Diagnostic yield of CT-guided sampling in suspected sternoclavicular joint infection.

Objective: To determine the prevalence of infection diagnosed by percutaneous computed tomography (CT)-guided sternoclavicular (SC) sampling in cases of suspected joint infection.
Materials and Methods: A retrospective search was performed in reports of SC joint CT-guided biopsies in adults from July 1992 to July 2012. We reviewed medical records, radiology, microbiology, laboratory, and surgical reports. A positive result was defined as demonstration of a pathogenic organism, either by microscopy or growth in culture, confirming the diagnosis of SC joint infection. A negative result was defined as the absence of such findings. Patients in whom sampling was unsuccessful or not subjected to microbiology were excluded. In addition, CT images were reviewed by the consensus of two musculoskeletal radiologists.
Results: A total of 41 patients (mean age 57.1 years) underwent CT-guided SC joint sampling, 27 of whom underwent microbiology studies. Sampling was performed using core biopsy alone in 19%, fine needle aspiration alone in 44%, and aspiration combined with core biopsy in 37%. Positive results were found in 52% (14/27) of patients. Related diseases and predisposing conditions for infection were found in 79% of positive patients. Negative results were found in 48% (13/27) of patients. There were no procedure-related complications. The dominant CT findings were soft tissue swelling (negative group), and effusion and/or capsular hypertrophy/distension (positive group).
Conclusions: CT-guided sampling is a safe procedure with positive microbiological cultures in slightly more than half of cases.