Tanya Pothini: No financial relationships to disclose
Purpose: The spectrum of sternoclavicular joint (SJC) infections is poorly understood. Most publications are single-institution series and none include non-surgically-treated patients. Our objective is to review our multi-institutional experience to better understand the disease spectrum by comparing outcomes between interventions and to better understand cases that evolve to requiring surgery. Methods: We performed an IRB-approved retrospective review of patients who were treated for SCJ infections at two large university hospitals with diverse patient populations including nine surgeons from 2010 to 2022. SSCJ cases were defined as those who had been diagnosed with sternoclavicular joint infection, sternoclavicular joint abscess, sternoclavicular joint septic arthritis, or sternoclavicular joint osteomyelitis in the electronic medical record. Patients were identified from a prospectively managed database, with additional cases being identified through surgeon operative log procedure names and the ICD codes. We performed a retrospective abstraction of the demographic, clinical, and outcome information from the electronic medical record, 54 variables and 7 total outcomes were studied. Pre-operative imaging reads were reviewed and key features were collected and tabulated for trends. Intraoperative details and extent of debridement were obtained from the surgeon operative report. All data was stored and analyzed using REDCAP data capture tools hosted at UT Southwestern. Results: Forty-nine patients (mean age, 50 (22-82); 78% male) underwent treatment for SSCJ infection, 36 underwent joint surgical and medical management, 5 patients underwent surgical intervention following failed antibiotic treatment, and 8 underwent antibiotic-only management. Patient characteristics and clinical features are summarized in Table 1. The decision to not intervene surgically was made based on the severity of symptoms, extent of bony destruction on imaging, and improvement on antibiotics. The overall average hospital length of stay was 22 days (range 0-105 days). Overall recurrence rate was 33% (15 patients). The rate of 30-day readmission was 19% (9 patients), with one 30-day mortality (2%). At a mean follow-up of 3 years, chronic pain at SCJ was present in 4 patients (10%). The mean time to return to preoperative status was 5 weeks (range 0-26). Non-surgically treated patients had shorter time to return to pre-operative status at 2 weeks, compared to 4 weeks in surgical patients. However, non-operative patients also had higher recurrence rates at 38% (5 patients), compared to 28% in surgery as an initial intervention (10 patients). There were no significant outcome differences between the initial surgical and non-surgical groups for HLOS, mortality, chronic pain, debility, and 30-day readmit rates. Conclusion: Our results suggest that if a patient has no other soft tissue infections, no IVDU, no history of osteomyelitis, and less significant imaging findings (mild erosions, no fluid collections), we recommend obtaining CT-guided biopsy, and then see if the patient improves on guided antibiotic therapy before making the decision for surgical intervention. However, if there is evidence of significant joint destruction, osteomyelitis, or abscess, then surgery should be the next best step. Surgical management to adequately debride the necrotic and infected tissue with thorough preoperative preparation and postoperative management appears to mitigate a poor outcome in the appropriate surgical candidates.
Identify the source of the funding for this research project: UT Southwestern Department of Thoracic Surgery
