Intestinal ultrasonography: a useful skill for efficient, non-invasive monitoring of patients with IBD using a clinic-based point-of-care approach

Inflammatory bowel diseases (IBD), comprising ulcerative colitis (UC) and Crohn’s disease (CD), are chronic inflammatory conditions reported to be increasing in incidence and prevalence in high-income and newly industrialised countries.¹ The diagnosis and management of IBD have evolved over decades. The diagnosis of UC is conventionally suspected by patient history, confirmed on endoscopy and supported by histological analysis. However, the diagnosis of CD depends on the location of the disease and thus ease of access for confirmatory histology. Some patients with endoscopically inaccessible disease rely on cross-sectional imaging such as CT or magnetic resonance enterography (MRE) and wireless capsule endoscopy, all of which have been reported to have high accuracy.^{2 3}

The sensitivity and specificity of MRE as a single modality are high, not only in the diagnosis of new and recurrent CD in terms of the longitudinal extent and location of the disease,⁴ but also correlate significantly with inflammation within the fibrotic segments, on histological assessment of CD-related strictures.⁵ It is also useful for detecting and quantifying the degree of established bowel damage.⁶ Therefore, the European Crohn’s and Colitis guidelines⁷ recommend MRE as a reliable test and is now routinely used in clinical practice. It does however have some limitations. It is expensive, requires advanced infrastructure, inconvenient for patients (requiring an additional hospital visit during which >1 L of oral contrast must be consumed followed by a 30–40 min scan, which some find claustrophobic), burdensome for healthcare systems and requires expertise to interpret the images.

In recent years, intestinal ultrasound (IUS) has emerged as a reliable, cost-effective, non-invasive, time-efficient test, making it an attractive alternative tool for diagnosis and follow-up of patients with CD. This technique is used extensively by non-radiologist clinicians in continental Europe, Canada and Australia to manage patients with CD, but has not been promoted in other parts of the world, including in the UK, where IUS remains largely the domain of gastrointestinal radiologists.

IUS has been reported to have a reasonable sensitivity of 79.7% (95% CI 71.9% to 87.5%) and a high specificity of 96.7% (95% CI 95.1% to 98.4%) in the diagnosis of suspected CD, as per review by Calabrese et al.⁸ The sensitivity was higher at 89% (95% CI 84.2% to 93.8%) when used in established CD, with a high specificity of 94.3% (95% CI 84.6% to 100%), including in the paediatric group. The expert group of authors also provided recommendations based on data that IUS can be used in other scenarios in CD, such as in the diagnosis of strictures, abscesses and intra-abdominal fistulae and in detecting postsurgical recurrence.⁸ The sensitivity and specificity may be increased by supplementary techniques, such as small intestinal contrast IUS, particularly for detection of jejunal lesions and small bowel strictures, but at some cost of time and patient comfort (due to introduction of oral contrast agent).

Several studies have reported the comparative accuracy of IUS with MRE and colonoscopy. A prospective study looked at consecutive patients with CD who were assessed using MRE, colonoscopy and IUS within 1 week. The investigators found high sensitivity, specificity and positive predictive value, along with high concordance with MRE and colonoscopy.⁹ A large multicentre UK study (METRIC study) prospectively compared MRE and IUS in patients with either newly diagnosed or relapsed CD. The authors reported that both modalities have excellent diagnostic accuracy in terms of the extent and activity of small bowel CD in newly diagnosed as well as relapsed disease, and can be considered as first-line investigations.⁴ MRE had slightly superior sensitivity and specificity for small bowel disease, although the two tests had similar accuracy for colonic disease. About 99% of patients rated IUS as an acceptable test compared with 88% for MRE, but it is important to mention that diagnostic accuracy was rated as the most important attribute of a test.⁴ In the UK, MRE is the preferred investigation due to its higher sensitivity and specificity in this setting.

IUS has been investigated as a tool for assessing response to therapy and predicting clinical outcomes. In a longitudinal observational study by Castiglione and group,¹⁰ the investigators used IUS to assess transmural healing in a group of 218 patients with CD treated with anti-tumour necrosis factor-alpha (TNF-α) therapy. They evaluated 1-year outcomes using steroid-free remission, hospitalisation and need for surgery. In this study, patients who achieved transmural healing as demonstrated by IUS had better clinical outcomes across all parameters used.¹⁰ Another study by Zorzi et al ¹¹ followed up 80 patients with CD using serial IUS using features of bowel wall thickness, extent of lesions and presence of complications before the start of anti-TNF therapy and at 18 months. Long-term outcomes were analysed using corticosteroid use, hospitalisation and surgery as adverse outcomes. It was noted that patients who were categorised as complete responders by IUS had fewer adverse outcomes compared with partial and non-responders.¹¹ Current evidence suggests that with right expertise IUS can be a reliable non-invasive investigation tool for assessing response to therapy and determining transmural healing in CD.¹² This is demonstrated by an example in figure 1.

Figure 1.

Selected intestinal ultrasound images from a young adult prior to (A and B) and at follow-up (C and D) after anti-TNF treatment. The pretreatment images show significant thickening of the terminal ileal wall (line in A), patchy mixed echo change within the submucosa of the bowel wall (arrow in A) and marked mesenteric expansion with oedema (double-headed arrow in A). There is considerable hypervascularity on the Doppler image (B). Following treatment, the bowel wall has returned to normal thickness (line in C) and normal mural architecture has been restored, with clear delineation of the outer echo-dark muscularis propria, the relatively echo-bright submucosa and the inner echo-dark deep mucosa (arrow in C). The deep mucosa (innermost echo-dark layer) is now of normal thickness (<1 mm). There is only mild residual mesenteric expansion (double-headed arrow in C). The Doppler signal is now normal (D). TI, terminal ileum, TNF, tumour necrosis factor.

A group of experts have recently published a consensus and developed a scoring system called the International Bowel Ultrasound Segmental Activity Score.¹³ This score provides users of IUS with a system that could bring more consistent and reproducible results when quantifying disease activity. The group used a Delphi consensus method to identify key parameters, which included bowel wall thickness, bowel wall stratification, hyperaemia of the wall (on colour Doppler) and inflammatory mesenteric fat. A total of 12 experts scored cases using all parameters, and the intraclass correlation coefficient for the score was found to be very high at 0.97.¹³ The development of a reliable scoring system such as this one further helps support the wider use of this modality. A list of commonly reported scoring indices and the criteria used are summarised in online supplemental table 1.

In a study by Maaser et al published in 2019,¹⁴ the investigators evaluated IUS as a monitoring tool in 224 patients with known UC. When patients were prospectively examined with IUS, they noted significant decrease in bowel wall thickness in the sigmoid and descending colon as early as 2 weeks post-therapy, which correlated with clinical response in about 90% of patients.¹⁴ These results are encouraging and indicate that use of IUS could potentially be extended as a point-of-care (POC) test to assess treatment response in UC. However, more studies are required to confirm these findings in UC.

The advantages of IUS are multiple (table 1). First, it can be deployed as a POC test at the bedside or in a clinic room, and hence can provide real-time answers and faster diagnosis, help in clinical decision-making and in longitudinal follow-up of lesions such as strictures and abscesses (except for those in the deep pelvis), and guide management in a single appointment. This prevents an unnecessary period of uncertainty, during which treatment may be withheld (or be incorrect), and is significantly more convenient for patients since they do not need to make a separate trip to the hospital for investigations. Second, the test is much cheaper than alternative cross-sectional imaging techniques (ie, CT enterography and MRE), and hence potentially has a direct cost benefit to trusts. Although no cost benefit has been shown in the National Health Service (NHS) yet, it is likely to be seen in other healthcare settings. Third, this is a virtually painless, non-invasive test that is extremely well tolerated and may obviate the need for invasive tests such as colonoscopy. This is particularly important in paediatric and young patients who have intense follow-up requirements. It can be performed without any oral or intravenous contrast medium, both of which are disliked by patients.

Table 1.

Comparison on key parameters between MRE and intestinal ultrasound

	MRE	Intestinal ultrasound
Access	Often long wait times	Quick
Monitoring	Expensive	Repeated assessments possible
Site of procedure	Radiology	Point of care
Operator	Radiologist	Gastroenterologist/radiologist
Decision	Return of report	Quick decision on therapy
Patient acceptance	Moderate	High
Cost	High	Modest
Delivery	No flexibility	Allows flexibility of set-up

MRE, magnetic resonance enterography.

Overall, the advantages of IUS are multiple. It is a test that carries enormous potential to make a difference to both direct clinical care as well as to the health system in general. With an increasing trend in the prevalence of IBD, the demand for resources is expected to be higher in the future. These increasing demands in service could be met by introducing reliable, time-efficient techniques such as IUS. In a health system where resources are already stretched, now made worse by the pandemic, techniques such as this may help alleviate some pressures.

An elegant summary of the advantages and situations in which IUS can be best used in IBD is given by experts in the field from around the globe in a review by Calabrese et al.⁸

Much like any test, there remain pitfalls that one needs to be aware of. Importantly, IUS has a steep learning curve which requires hours of training. This requires a substantial investment of time (and potentially money) by the trainee, as this is currently not part of the UK curriculum. Trainees must be able to access both skilled trainers and a consistent case load to develop and maintain their skills. Unlike in continental Europe where IUS is part of the gastroenterology training programme, to the best of our knowledge there are no centres in the UK that promote or advertise their interest to train interested clinicians. The training facilities are limited, and currently the only way to gain international accreditation in IUS is by attending training courses set up by the International Bowel Ultrasound (IBUS) Group.¹⁵ The training requires the completion of three modules within a 2-year timeframe. Within this training is a 4-week block in a specialist training centre outside the UK. This potentially may mean that this is a skill that is to be taken up post completion of training, or we must build centres in the UK that can offer this training.

As experience in Australia, Germany, Italy and Denmark has shown, it is possible for gastroenterologists to train and become proficient in IUS for IBD as a POC tool. Generally, short, intensive courses (such as offered by IBUS) provide familiarity with the technique and the machine, followed by stepwise accumulation of structured experience. IBUS has published a pathway for competency in IBD ultrasonography that can be practical for a busy clinical gastroenterologist, with national support.

In the UK, it is likely we will require some enthusiastic gastroenterologists to learn this procedure initially through IBUS accreditation and develop their own teaching course which could train others in the UK. This will involve support from our radiology colleagues, and relevant committees within the British Society of Gastroenterology, the British Society of Gastrointestinal and Abdominal Radiology, and international collaboratives. The training and monitoring structure of IUS could be similar to that followed by other specialties using ultrasonography, for example, respiratory medicine, with an agreed curriculum and sign-off.

We believe that in the right hands and training this could be a skill that improves the diagnostic capacity of UK clinicians with an interest in IBD. These will potentially speed up diagnostics, provide clinicians with real-time answers that can accelerate treatment decisions to get patients treated faster, reduce waiting list pressures for MRE scans, achieve financial savings for trusts and patients, and be more comfortable and acceptable to patients. For gastroenterology teams and NHS trusts aspiring to introduce IUS as part of their outpatient service, there are multiple models that can be considered. One model similar to some European centres is a clinic led by a gastroenterologist trained in IUS where suitable patients are selected for attendance. This works well for centres that have a sizeable and consistent load of patients with CD. This ensures a consistent steady flow of patients to fill appointments and make it economically feasible. While this can be an efficient model, it is likely to be limited to some tertiary care centres. For a larger proportion of trusts, the model that has the best chance of success will be one that is built in collaboration with gastrointestinal radiologists, for example, agreed ‘walk-in’ IUS appointments on specific days or an outpatient IBD clinic supported by the presence of a radiologist. For some other centres, a hybrid model may be necessary. Any delivery model must be sustainable and responsive to feedback from patients.

Given the enormous benefits and high level of patient acceptability that IUS brings, it is time to discuss how this skill can be incorporated into the UK training programme despite apparent initial challenges.

Ethics statements

Patient consent for publication

Not required.