Necrotising fasciitis of the breast: A rare but deadly disease

Highlights • 54 year old lady presenting with breast erythema and raised inflammatory markers.• Initially managed with IV antibiotics but deteriorated with worsening erythema a few days into her admission.• CT thorax showed a large volume of interstitial gas in the breast with histological samples confirming necrotising fasciitis.• We propose the use of cross-sectional imaging to investigate potential cases of necrotising fasciitis of the breast.


Introduction
Necrotising fasciitis (NF) is a bacterial inflammation of the fascia leading to necrosis of the underlying soft tissue and fascial layers. It is associated with a high mortality rate of approximately 25% [1]. Thankfully it is very rare, with approximately 500 reported cases annually in the UK [2]. It most commonly affects the superficial fascial layers of the extremities, the abdomen or the perineum and is commonly associated with an initial traumatic injury [3]. Furthermore, patient co-morbidities increase the risk of developing NF. These include diabetes, immunosuppression and frailty [1]. Cases of necrotising fasciitis of the breast (NFB) are even rarer, with very few reports in the literature (discussed below).
* Group A streptococcus +/− staphylococci infection and is the more severe [4].
The clinical features associated with NF are erythema, localised inflammation and pain at the infection site. Pain is often disproportionate to examination findings and serves as a warning sign to underlying NF [2]. The characteristic skin changes classically associated with NF are typically a late sign. The initial erythema of the skin, which can easily be confused for cellulitis, becomes increasingly purple before finally turning dusky blue. Necrosis then begins to develop and haemorrhagic bullae may form [2]. Its mostworrying complication is systemic shock which can lead to Multiple Organ Dysfunction Syndrome and death. With NFB the symptoms in their early stages can often mimic breast cellulitis, with difficulty in distinguishing clinically between the two.
The main principles of treatment are broad-spectrum antibiotics with debridement of the necrotic tissue. Traditionally the diagnosis is made clinically with little use of radiological imaging.
The following case report is written in accordance with the Surgical Case Report (SCARE) 2018 guidance to increase transparency and accuracy in case report publication [5]. Consent was achieved from the patient to publish this case report.

Case presentation
A 54 year-old lady with a background of poorly controlled tabletcontrolled type II diabetes mellitus (HbA1c = 108) and obesity was admitted with an atraumatic, painful right breast lump which had been present for five days to a medium-sized district general hospital. The lump was becoming increasingly painful and she had recently developed fevers. She was initially treated as a breast abscess with IV flucloxacillin due to raised inflammatory markers, with a white cell count (WCC) 15.25 and CRP 301.5. An USS was arranged which showed a "superficial infection with surrounding oedema but no collection or abscess". On the second day her WCC began to fall with intravenous antibiotic therapy but her CRP continued to rise with her pain intensifying and erythema spreading. Her antibiotics were therefore changed to include IV clindamycin on microbiology advice.
On day four of admission a repeat USS was requested due to worsening erythema. This showed "spreading skin thickening and oedema of the whole breast skin" but no demonstrable collection. Her blood results returned later that day with a CRP of 611 and an acute kidney injury. In light of the blood results and her deteriorating condition she was reviewed by the breast team. Her LRINEC score (a calculator of probability of NF) was calculated as 9 [19] and an urgent CT thorax was booked. This was completed within an hour and demonstrated a "large volume of interstitial soft tissue gas throughout the right breast with diffuse fat stranding" (see Fig. 1). This was thought to be consistent with necrotising fasciitis of the breast and once clinically stabilised, she was consented and taken to theatre for urgent debridement of the necrotic tissue by two breast consultants.
Intraoperatively, necrotic fat was found in the right lower inner quadrant extending to the subareolar region. This did not extend to the lateral breast or the abdominal wall and the pectoral muscle was healthy. Microbiology swabs were sent which came back positive for Enterococcus faecalis sensitive to amoxicillin and pieracillin/tazobactam (tazocin). This is consistent with Type I necrotising fasciitis. The histological specimen showed extensive necrosis and purulent inflammation in keeping with necrotising fasciitis. The wound was initially left open with a planned re-look two days later. Post-operatively she was transferred to ITU. Her antibiotics were changed to IV tazocin and clindamycin on microbiology advice, with the tazocin continuing for fourteen days in total. In total, she spent seven days on ITU requiring vasopressor support and temporary dialysis as well as ventilatory support.
Her second look operation, two days after the first, involved a washout plus excision of some medial necrotic skin. The wound was packed and the skin left loosely closed. She returned for a third look operation six days after the initial operation which showed no further necrotic tissue so the wound was washed and the skin was closed using an abdominal advancement flap (Fig. 2).

Discussion
Necrotizing fasciitis rarely develops in the breast, with its more common sites being the scrotum, abdominal wall, extremities and perineum. In the case described above it is the authors' consideration that the bacteria selectively infected Cooper's suspensory ligaments of the breast, leading to the delay in skin changes and relative preservation of both the superficial and deep fascial layers.  It is also likely that this lady's poorly controlled T2DM contributed to her presentation, leading to an immunocompromised state, predisposing her to spontaneous infection.
Due to the rarity of the condition, the diagnosis is often delayed. This results in an unclear management plan with uncontrolled disease progression prior to debridement. The LRINEC score can be calculated to aid in the probability of a case being NF [19]. A golden six point management plan for NFB was recommended by Shah et al. [8], which followed the management plan set out by Ward et al. [20] for general management of necrotising fasciitis.
(1) Early surgical referral for disproportionate pain and cellulitis (2) Broad-spectrum antibiotics and intravenous fluid resuscitation (3) Exploratory diagnostic incision to inspect the fascia (4) Radical debridement (5) Re-exploration of the wound within 24 h of first surgery (6) Involvement of plastic surgeons for reconstructive options.
The literature search suggests that both medical and surgical management is consistently the same with use of empirical antibiotics and debridement, but that problems arise in the diagnosis stage. Notably the use of imaging was incredibly variable with only three cases (23%) using any form of diagnostic imaging. Imaging is not mentioned by Shah et al. [8], most likely because CT imaging was not as widely available as it now is. This may also be due to a clearer initial diagnosis due to late presentation. In cases where the initial diagnosis is unclear this leads us to believe that further cross-sectional imaging, such as a CT scan, would allow prompt diagnosis of NFB. In our case described above this proved to be the diagnostic feature as our patient did not have the characteristic skin changes classically described in the literature. Reviewing the cases where imaging was used it often helped to confirm the diagnosis and allow a targeted approach to debridement. Marongiu et al. [16] initially diagnosed their patient with breast mastitis but changed their diagnosis once a CT scan was performed. This led to prompt debridement of the infected area. In the case of Keune et al. [17] it identified the extent of necrotising tissue and an initial more conservative debridement. This was ultimately unsuccessful and the patient had to have a mastectomy but it allowed the opportunity for a potentially less invasive surgery first of all.
We therefore think it is important that surgeons and clinicians hold a high index of suspicion when a patient presents with fever and disproportionate pain in the breast so that in these scenarios early cross-sectional imaging is carried out to ensure optimal Consequentially we have created a simple diagnostic (see Fig. 3) and management triad for NFB (see Fig. 4) which is a modification of the 6 point management plan devised by Shah et al. [8].

Conclusion
In conclusion, we believe that necrotising fasciitis of the breast is a very challenging diagnosis for a clinician to make due to its rarity and similarity with a simple breast infection. Clinicians must hold a high index of suspicion if pain is disproportionate to the signs or a painful swollen breast co-presents with sepsis. Further to this we believe that this case highlights the importance of urgent cross-sectional imaging of the breast, if there is doubt over the diagnosis, to avoid any delay in treatment and to allow a more targeted approach to the subsequent debridement.

Sources of funding
None.

Ethical approval
Exempt from ethical approval.

Consent
Written informed consent was obtained from the patient for publication of this case report and accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.

Author's contribution
Dr Bertram Marks -wrote the introduction, literature review, case report and helped edit and write the discussion.
Miss Tarannum Fasih -helped edit the introduction and write the discussion as well as coming up with the clinical diagnostic and management triads.
Mr Sunil Amonkar -helped edit the case report. Mr Mujahid Pervaz -helped edit the case report.

Registration of research studies
N/A.

Guarantor
Dr Bertram Marks and Miss Tarannum Fasih.

Provenance and peer review
Not commissioned, externally peer-reviewed.

Declaration of Competing Interest
No conflicts of interest.