A technically difficult endorectal ultrasound is more likely to be inaccurate

Endorectal ultrasound (ERUS) is well established as an accurate modality for local staging of rectal tumours. The aim of this study was to identify reasons for inaccurate staging of tumours, and to assess whether difficulties encountered during scanning are likely to influence accuracy.

ERUS was performed by a single operator using a 10 MHz rigid instrument. One hundred and seventeen patients that had both ERUS and surgery are included in this study (patients that had pre-operative radiotherapy were excluded). During ERUS, procedural conditions and limiting factors were recorded. Data was collected prospectively.

In 78 (66.7%) patients no technical difficulty was encountered during ERUS. In this group accuracy was 80% for T-stage and 77% for N-stage. Specific reasons for inaccuracy identified in this group were: inflammatory lymph nodes (from a tumour associated abscess and a colovesical fistula) and deep biopsy causing a submucosal defect with intramural haemorrhage in benign lesions (2 cases). In the remaining 39 (33.3%), the following problems were encountered: stenotic lesions (23), patient discomfort (8), poor bowel preparation (6), and scarring from previous surgery (2). In 11 patients from this group, the scan was considered inconclusive and no stage could be determined. For the other 28, the accuracy for T-stage was 68% and for N-stage 67%.

A technically difficult ERUS is likely to give an inconclusive or inaccurate result for both T-stage (P = 0.001) and N-stage (P = 0.003). In this situation a repeat scan may be considered (where appropriate). Alternatively, further assessment by MRI or flexible endoscopic ultrasound may be considered.