Factors affecting false-negative rates on ex vivo sentinel lymph node mapping in colorectal cancer

Abstract

Purpose Despite the increasing use of sentinel lymph node (SLN) mapping after colorectal cancer resection, reported node identification and false-negative rates vary considerably. The main aim of this prospective study was to quantify the false-negative rates on SLN mapping after resection and to evaluate factors influencing them. Methods Sixty-nine patients with biopsy-proven cancer of the colon and rectum underwent SLN mapping according to a protocol involving the ex vivo submucosal and peritumoral injection of 2–4 ml of Patent Blue V dye. All lymph nodes visualized were marked as SLN and totally embedded, then two 4 μm sections were cut for hematoxylin and eosin staining, and cytokeratin (AE1/AE3) immunostaining. A standard examination of the whole specimen and of the regional non-sentinel lymph nodes was also performed. Results SLNs were identified in 97.3% of the evaluable cases. A mean of 5.0 SLNs were removed per patient (SD ± 4.2). Nine false negatives were identified. Rectal cancer, tumor size > 60 mm, number of metastatic non-sentinel lymph nodes, and mucinous tumors were associated with false-negative SLNs. At multivariate analysis, a rectal location and mucinous differentiation were independently associated with false-negative SLNs. Conclusions Ex vivo SLN mapping after colorectal cancer surgery is technically feasible with a high identification rate. Tumor size and stage, rectal involvement and a mucinous histology seem to interfere with the reliability of SLN staging. It is mandatory to standardize the procedure and selection criteria in order to deal with the question of the reliability of SLN mapping in colorectal cancer.