Table 1.
Univariable and multivariable analysis of persistent circumferential resection margin (CRM) invasion on magnetic resonence imaging after preoperative chemoradiotherapy
| Variable | Univariate analysis |
P-value | Multivariate analysis |
P-value | ||
|---|---|---|---|---|---|---|
| Persistent CRM (n = 72) | Converted CRM (n = 40) | OR | 95% CI | |||
| Age (yr) | 64.0 ± 11.0 | 65.0 ± 10.5 | 0.617 | |||
| Sex | 1.000 | |||||
| Male | 51 (70.8) | 29 (72.5) | ||||
| Female | 21 (29.2) | 11 (27.5) | ||||
| Body mass index (kg/m²) | 22.6 ± 3.1 | 22.9 ± 2.6 | 0.602 | |||
| Hemoglobin (g/dL) | 12.7 ± 2.3 | 12.7 ± 2.4 | 0.925 | |||
| Tumor location | 0.055 | 0.007 | ||||
| Low | 17 (23.6) | 18 (45.0) | 1.000 | - | ||
| Mid/upper | 55 (76.4) | 22 (55.0) | 3.927 | 1.492–11.182 | ||
| Tumor size (cm) | 0.977 | |||||
| ≤5 | 29 (40.3) | 17 (42.5) | ||||
| >5 | 43 (59.7) | 23 (57.5) | ||||
| CEA at diagnosis (ng/mL) | 0.122 | |||||
| ≤5 | 26 (36.1) | 21 (52.5) | ||||
| >5 | 46 (63.9) | 19 (47.5) | ||||
| mrT category | 0.020 | 0.268 | ||||
| 2/3 | 47 (65.3) | 35 (87.5) | 1.000 | - | ||
| 4 | 25 (34.7) | 5 (12.5) | 1.936 | 0.628–6.771 | ||
| mrN category | 0.458 | |||||
| 0 | 6 (8.3) | 7 (17.5) | ||||
| 1 | 21 (29.2) | 10 (25.0) | ||||
| 2 | 45 (62.5) | 23 (57.5) | ||||
| mrEMVI | <0.001 | 0.030 | ||||
| No | 8 (11.1) | 16 (40.0) | 1.000 | - | ||
| Yes | 64 (88.9) | 24 (60.0) | 3.319 | 1.146–10.149 | ||
| Differentiation | 0.421 | |||||
| Well | 38 (52.8) | 25 (62.5) | ||||
| Moderately | 31 (43.1) | 15 (37.5) | ||||
| Poorly | 3 (4.2) | 0 (0) | ||||
| Pathologic CRM | 0.436 | |||||
| Involved | 26 (36.1) | 2 (5.0) | ||||
| Not involved | 46 (63.9) | 38 (95.0) | ||||
Values are presented as mean ± standard deviation or number (%).
CEA, carcinoembryonic antigen; EMVI, extramural venous invasion; OR, odds ratio; CI, confidence interval.