Incidence of breast abscess in lactating women: report from an Australian cohort


Dr L. Amir, Centre for the Study of Mothers' and Children's Health, 251 Faraday Street, Carlton 3053, Australia.


Objectives  To report the incidence of breast abscess in lactating women.

Design  Data were combined from two studies (a randomised controlled trial and a survey) to provide a cohort of women for this report.

Setting  Women were recruited from two hospitals on one site in Melbourne, Victoria, the Royal Women's Hospital (public) (1999–2001) and Frances Perry House (private) (2000–2001).

Participants  A total of 1193 of 1311 (91%) primiparous, English-speaking women from a diverse range of backgrounds, including those receiving public clinic care, private care and birth centre care.

Methods  A structured telephone interview was conducted on breastfeeding at six months postpartum.

Main outcome measures  Lactational mastitis and breast abscess.

Results  Two hundred and seven women experienced mastitis. Five women developed a breast abscess: 0.4% of women who commenced breastfeeding (95% CI 0.14–0.98); 2.9% of women who took antibiotics for mastitis (95% CI 1.0–6.7).

Conclusion  Although many authors estimate that 11% of women with mastitis develop a breast abscess, the incidence of lactating breast abscesses in Australia appears to be lower than reported in the past. Our estimate is that 3% of women with mastitis will develop a breast abscess.


Mastitis is an inflammation of breast tissue, which may or may not result from infection.1 Infective mastitis in lactating women is usually associated with Staphylococcus aureus (S. aureus),1 an organism which characteristically causes abscess development.2 A breast abscess is a localised collection of pus within the breast that usually occurs as a complication of mastitis.3

Lactating women and the clinicians caring for them are concerned about the possibility of breast abscesses. As some authors suggest that up to 11% of women with mastitis will develop a breast abscess,4–7 this is a potentially significant health issue.

We were unable to identify a recent population-based study of breast abscesses using Medline (key words: ‘breast’ and ‘abscess’, latest update 10 March 2004). The findings on breast abscesses from two breastfeeding studies in Melbourne, Victoria, Australia, are presented here.


Data from two studies have been combined for this paper, reflecting a diverse range of women attending the Royal Women's Hospital, a public, tertiary referral centre and Frances Perry House (private), located on the same site. Inclusion criteria in both studies included primiparity and ability to speak English. A randomised controlled trial (Attachment to the Breast and Family Attitudes to Breastfeeding [ABFAB]) to test whether breastfeeding education in mid-pregnancy could increase the duration of breastfeeding, recruited public patients at the Royal Women's Hospital at 18–20 weeks of gestation from May 1999 to August 2001.8 Two groups of women not included in the randomised controlled trial due to already high breastfeeding rates were recruited to a concurrent breastfeeding survey (the Survey) from August 2000 to March 2001: women from the Family Birth Centre (recruited at 36 weeks of gestation) and private patients from Frances Perry House (recruited two to four days postpartum).

A telephone interview at six months postpartum included questions about breastfeeding problems. Women with any symptoms of mastitis since birth (pain, hardness/lumps, redness of the breast, fever or ‘flu-like’ symptoms) were asked if they had experienced a breast abscess. If so, further questions about diagnosis and management of the abscess were asked. Stata 8 was used to calculate exact binomial 95% confidence intervals for proportions.

Both studies received approval from the Human Research Ethics Committees at the Royal Women's Hospital and La Trobe University. The Survey was also approved by the Medical Advisory Committee at Frances Perry House.


The interview at six months postpartum was completed by 889 (91%) of the 981 women enrolled in the randomised controlled trial, and 304 women, 92% of the 330 women recruited for the Survey. One hundred and seventy-one women were treated by antibiotics for at least one episode of mastitis (109 in the randomised controlled trial [12.3%], and 62 [20.4%], in the Survey, or 14.5% overall).

Five women reported having a breast abscess: four in the randomised controlled trial and one in the Survey. Each had one abscess. Of the 1183 women who commenced breastfeeding, 0.4% (5/1183) experienced a breast abscess. All five had received antibiotics prior to abscess development, thus 2.9% (5/171) of women who took antibiotics for mastitis experienced a breast abscess.

Table 1 lists the investigations, management and outcomes for each woman with a breast abscess. All women had their breast abscess drained by needle aspiration; one also required open drainage. It is considered safe to continue breastfeeding, during and after treatment of an abscess1; three of the five women were continuing to breastfeed at six months postpartum.

Table 1.  Investigations, management and breastfeeding outcome in women with a breast abscess.
Timing of abscessInvestigation(s)ManagementOutcome
Abscess at 3 weeks following mastitis at 5 daysNoneNeedle aspirationContinuing to breastfeed at 6 months
Abscess at 5 weeks associated with mastitisNoneNeedle aspiration and open drainageStopped breastfeeding at 6 weeks because of mastitis, flat nipples and low milk supply. History of abscesses under arms
Abscess at 7 weeks (had mastitis at 3 weeks—took antibiotics)Ultrasound, mammogramNeedle aspirationStopped breastfeeding at 6 weeks (baby could not suck effectively), thus weaning from breast occurred prior to abscess
Abscess at 8 weeks associated with mastitisUltrasoundNeedle aspirationContinuing to breastfeed at 6 months
Abscess after 2nd mastitis at 16 weeksUltrasoundNeedle aspiration (6 times)Continuing to breastfeed at 6 months


Breast abscesses were more common prior to the use of antibiotics. A prospective surveillance study conducted in one town in Scotland, from 1941 to 1943, found that 156 women developed a breast abscess, which was 8.9% of women giving birth (156/1751).9 In the 1950s and early 1960s, severe staphylococcal infections occurred worldwide due to the phage 80 ‘golden staph’.10 For example, during an outbreak of breast abscesses at Philadelphia General Hospital in late 1954, 16 abscesses occurred in one month.11 In Edinburgh in 1957, it was estimated that 3% to 4% of all women giving birth developed a breast abscess requiring hospital treatment.12

Although authors described an apparent fall in the incidence of breast abscesses in lactating women in the 1980s,3,13 as recently as 2002, Foxman et al.7 stated that ‘abscesses are reported to occur in 11% of all affected women’. However, the 1970 reference cited is a 20-year review of 53 patients seen by a private obstetrician, dating from 1948.14 The WHO review of mastitis (2000) concluded that 11% of women with mastitis develop an abscess from the same study,14 and also quoted 11% for the Thomsen study (when this should be 2.8% of women with clinical mastitis). 19Table 2 presents the incidence of breast abscess in developed countries published since 1970. Recently, the risks of nonlactational breast abscess development following nipple piercing have been highlighted.15

Table 2.  Reported incidence of breast abscess.
Author(s), year of publicationStudy populationRate of breast abscess of women with mastitisComments
Devereux 197014Patients seen by a private obstetrician, 1948–196815% (8/53)20 year study; antibiotic prescribed was sulfisoxazole. Has been reported as 11%, which indicates proportion of episodes of mastitis (rather than number of women).
Marshall et al. 197518Women attending a maternity hospital with mastitis, 1971–19734.6% (3/65)The three breast abscesses occurred in women who had stopped breastfeeding.
Thomsen et al., 198419213 women with non-infectious inflammation or infectious mastitis2.8% (6/213)Non-infectious mastitis defined as >106 leucocytes and <106 bacteria per mL of milk.
  Infectious mastitis defined as >106 leucocytes and >106 bacteria per mL of milk.
  In practice, these are indistinguishable, and would both present as clinical mastitis.
165 women with infectious mastitis3.6% (6/165)Women with infectious mastitis were randomized to 3 groups:
  (a) No treatment (n= 55)
  (b) Emptying the breast (n= 55)
  (c) Antibiotics and emptying the breast (n= 55)
55 women with infectious mastitis—Given no treatment (i.e. group ‘a’ above)11% (6/55)The only women who developed an abscess were in the group that received no treatment
Current cohort, 20041193 women at 6 months postpartum; 171 had mastitis (defined as took antibiotics for mastitis)2.9% (5/171) (95% CI 1.0–6.7)Of the total population, 5/1193 women giving birth developed a breast abscess 0.4% (95% CI 0.14–0.98)

Previously, standard treatment for breast abscesses was incision and drainage, followed by post-operative dressings, but needle aspiration has become more commonly practised. Two large case series (33 and 45 lactational breast abscesses, respectively) have concluded that aspiration is an effective treatment, which is also convenient for the woman and cost effective.16,17

Accurate estimates of the proportion of women who develop a breast abscess following mastitis are difficult due to the varied definitions of mastitis used. Estimates that 11% of women with mastitis develop a breast abscess are frequently published.1,4–7 However, the current incidence in developed countries appears to be much lower: our estimate is 3% of women with mastitis will develop a breast abscess. Current management by needle aspiration enables women to recover faster and facilitates continued breastfeeding.


The authors would like to thank all the women who participated in the study during a busy period in their lives, the childbirth educators, the student midwives who participated in piloting the intervention, the research midwives and the ABFAB study team.

Attachment to the Breast and Family Attitudes to Breastfeeding (ABFAB) funding was obtained from the National Health and Medical Research Council (NHMRC) and PhD scholarships (DF) from The Royal Women's Hospital (2002) and VicHealth (Victorian Health Promotion Foundation) (2003–2004). The Survey was funded by a grant from the Royal Australian College of General Practitioners. LA had a NHMRC Public Health Scholarship (2000–2002).

Contributions of authors

JL, DF and HM developed and conducted the ABFAB trial. LA conducted the Survey and wrote the first draft of the article.

Conflict of interest statement

None to report.

Accepted 26 April 2004