METRICMust knowCD · Imaging and monitoring
Diagnostic accuracy of magnetic resonance enterography and small bowel ultrasound for the extent and activity of newly diagnosed and relapsed Crohn's disease (METRIC): a multicentre trial
Taylor SA, et al.
Lancet Gastroenterol Hepatol 2018 · PMID 29914843 (PubMed, opens in a new tab) · doi 10.1016/S2468-1253(18)30161-4 (full paper, opens in a new tab)
| Measure | MR enterography | Ultrasound | Difference |
|---|---|---|---|
| Sensitivity for extent (primary endpoint) | 80% (95% CI 72–86) | 70% (95% CI 62–78) | Difference 10% (95% CI 1–18); P=0.027 |
| Specificity for extent | 95% (95% CI 85–98) | 81% (95% CI 64–91) | Difference 14% (1–27); P=0.039 |
| Sensitivity for presence | 97% (95% CI 91–99) | 92% (95% CI 84–96) | Difference 5% (1–9); P=0.025 |
| Specificity for presence | 96% (95% CI 86–99) | 84% (95% CI 65–94) | Difference 12% (0–25); P=0.054 |
Both tests detect small bowel disease in most patients; the gap opens when the question is how far the disease extends.
Source: Taylor 2018, abstract.
Structured abstract
BackgroundMR enterography and ultrasound were both used to image Crohn's disease, but their comparative accuracy for disease extent and activity was not known with certainty.
MethodsProspective, multicentre diagnostic accuracy trial (8 UK hospitals); every patient had both MR enterography and ultrasound; consensus panel reference standard. 284 patients. Primary endpoint: Difference in per-patient sensitivity for small bowel disease extent (correct presence and segmental location).
ResultsPrimary result. Sensitivity for small bowel extent: 80% (95% CI 72–86) with MR enterography vs 70% (62–78) with ultrasound; difference 10% (1–18); P=0.027.
Take-homeBoth MR enterography and intestinal ultrasound detect small bowel Crohn's disease with high sensitivity; MR enterography maps extent somewhat more accurately, while ultrasound is a valid first-line, point-of-care alternative.
| Patients | 284 |
|---|---|
| Population | Patients aged 16 or older with newly diagnosed Crohn's disease (133) or established disease with suspected relapse (151). By the reference standard, 233 (82%) had small bowel disease. |
| Comparison | MR enterography vs small bowel (intestinal) ultrasound |
| Published | Taylor SA, et al. Lancet Gastroenterol Hepatol 2018;3:548–58. |
Key points
- Sensitivity for small bowel disease presence was high with both: 97% (91–99) MR enterography vs 92% (84–96) ultrasound; difference 5% (1–9); P=0.025.
- Specificity for extent: 95% (85–98) vs 81% (64–91); difference 14% (1–27); P=0.039.
- Specificity for presence: 96% (86–99) vs 84% (65–94); difference 12% (0–25); P=0.054.
- No serious adverse events with either test.
- Observer agreement substudy (38 patients scanned twice): 82% agreement for small bowel disease presence in new diagnoses (kappa 0.64) and 81% in relapse (kappa 0.63) (PMID 32037466).
Limitations
- UK National Health Service setting; ultrasound expertise and equipment vary between centres and countries.
- Focused on diagnosis and relapse, not on monitoring response to treatment.
- Reference standard was a consensus panel rather than surgical histology for most patients.
Commentary
METRIC is the key head-to-head comparison of the two radiation-free tests used for small bowel Crohn’s disease. It found both highly sensitive for detecting disease, with MR enterography better at mapping its extent. The authors favoured MR enterography where available in the UK system.
Since METRIC, intestinal ultrasound has become central to monitoring rather than only diagnosis: it is repeatable at the bedside, shows response within weeks (see the STARDUST ultrasound substudy and TRUST&UC), and is acceptable in pregnancy (see intestinal ultrasound in pregnancy). A practical approach is MR enterography for baseline mapping of extent, penetrating disease or surgical planning, and ultrasound for serial monitoring.
Journal club
Design
Prospective, multicentre diagnostic accuracy trial (8 UK hospitals); every patient had both MR enterography and ultrasound; consensus panel reference standard. 284 patients.
Primary result. Sensitivity for small bowel extent: 80% (95% CI 72–86) with MR enterography vs 70% (62–78) with ultrasound; difference 10% (1–18); P=0.027.
Validity and applicability
- UK National Health Service setting; ultrasound expertise and equipment vary between centres and countries.
- Focused on diagnosis and relapse, not on monitoring response to treatment.
- Reference standard was a consensus panel rather than surgical histology for most patients.
Where it sits now
Both MR enterography and intestinal ultrasound detect small bowel Crohn's disease with high sensitivity; MR enterography maps extent somewhat more accurately, while ultrasound is a valid first-line, point-of-care alternative.
- STARDUST (2022): With ustekinumab, an endoscopy-driven treat-to-target algorithm did not improve 48-week endoscopic response over clinically driven care; treat-to-target only helps if escalation has something to add.
- TRUST&UC (2020): Intestinal ultrasound shows bowel wall thickening in most UC flares and tracks response within 2 weeks, so it is a practical point-of-care tool for monitoring UC treatment.
- Intestinal ultrasound in pregnancy (2025): In pregnancy, inflammation seen on intestinal ultrasound was associated with preterm birth, low birth weight and pre-eclampsia independently of symptoms and calprotectin; consider ultrasound to monitor IBD antenatally.
Questions for discussion
- The reference standard was a consensus panel. How might the way it was built affect the comparison between the two tests?
- Ultrasound accuracy depends on the operator. How well would these results transfer to your centre?
- For a newly diagnosed patient, when would you choose MR enterography over ultrasound?