Pseudomonal breast infection

Abstract

Breast infection and breast sepsis secondary to Pseudomonas aeruginosa is uncommon. We report two cases of pseudomonal breast infection leading to septic shock and abscess formation in women with non-responding breast infection. The management of breast infection is broad-spectrum antibiotics and ultrasound with aspiration of any collection. To treat breast infection effectively, the causative organism must be isolated to enable appropriate antibiotic therapy.

Pseudomonas aeruginosa is an opportunistic, Gram-negative pathogen, commonly causing urinary tract infections, lower respiratory tract infections and soft tissue infections in immunocompromised patients. Breast sepsis is not commonly recognised. We present two cases of breast infection secondary to P. aeruginosa infection in previously healthy women with a review of the current literature.

Case history 1

A 54-year-old woman presented with a week-long history of breast pain, fever and rigors. Prior to admission, she had received a course of co-amoxiclav which had been changed to erythromycin prior to presentation. There was no significant past medical history and she was a non-smoker.

Clinically, the patient showed signs of sepsis. Breast examination revealed a large, tender erythematous area at the 3–5 o’clock position of the left breast, with no associated palpable lesion or axillary lymphadenopathy.

Blood results revealed a leukocytosis and raised C-reactive protein. Intravenous co-amoxiclav and metronidazole were commenced. Ultrasound scan of the left breast revealed multiple, dilated, fluid-filled ducts in the subareolar region; however, there was no discrete collection amenable to drainage.

Some 48 h later, this patient became cardiovascularly compromised. An arterial blood gas analysis suggested a metabolic acidosis. Her antibiotic regimen was changed to intravenous clindamycin and gentamicin upon discussion with the microbiology department. Although initially responding well to fluid resuscitation, she further became significantly more hypotensive and oliguric.

The patient was admitted to the intensive care unit and the decision was made to proceed to theatre for an incision and drainage of the left breast to drain any sepsis.

Intra-operatively, a circumareolar incision was made with blunt dissection to the chest wall. Scanty, organised loculations of pus were seen, with no abscess cavity being identifiable.

The patient was commenced on ciprofloxacin empirically. Pus swabs grew P. aeruginosa, susceptible to ciprofloxacin, which was continued for a total of 14 days. Histology revealed acute inflammatory tissue with degenerate debris and no evidence of bacteria or malignancy.

Her postoperative recovery was unremarkable and she was discharged on day 10 of admission. Subsequently, she was reviewed clinically and radiologically and had made a complete recovery. Two years later, she is fit and well.

Case history 2

A 58-year-old post-menopausal woman presented with a 4-day history of left breast tenderness and erythema. She had been commenced on oral co-amoxiclav prior to being seen. There was no significant past medical history and she was a non-smoker.

Examination revealed a large area of cellulitis in the outer lower quadrant of the left breast but no associated palpable lesion. There was no evidence of skin necrosis, no axillary lymphadenopathy or signs of systemic sepsis.

Ultrasound scan of the left breast revealed a 2-cm abscess cavity, deep to the areola. This was aspirated under ultrasound guidance, and cultures subsequently grew P. aeruginosa, susceptible to ciprofloxacin. This patient was commenced on a week’s course of ciprofloxacin and metronidazole.

Subsequent clinical and radiological review revealed resolving sepsis and the patient was discharged.

Discussion

We report the first cases of pseudomonal breast infection in immunocompetent women.

Breast abscesses are typically classified as lactational and non-lactational infections, commonly affecting women in the 18–50 year age group.

Non-lactational breast infection can be central or peripheral. Central infection usually develops secondary to periductal mastitis, with cigarette smoking recognised as a major causative factor.¹ The association between cigarette smoking and periductal mastitis is unclear; however, the accumulation of toxic metabolites, such as cotinine, nicotine, epoxides and lipid peroxidise within the breast lead to damage of the subareolar ducts and may predispose to infection. Furthermore, smoking disrupts the complement of the normal bacterial flora, inhibiting Gram-positive bacteria and promoting the overgrowth of Gram-negative bacteria,² thus increasing their pathogenic potential to cause infection.

Peripheral, non-lactating, breast abscesses are less common than central abscesses and are often associated with premenopausal women, diabetes, rheumatoid arthritis, granulomatous lobular mastitis, steroid therapy and trauma.³

The treatment for all breast infection is broad-spectrum antibiotics and ultrasound scan and drainage of any collection. All breast infection should be assessed by ultrasound; this permits the differentiation between simple cellulitis and an abscess, allowing for assessment of loculation and permitting aspiration of pus. The effectiveness of this was demonstrated in both our cases, where one case clearly demonstrated an abscess and in the other case multiple loculations were identified. The advantage offered by ultrasound-guided aspiration is reduced scarring, continuation of breast-feeding and no requirement of general anaesthesia or hospitalisation.⁴ Furthermore, the low cost and easy accessibility of ultrasound has enabled effective out-patient management of breast infection.

To treat breast abscesses effectively, the spectrum of likely causative organisms must be considered. Staphylococcus aureus and Staphylococcus epidermidis are the commonest pathogens isolated from breast abscesses,^5–9 with Proteus spp. and anaerobic bacteria (including Bacteroides spp. and peptostreptococci) also commonly implicated, in mixed culture, particularly in non-lactating women.¹⁰ The isolation of a mixed culture may be due to partially treated infection with antibiotics prior to aspiration.

Moazzez et al.⁵ found S. aureus in 32% of cultures, with more than half of these isolates being methicillin-resistant S. aureus (MRSA). Walker et al.⁶ found S. epidermidis to be the commonest isolated aerobic bacteria in their study, with 60% of cultures yielding this organism.

P. aeruginosa has been documented to cause breast abscesses in the immunocompromised;^11,12 however, in a recent study only 8% of breast abscesses yielded this organism.⁵ Other atypical organisms include Escherichia coli, Salmonella typhi and anaerobic organisms.

Conclusions

The majority of lactating and non-lactating breast infections are initially treated with co-amoxiclav antibiotic therapy, as in both our cases. It is prudent to consider atypical organisms, such as MRSA and P. aeruginosa, in patients who show no improvement on conventional first-line antibiotic therapy. It is this cohort of patients that require referral to a breast specialist for appropriate ultrasonographic imaging and aspiration to enable appropriate, directed antibiotic therapy.