Table 1 Results of previous studies on risk factors for underestimation

From: A prediction model for underestimation of invasive breast cancer after a biopsy diagnosis of ductal carcinoma in situ: based on 2892 biopsies and 589 invasive cancers

Variablesa,b

Significance of potential risk factors, as described in literaturec,d

 

No

Yes, univariablee

Yes, multivariablef

Age

15, 16, 18, 22, 24, 25

14, 21

 

Detection modeg

 

14, 16

 

Palpable

22

16

15, 18, 23/28, 24, 25, 26

Clinical size of mass

 

18

 

BI-RADS g

21, 23

14, 15, 16

 

Maximum size on imaging

 

16

22

Maximum size on mammographyg

 

18, 21, 26

14, 20, 24, 25

Maximum size on ultrasonographyg

 

23

15, 30

Maximum size on MRI

 

26

30

Mass on imaging

  

14

Mass on mammography

 

16, 26

18

Mass on ultrasonography

30

26

23/28

Visibility on ultrasonography

16

 

24

Type of mammographic abnormalityg

14, 21, 24

15

25

Calcifications on mammography

23, 26

 

30

Calcifications on ultrasonography

  

23/28

Suspicious findings on ultrasonography or MRI

 

30

 

Multicentric

 

14

 

Breast density

 

14

 

Residual disease on mammogram after biopsy

  

21

Calcification % removed by CNB

 

14

 

Biopsy guidance techniqueg

 

14, 16, 24, 26

 

Biopsy type CNB, VAB

24

16, 25

15, 22, 23/28

Biopsy needle gauge

 

14

 

Number of cores obtainedg

14, 21

15, 25, 26

 

DCIS grade

26

16, 23, 24, 25

14, 15, 20

Nuclear grade

26, 30

 

22

Suspicious of invasion on biopsy

  

23/28, 24

Comedo-necrosis

16, 18, 23, 26

14, 22, 25

30

Intraductal structure

  

25

Cibriform

14

 

22

Sclerosing adenosis

  

30

Hormone receptor ER/PR

  

22

Progesterone receptor

 

30

 

HER2

22

 

30

  1. aVariables in bold are variables that were analysed in this study
  2. bVariables that were analysed but were not statistically significant in any study were: mass on MRI (30), mass on ultrasonography or MRI (30), abnormality on mammography; mass, asymmetry or distortion (23), calcifications on imaging (22), suspicious findings on ultrasonography (30), suspicious findings on MRI (30), solid (14, 22) papillary (14, 22), micropapillary (14, 22), necrosis (22), oestrogen receptor (14,30), period from breast biopsy to surgery (25), Van Nuys grouping (23), family history (21), menopausal status (21) and type of first resection (18)
  3. cListed are 12 studies with at least 100 cases of underestimation
  4. dReferences in bold are of the five studies that developed a prediction model
  5. eReference 16 presents results of random-effect logistic regression models in a meta-analysis
  6. fThe multivariable significant variables of ref. 23 were used in a prediction model, as described in ref. 28
  7. gThe categories of these variables were not uniformly defined between studies
Back to article page