Significance of secondary ultrasonographic endometrial thickening in postmenopausal tamoxifen-treated women

Authors

  • Ilan Cohen M.D.,

    Corresponding author
    1. Department of Obstetrics and Gynecology, Sapir Medical Center, Kfar-Saba, affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Ramat-Aviv, Israel
    • Department of Obstetrics and Gynecology, Sapir Medical Center, Kfar-Saba, 44281 Israel
    Search for more papers by this author
    • Fax: 011-44-972-9-747-2212

  • Ron Azaria M.D.,

    1. Department of Obstetrics and Gynecology, Sapir Medical Center, Kfar-Saba, affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Ramat-Aviv, Israel
    Search for more papers by this author
  • Jeremiah Shapira M.D.,

    1. Medical Oncology Unit, Sapir Medical Center, Kfar-Saba, affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Ramat-Aviv, Israel
    Search for more papers by this author
  • Dror Yigael M.D.,

    1. Medical Oncology Unit, Sapir Medical Center, Kfar-Saba, affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Ramat-Aviv, Israel
    Search for more papers by this author
  • Ron Tepper M.D.

    1. Department of Obstetrics and Gynecology, Sapir Medical Center, Kfar-Saba, affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Ramat-Aviv, Israel
    2. Gynecological Ultrasound Unit, Sapir Medical Center, Kfar-Saba, affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Ramat-Aviv, Israel
    Search for more papers by this author

Abstract

BACKGROUND

Ultrasonography has a limited value in endometrial assessment for identification of endometrial pathologies in postmenopausal tamoxifen-treated patients.

METHODS

We compared the rate of endometrial pathologies and the mean ± SD of endometrial thickness diagnosed after the first and second transvaginal ultrasonographic studies performed on 55 postmenopausal tamoxifen-treated patients with secondary endometrial thickening (Group I). This rate was also compared with 46 similar patients without secondary thickening (Group II). We also compared the mean ± SD of endometrial thickness detected in various ultrasonographic studies, as well as various clinical features.

RESULTS

A significantly higher rate of endometrial pathologies, including two cases of endometrial cancer identified in gynecologically asymptomatic patients (3.6%), was diagnosed in Group I after the second study compared with the first study (52.7% and 9.1%, respectively; P = 0.001) and compared with those diagnosed after the second study in Group II (30.4%; P = 0.03). There was a significant increase (74.7 ± 115%) in endometrial thickness after the second study compared with the first study performed on Group I (10.7 ± 5.53 mm and 16.59 ± 5.53 mm, respectively; P = 0.0001) and a significant difference in endometrial thickness demonstrated in the second study performed on Groups I and II (16.59 ± 5.53 mm and 11.4 ± 3.91 mm, respectively; P = 0.001).There were no significant differences in the time elapsed since the diagnosis of breast carcinoma and from the beginning of tamoxifen treatment to the performance of the first ultrasonographic study as well as the time elapsed between the first and second studies performed.

CONCLUSIONS

A significant increase (> 50%) in secondary endometrial thickening, measured ultrasonographically, in postmenopausal tamoxifen-treated patients, is associated with a high rate of endometrial pathologies, including endometrial cancer. Cancer 2002;94:3101–6. © 2002 American Cancer Society.

DOI 10.1002/cncr.10587

Transvaginal ultrasonography (TVS) has been used extensively to investigate and evaluate endometrial thickness among postmenopausal breast carcinoma patients receiving tamoxifen (TAM) treatment.1–29 However, there is no definitive conclusion with regard to the effectiveness of this diagnostic procedure.

Many studies have shown the same trend of a thicker endometrial echogenic line, measured by TVS, in postmenopausal breast carcinoma TAM-treated patients.2–5, 7–9, 13, 17–19, 21, 25–28 Some have recommended TVS as an effective diagnostic tool for screening postmenopausal TAM-treated patients.2, 19, 21, 25 Others have claimed that TVS is of limited value in the assessment of the endometrium in these patients, as there is a high-false positive sonographic appearance and findings of a thick endometrium.1, 6, 10–12, 14–16, 20, 22–24, 29 However, no one has yet looked at the effectiveness of TVS in identifying endometrial pathologies after secondary thickening of the endometrium in postmenopausal breast carcinoma TAM-treated patients.

In this study, we compared the rate of overall and various endometrial pathologies diagnosed after secondary thickening of the endometrium with the rate of pathologies diagnosed after the first TVS study of postmenopausal breast carcinoma patients receiving adjuvant TAM therapy. We also compared the results with those observed in similar patients in whom a second TVS study did not reveal any significant change in endometrial thickness. This comparison was performed to evaluate whether there is a significant increase in the rate of endometrial pathologies after secondary thickening of the endometrium in postmenopausal breast carcinoma TAM-treated patients. To the best of our knowledge, this is the first such series to be reported in the literature.

MATERIALS AND METHODS

Since September 1, 1989, all postmenopausal breast carcinoma patients who were under medical supervision and/or treatment at the Sapir Medical Center gynecological outpatient clinic were followed using an investigative protocol.

The study was approved fully by the Institutional and Israeli Health Ministry Helsinki Committees. Informed consent was obtained from each patient after the nature of the study was explained fully.

Among other periodic routine examinations, each patient's evaluation included repeated TVS endometrial assessments. According to our protocol, ultrasonographic evaluations were performed every 6 months for the first 2 years of follow-up and every 12 months thereafter.

TVS was performed using a 128XP10 Ultrasound system equipped with a 5–7-mHz HDI 5000 (ATL) high-resolution endovaginal transducer (Acuson, Mountain-View, CA). The patients were examined in the lithotomy position with an empty bladder. The uterus was scanned both sagittally and coronally to determine the regularity of the endometrium. Anteroposterior measurements of endometrial thickness and regularity were obtained from a long axis view between the outermost edges of the line separating the hyperechogenic endometrium from the myometrium. The maximal width was recorded. For the last 6 years, patients with an endometrial thickness of more than 8 mm, measured ultrasonographically, were also evaluated by transvaginal sonohysterography (SHG).17

In this study, we evaluated 328 postmenopausal breast carcinoma TAM-treated patients. Fifty-five patients (16.8%) had secondary endometrial thickening, as detected in a second TVS evaluation (Group I), and 46 patients (14.0%) had no considerable change in endometrial thickness in a second TVS study (Group II).

After, each ultrasonographic imaging mentioned, an endometrial sampling was obtained. At the beginning of the study, suction curettage was used in 10 (18.2%) patients in Group I and in 9 (19.5%) patients in Group II. Later in the study, an endometrial tissue sampling was obtained during diagnostic hysteroscopy from 20 (36.3%) patients in Group I and from 20 (43.5%) patients in Group II. An operative hysteroscopy was performed on 25 (45.5%) patients in Group I and on 17 (37.0%) patients in Group II, who were diagnosed by SHG as having an endometrial echogenic mass.

We compared the rate of the overall and various endometrial pathologies diagnosed after the first and second endometrial TVS studies that were performed on the Group I and Group II patients. We also compared the mean ± SD of endometrial thickness, detected by the various TVS studies, as well as various demographic characteristics, risk factors for endometrial cancer, and clinical factors related to the primary breast carcinoma.

All patients were treated with 20–30 mg TAM daily (ABIC, Chemical and Pharmaceutical Industries, Netanya, Israel).

Statistical analysis was performed using the Wilcoxon two-sample test, the Fisher exact test, and the McNemar test. A value of P < 0.05 was statistically significant.

RESULTS

Demographic and Clinical Characteristics

No significant differences were found between the groups in most of the various factors tested, including the time elapsed since the diagnosis of breast carcinoma and the length of TAM treatment (Tables 1–3). Significantly more patients in Group II than in Group I had undergone chemotherapy treatment..

Table 1. Comparison of Demographic Characteristics between Postmenopausal Breast Carcinoma Tamoxifen-Treated Patients with Secondary Endometrial Thickening (Group I) and Similarly Treated Patients with No Considerable Change in Endometrial Thickness (Group II)a
CharacteristicsGroup I (n = 55)Group II (n = 46)
  • a

    All parameters tested were statistically not significant.

Age (yrs) (mean ± SD)63.8 ± 9.262.8 ± 8.8
Parity (mean ± SD)2.8 ± 1.62.9 ± 2.5
Gravidity (mean ± SD)3.8 ± 2.04.8 ± 4.1
No. of Smokers (%)6 (10.9)5 (10.9)
Table 2. Comparison of Risk Factors for Endometrial Carcinoma between Postmenopausal Breast Carcinoma Tamoxifen-Treated Patients with Secondary Endometrial Thickening (Group I) and Similarly Treated Patients with no Considerable Change in Endometrial Thickness (Group II)a
CharacteristicsGroup I (n = 55)Group II (n = 46)
  • HRT: hormone replacement therapy; PMB: postmenopausal bleeding.

  • a

    All parameters tested were statistically not significant.

Age (yrs) at menarche (mean ± SD)12.3 ± 2.112.8 ± 1.8
Age (yrs) at first delivery (mean ± SD)22.6 ± 6.222.0 ± 8.3
Age (yrs) at menopause (mean ± SD)48.6 ± 4.148.5 ± 8.2
Weight (kg; mean ± SD)67.7 ± 11.269.8 ± 13.6
HRT (%)2 (3.6)1 (2.2)
Diabetes mellitus (%)4 (7.3)6 (13.0)
Hypertension (%)18 (32.7)12 (26.1)
PMB (%)5 (9.1)4 (8.7)
Table 3. Comparison of Clinical Parameters Related to the Primary Breast Carcinoma between Postmenopausal Breast Carcinoma Tamoxifen-Treated Patients with Secondary Endometrial Thickening (Group I) and Similarly Treated Patients with no Considerable Change in Endometrial Thickness (Group II)a
CharacteristicsGroup I (n = 55)Group II (n = 46)
  • TVS: transvaginal ultrasonography.

  • a

    All other parameters tested were statistically not significant (including time elapsed since diagnosis of breast carcinoma to the performance of the first TVS study and duration of breast carcinoma to the time of performance of the first TVS study).

  • bP value = 0.001.

Time (mos) elapsed since diagnosis of breast disease to the performance of first TVS study. (mean ± SD)37.9 ± 53.931.4 ± 31.8
Duration of tamoxifen treatment (mos) up to the time of performance of first TVS study (mean ± SD)33.3 ± 52.825.8 ± 21.8
Radiotherapy (%)29 (52.7)28 (60.7)
Chemotherapy (%)6 (10.9)a18 (39.1)a

TVS Studies

The first TVS study revealed a significantly thicker endometrium in Group II patients compared with Group I patients (Table 4). The second TVS study performed on Group I patients detected a significant increase in endometrial thickness compared with the first study, both in terms of millimeters and in percentage. There was a significant difference between Group I and Group II patients with regards to endometrial thickness detected in the second study was compared with measurements in the first study. There was no difference in endometrial thickness detected in the first and second studies performed on Group II patients (Table 4). A comparison of patients in both groups showed no significant differences in the time elapsed since the diagnosis of breast carcinoma and from the beginning of TAM treatment to the performance of the first TVS study, nor was there a significant difference in the time elapsed between the first and second studies for both groups of patients (Tables 3,4).

Table 4. Comparison of Endometrial TVS Findings between Postmenopausal Breast Carcinoma Tamoxifen-Treated Patients with Secondary Endometrial Thickening (Group I) and Similarly Treated Patients with no Considerable Change in Endometrial Thickness (Group II)
CharacteristicsGroup I (n = 55)Group II (n = 46)P valuea
  • TVS: Transvaginal ultrasonography.

  • a

    P value of the difference in TVS endometrial thickness between first and second studies in Group I is 0.001.

First TVS study (mm; mean ± SD)10.7 ± 5.413.5 ± 6.40.0087
Second TVS study (mm; mean ± SD)16.59 ± 5.5311.4 ± 3.910.001
Difference in endometrial thickness between first and second TVS studies (mm; mean ± SD)5.9 ± 5.5−2.1 ± 3.90.0001
Difference in endometrial thickness between first and second TVS studies (%; mean ± SD)74.7 ± 115.0−13.4 ± 23.90.0001
Time (mos) elapsed between first and second TVS studies (mean ± SD)23.2 ± 15.819.6 ± 13.3NS

Endometrial Histologic Findings

A histologic evaluation performed after the first TVS study revealed significantly more overall positive endometrial findings in Group II patients compared with Group I patients. All were benign pathologies: endometrial proliferation, simple endometrial hyperplasia, and endometrial polyps. A histologic evaluation of the specimens collected from Group I patients after the second TVS study revealed significantly more overall endometrial pathologies (including two cases of endometrial cancer [3.6%]: both patients were gynecologically asymptomatic) compared with specimens collected after the first TVS study and with those collected from Group II patients after the second sampling (Table 5).

Table 5. Comparison of Endometrial Histologic Findings between Postmenopausal Breast Carcinoma Tamoxifen-Treated Patients with Secondary Endometrial Thickening (Group I) and Similarly Treated Patients with No Considerable Change in Endometrial Thickness (Group II)a
Endometrial histologic findingGroup I (n = 55)Group II (n = 46)
First studySecond studyFirst studySecond study
  • a

    All other parameters tested were statistically not significant.

  • b

    P = 0.001.

  • c

    P = 0.016.

  • d

    P = 0.03.

Proliferation3123
Simple hyperplasia1411
Hyperplasia with atypia1
Polyp121810
Carcinoma2
Overall5 (9.1%)bc29 (52.7%)bd11 (23.9%)c14 (30.4%)d

Hormonal Evaluation

All patients in the study had serum 17 estradiol levels in the normal menopausal range (< 20 pg/mL; < 73.25 pmol/L).

DISCUSSION

In other studies on postmenopausal breast carcinoma TAM-treated patients, TVS has shown an endometrial echogenic line that is wider than 5 mm, which is the accepted endometrial thickness for healthy postmenopausal women.2–5, 7–9, 13, 17–19, 21, 25–28 Endometrial thickness, measured by TVS, was significantly thicker in asymptomatic, postmenopausal breast carcinoma TAM-treated patients compared with similar nontreated patients3, 7, 13 and with healthy postmenopausal patients.8 Most of the TAM-treated patients had a TVS complex endometrial echo of more than 5 mm.2, 3, 7, 8, 13

In addition, there was a significant increase in mean endometrial thickness measured before TAM treatment and up to more than 3 years of treatment,27 as well as a significant decrease in median endometrial thickness within 6 months after discontinuation of TAM treatment.27, 28 Other investigators have also found TVS to be effective in detecting endometrial abnormalities in postmenopausal TAM-treated patients and have recommended TVS as a useful diagnostic tool for screening postmenopausal TAM-treated patients.2, 19, 21, 25

Some investigators believe that TVS may be more effective and acceptable than office hysteroscopy in detecting endometrial abnormalities in women taking TAM.21 Others have claimed that TVS is of limited value in the evaluation of the endometrium of these patients.1, 6, 10–12, 14–16, 20, 22–24, 29 A false-positive sonographic appearance of a thick endometrium and a discrepancy between the sonographic and histologic findings were described in these patients. This may be attributed to stromal edema caused by TAM,10–12 due to the existence of subendometrial sonolucencies in adjacent myometrial tissue or to a thick miduterine structure resembling a thickened endometrium with no histologic evidence of endometrial neoplasm.1, 14, 15 Other researchers have suggested that assessing the single sonographic features of a single pathologic diagnosis is not possible.6 Cecchini et al.16 found an abnormal endometrial thickness of 6 mm or more, measured ultrasonographically, in only 28.4% of their postmenopausal TAM-treated patients.

Two large studies have shown the most prominent data on the usefulness of TVS in postmenopausal TAM-treated patients.23, 27 Love et al.23 used TVS to screen 357 TAM-treated patients and 130 controls. Women who were seen to have a thick endometrium underwent an outpatient hysteroscopy. Gerber et al.27 performed TVS studies every 6 months for up to 5 years in 247 TAM-treated patients and 98 controls. In both studies, the women treated with TAM had a significantly thicker endometrium than did the controls. However, there was a significantly low frequency of endometrial pathologic findings (46% and 73.1%, respectively)23, 27 and only one case of endometrial carcinoma was diagnosed in an asymptomatic patient.27 Based on these findings, both studies concluded that TVS is a poor diagnostic screening tool because of its high false-positive rate.23, 27

Some investigators have shown that extending the duration of TAM treatment in postmenopausal breast carcinoma patients does not cause a significant increase in TVS-measured endometrial thickness.26, 27 Others, however, have reported a highly significant positive correlation between the length of time of TAM treatment and endometrial thickness.23, 30 However, none of these studies examined the possible correlation between the thicker endometrium and the higher rate of endometrial pathologies.

We evaluated whether there is a significant increase in the rate of endometrial pathologies after secondary thickening of the endometrium in postmenopausal breast carcinoma TAM-treated patients. To do this, we compared the rate of overall and various endometrial pathologies diagnosed after the first and second endometrial TVS studies among the postmenopausal breast carcinoma TAM-treated patients with secondary endometrial thickening with similar patients who had no considerable change in endometrial thickness after a secondary ultrasonographic study.

There was a slight decrement in endometrial thickness in the second TVS study compared with the first TVS study performed on Group II patients. In addition, there was no significant change in the rate of endometrial pathologies recovered after the two TVS studies. There was a significant increment in endometrial thickness in the second TVS evaluation performed 23.2 ± 15.8 months after the first TVS study in Group I patients. This increment in endometrial thickness was accompanied by a significant increase in the rate of endometrial pathologic conditions identified (52.7) [including two cases of endometrial cancer (3.6%). Both of these were diagnosed in gynecologically asymptomatic patients]. This rate of endometrial pathologies is significantly higher than the rate (25–30%) reported in the literature for postmenopausal TAM-treated patients.2, 3, 7, 8, 13, 17–19, 21, 22, 24, 25, 29, 30

Although different methods of endometrial sampling were used in the current study, they were used in almost the same proportions in each group of patients. It is improbable that they would yield different histologic results. Furthermore, the rate of 52.7% of endometrial pathologies diagnosed after secondary thickening of the endometrium was significantly higher in Group I patients compared with 30.4% of pathologies found after a second TVS evaluation in Group II patients. However, there were no significant differences in the time elapsed since the diagnosis of breast carcinoma and from the beginning of TAM treatment to the performance of the first TVS study, nor were there significant differences in the time elapsed between the first and second studies that were performed on both groups of patients.

We cannot explain the relatively lower endometrial thickness diagnosed in the first TVS study, nor can we explain the subsequently lower rate of endometrial pathologies diagnosed in Group I patients compared with Group II patients. The patients in this study were selected from the same population, they had similar clinical features, and there were no significant differences in the time elapsed from the diagnosis of breast carcinoma and from the beginning of TAM treatment to the performance of the first TVS study. For these reasons, we do not believe that the Group I patients represent a specific group of patients.

We did not calculate the significance of the difference in the various endometrial pathologies diagnosed because of the relatively few endometrial pathologic entities. However, of importance, is the finding of two cases of endometrial carcinoma diagnosed after the secondary thickening of the endometrium in gynecologically asymptomatic patients, which were not diagnosed in any other session of the study.

Extending the duration of TAM treatment in postmenopausal breast carcinoma patients does not cause a significant increase in TVS endometrial thickness.26, 27 Despite a statistically significant positive correlation between length of time on TAM and endometrial thickness, no case of endometrial carcinoma was diagnosed.23 TVS is a poor diagnostic screening tool because of its high-false positive rate.23, 27 Therefore, there is no need for endometrial samplings.26, 27

We have demonstrated that TVS is a useful diagnostic tool in identifying endometrial pathologies in a specific high-risk subgroup of postmenopausal TAM-treated patients. Our findings emphasize the need to perform endometrial sampling after a significant secondary increase (> 50%) in endometrial thickness, as measured by TVS, due to the high rate of endometrial pathologies, including a relatively high rate of endometrial carcinoma that was diagnosed.

Significant secondary endometrial thickening, detected by TVS, in postmenopausal breast carcinoma TAM-treated patients is associated with a high rate of endometrial pathologies, including endometrial carcinoma. This finding justifies the requirement for endometrial sampling.

Acknowledgements

We thank Mrs. Ruth Sheridan for her editorial assistance.

Ancillary