Comparison of systematic mediastinal lymph node dissection versus systematic sampling for lung cancer staging and...

This self-controlled prospective study was designed to investigate the efficacy of systematic sampling (SS), compared with systematic mediastinal lymph node dissection (SMLD), for pathologic staging and completeness of surgery.

Over a period of 11 mo, 110 patients with lung cancer were enlisted and treated by pulmonary resection. Surgeons systematically sampled mediastinal lymph nodes prior to pulmonary resection, and after pulmonary resection SMLD was performed to each patient using Mountain's procedure [1].

After SMLD, pN status was classified as N0 in 57 cases, N1 in 27, and N2 in 26. SS detected 38.3% of pooled nodes and 37.6% of pooled positive nodes collected from SMLD. Pathologic diagnosis after SS was understaged in nine cases (8.2%) compared with staging after SMLD. However, surgery was incomplete in 24 cases (21.8%) if SMLD was not performed after sampling. Negative predictive value for SS was 86.8% on the right side, and 95.0% on the left. Three categories were generated according to pN status: negative nodes in SS and additional negative nodes from SMLD [S(-)D(-)], negative nodes in SS but additional positive nodes from SMLD [S(-)D(+)], and positive nodes in SS [S(+)D(+)]. cN2 (P=0.000) and CEA level (P=0.001) were correlated with pN status. There was significant overall survival difference between non-N2 group and N2 group (P=0.002).

SMLD may harvest about three times of mediastinal lymph nodes compared with SS. SS is more likely to affect the completeness of surgery instead of underrating pathologic stage.