Lymph node size as a predictor of lymphatic staging in colonic cancer.

Background: In colonic cancer, the number of harvested lymph nodes is associated with prognosis. The aim of this study was to determine the contribution of small lymph nodes to pathological staging, and to analyse the hypothesis that node size is a confounder in the relationship between prognosis and nodal harvest.

Methods: Nodal harvest and size were analysed in patients who underwent elective surgery for colonic cancer. Visible and palpable nodes were harvested without fat clearance techniques, and conventional histology was performed.

Results: Metastases were found in 99 of 2043 measured lymph nodes in 150 patients. Lymph nodes smaller than 3 mm were positive in 8.0 per cent of patients (12 of 150), but were the sole reason for upstaging in only 1.3 per cent (2 of 150). No metastases were found among 95 nodes of 1 mm or less. Metastatic nodes were larger than those without metastasis (median (i.q.r.) 5.0 (3.2-7.0) versus 3.8 (2.4-5.2) mm; P < 0·001), but a receiver operating characteristic (ROC) curve did not identify a relevant cut-off point to predict metastatic involvement. A hazard ratio of 0.71 (95 per cent confidence interval 0.50 to 1.01) was suggestive of an association between disease recurrence and increased node size, although not significant (P = 0.056). In patients with N0 disease, there was a correlation between node size and harvest (Pearson's correlation 0.317, P = 0.002), and a nodal yield of at least 12 was associated with a larger median node size (4.3 (3.3-5.0) versus 3.4 (2.7-4.0) mm; P = 0.015).

Conclusion: The contribution of lymph nodes smaller than 3 mm to nodal staging is limited. Increased node size is associated with increased nodal yield, and could be a confounder in the relationship between prognosis and nodal harvest in patients with N0 disease.