Breast Abscess Bacteriologic Features in the Era of Community-Acquired Methicillin-Resistant Staphylococcus aureus Epidemics

Hypothesis  Increasing rates of community-acquired methicillin-resistant Staphylococcus aureus (MRSA) infections have also affected the microbial profile of breast abscesses.

Objective  To update the decade-old bacteriologic description of breast abscesses to improve the choice of initial antibacterial drug therapy.

Design  Retrospective case series.

Setting  County hospital emergency department.

Patients  Forty-four women (mean age, 41 years; age range, 20-63 years) with breast abscesses.

Methods  All cultures from the breast abscesses of patients were reviewed.

Main Outcome Measures  The microbiologic features and sensitivities of breast abscesses.

Results  Of 46 specimens only 28 showed bacterial yield (61%). Of these, 11 (39%) were polymicrobial, for an average of 1.4 isolates per specimen. The most common organism was S aureus, present in 12 of 37 aerobic cultures (32%), with MRSA in 7 (58%). The remaining organisms included coagulase-negative Staphylococcus (16%), diphtheroids (16%), Pseudomonas aeruginosa (8%), Proteus mirabilis (5%), and other isolates (22%). All MRSA was sensitive to clindamycin, trimethoprim-sulfamethoxazole, and linezolid. Only 2 patients (29%) were sensitive to levofloxacin. Two anaerobic cultures were positive for Propionibacterium acnes and Peptostreptococcus anaerobius.

Conclusions  Staphylococcus aureus is the most common pathogenic organism in modern breast abscesses. Many breast abscesses have community-acquired MRSA, with more than 50% of all S aureus and 19% of all cultures being MRSA. This finding parallels the local and national increases in MRSA reported in other soft-tissue infections. With increasing bacterial resistance and more minimally invasive management of breast abscesses, understanding the current bacteriologic profile of these abscesses is essential to determining the correct empirical antibiotic drug therapy.

Nonpuerperal breast abscesses remain a significant disease owing to their underlying disability and incidence of recurrence.¹ Incision and drainage with postoperative antibiotic drugs has been suggested as the treatment of choice for these abscesses based on their pathogenesis.² However, recent studies^3-5 have suggested that ultrasound-guided aspiration plus systemic antibiotic drug therapy may be less invasive, with an improved cosmetic outcome and a higher cure rate. To increase the success rate of minimally invasive drainage techniques with adjunctive antimicrobial drug therapy, knowledge of the microbiologic spectrum of breast abscesses is necessary. Whereas the best drainage modality for the treatment of breast abscesses is being extensively studied, the bacteriologic features of nonpuerperal breast abscesses has not been systematically evaluated for a decade. This study reviews the cultures of these abscesses for the purpose of updating their bacteriologic profile and elucidating antimicrobial susceptibilities in an era of emerging bacterial resistance.

All female patients who underwent treatment for a breast abscess at Los Angeles County + USC Medical Center, Los Angeles, between May 1, 2003, and May 1, 2005, were retrospectively reviewed. They were identified based on a discharge diagnosis of breast abscess and on microbiologic laboratory records of specimens obtained from the breast. Institutional review board approval was obtained before acquisition of patient health information.

In 2 years, 44 female patients (mean age, 41 years; age range, 20-63 years) were treated for breast abscesses using either incisional or percutaneous drainage. Forty-six specimens were sent for aerobic culture, and 8 also had anaerobic cultures. Bacterial yield was seen in 28 of 46 specimens (61%) sent for culture (Table 1). Of these 28 cultures, 17 (61%) were positive for a single organism and 11 (39%) contained multiple organisms. Twenty-seven cultures (96%) yielded aerobic bacteria. Two of 8 cultures (25%) sent for anaerobic cultures had positive results. A total of 39 organisms were detected, 37 of which were aerobic and 2 of which were anaerobic, averaging 1.4 isolates per specimen. None of the 11 specimens sent for acid-fast bacilli or fungus culture yielded organisms.

Staphylococcus aureus was the most common aerobic organism, present in 12 cultures (32%), with 7 (58%) of these being methicillin-resistant S aureus (MRSA). The remaining positive cultures yielded coagulase-negative Staphylococcus (16%), diphtheroids (16%), Pseudomonas aeruginosa (8%), Proteus mirabilis (5%), and other bacteria (22%) (Table 1). Propionibacterium acnes and Peptostreptococcus anaerobius were the anaerobic bacteria.

The antibiogram shows that 74% (14 of 19 isolates) of the cultured bacteria were susceptible to levofloxacin (Table 2). All except P mirabilis were susceptible to gentamicin, and only Escherichia coli and P aeruginosa were resistant to trimethoprim-sulfamethoxazole. All MRSA was susceptible to clindamycin, trimethoprim-sulfamethoxazole, and linezolid. Only 2 MRSA strains (29%) were susceptible to levofloxacin, and the remaining were intermediately susceptible.

For more than a decade no studies have been published about the microbiologic features of breast abscesses. Our study found that S aureus is the most common aerobic organism, found in 32% (12 of 37) of cultures, with more than half of these isolates being MRSA. Only 2 of the 4 most recent studies,^6-9 published between 1988 and 1995, found S aureus to be the most common isolate. This new trend suggests that the emergence of MRSA in the community has reintroduced S aureus as the dominant isolate in breast abscesses.

For decades MRSA was a nosocomial pathogen seen predominantly in hospitalized patients. Recently it has caused infections in patients, including those without risk factors.¹⁰ A 2006 study¹¹ suggests that the prevalence of MRSA in patients presenting to emergency departments has increased to 59% of soft-tissue infections. β-Lactamase activity was previously reported in organisms cultured from breast abscesses,⁸ although none showed methicillin resistance. In our study, 58% of S aureus was methicillin resistant and, as seen with community-acquired MRSA,^10,11 was susceptible to most non–β-lactam antibiotic agents.

Walker et al⁷ found coagulase-negative Staphylococcus in 60% of cultures, which was the most common aerobic bacteria in their study. Edmiston et al⁸ reported that the same organism was the most common bacteria found in chronic breast abscesses. We found coagulase-negative Staphylococcus and diphtheroids to be the second most common organisms yielded in cultures (16%). The presence of coagulase-negative Staphylococcus reflects their ability to adhere to squamous epithelial cells,⁷ suggesting that these aerobic bacteria will continue to be prominent organisms in the pathogenesis of breast abscesses.

Pseudomonas aeruginosa and P mirabilis constitute the next most common isolates. They were found in 8% and 5% of the cultures, respectively. Alados et al⁹ identified their most common aerobic organism as P mirabilis, which was found in 31.8% of their cultures. In their study more than half of these cultures were from chronic and recurrent abscesses. It is important to identify this pathogen owing to its persistence in chronic infections, implying the need for aggressive early management.

Most studies^6-8 reported a high incidence of anaerobic bacteria in their cultures. In the present study only 8 specimens were sent for anaerobic cultures, and, of these, 2 had positive findings. This low incidence most likely represents a failure of collection of an anaerobic specimen or an inappropriate medium for transfer of the specimen to the laboratory. Because of the reported high incidence, consideration regarding the collection and transport of an appropriate anaerobic container needs to be made when a specimen from a breast abscess is sent for culture. This will ensure the detection of anaerobic bacteria and adequate antibiotic drug coverage.

Despite the diversity of bacteria found in breast abscesses, the antiobiogram demonstrates that all bacteria have some susceptibility to levofloxacin, although MRSA shows mostly intermediate sensitivity (Table 2). Levofloxacin was the only oral antibiotic with adequate activity against P aeruginosa. A previous study⁸ also showed that 96.7% of 213 isolates from breast abscesses were sensitive to ciprofloxacin. Although this suggests that fluoroquinolones may be the best empirical antibiotic agents for breast abscesses, it conflicts with the 2005 antibiogram from Los Angeles County + USC Medical Center, which shows that only 22% of MRSA is sensitive to levofloxacin. Based on additional microbiologic experience with fluoroquinolones, an intermediate minimum inhibitory concentration can rapidly become a resistant minimum inhibitory concentration.

The MRSA in breast abscesses was adequately covered by the following oral antibiotics: trimethoprim-sulfamethoxazole, clindamycin, tetracycline, and linezolid. Based on current knowledge of MRSA sensitivity, the predominance of community-associated strains of MRSA, and inferred data from soft-tissue infection studies, the best empirical initial therapies with standard adult doses are clindamycin (450 mg orally 4 times daily), trimethoprim-sulfamethoxazole (2 double-strength tablets orally twice daily), doxycycline (100 mg orally twice daily), and rifampin (600 mg orally every 24 hours or if nausea 300 mg orally twice daily) combination regimens.^11,12 Rifampin is no longer effective when used as monotherapy because of increased resistance, and it is more effective when combined with the previously listed oral antibiotic agents.¹³ Caution should be used with clindamycin because 50% of MRSA has inducible or constitutive resistance. Treatment failures have also occurred in up to 21% with tetracycline, so clinical progress bears monitoring. Linezolid (600 mg orally twice daily) compared with other available antibiotic agents is a less cost-effective choice for initial empirical therapy. Patients with severe infection requiring intravenous antibiotic drug therapy are also candidates for treatment with vancomycin (1 g intravenously every 12 hours), tigecycline (a 100-mg intravenous initial dose, then 50 mg intravenously every 12 hours), and daptomycin (6 mg/kg intravenously every 24 hours).¹⁴ Most important, antimicrobial drug therapy should be modified for adequate bacterial coverage based on individual cultures and sensitivity testing.

The trend toward increasing minimally invasive management of breast abscesses is based on algorithms with empirical antibiotic drug therapy. Antibiotic agents have consistently been used with aspiration or catheter drainage of breast abscesses.^3,5,15 Christensen et al³ treated 151 breast abscesses (puerperal and nonpuerperal) using ultrasound-guided aspiration (or catheter drainage if the cavity was >3 cm). They treated all patients with dicloxacillin and added metronidazole if the abscess was nonpuerperal. Erythromycin was used for penicillin-allergic patients. Of 62 nonpuerperal abscesses, 1 had recurrence of abscess and 11 required surgical incision and drainage, 5 of which had complex fistulas. Berna-Serna et al⁵ used a combination of amoxicillin with clavulanic acid and clindamycin as adjunctive therapy to aspiration. All abscesses smaller than 3 cm resolved without surgical intervention, although 3 needed repeated aspiration. Leborgne and Leborgne¹⁵ used ultrasound-guided aspiration of 73 breast abscesses and oral cephradine therapy. In addition, if the cavity was larger than 2.5 cm, irrigation of the cavity was performed using cephradine. They reported a 96% success rate and a potential role for local antibiotic drug therapy. These various treatment algorithms each report high success with percutaneous drainage while relying on an adjunctive antibiotic component. Although not proved yet, it seems that antibiotic agents may be more important in patients treated with minimally invasive techniques vs open techniques and may have an important role in source control.

Staphylococcus aureus and P aeruginosa are the most common pathogenic organisms associated with present-day breast abscesses. Of abscesses containing S aureus, 58% (7 of 12) were methicillin resistant. This increased prevalence of MRSA corresponds to the nationwide increases reported for other skin and soft-tissue infections. With increasing bacterial resistance and more minimally invasive management of breast abscesses, such as ultrasound-guided drainage plus systemic antibiotic drug therapy, understanding the current bacteriologic features of these abscesses is essential to determining the correct choice of empirical antibiotic drug therapy. Although all the bacteria showed sensitivity to levofloxacin, the intermediate sensitivity of MRSA combined with the local antibiogram and pharmacologic data suggests that levofloxacin is not an ideal choice of empirical therapy. Based on data from this study, current MRSA treatment guidelines, known effectiveness, known potential for resistance, cost, and patient compliance, the best empirical oral antibiotic drug therapy for patients with breast abscesses is trimethoprim-sulfamethoxazole.

Correspondence: Rodney J. Mason, MD, PhD, Department of Surgery, Los Angeles County + USC Medical Center, 1200 N State St, Room 10850, Los Angeles, CA 90035 (rmason@surgery.usc.edu).

Accepted for Publication: April 27, 2007.

Author Contributions:Study concept and design: Moazzez, Berne, and Mason. Acquisition of data: Moazzez and Mason. Analysis and interpretation of data: Moazzez, Kelso, Towfigh, Sohn, and Mason. Drafting of the manuscript: Moazzez and Mason. Critical revision of the manuscript for important intellectual content: Kelso, Towfigh, Sohn, Berne, and Mason. Study supervision: Kelso, Berne, and Mason.

Financial Disclosure: None reported.

Previous Presentation: This poster was presented at the 78th Annual Meeting of the Pacific Coast Surgical Association; February 18, 2007; Kohala Coast, Hawaii; and is published after peer review and revision.