Limitations and pitfalls of transrectal ultrasonography for staging of rectal cancer

PURPOSE: This study was designed to evaluate the accuracy of preoperative staging by transrectal ultrasonography (TRUS) and to clarify the limitations and pitfalls of TRUS by clinicopathologic analysis for staging errors. MATERIALS AND METHODS: Results of TRUS for 164 consecutive patients with rectal cancer were compared prospectively with histopathologic findings according to the newest TNM classification. Clinicopathologic factors that may influence staging errors were analyzed by reviewing both resected specimens and hard copies of TRUS. RESULTS: There were 13 patients histopathologically staged as pTis, 21 as pT1, 34 as pT2, 84 as pT3, 12 as pT4, 73 as pN0, and 91 as pN1–3. Of these, 85, 86, 56, 93, 75, 74, and 77 percent, respectively, were correctly staged by TRUS. Excluding 12 cases with incomplete examinations because of annular constricting tumors, overstaging of tumor invasion depth was mostly caused by tumor invasion close to the deeper uninvolved layer, inflammatory cell aggregation, desmoplastic change, and hypervascularity around the tumor, mimicking tumor invasion on TRUS. The understaging was mostly the result of microscopic invasion beyond the estimated layers and difficulties in examination because of the tumor location being close to the anal canal or on the Houston's valves or the tumor shapes being polypoid or bulky and fungating. Overstaging in lymph node status was caused by reactive lymph node swelling and understaging by the presence of only small involved node and metastasis in the extramesorectal nodes. CONCLUSIONS: An awareness of the limitations and pitfalls of TRUS, as demonstrated by the present study, should improve staging accuracy and contribute to optimum clinical decision-making.