Reply: Low-grade Cutibacterium acnes shoulder infections do exist!: In response to the Letter to the Editor by Reinier WA Spek, Job N Doornberg, David Ring and Michel PJ van den Bekerom

Septic Olecranon Bursitis With Osteomyelitis Attributed to Cutibacterium acnes: Case Report and Literature Overview of the Dilemma of Potential Contaminants and False-Positives

We report an unusual case of acute septic olecranon bursitis, with probable olecranon osteomyelitis, where the only organism isolated in culture was initially considered a contaminant, Cutibacterium acnes. However, we ultimately considered it the likely causal organism when treatment for most of the other more likely organisms failed. This typically indolent organism is prevalent in pilosebaceous glands, which are scarce in the posterior elbow region. This case illustrates the often challenging empirical management of a musculoskeletal infection when the only organism isolated might be a contaminant, but successful eradication requires continued treatment as if it is the causal organism. The patient is a Caucasian 53-year-old male who presented to our clinic with a second episode of septic bursitis at the same location. Four years prior, he had septic olecranon bursitis from methicillin-sensitive Staphylococcus aureus that was treated uneventfully with one surgical debridement and a one-week course of antibiotics. In the current episode reported here, he sustained a minor abrasion. Cultures were obtained five separate times because of no growth and difficulty eradicating the infection. One culture grew C. acnes on day 21 of incubation; this long duration has been reported. The first several weeks of antibiotic treatment failed to eradicate the infection, which we ultimately attributed to inadequate treatment of C. acnes osteomyelitis. Although C. acnes has a well-known propensity for false-positive cultures as typically reported in post-operative shoulder infections, treatment for our patient's olecranon bursitis/osteomyelitis was successful only after several surgical debridements and a prolonged course of intravenous and oral antibiotics that targeted it as the presumptive causal organism. However, it was possible that C. acnes was a contaminant/superinfection, and another organism was the culprit, such as a Streptococcus or Mycobacterium species that was eradicated by the treatment regime targeted for C. acnes.

Amplicon-Based Next-Generation Sequencing as a Diagnostic Tool for the Detection of Phylotypes of Cutibacterium acnes in Orthopedic Implant-Associated Infections

The diagnosis of orthopedic implant-associated infections (OIAIs) caused by the slow-growing anaerobic bacterium Cutibacterium acnes is challenging. The mild clinical presentations of this low-virulent bacterium along with its ubiquitous presence on human skin and human-dominated environments often make it difficult to differentiate true infection from contamination. Previous studies have applied C. acnes phylotyping as a potential avenue to distinguish contamination from infection; several studies reported a prevalence of phylotypes IB [corresponding to type H in the single-locus sequence typing (SLST) scheme] and II (SLST type K) in OIAIs, while a few others found phylotype IA₁ (more specifically SLST type A) to be abundant. However, phylotype determination has mainly been done in a culture-dependent manner on randomly selected C. acnes isolates. Here, we used a culture-independent amplicon-based next-generation sequencing (aNGS) approach to determine the presence and relative abundances of C. acnes phylotypes in clinical OIAI specimens. As amplicon, the SLST target was used, a genomic fragment that is present in all C. acnes strains known to date. The aNGS approach was applied to 30 sonication fluid (SF) samples obtained from implants removed during revision surgeries, including 17 C. acnes culture-positive and 13 culture-negative SF specimens. In 53% of the culture-positive samples, SLST types were identified: relative abundances were highest for K-type C. acnes, followed by H- and D-type C. acnes. Other types, including A- and C-type C. acnes that are more prevalent on human skin, had low relative abundances. The aNGS results were compared with, and confirmed by a culture-dependent approach, which included the isolation, whole genome sequencing (WGS) and phylotyping of 36 strains of C. acnes obtained from these SF samples. Besides serving as a powerful adjunct to identify C. acnes phylotypes, the aNGS approach could also distinguish mono- from heterotypic infections, i.e., infections caused by more than one phylotype of C. acnes: in eight out of nine culture-positive SF samples multiple C. acnes types were detected. We propose that the aNGS approach, along with the patient’s clinical information, tissue and SF cultures and WGS, could help differentiate C. acnes contamination from true infection.

Genomic Analysis of Cutibacterium acnes Strains Isolated from Prosthetic Joint Infections

Cutibacterium acnes is a common cause of prosthetic joint infections (PJIs). The C. acnes population can be divided into six main phylotypes (IA₁, IA₂, IB, IC, II and III) that are associated with different clinical conditions and normal skin. A single-locus sequence typing (SLST) scheme can distinguish ten main SLST types: A-E (all IA₁), F (IA₂), G (IC), H (IB), K (II), L (III). We genome-sequenced and compared 16 strains of C. acnes isolated from healthy skin (n = 4) and PJIs (n = 12), including six PJI cases with a good outcome (four shoulder PJIs, one hip PJI, one knee PJI) and six with infection relapse (three shoulder PJIs, three hip PJIs). The sequenced strains belonged to four different phylotypes (IA₁, IA₂, IB and II) and seven different SLST types. All five type IB strains (all SLST type H1) were PJI isolates (three hip PJIs, two shoulder PJIs), and four of these caused infection relapse (three hip PJIs, one shoulder PJI). Isolates from PJI cases with a good outcome belonged to three different phylotypes (IA, IB, II). Interestingly, four strains (three strains from PJI cases with good outcome and one strain from healthy skin) contained a linear plasmid; these strains belonged to different SLST types (A1, C1, F4, H1) and were isolated in three different hospitals. This study suggests that type IB strains have the potential to cause infection relapse, in particular regarding hip PJIs. Moreover, our study revealed that strains belonging to the same SLST type can differ in their accessory genome in different geographic locations, indicative of microevolution.