Introduction
Perianal abscesses are a frequent and serious complication of Crohn’s disease (CD), requiring timely and accurate diagnosis to guide appropriate treatment. Transperineal ultrasound (TPUS), especially when combined with contrast-enhanced ultrasound (CEUS), may offer a promising, real-time, bedside alternative.
Aims & Methods
This study aimed to evaluate the diagnostic performance and inter/intra-observer reliability of CEUS versus conventional TPUS B-mode in differentiating abscesses from inflammatory masses in CD patients with acute perianal symptoms. We retrospectively analyzed 52 examinations performed on 37 adult Crohn’s disease (CD) patients who underwent transperineal ultrasound (TPUS) at a tertiary IBD center between 2012 and 2021 for acute or painful perianal symptoms and suspected septic complications. Patients were included if TPUS identified perianal lesions or masses suggestive of abscesses, warranting further evaluation with contrast-enhanced ultrasound (CEUS) to better characterize the lesion. Exclusion criteria included contraindications to CEUS (e.g., congestive heart failure), the presence of simple fistulas, or large, clearly identifiable abscesses (>3 cm) with evident fluctuating fluid content already visible on B-mode imaging. TPUS B-mode and CEUS examinations were performed using standardized protocols, with intravenous administration of 4.8 mL of contrast agent (SonoVue®). All examinations were recorded, and three expert reviewers independently assessed anonymized video cases in two phases—B-mode followed by CEUS—randomized and blinded to clinical data. Lesions were classified as abscesses, inflammatory masses, or phlegmons. Intra- and inter-observer agreement was evaluated for maximum lesion size, maximum size of fluid content, and final diagnosis using intraclass correlation coefficients (ICC) and Cohen’s kappa (κ).
Results
On TPUS B-mode, 17 abscesses, 32 inflammatory masses, and 3 phlegmons were identified; CEUS reclassified several of these lesions, ultimately detecting 16 abscesses, 31 inflammatory masses, and 5 phlegmons. Among the 17 abscesses diagnosed on B-mode, only 4 were confirmed by CEUS, while 11 were reclassified as inflammatory masses and 2 as phlegmons. Conversely, 11 lesions initially classified as inflammatory masses on B-mode were reclassified as abscesses by CEUS.
Intra-observer agreement for B-mode was excellent in assessing lesion size (ICC 0.853) and fluid content (ICC 0.897), but only fair to substantial for final diagnosis (κ = 0.55). CEUS demonstrated superior reliability, with nearly perfect intra-observer agreement for diagnosis (κ = 0.95).
Inter-observer agreement for TPUS B-mode was moderate to good for lesion size (ICC 0.798) and fluid content (ICC 0.736), but poor for final diagnosis (κ = 0.31, 95% CI 0.00–0.78), mainly due to the limited reliability of most sonographic features (echogenicity, gaseous artifacts, vascularity), except for posterior echo enhancement (κ = 0.54, 95% CI 0.36–0.82). In contrast, inter-observer agreement for CEUS was good for lesion size (ICC 0.847, 95% CI 0.76–0.91), fluid collection size (ICC 0.925, 95% CI 0.88–0.96) and for final diagnosis (κ = 0.83, 95% CI 0.79–0.95).
Conclusion
Contrast-enhanced TPUS significantly improves diagnostic reliability in identifying perianal abscesses in Crohn’s disease, outperforming conventional B-mode TPUS in both intra- and inter-observer assessments. CEUS provides a valuable, accessible imaging modality for real-time evaluation of perianal complications, particularly in urgent clinical scenarios or when MRI is not feasible.
Disclosure
Giovanni Maconi consulted for Abbvie, Arena Pharmaceuticals, Fresenius Kabi, Galapaos, Janssen-Cilag, Gilead, Samsung Takeda. Cristina Bezzio received lecture fees from Takeda, AbbVie, and Janssen. Dan Carter has received speaker’s fees and/ or research support from Takeda, Janssen, AbbVie, Illy Lilly, Reckitt, and Lapidot. Consultancy fees from Takeda, AbbVie, and Taro. Bincy Abraham has received research funding from Takeda; consulted for AbbVie, Bristol Myers Squibb, Eli Lilly, Janssen, Medtronic, Pfizer, Samsung Bioepis, Celltrion, and Takeda; lectured for AbbVie, Bristol Myers Squibb, Eli Lilly, Janssen, Pfizer, and Takeda. Remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.