Research Article

Early Cervical Cancer: Predictive Relevance of Preoperative 3-Tesla Multiparametric Magnetic Resonance Imaging

Table 3

Univariable and multivariable analysis of factors associated with diagnostic inaccuracy of preoperative MRI staging.

Under-diagnosisOver-diagnosis
Univariable MultivariableUnivariableMultivariable
UOR 95 CIOver- all PAOR 95 CIOver- all PUOR 95 CIOver-all PAOR 95 CIOver- all P

Age(y)1.02
(0.99-1.05)
0.2540.99
(0.95-1.02)
0.409

Menopause0.6050.394

 No11

 Yes1.19
(0.62-2.27)
1.37
(0.66-2.83)

Parity1.08
(0.83-1.39)
0.5841.13
(0.82-1.55)
0.449

History of at least one normal delivery0.7510.314

 No11

 Yes0.86
(0.35-2.15)
0.62
(0.43-1.58)

Histology0.5210.159

 Squamous11

 Adenocarcinoma2.11
(1.07-4.16)
0.25
(0.074-0.87)

 Adeno-squamous1.72
(0.33-9.05)
1.38
(0.27-7.17)

0.0310.0340.0550.064

 Squamous1111

 Adenocarcinoma or adenosquamos2.07
(1.07-3.99)
2.07
(1.06-4.07)
0.38
(0.14-1.02)
0.34
(0.11-1.07)

SCC ag(ng/mL)0.96
(0.89-1.03)
0.2571.01
(0.99-1.04)
0.341

MRI tumor size0.76
(0.63-0.92)
0.0040.76
(0.63-0.92)
0.0051.46
(1.18-1.81)
0.0011.51
(1.06-2.16)
0.023

MRI parametrial invasion0.029<0.0001<0.0001

 No111

 Yes0.11
(0.14-0.79)
11.94
(5.21-27.35)
73.73
(8.89-611.38)

MRI uterine corpus invasion0.3470.3630.003

 No111

 Yes1.49
(0.65-3.49)
1.54
(0.61-3.90)
0.13
(0.03-0.49)

MRI deep stromal invasion0.8400.0060.051

 No invasion & partial invasion111

 Complete invasion1.07
(0.55-2.07)
2.92
(1.37-6.23)
0.19
(0.01-1.01)