Baseline and Postinduction Intestinal Ultrasound Findings Predict Long-term Transmural and Mucosal Healing in Patients With Crohn’s Disease

Abstract

Background

Intestinal ultrasound (IUS) is becoming a standard assessment tool in Crohn’s disease (CD), but limited data exist on its ability to predict long-term objective outcomes. Therefore, we aimed to investigate the predictive value of IUS findings for long-term transmural healing (TH) and mucosal healing (MH) in CD.

Methods

We prospectively included consecutive CD patients with active endoscopic disease and bowel wall thickness (BWT) >3.0 mm, initiating infliximab. Intestinal ultrasound parameters (ie, BWT, inflammatory mesenteric fat [i-fat], bowel blood flow and stratification) and International Bowel Ultrasound Segmental Activity Score (IBUS-SAS) were collected at baseline, after 14 to 26 weeks (visit 1, postinduction) and 44 to 56 weeks (visit 2). Transmural healing (normalization of all IUS parameters) and MH (SES-CD ≤2) were assessed at visit 2.

Results

One hundred twenty-nine patients were evaluated. At visit 2, 38.0% and 48.1% of patients achieved TH and MH, respectively. All the IUS parameters and IBUS-SAS showed improvement at visit 1 and visit 2 compared with the baseline (all P < .001). Multivariable analysis found that presence of i-fat at baseline (odds ratio [OR], 0.57; P = .008) and greater postinduction BWT (OR, 0.24; P < .001) were negative predictors for TH, while higher baseline (OR, 0.98; P = .013) and postinduction (OR, 0.94; P < .001) IBUS-SAS predicted negatively for MH. Postinduction BWT <4.5mm best predicted TH (AUC 0.85; P < .001), while postinduction IBUS-SAS <25.0 best predicted MH (AUC 0.82; P < .001). Moreover, colonic disease was associated with higher risk of TH (OR, 2.55; P = .027), and disease duration >24 months with lower risk of MH (OR, 0.27; P = .006).

Conclusions

Baseline and postinduction IUS findings reliably predict long-term TH and MH in patients with CD receiving infliximab.

Graphical Abstract

Open in new tabDownload slide

Introduction

In the era of biologics, mucosal healing (MH) has become a crucial therapeutic goal in Crohn’s disease (CD). Additionally, transmural healing (TH) has been proposed as an indicator of remission depth.¹ Mucosal healing and TH have been associated with improved long-term clinical outcomes.^1–3

Frequent ileocolonoscopies are invasive and not always well-tolerated by patients. Furthermore, in some cases, lesions may be situated beyond the scope of conventional endoscopies. Magnetic resonance enterography (MRE) can also be limited by lengthy acquisition times and high costs. However, intestinal ultrasound (IUS) is noninvasive, radiation-free, repeatable, cost-effective, and well-accepted by patients. Systematic reviews have reported that IUS has comparable accuracy to MRE in assessing disease activity in inflammatory bowel disease (IBD).⁴ Some IUS parameters are well-correlated with endoscopic activity.^5–7 Recent studies have also shown the ability of IUS to monitor treatment response in CD.^8,9 Although several IUS scores for use in CD exist, they have not been adequately validated.¹⁰ To our knowledge, the value of IUS parameters and International Bowel Ultrasound Segmental Activity Score (IBUS-SAS) in predicting MH and TH have not yet been clearly defined.^11,12

Therefore, the primary objective of this study was to investigate the predictive value of baseline and postinduction IUS parameters and IBUS-SAS for long-term TH and MH in patients with CD receiving infliximab (IFX). Additionally, the study aimed to assess the response of IBUS-SAS to IFX treatment.

Materials and Methods

Study Design and Patients

The study participants were derived from the MORE study (Multi-omics to Predict Responses to Biologics in IBD) conducted at our center since January 2020, which was a prospective observational cohort study (NCT05542459). This study was approved by the local ethical committee of Sun Yat-sen University Sixth Affiliated Hospital, and all patients provided their informed consent. From January 2020 to April 2022, patients who met the following criteria were recruited for the present study: (1) confirmed diagnosis of CD,¹³ (2) presence of at least 1 segment with a bowel wall thickness (BWT) >3.0mm in terminal ileum or colon, (3) presence of at least 1 segment with Simple Endoscopic Score for Crohn’s Disease (SES-CD) ≥3, (4) receiving IFX for at least 44 to 56 weeks in accordance with clinical practice guidelines,¹³ and (5) undergoing IUS before, at 14 to 26 weeks, and 44 to 56 weeks after initiating IFX, as well as undergoing ileocolonoscopy before and at 44 to 56 weeks after initiating IFX. Patients with obesity (body mass index >30 kg/m²), ileal or colonic stoma, bowel resection history, concomitant treatment including total enteral nutrition or other biologics, or who were pregnant were excluded.

Infliximab therapy was administered with 5 mg/kg at week 0, 2, and 6 for induction followed by every 8 weeks for maintenance. Treatment was intensified reactively on patients suffering from loss of clinical response with objective evidence of active disease (increased levels of serum C-reactive protein [CRP] or fecal calprotectin, or the presence of endoscopic or IUS lesions) and trough IFX levels below 3 μg/mL.

Demographic data were collected at baseline. At baseline, week 14 to 26 (visit 1, postinduction), and week 44 to 56 (visit 2), HBI score and CRP levels were collected, and IUS was performed. Ileocolonoscopy was performed at baseline and visit 2. The baseline IUS and ileocolonoscopy were performed within 6 weeks prior to initiating IFX. For visits requiring both IUS and endoscopy, a minimum 24-hour interval was mandatory, or the IUS examination could be conducted before bowel preparation for ileocolonoscopy. It aimed to minimize the potential bowel wall edema caused by bowel preparation, which might affect the measurement of BWT. The median interval between IUS and ileocolonoscopy was 2 days (range, 0-27) at baseline and visit 2. There were no major changes in treatment during the intervening period between the 2 examinations. Major treatment changes were defined as the initiation or discontinuation of steroids and immunomodulators, as well as the commencement of biologics.

Intestinal Ultrasound

Intestinal ultrasound examinations were performed by one of 3 experienced sonographers (W.C., S.Q., and Z.L.) with 5 to 8 years of experience in IUS. Patients were examined in a fasting state, and all the examinations were performed using the same equipment (LOGIQ E9; GE Healthcare, Milwaukee, WI, USA). Initial detection was performed using low-frequency curved array transducer (3.0-5.0MHz), followed by detailed examination using high-resolution linear array probe (8.4-11.0MHz). Intestinal ultrasound assessed the terminal ileum (the distal 15cm of ileum), cecum-ascending, transverse, descending-sigmoid colon.

The key IUS parameters of the target segment, including BWT, color doppler signal (CDS) of bowel blood flow, bowel wall stratification (BWS) and inflammatory mesenteric fat (i-fat), are assessed and classified as detailed in Table 1.¹⁴ To establish a consensus on the standardization of parameters, specific training sessions were conducted based on the guidelines for IUS.¹⁴ The specifications of IUS measurements are described in the Supplementary Material. The most affected segment in IUS was defined as the segment with the largest BWT at baseline. This target segment was monitored throughout the study and used for analysis. However, due to generally poor imaging quality, the rectum was not routinely examined and was not identified as the most affected segment in any case. Intestinal ultrasound data collected by the sonographers for each participant were recorded in a case report form (CRF). The sonographers were not blinded to other clinical data. We also calculated IBUS-SAS for the most affected segment using the raw data from the CRF (Table 1).¹¹

Ileocolonoscopy

Ileocolonoscopies were performed by experienced gastroenterologists with expertise in IBD. The SES-CD was utilized to score the endoscopic activity. The most affected segment in endoscopy was defined as the segment with the highest segmental SES-CD at baseline. Ileocolonoscopy data collected by the gastroenterologists for each participant were also recorded in a CRF. The gastroenterologists were not blinded to other clinical data.

End Points and Definitions

The primary end points of this study were TH and MH at visit 2. Transmural healing was defined as BWT ≤3.0 mm, preserved BWS, CDS 0-1, and the absence of i-fat in the most affected segment identified by IUS at baseline. Overall MH was defined as a total SES-CD ≤2, while segmental MH was defined as SES-CD of 0 and was only analyzed in segments with SES-CD ≥3 at baseline.¹

Statistics Analysis

Descriptive data were presented as the mean or median, depending on the data distribution. Categorical data are presented as frequencies. Descriptive data were compared using the Student t test or Mann–Whitney U test. Categorical data were compared using the χ² test or Fisher exact test. The agreement of the most affected segment identified by IUS and endoscopy was assessed using kappa statistics. The correlation between IBUS-SAS, SES-CD, HBI, and CRP was determined using Spearman correlation coefficient. Receiver operating characteristics (ROC) curve analysis was conducted to determine appropriate cut-off values for BWT and IBUS-SAS for predicting clinical outcomes. Logistic regression models were used to identify optimal IUS findings that could predict MH and TH. This was done by first fitting univariate models and then searching for the best model fit by using the stepwise approach. All models were validated using a multifold cross-validation (10 times 5-fold cross-validation) approach. For each final multivariable model of IUS findings, the accuracy (percentage of correct classification) and area under the ROC curve (AUC) were presented. The odds ratio (OR) values of the optimal IUS findings for predicting TH and MH were also adjusted for factors such as combination treatment, IFX optimization, disease duration, behavior, and location. We used SPSS V.26 for all descriptive and analytical procedures, and R V.4.2.2 was used for all modeling analyses. Statistical significance was considered at P < .05.

Results

Patient Characteristics

From January 2020 to April 2022, we included 129 consecutive CD patients with active disease and initiating treatment of IFX from the MORE study. The study flowchart is presented in Supplementary Figure 1. Clinical characteristics of patients enrolled are reported in Table 2. After inclusion, 12 (9.3%) patients were concomitantly receiving steroids and 47 (36.4%) patients receiving thiopurines/methotrexate. During the study period, 20 (15.5%) patients needed treatment intensification of IFX based on objective evidence of active disease. The most affected segments identified by IUS in 74 patients were located in the terminal ileum, while 55 were in the colon. Additionally, the most affected segments identified by endoscopy in 65 patients were located in the terminal ileum, while 64 were in the colon. The correlation of the most affected segments identified by IUS and endoscopy was good (Kappa = 0.829; P < .001).

IUS and Ileocolonoscopic End Points

At visit 2, 38.0% (49 of 129) and 48.1% (62 of 129) of patients achieved overall TH and MH. Rate of TH was numerically higher in the colon than the terminal ileum (45.5% [25 of 55] vs 32.4% [24 of 74]; P = .132; Figure 1). A total of 90 patients (69.8%) had a SES-CD ≥3 in the terminal ileum at baseline, 96 (74.4%) in the cecum-ascending colon, 64 (49.6%) in the transverse colon, 82 (63.6%) in the descending-sigmoid colon, and 76 (58.9%) in the rectum. For segment-level analysis, 50.0% (45 of 90) terminal ileum segments achieved segmental MH, and 57.5% (183 of 318) of colorectal segments achieved segmental MH (cecum-ascending colon 56.3% [54 of 96]; transverse colon 66.7% [42 of 64]; descending-sigmoid colon 53.7% [44 of 82]; and rectum 56.6% [43 of 76]; Figure 1). Four terminal ileums were inaccessible by endoscopy due to intestinal stenosis and were classified as non-MH in the segment-level analysis.

IUS Parameters and IBUS-SAS Changes

Supplementary Table 1 presents HBI, CRP level, SES-CD, IUS parameters, and IBUS-SAS at each visit. Notably, all markers of active inflammation mentioned previously exhibited a significant improvement at visit 1 and visit 2 compared with their baseline levels, respectively.

Patients with TH exhibited significantly lower BWT and IBUS-SAS at each visit compared with patients without TH (all P < .05; Figure 2). The decrease in BWT and IBUS-SAS at visit 1 (ΔBWT: −2.4 ± 1.4 mm vs −1.3 ± 1.3 mm, P < .001; ΔIBUS-SAS: −35.1 ± 21.9 vs −21.7 ± 24.2, P < .01) and visit 2 (ΔBWT: −3.2 ± 1.3 mm vs −1.2 ± 1.5 mm, P < .001; ΔIBUS-SAS: −44.5 ± 22.5 vs −18.4 ± 27.7, P < .001) from baseline were significantly greater in patients with TH compared with patients without TH (Figure 2). Additionally, the proportion of normal BWS (85.7% vs 50.0%; P < .001) and CDS (89.8% vs 57.5%; P < .001) were higher at visit 1 in patients with TH compared with patients without TH (Supplementary Figure 2). The proportions of i-fat at each visit were consistently higher in patients without TH (all P < .001; Supplementary Figure 2).

Patients with MH also had significantly lower BWT and IBUS-SAS at each visit compared with patients without MH (all P < .05; Figure 2). The decrease in BWT and IBUS-SAS at visit 1 (ΔBWT: −2.0 ± 1.4 mm vs −1.4 ± 1.5 mm, P = .003; ΔIBUS-SAS: −33.9 ± 22.0 vs −20.1 ± 24.3, P = .001) and visit 2 (ΔBWT: −2.7 ± 1.5 mm vs −1.3 ± 1.7 mm, P < .001; ΔIBUS-SAS: −42.0 ± 23.2 vs −15.7 ± 27.6, P < .001) from baseline were significantly greater in patients with MH compared with patients without MH (Figure 2). Additionally, the proportion of normal BWS (83.9% vs 44.8%, P < .001) and CDS (87.1% vs 53.8%, P < .001) were higher at visit 1 in patients with MH compared with patients without MH (Supplementary Figure 2). These trends were consistent at visit 2 (BWS: 93.5% vs 40.3%, P < .001; CDS: 98.4% vs 38.8%, P < .001; Supplementary Figure 2). The proportions of i-fat at each visit remained consistently higher in patients without MH (all P < .001; Supplementary Figure 2).

Baseline and Postinduction IUS Predictors of TH at Visit 2

Based on baseline IUS findings, univariate logistic model showed that presence of i-fat and higher IBUS-SAS were associated with a lower rate of TH at visit 2 (Table 3). Subsequently, the multivariate model showed that presence of i-fat was the optimal independent baseline IUS predictor associated with lower rate of TH at visit 2 (OR, 0.52; 95% CI, 0.35-0.75; P = .001), with a fair accuracy of 60.5% and AUC of 0.70 for validation of the model (Table 3). The adjusted OR value for the presence of i-fat was 0.57 (95% CI, 0.38-0.87; P = .008) in the multivariate logistic analysis (Supplementary Table 2). In this multivariate analysis, colonic disease (with ileum as a reference) was associated with a higher rate of TH at visit 2 (OR, 2.55; 95% CI, 1.11-5.86; P = .027), while treatment intensification of IFX during the study was associated with a lower rate of TH at visit 2 (OR, 0.15; 95% CI, 0.03-0.75; P = .021; Supplementary Table 2).

Based on postinduction IUS findings, the univariate logistic model showed that IBUS-SAS, BWT, BWS, CDS, i-fat, and decrease in IBUS-SAS and BWT were independent predictors for TH at visit 2 (Table 3). Subsequently, multivariate model showed that a greater BWT was the optimal independent postinduction IUS predictor associated with lower rate of TH at visit 2 (OR, 0.23; 95% CI, 0.13-0.39; P < .001), with an acceptable accuracy of 73.7% and AUC of 0.84 for validation of the model (Table 3). The adjusted OR value of postinduction BWT was 0.24 (95% CI, 0.14-0.42; P < .001) in the multivariate logistic analysis (Supplementary Table 2). This indicates that for every 1.0 mm increase in postinduction BWT at visit 1, the likelihood of achieving TH at visit 2 decreases by 76%. The ROC curve analysis revealed that a postinduction BWT <4.5 mm was the best predictor for achieving overall TH at visit 2 (sensitivity 85.7%, specificity 68.8%, AUC 0.85; 95% CI, 0.78-0.92; P < .001), as well as for both terminal ileum (sensitivity 91.7%, specificity 64.0%, AUC 0.86; 95% CI, 0.78–0.94; P < .001) and colonic segments (sensitivity 80.0%, specificity 76.7%, AUC 0.84; 95% CI, 0.73-0.94; P < .001; Figure 3).

Baseline and Postinduction IUS Predictors of MH at Visit 2

Based on the most affected segments identified by IUS at baseline, the univariate logistic model showed that all the IUS findings at baseline and postinduction were independent predictors for segmental MH at visit 2, except for a decrease in BWT (Table 4). Subsequently, multivariate model showed that a higher IBUS-SAS was the optimal independent IUS predictors associated with lower rate of segmental MH at visit 2, both at baseline (OR, 0.97; 95% CI, 0.95-0.99; P < .001) and postinduction (OR, 0.94; 95% CI, 0.91-0.96; P < .001). Baseline and postinduction models for predicting segment MH showed a fair to acceptable accuracy of 65.7% and 74.3%, and AUC of 0.73 and 0.80 in the validation process, respectively (Table 4). The adjusted OR value of baseline IBUS-SAS was 0.97 (95% CI, 0.96-0.99; P = .002) in the multivariate logistic analysis, while postinduction IBUS-SAS was 0.93 (95% CI, 0.90-0.96; P < .001; Supplementary Table 3). The ROC curve analysis showed that a baseline IBUS-SAS <66.5 (sensitivity 71.4%, specificity 69.7%, AUC 0.70; 95% CI, 0.60–0.79; P < .001) and a postinduction IBUS-SAS <25.0 (sensitivity 78.6%, specificity 74.2%, AUC 0.81; 95% CI, 0.74-0.89; P < .001) were the best predictors for achieving segmental MH at visit 2 (Figure 3).

The predictive value of baseline and postinduction IBUS-SAS for overall MH at visit 2 were presented in Supplementary Table 4. Multivariate analysis showed that a higher IBUS-SAS at baseline (OR, 0.98; 95% CI, 0.96-1.00; P = .013) and postinduction (OR, 0.94; 95% CI, 0.92-0.97; P < .001) were associated with lower risk of overall MH at visit 2. The ROC curve analysis demonstrated that a baseline IBUS-SAS <66.5 (sensitivity 64.5%, specificity 68.7%, AUC 0.66; 95% CI, 0.56-0.75; P < .001) and a postinduction IBUS-SAS <25.0 (sensitivity 75.8%, specificity 79.1%, AUC 0.82; 95% CI, 0.74-0.89; P < .001) were the best predictors for achieving overall MH at visit 2 (Figure 3). The multivariate analysis, including baseline variables, also revealed that disease duration >24 months (OR, 0.27; 95% CI, 0.11-0.69; P = .006) and treatment intensification of IFX during the study (OR, 0.11; 95% CI, 0.02-0.5; P = .008) were associated with lower risk of overall MH at visit 2 (Supplementary Table 4).

Correlation Among HBI, CRP, BWT, IBUS-SAS, and SES-CD

At baseline and visit 2, the correlation between BWT and segmental SES-CD was strong (rho = 0.82, P < .001), as was the correlation between IBUS-SAS and segmental SES-CD (rho = 0.82, P < .001; Supplementary Figure 3). However, BWT, IBUS-SAS and SES-CD demonstrated only moderate correlation with the HBI (rho = 0.64, rho = 0.67, and rho = 0.70, all P < .001) and CRP (rho = 0.66, rho = 0.66, and rho = 0.69, all P < .001). The ROC curve analysis indicated that BWT >4.0 mm (sensitivity 91.3%, specificity 89.2%, AUC 0.95; 95% CI, 0.93-0.98; P < .001) and IBUS-SAS >32.5 (sensitivity 88.8%, specificity 92.8%, AUC 0.96; 95% CI, 0.94-0.98; P < .001) were the best detectors for endoscopic activity (defined as SES-CD >2; Supplementary Figure 3).

Discussion

Transmural healing is currently the subject of intense investigation and has recently been proposed as a potential treatment goal in CD. In our study, TH was achieved in 38.0% of patients after 44 to 56 weeks of IFX therapy, which was comparable to previous studies.^3,15 The advancement of technology and growing knowledge of IUS have led to the wider use of this technique, particularly for monitoring TH in CD. Identifying baseline and postinduction IUS features that can predict long-term TH may provide valuable insights for optimizing treatment strategies.

A novel finding in our research was that presence of i-fat at baseline was an independent negative predictor for TH at visit 2. Few studies have previously explored the predictive value of IUS-measured i-fat for TH. Inflammatory mesenteric fat is linked to inflammatory factors and histological changes in the connective tissue of the bowel wall, such as muscular hypertrophy and intestinal fibrosis,¹⁶ which may influence the normalization of BWT. A previous study has also found that i-fat, as measured by MRE, is highly predictive of nonhealing of severe inflammatory lesions.¹⁷ In previous studies, we also observed that colonic disease was associated with higher rate of TH at visit 2.^8,9 The exact reason for the different ultrasonographic response to IFX in the ileum vs colon remains unclear. Anatomical differences, such as the predominant localization of Paneth cells and Peyer’s patches in the small intestine, or differences in gut microbiota, could potentially contribute to the distinct ultrasonographic response to IFX.¹⁸ A total of 20 (15.5%) patients needed treatment intensification of IFX during study period due to objective evidence of active disease and lower trough IFX level. While treatment intensification was associated with a lower rate of TH and MH at visit 2, a notable proportion of patients demonstrated disease improvement after treatment intensification (8 of 20 achieved endoscopic response and 7 of 20 achieved transmural response). Bowel wall thickness is the most crucial parameter in IUS for diagnosing and assessing disease activity in CD, and its predictive value for TH has been observed in a prior study.⁸ However, the optimal cut-off value of postinduction BWT for predicting long-term TH has not been reported in previous studies. Our study demonstrated that patients with a postinduction BWT <4.5 mm were more likely to achieve TH at visit 2. Thus, we propose that postinduction BWT <4.5 mm not only serves to define inflammatory status but also represents a potential new therapeutic target in the mid-term, as it is related to long-term TH.

The limitations of ileocolonoscopy for regularly monitoring treatment response are widely acknowledged. Despite prior studies having investigated the association between endoscopic response and IUS response, the relationship between baseline and postinduction IUS features and long-term MH has rarely been reported.^9,19 A recent study demonstrated that a decrease of 18% in BWT from baseline after 4 to 8 weeks of IFX treatment predicted endoscopic response at week 12 to 34.¹⁹ However, in our study, the decrease in BWT at visit 1 was not the optimal predictor for MH at visit 2. This discrepancy is likely attributable to the difference in the predicted outcomes (endoscopic response vs MH). We identified IBUS-SAS as the optimal IUS finding at baseline and postinduction for predicting MH at visit 2. The IBUS-SAS is a comprehensive index that integrates key parameters of IUS.¹¹ It potentially provides a richer understanding of bowel lesions and improves the prediction of subsequent clinical outcomes compared with using a single IUS parameter. These findings suggest that early IUS results can accurately predict long-term MH in terminal ileum and colon. Further investigation is required to determine its predictive role in small intestine segments that are not visualized by ileocolonoscopy.

Establishing a standardized and reproducible IUS score is of great importance for expanding its use in clinical trials and wider clinical adoption. While several IUS scores have been proposed, most studies had suboptimal methodology for development.¹⁰ Recently, an expert consensus identified key activity parameters and proposed IBUS-SAS using expert consensus and standardized approaches, which demonstrated excellent reliability.¹¹ However, external validation in monitoring treatment response remained insufficient. Our current study partially addressed this gap. We found that IBUS-SAS was responsive to the treatment of IFX, showing a significant decrease at visit 1 and visit 2, similar to HBI, CRP, and SES-CD. Additionally, IBUS-SAS exhibited a strong correlation with SES-CD, which was similar to the findings of a recent study conducted by Dragoni et al.²⁰ The accuracy of IBUS-SAS for predicting endoscopic activity was also comparable to that reported by Dragoni et al (AUC 0.96 vs 0.96).²⁰ Based on these findings, IBUS-SAS can be considered as a reliable tool for monitoring treatment response and predicting endoscopic activity. In this study, BWT also demonstrated high accuracy in predicting endoscopic activity, consistent with a previous study conducted by Voogd et al (AUC 0.95 vs 0.94).¹⁹ However, the correlation between BWT and SES-CD was moderate in their study, whereas in our study, the correlation was strong.¹⁹ We speculate that the longer interval between endoscopy and IUS (38 ± 38days) might have contributed to the decreased correlation observed in their study.¹⁹ Considering the complexity of assessing BWS and i-fat, the implementation of IBUS-SAS requires additional training. In centers with limited expertise in IUS, BWT remains a reliable tool for reducing the need for frequent endoscopy.

Our study has several limitations. Firstly, we did not implement central reading for IUS assessment. The interobserver variability has always been acknowledged as a concern in IUS. However, IUS is a part of standard care for CD evaluation in our center, and we routinely conduct a minimum of 400 IUS examinations per month specifically for patients with IBD. These examinations are performed by our 3 experienced sonographers (W.C., S.Q., and Z.L.). Before our research, specific training sessions were conducted to ensure consensus on standardization of all parameters. Similarly, the endoscopic examinations were conducted by several gastroenterologists with expertise in IBD, which could also lead to interobserver variability. Secondly, as this was an observational study, neither the sonographers nor the gastroenterologists were blinded to other clinical data. This lack of blinding may lead to an overestimation of the correlation between IUS and endoscopy due to observer bias. However, it is worth noting that these results remained consistent with previous study.²⁰ Furthermore, this suboptimal study setting is unlikely to impact the primary objective of our study. Thirdly, our study only included patients who received IFX treatment. Thus, the generalizability of our findings to other therapies requires further investigation.

In conclusion, the baseline and postinduction IUS findings are independent predictors of TH and MH after 44 to 56 weeks of IFX treatment. International Bowel Ultrasound Segmental Activity Score is responsive to IFX treatment and shows good agreement with SES-CD, making it suitable for adoption in centers with extensive expertise in IUS.

Supplementary Data

Supplementary data is available at Inflammatory Bowel Diseases online.

Acknowledgments

The authors would like to express gratitude to Zhuohua Liang and Si Qin for their valuable support of performing intestinal ultrasound examinations in this research.

Author Contributions

Conceptualization: X.G., G.L., Z.H., and W.C.; Methodology: Z.H. and W.C.; Formal analysis: Z.H.; Data curation and investigations: all authors. Writing: original draft preparation: Z.H.; Writing: review and editing: W.C., X.G. and G.L.; Funding acquisition: X.G.; Supervision: X.G. and G.L. Approval of final manuscript: all authors.

Funding

This work was financially supported by the National Natural Science Foundation of China (NO. 81870382) and the program of Guangdong Provincial Clinical Research Center for Digestive Diseases (2020B1111170004).

Conflicts of Interest

The authors declare that they have no conflict of interest.

References

Author notes