Ultrasound- and fluoroscopy-guided hip aspiration before revision total hip arthroplasty: procedural outcomes and diagnostic accuracy in sequential cohorts

BACKGROUND AND PURPOSE: Hip aspiration remains essential in the diagnostic work-up of a painful total hip arthroplasty (THA) to assess for periprosthetic joint infection (PJI). In 2020, our institution transitioned from fluoroscopy-guided to ultrasound-guided aspiration. Our study aimed to describe dry-tap proportions, aspirated volumes, and complications, and explore diagnostic accuracy in sequential cohorts. METHODS: This retrospective single-center study included 117 fluoroscopy-guided aspirations (December 2013-June 2020) and 90 ultrasound-guided aspirations (April 2020-March 2024) before revision THA. Procedural outcomes defined as dry taps, aspirated volumes, and complications within 14 days were assessed in the full cohorts, and diagnostic accuracy in procedures with a successful aspiration. Ultrasound-guided biopsy was performed in 39 patients. RESULTS: The prevalence of PJI was 54/117 (46%) in the fluoroscopy cohort and 25/90 (28%) in the ultrasound cohort (difference -18 percentage points, 95% confidence interval [CI] -31 to -5). Dry taps occurred in 38/117 (32%) and 31/90 (34%) procedures, corresponding to an absolute difference of 2 percentage points (CI -11 to 16). Median aspirated volumes were 5 mL in the fluoroscopy cohort and 4 mL in the ultrasound cohort. No complications were recorded in the fluoroscopy cohort; 2 occurred in the ultrasound cohort, both among patients who underwent biopsy. Sensitivity was 16/34 (47%, CI 30-65) for fluoroscopy and 8/13 (62%, CI 32-86) for ultrasound; specificity was 41/44 (93%, CI 81-99) and 37/40 (93%, CI 80-98). CONCLUSION: Dry-tap proportions were similar in these sequential cohorts. 2 complications were recorded, both in patients who underwent biopsy. Specificity was high, whereas sensitivity could be assessed only in patients with a successful aspiration and available culture results, resulting in less precise estimates. The findings do not establish equivalence or superiority.