How Accurate Is the Preoperative Diagnosis for Tem Procedures?

Theodore J Saclarides, MD, FACS, Maria C Mora Pinzon, MD, Amanda B Francescatti, BA, Bogdan Vidican, BS

Loyola University Medical Center, Rush University Medical Center

Background: The decision to perform TEM is based on patient, tumor characteristics, and preoperative data such as biopsy and imaging results. TEM is performed with an "excisional biopsy" mentality, understanding that the final diagnosis may be different than what was expected. The purpose of this study was to report the frequency of changes in the therapeutic approach secondary to the results of the TEM procedure.

Methods: Retrospective analysis of a prospectively collected database of all the patients that underwent TEM between 1990 – May 2012 for rectal tumors at Rush University Medical Center.

Results: 444 patients were included in the study, mean age was 65.40 years (21 – 98). 260 (58.6%) were male. Preoperative staging was done using biopsy results (97.8%) and transrectal ultrasound (26.3%), in 40 patients (9%) these results were not available. Preoperative results identified 176 adenomas, out of these 21.6% (38/176) had adenocarcinoma in situ (Tis), and 6.8% had invasive adenocarcinoma (T1=4%, T2=1.1%, T3=1.1%). Among the 86 patients with preoperative diagnosis of Tis, 18.6% were adenomas, and 31.4% (27/86) were invasive carcinoma, 41.9% were Tis, and the remaining 8.1% had no residual tumor. One hundred patients had a preoperative diagnosis of adenocarcinoma, among those 5.1% had a carcinoma in situ, 42.4% had a T1 tumor, 17.2% had a T2 tumor, 3% had a T3 and 27.3% had no residual tumor. The methods for preoperative staging for invasive carcinoma has a sensitivity of 77%, and a specificity of 95%.

Overall 22.3% of the preoperative results were inaccurate, neither age, sex, size of the lesion or stage were associated with the accuracy of the results.

Conclusions: Positive predictive value of adenoma was 71%, in the remainder upstaging occurred. Positive predictive value for Tis was 50%, in the remaining cases, upstage occurred. 28.4% of the presumed adenomas and 31.4% of the presumed in situ lesions are upstaged following complete excision. This suggests that all excisions should be full thickness and that patients should be counseled accordingly.

Key words: transanal endoscopic microsurgery, diagnosis, rectal tumor

Session: Poster Presentation

Program Number: P078