Skip to main content
. 2015 Oct 15;15:709. doi: 10.1186/s12885-015-1669-z

Table 2.

Multivariable analysis of predictors for level IB LNs metastases at diagnosis in 1438 patients

Variable HR 95 % CI P *
Age, ≧50 years vs. <50 years 1.51 0.78–2.94 .219
T classification, T4 vs. T1-3 1.16 0.53–2.52 .708
Nasal cavity involvement, (+) vs. (−) 1.31 0.65–2.64 .446
Oropharynx involvement, (+) vs. (−) 2.59 1.18–5.69 .018
Positive RLNs, (+) vs. (−) 2.85 0.86–9.50 .088
Positive CLNs, (+) vs. (−) 2.53 0.80–8.01 .113
LN necrosis, (+) vs. (−) 1.22 0.59–2.52 .594
LNs with ES, (+) vs. (−) 0.57 0.27–1.19 .131
DLN-IIa ≥ 20 mm or level IIa LNs with ES, (+) vs. (−) 2.21 1.10–4.46 .026
MAD of LNs ≥30 mm, (+) vs.(−) 1.51 0.70–3.25 .293
Positive bilateral CLNs, (+) vs.(−) 1.95 0.97–3.92 .061
Positive CLNs at supraclavicular fossa, (+) vs. (−) 2.04 0.87–4.82 .103

Abbreviations: LNs, lymph nodes; HR, hazard ratio; 95 % CI, 95 % confidence interval; RLNs, retropharyngeal lymph nodes; CLNs, cervical lymph nodes; DLN-IIa, greatest dimension of level IIa lymph nodes; MAD, maximal axial diameter; ES, extra-capsular spread

*P-values were calculated using a binary logistic regression model