Validation and Interobserver Reliability of a Simple Ultrasound Score for Crohn's Disease

Overview

Crohn's disease (CD) is a chronic inflammatory disease in the gastrointestinal (GI) tract. It is characterized by an alternating course between remission and relapse, which necessitates subsequent adjustments in medical therapy. Thus, it is important to have suitable tools to assess disease activity, and currently the reference standard is using ileocolonoscopy. However, ileocolonoscopy causes major patient discomfort and due to logistic reasons, it cannot be performed on a regular basis. High frequency ultrasound (US) is a potentially useful tool to evaluate changes in disease activity in affected bowel segments. Although there are some studies using US in CD patients to construct ultrasonographic scoring systems, there are no validated scores in daily clinical practice. In a recent study by Nylund et al, we used multiple linear regression to select which simple conventional ultrasound parameters that should be included in an ultrasound scoring system (Simple Ultrasound Score of Crohn's Disease (SUS-CD)) using endoscopy as a reference (unpublished data). In this study, we aim to validate this scoring system using an endoscopic score (Simple endoscopic activity score for Crohn's disease (SES-CD)) as reference standard and to assess interobserver variability. The scoring system might be a useful supplement to endoscopy to follow up CD patients.

Although there are some studies where they used conventional ultrasound and contrast enhanced ultrasound (CEUS) on CD patients to construct ultrasonographic scoring systems, there is currently no validated scores used in daily clinical practice. Even though CEUS provides potentially useful information, the method requires extensive sonographic training, high-end scanners and is time-consuming, making this technique less suitable in daily life. In contrast, conventional ultrasound can be performed using portable ultrasound system and requires less training. Furthermore, evidence suggests that bowel wall thickness predicts endoscopic remission better than CEUS. Thus, conventional ultrasound should be used to construct an ultrasonographic scoring system. In a recent study, we used multiple linear regression to select which simple conventional ultrasound parameters that should be included in an ultrasound scoring system (Simple Ultrasound Score of Crohn's Disease (SUS-CD)) using endoscopy as a reference. In this study, the primary objective is to validate the SUS-CD scoring system using a validated endoscopic score (Simple endoscopic activity score for Crohn's disease (SES-CD)) as reference standard and further to assess interobserver variability. Secondary objectives are to assess if changes in the SUS-CD score correlate to clinical and biochemical disease activity parameters used in CD, and to examine if SUS-CD could be used to predict treatment effect. Intervention: All patients will be examined with trans-abdominal ultrasound and ileocolonoscopy, and relevant blood and faeces samples will be collected. Ultrasound: The ultrasound examinations will be performed with a mid to high range ultrasound scanner equipped with a curvilinear abdominal probe and a high frequency probe. Frequency, focus and gain settings are optimised to get the best images of the patient. First, a curvilinear probe providing good depth penetration is used to get overview and to identify gross pathology. Next, a detailed examination of the large bowel is performed by scanning systematically from the terminal ileum and further distally, using a linear probe with high frequency. The colon is examined in both longitudinal and transverse views. The rectum is examined using the curvilinear probe because the bowel-segment is deeply located. The sonographic examination is scored according to the SUS-CD. Still images and cine loops in each segment are stored and the locations are identified using external markers obtained from the US-machine. The examiner is blinded to the findings on ileocolonoscopy and biochemical parameters when performing the ultrasound examinations. Color Doppler can be used to evaluate bowel wall vascularity. All bowel segments are scanned with color-coded Doppler sonography using standardized settings and velocity scale and wall filter adjusted to the lowest settings, enabling registration of lower velocities in the intestinal wall. The data is saved as a cine loop, and the color-Doppler scores are registered in SUS-CD. Main study parameter: Conventional ultrasonographic parameters in SUS-CD: bowel wall thickness, stratification, colour Doppler, fatty wrapping. Other parameters: Endoscopic disease activity score (SES-CD), blood and faecal inflammatory parameters, clinical disease activity (Harvey Bradshaw Index). Interobserver reliability: All patients included at Haukeland University Hospital are scanned by two observers to assess interobserver reliability. None of the observers are aware of the SES-CD, however, the Montreal classification is known.