Effectiveness of a combined arthroscopic and antibiotic-impregnated bead approach for septic shoulder arthritis management: a case series

Article information

Clin Shoulder Elb. 2025;28(1):31-39

Publication date (electronic) : 2025 February 17

doi : https://doi.org/10.5397/cise.2024.00584

Sung-Jin Park, Sam-Guk Park

Department of Orthopedic Surgery, Yeungnam University College of Medicine, Daegu, Korea

Correspondence: Sam-Guk Park Department of Orthopedic Surgery, Yeungnam University College of Medicine, 170 Hyeonchung-ro, Nam-gu, Daegu 42415, Korea Tel: +82-53-620-3640 Fax: +82-53-628-4020 E-mail: radiorth@ynu.ac.kr

Received 2024 July 28; Revised 2024 November 19; Accepted 2024 December 26.

Abstract

INTRODUCTION

Orthopedic infections, which can compromise surgical outcomes, often require long-term antibiotic therapy. The incidence of bacterial joint infection is reported to be 4 to 10 out of 100,000 per year, and the third most prevalent site is the native shoulder joint, followed by the hip and knee [1]. Conventional treatment involves early drainage of the joint along with intravenous antibiotic administration [2]. However, the best method for joint decompression remains controversial, as various techniques have yielded satisfactory results [3].

Arthroscopy is preferred because of its effectiveness in decompressing infected joints, particularly in the knee, with minimal residual complications [4]. The use of arthroscopy for managing septic shoulder arthritis has increased owing to the advantages of the technique over open procedures, such as small incisions, less postoperative pain, posterior joint visualization, and generally positive clinical outcomes [5,6]. However, the rate of reoperation after arthroscopic management is notably high at 26% to 32% [7,8]. This underscores the need for improved treatment strategies that effectively combat infection while reducing the likelihood of recurrence and subsequent surgeries [8].

Our treatment protocol for septic shoulder arthritis is arthroscopic synovectomy followed by strategic placement of antibiotic-impregnated polymethylmethacrylate (PMMA) beads in the subacromial space. This technique provides localized antibiotic delivery directly to the site of infection, enhancing treatment efficacy and potentially reducing the need for further surgery. In this study, we retrospectively reviewed the outcomes of 22 patients with septic shoulder arthritis who were treated using this method. We evaluated the effectiveness of antibiotic-laden PMMA beads for treating infectious arthritis in the shoulder joint and assessed the impact of this approach on the rate of reoperation.

METHODS

This study was approved by the Institutional Review Board of Yeungnam University Hospital (No. YUMC 2024-03-061). The informed consent for this retrospective study was waived.

Operative Technique

Confirmation of septic arthritis indicates immediate surgery. At our institution, arthroscopic shoulder surgery is performed under general anesthesia with patients in the lateral position. Before expanding the shoulder joint with saline, joint fluid was aspirated for culture and WBC count. For the initial examination of the shoulder joint, a 4-mm 30° arthroscope was inserted through the posterior portal. An anterior portal was established for instrumentation. Frozen and permanent biopsies were performed to confirm synovitis. The joint was lavaged using a 5.5-mm inflow cannula, and fibrous debris and inflamed synovial tissues were removed using a 4.2-mm powered shaver. The subacromial space was inspected through the posterior, lateral, and anterosuperior portals, with subsequent removal of inflamed synovial tissue. Irrigation was performed using 20 to 30 L of saline. Subsequently, 1,000 mg of vancomycin was mixed with 20 g of gentamicin mixed PMMA to create antibiotic-loaded cement beads on surgical wire (Fig. 1). Typically, seven beads were formed into long narrow strips and then linked together in a chain shape. After the cement had completely solidified, the beads were placed in the subacromial space during arthroscopy using a large inflow cannula (Fig. 2). A closed suction drain was inserted at the end of surgery (Fig. 3).

Fig. 1.

(A) Complete assembly of polymethylmethacrylate (PMMA) beads, with each strip composed of seven uniformly-spaced beads arranged in a long, narrow configuration. (B) Arthroscopic image of the shoulder joint after PMMA bead insertion.

Fig. 2.

Diagram of polymethylmethacrylate bead placement in the subacromial space.

Fig. 3.

Postoperative radiograph after arthroscopic synovectomy for septic shoulder arthritis and insertion of antibiotic-impregnated polymethylmethacrylate beads into the subacromial space.

The initial intravenous antibiotic was cefazolin. According to the results of culture and biopsy, antibiotic therapy was adopted after consultation with the Infectious Disease Department. In cases where patients had previously received antibiotics at other hospitals, experienced recurrence, or developed allergic reactions, the antibiotic regimen was adjusted accordingly in consultation with the Infectious Disease Department. Intravenous antibiotic therapy was continued for an average of 4.3 weeks (range, 1–10 weeks) until CRP levels normalized. Subsequently, oral antibiotics were administered to ensure a minimum treatment duration of 6 weeks [13-15].

Approximately 4.6 weeks (range, 3–8 weeks) later, we removed antibiotic-loaded cement beads when the CRP levels normalized. In patients with chronic conditions such as chronic kidney disease or chronic obstructive pulmonary disease, CRP levels may not normalize; hence, the beads were removed once CRP levels decreased significantly compared to baseline and when no clinical symptoms such as shoulder pain or redness were observed. If the patient’s condition permitted, diagnostic arthroscopy was performed after bead removal to confirm whether osteomyelitis was present at the rotator cuff insertion site of the greater tuberosity and whether purulent synovitis remained. If osteomyelitis was observed near the greater tuberosity during diagnostic arthroscopy, curettage and biopsy were performed [16]. Subsequently, drains were removed 1 to 2 days postoperatively, and no more antibiotics were administered (Fig. 4).

Fig. 4.

Patient no. 9: a 65-year-old female who presented with septic shoulder arthritis. (A) Initial magnetic resonance imaging (MRI) image. (B) Initial arthroscopy findings, where the insertion site of the supraspinatus was observed, but no osteomyelitis was identified. (C) The placement of antibiotic-impregnated beads, completing the surgery. C-reactive protein level subsequently normalized, and clinical symptoms resolved, leading to bead removal. (D) During bead removal, diagnostic arthroscopy showing pus at the supraspinatus insertion site. (E) Additional synovectomy, with bone suspected of osteomyelitis observed and subsequently shaved. (F) The site after the suspected osteomyelitis was removed. (G) Follow-up MRI findings at 1 year, showing no evidence of pyogenic arthritis.

RESULTS

In this study, we analyzed 22 patients who underwent a combined treatment approach for septic shoulder arthritis involving arthroscopic synovectomy and use of antibiotic-impregnated PMMA beads. Demographic characteristics and preoperative evaluation findings are summarized in Table 1. Laboratory analyses revealed an average CRP level of 10.3 mg/dL (range, 0.73–23.70 mg/dL) and ESR of 52.8 mm/hr (range, 12–120 mm/hr). Mean WBC count in the preoperatively aspirated joint fluid was 66,873/µL (range, 2,675–240,000/µL). Regarding infection severity, the distribution of Gächter stages was as follows: stage I, 1 shoulder (5%); stage II, 11 shoulders (50%); stage III, 6 shoulders (27%); and stage IV, 4 shoulders (18%) (Table 2). Moreover, MRI of all 22 shoulder joints revealed joint effusion. Important demographic data for the study group are listed in Table 3.

Table 1.

Demographic characteristics and preoperative and postoperative evaluations

Table 2.

Classification using the Gächter staging system [11]

Table 3.

Demographic data

Among the 22 patients, none showed symptoms of shoulder pain or redness before bead removal. CRP levels either normalized or significantly decreased and remained stable in all patients within an average of 3.4 weeks (range, 1–11 weeks) after the initial surgery. Consequently, beads were removed from all patients. Following bead removal, 21 patients remained recurrence-free. One patient exhibited signs of recurrence, including elevated CRP level 4 months later. This patient was infected with Candida species and had a Gächter stage IV infection. Open surgical debridement was performed, and no signs of infection were observed thereafter.

One patient experienced an infection at the site where the Hemovac was removed, which required open exploration and bead removal after the initial surgery. The infection was successfully managed after a second surgery. Another patient had osteomyelitis in the greater tuberosity of the humerus during diagnostic arthroscopy performed after bead removal. Consequently, curettage was performed using an arthroscope, thus controlling the infection. One patient underwent RCT repair after successful infection management; however, subsequent follow-up revealed re-rupture in this patient.

The causative organism was identified in 10 patients, with Staphylococcus aureus being the most common (isolated from seven patients). Methicillin-resistant Staphylococcus aureus (MRSA) was identified in one patient, resulting in an MRSA prevalence rate of 10% among culture-positive cases. The other two patients were infected with Streptococcus, and one patient was infected with Candida species (Table 4). Histopathological analysis of surgical biopsy specimens from the synovial tissues of all patients revealed acute active inflammation consistent with septic arthritis.

Table 4.

Causative organism on joint fluid culture

Except for one patient who experienced recurrence during initial follow-up, no further infections were observed. All patients eventually had well-healed wounds without reinfection until final follow-up. The average follow-up period was 20.2 months (range, 6–51 months). All patients experienced pain preoperatively, with an average pain level of 8.1 on the VAS (range, 3–10). At the time of follow-up, the average pain level had improved to 2.4 (range, 0–8). The average Constant-Murley score for functional outcomes at the final follow-up was 60.7 points (range, 17–81 points).

DISCUSSION

Although septic arthritis of the shoulder is less prevalent than that of the knee or hip joints, it poses significant management challenges and can result in considerable morbidity among affected patients [5]. Various treatment options include serial aspiration, open arthrotomy with irrigation and drainage, and arthroscopic irrigation and drainage, all of which have limitations. The infection recurrence rate following arthroscopic surgery for septic shoulder arthritis was approximately 30% [8].

Surgical treatment for septic arthritis aims to remove as much inflamed tissue as possible and to maintain high concentrations of antibiotics at the local site [17]. Therefore, we attempted this approach and aimed to report the outcomes of our procedure. Embedding antibiotics in methylmethacrylate cement to enhance therapeutic efficacy at infection sites is a longstanding practice, either as spacers in arthroplasty or during the fixation of arthroplasty components [18,19]. This method includes inserting PMMA antibiotic beads into the joint during surgery and collecting cultures to guide subsequent systemic antibiotic therapy based on these results.

The rationale behind this approach is the advantage of local drug delivery systems that can deliver high concentrations of antibiotics directly to the infection site while maintaining low systemic levels, thereby reducing potential systemic toxicity. This is crucial for achieving therapeutic concentrations in areas with poor blood supply, where the effectiveness of systemic antibiotics may be compromised [20]. Research has addressed concerns regarding chondrotoxicity, especially concerning antibiotics such as vancomycin, noting that concentrations of ≤1 mg/mL did not significantly affect cell proliferation and metabolism [21]. This indicates safe levels for antibiotic-loaded beads. In the current study, beads were prepared by mixing 1,000 mg of vancomycin with 20 g of gentamicin-infused PMMA cement, which is considered within safe limits for preventing chondrotoxic effects [17,22]. This method of using antibiotic-loaded cement beads aims to reduce the high recurrence rates observed with standard arthroscopic treatments. The infection recurrence rate was notably low in this study at 5%. The beads were removed approximately 5 weeks postoperatively to evaluate infection control and the potential for additional procedures, such as synovectomy, as necessary.

The limitations of this study include its retrospective observational design and relatively small sample size, which limit the generalizability of the findings. Additionally, the absence of a control group that was not treated with antibiotic beads makes it difficult to definitively conclude that the combined approach is superior to either arthroscopy or antibiotic beads alone. Furthermore, while the observed recurrence rate was low, the long-term implications compared to standard care over longer follow-up periods remain uncertain.

CONCLUSIONS

A combined arthroscopic debridement and antibiotic-impregnated bead approach for treating septic shoulder arthritis was effective with a very low infection recurrence rate. This method facilitated the direct delivery of antibiotics to infection sites, leading to good patient outcomes. Future research, particularly randomized controlled trials, is needed to confirm the efficacy of this approach. Further studies should also compare both short- and long-term outcomes to standard treatments.

Article information Continued