Cutibacterium Culture Isolation Following Total Hip and Total Knee Arthroplasty

1. Introduction

Infection following total hip arthroplasty (THA) and total knee arthroplasty (TKA) remains an important area of study given the significant costs related to these procedures as well as the morbidity associated with clinical management.[1,2,3] While studied more extensively in orthopaedics as a pathogen following shoulder surgery,[4,5] Cutibacterium following THA and TKA remains poorly understood with respect to unique characteristics of the organism and variable clinical presentations. Cutibacterium species therefore pose an interesting diagnostic challenge for orthopaedic arthroplasty surgeons.

Cutibacterium species, formerly known as Propionibacterium species, are commensal gram-positive skin bacterium which can cause implant-associated infections. They are ubiquitous organisms present in skin flora and are generally found in sebaceous glands. They are notably slow growing, facultatively anaerobic, and non-spore forming. Techniques such as prolonged agar cultures for 14 days as well as tissue sonication have increased the detection of these bacteria.[6] Although once thought to be a cutaneous bacteria of little pathologic significance, there has been increasing evidence for its role in both native as well as implant-associated infections.[7,8] Furthermore, Cutibacterium has been implicated in more systemic diseases such as sarcoidosis, benign prostatic hyperplasia, prostate cancer, and possibly Parkinson’s disease.[9,10,11]

Notable prior work has evaluated Cutibacterium in shoulder arthroplasty [12,13,14,15,16,17] as well as infections following anterior cruciate ligament reconstruction[18], shoulder arthroscopy [19,20,21], and clavicle non-unions[22]. However less has been reported about Cutibacterium in total joint arthroplasty. Elkins et. al assessed anterior and lateral thigh skin colonization of Cutibacterium prior to total hip arthroplasty and found that 14 of the 101 (14%) patients had a positive culture, with most localized to the anterior hip (65%)[23]. Prior work by Nodzo et. al compared infection characteristics for patients undergoing revision TKA for infection due to Cutibacterium acnes versus methicillin sensitive Staphylococcal aureus[24]. Although a small study of 16 patients with P. acnes, they did demonstrate that median ESR was significantly higher in the MSSA group (56.0 mm/h; interquartile range [IQR], 44.3-72.9 vs 23.0 mm/h; IQR, 18.5-52.0; respectively, p = .03). Similar findings were noted for CRP however synovial and serum WBC count did not demonstrate any differences between the two groups. This work also noted a mean time to culture of 8.3 ± 2.0 days in the Cutibacterium group compared to 1.8 ± 0.8 days in the MSSA cohort.

Given the potential for varied and perhaps less overt clinical symptoms in patients with Cutibacterium prosthetic joint infection (PJI), this study sought to evaluate TKA and THA Cutibacterium infections with respect to presenting characteristics, objective clinical workup, treatments, and outcomes of patients undergoing surgical intervention at the 1-year mark.

2. Results

Retrospective review identified a total of 29 patients with positive Cutibacterium cultures during the time period, 21 THAs and 8 TKAs. The average age of the THA patients was 65.8 years and 64.7 years for the TKA patients (p = 0.82). Of the 21 patients with THA infections, 10 preoperatively met MSIS criteria (47.6%), while 7 out of 8 patients with TKA infections preoperatively met MSIS criteria for infection (87.5%) (p= 0.13). Of the eight patients with Cutibacterium TKA infections, seven (87.5%) were female. Comparison of the THA and TKA cohorts did not demonstrate any significant differences with respect to age, sex, ASA, BMI, or prior revision procedure. Other demographic data is shown in Table 1.

Evaluation of the culture data demonstrated that 47.6% of the THA infections and 62.5% of the TKA infections were polymicrobial, with Cutibacterium as one of the identified species (Table 2). Average time to culture positivity was 6.8 days for THA and 7.4 days for TKA (p = 0.572). The rate of positive cultures was 0.53 for THA and 0.27 for TKA (p = 0.059). The number of sonicated tissue cultures that were positive for Cutibacterium was not statistically significantly different for THA and TKA, with 5 out of 21 sonicated samples testing positive in the THA group (23.8%), and 1 out of 8 sonicated samples testing positive in the TKA group (p=0.242).

Laboratory markers including ESR and CRP were also recorded pre-operatively or at the time of revision surgery and demonstrated a mean ESR of 36.4 mm/h for THA (range 9-106) and 51.5 mm/h for TKA (range 10-102) (p = 0.21). Mean CRP was 35.2 mg/dL for THA and 36.8 mg/dL for TKA (p = 0.95). Mean cell count and percentage polymorphonuclear cells from intra-articular aspiration was 27,055 cells (range of 34-128,205) with 68% (range 2-98%) PMNs for THA and 22,194 (range of 95-29,829) with 73.9% (range 8-87%) PMNs for TKA (p = 0.72 and p = 0.70, respectively). Mean cell count and percentage polymorphonuclear cells from intra-articular aspiration was 24,143 (range of 95 to 128,205) with 58.9% (range of 8-98%) PMNs for monomicrobial infections and 25,903 (range of 135 to 89,378) with 78.8% (range of 2-98%) PMNs for polymicrobial infections.

Of the 29 patients with THA and TKA Cutibacterium infections, 24 (82.7%) underwent a surgical intervention (Figure 1): 14 explant of arthroplasty implants, 3 revisions for loosening, 2 TKA polyethylene exchange procedures, 1 spacer exchange, 3 single stage revisions, and 1 head and liner exchange. Of the 8 TKA patients, 6 (75%) underwent an explant, while 8 out of the 21 THA patients underwent an explant of the components (38.1%). The other 5 patients were treated non-operatively with antibiotics.

For patients with positive Cutibacterium THA cultures, approach for the primary THA procedure was also further investigated (Table 3). Of the 21 patients in our study, 10 underwent prior THA via an anterior approach and 8 were performed via a posterior approach (p= 0.92). Five of the ten (50%) anterior approach patients had monomicrobial infections, and similarly four of the eight (50%) posterior approach patients had monomicrobial infections.

For those patients who did undergo subsequent surgical intervention following their positive periprosthetic Cutibacterium culture and had follow up at one year (n =22), outcomes for this cohort were evaluated using the MSIS ORT (Table 4). [26] Overall, 16 of the 22 (72%) had successful outcomes at one year. Only ASA was noted to be statistically significantly different between the two groups (3 for the failure group versus 2.75 for the successful outcome group, p = 0.03).

3. Discussion

Detection and management of Cutibacterium prosthetic joint infection can pose a diagnostic challenge for orthopaedic arthroplasty surgeons given more subtle clinical as well as laboratory changes compared to more virulent microbial species, particularly in monomicrobial communities.[27,28,29] Our study showed that half of the THA and half of the TKA patients with Cutibacterium presented with polymicrobial infections with the most common concomitant bacteria being coagulase negative Staphylococcus. This is consistent with prior work demonstrating Cutibacterium as an overall rare monomicrobial pathogen that is not infrequently found in conjunction with other microorganisms.[32,33] We compared laboratory values for monomicrobial and polymicrobial Cutibacterium infections in THA and TKA patients and did not find statistically significant differences. However, average CRP of the monomicrobial group was lower, with an average of 26.5 vs 44.2 mg/dL for the polymicrobial group, regardless of the joint, supporting the potentially indolent nature of the bacteria. There was no significant difference between the frequency of polymicrobial PJI in THA and TKA patients, with 5 out of 8 TKA patients and 10 out of 21 THA patients having polymicrobial Cutibacterium infections. Further, our work showed a delay in time to culture-positivity of approximately 1-week, necessitating the need for extended cultures to assist in diagnosis (mean of 6.5 days for THA and 7.39 days for TKA).

WBC, ESR, CRP, and percentage PNMs on average for both TKA and THA patients with Cutibacterium infections were found to be elevated. Average CRP and ESR values both met MSIS criteria of greater than 30 mm/h and greater than 10 mg/L, respectively.[31] However, it should be noted that when studied in isolation, only 47% of our THA patients met MSIS criteria pre-operatively when compared to 87% of TKA patients. We believe this point further strengthens the believe that Cutibacterium poses more of a threat to THA than TKA patients as commonly accepted diagnostic algorithms may miss the diagnosis.

Prior studies have shown Cutibacterium infections to be more common in males than females who have had shoulder surgery.[34] This is consistent with the fact that Cutibacterium is present in areas with greater concentration of sebaceous glands, and men have a greater concentration of sebaceous glands than women.[35] Our data does not show a statistically significant difference in men compared women with regards to TKA or THA infections with Cutibacterium. This discrepancy could be related to the relative abundance of Cutibacterium acnes around the axilla and shoulder region compared with the hip and knee. Using this theory, other studies have compared the difference between anterior and posterior approaches for THA and the relative rates of Cutibacterium infections. The theory was that perhaps there would be a higher rate of Cutibacterium culture positivity with the anterior approach given the larger concentration of sebaceous glands in this region. One study found increased rates of Cutibacterium avidum[36] in patients with an anterior compared to posterior THA. Our study did not find a statistically significant difference in the rates of Cutibacterium infections based on the approach of the initial arthroplasty procedure. Further studies have shown that the direct anterior approach to primary THA does not increase the risk of periprosthetic joint infection compared to the direct lateral approach. Additionally, there were not significant differences in infecting organisms between anterior and direct lateral approach groups[37]. Interestingly, of the 29 arthroplasty patients with Cutibacterium infections in this study, 21 were from THA’s, with just 8 from TKA’s. This is consistent with the study by Elkins et al, mentioned in the introduction, which showed that the anterior thigh has a high colonization rate of Cutibacterium. [23]

This study has several limitations as a retrospective analysis of a relatively uncommon organism culture in THA and TKA. As mentioned, Cutibacterium are ubiquitous organisms present in skin flora and some have suggested that isolation of the bacteria may represent a contaminant.[38,39] The present study sought to evaluate all patients with a positive culture, though we do acknowledge that of the 21 patients with THA infections, 10 preoperatively met MSIS criteria for PJI (47.6%) while 7 out of 8 patients with TKA infections preoperatively met current MSIS criteria for infection (87.5%) (p= 0.13). This finding may be explained by the less virulent nature of Cutibacterium to elicit common warning signs for PJI such as significant erythema, effusion, or sinus tract as well as the difficulty with culture of the bacteria. However, it is also possible that the Cutibacterium represents a contaminant or a non-pathologic organism in the polymicrobial infections. The purpose of this work was not to decipher such a difference but rather to report laboratory and other characteristics of those with a positive Cutibacterium prosthetic culture. Another limitation of this study is that it utilized conventional culture, which may have missed some positive identification of Cutibacterium possible with other methods (e.g., next generation sequencing), however we suspect the use of traditional cultures is more generalizable given the feasibility of this method. Furthermore, additional cases of Cutibacterium may be missed by cultures not being held for an extended time (14 days in our institution). As this study found, the average time to culture positivity of approximately one week (mean of 6.5 days with a range of 3-11 for THA and 7.3 days with a range of 5-12 for TKA). Lastly, there may have been other patient or clinical characteristics that influence the laboratory values and outcomes evaluated.

4. Methods

Retrospective chart review was performed to identify patients with culture-positivity Cutibacterium infection of a THA or TKA within the time period at our tertiary academic referral center. Using the electronic medical record, all Propionibacterium or Cutibacterium positive cultures from April 2014 to January 2021 were identified. Those that involved a native joint or an infection of a prosthetic joint that was not a hip or knee arthroplasty were excluded (Table S1) All procedures were reviewed to confirm that each had at least one culture with Cutibacterium or Propionibacterium. Using this criteria, 29 total cases were identified for further analysis. Importantly, this study was conducted as a retrospective review of de-identified data. No patient names, medical record numbers, or other identifiable health information were collected, stored, or reported. As such, this project did not meet the definition of human subjects research under 45 CFR 46.102(f) and therefore did not require Institutional Review Board (IRB) approval.

Patient demographics including gender, age, and BMI were collected. American Society of Anesthesiology (ASA) scores were used to evaluate each patient’s relative overall health. Preoperative lab values were collected, including ESR, CRP, WBC count, percentage of Polymorphonuclear leukocytes (% PMNs), and presence of alpha-defensin. Culture data was collected from either synovial fluid and/or from tissue in cases where the patient underwent subsequent operative intervention. Per institutional standards for periprosthetic joints, cultures were incubated for 14 days, with the exception of one sample which was held for seven days. Previous surgical interventions were recorded. Using the 2018 Musculoskeletal Infection Society (MSIS) criteria for prosthetic joint infection, each patient was dichotomized as infected with a score of ≥6 or not definitively infected with a score of ≤5.[25] One-year outcomes for patients who underwent subsequent surgical intervention and had documented clinical follow-up (n = 22) were categorized according to the Musculoskeletal Infection Society Outcome Reporting Tool (MSIS ORT, Table 4).[26] A single physician reviewer (TMB) retrospectively evaluated patient charts and assigned each patient into “success” (Tier 1 and Tier 2) and “failure due to either PJI or failure due to a secondary cause” (all other Tiers).

Statistical Analysis

Analysis of patient, laboratory, microbiologic and surgical characteristics was performed to evaluate the cohort as a whole and to compare THA to TKA. T-test and chi-squared analysis was utilized to compare patient baseline, demographic, and laboratory characteristics. Additional analysis evaluated outcomes at one year per the MSIS ORT for those who underwent surgical intervention and subsequent analysis compared those with “success” (Tiers 1 and 2) versus “failure” ((Tiers 3A, 3B, 3C, 3D, 3E, 3F, 4A, 4B).

5. Conclusions

This study contributes to the current literature on Cutibacterium culture positivity in TKA and THA patients. With longer culture times becoming more standard, Cutibacterium infections will continue to present itself as a factor in PJI. This research illuminates some of the characteristics of monomicrobial and polymicrobial Cutibacterium infections in the hip and knee and demonstrates that Cutibacterium must be considered as a pathogen when suspecting a PJI in the hip and knee. Further research is needed to identify just how prominent Cutibacterium infections in a THA and TKA are, and to identify the ideal treatments to optimize patient outcomes.