The significance of intrauterine lesions detected by ultrasound in asymptomatic postmenopausal patients


Dr A. Lev-Sagie, Department of Obstetrics and Gynecology, Hadassah University Hospital—Mt. Scopus, P.O.B. 24035, Jerusalem, Israel 91240.


A retrospective study on 82 women with an incidental sonographic finding suspected to be intrauterine polyps was undertaken to assess the histopathologic characteristics of such polyps utilising operative hysteroscopy. Endometrial polyps were found in 68 patients, submucousal myomas in 7, atrophic endometrium in 6 and thickened proliferative endometrium was found in 1 patient. Simple hyperplasia was found in one polyp but neither endometrial carcinoma nor complex hyperplasia was found. The total complication rate was 3.6%. It appears that the risk of endometrial carcinoma in postmenopausal women with asymptomatic endometrial polyps is low, although a larger series is required to confirm this finding.


Intrauterine endometrial polyps are relatively common findings among postmenopausal patients.1 The main concern that such lesions increase in this age group is the possibility of malignancy, which must be determined by histologic examination.2,3 Endometrial polyps can present with postmenopausal bleeding or can be an incidental finding on ultrasound examination done for other indications. Because vaginal bleeding is the most common presenting symptom of endometrial carcinoma and occurs in approximately 90% of affected patients, it always requires pathologic evaluation,4 whether an intrauterine polyp is present or not. But whether evaluation of asymptomatic polyps is mandatory is unclear. Because of the introduction of ultrasonographic examination in routine gynaecological practice, an increasing number of asymptomatic intrauterine lesions are identified in postmenopausal patients. Some of these findings are considered suspicious and require histopathologic evaluation. The latter include thickened endometrium,5,6 endometrial heterogenicity, focal mass and irregular endometrial margins.6 Among the asymptomatic intrauterine findings are endometrial polyps, whose pre-malignant potential remains unknown, although they are believed to be a risk factor for endometrial cancer.7 Further, the actual incidence of malignancy detected in endometrial polyps ranges between 0.5%8 and 4.8%3 in different series.

Endometrial biopsy (Pipelle) is usually sufficient for histologic diagnosis when there is no focal lesion in the uterine cavity,9 but such biopsy, or even curettage, are poor modalities for sampling focal lesions such as endometrial polyps or myomas.10,11 In cases of defined lesions, operative hysteroscopy and resection of the lesion identified are recommended.10,12 The entire lesion is then available for histologic examination, in addition to the procedure's therapeutic effect in cases of symptomatic bleeding.13 When asymptomatic polyps are identified in postmenopausal women, a similar approach is reasonable to outrule a malignant condition.

In our department, intrauterine polyps are removed by hysteroscopy whether postmenopausal bleeding is present or not. The purpose of this study was to determine the pathologic significance of asymptomatic intrauterine polyps in postmenopausal patients utilising operative hysteroscopy.


A retrospective review of the medical notes on patients who had undergone investigation for asymptomatic uterine lesions during 1998–2002 was undertaken. Ultrasound examination was performed for indications such as abdominal pain, recurrent urinary tract infection, tamoxifen use and for pelvic evaluation in obese patients in whom manual examination was not informative. Data collected included age, parity, postmenopausal years, use of hormonal therapy, tamoxifen treatment in cases of breast cancer, other medical conditions associated with endometrial carcinoma (hypertension and obesity) and the findings of diagnostic hysteroscopy.

Before the surgical procedure, every patient underwent general physical examination, pelvic examination and transvaginal sonogram.

Operative hysteroscopy was performed under general anaesthesia. The cervix was dilated to a Hegar number 10 dilator. Hysteroscope placement and the operative procedure were monitored by video. Operative hysteroscopy was performed with a continuous-flow 10 mm resectoscope (Karl Storz, Germany), with manitol 5% solution as the irrigating medium, up to a pressure of 100 mmHg. Polypectomy and myomectomy were performed with a wire loop electrode at a power setting of 140 W blended cutting current and a coagulation power of 40 W. When normal cavity (i.e. no polyp) or thickened endometrium was observed, curettage was performed. All tissues removed were examined histologically.

Patients were discharged from hospital as soon as they were medically fit. Three days of rest and two weeks of abstinence from sexual intercourse were recommended. Patients were instructed to return to the department in case of heavy or prolonged bleeding, fever or abdominal pain. All patients had a four-week follow up examination.


Eighty-two postmenopausal asymptomatic patients were referred to our department for operative hysteroscopy following an incidental finding of intrauterine polyp. Mean age of patients was 61 years (range 49–76) and mean parity was 3 (range 0–13). Mean postmenopausal duration was nine years (range 1–35); 21% of patients received hormonal therapy and 14% were using tamoxifen. Mean BMI of patients was 25.6 kg/m2 (mean weight 72 ± 13.11 kg) and 26% of patients had hypertension.

Of the 82 patients with suspected polyps, 73 lesions were confirmed by diagnostic hysteroscopy. Diagnostic hysteroscopy was not feasible in nine patients, owing to cervical stenosis in six patients and poor compliance in three. These patients were referred to both diagnostic and operative hysteroscopy under general anaesthesia.

Hysteroscopic findings and histopathologic results are summarised in Table 1. Among the 82 patients with suspected polyp on transvaginal sonogram, 68 had endometrial polyps and 7 had submucosal myomas. Six patients had atrophic endometrium on both operative hysteroscopy and pathologic examination. One patient in whom a polyp was suspected on both transvaginal sonogram and diagnostic hysteroscopy had thickened endometrium on hysteroscopy and proliferative endometrium on histopathologic examination.

Table 1.  Hysteroscopic and histopathologic results in patients with suspected polyp.
No. of patientsOperative hysteroscopyHistopathologic result
  • *

    In one polyp simple hyperplasia was found.

68PolypEndometrial polyp*
1PolypAdenomyoma–polypoid type
6Atrophic endometriumAtrophic endometrium
1Thickened endometriumProliferative endometrium

Out of six patients with atrophic endometrium, diagnostic hysteroscopy failed in four patients and found a polyp in two patients. Patients with atrophic endometrium had endometrial thickness of 6–31 mm on transvaginal sonogram, three of them had a myomatous uterus (endometrial thickness measurements 6, 13 and 31 mm) and one patient was using tamoxifen (endometrial thickness measurement, 10 mm).

Three of the patients underwent two operative hysteroscopies two to three years apart for recurrent asymptomatic intrauterine polyps. Each time a polyp was identified and entirely resected. These patients did not use hormonal therapy or tamoxifen.

In two hysteroscopies, a fundal perforation was identified following polyp resection. One case of difficult intubation occurred during the procedure. None of the patients developed fluid overload. Hospitalisation after the procedure ranged from a few hours (71 patients), one day (10 patients) to six days when anticoagulation therapy by intravenous heparin was indicated (one patient). The total complication rate was 3.6%.


The evaluation of asymptomatic findings, such as abnormal endometrial thickness, may permit the detection of pre-maligant conditions and earlier detection of endometrial carcinoma. Intrauterine polyps undergo the same evaluation although their malignant potential remains unclear. Malignancy in resected polyps was reported in one large study,7 but unlike this report, the series included a heterogeneous population (symptomatic and asymptomatic, post- and premenopausal patients). In deciding whether to resect asymptomatic polyps in postmenopausal patients, certain factors must be considered: the inherent risk of malignancy or pre-malignant pathology, the potential malignant transformation of benign incidental polyps over the years if left in situ and the risks of the procedure itself.

Suspected intrauterine polyp was the indication for hysteroscopy in 82 patients in this series, of whom 75 (91.5%) actually had polyps or myomas. In only one polyp was simple hyperplasia noted. There were no cases of any more sinister pathology. The incidence of atypical hyperplasia and malignancy in endometrial polyps (symptomatic and asymptomatic) found by Savelli et al.7 was 0.8% and 3.1%, respectively. These authors did not find a difference in the prevalence of atypical hyperplasia and endometrial carcinoma between symptomatic and asymptomatic patients.

Three patients in this series had repeated hysteroscopies for recurrent endometrial polyps. The recurrence of polyps may indicate a stimulatory process to the endometrium causing polyps, but whether subsequent malignant transformation will occur is unknown. It is also not known whether de novo polyp development in postmenopausal patients is oestrogenic-driven, which can also induce carcinomatosis. If the stimulation leading to polyp formation is the same as that causing endometrial malignancy, logically resection of the polyp will not prevent the development of endometrial carcinoma.

The complication rate of hysteroscopies in the present study was 3.6%. This is in accordance with the findings of Shushan et al.,2 who concluded that hysteroscopy is a safe procedure in this age group. However, the potential advantage of identifying malignancy in asymptomatic polyps must be balanced against possible complications, inconvenience and cost. Although the complication rate is low, some patients will suffer adverse operative and anaesthetic sequelae such as water intoxication, uterine perforation, bleeding, infection and allergic reaction.

In this series, no malignancies were detected. The study group though is small, and the true incidence of malignancy in such patients cannot be determined, although it seems to be low. Nevertheless, because malignancy occurs, albeit rarely, and because the operative risk is low, polyp resection can be recommended in patients with low operative risk. In those patients deemed at a high operative risk, conservative management may be a safer option.

Accepted 24 July 2004