Introduction Systematic reviews have reported wide variations in locoregional – ‘T’ and ‘N’ staging – accuracy of EUS¹for oesophageal cancer. This is attributed to multiple factors including the frequency of the echoendoscope (5–20 MHz), operator technique, non-traversibility of tumour and use of miniprobes.

Method Retrospective review of oesophageal cancer cases registered in the local upper GI cancer database and Somerset Cancer registry for the Lancashire and South Cumbria cancer network (January 2008–December 2013). Staging EUS was performed at a single centre (Blackpool Teaching Hospitals) by four operators using Olympus prosound alpha 5 radial (13.8 mm outer diameter) and linear (14.6 mm outer diameter) echoendoscopes. Demographic, endoscopic and histopathological data were analysed and EUS staging was compared with final histopathological (p) staging of tumours.

Results405 patients with oesophageal cancer identifed, 141 had EUS staging. 47 [39 male, mean age 70 (age range 45–85)] had final pathological staging available – surgical resection in 42 (89%) and endoscopic mucosal resection (EMR) in 5 (11%). The remaining 94 patients were unfit for resection or had palliative treatment only. Final ‘p’ staging confirmed adenocarcinoma in 38 (81%), sqaumous cell carcinoma in 7 (15%) and poorly differentiated in 2 patients (4%). 30 (64%) patients had neoadjuvant chemotherapy. 10 patients (21%) had endoscopically non-traversible tumours. Overall sensitivity and diagnostic accuracy of EUS for ‘T’ staging was 87% and 70% respectively, which falls to 67% and 49% respectively following neoadjuvant chemotherapy. Overall sensitivity and diagnostic accuracy of EUS for ‘N’ staging was 61% and 49% respectively, which falls to 40% and 38% respectively following neoadjuvant chemotherapy. None of the patients had FNA of local lymph nodes.

Conclusion These results confirm that EUS has an acceptable overall accuracy for locoregional staging. Accuracy for T4 is diminished, often due to non-traversable tumours (no dilatation pre-EUS or use of mini-probes). Overall accuracy of ‘N’ staging is less than that of published literature as routine FNA of suspicious nodes were not performed. Limitations of study include time lag (mean 16.4 weeks) between EUS and resection and neo-adjuvant chemotherapy down staging of tumours.

Disclosure of interest None Declared.

Reference

1. Puli SR, Reddy JBK, Bechtold ML, Antillon D, Ibdah JA, Antillon MR. Staging accuracy of esophageal cancer by endoscopic ultrasound: A meta-analysis and systematic review. World J Gastroenterol. 2008;14(10):1479–1490