Applied Radiology

Drs. Feigin, Rosenblatt, and Papanicolaou are in the Department of Radiology at The New York Hospital- Cornell Medical Center, New York, NY and Dr. Kutcher is with the Department of Radiology at White Plains Medical Center, White Plains, NY.

A search for anatomic disorders of the uterus, fallopian tubes, and ovaries is often indicated in infertile women. Traditional diagnostic methods have included hysterosalpingography (HSG), hysteroscopy, abdominal and endovaginal sonography, and magnetic resonance imaging (MRI). In general, HSG and hysteroscopy are used for evaluation of the endometrial cavity and fallopian tubes, while sonography has been widely used for evaluation of the myometrium and ovaries. The introduction of sonohysterography (SHG) over the last several years has contributed significantly to the evaluation of all components of the female genital tract.

Each of the traditional imaging techniques of the female genital tract has distinct advantages and disadvantages. HSG is a relatively noninvasive and widely used procedure, however, it involves the use of ionizing radiation to the pelvic organs, requires the use of iodinated contrast material, and is relatively costly. In HSG, proper technique is crucial; inadequate uterine distension may result in poor visualization and excessive instillation of contrast may obscure uterine pathology. In addition, evaluation for intramural and extrauterine pathology is limited to those cases in which the uterine cavity is distorted. Hysteroscopy is generally considered the gold standard of endometrial evaluation. It is more invasive than the other techniques, however, requiring the risks of anesthesia and costs of an operating room. Sonography is able to demonstrate myometrial and adnexal pathology, but its ability to evaluate the endometrial cavity is limited. MRI can render detailed images of the uterus and is accurate in the diagnosis of congenital uterine anomalies ^1-3 and uterine pathology, ⁴ however its widespread use has been restricted by lack of expertise in pelvic MR imaging and by the fact that its advantages are not well known to referring clinicians and third-party payers. ^5,6

Sonohysterography (SHG), also commonly known as saline contrast hysterosonography, hysterosalpingosonography, or endohysterosonography, is transvaginal ultrasonography (TVUS) of the uterus enhanced by instillation of sterile saline (or another contrast medium such as galactose, dextran, glycine, lactated Ringer's solution, or water) into the uterine cavity. The saline distends the uterine cavity and acts as a contrast medium, thereby improving visualization of the endometrium. Thus, SHG combines the advantages of sonography with improved endometrial evaluation. Common indications for SHG include abnormal or postmenopausal uterine bleeding; ^7-15 the presence of uterine abnormalities such as polyps, fibroids, or adhesions; ^8,9 and the assessment
of tamoxifen-related endometrial changes. ¹⁶ This article will discuss ano-ther common and important role of SHG, the evaluation of female infertility.

The technique of SHG was first described by Richman et al ¹⁷ in 1984, who used it with transabdominal sonography to determine tubal patency. In the infertile patient, SHG is usually performed within the first 10 days of the menstrual cycle to minimize the possibility of disrupting an early intrauterine pregnancy and to utilize the backdrop of a peri-
ovulatory endometrium, against which many pathologic conditions are best demonstrated.

Generally, a baseline TVUS examination with a 5 or 7.5 MHz probe is performed first to demonstrate the thickness of the endometrium and myometrium and to evaluate the ovaries. The transvaginal probe is then removed and a sterile speculum is inserted. The cervix is cleansed with an antiseptic solution and a hysterosalpingography or insemination catheter, pediatric feeding tube, or Foley catheter, prefilled with saline to exclude air, is introduced into the uterine cavity. Some operators prefer the use of a cervical tenaculum to provide tension while inserting the catheter. The speculum is then removed and the transvaginal probe, covered with a condom and sterile gel, is inserted into the posterior vaginal fornix. Sterile saline is then injected into the catheter under continuous sonographic visualization until resistance is encountered or until the entire uterine cavity is observed clearly. The uterus is surveyed from cornu to cornu in a long axis projection. Then the transducer is rotated 90 ^o into a coronal plane and the uterus is surveyed from the fundus to the endocervical canal. In the normal uterus, the endometrium appears symmetric, surrounding the anechoic contrast-distended endometrial cavity (figures 1 and 2). ^18-24 The entire procedure takes approximately 15 minutes ^14,21,23,24 and is generally well tolerated. The use of prophylactic antibiotics is generally not necessary, ^18-21 and most hystero-sonographers see no need for analgesics or anesthesia, other than non-steroidal anti-inflammatory medications as needed. ^18,19,22

Structural abnormalities within the uterine cavity, such as leiomyomata or polyps, uterine adhesions, and Muellerian abnormalities, are believed to play a role in obstetric complications, such as primary infertility, implantation failure, premature delivery, and malpresentation. ^25-28 SHG has been shown to be useful in the detection of all of these anatomic disorders.

Congenital disorders

Muellerian duct abnormalities (MDA) result from total or partial failure of fusion of the paired Muellerian ducts from which the uterus and fallopian tubes develop. Most of these congenital disorders are believed to be asymptomatic: ²⁹
however, congenital uterine malformations have been associated with primary infertility, spontaneous abortion, preterm labor and increased perinatal mortality, fetal growth retardation, and malpresentation. Obstetric prognosis varies with different types of malformations, with septate and bicornuate uteri carrying the highest rate of pregnancy loss. ^15,30 Arcuate uteri are considered normal variants and are not associated with infertility or obstetric difficulties. ³¹ The rare uterus didelphys represents a complete lack of fusion of the Muellerian ducts resulting in two completely separate uterine cavities and two cervices. The unicornuate uterus is also uncommon and results from the development of only one Muellerian duct. Vaginal atresia is usually associated with an abnormal or absent uterus.

The treatment of MDA is dependent on the type of abnormality, with septate and bicornuate uteri usually considered surgically amenable. Until recently, both septate and bicornuate uteri were repaired by means of transabdominal surgery. However, outpatient hysteroscopic metroplasty is now the treatment of choice for the septate uterus, making the preoperative distinction critical.

Traditionally, hysterosalpingography and/or TVUS have been used to diagnose congenital uterine anomalies, relying largely on the relatively indirect measure of the angle of divergence of the uterine cavities to distinguish septate from bicornuate uteri. Sonohysterography has the distinct advantage over both HSG and TVUS of being able to visualize both the uterine cavity and myometrium in the nonpregnant woman. In addition, septal height and thickness may be determined directly by SHG. ^20,32

Early studies reported that TVUS is sensitive for detecting MDA, but its ability to distinguish between different types of anomalies is limited. ^33-35 Salle et al ³² studied 14 patients with infertility problems and an abnormal uterus, based on the results of HSG or TVUS. These investigators concluded that SHG was preferable to TVUS and HSG in evaluation of septate uteri, as it allowed clear visualization as well as measurement of the height and maximal thickness of the septum. SHG detected all 14 septate uteri, while TVUS and HSG each detected 8 of 14, the other 6 were ascribed to bicornuate abnormalities with each technique. Following surgical resection of the septa, confirmation of successful repair was more accurate by SHG than by hysteroscopy, as hysteroscopy could mistake the natural spur of the uterine fundus as a remainder of a septum. ³² Similarly, in a large study of 264 patients with septate uteri, Kupesic and Kurjak ³⁶ demonstrated superior sensitivity and specificity of SHG to TVUS in making the appropriate diagnosis.

Recent reports demonstrate that MRI is accurate in the diagnosis of MDA ^1-3 and is indicated when conventional imaging techniques are
nondiagnostic or when additional information is needed for appropriate patient management. ⁵

Leiomyomata

Leiomyomata are the most common neoplasms of the uterus. They consist of smooth muscle and varying amounts of fibrous tissue. They may be submucosal, intramural, or subserosal in location. The myomata that generally interfere with fertility are submucosal, as these may distort the endometrial cavity, interfere with proper nutrition and function of the endometrium, and/or block the cervical os or fallopian tubes. Additionally, large or numerous submucosal or intramural myomata may be associated with spontaneous abortion and preterm labor. Hysteroscopic myomectomy has become the treatment of choice for preserving fertility, and this procedure requires precise knowledge of the number, location, and size of submucosal myomas.

Typically, leiomyomata are diagnosed by HSG and TVUS, with hysteroscopy being the gold standard for submucosal tumors. Unless calcified, leiomyomata are detected during HSG only to the extent that they may enlarge and distort the cavity of the uterus. Submucosal myomata produce filling defects within the uterine cavity, which must be distinguished from air bubbles, polyps, prominent folds, and blood clots. It is important to evaluate early HSG images for filling defects, as many lesions can be obscured by large amounts of contrast in the uterine cavity. On sonography, the fibroid uterus usually appears large and lobulated, with myomata generally appearing isoechoic or hypoechoic (figure 3). MRI has also been shown to be highly sensitive in detecting uterine myomata and is considered superior to sonography in
differentiating fibroids from adenomyosis. ⁵ This may have significant therapeutic implications, because symptomatic adenomyosis often requires hysterectomy for relief of symptoms.

Many studies have demonstrated comparable or superior sensitivity of SHG to HSG, TVUS, and hysteroscopy in detecting myomata. ^{14,20,22-24,37-39} Hoezinger ³⁷ compared SHG to HSG in vitro. He found SHG superior to HSG in determining the presence, number, and location of myomata. One limitation of SHG was its inability to detect very small myomata, secondary to the limited resolution of the technique. Also, acoustic shadowing behind larger myomata may obscure smaller myomata within the shadow. Lindheim and Sauer ²³ studied 50 patients undergoing ovum donation and compared HSG to SHG, confirming pathology with hysteroscopy. SHG resulted in one false positive in which an intramural myoma was misinterpreted as having a significant submucosal component. Lindheim suggests that false positive results of SHG could be minimized by performing studies during the follicular phase 1 to 2 days upon completion of a spontaneous or progestin-induced withdrawal bleed. In this phase of the menstrual cycle, the endometrium is thin, which reduces the likelihood of over-
reading pathology. In another study, SHG was found to be superior to TVUS in localizing the myomata and superior to hysteroscopy in estimating their sizes and degree of submucosal and intramyometrial extension. ²⁴

Polyps

Endometrial polyps consist of endometrial tissue attached to the mucosa by a stalk. They generally appear as filling defects on HSG without enlarging or deforming the uterine cavity, in contrast to larger submucosal myomata. Polyps are seen on SHG as pedunculated, hyperechoic soft-tissue masses projecting into the endometrial cavity and surrounded by anechoic injected fluid, with the point of attachment and thickness of the stalk clearly demonstrated (figure 4). They are usually asymptomatic, however, they have been associated with vaginal bleeding and infertility. Polyps are treated by hysteroscopic excision followed by curettage.

As with myomata, the sensitivity of SHG in the detection of polyps is remarkably good. Syrop and Sahakian ¹³ studied 14 patients with polyps seen on SHG. All of these were confirmed histologically. Parsons and Lense ¹⁴ report that during TVUS, small polyps may blend smoothly into secretory or atrophic endometrium, producing only an asymmetric thickening in contour and large polyps may not be distinguished from normal secretory endometrium. Using SHG, they found that 16 polyps were easily delineated, appearing as smooth-margined, echogenic masses with relatively homogeneous echotexture. These polyps were all verified at surgery. Lindheim's series ²³ revealed excellent sensitivity of SHG (100%) compared to HSG (80%) in detecting 10 pathologically proven polyps. Other investigators have supported these findings as well. ^22,38

Uterine adhesions

Intrauterine adhesions, or synechiae, usually result from vigorous therapeutic curettage of the uterine cavity. The denuded endometrium heals by forming fibrous bands that may adhere to one another (Asherman's syndrome). Adhesions may cause primary infertility, recurrent abortion, or premature labor. They are usually treated by hysteroscopic-guided lysis. On HSG, points of adherence form angular or linear filling defects around which contrast flows. On SHG, adhesions appear as thick bridging bands or thin membranes of echogenic tissue that may distort the uterine cavity (figure 5).

Most studies comparing various imaging methods include only a very small number of patients with adhesions. ^14,23,38,39 Regardless, there appears to be a trend suggesting that SHG is a reliable method for diagnosing clinically significant adhesions. Parsons and Lense ¹⁴ diagnosed four patients with adhesions on SHG and confirmed all four by hysteroscopy. Ayida's series ³⁸ had two patients with adhesions on hysteroscopy, one of whom was diagnosed by SHG, while none were detected using TVUS alone. In the case in which the adhesions were missed by SHG, they were deemed clinically insignificant. Similarly, Fleischer et al ³⁹ found that SHG missed three cases in which there were minimal (<2 mm) adhesions.

Tubal pathology

Fallopian tube pathology is the cause of infertility in up to 40% of women. ⁴⁰ Infection, endometriosis, and pelvic surgery are the leading causes of tubal scarring. ⁴¹ Evaluation of tubal patency is usually undertaken by HSG or laparoscopic chromopertubation. With SHG, if increased fluid is noted in the cul-de-sac at the end of the procedure, tubal patency, at least unilaterally, can be inferred. By this criterion, Richman et al, ¹⁷ using transabdominal sonohysterography, dem-onstrated bilateral tubal occlusion with a sensitivity of 100% (9 of 9) and tubal patency with a reported specificity of 96% (24 of 25).

While others report similar success with indirect measures of tubal patency, ^21,39 most investigators agree that SHG is not as effective as HSG for defining tubal anatomic structure in detail. Diechert et al ⁴² required that flow into the tubal isthmus or ampulla be demonstrated directly in their study for the tubes to qualify as patent. As expected, many tubes could not be adequately visualized to apply this criterion. Bonilla-Musoles et al ⁴³ also used direct tubal visualization as their criterion for diagnosing tubal patency with SHG. They found that in only 18 of 37 patients with laparoscopically--or HSG--proven tubal patency, could SHG demonstrate the tubal lumens. Color Doppler, pulsed wave Doppler, and various contrast agents are
currently under investigation for improving visualization of fallopian tubes with SHG. ^{21,40,42,44,45}

Theoretical risks of SHG are infection and perforation, ¹⁰ although no cases of perforation have been reported. Most investigators similarly report no infection and no significant pain; however, mild cramping and vaginal spotting are not uncommon. ^18-20,22 Potential contraindications of the procedure have been described. Cervical stenosis may prevent introduction of the endocervical catheter. Cervical dilatation may occasionally be helpful in such cases. A patulous cervix may lead to transcervical fluid leakage, but this does not severely limit the study, as more fluid may be instilled or a catheter balloon inflated to tamponade the cervical canal.

Conclusion

SHG enhances transvaginal ultrasound examination of the uterine cavity and is a valuable adjunct to HSG in the radiologic evaluation of female infertility. SHG is performed easily and rapidly at relatively low cost, and is virtually devoid of complications. Its use in selected infertile patients obviates the need for more invasive diagnostic procedures, such as hysteroscopy.

Currently, the integration of MRI in the work-up of female infertility and its complementary role to HSG, TVUS, and SHG are being investigated. It is likely that the use of MRI will continue to expand as healthcare providers realize its potential benefits.

AR

References

1. Pellerito J, McCarthy S, Doyle M, et al: Diagnosis of uterine anomalies: Relative accuracy of MR imaging, endovaginal sonography, and hystero-salpingography. Radiology 183:795-800, 1992.

2. Mintz M, Thickman D, Gussman D, Kressel H: MR evaluation of uterine anomalies. AJR 148:287-290, 1987.

3. Carrington B, Hricak H, Nuruddin R et al: Muellerian duct anomalies: MR imaging evaluation. Radiology 176:715-720, 1990.

4. Schwartz L, Panageas E, Lange R, et al: Female pelvis: Impact of MR imaging on treatment decisions and net cost analysis. Radiology 192:55-60, 1994.

5. Ascher S: MR imaging of the female pelvis: The time has come. RadioGraphics 18:931-945, 1998.

6. Hricak H: Widespread use of MRI in gynecology: A myth or reality? Abdom Imaging 22:579-588, 1997.

7. Dubinsky T, Parvey HR, Gormaz G, Maklad N: Transvaginal hysterosonography in the evaluation of small endoluminal masses. J Ultrasound Med 14:1-6, 1995.

8. Di Naro E, Bratta F, Romano F, et al: The diagnosis of benign uterine pathology using transvaginal endohysterosonography. Clin Exp Obstet Gynecol 23:103-107, 1996.

9. Bronz L, Suter T, Rusca T: The value of transvaginal sonography with and without saline instillation in the diagnosis of uterine pathology in pre- and post menopausal women with abnormal bleeding or suspect sonographic findings. Ultrasound Obstet Gynecol 9:53-8, 1997.

10. Dubinsky T, Parvey H, Gormaz G, et al: Transvaginal hysterosonography: Comparison with biopsy in the evaluation of postmenopausal bleeding. J Ultrasound Med 14:887-93, 1995.

11. Dubinsky T, Parvey H, Maklad N: The role of transvaginal sonography and endometrial biopsy in the evaluation of peri- and postmenopausal bleeding. AJR 169:145-149, 1997.

12. Goldstein S: Use of ultrasonohysterography for triage of perimenopausal patients with unexplained uterine bleeding. Am J Obstet Gynecol 170:565-570, 1994.

13. Syrop C, Sahakain V: Transvaginal sonographic detection of endometrial polyps with fluid contrast augmentation. Obstet Gynecol 79:1041-1043, 1992.

14. Parsons A, Lense J: Sonohysterography for endometrial abnormalities: Preliminary results. J Clin Ultrasound 21:87-95, 1993.

15. Cohen J, Luxman D, Sagi J, et al: Sonohysterography for distinguishing endometrial thickening from endometrial polyps in postmenopausal bleeding. Ultrasound Obstet Gynecol 4:227-230, 1994.

16. Achiron R, Grisaru D, Golan-Porat N, Lipitz S: Tamoxifen and the uterus: An old drug tested by new modalities. Ultrasound Obstet Gynecol 7:374-378, 1996.

17. Richman T, Viscomi G, deCherney A, et al: Fallopian tubal patency assessed by ultrasound following fluid injection. Radiology 152:507-510, 1984.

18. Cullinan, J, Fleischer A, Kepple D, Arnold A: Sonohysterography: A technique for endometrial evaluation. RadioGraphics 15:501-514, 1995.

19. Goldstein S: Saline infusion sonohysterography. Clin Obstet Gynecol 39:248-258, 1996.

20. Alatas C, Aksoy E, Akarsu C, et al: Evaluation of intrauterine abnormalities in infertile patients by sonohysterography. Hum Reprod 12:487-490, 1997.

21. Tufekci E, Girit S, Bayirli E, et al: Evaluation of tubal patency by transvaginal sonosalpingography. Fertil Steril 57:336-340, 1992.

22. Gaucherand P, Piacenza J, Salle B, et al: Sonohysterography of the uterine cavity: Preliminary investigations. J Clin Ultrasound 23:339-348, 1995.

23. Lindheim S, Sauer M: Upper genital-tract screening with hysterosonography in patients receiving donated oocytes. Int J Gynecol Obstet 60:47-50, 1998.

24. Cicinelli E, Romano F, Anastasio P, et al: Transabdominal sonohysterography, transvaginal sonography, and hysteroscopy in the evaluation of submucous myomas. Obstet Gynecol 85:42-47, 1995.

25. Musich J, Behrman S: Obstetric outcome before and after metroplasty in women with uterine anomalies. Obstet Gynecol. 52:63-66, 1978.

26. Fedele L, Zamberletti D, Ferrazzi E, et al: Reproductive performance of women with unicornuate uterus. Fertil Steril 47:416-419, 1987.

27. Jones T, Fleischer A, Daniell J, et al: Sonographic characteristics of congenital uterine abnormalities and associated pregnancy. J Clin Ultrasound 8:435-437, 1980.

28. Thi Tho P, Rogers Byrd J, McDonough P: Etiologies and subsequent reproductive performance of 100 couples with recurrent abortion. Fertil Steril 32:389-395, 1979.

29. Sorenson SS: Estimated prevalence of Muellerian anomalies. Acta Obstet Gynecol Scand 67:441-445, 1988.

30. Buttram V, Gibbons W: Muellerian anomalies: A proposed classification (an analysis of 144 cases). Fertil Steril 32:40-46, 1979.

31. Tulandi T, Arronet G, McInnes P: Arcuate and bicornuate uterine anomalies and infertility. Fertil Steril 34:362, 1980.

32. Salle B, Sergeant P, Gaucherand P, et al: Transvaginal hysterosonographic evaluation of septate uteri: A preliminary report. Hum Reprod 11:1004-1007, 1996.

33. Valdes C, Malini S, Malinak L: Ultrasound evaluation of female genital tract anomalies: A review of 64 cases. Am J Obstet Gynecol 149:285-290, 1984.

34. Reuter K, Daly D, Cohen S: Septate versus bicornuate uteri: Errors in imaging diagnosis. Radiology 172:749-752, 1989.

35. Nicolini U, Bellotti M, Bonazzi B, et al: Can ultrasound be used to screen uterine malformations? Fertil Steril 47:89-93, 1987.

36. Kupesic S, Kurjak A: Septate uterus: Detection and prediction of obstetrical complications by different forms of ultrasonography. J Ultrasound Med 17:631-636, 1998.

37. Hoetzinger H: Hysterosonography and hysterography in benign and malignant diseases of the uterus: A comparative in vitro study. J Ultrasound Med 10:259-263, 1991.

38. Ayida G, Chamberlain P, Barlow D, Kennedy S: Uterine cavity assessment prior to in vitro fertilization: Comparison of transvaginal scanning, saline contrast hysterosonography and hysteroscopy. Ultrasound Obstet Gynecol 10:59-62, 1997.

39. Fleischer A, Vasquez J, Cullinan J, Eisenberg E: Sonohysterography combined with sonosalpingography: Correlation with endoscopic findings in infertility patients. J Ultrasound Med 16:381-384, 1997.

40. Collins J, Woodward P: Radiological evaluation of infertility. Semin Ultrasound CT MRI 16:304-316, 1995.

41. Yoder IC: Diseases of the fallopian tubes. In: Yoder IC (ed): Hysterosalpingography and Pelvic Ultrasound: Imaging in Infertility and Gynecology. Boston, Little, Brown and Co., 1988.

42. Deichert U, Schlief R, van de Sandt M, et al: Transvaginal hysterosalpingo-contrast-sonography (Hy-Co-Sy) compared with conventional tubal diagnostics. Hum Reprod 4:418-424, 1989.

43. Bonilla-Musoles F, Simon C, Serra V, et al: An assessment of hysterosalpingosonography (HSSG) as a diagnostic tool for uterine cavity defects and tubal patency. J Clin Ultrasound 20:175-181, 1992.

44. Deichert U, Schlief R, van de Sandt M, Daume E: Transvaginal hysterosalpingo-contrast-sonography for the assessment of tubal patency with gray scale imaging and additional use of pulsed wave Doppler. Fertil Steril 57:62-67, 1992.

45. Schlief R, Deichert U: Hysterosalpingo-contrast sonography of the uterus and fallopian tubes: Results of a clinical trial of a new contrast medium in 120 patients. Radiology 178:213-215, 1991.