The management of the shoulder skin microbiome ( Cutibacterium acnes) in the context of shoulder surgery: a review of the current literature

Abstract

Background

This review aims to establish current knowledge of the shoulder skin microbiome and how to manage the bacteria that reside within it.

Methods

A review was undertaken of the current literature through OvidSP. All abstracts were reviewed by three independent researchers.

Results

Thirty-five studies met the inclusion criteria. With forward referencing an additional 14 were included. None commented on organisms specific to the shoulder microbiome other than Cutibacterium acnes. Therefore, this review is focussed on the current knowledge of C. acnes.

Discussion

C. acnes is a skin commensal within the pilo-sebaceous glands reported to be the primary pathogen in up to 86% of shoulder joint infections. Pre-operative culture of unprepared skin can be indicative of underlying joint infection in shoulder arthroplasty revision. Intra-articular biopsies may have a high false positive due to skin contamination. Correlating the number of positive samples and certain associated signs can give a greater than 90% probability of a true infection. Standard surgical skin preparation, peri-surgical intravenous antibiotics and oral pre-operative antibiotics do not reduce bacterial load within the skin. However, topical benzoyl peroxide and clindamycin have both demonstrated significantly reduced bacteria load. Phylogenetically there are six main types. Patients may have more than one phenotype present during infection.

Keywords

shoulder skin microbiome infection

Introduction

Study of the microbiome of the human body is increasingly yielding new causes of disease, the aetiology of which had previously been poorly understood. The role of Helicobacter pylori in peptic ulceration resulted in a complete revolution in the management of that condition.¹ Similarly, developments in the past decade evaluating infection of shoulder arthroplasty have demonstrated a frequency of Cutibacterium acnes as a common causative organism.² One recent systematic review of rates from arthroscopic surgery demonstrated a 37.3% positive biopsy rate for C. acnes corresponding to a 0.22% clinical infection rate.³ Prior to this C. acnes was not a recognised pathogen within the shoulder and there may be other low virulent organisms from the skin of the shoulder area that are still unrecognised.⁴

An infection is a potentially devastating complication of surgery. Shoulder joint infections with low virulent organisms are difficult to diagnose, often requiring debridement, implant revision, multiple operations and prolonged antibiotic therapy in order to successfully eradicate the infection. Therefore, reducing this risk should be a priority for the patient. It is also important to be able to interpret biopsy results and have a knowledge base to manage patients with low virulent organism shoulder infections.

As the most likely source of a deep surgical infection is from the skin flora or microbiome, knowledge of the normal microbiome is an essential first step to understanding and treating potential pathogens.⁵

Therefore, the aim of this review was to present the current literature relating to the known microbiome of the skin overlying the shoulder and to evaluate the best management of these organisms when they are encountered in surgical shoulder practice.

Methods

A systematic review of all available literature was undertaken in keeping with Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.⁶ MEDLINE, EMBASE and Cumulative Index to Nursing and Allied Health Literature databases were searched through OvidSP. The search strategy was kept broad to ensure capture of all available literature. The MeSH terms (shoulder AND Skin AND Infection) or (shoulder AND microbiome) or (shoulder AND skin AND flora) or (shoulder AND skin AND colonisation) were used on 27 April 2020. The references for each study were scrutinised to ensure other relevant studies not initially highlighted could be identified and duly included in the review. Forward referencing was also conducted in Google Scholar. Duplicates were identified and if papers were found to include the same patient sets, only the more recent was included for review.

Eligibility criteria

The only inclusion criterion was studies reported in English.

Study selection

All titles and abstracts were reviewed and evaluated by three independent researchers using a purposely designed systematic review programme (Rayyan, QCRI, Qatar Computing Research Institute, Doha, Qatar).⁷ Any disagreements were reviewed by a fourth independent author who was the final arbiter. The remaining texts were then reviewed in their entirety and further screened.

Results

Following the initial search, a total of 813 studies were identified and screened. After exclusions and review for relevance this was reduced to 35 papers. From these papers using forward referencing, a further 14 publications were identified giving a total of 49 articles for review.

The PRISMA flow diagram for the initial search including inclusion and exclusion is shown in Figure 1.

Of the final 49 articles identified all but four were specific to C. acnes. These remaining articles were deemed to contain valuable results pertinent to the review, with three related to prosthetic infection prevention and the final article was specific to a diagnostic technique for anaerobic infection. Our results precluded a review of the wider shoulder skin microbiome and have resulted instead to an analysis of the current knowledge and management of C. acnes.

Discussion

Background

C. acnes, formerly known as Propionibacterium acnes is a Gram positive, aero-tolerant, anaerobic bacillus bacterium frequently found in the skin flora as part of the normal human microbiome.⁸ It is also found in the mucosa of the mouth, nose, urogenital tract and large intestine.⁹ It can cause inflammatory acne vulgaris, shoulder infections and contamination of medical devices.^10,11 It has also been implicated in a variety of other conditions ranging from sarcoidosis and prostate cancer to being a possible cause of osteoarthritis.^12–15

The renaming of the bacteria to Cutibacterium was to allow specific reference to the strain of Propionibacteriacea that inhabits human skin. The genus Propionibacterium also includes Acidipropionibacterium and Pseudopropionibacterium, which are not normally found in human skin.

It resides in the pilo-sebaceous glands deep in the dermal layer of the skin.⁹ There are a larger number of hair follicles and sebaceous glands in the skin of the face, scalp, axilla, chest and back and therefore C. acnes is more prevalent in these areas.¹⁶ Pilo-sebaceous glands are thought to contain as many as 10⁵C. acnes organism per follicle.¹⁷ Known patient risk factors for C. acnes colonisation include male gender and the presence of hair. This may be due to increased testosterone and sebum production.¹⁸ Patients who report loss of hair have fewer C. acnes positive cultures from intraoperative samples in open shoulder surgery.¹⁹ Hair loss is linked to follicular downsizing, so the reduction in pilo-sebaceous glands around the shoulder could represent a step to reduce C. acnes infections.¹⁹

C. acnes has virulence factors that trigger an inflammatory response and promotes bacterial adhesion by forming a biofilm.⁹ This can happen rapidly after contamination in the presence of an implant. Phylogenetically there are six main types (IA1, IA2, IB, IC, II and III).²⁰

Recent advances in DNA sequencing technologies have allowed researchers to detect microbial genome sequences within tissues that were previously thought to be sterile. Qiu et al.²¹ found the presence of DNA from multiple species including Acinetobacter and Oxalobacteraceae in the rotator cuff. Acinetobacter has been found in other parts of the human body including skin and gut, which may have immune-regulatory roles. They found an absence of C. acnes in all shoulder tissues, except skin, reinforcing the theory that C. acnes infections are derived from skin contamination during surgery and not from the opportunistic expansion of resident C. acnes in the shoulder joint.²¹

Biopsy results

Several studies report an incidence of C. acnes found on biopsies taken from shoulder joints that had no previous shoulder surgery.^2,22 One found no correlation between previous steroid injection and positive biopsy results and so concluded their findings represented either false positives due to contamination or inadequate skin preparation, or may have represented colonisation.²² Another study suggested that previous injection strongly correlated with positive biopsy results, resulting in some equivocation as to the cause of such a positive biopsy result.² Results like these make it more complicated in interpreting patients with shoulder pain and a potential false positive result after surgery, especially as the consequences of treating a false positive are not benign.²²

Molecular epidemiologic information using multi-locus sequence typing, ribotyping and polymerase chain reaction (PCR) have been used to predict the association of C. acnes subtypes with certain diseases and infections.²³ As such, certain strains have been identified as being more likely to be pathogenic. A study by Johnson et al. showed that phylogroup type IA2 strains had an ability to produce more porphyrin than phylogroup type II. This suggests that type IA2 strains can trigger inflammatory responses as porphyrin can produce reactive oxygen free radicals.²⁴ Types IB and II have been suggested to be the most implicated in orthopaedic peri-prosthetic joint infections (PJIs).²⁰ The haemolytic characteristics of C. acnes could also be a clinical marker for orthopaedic infection.²⁵

The subtypes found have been shown to vary within the same patients. Forty-five percent of 11 patients had more than one subtype within the deep tissue recovered at the time of revision arthroplasty. Therefore, the clonality of C. acnes cultures from deep tissues cannot be assumed and the virulence and antibiotic resistance may require multiple deep specimen testing.²⁶

The biopsy technique may have an impact on the rate of positive cultures. When utilising the Oxford technique for specimen collection one study found only 3.1% of capsular biopsy sampled to be positive.²⁷ This contrasts rates of 17.1–42% in other studies investigating C. acnes presence during shoulder surgery.^15,28,29 The modified Oxford protocol involves obtaining six capsule/synovial specimens as soon as the glenohumeral joint is opened to minimise the risk of contamination. A no-touch technique is used, and separate sterile blades and forceps are used for each new sample. Five samples are sent for microbiology and one for histopathology in formalin.³⁰

Colonisation reduction

The role of topical agents to reduce or eliminate C. acnes has been extensively investigated. It is recognised that standard skin preparation does not decrease the C. acnes bacterial load as conventional agents cannot penetrate the deepest layer of the dermis leaving colonies in the pilo-sebaceous glands unaffected.^31–33 It is thought that shoulder joint inoculation then occurs as the surgical blade disrupts the integrity of the pilo-sebaceous glands as it traverses the dermis.³⁴ This releases C. acnes and all instruments that subsequently pass through this layer or any handling of the soft tissues by the surgeon may transmit the pathogen deeper and throughout the surgical field.³⁵ Consequently, the reported rate of deep tissue inoculation with C. acnes even after arthroscopy has been reported as high as 20%.³⁶

The use of topical benzoyl peroxide (BPO) has consequently been investigated. BPO is used in dermatology to treat acne vulgaris and is directly bactericidal. Its lipophilic properties permit penetration of the pilo-sebaceous duct. One study investigated its topical use in combination with clindamycin. Sixty-five shoulder arthroscopy patients had skin culture at pre-operative visit with C. acnes colonisation of 47.7%. The patients were then given topical benzoyl peroxide 5% and clindamycin 1.2% (BPO/C) gel to apply to the shoulder every night before surgery (average 2.3 (1–10)). Skin culture was repeated in the operating room before the formal operative skin preparation and a reduction in superficial colonisation to 21.1% of patients was demonstrated. Only 3.1% of patients with the BPO/C treatment had deep tissue C. acnes compared to 19.6% without treatment.³⁷

Another study of 80 patients was randomised to 5% BPO or 4% chlorhexidine gluconate (CHG) for three days before surgery. Shoulders treated with BPO had a statistically significant decrease in C. acnes colony counts compared to CHG.³¹

The bactericidal effects of hydrogen peroxide (H₂O₂) on C. acnes have also been investigated in the laboratory setting. One study investigated timings and concentrations and concluded that topical 3% H₂O₂ for 5 min before surgical skin prep prior to shoulder surgery may eradicate C. acnes from the skin. This was purely an in-vitro study and to date there have been no clinical studies demonstrating effectiveness. Additionally, it is unproven that H₂O₂ definitively penetrates the deep dermis and the pilo-sebaceous glands.³⁸

The use of systemic antibiotic therapy has also been studied. Standard surgical intravenous antibiotic prophylaxis given at anaesthetic induction, prior to the surgical incision, is ineffective at eliminating C. acnes from the surgical wound.³⁹ Whilst tetracycline antibiotics are used in dermatology and primary care for management of acne vulgaris, the administration of oral doxycycline for seven days pre-operatively has not been shown to have any significant impact on rate of colonisation with C. acnes.³⁴

As C. acnes colonises the subdermal and deep dermal layers of the skin, techniques should be used to minimise handling of the subdermal layer. Suggested techniques include changing gloves after dermal incision, working to ensure implants never touch the subdermal layer and repeated preparation of the exposed subdermal layer with an antibacterial agent once the wound is opened and throughout the case with the aim of reducing bacterial load.³⁵ Superiority between agents was not demonstrated for this when comparing 2% CHG/70% isopropyl alcohol, 0.7% iodophor/74% isopropyl alcohol, povidone-iodine (0.75% iodine).³⁵

Based on these findings we recommend pre-operative use of topical BPO 5% as part of a colonisation reduction process. The effect of duration of treatment has not been studied beyond those receiving treatment for one day compared to those receiving greater than one day. Eradication rates were higher with longer treatment (mean 2.3 treatments) suggesting three days minimum treatment may yield most benefit.³⁷

Identification of infected joints?

Identification of true C. acnes joint infection can be diagnostically challenging. Due to its indolence, the clinical features are often vague and non-specific but may include a low level pyrexia or mild soft tissue inflammation in the presence of normal or near-normal inflammatory markers.^40,41 Other patients may simply just present with increased pain and stiffness.^2,42,43 In the context of such implant-related pain, any early radiological loosening suggests infection, although, in many cases, imaging shows no signs of loosening.⁴³

The mainstay of diagnosis of a shoulder joint infection is arthroscopic biopsy tissue samples. Aspirates have not been shown to be reliable.^2,43 Specialised serum and synovial tests such as interleukin-6 and α-defensin are yet to be proven in detecting the likelihood of deep infection with C. acnes.⁴⁴ Biopsies should consist of at least five good quality deep samples, avoiding skin contamination with a volume of 0.5–1 cm.^3,45

It is recognised that cultures for C. acnes require extended enrichment testing. This is required with a chocolated agar medium and anaerobic environment with timeframes for interpretation reported between 8 and 21 days.^5,46 Any form of bacteriological culture can also be complicated by poly-microbial results and with such extended cultures this risk increases. The optimal time for extended culture has not been defined, but a shorter timeframe increases the risk of a false negative and a longer timeframe a false positive. There is also a delay of several days or weeks to identify a positive result, meaning patients have often been discharged home before any result is known. Prophylactic anti-microbial treatment during this time relies on clinical suspicion of infection.

PCR has been proposed as an alternative to diagnose the presence or absence of C. acnes; however, it does not differentiate dead from live cells and once again is associated with a delay to receiving results after the procedure.⁴⁵ Mass spectrometry provides an alternative option for C. acnes diagnosis and demonstrates good sensitivity and the possibility of undertaking genome amplification when compared to PCR. However, prior to undertaking mass spectrometry, colony growth is required and thus it does not provide any quicker result than conventional culture.⁴⁷

Diagnosing joint infection requires correlating positive deep culture with clinical findings. Contamination is a diagnosis of elimination based on one or more positive samples without clinical correlation. Classification systems exist for estimating the probability of a C. acnes infection. Asseray et al.⁴⁸ published the below criteria to differentiate the probability of infection from contamination in a positive culture (Table 1).

Table 1.

Diagnostic strategy for C. acnes osteoarticular infections (probability >90%).

Number of samples positive for C. acnes	Number of associated signs required	Including
≥2	At least one criterion	≥2 Previous surgery
		Presence of orthopaedic implant
		local signs of infection
		Peri-operative infectious findings
1	At least three criteria	≥2 Previous surgery
		Presence of orthopaedic implant
		local signs of infection
		Peri-operative infectious findings
		Inflammatory biological parameters

They considered peri-operative infectious findings as the presence of pus, dirty or necrotic tissues within the joint or loosening of the surgical implant. The mean number of samples taken for each of the 65 patients in this retrospective study was 2.75 (range 1–6). They reported a likelihood of C. acnes infection being greater than 90% with ≥2 positive samples and one criterion or one positive sample and at least three criteria.⁴⁸

One study investigated the correlation between pre-operative culture of unprepared skin and the bacterial load of C. acnes from samples taken from revision shoulder arthroplasty.⁵ Using semiquantitative scores, they found statistically significant correlation between skin C. acnes scores and the deep sample scores (p = 0.004). Therefore, skin cultures taken before surgery may be informative in planning prosthesis exchange, synovectomy, lavage and extended postoperative antibiotic therapy.⁵

Treatment

Where revision surgery for aseptic loosening is undertaken and positive biopsy samples are obtained, the implication is that this represents probable infection. Dual antibiotic therapy for 6–12 weeks with a multi-disciplinary approach and subsequent surveillance after cessation of antibiotics is recommended.⁴⁵

C. acnes is sensitive to various systemic antibiotics utilised in the treatment of bone and joint infection, including beta-lactams, quinolones, tetracyclines, rifampicin and clindamycin. It is typically resistant to macrolides and systemic clindamycin with reported increasing resistance to metronidazole.⁹ One study investigated resistance of cutaneous acnes to a variety of antibiotics, and whilst there was resistance to erythromycin, clindamycin and trimethoprim–sulfamethoxazole, all samples were sensitive to tetracycline and doxycycline.⁴⁹ Combination therapy with multiple antibiotics that have different mechanisms of action can increase efficacy of treatment and reduce the emergence of resistant organisms.⁵⁰

Treatment of definite PJI can be guided by time since implantation. Where infection occurs within four weeks of implantation, treatment with debridement, irrigation and synovectomy with preservation of implants can be considered.⁵¹

The formation of a biofilm makes implant revision mandatory where diagnosis and treatment starts after the first month post implantation. The surgeon should remove all components and cement. There is no consensus favouring single or two stage revision; however, frequency of single staged procedures is increasing.⁵² In all cases a broad-spectrum antibiotic should be started after biopsy harvest, and these should be tailored once the pathogen and sensitivities are known. Antibiotic therapy should endure for three months and an infection free period of two years is required before eradication can be considered.³⁵ Multiple case series report a greater than 90% eradication rate at two years after revision surgery.^40,41,53

Future research

C. acnes is not the only skin commensal in this area and there may be more yet unrecognised skin commensals, with yet unknown functions and potential to act as a pathogen. Future research must be targeted at understanding the full microbiome of the shoulder.

There is still a delay in diagnosis and obtaining sensitivities, and a more reliable and rapid diagnostic test would be of great benefit.

Delineating between contamination and infection remains problematic. Improved techniques and criteria to diagnose a true infection over a contaminant would also be a significant step forward.

Conclusion

This review confirms the current focus on C. acnes as an important part of the microbiome of the skin overlying the gleno-humeral joint. It is hoped that this can help surgeons reduce the likelihood of a C. acnes infection, increase the likelihood of detecting a C. acnes infection and understand the best practice of treating a C. acnes infection. It has demonstrated the paucity of knowledge of other potential pathogens within the normal microbiome and highlights the fact that a positive deep culture does not always reflect an infection.

Footnotes

Acknowledgements

This article is not based on a previous communication to a society or meeting.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Review and Patient Consent

Not applicable.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Guarantor

CS.

Contributorship

NM was principal author. TB undertook article screening and co-wrote the paper. CH undertook study registration and screened articles. WW undertook article screening. CS conceived the study, screened articles and authored the paper. All authors reviewed and edited the manuscript and approved the final version of the manuscript.

ORCID iDs

Timothy J Batten

William James White

References

Lanas

Chan

FKL.

Peptic ulcer disease. Lancet 2017; 390: 613–624.

Booker

Boyd

Gallacher

, et al. The colonisation of the glenohumeral joint by Propionibacterium acnes is not associated with frozen shoulder but is more likely to occur after an injection into the joint. Bone Joint J 2017; 99-B: 1067–1072.

Belk

Kraeutler

Smith

JR.

, et al. Prevention of Cutibacterium acnes in arthroscopic shoulder surgery: a systematic review. J Shoulder Elbow Surg 2020; 29: 867–873.

Brook

Frazier

EH.

Infections caused by Propionibacterium species. Rev Infect Dis 1991; 13: 819–822.

Hsu

Neradilek

Russ

SM.

Preoperative skin cultures are predictive of Propionibacterium load in deep cultures obtained at revision shoulder arthroplasty. J Shoulder Elbow Surg 2018; 27: 765–770.

Moher

Liberati

Tetzlaff

, et al. PRISMA – review guidelines. PLoS Med 2009; 6: e1000097–e1000097.

Ouzzani

Hammady

Fedorowicz

, et al. Rayyan – a web and mobile app for systematic reviews. Syst Rev 2016; 5: 210–210.

Taylor

Coe

Mata-Fink

Native joint Propionibacterium septic arthritis. Infect Dis Rep 2017; 9: 7185–7185.

Achermann

Goldstein

Coenye

, et al. Propionibacterium acnes: from commensal to opportunistic biofilm-associated implant pathogen. Clin Microbiol Rev 2014; 27: 419–440.

10.

Cogen

Nizet

Gallo

Skin microbiota: a source of disease or defence?

Br J Dermatol 2008; 158: 442–455.

11.

Fitz-Gibbon

Tomida

Chiu

BH.

, et al. Propionibacterium acnes strain populations in the human skin microbiome associated with acne. J Invest Dermatol 2013; 133: 2152–2160.

12.

Eishi

Suga

Ishige

, et al. Quantitative analysis of mycobacterial and propionibacterial DNA in lymph nodes of Japanese and European patients with sarcoidosis. J Clin Microbiol 2002; 40: 198–204.

13.

Fassi Fehri

Mak

Laube

, et al. Prevalence of Propionibacterium acnes in diseased prostates and its inflammatory and transforming activity on prostate epithelial cells. Int J Med Microbiol 2011; 301: 69–78.

14.

Perry

Lambert

Propionibacterium acnes: infection beyond the skin. Expert Rev Anti Infect Ther 2011; 9: 1149–1156.

15.

Levy

Iyer

Atoun

, et al.

Propionibacterium acnes: an underestimated etiology in the pathogenesis of osteoarthritis?

J Shoulder Elbow Surg 2013; 22: 505–511.

16.

Athwal

Sperling

Rispoli

DM.

, et al. Deep infection after rotator cuff repair. J Shoulder Elbow Surg 2007; 16: 306–311.

17.

Leeming

Holland

Cunliffe

WJ.

The microbial ecology of pilosebaceous units isolated from human skin. J Gen Microbiol 1984; 130: 803–807.

18.

Koh

Marsh

Drinkovic

Propionibacterium acnes in primary shoulder arthroplasty: rates of colonisation, patient risk factors, and efficacy of perioperative prophylaxis. J Shoulder Elbow Surg 2016; 25: 846–852.

19.

Hudek

Sommer

Abdelkawi

AF.

Propionibacterium acnes in shoulder surgery: is loss of hair protective for infection?

J Shoulder Elbow Surg 2016; 25: 973–980.

20.

Sampedro

Piper

McDowell

, et al. Species of Cutibacterium and Cutibacterium acnes phylotypes associated with orthopedic implants. Diagn Microbiol Infect Dis 2009; 64: 138–145.

21.

Qiu

Pena-Diaz

AM.

, et al. Cutibacterium acnes and the shoulder microbiome. J Shoulder Elbow Surg 2018; 27: 1734–1739.

22.

Sethi PM, Sabetta JR, Stuek SJ, et al. Incidence of culture positive Propionibacterium acnes in shoulder arthroscopy. Orthop J Sports Med 2014; 2(2 Suppl): 2325967114S00094.

23.

Kim

Song

Weon-Ju

Antimicrobial susceptibility and characterisation of Propionibacterium acnes by multilocus sequence typing and repetitive-sequence-based PCR. J Bacterial Virol 2016; 46: 135–141.

24.

Johnson

Kang

Barnard

, et al. Strain-level differences in porphyrin production and regulation in Propionibacterium acnes elucidate disease associations. mSphere 2016; 1: e00023–15.

25.

Nodzo

Hohman

Crane

JK.

, et al. Hemolysis as a clinical marker for Propionibacterium acnes orthopedic infection. Am J Orthop 2014; 43: E93–E97.

26.

Bumgarner

Harrison

Hsu

Cutibacterium acnes isolates from deep tissue specimens retrieved during revision shoulder arthroplasty: similar colony morphology does not indicate clonality. J Clin Microbiol 2020; 58: 121–119.

27.

Maccioni

Woodbridge

Balestro

JCY.

, et al. Low rate of Propionibacterium acnes in arthritic shoulders undergoing primary total shoulder replacement surgery using a strict specimen collection technique. J Shoulder Elbow Surg 2015; 24: 1206–1211.

28.

Horneff

III Hsu

Voleti

PB.

, et al. Propionibacterium acnes infection in shoulder arthroscopy patients with postoperative pain. J Shoulder Elbow Surg 2015; 24: 838–843.

29.

Mook

Klement

Green

CL.

, et al. The incidence of Propionibacterium acnes in open shoulder surgery: a controlled diagnostic study. J Bone Joint Surg Am 2015; 97: 957–963.

30.

Atkins

Athanasou

Deeks

JJ.

, et al. Prospective evaluation of criteria for microbiological diagnosis of prosthetic-joint infection at revision arthroplasty. J Clin Microbiol 1998; 36: 2932–2939.

31.

Kolakowski

Lai

Duvall

, et al. Benzoyl peroxide effectively decreases pre-operative Propionibacterium acnes shoulder burden: a randomised trial. J Shoulder Elbow Surg 2017; 27: 131–132.

32.

Lee

Matsen

III Pottinger

, et al. Propionibacterium persists in the skin in spite of standard surgical preparation. J Bone Joint Surg Am 2014; 96: 1447–1450.

33.

Phadnis

Gordon

Krishnan

, et al. Frequent isolation of Propionibacterium acnes from the shoulder dermis despite skin preparation and prophylactic antibiotics. J Shoulder Elbow Surg 2016; 25: 304–310.

34.

Numdari

Nicholson

Parvizi

Propionibacterium acnes from the skin of the shoulder: a randomised controlled trial. J Shoulder Elbow Surg 2017; 26: 1495–1499.

35.

Falconer

Baba

Kruse

LM.

, et al. Contamination of the surgical field with Propionibacterium acnes in primary shoulder arthroplasty. J Bone Joint Surg 2016; 98: 1722–1728.

36.

Chuang

Jancosko

Mendoza

, et al. The incidence of Propionibacterium acnes in shoulder arthroscopy. Arthroscopy 2015; 31: 1702–1707.

37.

Dizay

Lau

Nottage

Benzoyl peroxide and clindamycin topical skin preparation decreases Propionibacterium acnes colonisation in shoulder arthroscopy. J Shoulder Elbow Surg 2017; 26: 1190–1195.

38.

Hernandez

Sager

Liang

, et al. Bactericidal efficacy of hydrogen peroxide on Propionibacterium acnes. Bone Joint Res 2019; 8: 3–10.

39.

Matsen

Russ

Bertelsen

, et al. Propionibacterium can be isolated from deep cultures obtained at primary arthroplasty despite intravenous antimicrobial prophylaxis. J Shoulder Elbow Surg 2015; 24: 844–847.

40.

Zeller

Ghorbani

Strady

, et al. Cutibacterium acnes: an agent of prosthetic joint infection and colonization. J Infect 2007; 55: 119–124.

41.

Lutz

Berthelot

Fresard

, et al. Arthro-plastic and osteosynthetic infections due to Cutibacterium acnes: a retrospective study of 52 cases, 1995–2002. Eur J Clin Microbiol Infect Dis 2005; 24: 739–744.

42.

Achermann

Sahin

Schwyzer

H-K.

, et al. Characteristics and outcome of 16 periprosthetic shoulder joint infections. Infection 2013; 41: 613–620.

43.

Doherty

Furness

Batten

, et al. Arthroscopy of the symptomatic shoulder arthroplasty. J Shoulder Elbow Surg 2019; 28: 1971–1976.

44.

Frangiamore

Saleh

Grosso

MJ.

, et al. α-Defensin as a predictor of periprosthetic shoulder infection. J Shoulder Elbow Surg 2015; 24: 1021–1027.

45.

Boisrenoult

Cutibacterium acnes prosthetic joint infection: diagnosis and treatment. Orthop Traumatol Surg Res 2018; 104: S19–S24.

46.

Saper

Capiro

, et al. Management of Propionibacterium acnes infection after shoulder surgery. Curr Rev Musculoskelet Med 2015; 8: 67–74.

47.

Barreau

Pagnier

La Scola

Improving the identification of anaerobes in the clinical microbiology laboratory through MALDI-TOF mass spectrometry. Anaerobe 2013; 22: 123–125.

48.

Asseray

Papin

Touchais

, et al. Improving diagnostic criteria for Cutibacterium acnes osteomyelitis: a retrospective analysis. Scand J Infect Dis 2010; 42: 421–425.

49.

Schafer

Fich

Lam

, et al. Antimicrobial susceptibility and genetic characteristics of Propionibacterium acnes isolated from patients with acne. Int J Dermatol 2013; 52: 418–425.

50.

Leyden

New developments in topical antimicrobial therapy for acne. J Drugs Dermatol 2008; 7: s8–s11.

51.

Tsukayama

Estrada

Gustilo

RB.

Infection after total hip arthroplasty. A study of the treatment of one hundred and six infections. J Bone Joint Surg Am 1996; 78: 512–523.

52.

Hsu

Gorbaty

Whitney

IJ.

, et al. Single-stage revision is effective for failed shoulder arthroplasty with positive cultures for Propionibacterium. J Bone Joint Surg Am 2016; 98: 2047–2051.

53.

Jacobs

Van Hooff

Meis

JF.

, et al. Treatment of prosthetic joint infections due to Cutibacterium. Similar results in 60 patients treated with and without rifampicin. Acta Orthop 2016; 87: 60–66.