DISCUSSION

Endometrial carcinoma is the most common malignant tumor of the female genital tract in developed countries. The differences noted in epidemiology, presentation, and biological behavior of endometrial carcinoma suggest that there are two fundamentally different pathogenic types of the disease: type I (estrogen related, endometrioid type) and type II (non-estrogen related, non-endometrioid type).[5] Type 1 endometrial carcinoma represents an estrogen-related tumor, which usually originates in the setting of the endometrial hyperplasia, has an endometrioid histology, and tends to be biologically indolent.[5]

Patients with hyperplasia have anovulatory menstrual cycles; some of them may be on exogenous estrogen. Although anovulatory cycles occur at menarche and in postmenopausal women, hyperplasia is not usually encountered in younger women, probably because bleeding in menarchial women is seldom evaluated by an endometrial biopsy. In the present study, the minimum age of presentation was a 12-year-old child with SH. She was younger than the one reported in literature.[6] The most common age group was 41 – 50 years in each type of hyperplasia and endometrial polyp. Schroder et al. also documented this to be the most common age group for SH.[7] Endometrial hyperplasia occurs most often in older women in the climacteric as they commonly have anovulatory cycles. This is the age group when ovarian function is in transition as evidenced by pronounced fluctuating estrogen production. During the reproductive years, hyperplasia is relatively uncommon. Women who are obese and have polycystic ovarian disease may have hyperplasia, presumably as a result of the peripheral conversion of androstenedione to estrogen in the adipose tissue. Postmenopausal women develop hyperplasia due to an unopposed estrogen hormone replacement therapy.[8]

The relationship between hyperplasia and carcinoma has been an actively debated subject. Several studies have demonstrated a close relationship of endometrial hyperplasia and carcinoma. Gusberg et al., in 1961, observed that the cumulative risk of developing endometrial carcinoma for a patient with AH for 9 to 10 years is significantly higher than for women without hyperplasia.[9] Chamlian and Taylor, in a long term study, found that 14 % adenomatous and atypical hyperplasias subsequently developed into carcinoma.[10] Another view suggests that endometrial hyperplasia and adenocarcinoma may represent separate expression of endometrial pathology, which may occur side by side, but need not necessarily follow each other.[11] McBride studied cystic hyperplasia for a period ranging up to 24 years and found that less than 0.4 % developed carcinoma.[12]

The fate of a hyperplastic lesion of the endometrium in the present study showed a varied pattern. Tabata et al. also found a high percentage of regression including cases with features of atypia. However, they found only one case that progressed to adenocarcinoma.[13] Studies done by Yokosuka et al. and Kurman et al, also showed a higher percentage of regression as compared to ours, this may be due to the reason that most of the patients in their follow up were being treated with progesterone.[14,15] Our study showed regression to a milder lesion in 9.55% of the cases and reversion to normal pattern in 10.81% of the cases. A lower percentage of case regression in our study could be related to an economical background, illiteracy, and lack of compliance to treatment, as this is a referral government hospital with patient representation predominantly from rural areas. In most of our cases the primary lesions persisted on follow up. The persistence of the lesion could perhaps be explained by high level of estrogenic stimulus with possible intervals of fluctuation. The lesion, which had progressed to a higher grade, would have resulted from sustained high levels of estrogen. In our study, the lesion persisted even after five years in case of SH and for 13 years in case of a benign endometrial polyp, a finding that could be due to fluctuating estrogen levels, not allowing the lesion to progress or regress. On the other hand a sustained higher level could have resulted in progression to a higher grade.

Kurman et al. 1985, found that an average period of 4.1 years was needed for atypical hyperplasia to develop into carcinoma.[15] About 17 to 25% of women with atypical hyperplasia, diagnosed in curetting, could have a well-differentiated carcinoma in the uterus if a hysterectomy was performed within one month of curettage.[15] One case of a complex hyperplasia with atypia in our study progressed rapidly into carcinoma within a short span of one month. It therefore, needs to be emphasized that prevalence of endometrial carcinoma in patients harboring an atypical hyperplasia is high. Hence, while planning management strategies for women having atypical hyperplasia, clinicians and patients should take into account the considerable rate of the concurrent carcinoma.[16]

The numbers of prospective studies have been small and the true natural history of the disorder could not be observed due to the administration of various forms of therapies. Retrospective studies may on the other hand give a biased and unduly gloomy view of the risk of adenocarcinoma.

Endometrial hyperplasia of whatever type must be considered as a warning sign that an endometrium is noncycling and therefore susceptible to neoplastic events. The mere presence of hyperplasia is not a basis for hysterectomy. However, in general, the more severe the hyperplasia the more likely it is to be followed by invasive carcinoma. Timely treatment can help to provide an environment for the lesion to regress and avoid radical surgeries. Several investigators have found beneficial effects from treating hyperplasia and carcinoma with progesterone. Kistner, treated patients with atypical hyperplasia and patients with carcinoma in situ with progesterone and found that the lesions were reversible and none advanced.[17]

All forms of hyperplasia in the endometrium appear to be the result of excessive stimulation by estrogen, although such an association is somewhat clear for cystic hyperplasia than it is for atypical hyperplasia.[18] It would be expected from this that two foci of hyperplasia may often be seen in the same uterus. It is not surprising, therefore, to find an endometrium that shows widespread changes of cystic hyperplasia with superadded areas of atypical hyperplasia.[4,18] It is debatable whether the latter develops from the former. It is quite possible to suggest that endometrial glands, which have been constantly stimulated by estrogens for sometime may alter their response and eventually proliferate in an atypical rather than a cystic fashion.[4,18]

However, an alternative explanation could be that the two types of hyperplasia develop concurrently; variable effect being the result of the difference in response from one area of the endometrium to another.[18] In this study, there were 54 cases that showed the co-existence of various subtypes of hyperplasia / adenocarcinoma endometrium. This may be because they either express a common basic abnormality of growth or they form a continuous spectrum.[4] The combined lesions predominantly seen in our study were complex and atypical hyperplasias, suggesting a close relationship between them. According to Fox and Buckley, all types of hyperplasia can coexist with each other except adenomatous hyperplasia (now categorized as CH), which is usually seen in isolation.[4] However, in contrast we found a coexistence of CH with SH and atypical hyperplasias.

Endometrial polyps are a common cause of abnormal uterine bleeding. Rarely, is hyperplasia, either a complex or atypical type, identified within a polyp, in a biopsy or polypectomy specimen. Currently, it is not known whether the hyperplasia is likely to be confined to the polyp or also involves a non-polypoid endometrium.[19]

In our study, we had very few cases to assess this possibility, though we found progression to SH in one case, and a mixed pattern with SH and CH in 12 cases. This could draw our attention to the chances that all were related to each other. In a study done by Antunes et al, a low prevalence of premalignant and malignant lesions in endometrial polyps were found.[20] Their study also showed that older women and those with postmenopausal bleeding had a greater prevalence of malignancy and in these cases hysteroscopic resection of the endometrial polyps became mandatory.[20] Kelly et al. documented in their study that the risk of endometrial hyperplasia in a polyp, concurrently involving a non-polypoid endometrium was significant.[19]

The nature of the disease was such that patient would require several consultations. Although this was a retrospective study for a period of 16 years, there were limited patients who came for subsequent follow up at our center. Hence, there were chances that the traceability of patients had been lost either because they had lost their out-patient identity card and had re-registered at our center with a new identity number or had gone to a different centre for subsequent visits.