Skip to main content
. 2019 May 14;19:444. doi: 10.1186/s12885-019-5632-2

Table 3.

Univariate and multivariate logistic regression predictors of hilar lymph node metastasis

Independents Variables Univariate Predictors Multivariate Predictors
N Metastasis/N total Odds Ratio (95% CI), P-value B Odds Ratio (95% CI), P-value
Age (years)
 ≤ 60 31/243 (12.8%) Reference
 >  60 33/245 (13.5%) 1.07 (0.63–1.80), 0.816
Sex
 Female 23/238 (9.7%) Reference
 Male 41/250 (16.4%) 1.83 (1.06–3.16), 0.029*
Maximum diameter of tumor
 ≤1.6 cm 10/228 (4.4%) Reference
 > 1.6 cm 51/256 (19.9%) 5.42 (2.68–10.97), < 0.0001*
Position 0.122*
 Right Upper Lobe 15/157 (9.6%) Reference
 Right Middle Lobe 5/20 (25.0%) 3.16 (1.01–9.90), 0.049*
 Right Lower Lobe 13/68 (19.1%) 2.24 (1.00–5.01), 0.050*
 Left Upper Lobe 16/110 (14.5%) 1.61 (0.76–3.42), 0.213
 Left Lower Lobe 6/80 (7.5%) 0.77 (0.29–2.06), 0.600
 Ipsilateral Mixed lobes 7/33 (21.2%) 2.55 (0.95–6.86), 0.064*
 Bilateral Mixed lobes 2/16 (12.5%) 1.35 (0.28–6.53), 0.707
Pathological type
 Adenocarcinoma 51/370 (13.8%) Reference
 Others 13/118 (11.0%) 0.77 (0.41–1.48), 0.439
Tumor differentiation
 No 3/101 (3.0%) Reference
 Yes 60/368 (16.3%) 6.36 (1.95–20.74), 0.002*
Pathological morphology
 No 21/167 (12.6%) Reference
 Yes 42/306 (13.7%) 1.11 (0.63–1.94), 0.725
Pulmonary membrane invasion
 No 33/342 (9.6%) Reference
 Yes 14/55 (25.5%) 3.20 (1.58–6.47), 0.001*
Bronchial mucosa and cartilage invasion
 No 44/436 (10.1%) Reference
 Yes 20/52 (38.5%) 5.57 (2.94–10.56), < 0.0001* 1.13 3.11 (1.19–8.13), 0.021#
Vascular invasion
 No 45/445 (10.1%) Reference
 Yes 19/41 (46.3%) 7.68 (3.86–15.26), < 0.0001* 1.09 2.98 (1.14–7.81), 0.026#
Nerve invasion
 No 59/477 (12.4%) Reference
 Yes 5/11 (45.5%) 5.90 (1.75–19.95), 0.004*
CEA
 ≤2.21 ng/ml 5/197 (2.5%) Reference
 > 2.21 ng/ml 44/219 (20.1%) 9.66 (3.74–24.90), < 0.0001* 2.14 8.49 (2.49–28.97), 0.001#
Albumin
 ≤43.1 g/l 45/247 (18.2%) Reference
 > 43.1 g/l 16/227 (7.0%) 0.27 (0.11–0.64), 0.003*

*Only variables with P value less than 0.20 were included in the multivariate analysis. # P < 0.05 Sum might not always be in total because of missing data