Combinative evaluation of primary tumor and lymph nodes to predict pelvic lymphatic metastasis in cervical cancer: an integrated PET-IVIM MRI study

Table 4 For all patients, univariate and multivariate analysis of parameters to predict lympha metastasis

Method: Stepwise; Dependent Y: Lymphatic metastasis Multivariate logistic regression: Enter Variable< 0.005
Univariate logistic regression analysis
Variable	coeffcients	Std Error	Wald	P
MTV	0.065374	0.026612	6.0348	0.014
TLG	0.0044791	0.0019118	5.4889	0.019
D_min	−4.41339	1.98372	4.9498	0.026
PET in lymph node diagnosis	2.39356	0.60701	15.5488	< 0.001
MRI in lymph node diagnosis	1.39544	0.55797	6.2547	0.0124
Multivariate logistic regression analysis
Variable	coeffcients	Std Error	Wald	P
TLG	0.0083138	0.0028803	8.3317	0.0039
D_min	−10.14394	3.37033	9.0588	0.0026
PET in lymph node diagnosis	4.45475	1.10161	16.3528	0.0001

ISSN: 1470-7330