Limits of endorectal ultrasound in tailoring treatment of patients with rectal cancer.

BACKGROUND

Endorectal ultrasound (ERUS) is considered reliable in staging rectal cancer, but recently some critics have questioned its accuracy. The aim of this study was to evaluate how often an ERUS-based decision leads to an appropriate treatment.

METHODS

Two hundred and twenty patients with rectal cancer staged with ERUS who underwent a surgical resection or a local excision without neoadjuvant therapy from 1997 to 2012 were included. According to ERUS, patients were divided into three groups of indication: (a) local excision (Tis-1 N0), (b) direct surgery (T2 N0), (c) preoperative chemoradiation (T3-4 or N+). Accuracy was explored by the correlation established with the final pathology.

RESULTS

Accuracy for T and N staging was 65 and 64%, respectively. Indication to local excision and to chemoradiation was correct in 97 and 88% of patients staged by ERUS. Accuracy of indication to direct surgery was poor (37%), and 21% of patients were overtreated in this group.

CONCLUSIONS

ERUS seems not able to fulfill all the needs of ideal tailored therapeutic strategies. T2 diagnosis needs to be confirmed by an excisional biopsy before a final decision is made because overstaging of early tumors may occur in a not-so-negligible proportion of patients.