Abstract
BACKGROUND: The TOpClass classification for perianal fistulizing Crohn's disease (pfCD) facilitates tailored treatment to clinical state and patient goals. MRI is central to pfCD assessment, but existing indices are limited in predicting disease classification and trajectory. This study evaluated MRI fistula volume as a radiological biomarker and its longitudinal association with pfCD class. METHODS: We conducted a retrospective cohort study of 51 consecutive pfCD patients who underwent a baseline pelvic MRI in 2010 with ≥1 follow-up MRI. pfCD class was assigned at baseline, short- and long-term follow-up (median 10 years). A gastrointestinal radiologist, blinded to clinical data, measured active MRI fistula volume on T2-weighted axial sequences at each timepoint using ITK-Snap. The primary outcome was the association between volume and pfCD Class. Secondary outcomes included identification of volume thresholds discriminating clinical state/class via ROC analysis. RESULTS: Of 51 included patients (mean age 34.5), the majority had complex fistulae (92.1%) and 71% patients were TOpClass 2b, with 35% changing class during follow-up. MRI fistula volume measurement was feasible (median acquisition time 207 seconds, IQR 116-250). Volume was associated with disease severity, increasing across TOpClass strata (P < .001). ROC-derived volume thresholds effectively differentiated classes (AUROC 0.69-0.80). A ≥ 27% volume reduction was associated with clinical improvement (AUROC 0.78; sensitivity 64%, specificity 84%). CONCLUSIONS: MRI fistula volume is associated with pfCD class and disease trajectory. Volume thresholds are associated with classification shifts and clinical response, supporting their potential as objective quantitative tools. Prospective multicenter validation is warranted.