Background: We sought to elucidate the associations between the examined lymph node (ELN) count and survival after sublobar resection (SLR) in pathologic stage IA non-small-cell lung cancer and assess the survival efficacy for patients undergoing SLR with an adequate ELN count compared with lobar resection (LR).
Patients and methods: The newly released U.S. Surveillance, Epidemiology, and End Results database was queried. The association between the ELN count in SLR and survival was analyzed using Cox regression and propensity score matching analyses. The optimal cutoff for the ELN count in SLR was determined using the Chow test and "segmented" function. Survival differences between SLR with an adequate ELN count and LR were compared using the log-rank test.
Results: A total of 16,630 lobectomies, 3979 wedge resections, and 1119 segmentectomies were included in the present study. The median ELN count was 3 (interquartile range [IQR], 1-7), 1 (IQR, 0-4), and 7 (IQR, 4-12) for segmentectomy, wedge resection, and lobectomy, respectively (P < .001). The optimal cutoff for ELNs was 2.40 (IQR, 2.15-2.65) and 2.59 (IQR, 2.19-3.00) for patients undergoing wedge and segmental resection. SLR patients with ≥ 3 ELNs showed better overall survival and lung cancer-specific survival compared with those with < 3 ELNs. SLR patients with ≥ 3 ELNs had a noninferior prognosis compared with LR patients after matching of potential confounders.
Conclusions: In SLR, an adequate ELN count (≥ 3) could confer an improved prognosis compared with an inadequate ELN count (< 3). SLR patients with an adequate ELN count had noninferior survival outcomes compared with LR patients when the ELNs were well matched.
Keywords: Examined lymph nodes; NSCLC; Propensity score matching; SEER database; SLR.
Copyright © 2020 Elsevier Inc. All rights reserved.