Ultrasonographic detection and assessment of the severity of Crohn's disease recurrence after ileal resection

Abstract

Background: Recurrence and severity of Crohn's disease mucosal lesions after "curative" ileal resection is assessed at endoscopy. Intramural lesions can be detected as increased wall thickness at Small Intestine Contrast Ultrasonography (SICUS).

Aims: To assess after ileal resection whether: 1) SICUS detects recurrence of Crohn's disease lesions, 2) the intestinal wall thickness measured at the level of ileo-colonic anastomosis predicts the severity of endoscopic lesions, 3) the extension of intramural lesions of the neo-terminal ileum is useful for grading severity of the recurrence, 4) the combined measures of wall thickness of the ileo-colonic anastomosis and of the extension of intramural lesions at level of the neo-terminal ileum may predict the endoscopic Rutgeerts score

Methods: Fifty eight Crohn's disease patients (M 37, age range 19-75 yrs) were prospectively submitted at 6-12 months intervals after surgery to endoscopy and SICUS for a total of 111 observations.

Results: Six months or more after surgery wall thickness of ileo-colonic anastomosis > 3.5 mm identified 100% of patients with endoscopic lesions (p < 0.0001). ROC curve analysis, combining wall thickness of ileo-colonic anastomosis and the extension of intramural lesions of neo-terminal ileum, discriminated (0.95) patients with, from those without, endoscopic lesions. Performing two multiple logistic regression analyses only wall thickness of ileo-colonic anastomosis and extension of neo-terminal ileum intramural lesions were significantly associated with absence or presence of endoscopic lesions. An ordinal polychotomus logistic model, considering all investigated variables, confirmed that only SICUS variables were associated with endoscopic grading of severity.

Conclusions: In patients submitted to ileal resection for Crohn's disease non-invasive Small Intestine Contrast Ultrasonography 1) by assessing thickness of ileo-colonic anastomosis accurately detects initial, minimal Crohn's disease recurrence, and 2) by assessing both thickness of ileo-colonic anastomosis and extension of intramural lesions of neo-terminal ileum grades the severity of the post-surgical recurrence.

Figure 1

A. SICUS assessment of ileo-colonic anastomosis. Calipers indicate thickness of ileal (up) (1.9 mm) and colonic (down) (1.2 mm) limbs. Arrowhead indicates colon, arrow indicates neo-terminal ileum. B. SICUS assessment of ileo-colonic anastomosis and neo-terminal ileum. CD intramural involvement of ileo-colonic anastomosis (thickness 10 mm see arrows) and neo-terminal ileum (extending for 8 cm, see arrowhead).

Figure 2

A. Box-and-whiskers plots of wall thickness at the level of ileo-colonic anastomosis according to different Rutgeerts score. The boxes at each score extend from the 25^th percentile (x_[25]) to the 75^th percentile (x_[75]) [i.e., the interquartile range (IQ)]; the lines inside the boxes represent the median values. The line emerging from the boxes (i.e., the "whiskers") extend to the upper and lower adjacent values. The upper adjacent value is defined as the largest data point ≤x_[75] + 1.5 × IQ, and the lower adjacent value is defined as the smallest data point ≥x_[25] - 1.5 × IQ. Observed values more extreme than the adjacent values, if any, are individually plotted (circles). * p < 0.001 score 1 vs score 0, 2-4. The horizontal line indicates the cut-off value of 3.5 mm. B. Box-and-whiskers plots of extension of intramural lesion at the level of neo-terminal ileum according different Rutgeerts score. The boxes at each time unit extend from the 25^th percentile (x_[25]) to the 75^th percentile (x_[75]) [i.e., the interquartile range (IQ)]; the lines inside the boxes represent the median values. The line emerging from the boxes (i.e., the "whiskers") extend to the upper and lower adjacent values. The upper adjacent value is defined as the largest data point ≤x_[75] + 1.5 × IQ, and the lower adjacent value is defined as the smallest data point ≥x_[25] - 1.5 × IQ. Observed values more extreme than the adjacent values, if any, are individually plotted (circles). * p < 0.001 score 1 vs score 4, ** p = 0.03 score 1 vs score 2, *** p = 0.04 score 1 vs score 3

Figure 3

Scatter plot of pairs of US values of the extension of intramural lesions at the level of the neo-terminal ileum and the wall thickness at level of the ileo-colonic anastomosis in 111 evaluations (58 patients) and estimated correlation.

Figure 4

ROC curve analysis. Accuracy of the combined US values, i.e. intramural lesions extension at the level of the neo-terminal ileum and wall thickness values at the level of the ileo-colonic anastomosis, in discriminating patients with Rutgeerts score 0 vs 1-4 (section A) and 0 vs 1 (section B).

Figure 5

Predicted probabilities of having a score of 0 (section A), 1 (section B), and ≥2 (section C) from a polychotomous ordinal logistic model with ICA wall thickness and extension of neo-terminal intramural lesions as covariates. Section A. In absence of intramural lesion (extension 0) of neo-terminal ileum, the predicted probability to have score 0 is > 80% (a) when ICA wall thickness is ≤ 3.5 mm and progressively decreases to < 15% for ICA wall thickness ≥ 8 mm (b). The probability to have score 0 progressively decreases from 75% (c) to 23% (d) for intramural lesions of the neo-terminal ileum increasing from 3 cm to 20 cm. Section B. In absence of intramural lesion (extension 0) of neo-terminal ileum, the predicted probability to have score 1, progressively increases from 23% (a) to 66% (b) for wall thickness of ICA increasing from 3.5 mm to 8 mm. In absence of intramural lesion (extension 0) of neo-terminal ileum, the probability to have score 1 with ICA wall thickness > 9 mm is low (< 50%) (c). When the extension of intramural lesions at the level of neo-terminal ileum increases from 3 to 20 cm, the probability to have score 1 progressively increases from 24% (d) to 66% (e). Section C. In absence of intramural lesion (extension 0) of neo-terminal ileum, the predicted probability to have score ≥2 is < 1% (a) when ICA wall thickness is ≤ 3.5 mm and progressively increases to >80% (b) for ICA wall thickness >10 mm. With ICA wall thickness ≥8 mm and with intramural lesions of the neo-terminal ileum increasing from 3 cm to 20 cm, the probability to have score ≥2 progressively increases from 27% (c) to 80% (d).