Water enema computed tomography (WE-CT) in the local staging of low colorectal neoplasms: comparison with transrectal ultrasound.

BACKGROUND

To determine the accuracy of computed tomography performed with a water enema application (WE-CT) in the local staging of low colorectal neoplasms and to compare the results with those of transrectal ultrasonography (TRUS).

METHODS

Forty patients with low colorectal tumors were evaluated prospectively by CT with the simultaneous administration of a lukewarm rectal enema (0.5-1.5 L). Thin slices (5 mm) and intravenous application of iodinated contrast media were routinely used. TRUS was performed in 18 patients. Tumor size, location, and staging according to the TNM classification of the UICC were registered. Tumors were classified as < T3 (T1 or T2) or as T3 or T4. For staging peritumoral lymph node metastases on WE-CT, two criteria of positivity were tested: N+ if at least one peritumoral node > or 5 mm in diameter was seen (reading A); N+ if at least one peritumoral node > or = 5 mm or three peritumoral nodes < 5 mm were identified (reading B).

RESULTS

For the tumor staging, WE-CT showed a sensitivity of 90%, a specificity of 73%, a positive predictive value (PPV) of 90%, a negative predictive value (NPV) of 73%, and an accuracy of 85%. For TRUS, the results were sensitivity of 73%, specificity of 29%, PPV of 62%, NPV of 40%, and an accuracy of 39%. Concerning nodal staging with WE-CT, results were superior when reading A was used: sensitivity = 84%, specificity = 83%, PPV = 73%, NPV = 91%, and accuracy = 84%. TRUS showed a sensitivity of 29%, specificity of 100%, PPV of 100%, NPV of 67%, and an accuracy of 71%.

CONCLUSION

WE-CT is a reliable technique for the local staging of low colorectal tumors that can be superior to TRUS. For diagnosis of peritumoral metastatic lymph nodes on WE-CT, the 5-mm diameter cutoff value is the most appropriate size criterion.