Skip to main content
. 2022 Mar 30;11(3):233–246. doi: 10.1159/000521747

Table 3.

Univariate and multivariate analyses of MRI features in discriminating LD-type iCCA

MR imaging data Univariate
Multivariate
OR 95% CI p value OR 95% CI p value
Median size (cm) 1.130 0.995, 1.283 0.060
Location (center) 0.143
 Peripheral
 Midzone 0.959 0.278, 3.311 0.947
 Central/diffuse 5.395 1.001, 29.071 0.050
Location (perihilar border) 0.002
 Peripheral
 Midzone 2.760 1.007, 7.567 0.048
 Central/diffuse 4.601 1.985, 10.661 <0.001
Contour
 Round/lobulated
 Infiltrative 18.621 5.116, 67.775 <0.001 14.176 2.459, 81.708 0.003*
Liver surface change 0.654 0.299, 1.430 0.287
BD dilatation <0.001 0.066
 No dilatation
 Peritumoral dilatation 8.691 3.665, 20.610 <0.001 2.382 0.779, 7.279 0.128
 Diffuse dilatation 17.778 2.992, 105.641 0.002 9.673 1.217, 76.890 0.032*
Liver cirrhosis 1.034 0.487, 2.199 0.930
AP enhancement pattern <0.001 0.006*
 Nonrim APHE
 Rim APHE 0.977 0.381, 2.503 0.961 1.569 0.423, 5.824 0.501
 No APHE 14.844 3.886, 56.706 <0.001 17.848 2.685, 118.643 0.003*
Peritumoral arterial enhancement 2.917 1.264, 6.730 0.012
Delayed central enhancement 1.469 0.620, 3.479 0.382
Portal phase washout pattern 0.034
 No washout
 Peripheral washout 0.139 0.031, 0.630 0.011
 Nonperipheral washout 0.694 0.291, 1.652 0.409
TP appearance 0.003
 No hypointensitiy
 Targetoid appearance 0.143 0.046, 0.446 0.001
 Nontargetoid hypointensity 0.563 0.232, 1.363 0.203
HBP appearance
 Targetoid appearance
 Nontargetoid hypointensity 2.133 1.044, 4.357 0.038
Diffusion-weighted imaging appearance 0.007
 Targetoid restriction
 Marked diffusion restriction 0.965 0.334, 2.788 0.947
 Restriction, but nontargetoid nor marked 3.628 1.548, 8.504 0.003
Vascular invasion 8.667 3.354, 22.398 <0.001 4.477 1.298, 15.441 0.018*
Multiplicity 2.394 1.028, 5.572 0.043
Regional lymph node metastasis 2.206 0.958, 5.081 0.063
Necrosis or mucin 1.234 0.600, 2.537 0.568

OR, odds ratio; APHE, arterial phase hyperenhancement.

*

Statistically significant results from logistic regression analysis. Variables with p < 0.05 on univariate logistic regression analysis were utilized as input variables for multivariate logistic regression analysis.