Skip to main content
. 2016 Aug 5;18:82. doi: 10.1186/s13058-016-0742-0

Table 3.

Univariable Cox proportional hazard analysis of relationship between continuous MRI variables and recurrence-free survival

Recurrence-free survival
MRI variable Median P value Hazard ratio 95 % CI
Baseline (before NAC)
 Largest diameter MIP/initial enhancement (90 s) 43 mm 0.238 1,009 0.994, 1.024
 Largest diameter plateau/washout enhancement (450 s) 33 mm 0.027 1,017 1.002, 1.033
 Initial enhancement (90 s) % 152 % 0.326 0.997 0.99, 1.003
 Late enhancement (450 s) % −13 % 0.947 0.999 0.967, 1.032
During NAC
 Largest diameter MIP/initial enhancement (90s) 30 mm 0.155 1,011 0.996, 1.026
 Largest diameter plateau/washout enhancement (450 s) 17 mm 0.006 1,024 1.007, 1.041
 Initial enhancement % 135 % 0.993 1.00 0.995, 1.005
 Late enhancement % −4 % 0.600 0.995 0.975, 1.015
After NAC
 Largest diameter MIP/initial enhancement (90s)a 22 mm 0.140 1.01 1.00, 1.03
 Largest diameter plateau/washout enhancement (450 s) 0 mm 0.003 1.03 1.01, 1.051
 Initial enhancement % 100 % 0.057 1,005 1.00, 1.011
 Late enhancement % 9 % 0.815 0.998 0.983, 1.014
Percent change after NAC - baseline NAC %
 Largest diameter MIP/initial enhancement (90 s)b −40 % 0.280 1.01 0.99, 1.02
 Largest diameter plateau/washout enhancement (450 s)c −100 % 0.021 1,013 1.002, 1.024
Percent change after NAC - during NAC %
 Largest diameter MIP/initial enhancement (90 s) mmb −30 % 0.290 1.01 0.99, 1.02
 Largest diameter plateau/washout enhancement (450 s)c −61 % 0.066 1,008 0.999, 1.017

aPatients with largest diameter 0 mm were excluded; bpatients with change −100 % were excluded; cpatients without washout/plateau on both scans were excluded. MRI magnetic resonance imaging, CI confidence interval, NAC neoadjuvant chemotherapy, MIP maximum intensity projection. Numbers in bold are significant