STUDY POPULATION
In the period between February 2013 and January 2017,
in agreement with the Colleagues of Gynecology Department,
27 patients were selected who underwent MRI diagnostic investigation for deep pelvic endometriosis and subsequently underwent videlaparoscopic (VLS) therapeutic treatment.
The patients’ ages were distributed from a minimum of 26y to a maximum of 52y,
with an average of 35.85y.
The diagnosis of ureteral endometriosis was confirmed by histological examinations of the periureteral tissues and the nodules of endometriosis were surgically removed.
MRI TECHNIQUE
The MR examinations were performed with two 1.5 Tesla magnets (Magnetom Avanto,
Siemens Medical Solutions,
Achieva XR,
Philips Medical System) using a 4-channel surface coil (SENSE - body coil).
The examination protocol included two phases: the first one involves a MR imaging of the pelvis at high resolution (RM-HR),
after distension of the vagina with 50-60 ml of sterile gel and intravenous administration of 10 mg of Buscopan,
while the second phase was a Clisma-RM with intravenous administration of mdc.
The sequences executed in the first phase were: T2 (TSE) oriented in the three planes of the space and T2 Para-axial (Fig.1),
T1 TSE axial,
T1 VIBE axial.
The second phase involved colonic idric distension through an endorectal Foley catheter (20 F) and infusion of 1,500 ml of PEG solution at 37 °C,
after pharmacological intestinal hypotonization by further intravenous administration of 10 mg of Buscopan.
Once the bowel had been completely relaxed,
intravenous injection of 0.1 ml / kg paramagnetic contrast enhancement (C.E.) was performed.
The sequences acquired in this phase were: T1 VIBE in coronal after C.E.,
true-FISP in coronal and in axial,
T2 TSE in sagittal and para-axial as in the first phase,
T2 HASTE in coronal,
T1 VIBE coronal in late phase for the study of ureters (URO- RM).
In post-processing,
Maximum Intensity Projection (MIP) reconstructions of urographic sequences (T1 VIBE coronal and sagittal) (Fig.2) were performed.
ANALYSIS OF THE IMAGES
Two radiologists with different experience in pelvic MR imaging interpretation (expert reader: 15 years,
inexperienced reader,
with 4 years of experience) separately and independently analyzed the images of VLS positive patients for ureteral localization of disease and images of patients affected by deep pelvic endometriosis without intra-operative confirmation of ureteral disease.
A total of 27 examinations were randomly evaluated.
Each reader evaluated the presence or absence of endometriotic nodules at parametrial level and / or at the vaginal-rectum septum; its dimensions; the distance between the urinary excretory path and the nodulation; possible contact of the lesion with the ureter (Fig.3),
quantifying it as <180 °,
between 180 and 360 ° and equal to 360 °; the presence of hydronephrosis,
in order to evaluate not only the presence or absence of ureteral involvement but also indicate direct and indirect MR signs of such localization.
STATISTICAL ANALYSIS
Sensitivity,
Specificity,
positive and negative predictive value of disease localization were calculated for each reader,
considering each ureter as a possible lesion site and using the operative report as gold standard.
The student's t-test then calculated the mean distance from ureter nodulations in positive and negative VLS patients to assess the presence of a statistically significant difference between the two averages.
A ROC curve was then constructed to identify the distance below which the ureteral involvement should be considered with reasonable certainty.
Then we divided the patients into three groups,
based on the size of the lesion: <2cm,
between 2 and 3 cm,
greater than 3 cm,
to assess by chi-square test if there was a statistically significant prevalence of the positive patients in one of these groups.