Table 3.
Univariate and multivariate logistic regression analyses of disease underestimation based on core needle biopsy (CNB) (N = 56)
| Effect | Univariate | Multivariate | ||||||
|---|---|---|---|---|---|---|---|---|
| Odds ratio | 95% Wald confidence limits | p value | Odds ratio | 95% Wald confidence limits | p value | |||
| ADH | 1.44 | 0.40 | 5.20 | 0.574 | 1.51 | 0.14 | 16.89 | 0.738 |
| FEA | 0.68 | 0.20 | 2.29 | 0.534 | 0.68 | 0.13 | 3.60 | 0.651 |
| RS | – | – | ||||||
| LN | – | – | ||||||
| Extension | 1.14 | 0.39 | 3.38 | 0.813 | 0.94 | 0.30 | 3.03 | 0.923 |
| Multifocality | 1.40 | 0.36 | 5.47 | 0.628 | 1.48 | 0.35 | 6.26 | 0.596 |
| Calcification | 1.49 | 0.43 | 5.19 | 0.534 | 1.76 | 0.41 | 7.46 | 0.445 |
| Papillomas | 1.00 | 0.20 | 4.96 | 0.999 | 1.88 | 0.15 | 23.44 | 0.625 |
Associated B3 lesions, diameter of the lesion, multifocality, or calcification had no significant effect on upgrade rates in core needle biopsy samples (14G)
ADH atypical ductal hyperplasia, FEA flat epithelial atypia, RS radial scar, LN lobular neoplasia, classical type