AJR:174, June 2000 1589 aginal bleeding is one of the most frequent reasons women seek health care [1], and the cause, usually be- nign, is often atrophy, polyps, or fibroids. Women undergo endometrial biopsy to exclude endometrial carcinoma, which has a reported in- cidence of 5–17% in perimenopausal and post- menopausal women with vaginal bleeding [2]. A variety of biopsy devices are available, but of- fice-based biopsy procedures are uncomfortable [3], are nondiagnostic in 50% of cases [4], and can miss focal mass lesions including polyps and carcinoma [5]. In the last 10 years, endovag- inal sonography has been widely used to evalu- ate the endometrium of women with vaginal bleeding [6], and saline-infusion hysterosonog- raphy has helped in the further evaluation of women who have abnormal endometrial find- ings on endovaginal sonography [7]. Hys- terosonography improves the specificity of endovaginal sonography by excluding focal ab- normalities when the endometrium appears thickened [8]. If endometrial biopsy could be performed using the hysterosonography cathe- ter, images of the endometrium and tissue speci- mens could be obtained under visual guidance with a single catheterization of the cervix. To- ward this goal, we tested a variety of catheters and biopsy devices to determine the feasibility of performing hysterosonographically guided en- dometrial biopsy through a single catheter. Materials and Methods With our human subjects committee’s approval, 14 women (age range, 32–65 years; mean, 48 years) with abnormal uterine bleeding consented to un- dergo hysterosonographically guided endometrial biopsy at the time of hysterosonography. The proce- dures were performed from March 1998 to March 1999 by radiologists and gynecologists in an outpa- tient treatment room located in the women’s clinic of Harborview Medical Center. All women underwent endovaginal sonography with a Prima (Siemens, Redmond, WA), an HDI 3000 (Applied Technologies Laboratories, Bothell, WA), or a Logiq 700 (General Electric Medical Systems, Mil- waukee, WI) scanner and 5-MHz or multifrequency endovaginal transducers. The double-layer endome- trial thickness was measured in the sagittal plane. Peri- menopausal and postmenopausal women with a thick- ened endometrium, defined as greater than 5 mm, and premenopausal women with an endometrium thicker than 16 mm underwent hysterosonography using one of the following catheters: a 9-French Selecta-Cath System cervical-access catheter (Ackrad Laboratories, Cranford, NJ), a 12-French balloon cervical cannula (Cook OB/GYN, Spencer, IN), or a 9- or 12-French balloon cervical-access prototype designed specifi- cally for hysterosonographically guided endometrial biopsy procedures (Cook OB/GYN). The first two cer- vical-access catheters were tested with and without the inner coaxial-guidance catheters (packaged for cannu- lation of the fallopian tubes). We placed these catheters in the cervix using the standard procedure described for hysterosonography [5]. Although many women had endometrial thick- ening on endovaginal sonography, the only women we included in this series were those with focal or diffuse endometrial thickening on hysterosonogra- phy, and it is these women who subsequently under- went guided biopsy. We did not attempt to biopsy focal mass lesions such as polyps or fibroids, and we did not include women with these findings in this se- ries. We gave no antibiotics prophylactically. We biopsied the endometrium by placing a 5- French flexible biopsy forceps (Cook OB/GYN) co- axially through the cervix. Each combination of catheter and biopsy device was evaluated for ade- quacy of control of uterine distention, visibility un- der sonographic guidance, steerability, and success in obtaining an adequate specimen. All women undergoing a biopsy received a paracervical block using lidocaine. After the hys- terosonography and biopsy procedure, the gynecolo- gist performed aspiration curettage using a 7-mm sterile vacuum curette (Synemed, Berkeley, CA) to completely evacuate the uterus. Endometrial hys- terosonographically guided biopsy specimens were placed in formalin containers separate from those obtained with the vacuum curette so the specimens could be compared. The curettage specimens served as the control against which the flexible-forceps specimens were compared. We administered con- scious IV sedation, including 1 mg of midazolam and 50 μg of fentanyl citrate, to all the women. Results Adequate specimens were obtained in 13 of the 14 women. Three specimens were obtained Hysterosonographically Guided Endometrial Biopsy: Technical Feasibility Theodore J. Dubinsky 1 , Susan Reed 2 , Connie Mao 2 , Gayle M. Waitches 1 , Eric K. Hoffer 1 Received August 30, 1999; accepted after revision October 13, 1999. 1 Department of Radiology, University of Washington, Harborview Medical Center, Box 359728, 325 Ninth Ave., Seattle, WA 98104. Address correspondence to T. J. Dubinsky. 2 Department of Obstetrics and Gynecology, University of Washington, Harborview Medical Center, Seattle, WA 98104. AJR 2000;174:1589–1591 0361–803X/00/1746–1589 © American Roentgen Ray Society Technical Innovation V Downloaded from www.ajronline.org by 52.73.204.196 on 05/17/22 from IP address 52.73.204.196. Copyright ARRS. For personal use only; all rights reserved