Histopathological findings in hysterectomy specimens:
A retrospective study

......................................................................................................................................................................

Suad Mohamed O. Zaid (1)
Mazen Abood Ben Thabet (2)

(1) Department of Morphololgy Sciences, Faculty of Medicine, University of Aden
(2) Department of Paraclinic, Faculty of Medicine, University of Aden

Correspondence:
Suad Mohamed O. Zaid
Department of Morphololgy Sciences,
Faculty of Medicine,
University of Aden
Email: salamomer54@yahoo.com

Uterus, a vital reproductive organ is subjected to many benign and malignant diseases. Many treatment options are available including medical and conservative surgical but hysterectomy still remains the most common gynaecological procedure performed worldwide (1).

The procedure is not well embraced in developing countries, thus, the clinical indication for the procedure should be justifiable, for age and parity of the women (2).

In response to the consistent demand for this procedure, hysterectomy has been identified as a key health care indicator in recent reports, to measure and compare hospital performance (3).

It is the definitive cure for many of its indications which include dysfunctional uterine bleeding, fibroids, utero-vaginal prolapse, endometriosis and adenomyosis, pelvic inflammatory disease, pelvic pain, gynaecological cancers and obstetric complications. Ultimate diagnosis is only on histology, so every hysterectomy specimen should be subjected to histopathological examination (4).

In Yemen, histopathological examination of hysterectomy specimens carries diagnostic and therapeutic significance. Prevalence of uterine and adnexal pathologies varies from nation to nation and from region to region (5)

The present study is aimed at detailed histopathological evaluation of all lesions of hysterectomy specimens. It provides an intact uterus and consequent control over tissue sampling and hence enabling determination of origin of particular lesion and to compare the findings with other researchers.

Our study was a retrospective descriptive work analysis of 2,544 patients with hysterectomy, over a period of 6 years from January 2006 to December 2012. The information was gathered regarding age, and histological diagnosis and was analyzed by Excel program and tables performed according to the objectives of the study and compared to literature review.

A total of 2,544 hysterectomy specimens between January 2006 to December 2012 were analyzed. The age range of the patients was 16 to 80 years, with a mean of 47.6 years.

Of these 2,544 cases, most of the cases were in the 41-50 years age group 1431(56.3%), which is the most common age group for contracting various diseases as shown in Table 1.

Table 1: Distribution of patients according to age groups

Table 2 revealed that out of the total hysterectomy specimens 1,481(58.3%) were Endometrial hyperplasia, atrophied endometrium 396 (15.6%) Tumor was present in specimens out of which 10 were invasive complete hydatidiform mole, 45 were endometrial adenocarcinomas , Malignant mixed mullerian tumour (MMMT) 6 (0.2%) cases and one case of choriocarcinoma.

Table 2: Histopathological findings in Endometrium hysterectomy specimens

Most common histopathological abnormality in myometrium was adenomyosis followed by Leiomyoma. Adenomyosis in 793 (31.2%), followed by Isolated leiomyoma was seen in myometrium of 697 (27.4%) hysterectomies, where in 163 (6.3%) myometriums, both were present together. Tumor was present in specimens out of which 31 was invasive by malignant endometrial carcinoma as shown in Table 3.

Table 3: Histopathological findings in myometrium

In Table 4 cervix from 2,377 (53.6%) specimens showed chronic cervicitis. Cervical intraepithelial neoplasia (CIN) I, CIN II, CIN III with chronic cervicitis (0.8%, 0.4, 0.3%) and flat condyloma (0.1%,0.6%), squamous cell carcinoma were seen in 22 specimens (0.9%) and Adenocarcinoma were 17 cases. Uterovaginal prolapse were 132 cases (5.2%).

Table 4: Histopathology of cervix

Unremarkable Histopathology of the cervix were 655 cases (25.7%).

2,087 ovarian specimens were retrieved from the computerized database of the pathology department, from January 2006 to December 2013.

There were 1,386 (54.4%) non-neoplastic functional cysts.

The neoplastic were benign serous cystadenoma (2.5 %) and benign varian fibroma (0.9%), mucous cystadenoma (0.5%) and Mature cystic teratoma (Dermoid cyst) 0.5%.

The malignant were 21 cases of serous cystadeocarcinoma , 9 cases of mucinous cystadeocarcinoma , 2 cases undifferentiated carcinoma and 9 cases of Metastatic carcinoma, as appears in Table 5.

Table 5: Histopathology of ovaries

Hysterectomy is the commonest gynecological operation and the rate of hysterectomy varies according to geographic distribution, patient and physician related factors (1).

Hysterectomy is second only to cesarean section as the most frequently performed major operation in the United States. Approximately 600,000 hysterectomies are performed annually in the USA, and more than one third of US women have had a hysterectomy by the age of 60 (6).

In Pakistan, the rate of hysterectomy is quite high because it is the only surgical option available if the patient is not responding to medical treatment(7).

Many women in Africa and Nigeria in particular are reluctant to undergo this procedure because of the socio-cultural attachment to procreation and taboos associated with lack of menstruation(2).

Few studies have been performed describing the pathologic findings in hysterectomy specimen and examining the relationship between the preoperative clinical indication and pathologic diagnosis (8).

In the present study, the mean age of patients was 47.6 years and age range from 16 to 80 years which was nearly similar to findings by others (7,9,10).

The peak age for the procedure in our study was the fourth decade (41–50 years) as has been observed in other studies (9).

In the current work we found endometrial hyperplasia was the commonest histopathological finding with 58.3%.

Lee (11) reported that endometrial hyperplasia was confirmed in 95%, a somewhat higher figure than we found and less results 16% were found in Nepal by Ranabhat et al (5).

The exact pathogenesis of endometrial polyps is not fully elucidated, but they are thought to originate as a localized hyperplasia of the basalis, perhaps secondary to hormonal influences (13).

In our study the association between endometrial hyperplasia and hyperplastic endometrial polyp were 139 (5.5%) cases.

This figure approaches that seen by Kelly et al (14) with (3.1%) in all cases of endometrial hyperplasia in his study period. Other studies have found that incidence of endometrial polyps in endometrial hyperplasia range between 11 and 29% (15).

In the present study we found atrophic changes in 396 cases (15.6%), nearly approximate to that seen by Ranabhat et al (5) with 13% and that seen by Pity et al (16) from Iraq with 10.4 %.

A higher figure was seen in study by Thamilselvi et al (17) with 26%.

This may justify the sample size in our study.

Other authors were in discordance with our study, Gousia et al (18) reported 5.44%, and results reported by Sarawathi et al (19) with 2.44%.

In the present study inadequate secretory transformation were 216 (8.5%) cases, which was similar to other findings (20).

It is higher than that seen by Zeeba et al (21) with 1.8% and higher than our result was reported by Sarfraz et al (22) with 24%.

Chronic endometritis is commonly seen in the reproductive age due to either retained products of conception, pelvic inflammatory diseases or other pregnancy related conditions.

In our study 21 (0.8%) cases showed chronic endometritis of all hysterectomy samples, which approximate with the finding of Sajjad et al (23) which was 1% and completely lower than that seen by Ranabhat et al (5) which was 9.5%.

We found in our study endometrial adenocarcinoma were in 45 patients (1.7%). This finding is similar to that found by others (5,16), but it was completely lower than those reported by Patel (24) from Australia 10.5%, and by Gebauer et al (25) from Germany with 16%.

These regional variations have been reported with many speculative factors such as ethnic origin, blood group, age, parity, diet and nutrition, contraception, socio-economic status, immunologic factors and genetic constitution (27).

In our study we found 222 (8.7%) cases of GTD of the total hysterectomies samples, only 10 cases were invasive gestational trophoblastic disease at the time of pathological diagnosis and one case of choriocarcinoma.

In the Kingdom of Saudi Arabia (KSA) fifty-nine cases of hydatidiform mole, 36 complete hydatidiform mole (CHM) and 23 partial hydatidiform mole (PHM) and 2 cases of choriocarcinoma were observed, out of 64,762 pregnancies registered at Security Forces Hospital, Riyadh, KSA, during an 11 year period (27).

In a study in Nepal, there were 17 (37.8%) cases of hydatidiform mole, 6 (13.3%) of invasive mole and 22 (48.8%) patients of choriocarcinoma (28).

A malignant mixed Mullerian tumour (MMMT) of the uterine corpus is an extremely rare and aggressive malignancy, comprising only 1–2% of uterine neoplasms (29).

In our study there were 6 (0.2%) cases of MMMT. In the study of Rajshekar SK only four cases of MMMT were diagnosed representing 20% of his sample (30) and this variation in the frequencies may support our justification related to sample size and study design.

In the current study adenomyosis was the commonest lesion of the myometrial pathology and represented 31.1% followed by leiomyoma 27.4%. Adenomyosis appears also to be the commonest pathology and similar to our findings reported by others (5,22,31).

The present study revealed that leiomyoma was also the commonest pathology and it was 27.4%. Reported frequencies vary in different countries and it was 26% in KSA (32), and 36% and in Kurdistan/Iraq (16), in Nigeria 48%(33) and 17% in India (34) and only 8% in Sweden (35).

Some of the hysterectomy specimens show more than one lesion in the body of uterus, of which coexistence of adenomyosis and leiomyoma are the most common (34).

In the present study there was 6.4% showing coexistence of adenomyosis and leiomyoma. In other study increasing to 56% when adenomyosis with concomitant leiomyoma are included (31) and it was 19% reported by Sarfraz et al (21) and 5.6% reported by Qamar et al (7).

Leiomyoma was the commonest lesion of uterine corpus followed by adenomyosis.
This was similar to findings of other studies (16,32,33,36,37).

Geographical and racial influences are thus apparent on the prevalence of uterine leiomyoma and the prevalence of risk factors in terms of quantities and type. Early menarche, delayed menopause, decreased parity, obesity and lack of exercise are some of the risk factors of leiomyoma (5).

Among the cervix uteri, chronic cervicitis was the main pathological finding in the present study and accounted for 53.6%. This figure is nearly similar to that reported by Jamal et al (36) which was 41.5% and to that reported by Qamar et al (7) which was 31%.

A higher figure of chronic cervicitis seen in Nepal women by Jha et al was 96.4% (37); the variation may be related to different reproductive health procedures. Iin Yemen almost all males are circumcised which minimizes vaginal infection.

In our study 37 (1.5%) cases showed dysplasia of various degrees with chronic cervicitis and 20 (0.7%) cases showed cervical condyloma with dysplasia.

A premalignant lesion, Cervical intraepithelial neoplasia (CIN) was seen in 3.0% in a study by Thamilselvi et al (17) and 0.8 % reported by Ranabhat et al (5).

The low incidence of CIN in our study may related to the reproductive life style, where the women are restricted to single sexual partner , while the CIN is more common with sexually transmitted disease of HPV, which is more frequent in multiple sexual partner women .

The diagnosis of uterovaginal prolapse was based on clinical as well as pathological findings (38).

In our study Hysterectomies done for utero-vaginal prolapse were found in 132 (5.2%). This finding was higher than that reported by Pity et al (16) which was 2(0.5%), while less than the findings reported by Butt et al (39) with (11%) and less than 17% reported by Adelusola et al (33).

The present study revealed only 0.9% of all the samples of hysterectomy showing invasive squamous cell carcinoma at the pathological study.

This finding was nearly similar to that reported by Ranabhat et al (5), Gousia RR et al (18) and Bani et al (40) which were 0.6%, 0.3% and 0.6% respectively.

This low incidence may be related to reproductive health in Arab and Muslim countries where most of the women are restricted to one sexual partner and a Muslim habit for washing and vaginal douches after sexual intercourse and a high incidence of HPV infection in European countries play an important role in cervical dysplasia and carcinoma.

In the present study adenocarcinoma were 17 (0.7%) cases.

The most common lesions encountered in the ovary include functional or benign cysts and tumors and benign ovarian neoplasms occur at any age whereas malignant ovarian neoplasms are more common in the elderly (43,10).

Ovarian tumors are one of the major causes of gynaecological problems in females and present with marked variation in their histological types. Relative frequency of these lesions is different for Western and Asian countries (10).

We found in our current study, the most common pathological finding of the ovaries in all hysterectomy samples were benign (functional) cysts and were 54.4%.

Our finding was nearly similar to that reported by Mansour (44) in KSA where the benign non neoplastic ovarian cysts comprise 47.5%, while the data from South East Asia shows that 90.5% of ovarian cysts were benign (45), less results were reported by Gupta et al (46)with 2.77% and 20% by Ranabhat et al (5).

Surface epithelial tumours were the major histological type of ovarian tumours followed by germ cell tumours as the commonest ovarian cyst seen in most of the literature (8).

In our study, the most common surface epithelial tumors was benign serous cyst adenoma 2.5% followed by mucinous cystadenoma 0.5% , which approximate the finding seen by Jha et al (37) with 4.5% for benign serous cystadenoma, 3.1% for mucinous cystadenoma and 25.7% of benign surface epithelial tumors were serous cyst adenoma and 6.7 % were mucinous cyst adenoma reported by Pity et al (16) in their study, which was lower than that seen by Abdullah et al (38) where serous cystadenoma represented 44.6% and mucinous cystadenoma 13.6%. The low figure in our study may be related to the study sample, where we are selected only hysterectomy samples and excludes all cases with simple ovarian cystectomies.

In our study malignant serous cystadenocarcinoma were the most common malignant ovarian neoplasm and represented 0.8% of the cases followed by mucinous cystadeocarcinoma 0.4% and this figure approximates the data published by Jha et al (37) where 3.4% of his cases are malignant serous cystadenoma and 0.8% were malignant mucinous cystadenoma.

The higher result with data published by the others, and it’s 33.3% for malignant serous cystadenocarcinoma and 15.4% for malignant mucinous cystadenocarcinoma seen by Abdullah et al (38) and in Nepal malignant serous cystadenocarcinoma account for 21.1% and 22.2% of malignant mucinous cystadenocarcinoma found by Jha et al (37) and the low figure in our study related to the type of study sample.

Approximately 95.0% of ovarian germ cell tumors are mature cystic teratomas in the western world (47).

In this study mature cystic teratoma (Dermoid cysts) account for 12 (0.5 %) of all ovarian tumors. A study in Pakistan (48) reported a high figure of 38%.

A mature cystic teratoma is a benign neoplastic ovarian lesion that occurs during reproductive life and is more common in young females during active reproductive life and usually treated by simple cystectomy and this may justify the low incidence in our study where the hysterectomy is the sample study and not ovarian cystectomy.

Other ovarian tumours are rare in our study and it was 0.6% for ovarian fibroma which is similar to that reported by Jha et al (37) with 0.9%.

Granulose cell tumor was 6(0.2%) in our study and it is similar to other findings (37,49).

In the present study ovarian endometriosis accounted for 16 (0.6%), which was similar to that seen by Gousia et al (17) with (0.61%). Also, our finding was less than that observed by Randabhat et al (5) which was 8.9% and less than that seen by Ahsan et al (30) with 13%.

Ovarian endometriosis is a benign condition usually treated by simple ovariectomies, which justify the low figure in our study which is based on hysterectomy samples.

In the present study metastatic carcinoma to ovaries (secondary) accounted for 9(0.4%) which was lower than that seen by Abdullah et al (38) with 13(15.5%) and 4(2.4%) reported by Jha et al (10).

The low figure in our study does not reflect the low incidence in our patients but may be related to big sample size in our study (2,450 cases) as well as the type of sample study and most cases of metastatic carcinoma to ovaries are with advanced stages of either breast or GIT cancer, where there is no indication orfor hysterectomies.

Hysterectomy still remains the widely used treatment modality even in developed countries. The ultimate diagnosis is only on histology, so every hysterectomy specimen should be subjected to histopathological examination. Histopathological analysis correlates well with the pre-operative clinical diagnosis for hysterectomy.

Most of the pathologies are still benign; malignancies are also detected on hysterectomy specimens, but very rarely. A yearly audit should be conducted in every institute to collect data and to analyze the pattern of indications and types of histopathological lesions and pattern of diseases.

1. Vessey MP, Villard-Mackintosh L, McPherson K, CoulterA, Yeats D. (1992). The epidemiology of hysterectomy: findings in a large cohort study. Br J Obstet Gynecol.; 99: 402-7.
2. Ezem BU and Otubu JA. (1981) . Hysterectomy in the Hausa/Fulani population in Nigeria. Int J Gynecol Obstet;19:145-9.
3. Toma A, Hopman WM, Gorwill RH. ( 2004 ). Hysterectomy at a Canadian tertiary care facility: results of a one year retrospective review. BMC Women Health.:4(1):10
4. Nausheen F, Iqbal J, Bhatti FA, Khan AT, Sheikh S. Hysterectomy. (2004). The patient’s perspective. Annal Gynecol; 10: 339-41
5. Ranabhat SK, Shrestha R, Tiwari M, Sinha DP, Subedee LR. (2010) . A retrospective histopathological study of hysterectomy with or without salpingoophorectomy specimens. JCMC Journal of Chitwan Medical College; Vol. 1 (1): 24-29
6. Schaffer JI, Word A. (2002 ). Hysterectomy-still a useful operation. The New England Journal of Medicine; 347 (17): 1360-1362.
7. Qamar-ur-Nisa, Habibullah, Shaikh TA, Hemlata, Memon F, Memon Z. (2011). Hysterectomies; an audit at Tertiary Care Hospital. Professional Med J Mar;18(1):45-
8. Ahmed M, Malik TM, Afzal S, Mubarik A. (2004). Clinicopathological study of 762 ovarian neoplasms at Army Medical College Rawalpindi. Pak J Pathol;15(4):147-152.
9. Carlson KJ. Outcomes of hysterectomy. (1997). Clin Obstet Gymaecol;40:939-46.
10. Jha R, Karki S. (2008). Histological pattern of ovarian tumors and their age distribution. Nepal Med Coll J; 10: 81-85.
11. Lee NC. (1984). Confirmation of pre-operative diagnosis for hysterectomies. American Journal of Obstetrics & Gynaecology; 150 (3): 283-7.
12. Caspi E, Perpinial S, Reif A. (1977). Incidence of malignancy in Jewish women with postmenopausal bleeding. Israel Journal of Medical Sciences, 13(3):299–304.
13. McGurgan P, Taylor LJ, Duffy SR, O’Donovan PJ. (2006) Are endometrial polyps from pre-menopausal women similar to post-menopausal women? An immunohistochemical comparison of endometrial polyps from pre- and post-menopausal women. Maturitas ;54:277–84
14. Kelly P, Dobbs S, McCluggage W. (2007). Endometrial hyperplasia involving endometrial polyps: report of a series and discussion of the significance in an endometrial biopsy specimen. BJOG;114:944–950.
15. Bakour S, Khan S, Gupta JK. (2000) . The risk of premalignant and malignant pathology in endometrial polyps. Acta Obstet Gynecol Scand;79:317–20.
16. Pity IS, Jalal A. Jalal JA, Hassawi BA . (2011) . Hysterectomy: A Clinicopathologic Study . Tikrit Medical J 16 journal; 17(2): 7-16.
17. Thamilselvi R, Sinha P, Subramanium. (2011). Correlation between clinicopathological and ultrasonographical findings in hysterectomy. Journal of Clinical Diagnostic Research;vol.5(4):737-740.
18. Gousia RR, Yudhvir G, Subash B . (2013). Patterns of Lesions in Hysterectomy Specimens: A Prospective Study. JK Science. Vol. 15 No. 2.
19. Saraswathi D, Thank Johnson, Shalinee R, Rathi Rajkumar, Jaya Vijayaraghavan, Vinod Kumar Panicker. (2011). Study of Endometrial Pathology in Abnormal Uterine Bleeding. J Obstet Gynaecol India;61(4):426 - 430
20. Talat Mirza1, Saadia Akram, Aamir Mirza1, Sadiya Aziz, Tariq Mirzaand Tazeen Mustansar . (2012 ). Histopathological Pattern of Abnormal Uterine Bleeding. Journal of Basic & Applied Sciences, Vol. 8, No. 1. Pg 114-117.
21. Zeeba S. Jairajpuri, S. Rana and S. Jetley. (2013). Atypical uterine bleeding-Histopathological audit of endometrium. A study of 638 cases. Al Ameen J Med Sc i; 6 (1) :21-28.
22. Sarfraz R, Ahmed MM, Tariq M. Tahir And M. Sarfraz Ahmed. (2011). Benign Lesions In Abdominal Hysterectomies In Women Presenting With Menorrhagia. Biomedica Vol. 27 .
23. Sajjad M, Iltaf S, Qayyum S. (2011). Pathological findings in Hysterectomy specimens of patients presenting with menorrhagia in different age groups. Ann Pak Inst Med Sci;7:160-2.
24. Patel SRP. (1967) . Postmenopausal bleeding. MJA. 27:1080-82.
25. Gebauer G, Hafner A, Siebzehnrubl E,Lang N. (2001). Role of hysteroscopy in detection and extraction of endometrial polyps: results of a prospective study. Am J Obstet Gynaecol; 184:59-63
26. Mochizuki M, Maruo T, Matsuo H, Samoto T, Ishihara N. (1998). Biology of human trophoblast. Int J Gynecol Obstet; 60: 21-28.
27. Tariq Y. Khashoggi, MCPS, ABOG .(2003). Prevalence of gestational trophoblastic Disease. A single institution experience. Saudi Med J; Vol. 24 (12): 1329-1333
28. Pariyar J. (2009). Gestational trophoblastic disease in Nepalese women managed in B. P. Koirala Memorial Cancer Hospit. J Clin Oncol (Meeting Abstracts) vol. 27 no. 15S e16570
29. El-Nashar SA and Mariani A . (2011). Uterine carcinosarcoma Clin Obstet Gynecol 54(2) 292–304.
30. Rajshekar SK, Guruprasad B, Shakunthala PN, Praveen Rathod, Uma Devi and UD Bafna. (2013). Malignant mixed Mullerian tumour of the uterus. Ecancermedicalscience.; 7: 302.
31. Ahsan S, Naeem S, Ahsan A. (2001). A case note analysis of hysterectomies. performed for non-neoplastic indication. Liaquat National Hospital, Karachi. J Pak Med Ass; 51(10):346-9.
32. Sobande AA, Eskander M, Archibong EI, Damole IO. (2005). Elective hysterectomy: a clinicopathological review from Abha Catchment Area of Saudi Arabia. West Afr J Med24 (1): 31-35
33. Adelusola KA, Ogunniyi SO. (2001). Hysterectomies in Nigerians: histopathological analysis of cases seen in Ile-Ife . Niger Post grad Med J. ; 8(1): 37-40.
34. Talukder SI, Haque MA, Huq MH, Alam MO, Roushan A, Noor Z . (2007). Histopathological analysis of hysterectomy specimens, Mymensingh J; 16: 18- 24.
35. Borgfeldt C, Andolf E. ( 2000).Transvaginal ultrasonographic findings in the uterus and the endometrium: Low prevalence of leiomyoma in a random sample of women age 25-40. Acta Obstet Gynecol Scand;79:202-7.
36. Jamal S and Baqai S. (2001). A clinicohistopathological analysis of 260 Hysterectomies Pakistan. J Pathol.; 12(2): 11-4.
37. Jha R, Pant AD, Jha A, Adhikari RC, Syami G. (2006).The histopathological analysis of hysterectomy specimens. J Nepal Med Assoc; 45(163):283-290
38. Abdullah LS, Bondagji NS. (2011). Histopathological Pattern of Endometrial Sampling Performed for Abnormal Uterine Bleeding. Bahrain Medical Bulletin,;33(4):1-6.
39. Butt JL, Jeffery ST, Van der Spuy ZM. (2012). An audit of indications and complications associated with elective hysterectomy at a public service hospital in South Africa. Int J Gynecol Obstet.;116 (2):112-116.
40. Bani-Irshaid and Al-Sumadi A. (2011). Histological findings in women with postmenopausal bleeding: Jordanian figures.EMHJ . Vol. 17 No. 7.
41. Garud M, Lulla M,Patel K,Saraiya U, Mehataji S. (1981). Cervical borderline lesions. Ind Obstet Gynecol Ind:31:135.
42. Sanyal B, Pant GC, Sahni K, Koteshwar Rao K, Gupta S (1992). Non –epidermoid tumours of -uterine cervix. J . Obstete Gynecol Ind;32-591
43. Kumar, Vinay; Abbas, Abul K.; Fausto, Nelson; & Mitchell, Richard N. (2007) . Robbins Basic Pathology (8th ed.). Saunders Elsevier. pp. 718–721.
44. Mansour I. (2002 ). Ovarian diseases at King Abdul-Aziz University Hospital. Saudi Med J; 23: 1551-1552
45. Kayastha S. (2009). Study of ovarian tumors in Nepal Medical College Teaching Hospital. NepalMed Coll J;11:200-2.
46. Gupta G, Kotasthane D, Kotasthane V. (2009). Hysterectomy: A Clinico-Pathological Correlation Of 500 Cases. The Internet Journal of Gynecology and Obstetrics. Volume 14 Number 1.
47. Tavassoli FA, Devilee P (2003) .WHO Classification of Tumors. Pathology and Genetics, Tumors of Breast and Female Genital Organs. IARC Press; 35-36
48. Saeed M. Khawaja K. Rizwan 1, et al. (1991). Clinicopathological analysis of ovarian tumours. J. Pak. Med. Assoc.; 41: 161-64.
49. Schumer ST, Cannistra SA. (2003). Granulosa cell tumor of the ovary. J Clin Oncol; 21: 1180-1189.