Accuracy of biopsy and cytology for the preoperative diagnosis of colorectal adenocarcinoma

Abstract

BACKGROUND AND OBJECTIVES: Endoscopic biopsy for the diagnosis of colorectal adenocarcinoma is not accurate in every case. Brush cytology can increase the sensitivity for the diagnosis of gastroesophageal lesions when combined with biopsy, but very little information is available for these techniques in the diagnosis of colorectal adenocarcinoma. METHODS: A retrospective medical records review of 110 patients was performed. All patients underwent a colorectal resection for primary adenocarcinoma after a diagnostic endoscopy. Biopsy and brush cytology was evaluated for their respective sensitivity. Seventy-three patients had both biopsy and cytology. RESULTS: The sensitivity of biopsy was 83.6% (92/110); for cytology, 78.1% (57/73; P = 0.44). From the 73 patients who had both diagnostic techniques, 68.5% (50/73) had both positive biopsy and cytology, 12.3% (9/73) only a positive biopsy, and 9.6% (7/73) only a positive cytology. The two techniques combined were not significantly superior to biopsy alone (90.4%, 66/73, vs. 80.8%, 59/73, respectively; P = 0.16), but tended to be superior to cytology alone (P = 0.07). CONCLUSIONS: Cytology and biopsy have a comparable sensitivity. The combination of the two techniques compares favorably, but does not significantly increase the sensitivity of biopsy alone. Both techniques should be used whenever there are any uncertainties concerning the diagnosis of colorectal adenocarcinoma.