Shuichi Kuniyuki, Yuki Yoshida, Naoki Maekawa and Kazuhoshi Yamanaka

Department of Dermatology, Osaka City General Hospital, Osaka, Japan

The aim of this study was to determine whether bacterial infection plays a significant role in the inflammatory process of epidermal cysts. Samples from 152 patients (115 cases of inflamed and 37 of uninflamed epidermal cysts) were subjected to aerobic and anaerobic bacterial culture and the isolates were investigated. The rate of bacterial growth and the recovered anaerobes were significantly greater in the inflamed than the uninflamed epidermal cysts. However, it is difficult to determine whether recovered isolates from epidermal cysts represent ”infection” or ”colonization”. In conclusion, this study revealed the predominance of anaerobes in inflamed cysts, strongly suggesting that anaerobes play a role in the inflammatory process. Key words: epidermal cysts; bacterial culture.

(Accepted July 3, 2007.)

Acta Derm Venereol 2008; 88: 23–25.

Shuichi Kuniyuki, Department of Dermatology, Osaka City General Hospital, Miyakojima-Hondori 2-13-22, Miyakojima-ku, 534-0021, Osaka, Japan. E-mail: kuniyuki@ocgh.hospital.city.osaka.jp

Epidermal cysts are closed sacs with a definite wall that results from proliferation of surface epidermal cells. Production of keratin and lack of communication with the surface are responsible for cyst formation. Epidermal cysts can become infected, and when an abscess develops they may need to be surgically drained, followed by administration of systemic antibiotics. It is a matter of debate whether bacterial infection plays a role in the inflammatory process. Some authors reported that 2 groups of inflamed and uninflamed epidermal cysts did not differ significantly with respect to the number of bacterial isolates and aerobes, anaerobes or potential pathogens cultured (1, 2). In other reports, Staphylococcus aureus (3, 4) and indigenous anaerobes (4, 5) were predominantly isolated from inflamed epidermal cysts, suggesting that they may have played a role in the inflammatory process (3–5).

In this study, we obtained aerobic and anaerobic bacterial culture specimens from inflamed and uninflamed epidermal cysts. We then investigated the positive rate of culture growth and species of recovered isolates in order to determine whether bacterial infection plays a significant role in the inflammatory process of epidermal cysts.

Subjects and Methods

Subjects

The subjects were 152 patients clinically diagnosed with epidermal cysts at Osaka City General Hospital and Osaka City Juso Hospital between January 1998 and December 2005. They were non-immunocompromised adult patients with no history of recent systemic antibiotics. There were 115 patients with inflamed epidermal cysts (72 men and 43 women) aged 18–79 years (mean age 43.3 years) and 37 patients with uninflamed epidermal cysts (20 men and 17 women) aged 24–83 years (mean age 51.1 years).

Methods

An inflamed epidermal cyst was defined as a known cyst that developed a fluctuant soft-tissue swelling surrounded by erythema and contained a localized collection of purulent material. An uninflamed cyst was defined as an intradermal nodule with no evidence of inflammation. Each cyst was incised with a sterile surgical blade and then a sterile swab was inserted into it. After written informed consent was obtained, samples were collected by the swab method, before treatment with antibiotics. In the aerobic culture, samples were cultured on sheep blood agar plates and chocolate agar plates in 5.5% CO2 at 35ºC for 3 days. Colonies formed were identified based on colony morphology, Gram stain, and catalase and coagulase productivity. In the anaerobic culture, samples were immediately streaked on 7% sheep blood agar prepared with Brucella (Kyokuto Pharmaceutical Co., Tokyo, Japan) and incubated at 35ºC for 48 hours in an anaerobic jar with an atmosphere of 5% CO2 + 95% N2.The organisms grown anaerobically under such conditions were then identified using Rapid ANA II System (for biochemical identification of medically important anaerobic bacteria, RE-MEL Inc., Norcross, GA, USA) and Rapid ANA II Code Compendium (6). Institutional ethical review boards at the Osaka City General Hospital and Osaka City Juso Hospital approved the study protocol.

Results

Of the 115 inflamed cysts, 45 were from the head and neck, 38 from the trunk and 27 from the buttocks and inguinal area, whereas of the 37 uninflamed cysts, 16 were from the head and neck, 15 from the trunk, 3 from the gluteal and inguinal regions and 3 from the extremities (Table I).

Of the 115 inflamed cysts, 91 (79.1 %) yielded bacterial growth (a total of 116 isolates). Of the 37 uninflamed cysts, 20 (54.1%) cysts yielded bacterial growth (a total of 21 isolates), showing that isolates were significantly more frequently recovered from inflamed cysts than uninflamed cysts (χ2 test, p=0.003).

Aerobic bacteria were recovered from 67 of the 115 inflamed epidermal cysts and 19 of the 37 uninflamed cysts, with no significant difference (χ2 test, p=0.461). The predominant aerobes from the inflamed cysts, were coagulase-negative staphylococci (CNS) (43 isolates, including 18 S. lugdunensis, 6 S. epidermidis and 19 other CNS), followed by Corynebacterium spp. (11 isolates) and S. aureus (9 isolates). From the uninflamed cysts, 13 CNS isolates (2 S. lugdunensis, 5 S. epidermidis and 6 other CNS) and 4 Corynebacterium spp. were frequently isolated (Table II). S. aureus was more frequently isolated from the inflamed cysts than uninflamed cysts, but no significant difference was found (Fisher’s test, p=0.114)

Anaerobes were isolated from 35 (30.4%) of the 115 inflamed epidermal cysts and 2 (5.4%) of the 37 uninflamed epidermal cysts, showing that anaerobes were significantly more frequently isolated from the inflamed cysts (χ2 test, p=0.002). Peptostreptococcus spp. was the most frequently isolated from the inflamed cysts (26 isolates: 13 P. magnus and 9 P. asaccharolyticus), followed by Propionibacterium spp. (5 isolates), Bacteroides spp. (5 isolates) and Prevotella spp. (3 isolates). In contrast, only 2 isolates of P. magnus were isolated from the uninflamed cysts (Table II).

The distribution of the aerobic and anaerobic isolates, including CNS, Corynebacterium spp., S. aureus, Peptostreptococcus spp. and Propionibacterium spp. was equal in all the anatomic sites (Table III). However, 3 of 5 isolates of Bacteroides spp. were isolated from the gluteal region.Discussion

Epidermal cysts occasionally become inflamed, with an erythematous soft tissue swelling containing a purulent material. In surgical drainage, a purulent cheesy discharge is excreted. This inflammatory process may not be a simple bacterial skin infection; however, micro-organisms may have some influence on the inflammation mechanism of epidermal cysts. We have studied the bacterial isolates from inflamed and uninflamed epidermal cysts using techniques to recover anaerobic and aerobic bacteria. The number of recovered isolates and positive rate of bacterial growth in the inflamed group were significantly more frequent than in the uninflamed group, and the anaerobes were significantly more frequently recovered from inflamed cysts.

Diven et al. (1) investigated 25 inflamed and 25 uninflamed epidermal cysts, and Ohata et al. (2) investigated 40 inflamed and 32 uninflamed epidermal cysts, and they both concluded similarly that the dominant recovered isolates from both groups were CNS, Peptostreptococcus spp. and Propionibacterium spp., while the 2 groups did not differ significantly with respect to number of isolates, “no growth” cultures, and aerobic or anaerobic organisms cultured. They were unable to discern any pattern to the types of organisms or the quantities cultured from inflamed and uninflamed cysts (1, 2). In contrast, others reported that S. aureus (3, 4) and some anaerobes, such as Prevotella, Peptostreptococcus spp. and Bacteroides spp. (3–5), were more frequently isolated from inflamed cysts. Brook (3) investigated 231 infectious epidermal cysts, and isolated 153 isolates of aerobes (81 isolates of S. aureus and 7 isolates of CNS) and 162 isolates of anaerobes (55 isolates of Bacteroides spp., 85 isolates of Peptostreptococcus spp., and 7 isolates of Propionibacterium spp.), suggesting the involvement of S. aureus in inflammation (3, 4). Nishijima et al. (5) investigated 53 and 32 inflamed and uninflamed epidermal cysts, respectively, and found significant isolation of Propionibacterium, Prevotella, Veillonella and Bacteroides from the inflamed epidermal cysts (5).

Anaerobes were isolated significantly frequent from the inflamed, rather than from the uninflamed, epidermal cysts. Peptostreptococcus spp. accounted for 64% of anaerobic isolates, and was similarly predominant in other reports (57% (5), 56% (2) and 52% (3)). In our study, P. magnus and P. asaccharolyticus accounted for 33% and 21% of anaerobic isolates, respectively. Brook & Frazier (7) reported that 133 P. magnus and 84 P. asaccharolyticus isolates were found among 1260 isolates recovered from bacterial skin infections, suggesting that these Gram-positive cocci are important in the induction of skin soft tissue infection and inflammation of bones, joints and soft tissues (3, 8, 9). Similar to CNS, a cell wall component of gram-positive cocci, peptidoglycan, may act as a superantigen and induce the production of inflamed cytokines, including interleukin (IL)-1β and tumour necrosis factor (TNF)-α, leading to inflammation.

Other than Peptostreptococcus spp., a significant isolation of Bacteroides spp. (B. melaninogenicus and B. fragilis) from inflamed epidermal cysts was reported by Brook (3), and Propionibacterium, Prevotella, Veillonella and Bacteroides by Nishijima et al. (5). Higaki et al. (10) reported the predominance of Propionibacterium and Prevottella in an investigation of numerous bacterial skin infections (10). Lipopolysaccharides contained in the outer cell membrane of gram-negative bacilli, such as Bacteroides, Fusobacterium and Veillonella, have endotoxin-like properties, and exhibit pathogenic features (11). P. acnes releases lipase, which may produce free fatty acids, and protease, which may destroy hair follicle walls, thus contributing to the process of inflammation (12).

This study highlights the predominance of anaerobes in inflamed epidermal cysts, strongly suggesting that anaerobic bacterial infection may play a significant role in the inflammatory process (5). Questions regarding pathogenesis and appropriate treatment, including the proper selection of antimicrobial therapy, should be addressed in additional studies.

References

1. Diven DG, Dozier SE, Meyer DJ, Smith EB. Bacteriology of inflamed and uninflamed epidermal cysts. Arch Dermatol 1998; 134: 49–51.

2. Ohata C, Komatani M, Shirabe H, Takagi K, Kawatsu T. Bacteriological analysis of epidermal cysts, Skin Research 1996; 38: 305–309.

3. Brook I. Microbiology of infected epidermal cysts. Arch Dermatol 1989; 125: 1658–1661.

4. Brook I. Microbiology and management of infected neck cyst. J Oral Maxillofac Surg 2005; 63: 392–395.

5. Nishijima S, Higashida T, Oshima S, Nakaya H. Bacteriology of epidermoid cysts. Jpn J Dermatol 2003; 113: 165–168.

6. Celig DM, Schrechenberger PC. Clinical evaluation of the rapid ANA II panel for identification of anaerobic bacteria. J Clin Microbiol 1991; 29: 457–462.

7. Brook I, Frazier H. Aerobic and anaerobic bacteriology of wounds and cutaneous abscesses, Arch Surg 1990; 125: 1445–1451.

8. Murdoch DA. Gram-positive anaerobic cocci. Clin Microbiol Rev 1998; 11: 81–120.

9. Bourgault AM, Rosenblatt JE, Fitzgerald RH. Peptostreptococcus magnus: a significant human pathogen. Ann Intern Med 1980; 93: 244–248.

10. Higaki S, Kitagawa T, Morohashi M, Yamagishi T. Anaerobes isolated from infectious skin diseases. Anaerobe 1999; 5: 583–587.

11. Sveen K. The capacity of lipopolysaccarides from bacteroides, fusobacterium and veillonella to produce skin inflammation and the local and generalized Shwartzman reactions in rabbits. J Periodonal Res 1977; 12: 340–350.

12. Valentine MC. Bacteria in epidermal cysts. Arch Dermatol 1990; 126: 1103.