Histopathologic analysis of endometrial biopsies in a Northern Nigerian Teaching Hospital

Murtala Abubakar; Balarabe Kabir; Almustapha Aliyu Liman; Sani Muhammed Shehu; Bilkisu Lawal Kankia; Anisa Yahya; Zainab Adamu Ali

doi:10.4103/atp.atp_29_20

ORIGINAL ARTICLE

Year : 2020 | Volume : 11 | Issue : 2 | Page : 101-106

Histopathologic analysis of endometrial biopsies in a Northern Nigerian Teaching Hospital

Murtala Abubakar¹, Balarabe Kabir¹, Almustapha Aliyu Liman¹, Sani Muhammed Shehu¹, Bilkisu Lawal Kankia², Anisa Yahya², Zainab Adamu Ali¹
¹ Department of Pathology, Ahmadu Bello University Teaching Hospital, Zaria, Nigeria
² Department of Obstetrics and Gynaecology, Ahmadu Bello University Teaching Hospital, Zaria, Nigeria

Date of Submission	26-May-2020
Date of Decision	27-Jul-2020
Date of Acceptance	08-Aug-2020
Date of Web Publication	31-Dec-2020

Correspondence Address:
Dr. Murtala Abubakar
Department of Pathology, Ahmadu Bello University Teaching Hospital, Zaria
Nigeria

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/atp.atp_29_20

Abstract

Background: Endometrial biopsy is a reliable means of evaluating the physiologic and pathologic states of the endometrium. The aim of this study was to examine the histopathological pattern of endometrial biopsies as seen in Zaria, Nigeria and to compare the results with similar studies elsewhere. Method: A retrospective analysis was carried out based on the records of the Department of Pathology, Ahmadu Bello University Teaching Hospital, Shika, Zaria between January 2014 and December 2018. The neoplastic lesions were classified according to 2014 World Health Organization classification for endometrial neoplasms. Results: A total of 1518 endometrial tissue samples were analysed. Pregnancy related conditions (41.76%) and functional endometrial changes (27.24%) were the most common morphologic diagnostic category seen. Endometrial hyperplasia (7.13%) was the most common pathologic lesion diagnosed. Of the malignant endometrial tumours, endometrial carcinoma is the most common with 29 cases (58% of all malignant endometrial lesions). One Hundred and Nine cases (7.2%) of the total cases reviewed were found to be inadequate for histopathologic evaluation. Conclusion: Retained product of conception was the most common morphologic entity seen in endometrial biopsy in Zaria. The proportion of endometrial samples that are inadequate for histopathologic evaluation is high. There is a need, therefore, for retraining of staff and standardization of biopsy procedures.

Keywords: Biopsies, endometrium, morphology, Zaria

How to cite this article:
Abubakar M, Kabir B, Liman AA, Shehu SM, Kankia BL, Yahya A, Ali ZA. Histopathologic analysis of endometrial biopsies in a Northern Nigerian Teaching Hospital. Ann Trop Pathol 2020;11:101-6

How to cite this URL:
Abubakar M, Kabir B, Liman AA, Shehu SM, Kankia BL, Yahya A, Ali ZA. Histopathologic analysis of endometrial biopsies in a Northern Nigerian Teaching Hospital. Ann Trop Pathol [serial online] 2020 [cited 2021 Jan 23];11:101-6. Available from: https://www.atpjournal.org/text.asp?2020/11/2/101/305689

Introduction

Endometrial biopsy provides a safe and cost-effective means of obtaining sample for the evaluation of a wide spectrum of endometrial conditions.^[1] The procedure has acceptable degree of sensitivity and specificity.^[1],[2]

Endometrial biopsy enables morphologic assessment of a variety of endometrial conditions that may include functional changes; pregnancy-related states; inflammatory conditions; as well as benign, premalignant, and malignant lesions. Clinically, the leading indications for endometrial biopsies are abnormal uterine bleeding, assessment of infertility, and postmenopausal bleeding.^[3] These represent a significant proportion of all the presenting complaints during hospital consultations among women during all phases of their reproductive lives.

The aim of this study, therefore, was to characterize the histopathologic patterns of endometrial biopsies received at the pathology department of a tertiary hospital.

Materials and Methods

This retrospective analysis was carried out based on the histopathological reports of endometrial biopsy samples received at the Department of Pathology, of a teaching hospital, in northern Nigeria, over a 5-year period (January 2014 to December 2018).

The cases of interest were extracted from the surgical daybooks of the years under review. Other relevant data were obtained from the pathology request forms filled by the attending physicians and copies of the pathology reports issued on all the cases. The corresponding glass slides and stored paraffin tissue blocks (in cases of missing slides) for all cases were retrieved from the archives. Reviews of the initial diagnoses in the daybook were made only if at least three of the four general pathologists in the research team agreed that the review was necessary. All slides examined were 3–5-μm-thick sections made from 10% formalin-preserved tissue, processed according to standard protocols and embedded in paraffin wax. The stains employed in all cases were of hematoxylin and eosin.

Cases for which both the glass slides and the tissue blocks are missing as well as those from hysterectomy specimens were excluded from the study.

The neoplastic lesions were reclassified according to the 2014 World Health Organization classification of endometrial neoplasms.

The resulting data were analyzed using Microsoft Excel statistical package.

Ethical clearance for the study was obtained from the Health Research Ethics Committee of the teaching hospital.

Results

Endometrial biopsies constituted 9.2% (1518 of 16,500) of all specimens received in the laboratory over the 5-year study period.

The age range of the patients was 16–79 years, with a mean of 32 ± 7 years. The age range 21–30 years was, overall, the most common among our patients and constituted 36.4% (552/1518). [Figure 1] shows the age distribution of the patients. The frequency distributions of the morphologic diagnostic categories are shown in [Table 1]. In [Table 2], patients were categorized into premenopausal (<40 years), peri-menopausal (40–50 years), and postmenopausal (>50 years).

Figure 1: Age distribution of the patients

Click here to view

Table 1: Frequency distribution of the various morphologic diagnostic categories of endometrial biopsy

Click here to view

Table 2: Distribution of Patients by menopausal status

Click here to view

Pregnancy-related conditions were the most common histopathological diagnostic category, accounting for 42.1% (639) of cases. Retained products of conception (70%), was the most common morphologic entity under this category. Others in the category included decidual reaction (24%) and Arias Stella reaction (4%). These occurred mainly in patients aged 16–42 years, with a mean of 27 years. [Figure 2] shows a typical case of retained products of conception.

Figure 2: Photomicrograph of retained products of conception showing viable and ghosts of chorionic villi, hemorrhage, decidua, fibrin deposition, and necrosis

Click here to view

The functional endometrial changes accounted for 27.2% (413) of cases. The mean age of these patients was 31 years. Under this category, secretory-phase endometrium accounted for 150 cases, whereas secretory endometrium and hormonal imbalances constituted 128 and 60 cases, respectively. Atrophic endometrial biopsy pattern was seen in 75 women, 61.3% of whom were postmenopausal. [Figure 3] shows a photomicrograph of a typical case of secretory endometrium.

Inflammatory conditions involving the endometrium were diagnosed in 6.5% (98) of our cases. Nonspecific chronic endometritis accounted for 91 cases (6.0%), whereas granulomatous endometritis constituted 7 (0.04%) cases. Two (28%) of the cases of granulomatous endometritis stained positive for acid-fast bacilli (AFB) and were designated as tuberculous endometritis, while the remaining five tested negative for AFB and fungal stain (Grocott methenamine silver and periodic acid–Schiff).

Figure 3: Photomicrograph showing secretory-type endometrial glands in an edematous, loose stroma

Click here to view

Endometrial hyperplasia was the diagnosis in 108 (7.1%) of our cases. Approximately 2/3^rd of these patients had benign hyperplasia, while atypical hyperplasia (endometrial intra-epithelial neoplasia) [Figure 4] was seen in 35 patients. More than half (53.7%) of the patients with endometrial hyperplasia were postmenopausal and their mean age was 49 years.

Figure 4: Photomicrograph of endometrial hyperplasia without atypia showing crowded endometrial glands lined by relatively bland pseudostratified columnar epithelial cells

Click here to view

Benign endometrial tumors constituted 101 (6.7%) of all the morphologic entities seen in our study. Majority of these (59 cases) were gestational trophoblastic diseases comprising 31 partial [example shown in [Figure 5]] and 28 complete hydatidiform moles [example shown in [Figure 6]. The mean age of patients with hydatidiform moles was 26 years. Nearly 95% of our patients with molar gestations were premenopausal, with the remaining 5% in their peri-menopausal years (40–50 years). Other benign endometrial tumors diagnosed included endometrial polyps (41 cases) and endometrial adenomyoma (1 case).

Figure 5: Photomicrograph of atyical endometrial hyperplasia showing crowded endometrial glands lined by cells exhibiting nuclear pleomorphism, hyperchromasia, and loss of polarity

Click here to view

Figure 6: Photomicrograph of complete hydatidiform mole showing diffuse cystic enlargement and edema of chorionic villi and circumferential trophoblastic hyperplasia

Click here to view

The malignant endometrial tumors seen in our patients included endometrial carcinoma (29 cases), malignant mixed Mullerian tumor (MMMT) (9 cases), choriocarcinoma (7 cases), and endometrial stromal sarcoma (5 cases). They altogether constituted 3.3% of the total endometrial morphologic entities seen. Sixty percent of the malignant lesions were seen in postmenopausal women. The mean age of the patients with endometrial carcinoma was 59.8 years. Endometrioid adenocarcinoma was the most common subtype (26 cases), while clear cell carcinoma and serous carcinoma had 2 and 1 case, respectively. [Figure 7] shows a photomicrograph of a typical case of endometrioid adenocarcinoma.

Figure 7: Photomicrograph of endometrioid adenocarcinoma showing fused, atypical endometrial glands with no intervening stroma

Click here to view

Choriocarcinoma was seen exclusively in the pre- and peri-menopausal age groups. The mean age of the patients with this condition was 29 years. The mean age of the patients with endometrial stromal sarcoma and MMMTs were 49 and 52 years, respectively. Postmenopausal bleeding (52%) was the most common presenting symptom in all malignant lesions of the endometrium among our patients.

Discussion

In this study, endometrial biopsies constituted 9.2% of all surgical specimens received in our laboratory over the 5-year period. Other local studies have recorded about the same figures.^[4],[5],[6]

It is, however, considerably lower than a much older finding in Ibadan,^[7] South-western Nigeria, in which endometrial biopsies constituted about 50% of all surgical biopsies received over a 5-year study period. Factors that may explain the disparities include the inclusion of endometrial specimens obtained via hysterectomy in the Ibadan study, and the refinement of the selection processes of patients for endometrial biopsy (resulting into fewer women subjected to endometrial biopsies as part of evaluation for infertility and/or abnormal uterine bleeding) that may have occurred over the 25-year period between the two studies.

The age range (16–79 years), as well as the mean age (32 years) of patients undergoing endometrial biopsy in our study, is comparable to those found among women undergoing the procedure in Enugu, North-eastern Nigeria,^[8] and in a study from Pakistan.^[9]

Pregnancy-related conditions were the most common histopathological diagnostic category, accounting for 42.1% of our cases. Local studies by Dauda,^[6] Vhriterhire et al.,^[10] and Asuzu and Olaofe^[11] corroborated this finding. Other workers,^{[4],[12],[13]} however, found functional endometrial conditions as the most common morphologic category. Variations in study methodology may be responsible for these differences. The latter studies have mostly excluded products of conception from their analyses.

Inflammatory conditions involving the endometrium were diagnosed in 6.5% (98) of our cases. Nonspecific chronic endometritis accounted for 91 cases (6.0%) of total endometrial biopsies, which agrees with the figures obtained from some local studies,^[4],[5],[7] while granulomatous endometritis constituted seven (0.5%) cases. The figure for granulomatous endometritis (which in our environment is considered to, most probably, result from disseminated tuberculosis) also agrees with other local data from Ibadan^[4] and Lagos.^[5]

Endometrial hyperplasia was seen in 7.1% of our study participants. This is close to the figures obtained in other centers in South-western Nigeria^[4],[5] (6.7% and 8.1%). Studies from Jos^[6] and Makurdi^[7] (all in North-central Nigeria) reported higher proportions of endometrial hyperplasia among women who underwent endometrial biopsy (17.5% and 16.4%, respectively). This intra-country variation in the proportion of endometrial hyperplasia among women who underwent endometrial biopsy has also been noticed in India, where different authors reported significantly varied figures, such as Kau et al . (12%), Chitra et al . (16.2%), Kunda and Anupam (22%), and Soleymani et al . (2.5%).^{[14],[15],[16],[17]} This probably reflects the influence of local/cultural practices related to excessive estrogen exposure on the etio-pathogenesis of endometrial hyperplasia.

Majority of cases of endometrial hyperplasia in our study were benign hyperplasia. This is similar to the findings of many other studies from within Nigeria, Africa, and other parts of the world.^{[4],[9],[11],[15],[18]}

In our study, hydatidiform mole constituted the majority of the benign endometrial tumors, and the ratio of complete to partial moles was almost 1:1. These findings are consistent with that of Abdullahi et al.^[4] and Vhriterhire et al.^[10] In addition, similar to our finding, a significant majority of patients with hydatidiform moles in the aforementioned studies were premenopausal.

Endometrial polyps constituted 2.7% of the entities seen among our patients. This is similar to the findings of Dike and Ojiyi,^[8] Abdullahi et al.,^[4] and Asuzu et al.,^[11] who reported a higher proportion of cases of endometrial polyp (9.9%) among women who had endometrial biopsy as part of evaluation for infertility. They postulated that it may be related to the ingestion of hormonal medications for the treatment of infertility.

Malignant endometrial tumors constituted 3.3% of the total endometrial morphologic entities seen in our patients, and a significant majority of these were seen in postmenopausal women with a mean age of 60 years. Similar patterns were reported from a prior study from our center^[15] from other parts of our country^{[4],[10],[12]} and from several other parts of Africa.^[18],[19]

Endometrial adenocarcinoma was the most common of the malignant tumors constituting 58% of cases. Abdul et al. ,^[20] Mordi and Nnatu,^[21] and several other workers reported similar findings. Endometrioid adenocarcinoma constituted 90% of the histologic subtypes of endometrial carcinoma in our study. This is in tandem with the findings of other studies.^[4],[8],[10]

MMMT constituted 18% of all malignant endometrial lesions seen in our study. This figure has doubled what was found in a previous study in our center^[20] and also significantly higher than what was reported from Ibadan, South-western Nigeria.^[4] This may partly be due to exclusion of hysterectomy specimens from our study, which might have revealed malignant lesions other than MMMT with consequent drop in the proportion of MMMTs. Furthermore, the previous study has covered a longer period of time, which could result in it being more reflective of the real pattern.

In our study, five cases of endometrial stromal sarcoma were found. This represented 0.34% of all endometrial biopsies. Previous studies from our center,^[20] and others from Ibadan^[4] and Maiduguri,^[13] all in Nigeria, documented similar low frequency of occurrence of this tumor.

Conclusion

In this study, products of conception were observed to be the most common endometrial biopsy specimen received in our laboratory. The malignant lesions of the endometrium are more common among postmenopausal women.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Silverberg SG. “The endometrium. Pathologic principles and pitfalls.” Arch Pathol Lab Med 2007;131:372-82.
2.	Saadia A, Mubarik A, Zubair A, Jamal S, Zafar A. Diagnostic accuracy of endometrial curettage in endometrial pathology. J Ayub Med Coll Abbottabad 2011;23:129-31.
3.	Clark TJ, Mann CH, Shah N, Khan KS, Song F, Gupta JK. Accuracy of outpatient endometrial biopsy in the diagnosis of endometrial hyperplasia. Acta Obstet Gynecol Scand 2001;80:784-93.
4.	Abdullahi YM, Ajani MA, Iyapo O, Aramide KO, Okolo CA, Akang EE. Morphological pattern of endometrial biopsies in southwestern Nigeria. Ann Ibd Pg Med 2016;14:103-9.
5.	Iyare FE. Morphological patterns of endometrial disorder's histopathological survey of cases. FMC Path Dissertation 1995;67:53-6.
6.	Dauda A. Morphological patterns of endometrial biopsies as seen in Jos, Nigeria. Nig J Med 2011;20:151-4.
7.	Adewole IF, Babarinsa IA, Akang EE, Thompson MO. The value of routine endometrial biopsy in gynaecological practice in Nigeria. West Afr J Med 1997;16:242-5.
8.	Dike EI, Ojiyi EC. Histopathological characteristics of endometrial biopsies in the Igbos of Nigeria. JMR 2017;3:187-90.
9.	Abid M, Hashmi AA, Malik B, Haroon S, Faridi N, Edhi MM, et al . Clinical pattern and spectrum of endometrial pathologies in patients with abnormal uterine bleeding in Pakistan: need to adopt a more conservative approach to treatment. BMC Womens Health 2014;14:132.
10.	Vhriterhire RA, Ngbea JA, Ojo BA, Ugbaje B, Maanongun MT. Endometrial lesions in Makurdi, Nigeria: A histological review of curetting specimens. Int J Tro Dis Health 2017;22:1-6.
11.	Asuzu IM, Olaofe OO. Histological Pattern of Endometrial Biopsies in Women with Abnormal Uterine Bleeding in a Hospital in North Central Nigeria. International Journal of Reproductive Medicine. 2018 ;2018:2765927.
12.	Ojo BA, Aboyeji PA, Buhari MO. Endometrial pathology in a teaching hospital in north central Nigeria: A histopathological survey. Nig J Health Biomed Sci 2005;4:107-11.
13.	Idrisa A, Emeka O, Abimiku BM. Endometrial sampling at a teaching hospital in northern Nigeria. West Afr J Med 2000;19:212-5.
14.	Chitra T, Manjani S, Madhumittha R, Harke AB, Saravanan E, Karthik S, Shabana B. Histopathology of endometrial curettings in perimenopausal women with abnormal uterine bleeding. J Evolution Med Dent Sci 2016;5:1285-90.
15.	Kaur P, Kaur A, Suri AK, Sidhu H. A two year histopathological study of endometrial biopsies in a teaching hospital in Northern India. Indian Journal of Pathology and Oncology 2016;3:508-19.
16.	Kunda J, Anupam S. Histopathological study of endometrium in abnormal uterine bleeding in reference to different age groups, parity and clinical symptomatology. Int J Clin and Biomed Res 2015;1:90-5.
17.	Soleymani E, Ziari K, Rahmani O, Dadpay M, Taheri-Dolatabadi M, Alizadeh K, et al . Histopathological findings of endometrial specimens in abnormal uterine bleeding. Arch Gynecol Obstet 2014;289:845-9.
18.	Kasule J. The pattern of gynaecological malignancy in Zimbabwe. East Afr Med J 1989;66:393-9.
19.	Nkyekyer K. Pattern of gynaecological cancers in Ghana. East Afr Med J 2000;77:534-8.
20.	Abdul MA, Oguntayo AO, Samaila MO, Mohammed U, Sarkinpawa ZS, Abdullahi A. A diary of endometrial malignancies in Zaria, Northern Nigeria. Sub-Saharan Afr J Med 2017;4:43-6.
21.	Mordi VP, Nnatu SN. Endometrial carcinoma in Nigerians. A pathologic study. Cancer 1986;57:1840-1.