Table 1 Sensitivity and specificity of predicting disease type or subtype
CD type/subtype | MRE sensitivity (%) | MRE specificity (%) | Logistic regression | Important differentiating MRI feature |
|---|---|---|---|---|
Penetrating type | 63.7 | 60.3 | p < 0.001 | Presence of ulcer on T2-weighted images (T2WIs) [9] |
Nonstricturing/non penetrating | 58.2 | 48.4 | p < 0.001 | Presence of the classical CD features without complications on T2-weighted and gadolinium-enhanced T1-weighted sequences [8] |
Stricturing type | 63.9 | 63.2 | p < 0.001 | Presence of segment(s) of luminal narrowing (fibrotic or inflammatory cause) on unenhanced T2-weighted and gadolinium-enhanced T1-weighted sequences [8] |
Inflammatory subtype (Local inflammation, aphthoid and deep ulcers, frequent transmural inflammation with lymphoid aggregates, and granuloma development define the active inflammatory subtype of CD) | 73.8 | 64.9 | p < 0.001 | Acute inflammation is best seen in T2WIs in fat suppressed sequences; stratified contrast confirms CD is active [9] |
Fibro stenotic subtype (persistent intestinal damage, chronic inflammation of the gut wall tends to proceed to fibrostenotic and irreversible consequences (bowel strictures and blockage)) | 81.5 | 63.3 | p < 0.001 | Presence of stenotic segment that persist in the cine images and shows hypointense on T1WIs and T2WIs, and on contrast enhancement shows inhomogenous delayed enhancement [9] |
Reparative or regenerative subtype | 72.4 | 64.2 | p = 0.005 | Luminal narrowing may be seen without any sign of obstruction. Post contrast T1 images showed homogenous continuous enhancement [39] |
Fistulizing/perforating subtype (The presence of deep penetrating ulcers, which can lead to the formation of sinus tracts, fistulas, and/or abscesses, characterises it) | 61.0 | 68.9 | p = 0.008 | Presence of fistula or signs of perforation with adjacent collection, best seen on gadolinium-enhanced T1-weighted sequences [9] |