|
|
|
| The value of MRI in the histological grading of endometrial cancer |
| Ren Jie, Jin Yanfang, Cui Tingting, Fu Zichao, Zhang Chao, Yue Yunlong |
| Department of Magnetic Resonance, Beijing Shijitan Hospital, Capital Medical University, Beijing 100038, China |
|
|
|
|
Abstract Objective To investigate the value of magnetic resonance imaging (MRI) in the histological grading of endometrial cancer.
Method MR images of 64 patients with endometrial cancer confirmed by surgery from 2017 to 2020 were analyzed retrospectively from
Beijing Shijitan Hospital Capital Medical University. The maximum left and right diameter, anterior and posterior diameter and head and
foot diameter of tumor and uterus were measured on the maximum display level of the tumor and the median sagittal plane of the image
respectively. According to the simplified ellipsoid volume formula, the volume of tumor and uterus was calculated and their ratio neoplasm/
uterus (N/U) was calculated, taking pathological findings as the gold standard, the differences of uterine volume, tumor volume and N/
U at different histological grades (G1, G2 and G3) were compared by Nonparametric test (Kruskal⁃Wallis test) and one⁃way ANOVA.
According to the depth of tumor invasion and 5⁃year survival rate, G1, G2 and G3 tumors were divided into high⁃grade group and
low⁃grade group. Receiver operating characteristic curve（ROC curve）was used to analyze the value of N/U in different group in predicting
the histological grade of tumor before operation. Result Among the 64 cases of endometrial cancer, there were 20 cases of G1 grade,
24 cases of G2 grade and 20 cases of G3 grade. There was no significant difference in tumor volume among the three groups（H=3.528, P=0.171）. The N/U values of G1, G2 and G3 tumors were（0.13±0.07）,（0.19±0.06）,（0.21±0.08）, respectively. The difference among
the three groups was statistically significant（F=9.04, P<0.01）. There was no statistically significant between high⁃grade group and
low⁃grade group. Distinguishing G1 tumors from G2/G3 tumors with the cutoff value of 0.16, the area under curve (AUC) was 0.80, the
sensitivity was 79.5% and the specificity was 80%. Conclusion MRI has certain diagnostic value for histological grading of endometrial
cancer.
|
|
Received: 08 October 2021
|
|
|
|
|
|
|
