Transabdominal ultrasonography for small-bowel lesions in Crohn’s disease: a prospective comparison with balloon-assisted enteroscopy
Article information
Abstract
Background/Aims
Transabdominal ultrasonography (TUS) is increasingly used to assess small-bowel lesions in Crohn’s disease (CD); however, the diagnostic performance of prospective segment-specific data relative to balloon-assisted enteroscopy (BAE) remains limited.
Methods
In this prospective, single-center study, 94 patients with CD underwent TUS the day before BAE. The diagnostic accuracy of TUS for mucosal inflammation and stenosis was evaluated separately for the terminal ileum and proximal small bowel, using BAE as the reference standard and endoscopic criteria based on the Simple Endoscopic Score for Crohn’s Disease. We performed exploratory analyses of clinical factors associated with TUS detectability, including multivariable logistic regression for inflammatory lesions and exploratory comparisons for stenosis.
Results
For mucosal inflammation, TUS sensitivity and specificity were 69% and 94% in the terminal ileum (accuracy, 90%) and 91% and 82% in the proximal small bowel (accuracy, 84%), respectively. For stenosis, the sensitivity and specificity were 78% and 98% in the terminal ileum (accuracy, 96%) and 63% and 89% in the proximal small bowel (accuracy, 84%), respectively. In exploratory analyses of BAEconfirmed inflammatory lesions, anastomotic involvement, history of bowel resection, and small-bowel resection were associated with TUS detectability in univariate analyses. No independent predictors were identified in the multivariable logistic regression.
Conclusions
Compared to BAE as the reference standard, TUS showed high specificity and segment-dependent sensitivity for detecting small-bowel inflammation and stenosis. These findings support TUS as a complementary, practice-informing, and noninvasive assessment tool, indicating that additional evaluation may be needed when clinical suspicion persists despite negative or equivocal TUS findings.
INTRODUCTION
Crohn’s disease (CD) is an intractable disease that causes single or multiple lesions in any part of the gastrointestinal tract, particularly in the ileum and colon [1-3]. Disease progression frequently leads to complications such as strictures, fistulas, and abscesses, which impair the quality of life and complicate disease management [2,4]. Accurate assessment of disease activity and early detection of structural complications are essential for optimizing therapeutic strategies and improving long-term outcomes.
Multiple modalities are available for evaluating CD, including endoscopy, computed tomography enterography (CTE), and magnetic resonance enterography (MRE) [5-10]. Endoscopy allows direct visualization and tissue sampling but is invasive and has limited access to deep small-bowel lesions. CTE and MRE provide cross-sectional assessments of luminal and extraluminal diseases; however, CTE involves radiation exposure and MRE may be costly or less readily available [6-11].
Transabdominal ultrasonography (TUS) has emerged as an attractive noninvasive imaging tool for assessing CD. TUS enables real-time evaluation of bowel wall thickness (BWT), vascularity, motility, and complications, such as strictures or abscesses. It is widely available, well tolerated, repeatable, and free from radiation exposure, making it suitable for longitudinal assessment. Previous studies have demonstrated a strong correlation between TUS and colonoscopy in evaluating CD [12-14]. However, prospective direct comparisons with balloon-assisted enteroscopy (BAE), which can assess deep small-bowel segments beyond the reach of conventional endoscopy, remain limited [15-17]. Previous studies, including our 2023 study using intestinal ultrasound [17], mainly focused on stenosis detection rather than simultaneous assessment of both mucosal inflammation and stenosis across the terminal and proximal small-bowel segments.
In this prospective single-center study, we evaluated the diagnostic performance of TUS for small-bowel lesions in patients with CD using BAE as the reference standard. Specifically, we assessed the ability of TUS to detect (1) mucosal inflammation and (2) stenotic lesions in both the terminal ileum and proximal small bowel. Our aim was to provide prospective, segment-specific, practice-informing data on how TUS findings relate to deep small-bowel endoscopic assessment in a clinically indicated cohort.
METHODS
1. Patients
We prospectively recruited patients with CD who were admitted to our hospital for BAE with retrograde contrast evaluation of intestinal disease activity between January 2019 and December 2023. Although the primary focus was on small bowel lesions, a few patients with colonic-type CD were also included if small bowel involvement was suspected. Postoperative patients were eligible for inclusion if BAE was performed to evaluate postsurgical recurrence or guide further treatment.
The exclusion criteria were age <15 years and presence of stoma. All patients had a confirmed diagnosis of CD based on endoscopic and histological findings and were receiving medical therapy at the time of enrolment.
This was a pragmatic, prospective, diagnostic accuracy study embedded in routine clinical care without a formal a priori sample size calculation. All the enrolled patients were scheduled for clinically indicated BAE during the study period.
2. TUS Examination
TUS examinations were performed using the Aplio i800 system (Canon Medical Systems Corp., Otawara, Japan). Two experienced sonographers (with 3 and 8 years of experience in gastrointestinal ultrasonography, respectively) performed all examinations. TUS was conducted the day before BAE without bowel preparation such as laxatives. The patients fasted for approximately 3–5 hours before the examination. Prior to imaging, the sonographers were blinded to the BAE findings and were not provided with the patients’ clinical histories at the time of TUS. However, complete masking of the previous surgery was not possible in patients with visible abdominal surgical scars.
An initial survey of the entire abdomen was performed using a convex transducer (PVI-475BX), followed by detailed evaluation using a high-frequency linear array transducer (PLI-605BX). The small intestine was assessed in the following 2 segments: (1) the terminal ileum within 10 cm proximal to the ileocecal valve, and (2) the remaining small intestine proximal to the terminal ileum. The colon was examined in 3 segments: right, transverse, and left. The rectum was excluded from the evaluation due to technical limitations. Lesions at the ileocolonic anastomosis site were classified as being within the small intestinal segment.
Inflammatory mucosal findings were defined as BWT >3 mm [18], together with any intramural color Doppler signal (CDS present) using a high-frequency linear probe, which was chosen pragmatically to prioritize the detection of active mucosal inflammation while mitigating non-inflammatory thickening by requiring hyperemia. CDS was treated as a binary variable (present/absent). The CDS was optimized to detect low-velocity blood flow by minimizing the velocity range and filter values, while keeping the gain at the maximum level without generating artifacts.
The CDS was dichotomized (present/absent) rather than graded on a modified Limberg/IBUS scale [19], because a validated hyperemia grading for small-bowel segments was not routinely implemented at our center during the study period.
Stenosis was defined as 2 or more of the following criteria: (1) TUS-defined inflammation (BWT >3 mm with color Doppler); (2) proximal luminal dilation; and (3) a to-and-fro movement pattern of the intraluminal contents [17]. These sonographic definitions reflected our center’s pragmatic routine practice during study conduct and were established before publication of the 2024 STAR consensus on intestinal ultrasound for small-bowel strictures [20]. Accordingly, we did not prospectively apply standardized quantitative thresholds for luminal narrowing or prestenotic dilation. All findings, including the presence of inflammation and/or stenosis, were reviewed and confirmed through consensus to minimize interobserver variability.
3. BAE Examination
The details of BAE using retrograde contrast have been described previously [21]. Briefly, BAE was performed using a double-balloon enteroscope (EN-450T5; Fujifilm, Tokyo, Japan). The enteroscope was advanced into the small intestine as far proximally as possible via a retrograde approach.
In cases where stenotic lesions could not be traversed using this scope, balloon dilation was performed to facilitate further insertion. Following the deepest insertion, a contrast study was conducted using double-diluted meglumine diatrizoate (Bayer Aktiengesellschaft, Leverkusen, Germany) as the contrast medium. Balloon dilation to traverse strictures was performed at the endoscopist’s discretion, and long strictures judged to exceed approximately 50 mm (5 cm) were not dilated. When a stricture remained non-passable after attempted dilation, the lesion was classified as stenotic, and the mucosa beyond the stricture was not assessed endoscopically. Complete endoscopic assessment of mucosal inflammation in the corresponding proximal small-bowel segment was not possible in 1 case; in the remaining non-passable cases, mucosal inflammation could still be assessed within the endoscopically visualized extent. Contrast measurements (including stricture length and caliber) were recorded and used in the analysis.
The following parameters were measured during the contrast study: the number of strictures, minimum diameter of the strictures, maximum length of the strictures, maximum diameter of the pre-stenotic dilation, and diameter of the normal small intestine. Endoscopic findings were evaluated using the Simple Endoscopic Score for Crohn’s Disease (SES-CD). The SES-CD evaluates the following 4 endoscopic variables: (1) ulcer size, (2) ulcerated surface area, (3) affected surface area, and (4) presence of stenosis in 5 segments of the bowel (rectum, left colon, transverse colon, right colon, and terminal ileum). Each variable was graded from 0 (normal) to 3 (severe) per segment, and segmental and total endoscopic activities were quantified by summing the scores [22]. In this study, inflammatory mucosal findings at the terminal ileum and the proximal small bowel were defined as a total score of ≥ 4 in categories (1), (2), and (3) for each segment. Stenotic lesions were defined as those with a score of 3 in category (4) for each segment. In postoperative cases, the SES-CD was used for segmental scoring, with ileocolonic anastomotic ulcers counted as part of the small bowel segment.
The endoscopists were blinded to the TUS findings before the procedure. Segmental SES-CD scores (ulcer size, ulcerated surface, affected surface, and stenosis) were simultaneously recorded and no central reading was performed.
4. Statistical Analysis
All statistical analyses were performed using JMP software (version 17; SAS Institute Inc., Cary, NC, USA). The diagnostic performance of TUS and BAE was assessed by constructing 2 ×2 contingency tables for each segment. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy were calculated using BAE as the reference standard. For mucosal inflammation accuracy analyses, 1 proximal small-bowel segment was excluded from the 2×2 contingency tables because mucosal inflammation could not be assessed endoscopically because of a nonpassable stricture. In the remaining non-passable cases, the segments were classified as reference-positive or reference-negative according to the endoscopic findings within the visualized extent. For factors associated with an accurate diagnosis, univariate analyses were performed using the Shapiro–Wilk test for normality. Normally distributed continuous variables were compared using the Student’s t-test, non-normal variables were compared using the Mann–Whitney U test, and categorical variables were compared using Fisher exact test. Factors with P<0.05 were entered into a multivariate logistic regression. All P-values were two-sided, and P<0.05 was considered statistically significant.
5. Ethical Statements
This study was approved by the Institutional Review Board of Okayama University Hospital (IRB No. 1804-030). Written informed consent was obtained from all participants. All procedures were conducted in accordance with the principles outlined in the Declaration of Helsinki.
RESULTS
1. Patient Characteristics
In total, 102 patients with CD were enrolled in this study. Eight patients were excluded because of the presence of a stoma (n =6), inability to rule out intestinal tuberculosis (n =1), or technical difficulties in evaluating the bowel using TUS (n = 1). Therefore, 94 patients were included in the final analysis (Fig. 1).
Patient enrollment flowchart. Flow diagram showing the screening, inclusion, and exclusion of patients with CD enrolled in this study. A total of 102 patients were screened, of whom 8 were excluded because of the presence of a stoma, suspected intestinal tuberculosis, or technical limitations related to obesity or excessive intestinal gas. The final analysis included 94 patients with CD. CD, Crohn’s disease; TUS, transabdominal ultrasonography; BAE, balloon-assisted enteroscopy.
The baseline clinical characteristics of the study population are summarized in Table 1. The median patient age was 40 years (interquartile range, 28–48 years), and 66 (70.2%) patients were male. The median disease duration was 97.5 months (interquartile range, 36.3–215.0 months). Regarding the disease location, 33 patients (35.1%) had ileal-type disease, 57 (60.6%) had ileocolonic disease, and 4 (4.3%) had colonic-type disease. Patients with colonic-type disease were included if BAE was performed for suspected small-bowel lesions.
Characteristics of Patients with Crohn’s Disease
Forty patients (42.6%) had a history of small-bowel resection, including strictureplasty. A total of 64 patients (68.1%) received biological agents or Janus kinase inhibitors at the time of evaluation. No adverse events such as significant bleeding requiring transfusion or bowel perforation occurred during BAE with retrograde contrast.
This study cohort included patients with a broad spectrum of disease duration, location, and severity, providing an appropriate population for evaluating the diagnostic performance of TUS in detecting both inflammatory and stenotic lesions in the small intestine.
2. Diagnostic Performance of TUS for Inflammatory Lesions
Representative TUS and endoscopic images of normal and inflammatory terminal ileal lesions are shown in Fig. 2. The diagnostic performance of TUS for mucosal inflammation in the terminal ileum and proximal small bowel is summarized in Table 2. Sensitivity was 69% in the terminal ileum and 91% in the proximal small bowel, whereas specificity was 94% and 82%, respectively. The PPV and NPV were 69% and 94%, respectively, for the terminal ileum and 61% and 97%, respectively, for the proximal small bowel.
Normal and inflammatory lesions of the terminal ileum. (A) Case 1, normal terminal ileum on TUS showing a clear layered structure and wall thickness <3 mm. (B) Case 1, corresponding BAE image demonstrating intact mucosa. (C) Case 2, inflamed terminal ileum on TUS with wall thickening (>3 mm) and loss of stratification. (D) Case 2, color Doppler overlay highlighting hypervascularity in the inflamed segment. (E) Case 2, BAE image showing mucosal inflammation. Yellow arrows indicate a representative method for measuring the thickness of the bowel wall. TUS, transabdominal ultrasonography; BAE, balloon-assisted enteroscopy.
Diagnostic Accuracy of TUS for Mucosal Inflammation in the Terminal Ileum and Proximal Small Bowel
Endoscopic evidence of colonic inflammation was infrequent in all the segments. The diagnostic performance of TUS for colonic lesions is summarized in Supplementary Table 1.
These findings suggest that TUS is a useful noninvasive tool for detecting small-bowel inflammation in patients with CD. However, the present study did not directly evaluate whether TUS reduces the need for BAE, changes clinical decision-making, or improves patient outcomes.
3. Diagnostic Performance of TUS for Stenotic Lesions
Representative TUS and endoscopic images of proximal small-bowel stenosis are shown in Fig. 3. The diagnostic performance of TUS for detecting stenotic lesions in the terminal ileum and proximal small bowel is summarized in Table 3. Sensitivity was 78% in the terminal ileum and 63% in the proximal small bowel, whereas specificity was 98% and 89%, respectively. The PPV and NPV were 88% and 97%, respectively, for the terminal ileum and 56% and 92%, respectively, for the proximal small bowel.
Representative images of proximal small-bowel stenosis. (A) TUS demonstrating marked luminal narrowing with bowel wall thickening and prestenotic dilation suggestive of a stenotic lesion. (B) BAE showing a tight proximal small-bowel stricture with fibrotic changes consistent with advanced Crohn’s disease. Yellow arrows indicate stenotic lesions. TUS, transabdominal ultrasonography; BAE, balloonassisted enteroscopy.
Diagnostic Accuracy of TUS for Intestinal Stenosis in the Terminal Ileum and Proximal Small Bowel
Endoscopic stenotic lesions were not detected in any segment of the colon. The diagnostic performance of TUS for colonic stenosis is summarized in Supplementary Table 2.
The number of colonic stenoses was zero, and none of the cases met the predefined sonographic criteria for colonic stenosis.
4. TUS Detectability of Inflammatory Lesions
The clinical and disease characteristics of patients with inflammatory lesions detected by TUS and those not detected by TUS are summarized in Table 4. Among the 43 patients with BAE-confirmed small-bowel inflammation, 16 were classified as TUS-undetectable (misdiagnosis) and 27 as TUS-detectable (correct diagnosis). Baseline characteristics, including sex, age, disease duration, body mass index, and inflammatory location (terminal ileum vs. proximal small bowel), were comparable between the groups. In univariate analyses, anastomotic involvement was more frequent in the TUS-detectable group than in the TUS-undetectable group (74.1% vs. 31.3%; P=0.006). A history of bowel resection (74.1% vs. 37.5%; P=0.011) and prior small-bowel resection, including strictureplasty (70.4% vs. 25.0%; P=0.005), were also more common in the TUS-detectable group. However, in the multivariable logistic regression analysis, none of these factors remained independently associated with the TUS detectability of inflammatory lesions. These exploratory findings suggest that postoperative/anastomotic status may influence TUS detectability in univariate comparisons; however, given the limited sample size, correlated surgical variables, and likely technical/anatomical determinants of visualization, the absence of independent predictors should be interpreted cautiously and regarded as hypothesis-generating.
Comparison of Characteristics between TUS-Detectable and TUS-Undetectable Inflammation in 43 Patients with Inflammation Diagnosed by BAE
5. TUS Detectability of Stenotic Lesions
The clinical and stenotic characteristics of patients with stenotic lesions detected by TUS and those not detected by TUS are shown in Table 5. Among the 26 patients with BAE-confirmed small-bowel stenosis, 10 were classified as TUS-undetectable (misdiagnosis) and 16 as TUS-detectable (correct diagnosis). The baseline characteristics, including sex, age, disease duration, body mass index, stenosis location (terminal ileum vs. proximal small bowel), and presence of anastomotic stenosis, were comparable between the groups. The history of bowel resection, including ileocecal and small-bowel resections (including strictureplasty), was similar between groups. Regarding BAE-derived stenosis characteristics, the number of stenoses and stenosis length did not differ significantly between the groups, and the stenosis diameter did not vary significantly. However, smaller diameters were observed more frequently in TUS-undetectable cases. These exploratory findings suggest that the TUS detectability of small-bowel stenosis is not clearly explained by the patient background or stenosis location/length alone, while a narrower luminal diameter may contribute to missed detection. Given the limited sample size, these observations should be interpreted as hypothesis-generating.
Comparison of Characteristics between TUS-Detectable and TUS-Undetectable Stenosis in 26 Patients with Stenosis Diagnosed by BAE
DISCUSSION
This prospective study compared TUS with BAE to evaluate small-bowel lesions in patients with CD. Using BAE as the reference standard, TUS showed high overall accuracy for both mucosal inflammation and stenosis (84%–96% across segments), with consistently high specificity (82%–98%) and sensitivity (63%–91%), which varied by segment and phenotype. These findings provide prospective, practice-informing data supporting TUS as a complementary noninvasive modality alongside endoscopic assessment. Previous studies have shown that intestinal ultrasound findings correlate well with endoscopic activity in CD, particularly in accessible bowel segments [12-14]. However, prospective comparisons with deep small-bowel endoscopy remain limited [15-17]. By using BAE, which is capable of evaluating lesions beyond the reach of conventional ileocolonoscopy, as the reference standard, our study adds segment-specific data on the diagnostic performance of TUS in both the terminal ileum and more proximal small-bowel segments.
For inflammatory lesions, TUS demonstrated very high specificity and NPV (terminal ileum, 94% and 94%, respectively; proximal small bowel, 82% and 97%, respectively). Thus, a negative examination may be reassuring in selected clinical contexts, particularly when the pretest probability is not high; however, this study did not directly evaluate downstream management decisions or outcomes. The sensitivity varied by segment (69% in the terminal ileum vs. 91% in the proximal small bowel). This variability may reflect anatomical factors (e.g., bowel gas, depth, and loop mobility) and the fact that our sonographic definition required both bowel wall thickening (>3 mm) and hyperemia on color Doppler imaging.
Notably, anastomotic involvement, history of bowel resection, and prior small-bowel resection (including strictureplasty) were associated with TUS detectability of BAE-confirmed inflammation in univariate analyses; however, no independent predictors were identified in multivariable logistic regression. This analysis should be interpreted with caution because it was based on a relatively small subset of patients with BAE-confirmed inflammatory lesions (n=43), with a limited number of events relative to candidate variables and clinically related postoperative variables that may be collinear. Therefore, the absence of independent predictors may reflect limited statistical power, overfitting risk, and type II error, rather than the true absence of association. More broadly, TUS detectability is likely to be influenced by multifactorial technical and anatomical determinants, including bowel gas, lesion depth, loop mobility, lesion orientation, and postoperative anatomy, rather than by a single clinical characteristic.
For stenosis, TUS demonstrated excellent specificity, particularly in the terminal ileum (98%), supporting its role as a rulein test when the predefined sonographic criteria were met. The sensitivity was moderate and varied by segment (78% in the terminal ileum and 63% in the proximal small bowel), and the PPV was lower in the proximal segment (56%). Because our definition of stenosis requires at least 2 of the 3 criteria—TUS-defined inflammation, prestenotic dilation, and a to-and-fro movement pattern of intraluminal contents [17]—false negatives can occur when strictures are very tight or short without clear upstream dilation, and false positives may occur when functional stasis or spasm mimics impaired passage. In our exploratory analyses, stricture diameter was numerically smaller in TUS-undetectable cases, suggesting that luminal caliber and dynamic passage findings may influence sonographic visibility; these observations should be regarded as hypothesis generating.
Clinically, these results support TUS as a complementary, repeatable, and noninvasive assessment for small-bowel CD. TUS may be useful as an initial or interval test to screen for active inflammation and identify patients for whom further evaluation should be considered. However, the present study did not directly demonstrate a reduction in the need for BAE, changes in clinical decision-making, or improved patient outcomes. When clinical suspicion remains high despite negative or equivocal TUS findings, particularly for proximal disease, suspected high-grade stenosis, or discordant symptoms, cross-sectional imaging (MRE/CTE) and/or BAE should still be considered to avoid underestimation of the disease burden and to guide procedural or surgical planning [6,7,23-26].
The strengths of this study include its prospective design, short interval between TUS and BAE (1 day), and blinding of endoscopists to the TUS findings. In addition, we evaluated both inflammatory and stenotic phenotypes in clinically relevant small bowel segments using pragmatic sonographic definitions and a consistent endoscopic reference standard, enhancing the clinical relevance of our findings for real-world implementation.
This study has several limitations that warrant consideration. This was a single-center cohort enriched for patients undergoing clinically indicated BAE, which may have increased the disease prevalence and limited the generalizability to lower-risk populations. Ultrasound remains operator dependent; we did not perform formal interobserver reliability testing, and color Doppler hyperemia was dichotomized rather than graded. In addition, our pragmatic sonographic criteria for stenosis were not fully aligned with the subsequently published 2024 STAR consensus [20], which defines small-bowel strictures using bowel wall thickening, luminal narrowing, and prestenotic dilation with standardized quantitative thresholds. Specifically, our definition coupled bowel wall thickening with Doppler hyperemia and relied on proximal dilation and dynamic intraluminal to-and-fro movement rather than prospectively applying strict quantitative thresholds for luminal narrowing or prestenotic dilation. This approach may have reduced the sensitivity for fibrotic or relatively hypovascular strictures and limited the reproducibility across centers. Comprehensive sonographic still images and cine-loops were not systematically archived for all examinations during the study period; therefore, a reliable retrospective re-analysis applying the 2024 STAR-based criteria was not feasible. The analyses of TUS detectability were exploratory, and the multivariable logistic regression for inflammatory lesions was likely underpowered because it was performed in a relatively small subset with limited events and potentially collinear postoperative variables. Accordingly, the lack of independent predictors should not be interpreted as evidence that no such association exists. In addition, TUS detectability is likely to be influenced by technical and anatomical factors that are not fully captured in the clinical dataset. Although the sonographers were blinded to the BAE findings and were not provided with the clinical histories at the time of TUS, complete masking of the prior surgery was not possible in patients with visible abdominal surgical scars, which could have partially unblinded examiners in some postoperative cases. Furthermore, segmentation of the small bowel (terminal ileum vs. proximal small bowel) can be subjective, and BAE may not fully assess the mucosa beyond non-passable strictures; therefore, such segments were not used as reference-positive or reference-negative for mucosal inflammation accuracy analyses. Future multicenter studies using standardized STAR-based acquisition and scoring protocols, formal assessment of interobserver agreement, and external validation are warranted to define the optimal diagnostic pathways that combine TUS with cross-sectional imaging and endoscopy.
In conclusion, in this prospective comparison using BAE as the reference standard, TUS showed high specificity and overall accuracy for detecting both inflammatory and stenotic small-bowel lesions in CD, with sensitivity varying by segment and lesion characteristics. These findings support TUS as a complementary, practice-informing, noninvasive assessment tool, while emphasizing that the observed performance should be interpreted in light of the pragmatic sonographic definitions used, and that complementary investigations remain important when clinical suspicion is high or the disease is suspected in technically challenging segments.
Notes
Funding Source
The authors received no financial support for the research, authorship, and/or publication of this article.
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Data Availability Statement
Data analyzed in this study are available from the corresponding author upon reasonable request.
Author Contributions
Conceptualization: all authors. Data curation: Hiraoka S, Takeuchi K, Takei K, Inokuchi T, Aoyama Y, Igawa S, Ishiguro M, Toyosawa J, Yamasaki Y, Kinugasa H, Onishi H, Otsuka M. Formal analysis: Hiraoka S. Investigation: Hiraoka S, Takeuchi K, Takei K, Inokuchi T, Aoyama Y, Igawa S, Ishiguro M, Toyosawa J, Yamasaki Y, Kinugasa H, Onishi H, Otsuka M. Methodology: Takahara M, Hiraoka S. Project administration: Takahara M, Hiraoka S. Writing–original draft: Takahara M, Hiraoka S. Writing–review & editing: all authors. Approval of final manuscript: all authors.
Supplementary Material
Supplementary materials are available at the Intestinal Research website (https://www.irjournal.org).
Supplementary Table 1.
Diagnostic Accuracy of TUS for Colonic Mucosal Inflammation
Supplementary Table 2.
Diagnostic Accuracy of TUS for Colonic Stenosis