Preoperative Evaluation of Submucosal Myoma by Virtual Hysteroscopy

Size: px

Start display at page:

Download "Preoperative Evaluation of Submucosal Myoma by Virtual Hysteroscopy"

Frederick Wells
6 years ago
Views:

1 Virtual Hysteroscopy Takeda et al Preoperative Evaluation of Submucosal Myoma by Virtual Hysteroscopy Akihiro Takeda, M.D., Shuichi Manabe, M.D., Satoyo Hosono, M.D., and Hiromi Nakamura, M.D. Abstract (J Am Assoc Gynecol Laparosc 2004, 11(3): ) Study Objective. To assess the utility of a new technique called virtual hysteroscopy in the evaluation of the size and location of submucosal myomas before hysteroscopic myomectomy. Design. Retrospective analysis (Canadian Task Force classification II-1). Setting. D epartment of gynecology at a general hospital. Patients. Thirteen consecutive women. Intervention. Sixteen-slice computed tomography (CT) scanner. Measurements and Main Results. Thirteen women with submucosal myomas were examined by virtual hysteroscopy. The lesions were filmed by multislice CT scanner, immediately after C O 2 injection into the uterine cavity with an intravenous dosage of iodide contrast media. The filmed image was subsequently reconstituted and analyzed by endoscopy mode and volume mode using three-dimensional computer graphics software. The size and depth of invasion of the submucosal myoma were clearly identified by the procedure. Conclusion. Accurate preoperative evaluation of the size and location of submucosal myomas before hysteroscopic myomectomy is important for a safe surgical procedure. Virtual hysteroscopy can provide such information with good reproducibility and is superior to previously described diagnostic procedures. Uterine myomas are a common, usually benign neoplastic disorder. 1 Only 5% 10% of myomas are estimated to be submucosal, but these lesions can induce severe clinical symptoms such as iron-deficiency anemia due to hypermenorrhea as well as metrorrhagia. Before the availability of hysteroscopic myomectomy, 2 submucosal myomas frequently required laparotomic excision or even hysterectomy. Although with technical progress and development of surgical instruments, hysteroscopic myomectomy has become a standard procedure for minimally invasive treatment of submucosal myomas, 3 serious complications such as uterine perforation have been noted. 4 Therefore, preoperative evaluation of the size and location of the myoma, especially the degree of protrusion into the uterine cavity, is essential 4 for a safe surgical procedure. Preoperative evaluation methods such as hysterosalpingography, 5 magnetic resonance imaging (MRI), 6 transvaginal ultrasound, 7 transvaginal saline sonohysterography, 8 and three-dimensional sonohysterography 9 were previously reported. However, these procedures have limitations in imaging due to various degrees of sensitivity, specificity, and reproducibility. 10 The rapid development of the multislice helical computed tomography (CT) scanner in recent years has facilitated the rapid acquisition of scans with very thin slices, 0.5 mm in thickness. 11 The reconstituted images of coronal and sagittal sections, which were impossible to obtain by 404 conventional CT, can be expressed by multiplanar reformation (MPR) technology. Furthermore, 3-D observation of organs has become possible from every possible arbitrary angle and direction. Virtual endoscopy 12,13 is a noninvasive technology used to display the image of the cavity inside the organ by processing MPR image data using 3-D computer graphics (3DCG) software as if one is observing the organ by real endoscopy. Materials and Methods Women with persistent hypermenorrhea and/or irregular vaginal bleeding were first screened by conventional transvaginal ultrasound, and 13 patients with possible submucosal myomas were enrolled in the present investigation. After explanation of the procedure, including the potential complications related to use of contrast agents and radiation exposure to the pelvis, informed consent was obtained. Virtual Hysteroscopy Filming with a 16-slice CT scanner (Aquilion 16 TSX-101A, Toshiba Medical Co., Tokyo, Japan) was performed after menstruation ended when the endometrium was thin. An intrauterine injection catheter equipped with a balloon (Hyscath, Sumitomo Bakelite Co., Tokyo, Japan) was inserted and fixed (Figure 1). After ml of CO2 From the Department of Obstetrics and Gynecology, Gifu Prefectural Tajimi Hospital, Tajimi, Gifu, Japan (all authors). Corresponding author Akihiro Takeda, M.D., Department of Obstetrics and Gynecology, Gifu Prefectural Tajimi Hospital, Maebata-cho, Tajimi, Gifu , Japan. Submitted January 19, Accepted for publication April 21, Reprinted from the JOURNAL OF THE AMERICAN ASSOCIATION OF GYNECOLOGIC LAPAROSCOPISTS, August 2004, Vol. 11 No The American Association of Gynecologic Laparoscopists. All rights reserved. This work may not be reproduced in any form or by any means without written permission from the AAGL. This includes but is not limited to, the posting of electronic files on the Internet, transferring electronic files to other persons, distributing printed output, and photocopying. To order multiple reprints of an individual article or request authorization to make photocopies, please contact the AAGL.

2 August 2004, Vol. 11, N o. 3 The Journal of the American Association of Gynecologic Laparoscopists FIGURE 1. The method of manually injecting CO2 into the uterine cavity by syringe. was injected slowly and manually by syringe, a simple image was taken to examine the degree of dilatation of the uterine cavity and the presence of lesions. Enhanced images were obtained after intravenous injection of iohexol, an iodide contrast media (Omnipaque 300, Daiichipharm Co., Tokyo, Japan). In addition, transvaginal saline sonohysterography was done immediately after CT, using the same intrauterine injection catheter. The obtained images were transferred to a computer workstation and reconstituted by 3DCG software (Virtual Place, Office Azemoto, Tokyo, Japan). Virtual endoscopic analysis was done by the endoscopy mode (surface rendering) and/or the volume mode (volume rendering) of the 3DCG software. The degree of protrusion of the myoma into the uterine cavity (cavity intramural ratio [CIR]) was measured on the image obtained by volume mode. The thickness of the myometrium adjacent to the myoma was measured. The shape and number of myomas also were evaluated. In several early cases, MRI was done for comparison with the image obtained by virtual hysteroscopy. Based on these findings, the surgical procedure was selected in each case, and the findings obtained during hysteroscopic resection were compared with the image by virtual hysteroscopy. Hysteroscopic Surgery An osmotic cervical dilator (Lamicel, Medtronic, Mystic, CT) was inserted into the cervical canal to dilate and ripen the cervix 2 3 hours before surgery. Using spinal anesthesia, hysteroscopic surgery was performed with a continuous-flow resectoscope (Karl Storz Endoscopy Japan, Tokyo, Japan) with an outer diameter of 8 mm using sorbitol distension medium (Uromatic S, Baxter Ltd., Tokyo, Japan) by gravity flow. Results The representative images of a submucosal myoma are shown in Figures 2 and 3. Figure 2A shows the direction of observation (triangle mark) by endoscopy mode on sagittal section of an MPR image. As shown in Figure 2B, the myoma protruding into the uterine cavity is observed clearly on the image reconstituted by endoscopy mode. In Figure 2C, the actual finding of the myoma during hysteroscopic myomectomy is shown for comparison, indicating the good reproducibility of the image reconstituted by endoscopy mode. Although the endoscopy-mode image provides some useful information about the portion of the myoma projecting into the uterine cavity, evaluation of the 405

Virtual Hysteroscopy Takeda et al FIGURE 2. (A) An observation viewpoint (triangle mark) in the sagittal section of a reconstituted multiplanar reformation image of a submucosal myoma.

3 Virtual Hysteroscopy Takeda et al FIGURE 2. (A) An observation viewpoint (triangle mark) in the sagittal section of a reconstituted multiplanar reformation image of a submucosal myoma. (B) Myoma protruding into the uterine cavity viewed by endoscopy mode. (C) The myoma observed during hysteroscopic resection. degree of protrusion and tumor diameter, which are important factors in selecting the surgical procedure, was difficult to perform. Again, as distension of the uterine cavity was limited in most cases, the submucosal myoma was observed only from restricted angle by endoscopy mode. Therefore, we decided to evaluate the lesions mainly by volume mode as shown in Figure 3. Figure 3A shows the MR image of the same submucosal myoma for comparison. Figures 3B, 3C, and 3D show the images of the myoma reconstituted by volume mode, which were observed from various angles (Figure 3B: cross sectional view from the superior; 3C: coronal sectional view from the posterior with some rotation; 3D: sagittal sectional view from the left with some rotation). These images obtained by volume mode were more informative than those obtained by endoscopy mode for assessing the size and location of submucosal myomas. When compared with two-dimensional MR images, the size and location of submucosal myomas were easier to comprehend in reconstituted 3-D image by volume mode. In Figure 3E, processing to emphasize the inner surface membrane of the uterine cavity was added to indicate that future progress in graphic treatment technology may provide more information about the lesions. 406

August 2004, Vol. 11, N o. 3 The Journal of the American Association of Gynecologic Laparoscopists FIGURE 3. (A) Magnetic resonance image of the submucosal myoma shown in Figure 2.

4 August 2004, Vol. 11, N o. 3 The Journal of the American Association of Gynecologic Laparoscopists FIGURE 3. (A) Magnetic resonance image of the submucosal myoma shown in Figure 2. (B) and (C) Reconstituted images of the same myoma obtained by volume mode observed from various angles: cross-sectional view observed from the superior (B) (note the presence of left ovarian cyst); and coronal sectional view observed from the posterior (C) with some rotation. (D) Sagittal sectional view observed from the left with some rotation. (E) Sagittal sectional image further processed to emphasize the inner surface membrane. In Figure 4A, the method of assessing of CIR is shown in the coronal sectional image reconstituted by volume mode. By the differential effect of iodide contrast medium between the weakly enhanced myoma and strongly enhanced adjacent myometrium, the margin of the myoma was clearly identified. Diameter (X = 42 mm) of the myoma and length protruding into the uterine cavity (Y = 32 mm) were directly measured on the image. The CIR was calculated to be 76% in this particular case. Myometrial thickness (Z) also is shown to be 9 mm. In our department, CIR was classified into three categories depending on the degree of protrusion (Figure 4B). Submucosal myomas that had CIR of more than 50% and tumor diameter below 50 mm were treated by one-step hysteroscopic myomectomy, while myomas that fell outside these criteria were treated by either twostep hysteroscopic myomectomy or laparoscopic-assisted myomectomy depending on the patient s choice. Table 1 summarizes the results of 13 cases analyzed in the present study. Pretreatment by gonadotropin-releasing hormone analog was not done in the patients treated in our hospital from the beginning, except for one woman who received leuprorelin acetate for 4 months at another hospital. In one woman, virtual hysteroscopy was done 1 week after uterine arterial embolization was performed to stop TABLE 1. Cases Evaluated by Virtual Hysteroscopy Myoma Myometrium CIR D iameter Thickness Surgical Age (%) (mm) (mm) Procedure H M H M 36 a H M H M H M+TLC LAM H M H M 34 b H M H M H M H M H M CIR = cavity intramural ratio; H M = hysteroscopic myomectomy; LAM = laparoscopic-assisted myomectomy; TLC = total laparoscopic cystectomy. a After treatment by GnRH analogue. b After treatment by uterine arterial embolization. 407

Virtual Hysteroscopy Takeda et al FIGURE 4. (A) The analytical method for determining the cavity intramural ratio (CIR) on coronal image reconstituted by volume mode.

5 Virtual Hysteroscopy Takeda et al FIGURE 4. (A) The analytical method for determining the cavity intramural ratio (CIR) on coronal image reconstituted by volume mode. X = myoma diameter, Y = protruding length into the uterine cavity, Z = thickness of myometrium adjacent to the myoma node. The CIR (%) = Y X 100. The presence of a subserosal myoma was also noted. (B) Classification of submucosal myomas according to CIR in our department. massive bleeding. The lesions were mainly investigated by volume mode rather than endoscopy mode for the reason described above. Twelve myomas with CIR of more than 50% were treated by hysteroscopic myomectomy without any complications. In one woman, total laparoscopic cystectomy for an ovarian endometrioma was simultaneously performed. Another woman, whose myoma had a CIR of less than 50%, was treated by laparoscopic-assisted myomectomy by her choice. Some patients complained of minor lower abdominal discomfort due to expansion of the uterine cavity during CO2 injection, but the symptom disappeared immediately after the examination. Infection-related symptoms such as fever or abdominal pain were not noted after this examination. Discussion Since the first clinical application of CT in Britain in the 1970s, considerable improvement has been made. Although CT had been very influential in diagnostic imaging, there was a period during which CT was considered less informative than MRI, which can provide images in various directions while CT can obtain images only in cross section. After the appearance of the single-slice helical CT scanner, which can film the body while a high-speed x-ray detector device spins spirally, continuous volume data used to make MPR images became available, and the utility of CT was recognized again. However, as sufficient space-resolving power still was not available, the images obtained by this device were less satisfactory for diagnostic use. In the late 1990s, clinical introduction of the multislice helical CT scanner 14 represented a significant advance in diagnostic imaging, as this device possesses a plural number of x-ray detectors in the z axial direction and can provide clearer 3-D images of arbitrary sections having space-resolving power equal to that of cross section. 15 The number of detectors has grown since then from 4 to 16, and the precision of the images has improved markedly. Furthermore, in the 16-slice CT scanner used in the present study, the turn speed of the detector itself increased approximately two-fold with a 16-times greater scan speed when compared with scanners with one row. Therefore, the scan speed of the device increased more than 30 times with this improvement, and clearer images can be obtained with a markedly decreased quantity of radiation exposure because of increased scan speed (company information, aquilion16.htm). Accurate preoperative evaluation of the size and location of the myoma is essential for safe hysteroscopic resection. 10 In particular, CIR is thought to be the most important factor 4 in hysteroscopic resection of submucosal myomas. Previously, the protrusion degree of a submucosal myoma 408

6 August 2004, Vol. 11, N o. 3 The Journal of the American Association of Gynecologic Laparoscopists has been evaluated by MRI, 6 transvaginal ultrasound, 7 and sonohysterography. 8 However, image information with sufficient reproducibility could not be obtained by those diagnostic means. In the human body, organs are constituted threedimensionally. With conventional diagnostic imaging procedures based on two-dimensional imaging, the twodimensional image is reconstructed into a 3-D image in the head of the image reader before diagnosis. Therefore, compared with the conventional diagnostic process based on two-dimensional imaging, the 3-D image reconstituted by virtual hysteroscopy can provide more information about the structure of body organs, and the diagnostic process can be simplified further. Because of better image quality with sufficient reproducibility provided by a multislice CT scanner, significant interobserver differences in diagnosis decrease, and a more-accurate diagnosis can be made. Furthermore, virtual endoscopic technology 12,13 is a noninvasive diagnostic imaging procedure that uses the reconstituted image obtained by a high-speed CT scanner. In virtual hysteroscopy, two kinds of processing techniques, endoscopy mode and volume mode, mainly are used for analyzing CT-based images. For digestive organs, it was reported that volume rendering was superior to surface rendering in terms of imaging minute structures of the mucous membranes. 16 However, at present, there is no consistent evaluation standard for this procedure, and the virtual endoscopy technique in general is thought to be at the trial-and-error stage of choosing an expression method in each case. In this series, the analysis of submucosal myomas was initially performed using both endoscopy mode and volume mode. However, since observation of the whole image of the lesion was difficult in cases showing poor expansion of the uterine cavity and CIR could not be measured on images reconstituted by endoscopy mode, we processed the data by volume mode as cases were accumulated. Furthermore, the 3-D image obtained by volume mode was demonstrated to be extremely useful for preoperative evaluation of lesions since observation in an arbitrary section became possible while freely rotating the image on screen by manipulating the computer mouse. Conclusion Virtual hysteroscopy can provide important 3-D information on the status of submucosal myomas with good reproducibility and is superior to previously described preoperative diagnostic methods. At the present time, there are no previous reports in the literature on the usefulness of virtual endoscopy technology based on images obtained by multislice CT for preoperative evaluation of gynecologic disorders, and this trial is still in the beginning stages. Further studies must be done to examine the usefulness of this procedure. References 1. Buttram VC Jr, Reiter RC: Uterine leiomyomata: Etiology, symptomatology and management. Fertil Steril 1981, 36: Neuwirth RS, Amin HK: Excision of submucosal fibroids with hysteroscopic control. Am J Obstet Gynecol 1976, 126: Corson SL, Brooks PG: Resectoscopic myomectomy. Fertil Steril 1991, 55: Taylor PJ, Gordon AG: Prevention and management of complications. In Practical Hysteroscopy. London, Blackwell, 1993, pp Siegler AM: Hysterosalpingography and laparoscopy in infertility. In A Manual of Clinical Hysteroscopy. Edited by RF Valle. New York, Parthenon, 1998, pp Hricak H, Tscholakoff D, Heinrichs L, et al: Uterine leiomyomas: Correlation of MR, histopathologic findings and symptoms. Radiology 1986, 158: Dodson MG: Use of transvaginal ultrasound in diagnosing the etiology of menometrorrhagia. J Reprod Med 1994, 39: Laifer-Narin SL, Ragavendra N, Lu DS, et al: Transvaginal saline hysterosonography: Characteristics distinguishing malignant and various benign conditions. Am J Radiol 1999, 172: Bonilla-Musoles F, Raga F, Blanes J, et al: Three-dimensional hysterosonographic evaluation of the normal endometrium: Comparison with transvaginal sonography and threedimensional ultrasound. J Gynecol Surg 1997, 13: Cheng YM, Lin BL: Modified sonohysterography immediately after hysteroscopy in the diagnosis of submucosal myoma. J Am Assoc Gynecol Laparosc 2002, 9: Hsieh J: Analytical models for multi-slice helical CT performance parameters. Med Phys 2003, 30: Oto A: Virtual endoscopy. Eur J Radiol 2002, 42: Wood BJ, Razavi P: Virtual endoscopy: A promising new technology. Am Fam Physician 2002, 66: Hu H, He HD, Foley WD, et al: Four multidetector-row helical CT: Image quality and volume coverage speed. Radiology 2000, 215: Boiselle PM, Dippolito G, Copeland J, et al: Multiplanar and 3D imaging of the central airways: Comparison of image quality and radiation dose of single-detector row CT and multi-detector row CT at differing tube currents in dogs. Radiology 2003, 228: Hopper KD, Iyriboz AT, Wise SW, et al: Mucosal detail at CT virtual reality: Surface versus volume rendering. Radiology 2000, 14:

Frequency of menses. Duration of menses 3 days to 7 days. Flow/amount of menses Average blood loss with menstruation is 60-80cc.

Frequency of menses. Duration of menses 3 days to 7 days. Flow/amount of menses Average blood loss with menstruation is 60-80cc. Frequency of menses 24 days (0.5%) to 35 days (0.9%) Age 25, 40% are between 25 and 28 days Age 25-35, 60% are between 25 and 28 days Teens and women over 40 s cycles may be longer apart Duration of menses