Variability and performance of radiologic stricture parameters in Crohn's disease: a systematic review and meta-analysis.
systematic_review · Level I
Where this comes from
- Record sourced from PubMed, PMID 41127565.
- Also identified by DOI 10.1016/j.eclinm.2025.103541 and PMC identifier 12538911.
- Licence recorded as CC BY.
- The licence permits redistribution, so the abstract is shown in full and the full text is available from the publisher.
Abstract
Disease-related strictures are a common complication of Crohn's disease (CD). Cross-sectional imaging is widely used for their assessment, but definitions remain variable and non-standardised. This systematic review and meta-analysis aimed to identify commonly used imaging parameters, assess diagnostic performance, and evaluate consistency across studies. We conducted a systematic review of ultrasound (US), magnetic resonance imaging (MRI), and computed tomography (CT) studies on CD-associated strictures, searching MEDLINE/PubMed, Embase, and Cochrane to January 1, 2025. We included prospective and retrospective studies of small-bowel CD strictures with surgical histopathology as reference standard. Summary data were extracted from published reports and pooled using a random-effects bivariate meta-analysis, which jointly models sensitivity and specificity while accounting for between-study heterogeneity. Exclusion criteria were pediatric populations, colonic and upper-GI strictures. Main outcomes were stricture definitions and diagnostic performance. The study is registered with PROSPERO, CRD420251032918. Of the 9436 articles identified through the search, 30 met eligibility criteria and were included in the analysis, comprising 1866 patients with CD: 5 on US, 7 on CT, 8 on MRI, and the remaining assessed two techniques. Luminal narrowing (LN), bowel wall thickening (BWT), and pre-stenotic dilation (PSD) were the most common descriptors. 4 studies (13%) required all three; the remaining used LN or BWT alone, with PSD often considered optional (20/30 [77%]). Overall, 26.7% (8/30) of studies were judged at high or unclear risk of bias in at least one domain. Pooled sensitivity and specificity for US techniques to detect strictures were 0.88 (95% CI, 0.83-0.91) and 0.86 (95% CI, 0.79-0.91) (<i>I</i> <sup>2</sup> = 0%), respectively. Pooled sensitivity and specificity for MRE were 0.82 (95% CI, 0.69-0.90) and 0.80 (95% CI, 0.44-0.95) (<i>I</i> <sup>2</sup> = 61.2%), and for CTE were 0.83 (95% CI, 0.73-0.90) and 0.77 (95% CI, 0.47-0.93) (<i>I</i> <sup>2</sup> = 58.8%), respectively. High diagnostic accuracy was observed across imaging modalities, with no statistically significant difference among them, and only a minority of studies at risk of bias unlikely to affect these findings. Heterogeneity existed in cut-offs and parameter combinations used to define strictures. PSD does not appear essential for stricture diagnosis, which could simplify diagnostic protocols. Variations in histological criteria and limited evidence for some modalities may limit generalizability. None.