Skip to main content

Table 3 Univariable binary logistic regression analysis, based on generalized linear mixed - effects model, entire dataset

From: Factors influencing the detection of treatable epileptogenic lesions on MRI. A randomized prospective study

Factors associated with correct interpretation of the MRI scan

p - value

Odds ratio

95% CI

Lesion Category

 Nonlesional

< 0.001

0.395

0.248–0.629

 Hippocampal Sclerosis / Limbic Encephalitis

< 0.001

0.349

0.220–0.555

 Periventricular Nodular Heterotopia

< 0.001

0.076

0.047–0.121

 Gliotic scar / Hemosiderin / Cavernoma

0.005

0.505

0.315–0.810

 Focal Cortical Dysplasia

< 0.001

0.070

0.044–0.113

 Low Grade Epilepsy Associated Tumor

Reference

Medical Specialty

 Radiology

0.04

0.590

0.354–0.985

 Neurology

< 0.001

0.379

0.246–0.584

 Epileptology

0.32

0.774

0.467–1.283

 Neuroradiology

Reference

Postgraduate Experience

 <5 y

0.01

0.550

0.342–0.884

 5–10 y

0.03

0.634

0.420–0.958

 ≥10 y

Reference

Experience with MRI in Epilepsy

 None

0.001

0.333

0.175–0.634

 Yearly

0.03

0.509

0.282–0.918

 Monthly

0.008

0.466

0.265–0.818

 Weekly

0.28

0.707

0.378–1.326

 Daily

Reference

Experience with MRI field strength

 1 T

0.24

0.490

0.151–1.590

 1.5 T

< 0.001

0.524

0.374–0.734

 3 T

Reference

Experience with Focus Hypothesis in Epilepsy

 None

0.64

0.891

0.551–1.443

 <5 y

0.31

0.758

0.447–1.283

 5–10 y

0.53

1.187

0.699–2.017

 ≥10 y

Reference

Teaching Course in MRI completed

 No

0.07

0.716

0.497–1.032

 Yes

Reference

Focus Hypothesis provided

 No

0.001

0.682

0.544–0.855

 Yes

Reference