Intestinal Ultrasound—Ready to Use for Early Prediction of Disease Course in IBD?

This is the authors' abstract. We don't add key points for this paper.

The optimal treatment management of patients with inflammatory bowel disease [IBD] requires objective methods to monitor disease activity. Intestinal ultrasound [IUS] has become a promising imaging modality for monitoring patients with IBD because of its non-invasiveness, point-of-care availability, and cost. In addition to evaluating disease activity at a certain time point, IUS also seems to be a valuable tool for prediction of future treatment response. Recently, two studies focusing on the predictive role of bowel ultrasound in IBD have been performed.1,2 The first of these studies, published by Ellrichmann et al., is a prospective study evaluating the role of endoscopic ultrasound [EUS] coupled with contrast enhancement [dCEUS] in patients with active ulcerative colitis [UC] to predict the therapeutic response to adalimumab or infliximab.1 Reduction of bowel wall thickness [BWT] [mean value across measurements from three locations from the rectum to the sigmoid colon] already at week 2 was predictive of treatment response at week 14, as assessed by the Mayo score. Furthermore, parameters of bowel wall vascularity, such as time to peak [TTP] at week 2, showed significant normalization in responders versus non-responders. In contrast, early endoscopic assessment using the endoscopic Mayo score or faecal calprotectin levels did not demonstrate any changes in treatment response. The results of their study nicely confirm published evidence on the predictive role of IUS in early patient assessment regarding treatment outcome.3–5 Several studies have demonstrated that early IUS changes based on the reduction of BWT are predictive of clinical and endoscopic treatment response in patients with active UC.3–5 In a study by de Voogd et al., a reduction of BWT and colour Doppler signal at week 6 after the start of anti-inflammatory therapy predicted endoscopic remission and improvement between weeks 8 and 26.3 Furthermore, a significant decrease in BWT at week 2 was observed in patients treated with infliximab or tofacitinib. The most striking evidence of early prediction with ultrasound comes from a Danish study assessing the IUS response in patients with acute severe UC.4 The study demonstrated that a reduction of BWT within 48 ± 24 h of treatment initiation was predictive of therapeutic response and the need for rescue therapy. Despite the positive results of Ellrichmann et al.’s study regarding the usefulness of ultrasound in patients with UC, though confirming IUS the EUS method does not seem to add any additional value compared to transabdominal IUS. Indeed, EUS is not routinely available and may be considered an invasive procedure compared to transabdominal IUS. The only advantage of EUS seems to be proper visualization of the rectal wall, as the sensitivity of transabdominal IUS is limited in this location. However, this limitation can be overcome by using non-invasive transperineal ultrasound. In a prospective study, Sagami et al. showed that improvement in BWT measured by transperineal ultrasound at week 1 after treatment initiation was an accurate predictor of treatment response at week 8.5. The study by Dolinger et al. assessed early [week 8] IUS response to predict endoscopic remission in children with Crohn’s disease [CD] and terminal ileitis who started anti-tumour necrosis factor therapy or ustekinumab.2 The authors demonstrated that a reduction in BWT in the terminal ileum from baseline to week 8 [either as a percentage or absolute change] was significantly higher in children achieving versus not achieving endoscopic remission at a median time point of 8 months. Similar results were observed when the International Bowel Ultrasound Segmental Activity Score [IBUS-SAS] was used. In contrast, the Paediatric Crohn’s Disease Activity Index [PCDAI] had only a partial predictive role, and C-reactive protein [CRP] level showed no benefit. A ≥18% decrease in BWT or an absolute decrease of ≥1 mm in the terminal ileum from baseline to week 8 and an absolute BWT ≤ 5.2 mm at week 8 showed almost perfect accuracy in predicting endoscopic remission [AUROC of 0.99 and 0.96, respectively]. The prospective study by Dolinger et al. supports the findings from the adult population6–8 and highlights the role of IUS in predicting endoscopic remission in paediatric patients with CD, especially because clinical indices and laboratory markers do not reliably reflect disease activity. Changes in BWT on IUS predicted endoscopic remission with high sensitivity and specificity, thus increasing the opportunity for accelerated management of the disease and treatment decisions, especially in children, where treatment needs to be optimized early to avoid structural changes in the bowel and growth failure due to delayed intervention. It is important to stress that changes in IUS parameters in the current studies were superior to biomarkers such as CRP or faecal calprotectin in early prediction of treatment response including long-term endoscopic remission. These findings support the indispensable role of IUS as an accurate non-invasive imaging modality in paediatric patients as well as in the adult population. For the use of IUS in clinical practice, standardized measurements and reporting of IUS parameters are crucial, as recently suggested.9 Future studies need to determine the best time points for assessing early IUS response for different therapeutic agents and for different disease manifestations. If the current data are confirmed in a treat-to-target study using clearly defined IUS response parameters, broader implementation of early IUS in CD and UC will become standard of care in our daily clinical practice. No funding was received for this work. T.K. is current treasurer of the International Bowel Ultrasound Group [IBUS], K.M. is a member of the IBUS Pediatric committee and C.M. is IBUS past-president. The first draft was written by D.D. and C.M., which was then critically reviewed and further optimized by K.M. and T.K. No data were generated for this editorial.