Clinical Experience with Aspiration of Breast Abscesses Based on Size and Etiology at an Academic Medical Center Catherine S. Giess, MD,1 Mehra Golshan, MD,2 Karen Flaherty, MS, ANP-BC,2 Robyn L. Birdwell, MD, FACR1 1 2
Department of Radiology, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115 Department of Surgery, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115
Received 26 September 2013; accepted 3 June 2014
ABSTRACT: Purpose. Our purpose was to review needle aspiration of breast abscesses and identify factors associated with treatment by aspiration alone versus aspiration with surgical incision and drainage (I/D). Methods. This Institutional Review Board– approved, retrospective review of the breast ultrasound database from 2008 to 2010 identified 40 patients (41 abscesses) who underwent aspiration, with or without I/D. Demographics, imaging, number of aspirations, and microbiology were reviewed. Results. Twenty-two abscesses underwent aspiration only, 16 > 3 cm, 6 3 cm (mean 4.3 cm, range 0.9– 10 cm). Known risk factors included lactational (11), 3 weeks post partum (1), pregnancy (1), recent biopsy/ lumpectomy (5). Nineteen abscesses underwent aspiration and I/D, 15 > 3 cm, 4 3 cm (mean 4.1 cm, range 2.2–7.5 cm). Known risk factors included lactational (4), recurrent subareolar abscess (4), diabetes (3), hydradenitis suppuritiva (1), nipple piercing (2), smoking (1), pregnancy (1), HIV (1), and lumpectomy (1). Identified reasons for I/D included lack of improvement/recurrence (12), fistula (3), and one electively after clinical improvement of a recurrent subareolar abscess. Conclusions. Abscesses associated with pregnancy and lactation or breast biopsy are effectively managed with aspiration, even when large. Recurrence, chronicC 2014 ity, or fistula may require surgical intervention. V Wiley Periodicals, Inc. J Clin Ultrasound 42:513–521, 2014; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jcu.22191 Keywords: aspiration
breast;
ultrasound;
Correspondence to: C. S. Giess C 2014 Wiley Periodicals, Inc. V
VOL. 42, NO. 9, NOVEMBER/DECEMBER 2014
breast
abscess;
P
atients with mastitis present with a tender, focally or diffusely swollen breast. There is associated skin thickening and edema in the breast parenchyma, and there may be a focal fluctuant palpable mass when abscess is present. Breast ultrasound (US) is helpful in the diagnosis of breast abscess in this clinical setting. Breast abscesses can be broadly classified as puerperal and nonpuerperal, with nonpuerperal abscesses subclassified as peripheral or central subareolar.1,2 This broad classification is important for the breast imager and breast surgeon to be aware of, because patient demographics, inciting organisms, and clinical outcomes may be different between these groups. Traditional first-line treatment of breast abscesses had previously consisted of incision and drainage (I/D), coupled with antibiotic therapy. However, a number of studies have reported clinical success treating breast abscess with US-guided catheter placement or needle aspiration.3–9 Aspiration has therefore supplanted surgical I/D as first-line management in many clinical situations, although there is much variation in published technique.1 USguided percutaneous intervention offers a number of advantages over I/D; it is less invasive, is better tolerated by the patient, does not interfere with breast-feeding, may be performed repeatedly with minimal morbidity, and can be more cost-effective.5 US guidance enables the radiologist to direct a needle into loculated pockets with more precision than via palpation, and US is readily available in most radiology departments in the United States. Several years 513
GIESS ET AL
ago, our institution’s breast surgical team started referring abscess patients to breast imaging for serial US-guided aspirations during antibiotic therapy, in order to minimize surgical intervention. The purpose of this study was to review our success rate for needle aspiration in the treatment of large and small breast abscesses and to identify potential clinical and imaging factors associated with treatment by aspiration alone versus aspiration combined with surgical I/D.
MATERIALS AND METHODS
Institutional Review Board approval for retrospective review of patient medical records and images was given, with a waiver of informed consent. All consecutive breast sonographic examinations reporting an abscess or inflammatory fluid collection over a 3-year period (January 1, 2008 to December 31, 2010) were identified by an audit of the breast US database. Reports describing generalized inflammatory changes, such as skin thickening, soft tissue edema, cellulitis, or mastitis, but without a focal collection were excluded. Only patients in whom aspiration of a focal collection was performed at our institution, with or without additional surgical I/D during treatment, were included in the study. Cases were considered diagnostic of a breast abscess only when microbiology and or surgical pathology reports were reported as consistent with abscess. The study population included 40 patients with 41 abscesses (one patient had bilateral abscesses) who underwent aspiration with or without surgical I/D. All patients received antibiotics during the course of therapy. The longitudinal medical record was reviewed to obtain patient demographics, imaging findings, abscess size, number of aspirations, and microbiology and pathology findings. Diagnostic evaluations were done by staff radiologists at an academic medical center. US examinations were performed on Philips iu22 US machines (Bothell, WA) utilizing 7–15-MHz linear array transducers. US examinations were performed by attending radiologists and sometimes also radiologists in training. Digital mammograms, when performed, were done on Senographe DS, Senographe Essential equipment (GE Medical Systems, Milwaukee, WI). Standard imaging protocol at our institution for patients presenting with clinical signs suggestive of breast abscess is a targeted breast US 514
performed by the radiologist. Diagnostic mammography is not usually performed in this clinical setting, but may be done at the discretion of the diagnostic radiologist when the clinical or US presentation is atypical for abscess, and then only if the patient can tolerate mammographic compression. Breast abscess drainages are usually done under US guidance in the breast imaging section. The aspiration procedure at our institution includes local anesthesia using 2% unbuffered lidocaine hydrochloride in a 20 mg/ml concentration to the skin and superficial tissues. A straight needle (16–20-gauge) attached to a syringe is then inserted under US guidance into the collection, with aspiration of as much fluid as possible, for diagnostic and therapeutic purposes. Needle gauge is dependent on the viscosity of the fluid; an 18-G needle is most commonly used. Mild variations to the technique (needle gauge, buffered lidocaine, or tubing attached to the needle) may occur at the discretion of the radiologist, but lavage of the cavity with sterile saline or antibiotics is not part of our protocol. Fluid is sent to microbiology for Gram-stain, aerobic and anaerobic culture, and sensitivity. In atypical clinical presentations, such as an older patient without any risk factors, an indolent or subacute clinical course, or recurrent symptoms after treatment, fluid may also be sent for cytology, to exclude an unusual malignancy. Serial aspiration at our institution is defined as two or more US-guided aspirations of the same abscess collection, typically spaced 4–7 days apart, and occurring within the same clinical treatment period. Abscesses that recur following documented clinical and/or imaging resolution, and which undergo US-guided aspiration, would not be considered within the serial aspiration period, but would be considered a new clinical infection. At our institution, referral patterns for patients with suspected breast abscess can vary. Many patients with clinically suspected breast abscess are referred by the primary care providers or emergency room staff to the breast surgical team (including nurse practitioners and breast surgeons), who then refer the patient to radiology for US evaluation. Patients may be directly referred from primary care providers to radiology for imaging; when US findings are consistent with abscess, radiology performs same-day drainage and refers the patient to the surgical team for concurrent evaluation and ongoing clinical management. If the patient has clinical signs of fistula or skin breakdown, the JOURNAL OF CLINICAL ULTRASOUND
ASPIRATION OF BREAST ABSCESSES TABLE 1 Surgical versus Percutaneous Management of Breast Abscess According to Etiology Aspiration Only
Aspiration with I/D
12 1 0 0 0 0 0 5 0 4 22
4 1 4 1 1 1 3 1 2 1 19
Lactation/puerperal Pregnancy Recurrent subareolar abscess Primary subareolar abscess Chronic skin condition Immunocompromised Diabetes mellitus Recent breast intervention* Nipple piercing Unknown Total *Includes needle biopsy or excision.
radiologist refers the patient directly for surgical evaluation rather than performing aspiration, because the surgeon may elect to perform I/D in such cases. Patients are sent for serial sonographic examinations and, if necessary, serial drainages during antibiotic treatment in order to avoid surgical intervention whenever possible. The follow-up intervals for repeat breast sonographic examinations and additional aspiration are elected by the breast surgeon during follow-up clinical evaluation of the patient, every 4–7 days after initial presentation. The Fisher’s exact test was used to evaluate for a statistically significant difference in successful treatment by aspiration only for small (3 cm) versus large (>3 cm) abscesses, and for lactating or pregnant patients (the most common risk factor in this series), compared with other patients (SAS version 9.2, SAS Institute Inc, Cary, NC).
RESULTS
Aspiration-Only Group Twenty-two patients underwent aspiration only, with no surgical intervention. Patient age ranged from 15 to 59 years (mean 35 years). Risk factors for abscess development included the following (Table 1): lactation (N 5 11); 3 weeks post partum, not nursing (N 5 1); secondtrimester pregnancy (N 5 1); recent biopsy (N 5 3); post lumpectomy (N 5 2); and unknown (N 5 4). In the three patients with abscess following biopsy, one was lactating and reported developing a milk fistula at the recent core biopsy site, with subsequent abscess developVOL. 42, NO. 9, NOVEMBER/DECEMBER 2014
ment. This patient was classified as a post interventional abscess because, although lactating, clinically it was felt that the biopsy and milk fistula were the inciting events resulting in infection. One patient developed a superficial abscess after dermatologic biopsy of a skin mole. One patient developed an abscess after core biopsy at another facility. In the two patients who had undergone breastconservation surgery, one developed an infected seroma in the immediate postoperative period. The other patient had undergone lumpectomy 1 year previously, which was complicated by an infected seroma, with a more recent excision of the lumpectomy scar and subsequent redevelopment of an infected seroma. Abscess size in these 22 patients ranged from 0.9 to 10 cm (mean 4.3 cm), with 16 greater than 3 cm and 6 less than or equal to 3 cm in diameter. Fourteen patients underwent only one needle intervention (13 aspiration, 1 core biopsy), 6 had two aspirations (1 with later catheter placement), and 2 had three aspirations. The single abscess patient who underwent a core biopsy procedure diagnosing abscess had an unsuspected small abscess, presenting with a tender palpable complex cystic mass; the radiologist chose to perform core needle biopsy instead of fine needle aspiration. The majority of aspiration procedures were performed by radiology under US guidance, but three patients had palpatory-guided needle aspirations in surgery clinic (two with subsequent US-guided aspiration several days later, and one done in surgery clinic prior to any imaging), and another patient had an aspiration (details unknown) at an outside facility before seeking care at our institution. The single patient treated with an indwelling drainage catheter had the catheter placed by interventional radiology after two aspirations of a large recurring infected post lumpectomy seroma. US appearance of the abscesses ranged from mostly cystic to more solid-appearing masses. A complex, more solid appearance delayed aspiration in three patients (Figure 1), because the diagnostic radiologist reported that the findings were more consistent with phlegmon or inflammatory tissue rather than fluid. Three patients underwent diagnostic mammography: one showing an obscured mass, one an ill-defined mass, and one negative. Positive microbiology cultures in 19 patients demonstrated staphylococcus organisms (11 aureus, 4 coagulase negative, 4 multi-drugresistant [MRSA]); the remaining 3 patients 515
GIESS ET AL
FIGURE 1. A 35-year-old patient with lactational abscess. The patient had stopped breast-feeding 2 months previously and developed a swollen tender right breast with a focal palpable mass in the right upper inner quadrant. (A) US showed a complex collection (arrows) deep to the skin, which was initially thought to represent phlegmon/inflammatory tissue due to its small size and complex internal echotexture. The patient was placed on oral antibiotics. Symptoms worsened, and follow-up sonographic examination 5 days later (B) shows a larger, internally complex collection (arrows), which yielded 22 ml of pus on aspiration. Cytology and microbiology showed numerous neutrophils, reported as consistent with abscess; the patient had been on antibiotics for 5 days before aspiration, which resulted in no bacterial growth. Four days later, the patient had clinical improvement in symptoms, but a repeat sonographic examination (C) showed reaccumulation (arrows), with 16 ml of pus on repeat aspiration. Follow-up sonographic examination (D) 4 months later showed complete resolution.
had pathology findings reported as consistent with abscess. The four MRSA abscesses were community-acquired infections in lactating patients. All 22 patients were treated with antibiotics. Resolution of infection was documented clinically and by serial US in seven patients and clinically only (no follow-up sonographic examinations) in the remaining 15 patients. Aspiration with I/D Group Eighteen patients (19 abscesses; one patient had bilateral abscesses) underwent aspiration and I/D. Patient age in this group ranged from 17 to 61 years (mean 35 years). Risk factors for abscess development included the following (Table 1): lactation (N 5 4), recurrent subareolar abscess (Figure 2) (N 5 4, two with confirmed smoking history), smoking (N 5 1), diabetes 516
mellitus (DM) (N 5 3), hydradenitis suppuritiva (N 5 1), nipple piercing (N 5 2, one patient with type 1 DM), third-trimester pregnancy (N 5 1), HIV infection (N 5 1), breast-conservation surgery (N 5 1), and unknown (N 5 1). Fifteen abscesses were greater than 3 cm, and 4 abscesses were less than or equal to 3 cm (mean 4.1 cm, range 2.2–7.5 cm). Sixteen underwent one aspiration and 3 underwent two aspirations. The majority of aspirations were performed by radiologists, but three patients had an aspiration in surgical clinic (two patients with subsequent aspiration by radiology under US guidance). Abscess appearances on US ranged from predominantly cystic to complex or more solid in appearance. Two patients underwent concurrent diagnostic mammography, one showing a focal asymmetric density and the other a subareolar oval mass with associated skin thickening. JOURNAL OF CLINICAL ULTRASOUND
ASPIRATION OF BREAST ABSCESSES
FIGURE 2. A 49-year-old woman with subareolar abscess. The patient initially presented with right peri-areolar tenderness; cranio-caudal mammographic view (A) with metallic marker over area of focal tenderness shows peri-areolar thickening (arrows). (B) Initial US shows inflammatory change deep to the nipple (arrows); findings were deemed too small for aspiration. The patient was treated with oral antibiotics. A sonographic examination 2 months later (C) when the patient’s symptoms recurred shows small focal complex fluid collection (arrows) in the subareolar area. Aspiration yielded S. aureus. The patient had elective excision of the central lactiferous ducts after resolution, but she continued to have recurrent infections.
Positive cultures included five staphylococcus (two aureus, three MRSA), two beta hemolytic streptococcus, two proteus mirabilis, one pseudomonas aeruginosa, two corynebacterium (diphtheroids), one morganella morganii, and six anaerobic, all polymicrobial. The three patients with MRSA infections were community-acquired: two lactating and the third with no identifiable risk factors. All the patients were treated with antibiotics. Identifiable reasons for I/D in these 18 patients with 19 abscesses included a lack of clinical improvement and or recurrence in 12, fistula in 3, and electively after resolution in 1 (patient with a history of recurrent subareolar abscess). Review of the longitudinal medical record confirmed long-term resolution of the breast infection in 13 patients (two had subsequent nonmammary skin and soft tissue infections) and 5 patients with recurrent breast infections (4 with recurrent subareolar abscess and 1 lactating patient). No statistically significant difference in successful treatment by aspiration was found based on abscess size. Based on size less than or greater than 3 cm,4,7,8 60.0% (95% confidence interval: 26.2%, 87.8%) of abscesses 3 cm were treated by aspiration only. Evaluated with the Fisher’s exact test, this difference was not statistically significant (p 5 0.727). For those patients who were lactating or pregnant, abscess resolution with aspiration only was seen 72.2% of the time (95% binomial confidence interval: 46.5%, 90.3%). For those patients who were not lactating or pregnant, abscess resolution with aspiration only was seen 39.1% of the time (95% confidence interval: 19.7%, 61.5%). Evaluated with the Fisher’s exact test, this difference approached but did not meet statistical significance (p 5 0.058).
DISCUSSION
The classification of breast abscesses into puerperal and nonpuerperal types is clinically useful to the diagnostic radiologist, because patient factors, inciting organisms, and clinical outcomes vary and affect radiologic and clinical management. Puerperal abscesses develop when a milk duct becomes blocked, with resultant milk stasis and superimposed bacterial infection, either from skin contamination from the nursing mother’s skin flora or from the mouth 517
GIESS ET AL
of the nursing infant.2 Breast-feeding or pumping can and should continue provided the antibiotics given are safe for the nursing infant. In fact, it is important to ensure adequate drainage of the lactating breast to clear the infection and limit the bacterial culture medium. In the lactating patient, the most common organism found in abscess is Staphylococcus aureus.1,2,10,11 There are reports of an increasing incidence of community-acquired MRSA causing breast abscesses.12–14 Nonpuerperal breast abscess may be classified as central and subareolar (often recurrent), or peripheral.1 The central subareolar abscess is also called squamous metaplasia of lactiferous ducts and occurs in younger women, most of whom smoke cigarettes.1,2,15 The pathophysiology is believed to involve squamous metaplasia of the cuboidal epithelium in the central lactiferous ducts, followed by stasis and inspissation of secretions, the production of keratin plugs, and ductal obstruction. Ductal dilatation then occurs, followed by discontinuity or rupture of the epithelial lining, and a chemical inflammation from the luminal contents; colonization of the region by bacteria leads to a central or subareolar abscess.2,15–17 When advanced or chronic/recurrent, fistulas may occur to the areolar skin surface; this has been termed Zuska’s disease.1,18 Because of the distal obstruction of the lactiferous duct, abscesses tend to be recurrent.15–17 Recurrent subareolar abscesses are often polymicrobial and anaerobic.2,15–17 Peripheral breast abscesses may occur in patients with chronic conditions like DM, a history of recent breast interventions (biopsy, excision, or nipple piercing), or immune-compromised states.1 S. aureus remains a common infective organism in peripheral nonpuerperal abscesses, but Pseudomonas aeruginosa, Proteus mirabilis, and other less common organisms may be found.1,5,12 We routinely send abscess fluid from our patients for both aerobic and anaerobic cultures. Imaging predominantly involves targeted breast sonographic examination. The US appearance of breast abscess is an irregular or illdefined, internally heterogeneous, hypoechoic or anechoic mass, sometimes with fluid-debris levels or loculations, and posterior acoustic enhancement.1,10,19,20 Sometimes breast abscesses may appear circumscribed on US; internally, abscesses may appear solid or mixed solid and cystic.19 There is often associated skin thickening, edema in the surrounding parenchyma, and increased echogenicity of the fat lobules.1,7 It is important to note that many abscesses appear very complex 518
or even solid internally,19 particularly when the internal contents are thick, hemorrhagic, or very proteinaceous (lactational). In our series, three patients did not undergo aspiration at the first US evaluation, because the radiologist interpreted the US to be more consistent with phlegmon or inflammatory solid tissue than thick/ loculated pus fluid. We recommend a low threshold for attempting US-guided aspiration in suspected abscess, because even small amounts of aspirated pus can help tailor antibiotics by correct identification of the infectious organism. US findings in central subareolar abscesses demonstrate a peri-areolar or subareolar hypoechoic lesion, usually ill defined, at times hypervascular; occasionally a fistulous tract can be identified.15 In the acute setting, mammography is not commonly performed. Most patients with acute abscess have an exquisitely tender breast and are unable to tolerate mammographic compression. When performed, mammography typically shows skin thickening, a focal asymmetry, illdefined or circumscribed mass, and occasionally, architectural distortion.1,19,20 Mammography should be performed in older patients, patients with atypical clinical presentations, or patients with recurrent or nonimproving symptoms. The most serious differential diagnostic possibility is inflammatory breast cancer and should be considered for older patients and nonresolving inflammatory findings. US-guided percutaneous catheter placement to treat breast abscesses was described in the 1990s.3,21 At our institution, we prefer to avoid catheter placement, because it requires ongoing care and many patients dislike managing a catheter at home. We have found US-guided aspiration to be an effective treatment, particularly in the uncomplicated case of lactational or postbiopsy or postexcisional abscess. Some authors have suggested an abscess size less than 3 cm as a guideline for attempting aspiration rather than catheter placement,4,7,8,22 or for probable success of treatment from aspiration alone.4 Others have reported an abscess size greater than 5 cm to be a risk factor for failure to treat successfully via aspiration,23 although those authors did not use US guidance. In this study, we found no significant difference in successful treatment by aspiration based on abscess size. Patients in this series underwent surgical I/D due to lack of clinical improvement or recurrence, fistula formation, or electively after resolution in a patient with a history of recurrent abscess. However, the decision to proceed with JOURNAL OF CLINICAL ULTRASOUND
ASPIRATION OF BREAST ABSCESSES
FIGURE 3. Radiology management algorithm. PCP, primary care physician.
surgical I/D was made by the breast surgeon on a case-by-case basis rather than in a standardized fashion. Subareolar recurrent abscess, because the underlying ducts remain obstructed, usually requires resection of the involved ducts. Even with surgical treatment, recurrence has been reported in 28% of cases in one study.16 For patients with subareolar abscess, therefore, ongoing surgical oversight is necessary. Although our results endorse management of breast abscesses by serial US-guided aspiration regardless of abscess size, our experience also suggests that long-term, nonsurgical success is likely to be influenced by etiology. It is therefore important for the radiologist and surgeon to be in close communication regarding the management of patients with a breast abscess. Some authors have recommended irrigation of the abscess cavity with sterile saline or antibiotics after aspiration of pus. Leborgne and Leborgne6 irrigated cavities with sterile saline and instilled antibiotics into cavities >25 mm in diameter. Karstrup et al3 irrigated with sterile saline after placement of a pigtail catheter. Imperiale et al5 reported successful treatment of 27 abscesses in patients who had failed systemic antibiotics by injecting antibiotics into the abscess cavity and discontinuing systemic antibiotics. None of these studies was randomized or case-controlled. As Trop et al1 have pointed out, there are no studies in the radiology or surgical literature that clearly delineate the best management for breast abscesses. VOL. 42, NO. 9, NOVEMBER/DECEMBER 2014
Follow-up of abscess patients can be clinical as well as radiologic. Although Trop et al1 consider clinical follow-up for puerperal abscesses sufficient, they recommend clinical and US follow-up for nonpuerperal abscesses. In our institution, follow-up US examinations are ordered at the discretion of our surgical colleagues, depending on the clinical response. It is essential to ensure that patients with breast abscess are followed to clinical resolution in order to exclude the unlikely possibility of inflammatory breast cancer. In cases with an incomplete clinical response, US should be performed as well. Our study has limitations. Our study population was relatively small, and the study design was retrospective and observational rather than prospective and randomized. Both of these factors limit the validity of any statistical analysis. Our study reports a single institution’s clinical experience, and there may have been some variability among our breast surgeons in how patients were managed, particularly pertaining to the choice of repeated aspiration versus surgical I/D. Our data do not conclusively indicate the best method of treatment for breast abscess; as other authors have pointed out, a consensus on best treatment practices remains elusive.1 However, our observation that the success of US-guided aspiration is not dependent on the size of the collection is important for clinical care. The utility of repeated aspiration during treatment of breast abscess is likewise helpful for clinicians and radiologists managing 519
GIESS ET AL
patients with breast abscess. US-guided aspiration is minimally invasive, well-tolerated, and less costly than I/D, and repeated aspirations may avoid surgical intervention in some patients. We believe that our identification of clinical factors associated with additional I/D following US-guided aspiration will be practically useful to radiologists. In conclusion, this series reports practical, multidisciplinary clinical experience in managing breast abscess at a large urban academic center. Our results endorse serial US-guided aspirations in the diagnosis and treatment of breast abscesses, regardless of size, particularly when uncomplicated and in an immunecompetent patient. We advocate a low radiologic threshold for needle intervention in the setting of suspected breast infection, because even a small amount of pus can appropriately guide antibiotic therapy. Lack of clinical improvement after aspiration, or skin breakdown, or fistula formation should prompt surgical intervention. A radiology management algorithm is suggested in Figure 3. Although our results did not reach statistical significance, they suggest that serial aspiration for nonlactational abscesses may be less successful in completely resolving infection than for lactational abscesses, and surgical treatment may ultimately prove necessary in some cases. However, serial aspiration for diagnostic and therapeutic purposes in nonlactational abscesses may allow the surgeon to better plan surgical intervention for a time when the breast tissue is less inflamed and infected, in order to optimize healing. We believe that a multidisciplinary collaboration for treating breast abscess patients, which involves gynecology, primary care, surgery, and radiology, will provide the best clinical outcomes for these patients.
3.
4.
5.
6.
7.
8.
9.
10.
11. 12.
13.
14.
ACKNOWLEDGMENT We gratefully acknowledge Dorothy A. Sippo, MD for statistical assistance.
15.
16. REFERENCES 1. Trop I, Dugas A, David J, et al. Breast abscesses: evidence-based algorithms for diagnosis, management, and follow-up. Radiographics 2011;31: 1683. 2. Meguid MM, Kort KC, Numann PJ, et al. Subareolar breast abscess: the penultimate stage of the mammary duct-associated inflammatory disease sequence. In: Bland KI, Copeland EM III, editors.
520
17.
18. 19.
The Breast: Comprehensive Management of Benign and Malignant Disorders, 3rd ed. Philadelphia: Saunders; 2004, p 93. Karstrup S, Solvig J, Nolsoe CP, et al. Acute puerperal breast abscesses: US-guided drainage. Radiology 1993;188:807. Hook GW, Ikeda DM. Treatment of breast abscesses with US-guided percutaneous needle drainage without indwelling catheter placement. Radiology 1999;213:579. Imperiale A, Zandrino F, Calabrese M, et al. Abscesses of the breast: US-guided serial percutaneous aspiration and local antibiotic therapy after unsuccessful systemic antibiotic therapy. Acta Radiol 2001;42:161. Leborgne F, Leborgne F. Treatment of breast abscesses with sonographically guided aspiration, irrigation, and instillation of antibiotics. AJR Am J Roentgenol 2003;181:1089. Ulitzsch D, Nyman MKG, Carlson RA. Breast abscess in lactating women: US-guided treatment. Radiology 2004;232:904. Christensen AF, Al-Suliman N, Nielsen KR, et al. Ultrasound-guided drainage of breast abscesses: results in 151 patients. Br J Rad 2005;78:186. Ozseker B, Ozcan UA, Rasa K, et al. Treatment of breast abscesses with ultrasound-guided aspiration and irrigation in the emergency setting. Emerg Radiol 2008;15:105. Sabate JM, Clotet M, Torrubia S, et al. Radiologic evaluation of breast disorders related to pregnancy and lactation. Radiographics 2007;27:S101. Dener C, Inan A. Breast abscesses in lactating women. World J Surg 2003;27:130. Moazzez A, Kelso RL, Towfigh S, et al. Breast abscess bacteriologic features in the era of community-acquired methicillin-resistant staphylococcus aureus epidemics. Arch Surg 2007;142:881. Berens P, Swaim L, Peterson B. Incidence of methicillin-resistant staphylococcus aureus in postpartum breast abscesses. Breastfeed Med 2010;5:113. Chen CY, Anderson BO, Lo SS, et al. Methicillinresistant staphylococcus aureus infections may not impede the success of ultrasound-guided drainage of puerperal breast abscesses. J Am Coll Surg 2010;210:148. Lo G, Dessauvagie B, Sterrett G, et al. Squamous metaplasia of lactiferous ducts (SMOLD). Clin Radiol 2012;67:e42. Versluijs-Ossewaarde FNL, Roumen RMH, Goris RJA. Subareolar breast abscesses: characteristics and results of surgical treatment. Breast J 2005;11:179. Gollapalli V, Liao J, Dudakovic A, et al. Risk factors for development and recurrence of primary breast abscesses. J Am Coll Surg 2010;211:41. Zuska JJ, Crile G Jr, Ayres WW. Fistulas of lactifierous ducts. Am J Surg 1951;81:312. Lequin MH, van Spengler J, van Pel R, et al. Mammographic and sonographic spectrum of nonpuerperal mastitis. Eur J Radiol 1995;21:138. JOURNAL OF CLINICAL ULTRASOUND
ASPIRATION OF BREAST ABSCESSES 20. Crowe DJ, Helvie MA, Wilson TE. Breast infection: mammographic and sonographic findings with clinical correlation. Invest Radiol 1995; 30:582. 21. Berna JD, Garcia-Medina V, Madrigal M, Guirao J, et al. Percutaneous catheter drainage of breast abscesses. Eur J Radiol 1996;21:217.
VOL. 42, NO. 9, NOVEMBER/DECEMBER 2014
22. Berna-Serna JD, Madrigal M, Berna-Serna JD. Percutaneous management of breast abscesses. An experience of 39 cases. Ultrasound Med Biol 2004; 30:1. 23. Eryilmaz R, Sahin M, Tekelioglu MH, et al. Management of lactational breast abscesses. Breast 2005;14:375.
521
Department of Radiology, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115 Department of Surgery, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115
Received 26 September 2013; accepted 3 June 2014
ABSTRACT: Purpose. Our purpose was to review needle aspiration of breast abscesses and identify factors associated with treatment by aspiration alone versus aspiration with surgical incision and drainage (I/D). Methods. This Institutional Review Board– approved, retrospective review of the breast ultrasound database from 2008 to 2010 identified 40 patients (41 abscesses) who underwent aspiration, with or without I/D. Demographics, imaging, number of aspirations, and microbiology were reviewed. Results. Twenty-two abscesses underwent aspiration only, 16 > 3 cm, 6 3 cm (mean 4.3 cm, range 0.9– 10 cm). Known risk factors included lactational (11), 3 weeks post partum (1), pregnancy (1), recent biopsy/ lumpectomy (5). Nineteen abscesses underwent aspiration and I/D, 15 > 3 cm, 4 3 cm (mean 4.1 cm, range 2.2–7.5 cm). Known risk factors included lactational (4), recurrent subareolar abscess (4), diabetes (3), hydradenitis suppuritiva (1), nipple piercing (2), smoking (1), pregnancy (1), HIV (1), and lumpectomy (1). Identified reasons for I/D included lack of improvement/recurrence (12), fistula (3), and one electively after clinical improvement of a recurrent subareolar abscess. Conclusions. Abscesses associated with pregnancy and lactation or breast biopsy are effectively managed with aspiration, even when large. Recurrence, chronicC 2014 ity, or fistula may require surgical intervention. V Wiley Periodicals, Inc. J Clin Ultrasound 42:513–521, 2014; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jcu.22191 Keywords: aspiration
breast;
ultrasound;
Correspondence to: C. S. Giess C 2014 Wiley Periodicals, Inc. V
VOL. 42, NO. 9, NOVEMBER/DECEMBER 2014
breast
abscess;
P
atients with mastitis present with a tender, focally or diffusely swollen breast. There is associated skin thickening and edema in the breast parenchyma, and there may be a focal fluctuant palpable mass when abscess is present. Breast ultrasound (US) is helpful in the diagnosis of breast abscess in this clinical setting. Breast abscesses can be broadly classified as puerperal and nonpuerperal, with nonpuerperal abscesses subclassified as peripheral or central subareolar.1,2 This broad classification is important for the breast imager and breast surgeon to be aware of, because patient demographics, inciting organisms, and clinical outcomes may be different between these groups. Traditional first-line treatment of breast abscesses had previously consisted of incision and drainage (I/D), coupled with antibiotic therapy. However, a number of studies have reported clinical success treating breast abscess with US-guided catheter placement or needle aspiration.3–9 Aspiration has therefore supplanted surgical I/D as first-line management in many clinical situations, although there is much variation in published technique.1 USguided percutaneous intervention offers a number of advantages over I/D; it is less invasive, is better tolerated by the patient, does not interfere with breast-feeding, may be performed repeatedly with minimal morbidity, and can be more cost-effective.5 US guidance enables the radiologist to direct a needle into loculated pockets with more precision than via palpation, and US is readily available in most radiology departments in the United States. Several years 513
GIESS ET AL
ago, our institution’s breast surgical team started referring abscess patients to breast imaging for serial US-guided aspirations during antibiotic therapy, in order to minimize surgical intervention. The purpose of this study was to review our success rate for needle aspiration in the treatment of large and small breast abscesses and to identify potential clinical and imaging factors associated with treatment by aspiration alone versus aspiration combined with surgical I/D.
MATERIALS AND METHODS
Institutional Review Board approval for retrospective review of patient medical records and images was given, with a waiver of informed consent. All consecutive breast sonographic examinations reporting an abscess or inflammatory fluid collection over a 3-year period (January 1, 2008 to December 31, 2010) were identified by an audit of the breast US database. Reports describing generalized inflammatory changes, such as skin thickening, soft tissue edema, cellulitis, or mastitis, but without a focal collection were excluded. Only patients in whom aspiration of a focal collection was performed at our institution, with or without additional surgical I/D during treatment, were included in the study. Cases were considered diagnostic of a breast abscess only when microbiology and or surgical pathology reports were reported as consistent with abscess. The study population included 40 patients with 41 abscesses (one patient had bilateral abscesses) who underwent aspiration with or without surgical I/D. All patients received antibiotics during the course of therapy. The longitudinal medical record was reviewed to obtain patient demographics, imaging findings, abscess size, number of aspirations, and microbiology and pathology findings. Diagnostic evaluations were done by staff radiologists at an academic medical center. US examinations were performed on Philips iu22 US machines (Bothell, WA) utilizing 7–15-MHz linear array transducers. US examinations were performed by attending radiologists and sometimes also radiologists in training. Digital mammograms, when performed, were done on Senographe DS, Senographe Essential equipment (GE Medical Systems, Milwaukee, WI). Standard imaging protocol at our institution for patients presenting with clinical signs suggestive of breast abscess is a targeted breast US 514
performed by the radiologist. Diagnostic mammography is not usually performed in this clinical setting, but may be done at the discretion of the diagnostic radiologist when the clinical or US presentation is atypical for abscess, and then only if the patient can tolerate mammographic compression. Breast abscess drainages are usually done under US guidance in the breast imaging section. The aspiration procedure at our institution includes local anesthesia using 2% unbuffered lidocaine hydrochloride in a 20 mg/ml concentration to the skin and superficial tissues. A straight needle (16–20-gauge) attached to a syringe is then inserted under US guidance into the collection, with aspiration of as much fluid as possible, for diagnostic and therapeutic purposes. Needle gauge is dependent on the viscosity of the fluid; an 18-G needle is most commonly used. Mild variations to the technique (needle gauge, buffered lidocaine, or tubing attached to the needle) may occur at the discretion of the radiologist, but lavage of the cavity with sterile saline or antibiotics is not part of our protocol. Fluid is sent to microbiology for Gram-stain, aerobic and anaerobic culture, and sensitivity. In atypical clinical presentations, such as an older patient without any risk factors, an indolent or subacute clinical course, or recurrent symptoms after treatment, fluid may also be sent for cytology, to exclude an unusual malignancy. Serial aspiration at our institution is defined as two or more US-guided aspirations of the same abscess collection, typically spaced 4–7 days apart, and occurring within the same clinical treatment period. Abscesses that recur following documented clinical and/or imaging resolution, and which undergo US-guided aspiration, would not be considered within the serial aspiration period, but would be considered a new clinical infection. At our institution, referral patterns for patients with suspected breast abscess can vary. Many patients with clinically suspected breast abscess are referred by the primary care providers or emergency room staff to the breast surgical team (including nurse practitioners and breast surgeons), who then refer the patient to radiology for US evaluation. Patients may be directly referred from primary care providers to radiology for imaging; when US findings are consistent with abscess, radiology performs same-day drainage and refers the patient to the surgical team for concurrent evaluation and ongoing clinical management. If the patient has clinical signs of fistula or skin breakdown, the JOURNAL OF CLINICAL ULTRASOUND
ASPIRATION OF BREAST ABSCESSES TABLE 1 Surgical versus Percutaneous Management of Breast Abscess According to Etiology Aspiration Only
Aspiration with I/D
12 1 0 0 0 0 0 5 0 4 22
4 1 4 1 1 1 3 1 2 1 19
Lactation/puerperal Pregnancy Recurrent subareolar abscess Primary subareolar abscess Chronic skin condition Immunocompromised Diabetes mellitus Recent breast intervention* Nipple piercing Unknown Total *Includes needle biopsy or excision.
radiologist refers the patient directly for surgical evaluation rather than performing aspiration, because the surgeon may elect to perform I/D in such cases. Patients are sent for serial sonographic examinations and, if necessary, serial drainages during antibiotic treatment in order to avoid surgical intervention whenever possible. The follow-up intervals for repeat breast sonographic examinations and additional aspiration are elected by the breast surgeon during follow-up clinical evaluation of the patient, every 4–7 days after initial presentation. The Fisher’s exact test was used to evaluate for a statistically significant difference in successful treatment by aspiration only for small (3 cm) versus large (>3 cm) abscesses, and for lactating or pregnant patients (the most common risk factor in this series), compared with other patients (SAS version 9.2, SAS Institute Inc, Cary, NC).
RESULTS
Aspiration-Only Group Twenty-two patients underwent aspiration only, with no surgical intervention. Patient age ranged from 15 to 59 years (mean 35 years). Risk factors for abscess development included the following (Table 1): lactation (N 5 11); 3 weeks post partum, not nursing (N 5 1); secondtrimester pregnancy (N 5 1); recent biopsy (N 5 3); post lumpectomy (N 5 2); and unknown (N 5 4). In the three patients with abscess following biopsy, one was lactating and reported developing a milk fistula at the recent core biopsy site, with subsequent abscess developVOL. 42, NO. 9, NOVEMBER/DECEMBER 2014
ment. This patient was classified as a post interventional abscess because, although lactating, clinically it was felt that the biopsy and milk fistula were the inciting events resulting in infection. One patient developed a superficial abscess after dermatologic biopsy of a skin mole. One patient developed an abscess after core biopsy at another facility. In the two patients who had undergone breastconservation surgery, one developed an infected seroma in the immediate postoperative period. The other patient had undergone lumpectomy 1 year previously, which was complicated by an infected seroma, with a more recent excision of the lumpectomy scar and subsequent redevelopment of an infected seroma. Abscess size in these 22 patients ranged from 0.9 to 10 cm (mean 4.3 cm), with 16 greater than 3 cm and 6 less than or equal to 3 cm in diameter. Fourteen patients underwent only one needle intervention (13 aspiration, 1 core biopsy), 6 had two aspirations (1 with later catheter placement), and 2 had three aspirations. The single abscess patient who underwent a core biopsy procedure diagnosing abscess had an unsuspected small abscess, presenting with a tender palpable complex cystic mass; the radiologist chose to perform core needle biopsy instead of fine needle aspiration. The majority of aspiration procedures were performed by radiology under US guidance, but three patients had palpatory-guided needle aspirations in surgery clinic (two with subsequent US-guided aspiration several days later, and one done in surgery clinic prior to any imaging), and another patient had an aspiration (details unknown) at an outside facility before seeking care at our institution. The single patient treated with an indwelling drainage catheter had the catheter placed by interventional radiology after two aspirations of a large recurring infected post lumpectomy seroma. US appearance of the abscesses ranged from mostly cystic to more solid-appearing masses. A complex, more solid appearance delayed aspiration in three patients (Figure 1), because the diagnostic radiologist reported that the findings were more consistent with phlegmon or inflammatory tissue rather than fluid. Three patients underwent diagnostic mammography: one showing an obscured mass, one an ill-defined mass, and one negative. Positive microbiology cultures in 19 patients demonstrated staphylococcus organisms (11 aureus, 4 coagulase negative, 4 multi-drugresistant [MRSA]); the remaining 3 patients 515
GIESS ET AL
FIGURE 1. A 35-year-old patient with lactational abscess. The patient had stopped breast-feeding 2 months previously and developed a swollen tender right breast with a focal palpable mass in the right upper inner quadrant. (A) US showed a complex collection (arrows) deep to the skin, which was initially thought to represent phlegmon/inflammatory tissue due to its small size and complex internal echotexture. The patient was placed on oral antibiotics. Symptoms worsened, and follow-up sonographic examination 5 days later (B) shows a larger, internally complex collection (arrows), which yielded 22 ml of pus on aspiration. Cytology and microbiology showed numerous neutrophils, reported as consistent with abscess; the patient had been on antibiotics for 5 days before aspiration, which resulted in no bacterial growth. Four days later, the patient had clinical improvement in symptoms, but a repeat sonographic examination (C) showed reaccumulation (arrows), with 16 ml of pus on repeat aspiration. Follow-up sonographic examination (D) 4 months later showed complete resolution.
had pathology findings reported as consistent with abscess. The four MRSA abscesses were community-acquired infections in lactating patients. All 22 patients were treated with antibiotics. Resolution of infection was documented clinically and by serial US in seven patients and clinically only (no follow-up sonographic examinations) in the remaining 15 patients. Aspiration with I/D Group Eighteen patients (19 abscesses; one patient had bilateral abscesses) underwent aspiration and I/D. Patient age in this group ranged from 17 to 61 years (mean 35 years). Risk factors for abscess development included the following (Table 1): lactation (N 5 4), recurrent subareolar abscess (Figure 2) (N 5 4, two with confirmed smoking history), smoking (N 5 1), diabetes 516
mellitus (DM) (N 5 3), hydradenitis suppuritiva (N 5 1), nipple piercing (N 5 2, one patient with type 1 DM), third-trimester pregnancy (N 5 1), HIV infection (N 5 1), breast-conservation surgery (N 5 1), and unknown (N 5 1). Fifteen abscesses were greater than 3 cm, and 4 abscesses were less than or equal to 3 cm (mean 4.1 cm, range 2.2–7.5 cm). Sixteen underwent one aspiration and 3 underwent two aspirations. The majority of aspirations were performed by radiologists, but three patients had an aspiration in surgical clinic (two patients with subsequent aspiration by radiology under US guidance). Abscess appearances on US ranged from predominantly cystic to complex or more solid in appearance. Two patients underwent concurrent diagnostic mammography, one showing a focal asymmetric density and the other a subareolar oval mass with associated skin thickening. JOURNAL OF CLINICAL ULTRASOUND
ASPIRATION OF BREAST ABSCESSES
FIGURE 2. A 49-year-old woman with subareolar abscess. The patient initially presented with right peri-areolar tenderness; cranio-caudal mammographic view (A) with metallic marker over area of focal tenderness shows peri-areolar thickening (arrows). (B) Initial US shows inflammatory change deep to the nipple (arrows); findings were deemed too small for aspiration. The patient was treated with oral antibiotics. A sonographic examination 2 months later (C) when the patient’s symptoms recurred shows small focal complex fluid collection (arrows) in the subareolar area. Aspiration yielded S. aureus. The patient had elective excision of the central lactiferous ducts after resolution, but she continued to have recurrent infections.
Positive cultures included five staphylococcus (two aureus, three MRSA), two beta hemolytic streptococcus, two proteus mirabilis, one pseudomonas aeruginosa, two corynebacterium (diphtheroids), one morganella morganii, and six anaerobic, all polymicrobial. The three patients with MRSA infections were community-acquired: two lactating and the third with no identifiable risk factors. All the patients were treated with antibiotics. Identifiable reasons for I/D in these 18 patients with 19 abscesses included a lack of clinical improvement and or recurrence in 12, fistula in 3, and electively after resolution in 1 (patient with a history of recurrent subareolar abscess). Review of the longitudinal medical record confirmed long-term resolution of the breast infection in 13 patients (two had subsequent nonmammary skin and soft tissue infections) and 5 patients with recurrent breast infections (4 with recurrent subareolar abscess and 1 lactating patient). No statistically significant difference in successful treatment by aspiration was found based on abscess size. Based on size less than or greater than 3 cm,4,7,8 60.0% (95% confidence interval: 26.2%, 87.8%) of abscesses 3 cm were treated by aspiration only. Evaluated with the Fisher’s exact test, this difference was not statistically significant (p 5 0.727). For those patients who were lactating or pregnant, abscess resolution with aspiration only was seen 72.2% of the time (95% binomial confidence interval: 46.5%, 90.3%). For those patients who were not lactating or pregnant, abscess resolution with aspiration only was seen 39.1% of the time (95% confidence interval: 19.7%, 61.5%). Evaluated with the Fisher’s exact test, this difference approached but did not meet statistical significance (p 5 0.058).
DISCUSSION
The classification of breast abscesses into puerperal and nonpuerperal types is clinically useful to the diagnostic radiologist, because patient factors, inciting organisms, and clinical outcomes vary and affect radiologic and clinical management. Puerperal abscesses develop when a milk duct becomes blocked, with resultant milk stasis and superimposed bacterial infection, either from skin contamination from the nursing mother’s skin flora or from the mouth 517
GIESS ET AL
of the nursing infant.2 Breast-feeding or pumping can and should continue provided the antibiotics given are safe for the nursing infant. In fact, it is important to ensure adequate drainage of the lactating breast to clear the infection and limit the bacterial culture medium. In the lactating patient, the most common organism found in abscess is Staphylococcus aureus.1,2,10,11 There are reports of an increasing incidence of community-acquired MRSA causing breast abscesses.12–14 Nonpuerperal breast abscess may be classified as central and subareolar (often recurrent), or peripheral.1 The central subareolar abscess is also called squamous metaplasia of lactiferous ducts and occurs in younger women, most of whom smoke cigarettes.1,2,15 The pathophysiology is believed to involve squamous metaplasia of the cuboidal epithelium in the central lactiferous ducts, followed by stasis and inspissation of secretions, the production of keratin plugs, and ductal obstruction. Ductal dilatation then occurs, followed by discontinuity or rupture of the epithelial lining, and a chemical inflammation from the luminal contents; colonization of the region by bacteria leads to a central or subareolar abscess.2,15–17 When advanced or chronic/recurrent, fistulas may occur to the areolar skin surface; this has been termed Zuska’s disease.1,18 Because of the distal obstruction of the lactiferous duct, abscesses tend to be recurrent.15–17 Recurrent subareolar abscesses are often polymicrobial and anaerobic.2,15–17 Peripheral breast abscesses may occur in patients with chronic conditions like DM, a history of recent breast interventions (biopsy, excision, or nipple piercing), or immune-compromised states.1 S. aureus remains a common infective organism in peripheral nonpuerperal abscesses, but Pseudomonas aeruginosa, Proteus mirabilis, and other less common organisms may be found.1,5,12 We routinely send abscess fluid from our patients for both aerobic and anaerobic cultures. Imaging predominantly involves targeted breast sonographic examination. The US appearance of breast abscess is an irregular or illdefined, internally heterogeneous, hypoechoic or anechoic mass, sometimes with fluid-debris levels or loculations, and posterior acoustic enhancement.1,10,19,20 Sometimes breast abscesses may appear circumscribed on US; internally, abscesses may appear solid or mixed solid and cystic.19 There is often associated skin thickening, edema in the surrounding parenchyma, and increased echogenicity of the fat lobules.1,7 It is important to note that many abscesses appear very complex 518
or even solid internally,19 particularly when the internal contents are thick, hemorrhagic, or very proteinaceous (lactational). In our series, three patients did not undergo aspiration at the first US evaluation, because the radiologist interpreted the US to be more consistent with phlegmon or inflammatory solid tissue than thick/ loculated pus fluid. We recommend a low threshold for attempting US-guided aspiration in suspected abscess, because even small amounts of aspirated pus can help tailor antibiotics by correct identification of the infectious organism. US findings in central subareolar abscesses demonstrate a peri-areolar or subareolar hypoechoic lesion, usually ill defined, at times hypervascular; occasionally a fistulous tract can be identified.15 In the acute setting, mammography is not commonly performed. Most patients with acute abscess have an exquisitely tender breast and are unable to tolerate mammographic compression. When performed, mammography typically shows skin thickening, a focal asymmetry, illdefined or circumscribed mass, and occasionally, architectural distortion.1,19,20 Mammography should be performed in older patients, patients with atypical clinical presentations, or patients with recurrent or nonimproving symptoms. The most serious differential diagnostic possibility is inflammatory breast cancer and should be considered for older patients and nonresolving inflammatory findings. US-guided percutaneous catheter placement to treat breast abscesses was described in the 1990s.3,21 At our institution, we prefer to avoid catheter placement, because it requires ongoing care and many patients dislike managing a catheter at home. We have found US-guided aspiration to be an effective treatment, particularly in the uncomplicated case of lactational or postbiopsy or postexcisional abscess. Some authors have suggested an abscess size less than 3 cm as a guideline for attempting aspiration rather than catheter placement,4,7,8,22 or for probable success of treatment from aspiration alone.4 Others have reported an abscess size greater than 5 cm to be a risk factor for failure to treat successfully via aspiration,23 although those authors did not use US guidance. In this study, we found no significant difference in successful treatment by aspiration based on abscess size. Patients in this series underwent surgical I/D due to lack of clinical improvement or recurrence, fistula formation, or electively after resolution in a patient with a history of recurrent abscess. However, the decision to proceed with JOURNAL OF CLINICAL ULTRASOUND
ASPIRATION OF BREAST ABSCESSES
FIGURE 3. Radiology management algorithm. PCP, primary care physician.
surgical I/D was made by the breast surgeon on a case-by-case basis rather than in a standardized fashion. Subareolar recurrent abscess, because the underlying ducts remain obstructed, usually requires resection of the involved ducts. Even with surgical treatment, recurrence has been reported in 28% of cases in one study.16 For patients with subareolar abscess, therefore, ongoing surgical oversight is necessary. Although our results endorse management of breast abscesses by serial US-guided aspiration regardless of abscess size, our experience also suggests that long-term, nonsurgical success is likely to be influenced by etiology. It is therefore important for the radiologist and surgeon to be in close communication regarding the management of patients with a breast abscess. Some authors have recommended irrigation of the abscess cavity with sterile saline or antibiotics after aspiration of pus. Leborgne and Leborgne6 irrigated cavities with sterile saline and instilled antibiotics into cavities >25 mm in diameter. Karstrup et al3 irrigated with sterile saline after placement of a pigtail catheter. Imperiale et al5 reported successful treatment of 27 abscesses in patients who had failed systemic antibiotics by injecting antibiotics into the abscess cavity and discontinuing systemic antibiotics. None of these studies was randomized or case-controlled. As Trop et al1 have pointed out, there are no studies in the radiology or surgical literature that clearly delineate the best management for breast abscesses. VOL. 42, NO. 9, NOVEMBER/DECEMBER 2014
Follow-up of abscess patients can be clinical as well as radiologic. Although Trop et al1 consider clinical follow-up for puerperal abscesses sufficient, they recommend clinical and US follow-up for nonpuerperal abscesses. In our institution, follow-up US examinations are ordered at the discretion of our surgical colleagues, depending on the clinical response. It is essential to ensure that patients with breast abscess are followed to clinical resolution in order to exclude the unlikely possibility of inflammatory breast cancer. In cases with an incomplete clinical response, US should be performed as well. Our study has limitations. Our study population was relatively small, and the study design was retrospective and observational rather than prospective and randomized. Both of these factors limit the validity of any statistical analysis. Our study reports a single institution’s clinical experience, and there may have been some variability among our breast surgeons in how patients were managed, particularly pertaining to the choice of repeated aspiration versus surgical I/D. Our data do not conclusively indicate the best method of treatment for breast abscess; as other authors have pointed out, a consensus on best treatment practices remains elusive.1 However, our observation that the success of US-guided aspiration is not dependent on the size of the collection is important for clinical care. The utility of repeated aspiration during treatment of breast abscess is likewise helpful for clinicians and radiologists managing 519
GIESS ET AL
patients with breast abscess. US-guided aspiration is minimally invasive, well-tolerated, and less costly than I/D, and repeated aspirations may avoid surgical intervention in some patients. We believe that our identification of clinical factors associated with additional I/D following US-guided aspiration will be practically useful to radiologists. In conclusion, this series reports practical, multidisciplinary clinical experience in managing breast abscess at a large urban academic center. Our results endorse serial US-guided aspirations in the diagnosis and treatment of breast abscesses, regardless of size, particularly when uncomplicated and in an immunecompetent patient. We advocate a low radiologic threshold for needle intervention in the setting of suspected breast infection, because even a small amount of pus can appropriately guide antibiotic therapy. Lack of clinical improvement after aspiration, or skin breakdown, or fistula formation should prompt surgical intervention. A radiology management algorithm is suggested in Figure 3. Although our results did not reach statistical significance, they suggest that serial aspiration for nonlactational abscesses may be less successful in completely resolving infection than for lactational abscesses, and surgical treatment may ultimately prove necessary in some cases. However, serial aspiration for diagnostic and therapeutic purposes in nonlactational abscesses may allow the surgeon to better plan surgical intervention for a time when the breast tissue is less inflamed and infected, in order to optimize healing. We believe that a multidisciplinary collaboration for treating breast abscess patients, which involves gynecology, primary care, surgery, and radiology, will provide the best clinical outcomes for these patients.
3.
4.
5.
6.
7.
8.
9.
10.
11. 12.
13.
14.
ACKNOWLEDGMENT We gratefully acknowledge Dorothy A. Sippo, MD for statistical assistance.
15.
16. REFERENCES 1. Trop I, Dugas A, David J, et al. Breast abscesses: evidence-based algorithms for diagnosis, management, and follow-up. Radiographics 2011;31: 1683. 2. Meguid MM, Kort KC, Numann PJ, et al. Subareolar breast abscess: the penultimate stage of the mammary duct-associated inflammatory disease sequence. In: Bland KI, Copeland EM III, editors.
520
17.
18. 19.
The Breast: Comprehensive Management of Benign and Malignant Disorders, 3rd ed. Philadelphia: Saunders; 2004, p 93. Karstrup S, Solvig J, Nolsoe CP, et al. Acute puerperal breast abscesses: US-guided drainage. Radiology 1993;188:807. Hook GW, Ikeda DM. Treatment of breast abscesses with US-guided percutaneous needle drainage without indwelling catheter placement. Radiology 1999;213:579. Imperiale A, Zandrino F, Calabrese M, et al. Abscesses of the breast: US-guided serial percutaneous aspiration and local antibiotic therapy after unsuccessful systemic antibiotic therapy. Acta Radiol 2001;42:161. Leborgne F, Leborgne F. Treatment of breast abscesses with sonographically guided aspiration, irrigation, and instillation of antibiotics. AJR Am J Roentgenol 2003;181:1089. Ulitzsch D, Nyman MKG, Carlson RA. Breast abscess in lactating women: US-guided treatment. Radiology 2004;232:904. Christensen AF, Al-Suliman N, Nielsen KR, et al. Ultrasound-guided drainage of breast abscesses: results in 151 patients. Br J Rad 2005;78:186. Ozseker B, Ozcan UA, Rasa K, et al. Treatment of breast abscesses with ultrasound-guided aspiration and irrigation in the emergency setting. Emerg Radiol 2008;15:105. Sabate JM, Clotet M, Torrubia S, et al. Radiologic evaluation of breast disorders related to pregnancy and lactation. Radiographics 2007;27:S101. Dener C, Inan A. Breast abscesses in lactating women. World J Surg 2003;27:130. Moazzez A, Kelso RL, Towfigh S, et al. Breast abscess bacteriologic features in the era of community-acquired methicillin-resistant staphylococcus aureus epidemics. Arch Surg 2007;142:881. Berens P, Swaim L, Peterson B. Incidence of methicillin-resistant staphylococcus aureus in postpartum breast abscesses. Breastfeed Med 2010;5:113. Chen CY, Anderson BO, Lo SS, et al. Methicillinresistant staphylococcus aureus infections may not impede the success of ultrasound-guided drainage of puerperal breast abscesses. J Am Coll Surg 2010;210:148. Lo G, Dessauvagie B, Sterrett G, et al. Squamous metaplasia of lactiferous ducts (SMOLD). Clin Radiol 2012;67:e42. Versluijs-Ossewaarde FNL, Roumen RMH, Goris RJA. Subareolar breast abscesses: characteristics and results of surgical treatment. Breast J 2005;11:179. Gollapalli V, Liao J, Dudakovic A, et al. Risk factors for development and recurrence of primary breast abscesses. J Am Coll Surg 2010;211:41. Zuska JJ, Crile G Jr, Ayres WW. Fistulas of lactifierous ducts. Am J Surg 1951;81:312. Lequin MH, van Spengler J, van Pel R, et al. Mammographic and sonographic spectrum of nonpuerperal mastitis. Eur J Radiol 1995;21:138. JOURNAL OF CLINICAL ULTRASOUND
ASPIRATION OF BREAST ABSCESSES 20. Crowe DJ, Helvie MA, Wilson TE. Breast infection: mammographic and sonographic findings with clinical correlation. Invest Radiol 1995; 30:582. 21. Berna JD, Garcia-Medina V, Madrigal M, Guirao J, et al. Percutaneous catheter drainage of breast abscesses. Eur J Radiol 1996;21:217.
VOL. 42, NO. 9, NOVEMBER/DECEMBER 2014
22. Berna-Serna JD, Madrigal M, Berna-Serna JD. Percutaneous management of breast abscesses. An experience of 39 cases. Ultrasound Med Biol 2004; 30:1. 23. Eryilmaz R, Sahin M, Tekelioglu MH, et al. Management of lactational breast abscesses. Breast 2005;14:375.
521
