Between August 2012 and November 2012 12 patients affected by hystologically proved EC were enrolled in our prospective study.
MR examination was performed on a 1,5T scanner (Symphony,
Siemens) equipped with high-performing gradients (amplitude 30 mT/m),
with patient in supine position; before exam beginning,
20 mg of N-butile-scopolamine were injected i.v.
to reduce bowel movements artifacts.
Then we started with TSE T2-weighted sequences on multiple planes (FOV:220X220,
slice thickness:4mm,effective matrix:384x75,
FA:180°,
TR:4400,
TE:104 both for sagittal and coronal planes; slice thickness:3mm,
effective matrix:320x75,
FA:180°,
FOV:280x280,
TR:5452,
TE:103 for axial oblique plane),
TSE T1-weighted on axial oblique plane,
with and without fat saturation; DCE-MR imaging was performed after administration of 01.mmol/kg of a gadolinium chelate at a rate of 2mL/sec by using a three-dimensional spoiled gradient-recalled echo (GRE) T1-weighted (T1w) sequence on the axial oblique plane (thickness:3mm, FOV: 350x87.5mm,
TR:291,
TE:414).
Images were acquired prior to contrast medium injection and then during multiple phases (9) of enhancement in axial oblique and sagittal plane (precontrast and post-contrast sequences until 120 seconds after i.v.
contrast medium injection in the axial oblique plane,
post-contrast sequences at 180 seconds in both the sagittal and the axial oblique planes and a late sequence at 240 seconds in the axial plane).
Tumors were defined as lesions of heterogeneous intermediate signal intensity relative to the hyperintense normal endometrium and mildly hyperintense relative to the normal myometrium on T2-weighted images (figures 1-2); the images obtained at 120 seconds after gadolinium administration were the most accurate in the assessment of the depth of myometrial invasion (figures 3-4); delayed-phase images obtained 3-4 minutes after gadolinium administration were useful in detecting cervical stroma invasion.
The depth of myometrial invasion was calculated both on TSE T2-weighted sequences and on DCE-MRI ones on the axial oblique and the sagittal plane.
It was defined as the distance between inner myometrial interface and deepest myometrial invasion point.
The myometrial invasion ratio was invasion depth divided by myometrial thickness,
defined as <50% vs >50%,
according with FIGO new staging.
The standard of reference were the hystologic results.
Cases with imaging myometrial involvement more or equal than 50% were considered as true positive (TP) and false positive(FP); true negative (TN) and false negative (FN) cases were those with imaging myometrial invasion less than 50%.
Diagnostic performance of both T2W and DCE sequences in the assessment of myometrial invasion were evaluated,
considering sensibility,
specificity,
positive and negative predictive value and diagnostic accuracy for each one,
with a significative statistical correlation of p value <0,05.