The 5^th century B.C. renowned physician, Hippocrates, vividly described necrotizing fasciitis as a complication of erysipelas [8]. Erysipelas has been defined as an infection of the upper dermis and superficial lymphatics as opposed to cellulitis, which is in the deeper dermis and subcutaneous fat; especially in Europe, these are synonymous terms [9]. Both cause rapidly spreading erythema, swelling, pain, and edema that may progress to vesicles, bullae and ecchymoses.

Loudon provided a detailed description of necrotizing fasciitis, the “once-see-never-forgotten-disease,” as it traversed through Europe in the 18^th and 19^th century, especially during the wars, (Peninsular, Battle of Waterloo, Napoleonic, Crimean), and on transport ships, where it was known as phagedena/hospital gangrene [10]. Although seen mainly in the military, “phagedena” did occur in England and Wales and had a high mortality of 30,475 patients in 1863 during an epidemic of scarlet fever, and another smaller occurrence in 1893 (78 patients) when there was an increase of deaths due to erysipelas and puerperal fever (uterine postpartum infection) [10]. The first English description of NSTIs was by three British physicians: Leonard Gillespie, Gilbert Blaine and Thomas Trotter, who called these infections phagedaenic ulcer, phagedena gangrenous, hospital gangrene, etc. in the late 18^th century [11].

In 1871 during the American Civil War, an Army surgeon, James Jones produced a treatise on hospital gangrene, reporting on 2642 cases with a 46% mortality [12] [13]. In World War I, the Belgian physician, Antoine Depage and others debrided the necrotic tissue and delayed primary closure; this maneuver along with newly developed antimicrobial interventions increased survival [14]. Mortality rates from gangrene decreased from 28% in World War I to 15% in World War II (1939-1945) [13]. As a result of rapid evacuation from the battlefield and antibiotics, none of the American soldiers developed gangrene during the Vietnam War [14]. The Joint Theater Trauma Registry (TTTR) established as a central repository of clinical casualty injury data in Operation Iraqi and Enduring Freedom reported that of the 562 completed records (2003-2006), there was only one patient with an ICD-9 code for necrotizing fasciitis [15]. Manring et al. compiled an interesting review on the treatment of war wounds throughout the centuries [16]. Although frequently quoted whenever an article about NF or NSTI is published, few reports indicate the relevance of the Meleny and Wilson observations about this disease process.

Historically, Wilson and Meleny were the most instrumental in identifying and describing NF and NSTI as we know them today. In 1924, Meleny documented the connection between 19 cases of β hemolytic Streptococcus (19 cases) and gangrene in the extremities and trunk [17]. Wilson defined the disease severity and characteristics that required an acute observation, and a rapid decision to provide surgical debridement, removal of necrotic subcutaneous tissue, and fascia of the afflicted body part [18]. Wilson compared his own patients to those seen by Meleny, and in 1952 published a description of the clinical presentation and treatment of necrotizing fasciitis, Table 1. Current patient diagnostic parameters and treatment have not changed significantly from those patients encountered by Wilson and Meleny, especially the necessity for rapid surgical intervention at the time of admission.

Many of the breast NF and NSTI case reports in the literature credited Shah et al. (2001) with the first report of NF occurring in the adult female breast [19] There were, however, earlier reports in 1924 by Probstein of an NSTI reported as gangrene of the breast [20]. In their case report, Shah et al. did not describe fascial involvement of the pectoral muscle or chest wall (the pre-operative diagnosis was necrotizing fasciitis): “… On gross examination the breast showed two haemorrhagic areas, each up to 30 mm across. Microscopically there was extensive necrosis of the breast tissue, subcutaneous fat and skin. Around the margin of necrosis, a rim of acute inflammation was present with associated focal vasculitis. Throughout the necrotic tissue numerous bacteria were seen without attendant inflammation; these were positive on Gram staining (Figure 1). The appearances were typical of necrotising fasciitis.” [19] In the discussion, NF was

Median (IQRT) = (lower 25% interquartile range, upper 75% interquartile range); Number (percent) = # (%); Meleny: 19 of 23 patients had cultured Streptococcus; Wilson: 5 (26%) had cultured Streptococcus, 8 (42%) had Staphylococcus, and 6 (32%) had both Streptococcus and Staphylococcus. Some patients in the literature also complain of numbness of the breast, as noted by the diminished sensation above.

described as “a rare but serious progressive infection characterized by widespread necrosis of subcutaneous tissue and fascia with associated systemic toxicity.” [19] This report, however, did note that: “Diagnostic incision of the affected site to inspect the underlying fascia should be performed at the earliest opportunity, with pus sent for urgent Gram stain and culture.” [19] The keen observations of a case with NSTI defined the characteristics of necrotizing soft tissue infections to which NF belongs. It was emphasized in this case report that patients with breast NF and NSTI required the same rapid clinical observations and treatment as NF in other sites; otherwise, the rapid spread of an infection could evolve into an NF and increase mortality [19]. For the diagnosis of breast NF, it is important to document whether the breast and fascia of the underlying muscles and chest wall are affected by the infection or necrosis. If the fascia is not involved, then the diagnosis is a necrotic soft tissue infection (NSTI) of the breast. Since 1924, there have been abbreviated literature reviews with multiple case reports. The most recent NF and NSTI breast reviews have limited selected cases and time periods: Cai et al. with 40 cases between 1997-2018 and Konik et al. with 25 primary breast cases reported between 2001-2017 [21] [22]. Our literature review included all the cases reported in these two reviews; in addition, our review encompassed 1924 through 2021, and comprised patients with primary and secondary breast involvement: females, males, infants, in addition to women with lactational mastitis, abscesses, and surgical cases that progressed to NSTI and NF.

There were 123 cases gleaned from 96 articles (1924 through 2020) that reported NF and NSTI of the breast. Literature searches were conducted through PubMed of the National Library of Medicine, Google Scholar, and from article reference sections [5] [6] [9] [10] [19] [20] [21] - [117]. All accessible English, Spanish, German, and French printed journal articles as case reports or series were identified if they pertained to breast gangrene, breast necrosis, necrotizing fasciitis of the breast, and necrotizing soft tissue infections. The specific terms searched for were: breast gangrene, necrosis, NF, NSTI, synergistic gangrene of the breast, and NF of the chest wall. Case reports which did not have “necrosis” or “gangrene” reporting abscesses, ulcerations, surgical soft tissue infections were not included in the title or case description. As noted in Figure 1, more than half of the case reports were published within the past decade. The landmark article recognized for NF was by Wilson in 1952, and for breast NF, Shah in 2001 [17] [19]. There may have been earlier publications than in 1924 about breast NF and NSTI, but they were not included in this literature review.

The major specialty journals were: 34 (35.4%) Medicine (included Pharmaceuticals, Radiology, Tropical Medicine); 30 (31.3%) Surgery (included Plastic Surgery and Trauma); 10 (10.4%) in Breast; nine (9.4%) in Case Reports; four (4.2%) each in Pediatrics, Infectious Diseases, and Obstetrics/Gynecology; and one (1%) in Pathology. Most of the articles originated from the United States 21 (22%); the United Kingdom 16 (16.7%); India 15 (15.6%); four (4.2%) each from Brazil, Korea, France; three (3.1%) from Australia; two (2.1%) each from Japan, Nigeria, Spain, Switzerland, Italy, Singapore, Saudi Arabia; and one (1.0%) each from Germany, Belgium, Netherlands, Canada, Colombia, Argentina, China, Nepal, New Guinea, Morocco, Oman, Yemen, Kuwait, Egypt, and Israel.

The six study inclusion criteria for accepting a case as NF or NSTI have been the following variables: 1) patient age; 2) gender; 3) surgery or other; 4) tissue “necrosis or gangrene;” 5) presence or absence of fascial disease; and 6) mortality. For the diagnosis of necrotizing fasciitis, documentation of fascial involvement either through the surgical procedure, operative report, or histopathologic examination was required. Reports with incomplete or no reference to fascial involvement (that the fascia of muscles or chest wall were affected) were placed into the NSTI group. Figure 2 depicts the different articles titles in case reports that described either NF or NSTI of the breast and are part of this literature review. The article titles did not always reflect the appropriate disease state reported.

Statistical analysis was performed with Statistica^® (StatSoft, Tulsa, OK) descriptive statistics, one-way ANOVA, in post hoc comparisons, the Tukey test for unequal numbers was used. The nonparametric Mann Whitney U test was used for numeric results. Maximum Likelihood chi-squared was used for categorical variables. Patient demographics included: age, gender, primary or secondary involvement, NF, NSTI, location, presentation, admission sepsis, chest wall and

muscle involvement, source and location of initial injury, laterality, surgery, mastectomy, debridement, initial white blood count, C-reactive protein, sodium chloride, creatinine, glucose, hemoglobin, platelets, microorganism (Type I polymicrobial and Type II monomicrobial), admission values for blood pressure, tachycardia, temperature, microorganisms, antibiotics, symptom duration prior to admission, length of stay (LOS), days in the intensive care unit (ICU), days ventilated, operations, reconstructive procedures, death, negative pressure wound therapy (NPWT), and hyperbaric oxygen therapy. Included were comorbidities such as diabetes, obesity, renal failure, hypertension and cardiovascular issues. Due to the small numbers, only the basic statistical analyses were performed. Primary breast NF was defined as the presenting disease process. Secondary breast NF was defined as breast tissue involvement, due to proximity of the spreading necrotic tissue, muscle, and fascia spread from an adjacent anatomical region. Comparisons were made between the NF and NSTI groups in terms of patient demographics, treatment, Type I and Type II infections, chest wall and muscle involvement, microorganisms and antibiotic use. The study was approved by the Institutional Review Board.

Although there were more female patients, both sexes presented with similar sign and symptoms. In the NF group, male patients had longer LOS (median 44 versus 20 days, P = 0.002); ICU (median 45 versus 8 days, P < 0.0002); higher number of operations (3.5 versus 2.0, P = 0.005); and lower BMI indices (median 27 versus 35, P = 0.03) than the female patients. In contrast, 9 of 12 men (75%) were more likely to have secondary breast involvement than the women (9 of 111, 8%). There was no significant difference in mortality, which was lower than that reported in other locations of NF and NSTI in the literature.

The microorganisms of breast NF and NSTI were the same as seen in other locations. Table 4 lists the most frequent microorganisms: β hemolytic Group A Streptococcus (GAS), Staphylococcus aureus and Pseudomonas aeruginosa. Type I infections were polymicrobial and had the full gamut of aerobic and anaerobic species seen (Streptococcus, Staphylococcus, Proteus, Candida, et cetera). Type II infections were usually β hemolytic Group A Streptococcus, Staphylococcus aureus, and methicillin resistant Staphylococcus aureus (MRSA). Unusual microorganisms were frequently listed in the article titles. A virulent deadly strain of Klebsiella species with a 40% mortality was seen in Asia and other areas; it was not reported in patients with breast NF or NSTI, but rather in other body locations [116] [117]. The preoperative antibiotics for breast cases were documented in 50 (56.2%) NF and 39 (43.8%) NSTI; postoperative antibiotics were recorded in 59 (60.2%) NF and 39 (39.8%) NSTI cases, Table 5. There

Number (percent) = # (%); GAS = β hemolytic Group A Streptococcus; NF = necrotizing fasciitis; NSTI = necrotizing soft tissue infection.

Number (percent) = # (%); GAS = β hemolytic Group A Streptococcus; NF = necrotizing fasciitis; NSTI = necrotizing soft tissue infection.

were multiple combinations of antibiotics used on admission: penicillin, piperacillin, vancomycin, clindamycin, amoxicillin-clavulanic acid, metronidazole, and gentamicin. The most frequent antibiotics administered after culture and sensitivities were: clindamycin, metronidazole and vancomycin. There was also a variety of other antibiotics and fungicides. Fungi were rare: there were only a few cases of Candida albicans and Fusarium species.

During this review, there were several other disease processes that had necrosed breast tissue and gangrene: anticoagulant-induced gangrene of the breast, pyoderma gangrenosum, purpura fulminans, and Sweet syndrome. Flood et al. described a 40-year-old female who developed gangrene of the right breast several days after starting Dicumarol for thrombophlebitis of the right saphenous vein in her right leg [118]. The patient required debridement of the necrotic breast tissue; pathology revealed hemorrhage, thrombosed veins with many leukocytes and fibroblasts [118]. This case was considered thrombophlebitis migrans dissemina because of the involvement of the lung and extremity [118]. Although rare, a number of these cases with necrosis of the breast have been reported after treatment with warfarin (coumadin) [119].

Other conditions that presented as NSTI were pyoderma gangrenosum cases, a “rapid spreading ulceration and necrosis similar to synergistic gangrene.” [120] A patient presented four days after a breast reduction with fever, painful, swollen breasts and erythema not affecting the suture lines; with no response to antibiotics, she responded quickly to immunosuppressive treatment with cyclosporin or tacrolimus [120]. Pyoderma gangrenosum, a neutrophilic dermatosis, is a hallmark of pathergy and can occur spontaneously or after slight injury; since it can present as ulceration after breast surgery, it must be distinguished from NF and NSTI; the treatment is steroids or immunosuppressive therapy rather than debridement or mastectomy [121] [122]. It usually occurs after bilateral mammoplasty or autologous reconstruction [121] [122].

Gesakis et al. reported a 45-year-old woman who presented with a septic picture and an enlarged left breast with erythema, blisters, necrosis and tissue air on ultrasound [123]. The LRINEC score was 8, and she underwent a mastectomy. Histopathologic examination of the tissue revealed primary spindle cell sarcoma of the breast [123]. Previous radiation treatment, especially for breast cancer and non-Hodgkin’s lymphoma, is the main risk factor for developing breast sarcomas; this patient did not have radiation therapy previously [123].

Young et al. reported on Sweet syndrome, which was a case of breast hidradenitis with multiple breast abscesses and mislabeled as NF [124]. Necrotizing Sweet syndrome (a variant of neutrophilic dermatosis) was described as an NSTI in four patients; one case involved the pectoralis major muscle in a transsexual male and three others had their disease manifest in the lower extremity [125]. [126]. This condition manifested pathergy, mimicked NF and was successfully treated with corticosteroids and cyclosporin [125] [126].

Table 6 indicates the laboratory values that were frequently reported. Most

Median (IQRT) = (lower 25% interquartile range, upper 75% interquartile range); statistics were based on 20 or more cases reporting laboratory results. NF = necrotizing fasciitis; NSTI = necrotizing soft tissue infections.

patients had a neutrophilia, C-reactive protein, BUN, and creatinine. Glucose levels were reported mainly for the patients with diabetes; of the five HgA1c levels reported (median 10.8%), four were in the NF group. The literature laboratory results were few with mainly the WBC count reported, 38% of the time for NF and 18% for NSTI. With clinically painful, swollen, gangrenous, and discolored breasts (with and without sepsis), most cases reported rapid diagnosis, surgical intervention within 24 hours of presentation, and broad-spectrum antibiotics for adequate treatment and survival.