Non-invasive monitoring of inflammatory bowel disease using intestinal ultrasound

Abstract

BACKGROUND

Intestinal ultrasound (IUS) is a safe and effective way for the diagnosis and surveillance of patients with inflammatory bowel disease (IBD). It allows a noninvasive and reproducible follow-up for patients with IBD.

AIM

To compare the outcomes of colonoscopy and IUS in diagnosing and monitoring patients with IBD.

METHODS

A prospective study was conducted over a three-year period (January 2021 to April 2024) comparing endoscopic and IUS findings. A total of 101 patients were included in the study (68 with Crohn’s disease and 33 with ulcerative colitis). All patients underwent both IUS and colonoscopy within a 10-day period.

RESULTS

The study found a strong correlation between bowel thickening on IUS and inflammatory activity (P = 0.004), IUS remission and endoscopic remission (P = 0.03), IUS and endoscopic location (P = 0.04), as well as IUS and computed tomography scan findings for collection diagnosis (P < 0.01).

CONCLUSION

The study’s findings demonstrated excellent results for using IUS in the diagnosis and follow-up of IBD patients.

Key Words: Inflammatory bowel disease; Ulcerative colitis; Crohn’s disease; Intestinal ultrasound; Endoscopy

Core Tip: This is the first single-blind prospective study conducted to compare the results of ultrasound and colonoscopy. Our goal is to present an objective comparison between intestinal ultrasound (IUS) and colonoscopy. All the patients with inflammatory bowel disease (Crohn’s or ulcerative colitis) were contacted by the same investigator (Dr. Hajar Cherkaoui) in a delay of 10 days of colonoscopies. The operator (Dr. Hakima Abid) did not know the results of colonoscopies while performing the IUS. This ensured objective, unbiased evaluation of IUS in the diagnosis and follow-up of inflammatory bowel disease patients.

INTRODUCTION

Intestinal ultrasound (IUS) is a safe tool for the diagnosis and surveillance of inflammatory bowel disease (IBD). In addition to being non-invasive, non-irradiating, easy to perform at the patient’s bedside, and inexpensive, intestinal ultrasonography enables rapid results at the patient’s bedside, and equally rapid decision-making[1]. Among the many advantages of IUS are its safety, reproducibility, and accessibility. It can be easily integrated into the close follow-up of IBD patients and treat-to-target. The rapid improvement in ultrasound results and the possibility of frequent follow-up will revolutionize the monitoring of IBD patients, enabling therapeutic decisions to be made more rapidly[2].

MATERIALS AND METHODS

Type and location of study

This prospective, monocentric, descriptive, and analytical single-blind study was conducted over a 3-year period (February 2021 to April 2024) in the Gastroenterology Department of the Hassan II University Hospital. A total of 101 patients with chronic IBD were recruited and monitored during the study. All patients underwent colonoscopy and intestinal ultrasonography. Of these 101 patients, 7 were followed up with IUS.

Inclusion criteria

Patients over 18 years old with chronic IBD [Crohn’s disease (CD) or ulcerative colitis (UC)] who consented to participate in the study were included. Eligibility required follow-up in our unit through hospitalization, consultation, or day hospital visits, as well as undergoing colonoscopy within 10 days before or after an abdominal ultrasound.

Exclusion criteria

Exclusion criteria included patients under 18 years of age, those who did not consent to the study, individuals without IBD, and those who underwent IUS and colonoscopy more than 10 days apart.

Study design

Patients meeting the inclusion criteria and providing consent were contacted by telephone and scheduled to attend the abdominal ultrasound unit of the Gastroenterology Department within 10 days before or after their digestive endoscopy. Patients underwent a six-hour fasting period. All abdominal ultrasounds were performed and validated by the same experienced operator, who was unaware of the results of the digestive endoscopy (single-blind study). This operator performed more than 200 IUS prior to the study. This approach minimized information bias. To reduce selection bias, all patients were selected by the same investigator. The imaging equipment used was a Toshiba ultrasound scanner for abdominal ultrasounds and an Olympus 190 colonoscope for endoscopic procedures.

Data collection

The collected data were recorded on a digital data sheet and compiled into an Excel file. The dataset included epidemiological, biochemical, ultrasonographic, endoscopic, radiological, and therapeutic information.

IUS criteria

IBD remission criteria used in the study are defined as a reduced bowel wall thickness (BWT) of less than 3 mm. Additional remission criteria included a reduced or absent Doppler signal within the bowel wall, preservation of normal bowel wall stratification, normalization or improvement of peristalsis, absence of inflamed mesenteric fat, and the absence of complications, such as abscesses, fistulas, or strictures. To measure BWT using IUS, the ultrasound probe was positioned over the abdomen to locate the bowel segment, and the thickness was measured perpendicularly from the inner mucosal layer to the outer serosal layer, ensuring the bowel was not compressed, with Doppler used to assess vascularity.

Statistical analysis

Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS; version 22) software, with quantitative data expressed as median and mean ±SD. Qualitative data were expressed as bimodal values (0 or 1). The association between endoscopic and ultrasonographic findings was analyzed using the χ² statistical test, with a P value of < 0.05 considered statistically significant. Pearson’s statistical test was used to express the correlation between ultrasound and endoscopic aspects.

RESULTS

After statistical analysis of data, the results obtained are presented in Table 1.

Table 1 Statistical analysis using the Pearson correlation test.

Category	Measurement	Crohn’s disease (n = 67)	Ulcerative colitis (n = 34)
Baseline characteristics	Gender	Male: 45%; Female: 55%	Male: 50%; Female: 50%
	Age (years), mean ± SD	35.4 ± 12.7	40.2 ± 14.5
	BMI (kg/m²), mean ± SD	23.1 ± 3.5	24.0 ± 4.1
	Disease duration, median (IQR), year	7 (3-12)	5 (2-10)
	Smoking status	Smokers: 30%; Non-smokers: 70%	Smokers: 20%; Non-smokers: 80%
	Medication use	Biologics: 60%; Immunomodulators: 40%	Biologics: 50%; Immunomodulators: 50%
	Disease location	Ileum: 45%; Colon: 35%; Ileocolon: 20%	Left-sided: 60%; Pancolitis: 40%
IUS and endoscopy inflammation signs	IUS inflammation vs endoscopic inflammation	P: 0.004; R: 0.851 (0.70-0.94); N: 56, 83%	P: 0.007; R: 0.842 (0.78-0.96); N: 24, 72%
	Limberg endoscopic score > 2 and deep ulcers	P: 0.079	P: 0.089
IUS inflammation and clinical scores	Intestinal BWT vs HBI score > 4 (CD). Intestinal BWT vs Mayo score > 7 (UC)	P: 0.6; R: 0.18 (0.02-0.35); N: 56, 83%	P: 0.27; R: 0.230 (0.117-0.527); N: 23, 70%
IUS and endoscopic remission	Endoscopic remission vs IUS remission	P: 0.03; R: 0.656 (0.47-0.87); N: 9, 13%	P: 0.04; R: 0.62 (0.44-0.71); N: 6, 19%
IUS and endoscopic location	Same location in endoscopy and IUS	P: 0.04; R: 0.53 (0.31-0.65); N: 48, 71%	P: 0.035; R: 0.647 (0.434-0.772); N: 25, 76%
Collection in IUS and CT scan	Collection in IUS vs collection in CT scan	P: < 0.01; R: 0.835 (0.699-0.969); N: 6, 9%	P: < 0.01; R: 0.835 (0.699-0.969); N: 10, 14%
Bowel thickness and endoscopic scores	BWT > 7 mm vs CDEIS > 15 (CD). BWT > 7 mm vs UCEIS > 6 (UC)	P: 0.053; R: 0.452 (0.2-0.61); N: 22, 47%	P: 0.008; R: 0.87 (0.759-0.94); N: 16, 47%
Inflammatory activity in IUS and histology	Inflammatory activity in IUS vs histology	P: 0.012; R: 0.72 (0.68-0.84); N: 53, 87%	P: 0.21; R: 0.25 (0.13-0.317); N: 24, 72%
Detailed IUS and endoscopic findings	Bowel wall thickness, mean ± SD	5.4 ± 1.2 mm	4.8 ± 1.1 mm
	Vascularity (Doppler)	Increased: 65%; Normal: 35%	Increased: 55%; Normal: 45%
	Mucosal ulceration	Present: 40%; Absent: 60%	Present: 35%; Absent: 65%
	Stratification loss	Present: 30%; Absent: 70%	Present: 25%; Absent: 75%
	Fistulas/abscesses	Present: 15%; Absent: 85%	Present: 10%; Absent: 90%

BMI: Body mass index; IQR: Interquartile range; IUS: Intestinal ultrasound; BWT: Bowel wall thickness; HBI: Harvey-Bradshaw index; CD: Crohn’s disease; US: Ulcerative colitis; CT: Computed tomography; CDEIS: Crohn’s Disease Endoscopic Index of Severity; UCEIS: Ulcerative Colitis Endoscopic Index of Severity.

Epidemiological characteristics

During the study period (February 2021 to April 2024), 101 patients with chronic IBD at the Hassan II University Hospital in Fez were recruited. All these patients benefited from a diagnostic IUS. Among these 101 patients, 7 of them benefited from a follow-up IUS. The mean age of the patients was 40 years, with extremes ranging from 18 to 68 years, and a standard deviation of 15 years. The female/male sex ratio was 1.3. The majority of patients (90%) were undergoing outpatient treatment at the time of the ultrasound scan. The remaining patients (10%) were hospitalized at the time the ultrasound was performed (digestive abscess, and severe acute colitis). All outpatients were contacted by telephone to report to the department on the day of the abdominal ultrasound scan.

Medical history

Among the 101 patients included in the study, the majority had CD (n = 68; 67.5%) and 32.6% (n = 33) had UC. Sixty-eight patients had CD, of whom 50 patients (73%) had ileocolonic localization. Among these patients, 9 patients had isolated localization at the ileocecal junction, and one patient had associated high localization. Pancolonic localization was found in 9 patients (13%). Isolated left colonic localization was observed in one patient. Ten patients had localized disease in the small intestine. The presence of anoperineal lesions was found in 13 patients (18%). Isolated left localization was found in one patient. Inflammatory phenotype was the predominant phenotype, accounting for 80% of cases. The stenosing phenotype was observed in 9 patients (13%), and the fistulizing phenotype was found in 5 patients (7%). The average follow-up period was 3.1 ± 4 years with extremes of 3 months and 20 years. UC was found in 33 patients (32.6%). Pancolonic localization was predominant (n = 25, 76%), left colonic in 7 patients (24%), and isolated rectal in one patient. The average follow-up of patients was 1.76 ± 3.89 years, with extremes ranging from 2 months to 26 years. Other history was psoriasis in 2 patients and ankylosing spondylitis in 6 patients. Family history of IBD was present in 3 patients, all first-degree relatives. Ten patients had a history of surgery: One patient underwent right hemicolectomy, three patients had intestinal resection, two patients underwent subtotal colectomy, and three patients had appendectomy.

Clinical signs

Patients with CD had watery diarrhea (n = 50; 76%). Seven patients exhibited mucus-tinged blood, and rectal symptoms were noted in three patients. Abdominal pain was found in 37 patients (54%). These pains were mostly localized in the right iliac fossa (65% of cases), with the rest in the colonic area. Among patients with right iliac fossa pain, 6 patients had a Koening syndrome. Other clinical signs included chronic vomiting associated with upper localization of CD (n = 1) and an obstructive syndrome (n = 1). Pus discharge through the anus was found in one patient. Patients with UC had bloody mucus diarrhea (n = 21, 63%), with watery diarrhea in 7 patients (21%). Abdominal pain was present in the majority of patients (n = 24; 72%). These pains were intermittent in most cases (80%), peri-umbilical (n = 7; 29%), in the colonic area (n = 7; 29%), and in the left iliac fossa (n = 10; 42%). Small-volume rectal bleeding was present in 2 patients (n = 2; 8.3%).

Physical examination

Physical examination in patients with CD found an average weight of 55.77 ± 8 kg. The average body mass index was 21 ± 3 kg/m². At the time of IUS, all patients were afebrile. The abdomen was soft in all patients, with abdominal tenderness observed in 24 patients (35%). The examination revealed an enterocutaneous fistula in 2 patients. The average Harvey-Bradshaw index among patients was 5.49 ± 4.11, with extremes ranging from 0 to 30. Physical examination of patients with UC revealed an average weight of 61.37 ± 8.11 kg. The average body mass index was 22 ± 2.4 kg/m². The abdomen was soft in all patients, with abdominal tenderness noted in 7 patients (21%). The Mayo score was 5.24 ± 4.05.

Biochemical information

The average time to complete the laboratory assessment was 2 ± 1 weeks. The average hemoglobin level was 11.8 ± 1.49 g/dL in patients with CD, with extremes ranging from 4.8 to 14.8 g/dL. The average white blood cell count was 7820 ± 2633 cells/L. The average C-reactive protein (CRP) level was 45.57 ± 48.23. Fecal calprotectin was measured in 54% of patients (n = 37). The average fecal calprotectin level was 420 ± 323 μg/g, with extremes ranging from 0.6 to 1235 μg/g. In UC, the average hemoglobin level was 12.11 ± 2.17 g/dL, with extremes ranging from 8.3 to 16.3 g/dL. The average white blood cell count was 7732 ± 2500 cells/L. The average CRP level was 43.19 ± 45.20, with extremes ranging from 0 to 140. Fecal calprotectin was measured in half of the patients (n = 17, 51%) with an average of 370 ± 432 μg/g and extremes ranging from 8 to 3170 μg/g.

IUS findings in CD

Wall thickening was found in 82% of patients (n = 56). Thickening of the terminal ileum was found in 52% of patients (n = 32), measuring 7.13 ± 1.98 mm. Gastric thickening was detected in one patient, measuring 8 mm in thickness. Thickening of the right colon was observed in 20 patients (29%). Isolated left colonic thickening was found in 4 patients (5.8%), averaging 5.2 ± 2.49 mm. Isolated rectal thickening was found in 2 patients (3%) with an average of 7 ± 1.5 mm. Pancolonic thickening was observed in 6 patients (8%), with an average of 10 ± 2.66 mm and extremes ranging from 6 to 26 mm. The last ileal loop was fixed without thickening in 2 patients with ileocecal CD. Loss of stratification was recorded for 46 patients (67%) with CD. Comb sign of the mesentery was found in 3 patients, accounting for 4% of patients. Hyperdense mesenteric fat was observed in 10 patients with CD, accounting for 14% of patients. The Doppler signal was detected in 20 patients, representing 30% of those with CD. The Limberg score was 1.67 ± 1.22. A Limberg score of 4 was observed in only one patient. A collection was identified in 6 patients, located in the right iliac fossa in 5 patients and in the pelvis in 1 patient. The collection averaged 4 cm × 3 cm with extremes ranging from 2 to 12 cm in length. The presence of a fistula was observed in 6 patients (8%), with 5 being enteric-enteric and one being enteric-parietal. Stenosis was noted in 8 patients (11%), with an average upstream ileal dilation of 20 mm. This stenosis was located in the terminal ileum in 6 patients (8%), at the ileocecal valve in one patient (1.14%), and in the right colon in 2 patients (2.9%). An inflammatory mass was found in the terminal ileum in one patient. Extraintestinal signs, such as gallstones, were recorded for 2 patients (2.9%). Hepatic steatosis was detected in 8 patients (11.7%). Lymphadenopathy was observed in 5 patients, all of whom had sub-centimeter, benign-appearing lymph nodes located in the mesenteric region.

IUS findings in UC

Digestive wall thickening was observed in 69% of patients (n = 23), with 66% of cases (n = 16) exhibiting pancolonic involvement, and the average thickness measured 6.8 ± 3.8 mm. Right colonic thickening was found in 70% of patients (n = 17), measuring 6.87 ± 3.45 mm. Isolated rectal thickening was observed in 16 patients, averaging 6.81 ± 3.52 mm with extremes ranging from 5 to 10 mm. Loss of stratification was observed in 17 patients (51%). Pseudopolyps were found in one patient, while the comb sign was present in 4 patients, representing 12% of the cases. Hyperdense mesenteric fat was noted in 5 patients with UC, accounting for 15%. Parietal Doppler uptake was detected in 15 patients, constituting 45% of UC patients. Thus, the colonic wall flow was 1 in 45% of patients and 0 in 55% of patients. The average Limberg score was 1.51 ± 1.16. The Milan ultrasound criteria (MUC) score was calculated for all patients, with an average of 8.82 ± 5.17. Twenty patients (60%) had an MUC score greater than 6.2. Stenosis was observed in one patient, with biopsy results showing no signs of malignancy. Additionally, gallstones were found in one patient (3%), hepatic steatosis in 5 patients (15%), and lymphadenopathy in one patient (3%). The lymph nodes were oval, sub-centimeter, and suggestive of inflammation.

Colonoscopy findings in CD

Mucosal erythema was found in 48 patients, accounting for 70% of cases. Partial disappearance of the vascular pattern was observed in 23 patients (33% of patients). Mucosal fragility was observed in 23 patients, accounting for 33%. Erosions were found in 33 patients (48%), while mucosal bleeding was present in 17 patients (25%). Aphthous ulcers were identified in 14 patients (20%). Ulcers, in general, were found in 51 patients, representing 75% of those examined. They were superficial in 41 patients and deep in the remaining 10 patients. Mucosal detachment was detected in 9 patients (13%). A pavement-like appearance of the mucosa was described in 2 patients (2.9%). The mean Crohn’s Disease Endoscopic Index of Severity score was 10.29 ± 8.77. The Rutgeerts score was performed in 4 patients, showing a Rutgeerts i0 in one patient, i2 in one patient, and i4 in 2 patients.

Colonoscopy results were normal in 9 patients (13%), showed ileal involvement in 11 patients (16%), and ileocolonic involvement in 30 patients (44%). Among those with ileocolonic involvement, 13 patients had involvement at the ileocecal valve. Pure colonic involvement was observed in 21 patients (30%). Among these patients, lesions were pancolonic in 12 patients (18%), right colonic in 3 patients (4%), and left colonic in 6 patients (8%). Stenosis was found in 24 patients (35%) of the patients. The stenosis was located at the ileocecal valve in 11 patients (16%), at the terminal ileum in 6 patients (8%), in the right colon in 3 patients (4%), in the transverse colon in 1 patient (1.4%), and in the sigmoid colon in 3 patients (4%). Pseudopolyps were detected in 5 patients (7%). They were ileal in 2 patients, at the ileocecal valve in one patient, and pancolonic in 2 patients. Fistulas were found in 2 patients at the ileal level.

Colonoscopy findings in UC

Mucosal erythema was found in 25 patients, accounting for 75% of cases. Partial disappearance of the vascular pattern was observed in 13 patients (39% of patients). Mucosal fragility was detected in 18 patients, accounting for 54% of patients. Erosions were found in 16 patients (48% of patients). Mucosal bleeding was present in 16 patients (48% of patients). The presence of ulcers was noted in 25 patients (75% of the patients) examined. They were superficial in 20 patients and deep in the remaining 5 patients. Mucosal detachment was observed in 6 patients (46%). Mucosal bridges were recorded for one patient. Left colonic involvement was found in 10 patients (30%), with exclusive involvement at the rectal level in 2 patients (6%). Pancolonic involvement was observed in 15 patients (45%). Colonoscopy results were normal in the remaining 6 patients (19%). Stenosis was observed in 2 patients (6%), with one case in the right colon and the other in the left colon. Biopsies of the strictures were benign. Pseudopolyps were found in 4 patients (12%), located in the transverse colon (n = 2), pancolonic (n = 1), and rectal (n = 1). The mean Ulcerative Colitis Endoscopic Index of Severity score was 3.57 ± 2.83.

Biopsy findings in CD

Systematic biopsies for CD were not performed; however, 80% of colonoscopies were supplemented with ileocolic biopsies. Among these biopsies, 59% were non-specific, 30% were consistent with CD, and 11% showed signs of endoscopic remission. In UC, biopsies were not performed systematically in 75% of cases (n = 25). Among those who underwent biopsies, 40% (n = 10) were consistent with active UC, 8% (n = 2) showed signs of IBD in remission, and 39% (n = 13) were non-specific.

Computed tomography scan findings in CD

The enterography computed tomography (CT) scan was not systematically performed in all patients with CD (n = 50; 77%). It was conducted within an average period of 14 ± 5 weeks from the IUS. Intestinal thickening was found in 37 patients (74%). This thickening was ileocolonic in the majority of cases (n = 30; 81%), with an average thickness of 5.75 ± 4.99 mm (ranging from 4 to 14 mm). Pancolonic thickening was present in 3 patients (5%), with an average thickness of 6.54 ± 6.23 mm (ranging from 5 to 13 mm). Jejunal thickening was present in 2 patients (5%), with an average thickness of 5.5 ± 0.5 mm. Left colonic thickening was found in 7 patients (21%), with an average thickness of 5.63 ± 5.11 mm (ranging from 6 to 11 mm). A collection was identified in 10 patients (20%), with ileocecal involvement in 8 patients (16%), jejunal in 1 patient (2%), and pelvic in 1 patient (2%). Stenosis was observed in 7 patients (14%), with 6 cases being inflammatory and 1 fibrotic. The stenosis was located in the ileum in 5 patients (10%), the rectum in 1 patient (2%), and the jejunum in 1 patient (2%). Enteric-enteric fistula was detected on enterography CT scan in 7 patients (14%). Contrast enhancement of the digestive wall on CT scans was observed in 20 patients (40%). Lymphadenopathy was identified in 10 patients (20%), with all detected lymph nodes being sub-centimeter and oval-shaped. The mesenteric comb sign was present in 17 patients (34%). It was meso-cecal in 15 patients (30%), meso-rectal in one patient (2%), and peri-mesenteric in one patient (2%). Mesenteric fat hypertrophy was found in 16 patients (32%).

Magnetic resonance imaging findings in CD

Magnetic resonance imaging (MRI) was performed in 8 patients (11% of all patients with CD). All the MR enterographies conducted were done using T1-weighted sequences with gadolinium injection and T2-weighted sequences. Intestinal thickening was observed in 7 patients, presenting as hyposignal in 2 patients and hypersignal in 3 patients. The thickening was localized to the ileum in 6 patients and the right colon in 1 patient. Stenosis was identified in 3 patients, with 2 cases being fibrotic and 2 inflammatory. Enteric-enteric fistulas were detected in 3 patients. One patient also had an anal fistula communicating with an enteric-enteric fistula, another had an enterovesical fistula, and two patients presented with Saint James classification grade 1 and grade 3 fistulas, respectively. Among the 11 patients who underwent MR enterography, no collections were identified. Contrast uptake with gadolinium injection was observed in 4 patients. The comb sign of the mesentery was found in 3 patients. Mesenteric fat proliferation was observed in 3 patients. Mesenteric lymphadenopathies were detected in 3 patients. All lymph nodes were oval-shaped and sub-centimeter.

CT scan findings in UC

CT scan was not systematically performed in all patients with UC (n = 24, 72%). It was conducted within an average period of 13 ± 6 weeks. CT enterography was performed after patients drank one liter of water one hour before the scan, following an overnight fast. All CT enterographies were done with contrast injection. Intestinal thickening was found in 18 patients (52%). This thickening was pancolonic in 10 patients (55%), with an average thickness of 11.73 ± 2.41 (ranging from 9 to 14 mm), and in the left colon in 8 patients (44%), with an average thickness of 10.72 ± 2.94 mm (ranging from 5 to 18 mm). Rectal stenosis was found in one patient, extending over 3 cm. Contrast uptake in the intestinal wall was observed in 10 patients (55%). The presence of lymphadenopathy was detected in 6 patients (34%). All detected lymph nodes were sub-centimeter and oval-shaped. The comb sign of the mesentery was present in 6 patients (34%). It was mesorectal in 2 patients, mesocolic in 3 patients, and perimesenteric in one patient. Mesenteric fat hypertrophy was detected in 6 patients (34%).

DISCUSSION

Intestinal IUS analysis of the digestive loops involves evaluating four key parameters: Wall thickness, wall differentiation, presence of Doppler activity, and assessment of the peri-digestive space, including sclerosis, lymph nodes, and the hyper-echogenic appearance of mesenteric fat. Among these, intestinal wall thickness is the most critical ultrasound criterion. The threshold retained was 3 mm for the small bowel and 4 mm for the rectum. The measurement should be performed in both longitudinal and transverse sections and repeated at two points along the intestinal segments being studied. This parameter is highly reproducible, with strong inter-observer agreement, reflected by an intraclass correlation coefficient of 0.96 (95% confidence interval: 0.94-0.98)[1]. Wall differentiation enables the identification of the distinct layers of the intestinal wall: The mucosa (hypoechoic), submucosa (hyperechoic), muscularis (hyperechoic), and serosa (hyperechoic). Loss of differentiation between these layers is a characteristic finding in CD[2]. Doppler imaging indicates parietal hyperemia and is used to define the Limberg classification. It requires high-quality ultrasound probes with good settings allowing artifact-free waveform acquisition. Limberg classification greater than 1 is associated with the presence of endoscopic lesions with a sensitivity of 67% and a specificity of 92% with a strong predictive value of 97%. The examination of the peri-digestive space includes examination of the fat, with its hyperechoic appearance being compatible with inflammation. The presence of lymph nodes is correlated with intestinal inflammation[3].

Several studies have shown a correlation between ultrasound scores and endoscopic scores, leading to the publication of numerous studies that include IUS as a valuable tool in the management of patients with IBD[4]. Current indications for IUS in CD include: (1) Differentiating between irritable bowel syndrome and IBD; (2) Monitoring therapeutic response in IBD patients, as validated by the European Crohn’s and Colitis Organization and the European Society of Gastrointestinal and Abdominal Radiology; (3) Integrating into therapeutic management algorithms for IBD patients; and (4) Serving as a valuable diagnostic tool in suspected cases of CD. CD is a chronic, transmural IBD primarily affecting young individuals (ages 15-25)[5]. Systematic literature reviews report a sensitivity of 85%-97% and a specificity of 83%-97% for IUS in diagnosing CD. The METRIC study concluded that MRI is more effective in estimating the extent of intestinal thickening, but ultrasound remains the first-line approach due to its good negative predictive value and non-ionizing nature. It is also the first diagnostic tool used in children. In cases of strong suspicion of CD, the least expensive and most accurate diagnostic strategy would be to perform IUS coupled with a colonoscopy[6]. In CD, the METRIC study showed an excellent correlation between endoscopy and IUS in lesion localization. The METRIC study also showed an excellent correlation between the location found on IUS and cross-sectional imaging. Exceptions include the jejunal location, where intestinal MRI is more effective, and the rectal location, where endoscopy is more effective[7].

Given the intermittent nature of the disease (evolution by flare-remission), close therapeutic monitoring is recommended. This enables the evaluation of transmural activity and the detection of complications, such as abscesses, stenosis, collections, inflammatory flares of CD, and fistulas[8]. In our series, IUS identified complications in 20 patients, with perfect scanographic correlation, consistent with the existing literature[9]. Transmural remission is a key therapeutic goal in the treat-to-target approach for CD. This evaluation cannot be performed endoscopically, highlighting the importance of IUS. The STRIDE 2 study validates mucosal healing as a therapeutic goal, and IUS is one of the available means to assess this endoscopic healing[10]. Meta-analyses have shown that the results of IUS are strongly correlated with cross-sectional imaging findings, as confirmed by the European Crohn’s and Colitis Organization and the European Society of Gastrointestinal and Abdominal Radiology recommendations[11]. These same results were observed in our study, further emphasizing the value of using IUS to assess endoscopic remission.

The intestinal thickening is the most specific sign for diagnosis and assessment of disease activity. It is the ultrasound sign with the least inter-observer variability. In addition to the correlation between endoscopic activity and histological activity, ileal thickening of more than 3 mm is associated with histological activity[12], and also with elevated fecal calprotectin[13]. These results are consistent with the findings of the present study. The presence of inflammation in CD is manifested by parietal hyperemia and accentuation of hypoechoic submucosa. This is characteristic of inflammatory disease activity and is correlated with the presence of endoscopic ulcers[14]. Parietal hyperemia is also associated with a higher risk of surgery[15]. Additionally, the absence of parietal stratification may indicate significant parietal inflammation[16]. Parietal hyperemia, defined by Doppler uptake in the intestinal wall (Limberg stage 2 or 3), is associated with severe disease activity and correlates with both endoscopic and histological activity[17]. The presence of parietal hyperemia is associated with the presence of ulcers on endoscopy in 97% of cases[18]. The persistence of parietal hyperemia is associated with the onset of IBD flare-ups[19]. Thus, a Limberg score greater than 1 is associated with histological inflammation, showing a specificity of 73%. When the score exceeds 2, the specificity increases to 91%[12]. Parietal contrast enhancement performs better than Doppler in estimating parietal activity[20]. This sign is particularly significant when no Doppler uptake is observed in IUS[18]. It holds greater importance, with a sensitivity of 97% and a specificity of 91% in predicting endoscopic activity. The consensus threshold for detecting contrast uptake is 15 dB for disease activity and 18 dB for moderate activity[21].

Wall ulcers are irregularities in the hypoechoic submucosal layer or hyperechoic buildup[14]. A recent meta-analysis shows better correlation between IUS and colonoscopy in detecting wall ulcers than the combination of MRI and colonoscopy[22]. In our series, the presence of inflammatory activity in IUS is associated with the presence of inflammatory activity on colonoscopy. IUS also includes examination of the intestinal wall and the parietal fat. The hyperechoic appearance of parietal fat is correlated with intestinal inflammation[23]. This appearance is also linked to the presence of intestinal thickening and remission[23]. While non-specific and present in various intestinal inflammatory processes, in CD, it provides valuable information regarding the location of the inflammation[14]. Mesenteric lymph nodes are present in 25% of patients with CD[14]. Additionally, one-third of patients with CD exhibit a fistulizing phenotype[24]. Fistulas can be enteric-enteric, enterocolic, enterocutaneous, enterovesical, or enterovaginal[25]. In our series, IUS allowed the diagnosis of fistulas in 6 patients (8%). These results correlate with the findings of enter-MRI. Intestinal abscesses are often associated with a fistulous tract[26]. The use of contrast-enhanced ultrasound can help differentiate these abscesses. In cases of intestinal abscesses, IUS typically shows peripheral contrast uptake with a central avascular area[18]. The sensitivity of IUS in diagnosing abscesses is 92%, although CT scan sensitivity is higher, particularly for abscesses located in the pelvis[27].

Inflammatory activity is assessed within 6 months of starting a new treatment. The most significant study evaluating the value of IUS in CD is the TRUST study, a prospective multicenter trial involving 234 patients who were followed for 12 months after beginning treatment[28]. The first signs of improvement were observed as early as the second week of treatment. Normalization of ultrasound parameters is correlated with mucosal remission[29]. Improvement in ultrasound signs by the 12^th week of anti-tumor necrosis factor therapy was predictive of endoscopic and transparietal remission, as well as a reduced need for surgery at one year[30]. In our series, follow-up by IUS showed a perfect correlation with endoscopic appearance. The treatment of UC requires tight monitoring and the use of a treat-to-target strategy using biomarkers (CRP, fecal calprotectin) and endoscopic remission as the primary therapeutic goal[31]. However, colonoscopy is an invasive and unattractive means of tight monitoring, which highlights the interest in IUS[32]. Since UC is a mucosal disease, IUS has long been considered unsuitable for this condition. However, numerous studies have demonstrated the valuable role of IUS not only in diagnosis but also in the follow-up of patients with UC[5]. There are both transmural and extramural changes in UC[33]. IUS demonstrates excellent correlation with endoscopy in UC[34], with a strong association between endoscopic activity and ultrasound signs of inflammation[35]. This same result is confirmed in the present study. Since the disease is only luminal, loss of stratification and hyperechoic mesenteric fat are less frequent than in CD[36].

The pilot study that allowed correlation between endoscopic and ultrasound activity in UC is the study conducted by the Milan team, which shows an excellent correlation between ultrasound and endoscopic activity. It is a prospective study including 53 patients comparing their ultrasound and endoscopic results. Thus, thickening of more than 3 mm is correlated with a Mayo score greater than 2 with a sensitivity of 89%[3]. Another prospective study involving 60 patients confirmed an excellent correlation between ultrasound and endoscopy. It showed that colonic thickening of less than 2.1 mm could reliably exclude endoscopic activity with a Mayo score of 0, achieving a sensitivity of 93%. The cutoff of 3.2 allowed differentiation between a Mayo sub score of 0-1 and 2-3 with a sensitivity of 90%[36]. The adapted ultrasound activity score is the MUC, which is a combination of intestinal thickening and parietal hyperemia. There are other factors to evaluate intestinal activity, such as loss of colonic haustra and mesenteric fat hypertrophy. The formula used to calculate the MUC is: MUC = 1.4 × colonic wall thickness + 2.0 × colonic wall flow. The cutoff used to predict ultrasound activity is 6.2[36].

CONCLUSION

IUS is a non-invasive, reproducible means that allows the diagnosis and monitoring of IBD patients with a good correlation between endoscopic and ultrasound results. The accessible learning curve of this technique will facilitate its integration into gastroenterology practices and pave the way for its broader adoption in medical practice. The future direction of the present study is to conduct a study on the learning curve of IUS to determine realistic and pedagogical learning objectives for this technique. The present study shows an excellent correlation between abdominal ultrasound and colonoscopy. This is a Moroccan pilot study that will enable the conduct of larger-scale studies. Prospective multicenter and randomized studies will be useful to establish IUS as a cornerstone in the management of patients with IBD.