Table 4.
Univariate and multivariate Cox regression models for predicting structural recurrence.
| Univariate Analysis | Multivariate Analysis (Model 1) | Multivariate Analysis (Model 2) | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | p | HR (95% CI) | p | HR (95% CI) | p | |
| Age | 0.98 (0.95–1.00) | 0.057 | ||||
| Sex, men | 1.39 (0.73–2.64) | 0.314 | ||||
| Size | 1.19 (0.99–1.42) | 0.062 | ||||
| Gross ETE, positive | 5.17 (2.74–9.77) | <0.001 | 2.89 (1.47–5.69) | 0.002 | 1.56 (0.75–3.21) | 0.232 |
| N stage 0 1a 1b |
1.00 (reference) 12.96 (2.78–60.4) 15.42 (3.66–65.2) |
0.001 <0.001 |
8.94 (1.79–44.8) 11.0 (2.52–48.1) |
0.008 0.001 |
4.76 (0.91–24.7) 1.16 (0.20–6.75) |
0.064 0.868 |
| LNR (%) | 1.05 (1.04–1.06) |
<0.001 | 1.04 (1.02–1.06) |
<0.001 | ||
| Postoperative calcitonin ≥ 5 pg/mL |
15.76 (3.80–65.40) |
<0.001 | 5.43 (1.13–26.1) |
0.035 | ||
| Postoperative CEA ≥5 ng/mL |
7.79 (4.11–14.77) |
<0.001 | 2.83 (1.39–5.78) |
0.004 | ||
HR, hazard ratio; CI, confidence interval; ETE, extra thyroidal extension; LNR, metastatic lymph node ratio; CEA, carcinoembryonic antigen.