Skip to main content
. 2020 Aug 12;20:95. doi: 10.1186/s12880-020-00494-z

Table 3.

Univariate and multivariate analysis using a cumulative logistic regression method to predict histologic upgrade to microinvasion and invasion

Univariate analysis Multivariate analysis
Odds ratio (95% CI) P value Adjusted Odds ratio (95% CI) P value
Age (old vs. young) 1.02 (1.00–1.05) 0.070
Biopsy method (US-CNB vs. SVAB) 2.78 (1.28–6.01) 0.009
Nuclear grade (grade 3 vs. grade 1–2) 7.51 (2.38–23.75) < 0.001
Necrosis (present vs. absent) 1.74 (1.02–2.98) 0.044
ER (absent vs. present) 2.97 (1.63–5.38) < 0.001
PR (absent vs. present) 3.53 (2.02–6.17) < 0.001 2.40 (1.29–4.44) 0.006
HER2 (present vs. absent) 2.81 (1.62–4.87) < 0.001
Ki-67 (high vs. low) 3.44 (1.98–6.00) < 0.001 2.42 (1.30–4.50) 0.005
Mammographic calcification (present vs. absent) 1.70 (0.96–3.02) 0.072
BPE on MRI (minimal or mild vs. moderate or marked) 1.88 (1.00–3.55) 0.051
MR type of lesion (mass or NME vs. non-visualization)
  Mass 8.81 (1.10–70.93) 0.041 8.84 (1.05–74.04) 0.045
NME 11.95 (1.50–95.20) 0.019 11.17 (1.35–92.36) 0.025

CI confidence interval; US-CNB US-guided core needle biopsy; SVAB stereotactic vacuum-assisted biopsy; ER estrogen receptor; PR progesterone receptor; HER2 human epidermal growth factor receptor 2; BPE background parenchymal enhancement; NME non-mass enhancement