cohort study analyzed the charts of all women who underwent a hysterectomy for various
indications at the department of obstetrics and gynecology, Benha University,
between the January 2010 and June of 2017 after getting the legitimate ethical
approvals from the local ethics committee.
hysterectomy indications had a wide range of complaints such as abnormal
uterine bleeding, chronic pelvic pain, myomas, endometriosis, adenomyosis,
cervical, endometrial and ovarian cancers. The histopathological diagnosis of
the D&C samples was compared with the postoperative one of the hysterectomy
specimen’s without any considering the hysterectomy type or indication.
excluded incomplete medical records and women with an endometrial sample
obtained before the hysterectomy by an interval of more than two months. After
exclusion, 829 specimens obtained from 531 premenopausal and 298 postmenopausal
women were enrolled in the study.
endometrial samples were collected by D technique performed by certified
physicians under general anesthesia. After the woman was placed in the
lithotomy position, the vagina and perineum were cleaned by betadine. Bimanual pelvic
examination was used to locate the position of uterus after evacuating the
urinary bladder. Then, Hegar dilators were used to gradually dilate the cervix
before starting curettage of the uterine cavity walls.
in paraffin, was used for fixation of all endometrial histological sections before
cutting them into 4-?m-thick sections. Also, hematoxylin and eosin was the
standard stain of choice for histopathological examination which was carried by
an experienced gynecological histopathologist who evaluated sections by using
(Leica RM2125 / RM2125RT Rotary microtome, Leica Biosystems©, Nussloch GmbH,
lesions were evaluated according to world Health Organization (WHO)
classification of tumors of female reproductive organs (2014) that stratify
endometrial epithelial lesions into; tumor like lesions, precursors of endometrial
carcinoma, leiomyoma and endometrial carcinomas. WHO considered endometrial
polyp, Arias-Stella reaction and metaplasias as tumor-like lesions and
endometrial hyperplasias (hyperplasia without atypia and atypical hyperplasia)
as precursors of endometrial carcinoma.
not to miss other categories of endometrial lesions we classified the samples into
three groups. The benign pathology group included the normal and benign
endometrial pathology such as atrophic endometrium, chronic endometritis,
endometrial polyps, submucous myomas, glandular-Stromal breakdown and
disordered proliferative endometrium. Premalignant (precursor) endometrial
pathology group that reported samples of simple, complex and atypical
endometrial hyperplasia and malignant endometrial pathology (adenocarcinoma or
others) were also included.
analyzed using IBM SPSS Statistics version 23 (IBM© Corp., Armonk, NY, USA).
Contingency tables were constructed in which the result of hysterectomy biopsy
(gold-standard test) was cross-tabulated against that of D biopsy. The
following diagnostic indices together with their 95% confidence limits (95% CI)
were calculated: sensitivity, specificity, positive and negative predictive
values (PPV and NPV). Overall accuracy or correct classification rate was also
accuracy of D biopsy was evaluated using the whole study population as
well as subgroup analysis was conducted for premenopausal and postmenopausal
women separately. Inter-method agreement was examined using the weighted kappa
legible specimens were analyzed for the conformity of preoperative D
tissue samples and the more conclusive hysterectomy specimens regarding their histopathological
There were 179(21.6%) women who had
“insufficient tissue samples” for adequate pathological reporting. Getting
insufficient endometrial tissue samples was more frequent in 147 postmenopausal
than 32 premenopausal women. The findings of most of the “insufficient tissue”
group were atrophic endometrium (90 cases, 61.2%) among postmenopausal women
and normal endometrium (22 cases, 68.75%) in premenopausal women. (Table 1)
histopathological results for the study group with sufficient samples (n=650) are
compared in Tables (2, 3
and 4) as well as the subgroups of premenopausal (n=499) and
postmenopausal (n=151) women. Comparing the findings showed that endometrial
polyps and atrophic endometrium were the commonly missed lesions for
premenopausal and postmenopausal women respectively.
insufficient samples, sensitivity, specificity, PPV, NPV and correct
classification rate (accuracy) with their 95% confidence limits of the
curettage results of all the recorded endometrial lesions collectively and separately
are listed in Tables (5&6) respectively.
Table (7) shows a
good agreement between D biopsy and hysterectomy diagnoses regarding the
ultimate classification of women in the whole study population (weighted ?
= 0.756, 95% CI = 0.692-0.819) and postmenopausal women (weighted ? =
0.834, 95% CI = 0.761-0.906). However, the agreement is moderate in the
premenopausal group (weighted ? = 0.568, 95% CI = 0.433-0.704).