Echogenic Endometrial Fluid Collection in Postmenopausal Women Is a Significant Risk Factor for Disease

Transcription

1 Article Echogenic Endometrial Fluid Collection in Postmenopausal Women Is a Significant Risk Factor for Disease Peter Takacs, MD, PhD, Timothy De Santis, MD, M. Catherine Nicholas, MS, RDMS, Usha Verma, MD, Richard Strassberg, MD, Lunthita Duthely, MS Objective. The purpose of this study was to assess postmenopausal women with endometrial fluid collection and the risk of significant endometrial or cervical disease. Methods. A retrospective chart review was conducted of 343 postmenopausal women with endometrial fluid collection on pelvic sonography. Medical records were reviewed to identify women who underwent an evaluation of the endometrium with endometrial biopsy, hysteroscopy, or hysterectomy after the sonographic examination. Clinical and sonographic characteristics were compared between women with diagnoses of cervical or endometrial cancer or hyperplasia (nonbenign group) and women with benign conditions (benign group). Results. The endometrium was significantly thicker in the nonbenign group compared with the benign group (mean ± SD, 9.9 ± 7.4 versus 5.9 ± 4.1 mm; P =.016). None of the patients with adenocarcinoma of the endometrium had endometrial thickness of 3 mm or less, but 2 with endocervical cancer did. Echogenic fluid in the endometrial cavity was significantly more likely to be found in the nonbenign group compared with the benign group (45.8% versus 4.8%; P <.01). Multivariate logistic regression analysis revealed that echogenic fluid in the endometrial cavity was the only significant risk factor for nonbenign conditions (odds ratio, 10.94; 95% confidence interval, ; P <.01). Conclusions. Postmenopausal women with endometrial fluid collection on sonography should undergo endometrial sampling if the endometrial lining is thicker than 3 mm or the endometrial fluid is echogenic. If the lining is 3 mm or less and the endometrial fluid is clear, endometrial sampling is not necessary, but we recommend endocervical sampling to rule out endocervical cancer. Key words: echogenicity; endometrial cancer; endometrial fluid; menopause. Abbreviations CI, confidence interval; HRT, hormone replacement therapy; OR, odds ratio Received April 13, 2005, from the Department of Obstetrics and Gynecology, Jackson Memorial Hospital, University of Miami, Miller School of Medicine, Miami, Florida USA. Revision requested May 16, Revised manuscript accepted for publication June 13, Address correspondence to Peter Takacs, MD, PhD, Department of Obstetrics and Gynecology, D- 50, Jackson Memorial Hospital, University of Miami School of Medicine, PO Box , Miami, FL USA. ptakacs@med.miami.edu Advancement of sonographic equipment and the widespread use of vaginal probe sonography have resulted in detection of endometrial fluid in many postmenopausal patients. The reported rate of endometrial fluid collection found on pelvic sonography in postmenopausal women varies from 4% to 18%. 1 4 Endometrial fluid collection in postmenopausal women has been thought to be a sign of serious endometrial or cervical disease, hyperplasia, or cancer. 5 7 Contrary to this, more recently, several investigators reported a low incidence of significant endometrial disease in postmenopausal women with endometrial fluid 2005 by the American Institute of Ultrasound in Medicine J Ultrasound Med 2005; 24: /05/$3.50

2 Echogenic Endometrial Fluid Collection in Postmenopausal Women collection. 2,8 10 These investigators concluded that postmenopausal intrauterine fluid collection is a common, benign finding and may be the result of the naturally happening atrophic process. 3,8,11 Goldstein 8 proposed that postmenopausal fluid collection is a transudate attributable to cervical stenosis. He has proposed that if the endometrial tissue surrounding the fluid is thin ( 3 mm), the endometrium is inactive, and sampling is not necessary. The aim of this retrospective study was to evaluate postmenopausal women with endometrial fluid collection and the risk of significant endometrial or cervical disease. Materials and Methods Before the onset of data collection, approval was obtained from the Institutional Review Board to perform a chart review of all postmenopausal women undergoing pelvic sonography for any reason at Jackson Memorial Hospital between July 1, 1997, and June 30, Postmenopausal women with endometrial fluid collection were identified, and a retrospective chart review was then performed with the use of medical records, emergency department records, and sonography records. The data were then entered into a database for ease of retrieval and filtering of data sets. Certified sonographers from our department performed all sonographic examinations. Endometrial stripe thickness was measured during transvaginal sonographic examination (Acuson Sequoia 512; Siemens Medical Solutions, Mountain View, CA). The endometrial thickness was measured on a sagittal view of the uterus and included both the anterior and posterior walls of the endometrial lining (double-wall thickness). In cases in which fluid collection was found within the endometrial cavity, the fluid was not included in the measurements of the endometrium. The endometrial fluid was assessed for echogenicity and quantity (subjective assessment). If a polyp or fibroid was noted, that structure was measured and reported. Once postmenopausal women found to have endometrial fluid collection on sonography were identified, the medical records were reviewed for whether the identified women underwent an evaluation of the endometrium with endometrial biopsy, hysteroscopy, or hysterectomy. Demographics, indication for the sonographic examination, endometrial stripe thickness, texture of the endometrium, echogenicity and quantity of the endometrial fluid, symptoms, history of cancer, and use of hormone replacement therapy (HRT) or tamoxifen were collected. Statistical techniques included independent sample t tests for the comparison of means of variables with normal or approximately normal distributions and Mann-Whitney tests for variables with skewed distribution. The Fisher exact test was used to calculate crude odds ratios (ORs). Multivariate logistic regression analysis was used to calculate the adjusted ORs. Results with P <.05 were considered statistically significant. Results During the 7-year period, 343 postmenopausal women were identified with endometrial fluid collection on vaginal probe sonographic examination. One hundred forty-two had undergone some form of evaluation of the endometrial lining, whereas 201 had no tissue evaluation of the endometrium because either these patients did not return for follow-up or the treating physician did not think that further evaluation of the endometrium was necessary. Of the 142 patients who underwent evaluations, 13 had insufficient tissue on endometrial biopsy for diagnosis. The remaining 129 women had adequate tissue evaluation of the endometrium by endometrial biopsy, hysteroscopy, or both, or hysterectomy. Of the 129 women, 25 had diagnoses of cancer or hyperplasia (nonbenign group) and 104 had benign findings (benign group). In the nonbenign group, 19 patients had cancer (11 adenocarcinoma and 8 squamous cancer) and 6 had hyperplasia (4 complex and 2 simple). Of the 11 patients with adenocarcinoma, 2 had adenocarcinoma of the endocervix and 9 had adenocarcinoma of the endometrium. Of the 8 patients with squamous cancer, 2 had squamous cell carcinoma of the endocervix and 6 had cervical cancer. In the benign group, 85 patients had an atrophic or inactive endometrium, 17 had polyps (9 symptomatic and 2 asymptomatic), and 2 had endometritis. Clinical and sonographic findings were compared between these groups, nonbenign versus benign (Table 1). Seventy-one women had pelvic sonographic examinations and endometrial sampling done secondary to postmenopausal bleeding, and 58 women had the sonographic examinations done 1478 J Ultrasound Med 2005; 24:

3 Takacs et al Table 1. Characteristics of Postmenopausal Women With Endometrial Fluid Collection Nonbenign Benign Crude OR Characteristic Statistic (n = 25) (n = 104) P (95% CI) Age, y Mean ± SD 62.4 ± ± 8.0 NS ND Endometrial stripe, mm Mean ± SD 9.9 ± ± ND Endometrial stripe >3 mm n (%) 23 (95.8) 76 (75.2) ( ) Endometrial stripe >5 mm n (%) 16 (66.7) 40 (39.6) ( ) Postmenopausal bleeding n (%) 17 (68) 52 (50) NS ND Moderate to large endometrial fluid n (%) 7 (28) 10 (9.6) ( ) Heterogeneous endometrium n (%) 16 (66.7) 50 (48.1) NS ND Echogenic endometrial fluid n (%) 11 (45.8) 5 (4.8) < ( ) HRT use n (%) 1 (4.2) 11 (11.6) NS ND Tamoxifen use n (%) 1 (4.2) 9 (9.7) NS ND History of nongenital cancer n (%) 3 (12) 19 (18.3) NS ND ND indicates not determined; and NS, not significant (P.05). secondary to others reasons (eg, pelvic pain and suspected adnexal mass). Of the 19 women with cancer, 12 had vaginal bleeding and 7 did not. Of these 7, 4 patients had echogenic fluid on sonography (Figure 1). There was no significant difference between the groups in age, rate of postmenopausal bleeding, texture of the endometrium (homogeneous or heterogeneous), HRT or tamoxifen use, or history of cancer. However, the endometrium was significantly thicker in the nonbenign group compared with the benign group (mean ± SD, 9.9 ± 7.4 versus 5.9 ± 4.1 mm; P =.016). Significantly more patients had moderate to large amounts of fluid in the endometrial cavity in the nonbenign group compared with the benign group (28% versus 9.6%; P =.02). In addition, echogenic fluid in the endometrial cavity was significantly more likely to be found in the nonbenign group compared with the benign group (45.8% versus 4.8%; P <.01). The rate of echogenic fluid in those patients who were lost to follow-up was even lower than the rate found in the benign group (1.5% versus 4.8%; P =.04). Multivariate logistic regression analysis was performed, introducing all significant and nearsignificant variables into the model (echogenic fluid, endometrial stripe thickness, moderate to large amount of fluid in the endometrial cavity, texture of the endometrium, and postmenopausal bleeding). The logistic regression analysis revealed that echogenic fluid in the endometrial cavity was the only significant risk factor for nonbenign disease (OR, 10.94; 95% confidence interval [CI], ; P <.01). None of the patients with adenocarcinoma of the endometrium had endometrial thickness of 3 mm or less, but 2 patients with squamous cell cancer of the endocervix had endometrial thickness of 3 mm or less. Discussion Widespread use of sonography has resulted in detection of endometrial fluid in many postmenopausal patients. The reported rate of endometrial fluid collection found on pelvic sonography in postmenopausal women varies from 4% to 18%. 1 4 Conflicting results have been reported regarding the prevalence of endometrial and cervical disease in postmenopausal women with endometrial fluid collection on sonography. Several investigators reported that endometrial fluid collection in postmenopausal Figure 1. Transvaginal sonogram from a postmenopausal woman with echogenic endometrial fluid collection. Endometrial biopsy revealed adenocarcinoma of the endometrium. J Ultrasound Med 2005; 24:

4 Echogenic Endometrial Fluid Collection in Postmenopausal Women women is a sign of serious endometrial or cervical disease, hyperplasia, or cancer. 5 7 However, other investigators reported a low incidence of significant endometrial disease in postmenopausal women with endometrial fluid collection. 2,8 10 Cervical stenosis has been suggested as a cause of endometrial fluid collection because it prevents the physiologic drainage of the uterine cavity. 12 In addition, Goldstein 8 proposed that postmenopausal endometrial fluid collection is a transudate attributable to cervical stenosis. He has proposed that if the endometrial tissue surrounding the fluid is thin ( 3 mm), the endometrium is inactive, and sampling is not necessary. If the endometrium is thicker than 3 mm, sampling is mandatory. Our findings are in agreement with the 3-mm cutoff for the purpose of ruling out endometrial cancer. However, 2 of our postmenopausal patients with endometrial fluid collection had endometrial thickness of 3 mm or less and later had a diagnosis of cervical (endocervical) cancer. Previously, it was reported by Epstein and Valentin, 13 on the basis of their personal opinion, that the presence of echogenic fluid should necessitate further evaluation irrespective of the endometrial thickness. On the basis of our study, we report that echogenic fluid collection in postmenopausal women is a significant risk factor for cancer, and patients should undergo a thorough evaluation of the endometrium to rule out cancer. In our study, we found that echogenic fluid in the endometrial cavity was the only significant risk factor for predicting serious disease in postmenopausal women with endometrial fluid collection (OR, 10.94; 95% CI, ; P <.01). We are fully aware of the limitations of a retrospective study and the fact that many of our patients did not have endometrial sampling. Most of our patients have no access to other medical facilities than our hospitals and clinics. We think that patients who did not have endometrial sampling soon after their sonographic examinations would have been identified later if endometrial or cervical cancer developed because we cross-matched the sonography database with the computerized pathology reporting system, and most of these patients did have a clinic visit at our institution but with a different specialty. In addition, the rate of echogenic fluid in those patients who were lost to follow-up was even lower than the rate found in the benign group. In summary, we recommend that postmenopausal women with endometrial fluid collection on sonography should undergo endometrial sampling if the endometrial lining is thicker than 3 mm or the endometrial fluid is echogenic. If the lining is 3 mm or less and the endometrial fluid is clear, endometrial sampling is not necessary, but we recommend endocervical sampling and a Papanicolaou test to rule out cervical (endocervical) cancer. References 1. Carlson JA Jr, Arger P, Thompson S, Carlson EJ. Clinical and pathologic correlation of endometrial cavity fluid detected by ultrasound in the postmenopausal patient. Obstet Gynecol 1991; 77: Gull B, Karlsson B, Wikland M, Milsom I, Granberg S. Factors influencing the presence of uterine cavity fluid in a random sample of asymptomatic postmenopausal women. Acta Obstet Gynecol Scand 1998; 77: Vuento MH, Pirhonen JP, Makinen JI, et al. Endometrial fluid accumulation in asymptomatic postmenopausal women. Ultrasound Obstet Gynecol 1996; 8: Epstein E, Ramirez A, Skoog L, Valentin L. Transvaginal sonography, saline contrast sonohysterography and hysteroscopy for the investigation of women with postmenopausal bleeding and endometrium > 5 mm. Ultrasound Obstet Gynecol 2001; 18: Nasri MN, Shepherd JH, Setchell ME, Lowe DG, Chard T. The role of vaginal scan in measurement of endometrial thickness in postmenopausal women. Br J Obstet Gynaecol 1991; 98: Breckenridge JW, Kurtz AB, Ritchie WG, Macht EL Jr. Postmenopausal uterine fluid collection: indicator of carcinoma. AJR Am J Roentgenol 1982; 139: McCarthy KA, Hall DA, Kopans DB, Swann CA. Postmenopausal endometrial fluid collections: always an indicator of malignancy? J Ultrasound Med 1986; 5: Goldstein SR. Postmenopausal endometrial fluid collections revisited: look at the doughnut rather than the hole. Obstet Gynecol 1994; 83: Seckin NC, Sener AB, Gozen A, Kutlay L, Cobanoglu O, Gokmen O. The importance of endometrial fluid 1480 J Ultrasound Med 2005; 24:

5 Takacs et al collection in postmenopause. Gynecol Obstet Invest 1996; 41: Brooks SE, Yeatts-Peterson M, Baker SP, Reuter KL. Thickened endometrial stripe and/or endometrial fluid as a marker of pathology: fact or fancy? Gynecol Oncol 1996; 63: Krissi H, Bar-Hava I, Orvieto R, Levy T, Ben-Rafael Z. Endometrial carcinoma in a post-menopausal woman with atrophic endometrium and intra-cavitary fluid: a case report. Eur J Obstet Gynecol Reprod Biol 1998; 77: Scott WW Jr, Rosenshein NB, Siegelman SS, Sanders RC. The obstructed uterus. Radiology 1981; 141: Epstein E, Valentin L. Managing women with postmenopausal bleeding. Best Pract Res Clin Obstet Gynaecol 2004; 18: J Ultrasound Med 2005; 24: