HIGH-RESOLUTION MR-IMAGING OF STAGE-I CERVICAL NEOPLASIA WITH A DEDICATED TRANSVAGINAL COIL - MR FEATURES AND CORRELATION OF IMAGING AND PATHOLOGICAL FINDINGS
Nm. Desouza et al., HIGH-RESOLUTION MR-IMAGING OF STAGE-I CERVICAL NEOPLASIA WITH A DEDICATED TRANSVAGINAL COIL - MR FEATURES AND CORRELATION OF IMAGING AND PATHOLOGICAL FINDINGS, American journal of roentgenology, 166(3), 1996, pp. 553-559
OBJECTIVE. The purposes of this study were to assess the appearance of
stage 1 neoplasia of the cervix by high-resolution MR imaging with an
enveloping transvaginal receiver coil and to correlate the imaging fi
ndings with the pathologic findings. SUBJECTS AND METHODS. Fifteen pat
ients (25-73 years old; mean, 40 years old) with clinical stage I dise
ase were examined with a 37-mm-diameter ring design solenoid receiver
coil placed around the cervix. Axial 2.5-mm contiguous slices were obt
ained with a field of view of 10-15 cm on a 1.0-T HPQ Vista scanner wi
th T1-weighted (660/20 msec [TR/TE]) and T2- weighted (2500/80 msec) s
pin-echo sequences and dynamic gradient-echo sequences during injectio
n of gadopentetate dimeglumine (0.1 mmol/kg). Ten patients subsequentl
y underwent Wertheim's hysterectomy, two underwent radiotherapy, two u
nderwent extended cone biopsy for microinvasive disease, and one under
went a punch biopsy. For seven of 10 patients who had a hysterectomy,
the widths of the tumor and the residual stroma were measured at eight
radial points on the transverse images and at corresponding points on
the histologic specimens at 5, 10, 15, 20, and 25 mm from the ectocer
vix. We then compared the widths of the tumor and the stroma on images
and histologic specimens at each of these 40 points. Tumor volumes we
re calculated from the MR imaging and pathologic data and compared. Fo
r the other three patients, detailed MR imaging-pathology correlation
was not possible because of multifocal tumor distribution (two patient
s) and insufficient detailed pathologic data (one patient). RESULTS. T
hree carcinoma types were recognized. Squamous carcinoma (nine cases)
was seen as a centrally expanding intermediate-signal-intensity mass,
whereas oat (small)-cell carcinoma (one case) and clear cell carcinoma
(one case) showed a multifocal distribution. For patients who had a r
adical hysterectomy, we noted good agreement between the widths of the
tumor and the stroma determined by MR imaging and histology. Tumor vo
lumes were determined to be 0 -28.2 cm(3) by MR imaging and 0-18.4 cm(
3) by pathology. We observed tumor extension into the immediate parame
trium in four patients by MR imaging; one of these cases was not confi
rmed at surgery. Parametrial extension was not underestimated by MR im
aging in any case. CONCLUSION. High-resolution imaging of the cervix w
ith a transvaginal coil provides accurate assessment of the intra- and
extracervical extents of tumors in clinical stage 1 cervical neoplasi
a.