Management Considerations for Prolapsed Submucosal Leiomyomata: A Retrospective Cohort Study

Background

Uterine leiomyomata are benign monoclonal tumors of the smooth muscle of the myometrium. ¹ The exact prevalence is unknown, since not all leiomyomata are symptomatic, however, estimates range from 40% to 70% depending on age, race, and other risk factors. ² Submucosal leiomyomata are known to prolapse beyond the cervix^3,4 and usually appear as a mass protruding from the cervix. Clinically, it can be difficult to distinguish these prolapsed leiomyomata from other lesions, such as cervical polyps, endometrial polyps, ⁵ uterine inversion, ⁶ uterine cancers, or cervical cancers.^7,8

Our goals were to describe (1) sensitivity and specificity of physical examination to correctly diagnose a prolapsed vaginal mass as benign, (2) success rate for transvaginal myomectomy, and (3) risk of associated malignancy.

Materials and Methods

This was a retrospective cohort study of all women seen at the LAC+USC Medical Center from March 2007 to June 2012, with an initial diagnosis, including prolapsing leiomyoma. Since these patients were evaluated and treated in a variety of clinical settings, subjects were identified from the following four sources: (1) Pathology Department database, using the search terms “aborting,” “transcervical myomectomy,” “prolapsed myoma,” “prolapsed fibroid,” and “prolapsed leiomyoma,” (2) departmental records of all surgical cases performed to identify patients undergoing transvaginal myomectomy, (3) Operating Room Scheduling Office System to identify any cases of transvaginal myomectomy by the CPT code, and (4) obstetrics and gynecology log for all patients evaluated for prolapsed leiomyoma in the outpatient procedure area. The Institutional Review Board at the University of Southern California approved this protocol.

Medical records of all subjects identified were reviewed for demographic information, clinical presentation, initial physical examination findings, diagnostic impression, operative findings, final pathology, and results of subsequent biopsy specimens and hysterectomy specimens. Comparisons between groups were made using a Student's t-test, Fisher's exact test, Mann–Whitney test, or χ² analysis as appropriate. For the purposes of data analysis, a p-value of <0.05 was considered statistically significant. All statistics were performed using SPSS v20.0.0.

Results

A total of 389 charts were reviewed, and 226 cases were confirmed to have an initial diagnosis of prolapsed submucosal leiomyoma. Basic demographics are shown in Table 1. The most common presenting symptoms were bleeding (79.2%) and pain (28.8%). Fifty-two (23%) women reported both bleeding and pain, and 34 (15%) women were asymptomatic. The median size of the mass on examination was 4.0 cm (range 1–11, IQR 2.0) (n=203), with a median stalk size of 1.0 cm (range 0.3–5.0, IQR 1.0) (n=87). The median hemoglobin on the day of presentation was 9.6 g/dL (range 3.2–15.6, IQR 4.5) (n=212). Sixty-one percent of women were anemic (hemoglobin level ≤11 g/dL) and 28.8% were severely anemic (hemoglobin ≤8 g/dL). There were no cases of uterine infection diagnosed at the time of presentation.

Of the 226 cases, 208 (92.0%) had surgical management on the day of presentation. Eighteen cases were delayed for biopsy or transfusion; 13 women had definitive treatment within 2 weeks of presentation, and five women had treatment, a median of 98 days after presentation. Of the 18 cases that were delayed, one woman had an acute episode of vaginal bleeding 2 days later and underwent transvaginal myomectomy, one woman required transfusion 16 months after initial presentation, and one woman spontaneously expelled the leiomyoma 9 days after initial presentation and required no further intervention. The median follow-up after treatment was 26.0 days (range 0–1110, IQR 12.7).

Of those who had surgical management for their mass, 195 had transvaginal myomectomy and 14 underwent hysterectomy (7 total abdominal, 5 supracervical, and 2 vaginal). Complications associated with these procedures are listed in Table 2. Overall, women who had a transvaginal myomectomy had a significantly lower blood loss than those who had an immediate hysterectomy. Two cases were associated with uterine inversion; both cases were in older (38 and 48 years old, respectively) parous women with large masses (8.0 and 8.2 cm, respectively) and significant anemia. Two transvaginal myomectomies were aborted due to bleeding and returned malignancy on biopsy. Three (1.5%) women returned within 30 days for acute vaginal bleeding after transvaginal myomectomy, a median of 14 days (range 1–29, IQR 28) after the procedure.

Final histopathology was available for all cases (n=226). Histopathology was confirmed as follows: leiomyoma in 140 (61.9%), adenomyoma in 17 (7.5%), polyp in 60 (26.5%), squamous cervical cancer in 5 (2.2%), small-cell neuroendocrine cancer in 1 (0.4%), prolapsing endometrial cancer in 1 (0.4%), polyp containing atypical hyperplasia in 1 (0.4%), and benign in 1 (0.4%) subject(s). Of the women confirmed to have a benign mass, additional endometrial sampling was performed in 166 cases (75.8%). Of these endometrial samples, 124 (74.6%) were benign (atrophic, proliferative, secretory, or menstrual phase), 27 (16.2%) were polyps, 8 (4.8%) were insufficient for evaluation, and 7 (4.2%) returned endometrial hyperplasia or endometrial carcinoma.

The initial primary diagnosis included possible malignancy in 14 cases (6.2%). When the initial impression was benign, the final pathology was malignant in 4 of 212 cases (1.9%); when the initial impression was possible malignancy, the final histology was malignant in 3 of 14 cases (21.4%) (p=0.006). The positive predictive value (PPV) of a suspicious examination was 21.4% (95% CI: 5.7%–51.2%) and the sensitivity and specificity of examination to detect cancer were 42.8% (95% CI: 11.8%–79.8%) and 95.0% (91.0%–97.3%), respectively. Initial descriptions of the masses included the following: smooth (11.9%), friable (7.5%), necrotic (7.1%), hard (4.4%), degenerating (4.0%), and irregular (3.5%). The only descriptor significantly associated with a final cancer diagnosis was degenerating (p=0.013), although friable approached a significant association (p=0.08). None of the other descriptors was associated with a final diagnosis of cancer. When the initial impression was probable leiomyoma, 120 of 157 (76.4%) were confirmed as leiomyomata for a PPV of examination to detect leiomyoma of 76.4% (95% CI: 68.9%–82.7%), a negative predictive value of 69.6% (95% CI: 57.2%–79.8%), sensitivity of 85.1% (95% CI: 77.9%–90.3%), and specificity of 56.5% (95% CI: 45.3%–67.0%).

Of all the women who underwent transvaginal resection of their mass (n=195), two were diagnosed with cervical cancer. Twenty-one of the remaining 193 (10.9%) later required hysterectomy for persistent bleeding at a median (range) of 142.5 (7–1103, IQR 191.2) days after their initial presentation. Of the patients who underwent transvaginal resection of their mass that was known to have additional leiomyomata, 24.6% later required hysterectomy, while only 3.4% of those who did not have additional leiomyomata eventually required hysterectomy (p=0.004).

Discussion

We reviewed our experience with prolapsing submucosal leiomyomata and polyps at a single institution over a 5-year period. Prolapsing uterine masses represent a diverse spectrum of diseases, with benign leiomyomata being the most common. Other benign lesions include adenomyomas and endometrial polyps. We found that prolapsing leiomyomata, adenomyomas, and endometrial polyps were not always distinguishable clinically and, thus, included them as one clinical entity.

Transvaginal excision appears to be the treatment of choice for most women with prolapsing leiomyomata and polyps.^3–5,9,10 The procedure is associated with low risk of complications and results in cure of abnormal bleeding and/or pain in the majority of women. Although blood loss associated with the procedure is usually low, 6.1% of women required perioperative blood transfusion due to anemia. In contrast, immediate hysterectomy is associated with a greater blood loss and a higher risk of blood transfusion. This may be due to the increased complexity of a hysterectomy and difficulty in securing the uterine vessels in these women or it may reflect a selection bias of women who require a hysterectomy in this setting. Due to the retrospective nature of these data, it is difficult to elucidate the exact reasons. Only 14.4% of women who had a successful transvaginal excision of a prolapsing leiomyoma or polyp required subsequent hysterectomy; women with additional leiomyomata were significantly more likely to require subsequent hysterectomy for persistent symptoms. Women who require subsequent hysterectomy should have their aborting lesions excised to reduce the risks of persistent bleeding. Furthermore, removal of the aborting lesion will allow the cervix to return to the undilated state, which facilitates surgical access and decreases the risk of ureteral injury.

Immediate treatment is usually recommended, and the majority of women in our study had immediate surgical excision; however, a short delay in treatment is probably safe. Only 1 (5.9%) of 17 women who had delayed management returned with acute bleeding in the first 2 weeks after diagnosis. There may be indications to delay surgical treatment if the diagnosis is unclear and biopsy of the mass is required or when the patient is unwilling or unable to tolerate immediate surgery.

Our data suggest that in women with benign prolapsing uterine masses, concomitant endometrial cancer or hyperplasia is not rare. Endometrial sampling was not performed in all women, thus potentially artificially increasing the apparent prevalence of concomitant endometrial pathology. It is possible that the women who had endometrial sampling were at a higher risk of endometrial pathology due to known risk factors and, therefore, had endometrial sampling. However, even assuming that the unbiopsied women did not have any endometrial pathology, 7 of 226 women (3.1%) had concomitant endometrial cancer or hyperplasia. The presence of a prolapsing leiomyoma does not preclude the possibility of endometrial pathology, and endometrial sampling should be considered in these women, either at the time of surgical excision of the prolapsing lesion or soon thereafter.

Squamous cell cervical cancer, endometrial adenocarcinoma, a polyp containing atypical hyperplasia, and small-cell neuroendocrine cancer were identified in 8 (3.5%) women who were initially thought to have a prolapsing leiomyoma. The sensitivity of the clinical examination to diagnose malignancy was low. In fact, two women with cervical cancer underwent attempted transvaginal myomectomy associated with excess bleeding before the diagnosis was made. These findings are consistent with other observational studies, where presumed leiomyomata were actually found to be cervical cancer, ³ uterine carcinosarcoma, ³ leiomyosarcoma, ⁸ uterine inversion,^6,11,12 and even enterocele. ¹³ These results are limited by the retrospective nature of the study, which relies heavily on the existing clinical documentation that may be incomplete. Furthermore, the majority of these women were initially examined by a resident or a nurse practitioner. The initial clinical impression of experienced practitioners may be more reliable and sensitive. We feel that while routine biopsy of the prolapsing mass is probably not indicated, there should probably be a low threshold for considering the biopsy of unusually appearing masses before surgical management.

This study is the largest case series of prolapsed uterine masses in the literature. Not all prolapsing masses are leiomyomata, although the majority of masses are benign. Physical examination findings do not appear to be sensitive to exclude malignancy; therefore, cancers masquerading as prolapsing tumors should be considered in the differential diagnosis when evaluating these women. Concomitant endometrial hyperplasia or malignancy is not rare and should be evaluated in these women. Transvaginal excision is curative in the majority of women, especially when there are no additional uterine leiomyomata.