Table 2.
Synopsis and key recommendations (based on items for which ≥ 80 % consensus was reached)
| I - Recommendations for MR image acquisition |
| a. hardware |
| MRI should routinely be performed for primary staging and restaging of rectal cancer |
| Endorectal ultrasound is the preferred technique for the differentiation and staging of T1 tumours |
| MRI should be performed with an external surface coil on a 1.5T or 3.0T MRI system |
| b. patient preparation |
| Use of an enema is not routinely recommended |
| (Use of spasmolytics may be useful to reduce bowel movement artefacts (no consensus: 57 % recommended/mandatory)) |
| (Use of endorectal filling is not routinely advised (no consensus: 71 % not recommended)) |
| c. sequences and sequence angulation |
| A routine protocol should (at least) include 2D T2-weighted sequences in 3 planes and a diffusion-weighted sequence (including at least a high b-value of ≥ 800) |
| Diffusion-weighted images (including Apparent Diffusien Coefficient maps) should mainly be assessed visually; quantitative ADC measurements are not routinely advised |
| Diffusion-weighted imaging is recommended for restaging of the yT-stage. |
| Fatsuppressed, T1-weighted (non-enhanced and contrast-enhanced) and dynamic contrast enhanced (DCE) sequences are not routinely recommended |
| Slice thickness (for the axial and coronal T2-weighted sequences) should be ≤ 3 mm |
| Transverse and coronal sequences should be angulated perpendicular and parallel to the rectal tumour axis, respectively. |
| In distal tumours a coronal sequence angulated parallel to the anal canal should be included to assess the relation between tumour and anal sphincter |
| II - Recommendations for MR image evaluation and reporting |
| a. primary staging |
| Structured reporting is recommended and should include the items described in the report template in Fig. 1 |
| For nodal staging the criteria described in Table 4 are recommended |
| Stranding into the mesorectal fat is an equivocal sign that may indicate either a T2 or T3 tumour |
| The mesorectal fascia (MRF) is 'involved' if the distance between MRF and tumour is ≤1 mm |
| When a tumour shows stranding into the MRF, the MRF should be considered involved |
| A tumour that involves the MRF should be considered a T3 (and not a T4) tumour |
| Tumour invasion above the level of the peritoneal reflection (at the anterior side) should be considered at risk for peritoneal rather than MRF invasion |
| A tumour that invades the pelvic floor or pelvic side wall muscles should be considered a T4 tumour |
| A tumour that grows into the internal anal sphincter muscle should be considered a T3 (and not a T4) tumour |
| b. restaging after neoadjuvant treatment |
| Structured reporting is recommended and should include the items described in the report template in Fig. 1 |
| For nodal restaging the criteria described in Table 4 are recommended |
| On T2-weighted MRI, a normalised, two-layered wall after CRT is suggestive of a complete response |
| On T2-weighted MRI, a completely hypointense (fibrotic) residue without an isointense mass indicates a complete or near-complete response |
| When considering organ preservation (watchful waiting) after CRT, MRI findings should be correlated with clinical examination (endoscopy / digital rectal examination) |
| If a fatpad re-appears between the tumour and MRF after CRT, the MRF should be considered uninvolved/cleared. |
| Persistent stranding of tumour into the MRF should be considered an equivocal sign that may or may not indicate persistent MRF involvement |
| III - MRI performance |
| a. T2-weighted MRI |
| Primary staging |
| 2D T2-weighted MRI can be used to reliably (≥80 % accurate): |
| Differentiate between T2 and T3 tumours |
| Differentiate between non-involved and involved mesorectal fascia |
| 2D T2-weighted MRI is not accurate to differentiate between T1 and T2 tumours |
| b. Diffusion-weighted MRI |
| Primary staging |
| Diffusion-weighted MRI is not accurate to: |
| differentiate between T1 and T2 tumours |
| differentiate between T2 and T3 tumours |
| differentiate between N0 and N+ stage |
| differentiate between non-involved and involved mesorectal fascia |
| assess EMVI |
| Restaging |
| Diffusion-weighted MRI is not accurate to: |
| differentiate between T1 and T2 tumours |
| differentiate between N0 and N+ stage |
| assess EMVI |