5 weeks. Plugged duct outcome. Plugged duct severity significantly
Both the engorgement and MPD groups had similar base- decreased following treatment. Only 7% of mothers had a
line characteristics with regard to maternal age, parity, plugged duct measuring greater than 3 cm posttreatment as
breastfeeding intent, return to work, education, insurance, compared to before massage (68% vs 7%), and for 57% of
and previous instruction on hand expression and massage. mothers, the plugged ducts were resolved (P < .001).
Mothers presenting with MPD were significantly more
likely to be Caucasian (57% vs 89%, P = .044) and have
Case Control Engorgement Severity
older infants (0.7 vs 10 weeks, P < .001). Most mothers
had tried home remedies, as stated above, before present- To better understand the sample of breastfeeding mothers
ing at the initial visit (Table 1). who received in-office treatment, we performed a nested

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
Witt et al 5

Table 2.  Treatment Response to Therapeutic Breast Massage in Lactation.

Premassage Postmassage

Pain (n = 42) Mean (SD) Mean (SD) Change P Valuea

Breast pain level (maternal report) 6.4 (2.5) 2.8 (1.5) 3.6 (2.2) < .001b
Nipple pain level (maternal report) 4.6 (3.9) 2.8 (2.4) 1.8 (3.2) .001b
Breast tenderness (on exam) 7.14 (2.7) 3.26 (1.8) 3.9 (2.4) < .001b
Nipple tenderness (on exam) 5.2 (3.7) 3.1 (2.4) 2.1 (3) < .001b

Engorgement (n = 30) No. (%)  

1. Soft, no change 0 (0) 1 (2) 1.00
2. Slight change 0 (0) 3 (10) .237
3. Firm, nontender 2 (7) 5 (17) .424
4. Firm, beginning tenderness 3 (18) 14 (47) .002
5. Firm, tender 11 (37) 5 (17) .08
6. Very firm and very tender 16 (53) 0 (0) < .001
Engorgement severity, mean (SD) 5.31 (1.2) 3.48 (1.2) 1.82 (2.6) < .001b

Plugged Duct (n = 28) No. (%)  

Resolved/gone 0 (0) 16 (57) < .001
1. 1-3 cm 9 (32) 10 (36) .778
2. > 3 but < 5 cm 5 (18) 2 (7) .422
3. > 5 cm 11 (39) 0 (0) < .001
Plugged duct level, mean (SD) 2.29 (1) 0.5 (.3) 1.8 (1.1) < .001b
a
Pearson chi-square for dichotomous variables unless noted by superscripted b for paired t test.

case-control study on engorged mothers. During this arm, Eight mothers (17%) had a bleb unroofed. Fourteen moth-
95 mothers were screened, 20 declined, and 2 were ers (33%) were prescribed oral antibiotics; however, at the
excluded because they received TBML. Seventy-three 2-day survey, only 8 mothers had taken the antibiotics. Two
were enrolled for the final analysis. We compared the mothers had ultrasounds, 1 revealing no mass and the second
mothers treated in-office with TBML for breast engorge- a benign solid mass. Seventeen mothers (40%) had a breast
ment with the control group. There was no difference milk culture sent, with 5 (30%) of the cultures growing
between case and control for baseline characteristics, Staphylococcus aureus.
delivery, or hospital information. However, cases were less
likely to drain 1 side before switching to the second side
Follow-Up
(67% vs 89%, P = .038) (Table 3). There was no difference
between groups for exclusive breastfeeding (67% vs 88%, Thirty-eight women completed the 2-day survey. Ninety-
P = .06), although the lack of difference could represent a two percent reported pain improvement and 43% reported
type II error given the low power (0.40). Cases were less pain resolution. Forty-one women completed the final sur-
likely to be exclusively direct breastfeeding (40% vs 82%, vey. At 12 weeks, 65% of mothers were exclusively breast-
P = .002) and were more likely to cite pain as the reason feeding as compared to 71% at the initial visit, and 5 (14%)
for not doing so (27% vs 4%, P = .015). Those mothers had weaned. Two weaned because of low milk supply, 2
treated in-office for engorgement were significantly more planned on weaning, and 1 weaned because of both low
likely to have very severe engorgement (47% vs 7%, milk supply and continued pain. At 12 weeks, 6 women
P < .001), cracked nipples (73% vs 37%, P = .01), and reported a new episode of mastitis (17%) and 10 women
more severe nipple tenderness (6.9 vs 3.2, P < .001) and reported a new episode of plugged duct (27%) occurring at
breast tenderness on exam (8.3/10 vs 2.1/10, P < .001). some point during the 12-week follow-up. One mother,
originally seen for engorgement, reported treatment for an
abscess.
Massage and In-Office Treatment Follow-up between the case and control engorgement
Median length of massage time was 30 minutes with a range groups was similar. Whereas the TBML treatment group pre-
of 15 to 60 minutes with no significant difference between sented with significantly more severe engorgement at the ini-
MPD and engorgement groups (P = .137). tial visit, by the 2-day survey, there was no difference in pain

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
6 Journal of Human Lactation 

Table 3.  Engorgement Case Control Comparisons.

Sample Characteristic Case (n = 15), No. (%) Control (n = 73), No. (%) P Valuea
Maternal age, years (range) 33 (23-40) 31 (23-42) .571
Infant age, weeks (range) 0.71 (0.43-4.86) 0.57 (0.43-1.43) .13
Private insurance 13 (93) 64 (88) 1.00
College graduate 11 (73) 63 (86) .247
Caucasian 8 (57) 59 (81) .080
Working mother 11 (73) 56 (77) .780
Multiparous 11 (73) 36 (49) .089
Breastfeeding goal > 12 months 7 (47) 44 (61) .301
Delivery information
  Vaginal delivery 12 (80) 60 (80) 1.00
 Natural 3 (25) 11 (18) .687
  Intravenous fluids during delivery 12 (80) 68 (93) .126
Hospital experience
 Pacifier 6 (40) 23 (32) .552
  Room-in with infant 14 (93) 68 (94) .866
  Hand expression taught 10 (67) 44 (60) . 643
  Formula supplement 4 (27) 10 (14) .271
Feeding patterns
  Exclusive breastfeeding 10 (67) 64 (88) .06
  Direct breastfeeding exclusively 6 (40) 60 (82) .002
  Drain 1 side first 8 (67) 61 (88) .040
  Offer both 10 (77) 54 (81) .72
 Alternate 10 (83) 65 (97) .105
Initial exam
  Engorgement very severe 7 (47) 4 (7) < .001
  Breast tenderness on manual 8.3 (2.7) 2.3 (3) < .001
expression, mean (SD)
  Cracked nipples 11 (73) 27 (37) .01
  Periareolar swelling 14 (93) 26 (36) .01

2-Day Outcome Case (n = 15) Control (n = 56)  

Engorgement very severe 1 (8) 3 (5) .563
Breast pain 3.0 (2.6) 2.6 (2.4) .61
Nipple pain 2.7 (2.8) 3.7 (2.4) .183
Pain-free breasts 5 (39) 24 (36) 1.00

12-Week Outcome Case (n = 13) Control (n = 64)  

Pain-free breasts 11 (85) 58 (92) .343
Current exclusive breastfeeding 7 (58) 44 (67) .536
Weaned 3 (25) 7 (10) .168
Breastfeeding complications
 Mastitis 3 (21) 4 (6) .15
  Plugged duct 3 (25) 12 (19) .70
 Abscess 0 3 (5) .07
a
Pearson chi-square used except in cases when cell count less than expected and then Fisher exact t test used.

or engorgement severity between groups. Furthermore, at 12 outcome was only 0.295). To further check whether our non-
weeks, there was no difference in pain, exclusive breastfeed- significant results were due to a lack of statistical power, we
ing, or breastfeeding complications (Table 3). However, the conducted post hoc power analyses (with β = 0.80 and α =
possibility of a type II error remains given the low power to 0.05, two-tailed), which indicated that sample size would
detect differences at the 12-week follow-up due to a small have to increase to 415 (n = 83 for case and n = 332 for con-
sample size (eg, the power to detect a difference in mastitis trol) to be powered to evaluate 12-week outcomes.

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
Witt et al 7

Table 4.  Mothers’ Impressions on Therapeutic Breast Massage in Lactation.

Question Impression
What was helpful about the visit and the massage? “Pain relief was amazing.”
  “The immediate relief and learning the techniques.”
  “It was great to learn how to proceed to do it and how useful it is.”
  “Massage and resolving mastitis helped me meet my goal of exclusive
breastfeeding. I would not have succeeded without this help.”
  “Massaging your own breast while nursing can sometimes be
challenging. Having a professional assist was more effective.”
Were the treatments learned during the visit helpful “I felt like I was getting a plugged duct in my right breast a week ago
for future episodes of mastitis or plugged duct? and I did the massaging and it cleared up!”
  “I learned how to do it, and when I had a new episode of plugged
duct, I was able to resolve it by myself quickly using hand
expression and breast massage techniques that I learned.”

To further evaluate potential benefits from in-office treat- and plugged ducts. To our knowledge, this is the first study
ment and whether the techniques could be taught for use at to quantitatively examine clinical response to therapeutic
home, we asked mothers in the 12-week follow-up about what breast massage along with mothers’ perception of care.
home treatments they tried if they developed a subsequent Our study shows that TBML provides an important addi-
plugged duct or mastitis and if massage and hand expression tional treatment option for the clinician facing a case of
were helpful. Among those mothers experiencing an episode engorgement. It is common for clinicians to recommend
of plugged duct or mastitis, all found the techniques helpful feeding more frequently and applying warm or cold com-
for home treatment, with 60% of mothers with mastitis and presses and cabbage leaves for the treatment of engorge-
80% with plugged ducts finding them “very helpful.” ment.8,25 In our study, many women had already tried home
treatments and continued to have severe symptoms.
Therapeutic breast massage in lactation offered an immedi-
Mothers’ Impressions ate decrease in engorgement severity which, as confirmed by
In-office immediately following TBML treatment, mothers our 2-day survey, continued beyond the office visit.
were asked to rate how their pain changed. All mothers An additional strength of the study is that we put our sam-
reported less pain. Eighty-six percent of mothers at the 2-day ple in context by comparing the mothers receiving TBML for
survey and 82% at the 12-week survey found the massage their engorgement with a control group of postpartum moth-
helpful, with 65% of mothers at 12 weeks finding the mas- ers from a similar patient population. Our results suggest that
sage very helpful. although the majority of mothers do not have severe engorge-
When asked to describe what they found helpful, mothers ment postpartum, those mothers who do are likely to find
were likely to mention at least 1 of 3 themes: immediate immediate pain relief from TBML. Furthermore, there was
relief, learning specific techniques, or support received no difference in exclusive breastfeeding or breastfeeding
(Table 4). Mothers saw value in the in-office TBML and complications between the groups at the 12-week follow-up,
were pleased with this service. No mothers found the service although this should be interpreted with caution given the
unhelpful. One mother noted that the massage did not help sample size.
relieve her symptoms because later ultrasound evaluation This was a pilot study and, thus, long-term follow-up was
revealed that she had a benign solid mass rather than a a study limitation. Our study was not adequately powered to
plugged duct. Nevertheless, she noted, “I believe the breast evaluate the effect of treatment on weaning, breastfeeding
massage helped to increase my milk production.” rates, or mastitis or plugged duct recurrence. Future studies
could identify whether TBML has an effect on long-term
breastfeeding complications or success.
Discussion Empowering mothers to resolve complications at home
The primary findings of our study reveal that TBML pro- may be a critical skill for extending breastfeeding duration.
vides immediate significant pain reduction for engorgement, The 12-week follow-up was a study strength because it iden-
plugged ducts, and mastitis. We also demonstrate that TBML tified the potential benefit of training mothers in the hands-on
reduces engorgement severity and provides immediate size techniques. Our study is the first to ask if mothers found the
reduction—and, at times, complete resolution—for plugged techniques helpful both immediately and in the future. Our
ducts. Furthermore, mothers reported that TBML was results were encouraging because mothers with repeat epi-
helpful both immediately and for future episodes of mastitis sodes of milk stasis reported an ability to use the techniques

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
8 Journal of Human Lactation 

after in-office instruction. This suggests that TBML can be Funding

taught to mothers and other professionals. Further study will The authors disclosed receipt of the following financial support for
need to evaluate the most effective way to teach the massage the research, authorship, and/or publication of this article: This
technique, in particular how hands-on demonstration com- publication was made possible by the Clinical and Translational
pares to verbal or written teaching. Whereas teaching the Science Collaborative (CTSC) grant support (UL1TR 000439). The
TBML techniques may be empowering to mothers, it is project was also supported in part by the Dean’s Summer Research
important to emphasize that TBML complements other inter- Award from Case Western Reserve University School of Medicine.
ventions, and mothers and health care providers should con-
tinue to use established protocols for treatment.9,10 Other Supplementary Material
treatments are especially important in cases of infectious Supplementary material for this article is available online at http://
mastitis where milk stasis, although a contributing factor, is jhl.sagepub.com/supplemental.
not the only etiology, and antibiotics and other interventions
may be required.10 References
Given that our study was a descriptive pilot study, we
could not compare TBML with other reported treatments for 1. Eidelman A, Schanler RJ. Breastfeeding and the use of human
milk. Pediatrics. 2012;129(3):e827-e841.
milk stasis13 and therefore cannot assess relative efficacy.
2. Fein SB, Grummer-Strawn LM, Raju TNK. Infant feeding and care
Also, given that it was not a randomized controlled trial, we practices in the United States: results from the Infant Feeding
cannot assess placebo effect on pain reduction. Like other Practices Study II. Pediatrics. 2008;122(suppl 2):S25-S27.
studies, we found clinical improvement with hands-on treat- 3. Sheehan D, Krueger P, Watt S, Sword W, Bridle B. The
ment for plugged duct and engorgement.12,13,16 Similar to the Ontario Mother and Infant Survey: breastfeeding outcomes.
Thomsen et al13 study, we found symptomatic improvement J Hum Lact. 2001;17(3):211.
in mastitis with breast emptying. Future randomized studies 4. Declercq ER, Sakala C, Corry MP, Applebaum S. Listening
are needed to compare TBML with other hands-on treat- to Mothers II: report of the second national U.S. Survey
ments for milk stasis. of Women’s Childbearing Experiences. J Perinat Educ.
The study was conducted at a breastfeeding medicine prac- 2007;16(4):9-14.
tice where teamwork between lactation consultants and physi- 5. Li R, Fein SB, Chen J, Grummer-Strawn LM. Why mothers
stop breastfeeding: mothers’ self-reported reasons for stopping
cians allows the ability to clinically evaluate for serious
during the first year. Pediatrics. 2008;122(suppl):S69-S76.
complicating conditions such as abscess. We did not use 6. Schwartz K, D’Arcy H, Gillespie B, Bobo J, Longeqy M,
TBML in isolation but, rather, in the context of full breastfeed- Foxman B. Factors associated with weaning in the first 3
ing support. All practitioners in this study received specific months postpartum. J Fam Pract. 2002;51(5):439-444.
training on TBML techniques. Therefore, the reproducibility 7. Stuebe AM, Horton BJ, Chetwynd E, Watkins S, Grewen K,
of this study may be limited by breastfeeding support, experi- Meltzer-Brody S. Prevalence and risk factors for early, unde-
ence, skill level of practitioners, and training availability. In sired weaning attributed to lactation dysfunction. J Womens
addition, the study population was an educated, primarily Health. 2014;23(5):404-412.
Caucasian group committed to breastfeeding. The benefits of 8. Mangesi L, Dowswell T. Treatments for breast engorge-
TBML should not be extrapolated to other settings. ment during lactation. Cochrane Database Syst Rev.
2010;(9):CD006946.
9. World Health Organization. Mastitis: Causes and Management.
Conclusion Geneva, Switzerland: World Health Organization; 2000.
10. Amir L. ABM clinical protocol #4: mastitis. Breastfeed Med.
When treating mothers for acute breast pain associated with
2014;9(5):239-243.
milk stasis, our study supports a role for TBML in providing 11. Storr GB. Prevention of nipple tenderness and breast engorge-
immediate clinically significant improvements in pain, ment in the postpartal period. J Obstet Gynecol Neonatal Nurs.
engorgement, and plugged duct severity. Mothers find this 1988;17(3):203-209.
treatment helpful and continue to use the techniques learned 12. Zhao C, Tang R, Wang J, et al. Six-step recanalization manual
for future episodes. therapy: a novel method for treating plugged ducts in lactating
women. J Hum Lact. 2014;30(3):324-330.
Acknowledgments 13. Thomsen AC, Espersen T, Maigaard S. Course and treatment
of milk stasis, noninfectious inflammation of the breast and
Special thanks to Shelly Senders, MD, and the staff at Senders
infectious mastitis in nursing women. Am J Obstet Gynecol.
Pediatrics for their continued support in improving breastfeeding
1984;149(5):492-495.
practices. The authors thank Kristen Auletta and Mara Uguccini for
14. Bolman M, Saju L, Oganesyan K, Kondrashova T, Witt AM.
their assistance with data collection.
Recapturing the art of therapeutic breast massage during
breastfeeding. J Hum Lact. 2013;29(3):328-331.
Declaration of Conflicting Interests 15. Chiu JY, Gau ML, Kuo SY, Chang YH, Kuo SC, Tu HC.
The authors declared no potential conflicts of interest with respect Effects of Gua-Sha therapy on breast engorgement: a random-
to the research, authorship, and/or publication of this article. ized controlled trial. J Nurs Res. 2010;18(1):1-10.

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
 Witt et al 9

16. Kabir S, Tasnim S. Oketani lactation management: a new

20. Giugliani E. Common problems during lactation and their

method to augment breast milk. J Bangladesh Coll Physicians management. J Pediatr (Rio J). 2004;80(8):147-154.
Surg. 2009;27(3):155-159. 21. Witt AM, Smith S, Mason MJ, Flocke SA. Integrating routine
17.
International Lactation Consultant Association. ILCA lactation consultant support into a pediatric practice. Breastfeed
Clinician’s Breastfeeding Triage Tool. Morrisville, NC: Med. 2012;7(1):38-42.
International Lactation Consultant Association; 2014. 22. Witt AM, Mason MJ, Burgess K, Flock S, Zyzanski S. A case
18. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde control study of bacterial species and colony count in milk of
JG. Research electronic data capture (REDCap)—a metadata- breastfeeding women with chronic pain. Breastfeeding Med.
driven methodology and workflow process for providing 2014;9(1):29-34.
translational research informatics support. J Biomed Inform. 23. Ferreira-Valente MA, Pais-Ribeiro JL, Jensen MP. Validity
2009;42(2):377-381. of four pain intensity rating scales. Pain. 2011;152(10):
19. Hjermstad MJ, Fayers PM, Haugen DF, et al. Studies compar- 2399-2404.
ing numerical rating scales, verbal rating scales, and visual 24. Humenick SS, Hill PD, Anderson MA. Breast engorgement:
analogue scales for assessment of pain intensity in adults: a patterns and selected outcomes. J Hum Lact. 1994;10(2):87-93.
systematic literature review. J Pain Symptom Manage. 2011; 25. Berens P. ABM clinical protocol #20: engorgement. Breastfeed
41(6):1073-1093. Med. 2009;4(2):111-113.

Downloaded from jhl.sagepub.com at International Lactation Consultant Association on December 29, 2015
View publication stats