Adding Endometrial Biopsy to Polypectomy: Is It Worthwhile?

Introduction

Endometrial polyps are localized overgrowths of endometrial tissue that is composed of variable amounts of glands, stromas and blood vessels, that is, covered by epithelium. ¹ They can be solitary or multiple or can occur in association with other intrauterine abnormalities,^2,3 Endometrial polyps may be asymptomatic and found incidentally during routine scanning or be associated with abnormal uterine bleeding. The majority of polyps are benign with a low risk of premalignant or malignant transformation (0.8–4.9%).^2,3 The pathogenesis of polyps is not fully understood but they are believed to be a risk factor for the development of hyperplastic changes in the endometrium as well câncer.

Hysteroscopy is the gold standard for both the diagnosis and treatment of endometrial polyps,^3,4 but some authors question the need for endometrial biopsy in every diagnostic hysteroscopy, mainly when its results are normal and the endometrium is atrophic. It should be noted that hyperplasia within an endometrial polyp in a polypectomy or biopsy specimen is uncommon but not rare.^5–7

Hysteroscopic resection of endometrial polyps is the gold standard method of polyp removal. It ensures complete removal of the polyp as well as assessment of the endometrial cavity but is there a need for additional endometrial evaluation in women with these polyps?^2,3

A retrospective study of 133 consecutive hysteroscopic polypectomies at a private tertiary hospital (Biocor Hospital), to establish the role of adding endometrial biopsy to polypectomy in the diagnosis of intrauterine pathology, was performed.

Methods

One hundred and thirty-three consecutive patients who underwent hysteroscopic resection of endometrial polyps with additional endometrial biopsy from July 1997 through June 2004 with conclusive histologic results were included. All women were >40 years of age, and those with abnormal uterine bleeding or an abnormal pelvic scan were referred for investigation. Most women had had a pelvic scan before referral to hysteroscopy. None of them had undergone office endometrial sampling prior to hysteroscopy. Hysteroscopy was performed irrespective of the ultrasound results or menstrual cycle phase in the case of abnormal uterine bleeding.

Because many patients had undergone diagnostic hysteroscopy or other diagnostic methods elsewhere using different diagnostic criteria, a diagnostic hysteroscopy was performed in all patients before starting the surgical procedure, to make hysteroscopic examination and diagnostic criteria uniform.

Hysteroscopic diagnosis was defined by the appearance of the surface of the uterine cavity before biopsy or surgical exeresis. Hysteroscopic surgery was performed by two experienced gynecologists and was preferably scheduled to take place in the early proliferative period, mostly 2–3 days after menstruation, and all patients were instructed to use barrier methods for contraception in the cycle preceding surgery unless they were menopausal. The type of anesthesia (general or epidural) used depended upon the patient's age and clinical conditions assessed by the anesthesiologist.

The procedure included a pelvic examination, speculum placement, cervical tenaculum application, and measurement of the uterine cavity. Diagnostic hysteroscopy was performed using a 4-mm hysteroscope under continuous saline flow. The examination was defined as completed when the entire uterine cavity was visualized. Hysteroscopic findings were defined as the diagnostic impression based on the surface of the uterine cavity before biopsy or resection according to the criteria described by Hamou. ⁸ All surgical procedures were recorded by VHS. The operation was performed as described previously. ⁸ Briefly, diagnostic hysteroscopy was performed with a standard rigid hysteroscope (Karl Storz GbmH & Co., Tuttlingen, Germany) with a 5.5-mm diagnostic sheath. The hysteroscope was guided through the endocervical canal into the uterine cavity under visual control, the tubal ostia identified, and the endometrial surfaces systematically inspected. The cervical canal was then viewed in its entire length during withdrawal of the hysteroscope. If the operative sheath was required, the cervix was dilated 7–8 mm. Operative procedures were performed either using an operative hysteroscope (Karl Storz) with a 9-mm operative sheath or a resectoscope (Karl Storz) with an outer sheath 9 mm in diameter. Intrauterine polyps were removed hysteroscopically with electrosurgery using either a monopolar probe or a resectoscope. For small or narrow stalk polyps, polypectomy was performed using a monopolar electrode through an operative hysteroscope. For large polyps and those with a broad base, a resectoscope was used. Complete removal of intrauterine polyps was attempted in all patients. Polypectomy was continued until a normal panoramic view of the endometrial cavity was obtained and both tubal ostia were visualized. Hysteroscopically targeted endometrial biopsy far from the base of the polyp was added to polypectomy in all cases. The area biopsied looked normal at hysteroscopy. Sterile 2.7% sorbitol and manitol 0.54% solution was used for uterine distention and irrigation. To prevent excessive intravasation, fluid balance was recorded strictly by measuring the infused and drained fluid from the continuous flow hysteroscope, taking into account the fluid irrigated separately from the operative field into a collecting bag.

The material for histologic examination was obtained through endometrial biopsy far from the base of the polyp or through the resection of the entire lesion in patients who underwent surgery. Endometrial tissue obtained during hysteroscopy was examined by one experienced pathologist under standard paraffin cuts and a special hematoxylin stain. The criteria used for histopathologic diagnosis were defined by Kurman ⁹ and Rosai and Ackerman. ¹⁰

Standard histopathologic criteria were used for diagnosing the lesions and as the gold standard against which the hysteroscopic view was compared. Description of the surgical procedure and the morphologic appearance of the endometrium and polyps, as well as histologic diagnosis, were recorded for all patients. The study protocol was approved by the local Ethics Committee. Sensitivity and specificity were calculated using the Statistical Package for the Social Sciences (SSPS). McNemar test was used to test the value of adding biopsy to polypectomy.

Results

One hundred thirty-three polypectomies with conclusive histologic results were analyzed. Mean patient age was 59 years of age (range, 30–85 years). Only 22 (16,5%) of the women studied were premenopausal. Endometrial polyp (n = 79; 59.3%) was the main surgical indication followed by abnormal uterine bleeding (n = 29; 21.8%), abnormal endometrial thickness on ultrasound (n = 20; 15,1%), and miscellaneous (n = 5; 3.7%). Endometrial polyp was the sole diagnosis in 44 (33,1%) whereas in 89 an additional diagnosis was made, the majority being endometrial hyperplasia (n = 66; 49,6%) and atypical endometrial hyperplasia (n = 10; 7,5%). Endometrial cancer was identified in three cases (Table 1). The most frequent additional diagnoses obtained by random biopsy in patients with false negative hysteroscopies were simple endometrial hyperplasia (n = 66), followed by atypical endometrial hyperplasia (n = 10), endometrial cancer (n = 3) and myoma (n = 2). The patients with atypical endometrial hyperplasia presented endometrial polyp on ultrasound (n = 5), abnormal endometrial thickness on ultrasound (n = 2) and abnormal uterine bleeding (n = 2). Only 2 patients were premenopausal. Of the patients with endometrial cancer (n = 3) only 2 complained of abnormal uterine bleeding and one was premenopausal. The presence of clinical symptoms (bleeding or abnormal endometrium on ultrasound) showed a low sensitivity (36.9% 95% CI:27.3–47.5) and specificity (61.2% 95% CI:47.2–73.5) in the diagnosis of endometrial hyperplasia or cancer. Hysteroscopic view also revealed a low sensitivity and specificity in the diagnosis of atypical endometrial hyperplasia and cancer (Table 2). The addition of endometrial biopsy significantly increased the diagnosis of atypical hyperplasia and cancer (McNemar test p = 0.000) as well as of simple hyperplasia (p = 0.008).

Discussion

Hysteroscopy allows thorough inspection of the uterine cavity and endocervical canal in addition to the assessment of endometrium under magnification. Nevertheless, the need for endometrial biopsy in every diagnostic hysteroscopy is frequently questioned. The results in this study show that the combination of diagnostic hysteroscopy with endometrial biopsy improved the diagnostic accuracy and significantly increased the diagnosis of atypical hyperplasia and cancer as well as of simple hyperplasia. Clinical symptoms and hysterocospic view showed low sensitivity and specificity in the diagnosis of endometrial hyperplasia and cancer, as expected.^5,11

In this study population the mean age was 59 years and it is in this age group that concerns about endometrial cancer and hyperplasia should be raised. ¹¹ This could account for the high incidence of endometrial hyperplasia in this study.

Although the risk of malignancy in a polyp is low, the risk of concurrent endometrial hyperplasia and cancer in the surrounding endometrium is significant and could be otherwise missed by hysteroscopy.^6,7,11 There appears to be only one published study evaluating the occurrence of hyperplasia in the endometrium surrounding a polyp, which found hyperplasia in 52% of endometrial biopsy specimens obtained after polypectomy. ⁷ The findings of the current study highlight the need to add endometrial biopsy to polypectomy as 66 cases (49%) of simple hyperplasia and 10 cases of atypical hyperplasia (7.5%) were identified. As hysteroscopy alone cannot accurately predict endometrial histopathology, especially when hyperplasia is considered,^5,11 and random endometrial biopsy is not effective for diagnosing focal intrauterine lesions, both techniques should be combined to increase diagnostic performance.^7,12

This study is subject to the limitations of any retrospective study. As patients' charts were reviewed, some data were simply not found. Histopathologic reports were used and tissue samples were not reviewed to validate the histopahologic results, although all histopathologic specimens were evaluated by an experienced pathologist using well-accepted criteria.^9,10 As the study was performed at a tertiary referral center, patients came from various parts of the state having had different diagnostic workups. Most patients had a pelvic ultrasound scan but none underwent endometrial sampling. In an attempt to make hysterococpic diagnostic criteria uniform, a diagnostic hysteroscopy was performed in all patients immediately before starting the surgical procedure. Patients were sent back to their gynecologists to complete treatment in cases of atypical endometrial hyperplasia or cancer. The hysterectomy specimens have not been retrieved to date.

The addition of endometrial sampling to polypectomy significantly increased the diagnosis of concomitant endometrial abnormalities, mainly endometrial hyperplasia and cancer, which could have been missed otherwise. Therefore, endometrial biopsy should be added to polypectomy.