Table 1.
Characteristics | Training Set (N = 423) | Internal test set (N = 182) | External test set (N = 73) | ||||||
---|---|---|---|---|---|---|---|---|---|
CLNM (+) (N = 218) | CLNM (-) (N = 205) | p-value | CLNM (+) (N = 94) | CLNM (-) (N = 88) | p-value | CLNM (+) (N = 46) | CLNM (-) (N = 27) | p-value | |
Age, mean ± SD, years | 44± 12 | 44 ± 12 | 44± 12 | 44 ± 12 | 44 ± 12 | 44 ± 12 | |||
< 45, No. (%) | 131 (60.09) | 84 (40.98) | <0.001 | 52 (55.32) | 41 (46.59) | 0.239 | 21 (45.65) | 12 (44.44) | 0.92 |
≥ 45, No. (%) | 87(39.91) | 121 (59.02) | 42 (44.68) | 47 (53.41) | 25 (54.35) | 15 (55.56) | |||
Sex, No. (%) | |||||||||
Male | 70 (32.11) | 45 (21.95) | <0.001 | 24 (25.53) | 19 (21.59) | 0.532 | 18 (39.13) | 2 (28.57) | 0.793 |
Female | 148 (67.89) | 160 (78.05) | 70 (74.47) | 69 (78.41) | 28 (60.87) | 25 (71.43) | |||
CT-maximum diameter*, No. (%) | |||||||||
<1cm | 88 (40.37) | 120 (58.54) | <0.001 | 35 (37.23) | 55 (62.50) | <0.05 | 19 (41.30) | 12 (44.44) | 0.793 |
≧1cm | 130 (59.63) | 85(41.46) | 59 (62.77) | 33 (37.50) | 27 (58.70) | 5 (55.56) | |||
CT-reported LN status, No. (%) | |||||||||
LN-negative | 114 (52.29) | 147 (71.71) | <0.001 | 45 (47.87) | 68 (77.27) | <0.05 | 10 (21.74) | 4 (14.81) | 0.468 |
LN-suspicious | 39 (17.89) | 45 (21.95) | 23 (24.47) | 12 (13.64) | 0 (0.00) | 0 (0.00) | |||
LN-positive | 65 (29.82) | 13 (6.34) | 26 (27.66) | 8 (9.09) | 36 (78.26) | 23 (85.19) |
Data are n (%) unless otherwise indicated. No significant divergences were found between CLNM (+) and CLNM (-) in terms of age, sex, CT-maximum diameter, CT-reported LN status in the training set (p < 0.05). CT-longest diameter and CT-reported LN status were significantly different between CLNM (+) and CLNM (-) in the internal test set (p < 0.05), but not in the external test set.
CT, computed tomography; LN, lymph node; CLNM, central lymph node metastasis.
*Largest diameter of the target primary lesion.