Skip to main content
. Author manuscript; available in PMC: 2013 Dec 6.
Published in final edited form as: Ann Surg Oncol. 2012 May 24;19(12):10.1245/s10434-012-2413-4. doi: 10.1245/s10434-012-2413-4

TABLE 3.

Univariate and multivariate Cox proportional hazard analysis of clinicopathologic and treatment factors associated with local relapse after BCS for DCIS (n = 1216)

Variable 5 year LRR rate (%) Univariate P Hazard ratio 95 % Confidence interval Multivariate P
Age
 ≥40 years 3.2 0.005 1.00 (reference)
 <40 years 10.1 2.70 1.06–6.93 0.038
Initial presentationa
 Clinical 5.6 0.026 1.00 (reference)
 Radiological 3.2 0.58 0.28–1.24 0.160
Pathological sizeb
 <1.5 cm 2.2 0.018 1.00 (reference)
 ≥1.5 cm 5.6 2.72 1.23–6.01 0.013
Margin statusc
 Negative 3.7 0.270 1.00 (reference)
 Close/positived 2.2 0.51 0.21–1.23 0.134
ER statuse
 Positive 2.6 0.078 1.00 (reference)
 Negative 5.3 2.23 0.71–7.04 0.171
Gradef
 I 2.2 0.340 1.00 (reference)
 II/III 3.7 1.80 0.54–6.03 0.340
Radiotherapy 1.00 (reference)
 Yes 2.3 0.018
 No 7.9 2.68 1.36–5.26 0.004
Endocrine use
 Yes 2.0 0.087 1.00 (reference)
 No 4.6 1.29 0.62–2.68 0.490

BCS breast-conserving surgery, DCIS ductal carcinoma-in-situ, LRR locoregional recurrence, ER estrogen receptor

a

Initial presentation signs were not available for 7 patients

b

Largest pathological sizes were recorded in 900 patients

c

Margins were not available for 12 patients

d

Positive in 4 patients; close (<2 mm) in 99 patients

e

ER status was determined in 743 patients

f

Grade was not recorded in 29 patients