Chapter 37: TRANSVAGINAL SONOGRAPHY IN GYNECOLOGIC INFERTILITY

Infertility, the inability of a couple to achieve pregnancy with unprotected intercourse for 1 year, affects approximately 10% to 15% of couples and is increasing in incidence. The evaluation and treatment of infertility has undergone dramatic advances in the recent decade. Concurrently, transvaginal sonography (TVS) has a vital role in the evaluation and management of infertility related to a variety of gynecologic disorders.^1,2 Transvaginal ultrasound has revolutionized how we practice infertility over the last 2 decades.

Specifically, TVS has its greatest clinical applications in follicular monitoring and guided follicular oocyte aspiration. Transabdominal sonography (TAS) and TVS are also significantly involved in guiding embryo transfers and in the initial evaluation of infertility. The initial evaluation of infertility consists of the assessment of pelvis, uterine cavity, and tubal patency. This baseline pelvic sonogram identifies any existing pathology such as fibroids, congenital uterine anomalies, and adnexal masses of ovarian or tubal etiologies (ie, endometriomas, dermoids, other neoplasms, hydrosalpinges, etc) that may impact fertility. Additionally, this baseline evaluation has become very helpful in determining the ovarian reserve for infertility counseling and for guiding the ovulation induction dosing. Sonography with the instillation of fluid can determine the presence of uterine filling defects and can assess tubal patency. Follow-up of disorders that may be related to infertility, such as endometriosis, fibroids, and ovarian cysts, can be monitored with TVS. This chapter emphasizes the most frequently used applications and, in particular, stresses the role of transvaginal transducer/probes.

Transvaginal sonography is the method of choice for initial evaluation of the uterus and ovaries in most situations. The transvaginal transducer/probe allows a detailed depiction of the uterus and ovary because of the proximity of these structures to the transducer. In contrast to transabdominal scanning, however, transvaginal scanning displays images in nonconventional imaging planes. In addition, the regions of interest are limited to approximately 6 to 10 cm from the probe and do not afford as global a delineation of the pelvic structures in some cases as transabdominal scanning. TVS is typically performed with the bladder empty; in fact, a full bladder pushes the gynecologic structures away from view and may distort the anatomy. In TVS, surrounding bowel loops usually are not interposed between the probe and the adnexa. If they are, gentle manual abdominal palpation or manipulation of the probe, or both, can be used to displace adjacent bowel. Additionally, this technique can be used as an extension of the pelvic exam and may be coined as an "ultrasound-guided pelvic examination." This approach can help delineate whether adhesions are present between the uterus and the ovaries. The ovary should be mobile and free from the uterus. However, if the ovary is in contact with the uterus and abdominal palpation performed during the TVS does not move the ovary, then the ovary is likely adherent to the uterus. Sometimes this technique causes some mild discomfort for the patient if adhesions are present.

Appendix 37-1 lists the equipment and documentation guidelines, and guidelines recommended by the American Institute of Ultrasound (AIUM) for performance of transvaginal sonography in infertility patients. There are several types of transvaginal transducer/probes available, including curved linear array, phased array, and mechanical sector. The field of view of most of these probes is 100 degrees. The curved liner array probes afford a large sector field of view. Phased array probes may have resolution problems in the lateral aspects of the image due to side lobe artifacts. Color Doppler capabilities have been added to most transvaginal scanners. The addition of color Doppler sonography (CDS) does not significantly increase the thermal intensities used for imaging in most cases, and the color Doppler reveals vascular patterns. CDS affords specific selection of the vessels and areas to be interrogated by pulsed Doppler techniques. Flow can be assessed by waveform analysis, which reveals relative impedance to flow, and by overall assessment of the color shown in the area of the uterus, which is related to the relative vessel density of a particular area.

Three-dimensional (3D) sonography is available on some transvaginal ultrasound probes. Three-dimensional sonography is useful in the assessment of uterine malformations and fibroid mapping, and as an adjunct to sonohysterograms. Other applications are under investigation.

Infertility, as defined by the inability to conceive after 1 year of unprotected intercourse, affects approximately 10% to 15% of couples in the United States. Whereas 80% of couples who are between 18 and 28 years of age will conceive during a 1-year period and another 10% will the following year, 10% of couples in the United States (approximately 2.4 million couples) have a fertility disorder. In approximately two-thirds of these cases, the cause for infertility can be related to female factors; in one-third, to male factors; and about 25% of the time there are multiple factors. The most common gynecologic disorders related to infertility include occlusive tubal disease or endometriosis (30% to 50%), ovulation disorders (40%), and uterine and cervical factors (10%).³ In 10% to 15% of cases the cause of infertility is unexplained.^3,4

The initial evaluation of infertility includes an evaluation of each of these factors.^5,6 After the medical histories of the couple are reviewed and the physical examination is done, the testing usually includes a semen analysis, some evidence of ovulation via an ovulation predictor kit or mid-luteal progesterone level, an assessment of the uterine cavity and tubal patency (traditionally by hysterosalpingogram), and hormonal evaluation (thyroid-stimulating hormone [TSH], prolactin, and sometimes gonadotropins). In the early 1990s this initial evaluation also included a diagnostic laparoscopy to evaluate the pelvis for endometriosis and pelvic adhesions. However, with the improvement in sonographic technology and with more effective treatments such as ovulation induction with gonadotropins and in vitro fertilization (IVF), the diagnostic laparoscopy has dropped out of the initial infertility evaluation unless there is a medical indication (ie, pelvic pain, adnexal mass, etc).

Sonography has taken a greater role in both the evaluation and treatment of infertility. The most common indications for sonography in infertility include:

1. Baseline pelvic sonography

   1. Rule out adnexal and uterine pathology

   2. Assessment of ovarian reserve

2. Assessment of the uterus, endometrium, uterine cavity, and tubal patency

3. Serial monitoring of follicular development during treatment

4. Assessment of endometrial development⁷

5. Guided follicular oocyte aspiration

6. Guided embryo transfer⁸

7. Confirmation of intrauterine pregnancy (discussed further in Chapters 3 and 4)

Transvaginal sonography is occasionally used for guiding transluminal gamete intrafallopian tube transfer (GIFT) procedures and evaluation of endometriotic cysts.

Three-dimensional sonography is helpful in diagnosing congenital uterine malformations and in evaluating the location of intrauterine masses during the baseline sonogram.⁹ For greatest accuracy with a noninvasive sonographic technique, 3D sonography is best performed in the luteal phase when the endometrium, which is more hyperechoic than the myometrium, can act as an endogenous contrast. Additionally, a saline or sterile water infusion sonohysterogram can be performed with 3D sonography to aid in the mapping of identified intrauterine abnormalities in the early follicular phase. Recently, a new technological advance with 3D sonography was introduced by General Electric (GE), called "sonoAVC." It is a method of using 3D sonography with inversion mode imaging and additional software, to automatically measure all follicles within the ovary within a few moments. This technology is currently available. It may reduce the time to measure follicles and may more accurately measure follicles from day to day during stimulation, but it is too new to determine the impact on infertility centers, or whether this technology will really improve pregnancy outcomes.

Color Doppler sonography affords physiologic assessment of the ovaries and endometrium. Lack of the typical low-impedance, high diastolic flow within a corpus luteum, coupled with poor uterine or endometrial perfusion, may suggest luteal phase inadequacy.¹⁰ When endometrial abnormalities are noted on gray scale or on sonohysterogram, CDS may be able to identify a polyp or fibroid with its vascular pattern. Color Doppler sonography may also be helpful in the assessment of tubal patency because it is quite sensitive to the flow created when saline or contrast is injected through the tube.¹¹

Spontaneous Cycles

To better comprehend the value of the baseline sonogram in infertility, an understanding of the physiology of natural, spontaneous cycles is essential. At the time of birth, the female neonate is estimated to have approximately 2 million primary oocytes within each ovary. When menarche begins, approximately 200,000 oocytes remain per ovary. During the child-bearing years, approximately 200 oocytes will be ovulated. This indicates that approximately 99.9% of primary oocytes become atretic or do not develop at all. Maturation of the oocyte and follicle is responsive primarily to circulating levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), and circulating levels of estradiol (E₂). With the cascade of events initiated by elaboration and release of FSH in the late secretory phase, there is development of 1 and sometimes 2 dominant follicles in the subsequent spontaneous cycles. It is not uncommon to observe 2 follicles developing to approximately 10 mm in size, with one becoming dominant and growing and the other regressing. Estradiol is synthesized by the granulosa cells and provides important feedback to the pituitary in the production of FSH and LH. Elevated estradiol levels are thought to be part of the signaling process to trigger the LH surge. Luteinizing hormone reinitiates meiosis of the oocyte, and ovulation typically occurs within 36 to 40 hours of its "surge" in circulating LH levels. After ovulation, the remaining follicle becomes a corpus luteum. With the development of a corpus luteum, there is an increase in vascularity within the ovarian cyst wall and it can be observed on TV-CDS; some describe it as a "ring of fire."

Beginning with menarche, spontaneous cycles start with the development of a cohort of follicles, beginning at the time they measure between 3 and 5 mm. In the spontaneous cycle, there is usually 1 or at most 2 follicles that develop to approximately 10 mm in size. As the follicle matures, it enlarges an average of 1.5 to 2 mm in diameter per day due to more fluid collecting in its center in association with an accumulation of granulosa cells lining the inside wall of the follicle. The oocyte itself, which is smaller than 0.1 mm, is surrounded by a cluster of granulosa cells. This complex is termed the cumulus oophorus. It measures approximately 1 mm and can be depicted occasionally inside mature follicles (Figures 37-1 and 37-2B). As the follicle reaches maturity, its inner dimensions range from 17 to 25 mm.¹² Within the same individual, however, the size of a mature follicle is relatively constant from cycle to cycle. Intrafollicular echoes may be observed within mature follicles, probably rising from clusters of granulosa cells that shear off the wall near the time of ovulation. After ovulation, the follicular wall becomes irregular as the follicle becomes "deflated." The fresh corpus luteum usually appears as an echogenic structure approximately 15 mm in size, but may vary. Occasionally, the corpus luteum refills with serum and can look similar to a nonovulated follicle on gray scale.

On TV-CDS, folliculogenesis can be assessed by changes in impedance and velocity in intraovarian arterioles, and on 3D CDS and power Doppler with the percentage of volume showing a flow signal (VFS) and the perifollicular vascular network.¹³ With formation of a corpus luteum, impedance drops, a reflection of the vascular arcade that develops within the wall of the corpus luteum. Velocities increase up to 40 cm/s as the corpus luteum becomes functional.

In addition to delineation of changes in follicle size, morphology, and blood flow, sonography can depict the presence of intraperitoneal fluid (see Figure 37-1C) and endometrial changes (Figure 37-3). It is not uncommon to have approximately 1 to 3 mL in the cul-de-sac before ovulation. When ovulation occurs, there is typically between 4 and 5 mL within the cul-de-sac. When the patient is scanned with a fully distended bladder, the fluid may be located outside the cul-de-sac, surrounding bowel loops in the lower abdomen and upper pelvis. Although it is informative to learn about the sonographic changes of the natural or spontaneous cycle, it is not cost-effective to regularly monitor natural cycles on a daily or every other day basis to confirm ovulation or to time an intrauterine insemination treatment. TVS may be used more cost-effectively as an adjunct to ovulation predictor kits (OPKs) with natural cycles, particularly when there is ambiguity in the OPK result.

Besides the factors involved in obtaining a mature ovum, the developmental state of the endometrium may also be a factor that influences the probability that conception will occur (see Figure 37-3).¹⁴ Because the endometrium can be delineated on scans performed for follicular monitoring, several investigators have evaluated this specialized mucous membrane in an attempt to study whether or not there is an optimal thickness or texture.^{15,16,17, and 18} Clearly, there is an association of the texture of the endometrium as depicted sonographically and the circulating levels of estrogen and progesterone.¹⁴ In spontaneous and induced cycles, the sonographic appearance of the endometrium differs according to its specific phases of development. In the menstrual phase, the endometrium appears as a thin, broken echogenic interface. In the early proliferative phase, it thickens and becomes isoechoic and measures 3 to 5 mm in anterior–posterior width. Its relative hypoechogenicity is related to the relatively orderly organization of the glandular elements within the endometrium.

As ovulation approaches, the endometrium becomes more echogenic and thickens, probably related to development of secretions within the endometrial glands and the numerous interfaces that arise from distended and tortuous glands. A trilaminar appearance is typically described as the pattern in the late proliferative phase near ovulation. This pattern is described as a hypoechoic area within the most central portion of the endometrium, probably within the compactum layer with a hyperechoic line at the periphery and centrally. This finding has been described as a means of confirming ovulation has occurred. With TVS, however, we have observed this finding both before and immediately after ovulation. During the secretory phase, the endometrium becomes uniformly hyperechoic and maintains a maximal thickness (between 6 and 12 mm). Stromal edema also causes the endometrium to become more echogenic during the luteal phase. In addition to the echogenic endometrium, a hypoechoic band beneath the endometrium may be identified, probably arising from the inner layer of the myometrium.

The fact that medications used for ovulation induction may alter the development of the endometrium has been shown by both sonographic and histologic studies.¹⁶ The relative importance of these changes to the success or failure of achieving pregnancy is only speculative, however. One study, which evaluated the endometrial thickness in the secretory phase, has shown that conception was unlikely in endometria that measured less than 13 mm at 11 days postovulation.^17,18 Other studies have indicated that the texture of the endometrium during the proliferative phase may be related to the success or failure of pregnancy. Specifically, the presence of a multilayered endometrium around ovulation to within 1 to 2 days of embryo transfer was associated with a high postovulatory conception; yet, an endometrium too thin (<6 or <4 mm) around ovulation is associated with lower pregnancy rates.^19,20

Color Doppler sonography has shown that uterine blood flow can also predict pregnancy attainment.²¹ In one study, pregnancies did not occur in women with high-impedance (pulsatility index >3.0) uterine flow. Others have shown decreased pregnancy rates when intraendometrial spiral arterioles were not seen.²² Three-dimensional ultrasonographic endometrial volumes and vascular patterns may be important in the future to assess endometrial receptivity.²³

Given the above information regarding the natural changes of the pelvis during a cycle, one can get much information during the baseline evaluation. Many infertility specialists recommend doing the baseline evaluation during the initial follicular time period, such as around cycle day 3 because it is the most informative regarding the ovarian reserve. However, if one understands the basic physiology of the menstrual cycle and its corresponding sonographic changes, a baseline sonogram can be meaningful at any time the patient comes into the office for the initial evaluation of infertility. This approach may be more convenient for the patient than separate visits. Yet, insurance billing currently does not allow for separate billing of evaluation and management visits and sonographic procedural visits on the same day with the same diagnosis code. So at this time, a sonogram done at the time of the first visit may be done and thought of as an extension of the physical exam; it can be very informative for counseling, but not billed separately unless there is a separate indication for the procedure. The cycle day 3 sonogram may provide more information regarding the ovarian reserve, which is important for counseling and guiding treatment.

The components of the baseline sonogram for infertility are very similar to the components of a pelvic sonogram done at any time. The important components are to identify, measure, and describe the uterus and ovaries and any pathologic findings. The uterus tends to be very sensitive to estrogen. So an individual who presents with amenorrhea, whose sonogram shows a smaller uterus with a thin endometrial echo and small ovarian volumes, is suggestive of premature ovarian failure or other hypoestrogenic state. The sonogram of the uterus may reveal an abnormal shape or separate endometrial echoes on a transverse view of the uterus, and the addition of 3D sonography may confirm a uterine anomaly, such as a bicornuate or septate uterus. More commonly, fibroids may be detected within the uterus or distorting the uterine cavity, which may impact fertility and miscarriage rates, and cause other pregnancy complications.

Sonography can identify the presence of polycystic ovaries and adnexal masses, such as endometriomas, dermoids, hydrosalpinges, and tubo-ovarian abscesses that may be associated with gynecologic infertility (Figure 37-4). Follow-up sonograms may be used to document the progression or regression of endometriomas, ovarian cysts, or postsurgical recovery or recovery after drainage of an abscess.²⁴ Most cases of endometriosis consist of milder states that have mainly endometriotic implants that are small (approximately 5 mm or less) and are attached to parietal peritoneum or the serosal surfaces of bowel or other intraabdominal/pelvic organs or ligaments. These small lesions may not be detectable on sonography. Larger ones that are located separate from bowel can be delineated and followed during and after treatment.²⁵ Endometriosis may invaginate into the ovary and cause the formation of endometriomas. Endometriomas can be seen with sonography and diagnosed sonographically with about 90% accuracy when the classic sonographic findings of low-level echoes throughout the ovarian cyst and smooth borders are present.²⁶ The presence of an endometrioma automatically classifies the patient at stage 3 or higher on the American Society of Reproductive Medicine (ASRM) endometriosis classification system. This stage is associated with moderate endometriosis and is thought to have a significant impact on fertility.

If the TVS shows enlarged and rounded ovaries containing multiple immature, subcapsular follicles and increased amounts of stroma, then this pattern is suggestive of polycystic-appearing ovaries. Many sonologists use the Rotterdam criteria of more than 12 follicles in one or both ovaries that are less than 10 mm in size, or an ovarian volume greater than 10 cm³ to make the sonographic diagnosis of "polycystic-appearing ovaries." The polycystic ovary syndrome (PCOS) as described by the Rotterdam criteria consists of identifying 2 of 3 diagnostic criteria (anovulation or amenorrhea, evidence of hyperandrogenism by clinical or laboratory criteria, and sonographic findings consistent with polycystic-appearing ovaries) after ruling out other clinical scenarios that may have a similar appearance (thyroid disease, adrenal disorders, etc).²⁷ In up to 30% of women with this disorder, the ovaries may not be abnormally enlarged.^24,28 In these cases, it is also important to look at the endometrium. Patients with chronic anovulation, as in PCOS, are at increased risk of endometrial cancer. So if the endometrial echo is unusually thick, one needs to consider a further evaluation of the patient with an endometrial biopsy or hysteroscopy/dilation and curettage to obtain a sample for histology.

Dermoids, simple cysts, and paraovarian/paratubal cysts are very common in women of reproductive age. Occasionally, there is a more tubular cyst that may initially appear to be in the ovary. However, as one performs the TVS, changing position or changing pressure of the transvaginal probe or adding some abdominal pressure while TVS scanning, this tubular structure may appear separate from the ovary. One needs to consider whether a hydrosalpinx is present. Often the hydrosalpinx may have a tubular hypoechoic shape, incomplete septa, or "cogwheel" appearance that helps in making the diagnosis.²⁹ Hydrosalpinges imply occluded or diseased tubes that may be the cause of the couple's infertility and may put them at increased risk for ectopic pregnancy if conception occurs on their own (with a partially open tube) or with assisted reproductive technology (ART).

The infertility sonographic evaluation not only includes an assessment of normal or pathologic findings within the pelvis, but also a close assessment of the ovaries and a count of the antral follicles in each ovary.

Ovarian Reserve Assessment

As women age their ability to conceive decreases; the idea of a test for ovarian reserve assessment is to identify women who are unlikely to respond to therapy prior to starting treatment. There is no test available to date that can quantify the number of oocytes that a woman has remaining in the ovaries. However, there are several laboratory tests and sonographic criteria that can help in counseling couples regarding their likelihood of success. The most common laboratory evaluations consist of a cycle day 3 FSH level with or without an estradiol level. A high FSH and/or estradiol level around cycle day 3 is a poor test result and represents a likely poor response to fertility therapy. A more dynamic laboratory evaluation is a clomiphene challenge test. This test evaluates the FSH on cycle days 3 and 10 and the patient takes two 50-mg tablets of clomiphene citrate on cycle days 5 to 9. Typically, the 2 FSH values are added together and the cutoff value of this sum is determined by each laboratory. The cutoff value is the value above which no pregnancies have occurred. Other laboratory criteria are also being used either separately or together to determine the ovarian reserve, and these consist of inhibin B and anti-müllerian hormone.

The sonographic criteria for ovarian reserve testing consists of an assessment of ovarian volume and/or the number of basal antral follicles.³⁰ The ovarian volume may be obtained with either two-dimensional (2D) or 3D sonography. The ovarian volume is done in the early follicular time period (ie, around cycle day 3) to avoid the presence of an ovarian cyst, which would interfere with the results. A basal antral follicle count consists of the number of follicles in each ovary under the size of 10 mm. Some suggest that the presence of a follicle or cyst larger than 10 mm may impact this basal antral follicle count and make it less meaningful; others disagree. This antral follicle number does vary from cycle to cycle within women, but not significantly in most cases. The antral follicle count is done early in the menstrual cycle and can be predictive of therapeutic success. Typically, basal antral follicle counts under 3 to 5 bode poorly with treatment and those over 10 do well. However, the numbers in between tend to be less predictable. There appears to be a clear relationship between the decreased ovarian volume and antral follicle counts and advancing age and increasing cycle day 3 FSH levels. However, there are other factors that may affect the ovarian volume such as hormonal contraceptives, smoking, etc. CDS may also be helpful in predicting follicular response.

Applications of Color Doppler Sonography

TV-CDS can provide information regarding the flow within the ovary, uterus, and endometrium, as well as its use in assessment of the uterine cavity and tubal patency (Figure 37-5). Transvaginal CDS may be helpful to confirm the presence of small follicles and later of a dominant follicle.^31,32 CDS and 3S power Doppler may be used to assess vascular patterns of growing or dominant follicles.¹³ Another study reported that CDS could be used in the early follicular time period (cycle day 2 or 3) to assess ovarian reserve or follicular response to ovulation stimulation. This group reported on the correlation of the mean ovarian stromal peak systolic blood flow velocity with follicular response and IVF aspirated egg count, but not pregnancy outcome.³³ Lower velocities correlated with fewer follicles and, conversely, higher velocities with a higher number of follicles. Later in the cycle, transvaginal CDS can confirm the functionality of a corpus luteum by demonstrating characteristic low-impedance, high diastolic flow in vessels within its walls. In fact, corpora lutea that may not be apparent on gray scale may be detectable by their characteristic rim of vascularity, also referred to as the "ring of fire" (see Figure 37-1F).³⁴

Uterine blood flow can also be assessed with TV-CDS. It has been shown that higher impedance of uterine flow has a high association with infertility.³⁵ Transvaginal CDS of some equipment can assess vascularity within the endometrium. The presence of flow within the endometrium, which represents the spiral vessels, is a good sign that the endometrium is primed for implantation.²² Three-dimensional endometrial volumes are reproducible from cycle to cycle, and the endometrial vascular parameters and patterns may provide information regarding endometrial receptivity.²³

Another recently observed parameter that may have clinical significance is the depiction of "endometrial peristalsis" on TVS. This is particularly well seen if the examination is recorded on videotape and played back at fast forward. Endometrial peristalsis is directed toward the cervix during menses, but toward the fundus during the periovulatory period.³⁶ Stronger directional endometrial peristalsis may be associated with a higher implantation rate.³⁷

Infertility Assessment with Sonohysterography

Typically, the next evaluation for infertility has been the hysterosalpingogram (HSG), which is a radiologic test where iodinated contrast material is injected into the uterine cavity during fluoroscopy and x-rays are obtained (Figure 37-6). The HSG shows the contrast material within the uterine cavity and within the tubal canals and spilling into the peritoneal cavity when the result is normal. The HSG is done during cycle days 6 to 12, right after the menses is completed and prior to ovulation, to avoid early radiation exposure to an unsuspected pregnancy and to better evaluate the uterine cavity when the endometrial tissue is not too thick. TVS with saline infusion or sterile water infused into the uterine cavity can provide the same type of evaluation, without the radiation exposure. In addition, any filling defect detected during this procedure can typically be diagnosed with sonohysterography (SHG), unlike with HSG.

Some of the surgically correctable causes of infertility include intrauterine adhesions (synechiae), submucosal fibroids, and polyps. Each of these is detectable using SHG. Adhesions differ, from echogenic to hypoechoic. On SHG they appear as interfaces that course between endometrial layers. Some have a free edge. Submucosal fibroids may extend into the lumen and displace the overlying endometrium. These can be removed by wire loop resection if they do not extend into the outer layers of myometrium. Polyps may create a hostile environment for implantation and should usually be removed.

We advocate the use of SHG followed by sonosalpingography (SSG) for the evaluation of most infertility patients. The SHG portion evaluates the uterine lumen, and the SSG assesses the tubes (further discussion of tubal patency is in the next section). A bulbed/balloon-ended catheter or catheter with an acorn is recommended to block fluid from exiting out the cervix. This type of reflux would be detrimental to obtaining maximal distention of the endometrial lumen and subsequent evaluation of the tube. SHG is important in the assessment of the uterine cavity for filling defects and for the evaluation of uterine anomalies. Sonohysterography provides anatomic information concerning the relation of the lumen to the surrounding myometrium. Bicornuate uteri can often be distinguished from septated ones by delineation of the fundus. If there is a fundal cleft, bicornuate is more likely than septated uterus. This feature is particularly well delineated using 3D sonography, since the 3D allows for depiction of the fundus in the coronal plane.

In studies where SHG, HSG, and hysteroscopy are compared, SHG is more accurate in identifying the presence, size, and location of uterine filling defects as well as diagnosing the defect (polyp, fibroid, adhesion, etc).^38,39 More recently 3D sonohysterograms (3D-SHG) with contrast agents have been compared with x-ray HSG, and 3D-SHG also outperformed HSG in assessing the uterine cavity.⁴⁰

Additionally, intrauterine filling defects found on HSG or on SHG can be identified with the help of TV-CDS. A round filling defect that appears to be of similar echogenicity as the endometrium is likely to be a polyp. And TV-CDS may show the presence of a single vessel going to the center of this endometrial protrusion, confirming a polyp. Occasionally, one may find multiple vessels pushed to the side in a finding like this, making the diagnosis more likely to be a fibroid. Typically, fibroids are more hypoechoic than endometrium and exhibit shadowing, but these findings are not highly specific and the TV-CDS aids in the appropriate diagnosis.

Tubal Patency Assessment

Transvaginal sonography has been used to detect intraperitoneal spillage of saline or contrast injected into the uterus and tubes. These techniques currently are performed at select centers, and with time will likely be more widely performed. Some have used a specific contrast material consisting of a suspension of galactose monosaccharide microparticles (Echovist, Schering, Berlin, Germany; not US Food and Drug Administration [FDA] approved in the United States) and CDS as a means to assess tubal patency. The advantage of this technique includes avoidance of radiation, the potential to conduct the procedure in the gynecologist's office, and high tolerance by the patient. One study showed that sonographic evaluation of the tube was less painful than standard hysterosalpingography.⁴¹ Patency of at least one tube can be inferred when there is spillage into the cul-de-sac, whereas tubal obstruction may result in the filling of the uterine lumen without spillage. Often cramping is more prevalent in the latter situation.

In the last few years there has been great interest in the evaluation of tubal patency by sonographic methods. The first reported studies assessed tubal patency with transabdominal sonographic methods.^42,43 When they observed the collection of instilled fluid in the cul-de-sac, the investigators indirectly concluded that one of the fallopian tubes or both were patent. More recently, other investigators⁴⁴ have compared transvaginal sonosalpingography (TVSS) with chromopertubation by laparoscopy in patients with unknown tubal function. In their study, TVSS and laparoscopy were completely consistent in 29 cases (76.3%) and partially consistent in 8 cases (21.05%). Transvaginal sonosalpingography accurately showed patency in 26 patients and bilateral nonpatency in 3 patients. Thus, they concluded that TVSS is a safe and accurate screening and diagnostic technique in the evaluation of tubal patency.

Other investigators have used pulsed-wave (PW) Doppler to improve the sensitivity to flow through the tube after injected fluid.⁴⁵ They used transvaginal hysterosalpingocontrast sonography (THCS). They administered SH U 454 (Echovist) transcervically, then performed a THCS, and followed up with either chromopertubation with laparoscopy or hysterosalpingography. The diagnostic efficacies of gray-scale and PW Doppler were then compared with each other and against one of the conventional control procedures. Deichert et al⁴⁵ concluded that PW Doppler in THCS is recommended as a supplement to gray-scale imaging in cases of suspected tubal occlusion and if there is intratubal flow demonstrable only over a short distance.

Color Doppler and power Doppler have been studied to assess tubal patency. Stern et al⁴⁶ compared the results of color Doppler (CD) sonographic hysterosalpingography (US-HSG) and x-ray–HSG with chromopertubation at the time of laparoscopy. Saline was administered transcervically during TV-CDS in 238 women. Traditional x-ray–HSG was performed in 89 women, and laparoscopy and chromopertubation in 121 women. Forty-nine women had all 3 procedures performed. Correlation between CD US-HSG and x-ray findings with chromopertubation occurred in 81% versus 60% (P < .001) of all women studied. In the 49 women who had all 3 procedures, CD US-HSG results correlated with chromopertubation more often than x-ray–HSG (82% vs 57%; P < .05). Thus, color Doppler US-HSG appears to be an efficacious method of establishing fallopian tube patency in patients with infertility. Sladkevicius et al and Kiyokawa et al have evaluated the use of 3D sonography with contrast or power Doppler imaging to assess tubal patency and found advantages. Three-dimensional power Doppler imaging (3D-PDI) was compared with contrast SHG (using hysterosalpingo-contrast [HyCoSy]) for tubal patency. Color-coded 3D-PDI with surface rendering enabled the visualization of the entire length of the fallopian tube and documented free spill in most cases.^40,47 Advantages of this approach include that it can be performed in an outpatient setting, and it has high patient acceptance and no radiation exposure.

Future infertility therapeutic approaches that may involve TVS include early tubal screening in the diagnosis of infertility, visualization and documentation of transcervical tuboplasties, and gamete or embryo tubal transfers following transvaginal ultrasound-guided ova retrievals.

Sonosalpingography (SSG) is a technique that uses TVS to evaluate tubal patency and morphology (Figure 37-7). It involves the use of either saline or a contrast media for assessment of tubal patency. As noted, several multicentered studies have indicated that it is as accurate as the established techniques for assessment of tubal patency, such as HSG.^48,49 This subsection describes the current use of SSG and its limitations and future refinements.

Technique

Sonosalpingography is an addendum to SHG with the instillation of either saline or contrast media into the uterine lumen during TVS. It is best performed after sonohysterography (SHG) because SHG is vital in delineating the endometrial surfaces for the presence of synechiae or intraluminal lesions such as polyps or submucosal fibroids.^50,51 Another advantage of performing SHG before SSG is that the fluid introduced by SHG tends to surround the ovary, giving a hypoechoic background for assessment of the echogenic contrast, which is expressed from a patent tube after intrauterine injection (Figure 37-8). The mobility of the tube and ovary can be assessed during the examination by observing the motion of the ovary and tube while gentle pressure is applied by the probe and/or the examiner's palpating hand on the lower abdomen.

The study must be performed in real time, and may be recorded on videotape or digital format. The examiner is encouraged to apply gentle pressure with the probe while sonographically assessing the mobility of the uterus, ovary, and tubes. Adhesed tubes or ovaries do not have a normal excursion upon ballottement with the transducer.

Saline can be initially used for assessment for tubal patency using an agitated saline technique. However, if there is any doubt of tubal spill, contrast should be used. It is necessary to have a bulbed/ballooned catheter or a catheter with an acorn to block cervical egress of fluid from the uterine lumen during injection. After SHG is performed, the saline should be removed before instillation of positive contrast.

We have found that in about one-half of the patients in our series, tubal patency can be assessed by repeated injections of 3 to 5 cc of sterile saline. However, contrast afforded accurate assessment of tubal patency in almost (92%) of the patients.⁵² Therefore, when there is no definite spillage or when the initial findings are equivocal, a positive contrast is needed to establish tubal patency.

Some have advocated the use of CDS with saline instillation to best depict tubal patency.⁵³ This technique can also be used with contrast. Preliminary work suggests that three-dimensional and harmonic imaging may enhance sonographic depiction of the tube (see Figure 37-8D).

Albunex has been used for multiple applications including evaluation of tubal patency. It provided an echogenic contrast to document tubal patency and spillage. The safety of Albunex has been established in multiple studies, but its cost ($83/10-cc vial) makes selected use necessary. Albunex consists of microbubbles suspended in human albumen. It must be stored and used properly for its best effect. The production of Albunex has been discontinued. However, a similar contrast agent (Optison®) is now available and has been approved by the FDA for cardiac use. Another contrast material consisting of 1- to 3-μm microbubbles (Definity) is commercially available (Lantheus Medical Imaging, Billerica, MA, USA). It is also not approved for general use by the FDA but can be used off label.

Normal and Abnormal Anatomy

Before contrast or saline is introduced, it is recommended that the sonographer confirm the presence or absence of cul-de-sac fluid or fluid around the ovaries. Then the sonographer should identify the approximate location of the tube by identifying the ovary and endometrium that invaginates into the uterine cornua (see Figure 37-7B). Once this scan plane is established, contrast spillage can be followed around the ovary because the fimbriated end of the tube is normally near the ovary. In some cases, the contrast can be "chased" by reinjection of 5 to 10 cc of saline to maximize tubal delineation.

If there is pain in the cornual regions during the baseline ultrasound or a history of hydrosalpinx, we recommend pretreating with antibiotics (such as doxycycline) prior to performing the SHG. However, if the pelvic exam or baseline sonogram revealed no tenderness prior to performing the SHG and there is pain during injection of saline or contrast, this may be a sign of tubal obstruction, either intrinsic or extrinsic adhesions, or compression. Spasm may be present and cause transitory lack of filling of the proximal portion of the tube (Figure 37-9). Sonosalpingography is associated with less discomfort than HSG.⁴¹

With the use of saline or contrast, the normal tubal lumen is easily identified as a thin (approximately 1 mm) serpiginous adnexal structure. The motion of saline and/or contrast also helps identify it particularly when CDS is used.

Hydrosalpinges appear as fusiform cystic structures on TVS (Figures 37-10 and 37-11). Other characteristic sonographic findings include tubular structure, incomplete septa, and small clumps around the internal periphery of the cystic structure described as a "cogwheel" appearance. Occasionally, the contrast may become diluted or bubbles may come out of suspension, making complete delineation difficult to recognize (see Figure 37-10). If a hydrosalpinx is seen, patients may be given 200 mg of doxycyline initially and then 100 mg orally twice a day for 5 days for prophylaxis against exaggeration of preexisting (dormans) pelvic inflammatory disease.

Comparison With Hysterosalpingography

A large multicentered study in Europe has indicated that SSG with Levovist (which is a suspension of microbubbles in a galactose solution) is as accurate for assessment of tubal patency as HSG.⁴⁸ In many cases, SSG is preferable because it does not produce significant pain during the procedure, and does not require fluoroscopic evaluation. One limitation, however, is the inability of SSG to confidently diagnose salpingitis isthmica nodosa. If this condition is suspected, HSG is the preferred modality for delineation of the anatomic detail. Salpingitis isthmica nodosa is associated with multiple diverticule, usually the result of chronic infection. Also, HSG is preferred for determining the amount of proximal fallopian tube present after a bilateral tubal ligation when a patient wants a tubal anastamosis procedure.

Other problems can occur with contrast use. One is difficulty in establishing tubal patency in patients with obstructing lesions such as cornual fibroids. Another limitation is that the contrast can become diluted in hydrosalpinges that contain a significant amount of fluid. And finally improper contrast storage or faulty administration of the contrast can result in less than optimal echogenicity. It is rare to have a reaction to the contrast material.

Sonosalpingography Findings

In patent tubes, contrast can be seen exiting the fimbriated end of the tube, which usually surrounds and is adjacent to the ovary (see Figure 37-8). Occasionally, fimbria can be seen floating in the saline or contrast material. Fibroids may limit visibility due to shadowing on the ultrasound or may obstruct flow within the tube, as well as intraluminal processes due to adhesions or previous surgery (see Figure 37-11). True obstruction is associated with lack of egress of saline or contrast from the fimbriated end of the tube after an appropriate length (15 to 20 minutes) of observation (Figure 37-12). Hydrosalpinges appear as fusiform structures that are filled with fluid. Occasionally, the tubal endosalpingeal folds can be recognized (Figure 37-13).⁵⁴

One must be cautious in diagnosing tubal obstruction versus tubal spasm. In some patients, distension of the uterus with saline or more commonly with the introduction of contrast material may be associated with tubal spasm and a follow-up examination after the spasm relaxes may indicate tubal patency.

Some advocate the use of saline with CDS. This technique provides detection of motion of fluid or contrast within and out of the tube.⁵⁵ In our series, contrast was needed in about half of the cases to ensure that the initial sonographic evaluation with saline for tubal patency was accurate.⁵²

Sonosalpingography provides accurate assessment of tubal patency. In addition, the mobility and relative position of the tube and ovary can be assessed. Its diagnostic accuracy is comparable to HSG and chromopertubation, and can be performed in an office setting. It is our expectation that sonosalpingography will become the initial study of choice for evaluation of tubal patency, with HSG reserved for specific conditions such as salpingitis isthmica nodosa or presurgical evaluation of the proximal fallopian tube length prior to tubal reanastomosis. Transvaginal sonography can also be used for selective tubal dilatation by ensuring proper location of the catheter tip.

Sonographic Assessment of the Intrauterine Lumen

Sonohysterography is being used more extensively for assessment of intraluminal disorders, such as endometrial polyps, submucosal fibroids, intrauterine synechiae, and uterine malformations. The procedure uses a thin, flexible catheter that is placed through the cervix into the lumen with its tip near the fundus. Other catheters with a balloon near the tip are designed for placement in the cervical canal, which usually causes less discomfort and less interference with assessing the uterine cavity during this procedure. Slow injection of 3 to 10 mL of sterile saline usually produces adequate distention of the lumen for the assessment of endometrial polyps, submucous fibroids, or synechiae (see Figure 37-5). This procedure should be performed in the follicular phase of the menstrual cycle when the endometrium is thin and the most echogenic. Endometrial polyps and submucosal fibroids appear as echogenic structures that project into the lumen where synechiae typically are echogenic linear interfaces.

The addition of CDS will allow for the differentiation of certain diagnoses, such as polyps and fibroids. Typically, a polyp will have a single central vessel and an echogenicity similar to endometrium, whereas a fibroid will have the blood vessels displaced to the periphery of the fibroid and may have some shadowing. This distinction is important since it has an impact on the surgical approach. Polyps tend to be soft and can be easily removed with a direct visualized hysteroscopic biopsy, graspers, or excision with scissors. However, fibroids typically are much firmer, and it is important to realize which type of fibroid is present. (Type 0, completely in the uterine cavity; type 1, at least 50% of the submucosal fibroid is within the uterine cavity; and type 2, more than 50% of the fibroid is in the myometrium. It is also important to evaluate how close type 2 fibroids are to the serosal surface.) For a type 0 submucous fibroid, the surgeon should set up a resectoscopic procedure, often with a fluid management system. For a type 1 submucosal fibroid, the procedure may be done similarly, but it may be prudent to also have the potential to add a concurrent laparoscopy to avoid complications, such as uterine perforation, particularly in less experienced hands. Type 2 may not only require all that is done for type 1, but may also take more than one procedure with the resectoscope over several months, or the surgeon may choose to use an abdominal approach for a single procedure.

Three-dimensional sonography is very helpful in assessing the uterine lumen and is better in mapping the location of intraluminal pathology (polyps, fibroids, adhesions). This mapping may be helpful in planning surgery. Additionally, 3D ultrasound can be used to delineate distended fallopian tubes.

This procedure requires a balloon-tipped catheter, which is inflated with saline and placed at the internal cervical os, or a cone-tipped catheter placed flush at the external cervical os to prohibit retrograde flow, and to encourage flow into the proximal tube. As 10 to 15 mL of saline is injected into the endometrial lumen, the uterus is automatically swept with 3D sonography and images are obtained as well as multiplanar reconstruction. The volume can then be viewed in any selected scan plane depending on the area of interest. This technique can also be done with 2D ultrasound; however, it is critical to actively scan the uterus in the longitudinal and transverse planes systematically. To evaluate tubal patency, the cornual areas need to be observed in real-time 2D, and bubbles from agitated saline need to be seen exiting the cornua, with free fluid in the cul-de-sac.

Over the past 10 to 15 years, there have been significant advances in the medical treatment of infertility, as well as more refined techniques for the surgical management of these disorders. Improvements in infertility medications, culturing techniques for in vitro fertilization (IVF), and imaging have led to higher pregnancy rates. TVS has become a more important tool in infertility treatments. It is utilized for follicular monitoring and ultrasound-guided procedures (such as follicular aspiration of oocytes and embryo transfer).

Treatment is determined by the diagnosis. Anovulation or other ovulatory disorders are treated with ovulation induction. Mild male factor and unexplained infertility are initially treated with superovulation with intrauterine inseminations. Uterine disorders (ie, synechiae, submucosal myomas, and uterine septums) are treated surgically. Tubal disorders are either corrected surgically or bypassed using IVF. Additionally, couples with severe male factor contributing to their inability to conceive may go directly to IVF with intracytoplasmic sperm injection (ICSI) if sperm are available. If milder therapy is successful in ovulation induction response, but not successful in pregnancy, then the couple may also consider IVF therapy.

The most common medical treatments for ovulation induction consist of essentially 3 classes of medications: clomiphene citrate (CC), aromatase inhibitors, or gona-dotropins (both urinary and recombinant). Although these medications may result in the development of multiple follicles, they act by different mechanisms. Of note, aromatase inhibitors are not currently approved by the FDA for indication of ovulation induction. Sonography has a vital role in monitoring follicular development in women receiving these ovulation-induction medications.⁵⁶

The first 2 classes are oral medications. Clomiphene citrate is considered an estrogen agonist antagonist and exerts its effect by binding estrogen receptor sites in the pituitary and hypothalamus. This in turn influences the pituitary to synthesize more FSH, thereby stimulating the ovary to recruit more than one follicle. One to five follicles may become mature and ovulate on CC therapy. Most commonly 2 to 3 oocytes ovulate on CC therapy.

Aromatase inhibitors also have an indirect method of stimulating the follicles to grow to maturity and ovulate, and with this medication the average response is 1.5 mature follicles that ovulate. The pregnancy rates tend to be similar to CC therapy with similar or slightly lower multiple gestation rates. Both of these oral methods for ovulation induction work through indirect mechanisms and require an intact and functional pituitary in order to be effective.

As opposed to CC, treatment with gonadotrophin does not require an intact hypothalamus or pituitary. Only gonadotropin therapy directly stimulates the ovary with FSH, with or without a small amount of LH. Gonadotropins are injectable medications that are given on a daily basis and can be adjusted to tailor the response to the individual in that particular cycle. Gonadotropin medications may be given with or without a second medication to block the pituitary's LH surge, such as gonadotropin-releasing hormone agonists (ie, Lupron [TAP] Pharmaceuticals, Deerfield, IL, USA) or gonadotropin-releasing hormone antagonists (ie, Ganirelix, Organon, a part of Schering Plough, Newhouse, UK, or Cetratide, Merck Serono International, Geneva, Switzerland).

Sonography has a vital role in depicting follicular development in patients treated for infertility who receive ovulation-induction medication.⁵⁷ Although the maturity of the oocyte can only be indirectly inferred by the size of the follicle, the sonographic information can be coupled with estradiol values to provide an accurate assessment of the presence or absence, as well as the number of mature follicles.^14,58,59 The anatomic information obtained with sonography concerning the size and development of maturing follicles and corpora lutea can be used to distinguish physiologic from insufficient or abnormal cycles.⁶⁰ For example, the maximal follicle size in insufficient cycles has been reported to be significantly smaller than that in normal ones, and the absence of a corpus luteum has been found more often in insufficient cycles.⁶⁰ In addition, the undesirable development of multiple immature follicles rather than development of a single dominant follicle can be recognized in patients with polycystic ovaries.⁶⁰

Although its actual contribution to infertility is controversial, some infertility specialists describe an abnormality in ovulation, termed luteinized unruptured follicle syndrome, as a cause of unexplained infertility.⁶¹ In this disorder, there is failure of ovulation of the oocyte, which remains trapped within the follicle. The presence of this abnormality can only be confirmed by observation at laparoscopy of the absence of a stigma (healed rent in the ovarian capsule where ovulation occurred) on the ovarian capsule, and/or by failure to identify elevated peritoneal fluid steroid level, alter presumed ovulation. Because this stigma of ovulation may no longer be present as early as 2 hours postovulation, the presence of this syndrome is difficult to confirm. With sonography, one can observe failure of the follicle to deflate and the absence of intraperitoneal fluid associated with ovulation or, theoretically, a luteal cyst could be aspirated and yield an oocyte. This syndrome may be more common in women with endometriosis and may not be present on consecutive cycles.^56,62

Abnormalities in folliculogenesis can be further assessed with TV-CDS. This technique depicts physiologic changes in flow impedance and velocity associated with follicular enlargement and corpus luteum formation. A significant drop in impedance occurs with the development of the vascular "arcade" that develops within the follicular wall of a corpus luteum. In addition, a slow increase in velocity and blood flow within the intraovarian arterioles occurs during folliculogenesis. This steady increase is interrupted in patients with luteinized unruptured follicles.

Follicular Monitoring by Type of Ovulation Induction

TVS has an important role in documenting both normal or abnormal follicular development.⁶³ Although TVS is preferred for follicular monitoring, transabdominal scanning is sometimes helpful when the ovaries are located high in the pelvis. For indirect methods of stimulating the ovaries, oral medications (CC or aromatase inhibitors) are given for 5 days in the cycle (usually cycle days 3 through 7 or 5 through 9), and a midcycle TVS (usually between cycle days 11 to 14) will document the follicular response.

Follicular development with CC can be quite different than that observed in spontaneous cycles (see Figure 37-2). Specifically, each follicle seems to develop at an individual rate and at times may be accelerated or slowed down. Therefore, the largest follicle on a given date may not be the same one that is the largest 2 days later, and it may not even be the same one that is most mature. Furthermore, correlation of E₂ and follicle size is poor, and the maximum preovulatory diameter can range from 19 to 24 mm. TVS can be used alone or in combination with other tests such as an ovulation predictor kit, which measures urinary LH to better define the time of ovulation and timed intercourse or insemination.⁶⁴ Some patients are given human chorionic gonadotropin (hCG) to induce final oocyte maturation and ovulation when the follicles are mature (usually 18 to 20 mm in average diameter). This hCG trigger may also assist in timing for the intrauterine insemination. Unfortunately, these indirect medications can not be modified in dosage during the treatment cycle to effect a change in follicle growth and number, but the dose can be adjusted in the following cycle.

Gonadotropin therapy is an ovulation-induction medication that can result in the development of multiple follicles. Sonography also has a vital role in monitoring follicular development in women receiving these ovulation-induction medications.⁵⁶ Patients undergoing ovulation induction with gonadotropins are usually examined every other day, and occasionally daily, beginning at about cycle day 9. Performance of follicular monitoring by the transvaginal technique not only allows assessment of follicular response in order to adjust the ovarian stimulation regimen, but also to identify women at high risk for multiple births and ovarian hyperstimulation syndrome if there is overstimulation or hyper-response to the treatment.

In gonadotropin-treated patients, there seem to be 2 distinct patterns of follicular development.⁶⁵ In those amenorrheic women with no exogenous estrogenic activity and dormant ovaries, the response to endogenous gonadotropins is to develop a small number of large follicles (see Figure 37-2). The growth rate and E₂ secretion are linear, correlate well, and are of equal predictive value. A higher pregnancy rate is achieved in this group. In contrast, patients with estrogenic activity who harbor antral follicles at different stages of development react very differently. Stimulation of these patients requires less human menopausal gonadotropin (hMG) and usually results in the rapid recruitment of many follicles, with different growth rates with different degrees of E₂ secretory capacity. Also, the rate at which E₂ increases is exponential, increasing the risk of hyperstimulation. Thus, there is a dissociation between follicle size and E₂ levels, suggesting that growth rate and functional maturity are asynchronous. This group of women particularly benefits from combined E₂ levels and sonographic follicular monitoring.

In gonadotrophin-induced cycles, sonographic delineation of follicle size is crucial because hCG is best administered once follicles reach 15 to 18 mm in size. For ovulation to occur in gonadotropin therapy, either hCG, which is most common, or a gonadotropin-releasing hormone (GnRH) agonist is required. Intercourse or timed intrauterine insemination may be chosen to accompany this therapy depending on the diagnosis.

Ovulation induction and follicular monitoring are also used in in vitro fertilization and embryo transfer (IVF-ET) to increase the number of oocytes aspirated. This increase in turn increases the number of fertilized oocytes and embryos that may be transferred, thereby increasing the chance of pregnancy (but also the potential for multiple gestations). Patients undergoing IVF-ET are examined by sonography starting earlier in their cycles and usually daily in an attempt to carefully monitor their follicular development. Further description is noted later in the chapter.

Transvaginal sonography has an important role in detecting cysts on an early follicular sonogram (usually cycle day 3) prior to starting therapy. Typically, cysts may be associated with Lupron pretreatment or residual cysts from prior treatment, which may impair the ability to induce ovulation with CC or hMG, particularly if they remain hormonally active as assessed by an elevated estradiol level.

Assisted reproductive technology (ART) includes all forms of in vitro fertilization and embryo transfer, and gamete intrafallopian tube transfer (GIFT). The use of TVS has afforded the development of several minimally invasive procedures, such as transvaginal aspiration of follicles, ultrasound-guided embryo transfer, and cannulation of the fallopian tube for GIFT procedures.

In vitro fertilization and embryo transfer (IVF-ET) was originally offered in centers primarily for patients who have absent or significantly diseased tubes. Now it is also used for other infertility disorders related to cervical factors, oligospermia, immune factors, ovulatory disorders, and unexplained infertility. For this procedure, TVS not only has an important role in follicular monitoring, but it plays a vital role in guiding follicular aspiration.²

The actual success rate (as defined by the percentage of "take-home babies" divided by the number of stimulated cycles) for IVF-ET continues to increase and now reaches 50% or more in selected populations. Thus, in some subpopulations, this rate is higher, exceeding the "natural rate" at which fertilized ova progress to delivering a live neonate (approximately 18% to 25%).⁶² The primary factors that influence the success or failure of IVF-ET include the number of mature oocytes retrieved and the number and quality of embryos transferred.⁶⁶ Other factors that may influence the success or failure of this procedure include the endometrial receptivity to the implanting conceptus.

Another procedure has been developed that involves transfer of sperm and ova directly into the tube (GIFT, or zygote intrafallopian tube transfer [ZIFT], or tubal embryo transfer [TET]) for some couples with unexplained infertility. This procedure involves follicular aspiration and mixing of ovum and sperm, with replacement of the mixture (GIFT) or fertilized zygote (ZIFT) or cleaved embryos (TET) back into the fallopian tube. Replacement is done either by laparoscopic cannulation of the fallopian tube or, more recently, by sonographically guided catheterization of the fallopian tube through the endocervical and endometrial lumina.⁶⁴ These procedures account for less than 1% of all reported ART procedures (Centers for Disease control, 2005 data, www.cdc.gov/ART; and Society of Assisted Reproductive Technologies, 2006 data, www.sart.org).

The ovulation induction for IVF is different from other forms of ovulation induction. The objective of most forms of ovulation induction is to ovulate a few oocytes. With IVF we want to induce many follicles to grow, to obtain many oocytes, to have many fertilized oocytes that grow to yield many embryos, with the best quality embryos then transferred to improve the pregnancy rate. In order to increase the number of mature follicles, a medication is given to block the LH surge from the pituitary. This block can be obtained with either a gonadotropin-releasing hormone (GnRH) agonist (ie, Lupron) or a GnRH antagonist (ie, ganirelix or cetratide). When a GnRH agonist is given in a long protocol, the Lupron is given in the luteal phase or with an oral contraceptive overlap to avoid the occurrence of an escape ovarian cyst. After about 10 days of Lupron, a TVS and laboratory evaluation (E₂) are done to confirm that the ovaries are suppressed, implying that the pituitary is blocked, and no ovarian cysts are present. Then gonadotropin therapy is given in addition to the Lupron to stimulate multiple follicles to grow. TVS and E₂ monitoring occurs about 5 days after starting the ovarian stimulation and then daily or every other day until mature follicles are identified (that is, follicles with a size approximately 16 to 20 mm in diameter). Alternatively, the GnRH antagonist protocol is shorter. This protocol starts with daily gonadotropin therapy, and the GnRH antagonist is begun when the follicles reach the 12- to 14-mm size to block the LH surge. Again, the follicle number and sizes as well as estradiol levels are monitored with both IVF protocols.

For IVF, follicles are typically triggered with hCG when the follicles reach approximately 16 to 20 mm in average dimension, with a minimum of 2 at 18 mm, and when there is evidence by estradiol values of multiple mature follicles (approximately 200 to 400 pg/mL per mature follicle).⁵⁸ Another sonographic sign of a mature follicle is the presence of low-level, intrafollicular echoes. These echoes probably arise from sheets of granulosa cells that have separated from the follicular wall. In one study involving patients who underwent ovulation induction, a greater pregnancy rate was achieved in those patients whose follicles demonstrated such intrafollicular echoes.²

Transvaginal sonography–guided follicular aspiration has now become the preferred procedure for oocyte retrieval over the previously used laparoscopic technique (Figure 37-14). The major advantages to this technique include decreased exposure to general anesthesia, lower chance for operative complications, and feasibility of performing this procedure on an outpatient basis. The success rate as determined by the number of fertilizable oocytes retrieved and pregnancies produced is comparable to or better than that of the laparoscopic technique.⁶⁷ The procedure is also advantageous in patients with pelvic adhesions because laparoscopic access to the ovary could be hampered in those cases.⁶⁸ Most importantly, acceptance of the procedure by patients is high, and it can be performed as an outpatient procedure with only intravenous (IV) sedation for anesthesia and not general anesthesia.^69,70

Several methods for follicular aspiration that involve sonographic guidance have been described. These include TVS for guidance of transvesicular aspiration, and transabdominal sonographic guidance for transurethral aspiration. TVS with transvaginal aspiration has essentially replaced these other methods since the stimulated ovary typically falls into the posterior cul-de-sac, such that it is usually easy to do the TVS approach for follicle aspiration. Occasionally, adhesions inhibit the mobility of the ovary and the ovary is located high in the pelvis or sometimes in the abdomen; it is these ovaries that are the only exception to transvaginal aspiration and require transabdominal aspiration to obtain the oocytes.

For transvaginal aspiration with transvaginal transducers, a needle guide is directly attached to the transducer/probe, which is covered by a sterile sheath or condom containing sterile gel. A long (30-cm) 16- to 18-gauge needle with a scored tip for easy visualization on sonography is used. Some report using a smaller-gauge needle (21 gauge), which may reduce the discomfort. The needle guide and needle are set up to traverse in the beam path of the transducer with the biopsy mode screen. Prior to performing an IVF procedure, the biopsy screen and transvaginal ultrasound probe with needle guide attached and needle need to be calibrated. The calibration is typically done by placing the TV ultrasound probe with needle guide into a bucket of water and then sliding the needle through the guide while it is under water and checking the biopsy screen to see if it is accurately aligned. If not, the biopsy screen can be adjusted.

After calibration this setup is ready for TV aspiration. The cursor is displayed on the scanning screen, which indicates the path of the needle (see Figure 37-14). After a condom containing sterile gel is placed over the transducer and the sterile needle guide is attached, the operator manipulates the transducer to optimally delineate the ovary. The desired follicle is brought into the "line of sight" and the needle is introduced through the needle guide and then transvaginally into the ovary. The needle is typically placed in the center of the follicle and aspirated. The fluid collects in the test tubes that are transferred to the ART laboratory to identify oocytes. This process is continued until all of the follicles have been aspirated.

This aspiration technique has been associated with a low complication rate (<0.01%). One complication that has been described is the inadvertent introduction of the needle into a vessel in the pelvis.⁷¹ This problem can be avoided if the operator carefully examines any round structure in both long and short axis to distinguish a pulsatile vascular structure from a follicle.

Whether or not the route of follicular aspiration actually affects the outcome of IVF-ET can be questioned.⁷² Most studies, however, have indicated a conception rate that is higher than or equal to that achieved using the laparoscopic technique.⁶⁷

Sonography is also being used now for transcervical cannulation of the uterine lumen.¹⁵ This can be used for 2 procedures, including embryo replacement and cannulation of the fallopian tube required in GIFT procedures.

Sonographic guidance for embryo replacement has been described in one series.⁸ This procedure involves transabdominal sonographic monitoring of the catheter in which the embryo is loaded. Optimally, the embryo should be instilled when the tip of the catheter extends near to the top of the endometrial lumen. Most suggest that the embryos are placed at least 1.0 to 1.5 cm from the fundal endometrium. It is also important to atraumatically place embryos in the uterine cavity. Embryo placement into the proper position may be identified by observing the echogenic air bubble, which is loaded before the embryo at the tip of the catheter, within the uterine cavity. Although not all institutions use sonographic monitoring for embryo replacement, it has been shown that this technique can facilitate more accurate placement of the transferred embryo and improve pregnancy rates.⁸ Three-dimensional sonography is a better technique to determine where the embryo transfer catheter and fluid with bubble are located after transfer. Perhaps, the 3D real-time (or four-dimensional [4D]) ultrasound will become more refined in the future to allow for a direct 3D real-time or 4D ultrasound-guided embryo transfer.

A technique for sonographic guidance of placement of a catheter into the fallopian tubes for GIFT procedures has been described.⁷³ For this procedure, a catheter is placed transcervically and manipulated into the area of the uterine cornu. The catheter is slowly introduced, under sonographic guidance, into the tubal ostium. Once the catheter is in the distal isthmic portion of the fallopian tube, the sperm and ova may be introduced through the cannula directly into the tube.

Recovery after Ovulation Induction/IVF and Complications

After induced ovulation, the stimulated follicles usually undergo regression, but they may persist and enlarge over the remainder of the cycle. The presence of physiologic ovarian cysts (>3 cm) may preclude ovulation induction during the next cycle, because the previously induced follicles may not have totally regressed, and the remaining ovarian tissue may not be as responsive to ovulation-induction medication. Theoretically, the risk of torsion, rupture, or both may also be increased in these women. Usually, the ovaries are back to normal size within 1 month of ovulation-induction therapy.

Sonography has an important role in decreasing the likelihood of complications such as ovarian hyperstimulation syndrome (OHSS) and multiple gestations. Ovarian hyperstimulation disorder occurs in various degrees of severity in many patients who undergo ovulation induction (15% to 30%), ranging from mild abdominal discomfort, probably due to the distention of the ovarian capsule, to severe circulatory compromise and electrolyte imbalance, probably secondary to ascites or pleural effusions that may develop. This more severe form of OHSS occurs in less than 1% of cases and is usually associated with massive stromal edema of the ovary, ascites, electrolyte imbalances, intravascular dehydration, increased risk of hypercoagulability, and sometimes pleural effusions. The enlarged ovaries may be prone to torsion. The symptoms associated with OHSS usually begin 5 to 10 days after the hCG trigger shot is given, but they can be most severe in patients who actually achieve pregnancy. There is a late-onset form of OHSS that begins around the time of the first positive pregnancy test, which may be triggered by the rising hCG levels of pregnancy. Recent studies have shown that hyperstimulation is unlikely in women whose ovaries contain several large (>15 mm) follicles and few or no small follicles (11 to 15 mm), and tends to occur more often when there are several small- or intermediate-sized follicles in addition to the large follicles.⁷⁴

Sonographic findings of enlarged ovaries with multiple immature follicles may suggest the possibility of hyperstimulation. Although OHSS is hard to predict, this syndrome does occur more commonly when E₂ levels are high (>3000 pg/mL). For ovulation induction and IVF, the estradiol levels are typically less than 3000 pg/mL and the incidence of OHSS is approximately 15% to 30% for mild symptoms and less than 1% for severe OHSS symptoms. However, if the estradiol level exceeds 5000 pg/mL at its peak prior to the hCG trigger, then the OHSS incidence may be as high as 80%. Despite the superovulation required for IVF, severe hyperstimulation is only rarely encountered. This is probably a reflection of the close monitoring with frequent adjustments in gonadotropin dosing. But IVF may have an added benefit due to the drainage of follicles containing oocytes and granulosa cells via TV ultrasound-guided aspiration and the resultant collapse of the aspirated follicles, which may lower the risk of OHSS.

On sonography, patients with OHSS usually have enlarged ovaries (>10 cm) that may contain several hypoechoic areas (Figure 37-15). The hypoechoic areas may correspond to atretic follicles or to regions of hemorrhage within the ovary. Color Doppler sonography usually shows increased diastolic flow within the intraovarian arteries. In addition, venous flow may become nonphasic, which is a sign of poor venous return. The pregnancies that occur with OHSS may be very early (<4 weeks), and no definitive sonographic findings may be found. Intraperitoneal fluid is commonly present. With supportive therapy, this syndrome usually regresses spontaneously.

Sonography is also helpful in decreasing the likelihood of multiple gestations that may occur with fertilization of multiple ova. It is difficult, however, to predict which pregnancies will result in multiple births. Clearly, however, when there are more than 4 mature follicles in gonadotropin therapy, the chance for multiple gestation beyond twinning is greater than if only 2 or 3 mature follicles are induced. However, with IVF the risk of multiple gestations is controlled by limiting the number of embryos transferred into the uterus, after the oocytes were retrieved and fertilized in culture.

Sonography has a vital role in the evaluation and management of certain gynecologic causes of infertility.³² The baseline pelvic ultrasound yields information about the ovarian reserve and is indicative of the likely response to therapy, as well as identifying any underlying pathology that may impact fertility (ie, fibroids, polyps, adnexal masses, endometriomas, hydrosalpinges, etc). By adding 3D ultrasound to this baseline ultrasound, congenital uterine anomalies can be accurately diagnosed. The SHG may be used not only to assess the uterine cavity for filling defects, but also to diagnose the specific filling defect (such as adhesions, polyps, fibroids) with the assistance of typical ultrasound characteristics on gray scale or CDS. SHG with SSG can identify tubal patency without radiation exposure, and with less discomfort. SHG/SSG may replace HSGs in most cases, except in those patients suspected of Salpingitis Isthmica Nodosa SIN or those needing presurgical evaluation of the proximal tubal stump prior to tubal reanatomosis surgery.

In the treatment of infertility, the 2 major clinical applications involve follicular monitoring and ultrasound guidance of follicular aspiration and embryo transfer. The use of transvaginal transducer/probes has also enhanced their depiction sonographically and now allows a more in-depth knowledge of the processes leading to achievement of pregnancy in infertile or subfertile women, and in monitoring treatments.

Three-dimensional ultrasound is helpful in the evaluation of patients for uterine malformations and may have more applications in infertility patients in the future. There is a recent development of 3D ultrasound used in the inversion mode to identify and measure ovarian follicles automatically by GE (called sonoAVC). Its benefits are currently under investigation. Also, 3D ultrasound can currently be performed to evaluate where the IVF embryos are placed in the uterine cavity. However, in the future, as real-time 3D ultrasound (aka 4D ultrasound) improves, we may be able to do 4D ultrasound-guided embryo transfers for better accuracy, and possibly better pregnancy rates and lower ectopic rates.

Thus, in the last 2 decades, ultrasound has become a mainstay in infertility evaluation and management. We expect ultrasound will have even greater roles in the future as the equipment and software continue to evolve, and as clinician familiarity and experience with these evolving technology increases.

The following are proposed guidelines for ultrasound evaluation of the female pelvis. The document consists of 2 parts:

• Part I: Equipment and Documentation Guidelines

• Part II: Guidelines for Performance of the Ultrasound Examination of the Female Pelvis for Infertility and Reproductive Medicine

This guideline has been developed to provide assistance to practitioners performing ultrasound studies of the female pelvis. In some cases, additional and/or specialized examinations may be necessary. Although it is not possible to detect every abnormality, adherence to the following will maximize the probability of detecting most of the abnormalities that occur.

This guideline includes excerpts from various previously published guidelines of the AIUM. The latest versions of all AIUM guidelines are available at www.aium.org.

PART I: EQUIPMENT AND DOCUMENTATION GUIDELINES

Equipment

The sonographic examination of the female pelvis should be conducted with a real-time scanner, with the availability of multiple types of transducers. The transducer or scanner should be adjusted to operate at the highest clinically appropriate frequency, realizing that there is a trade-off between resolution and beam penetration. With modern equipment, studies performed from the anterior abdominal wall can usually use frequencies of 3.5 MHz or higher, while scans performed from the vagina should use frequencies of 5 MHz or higher.

Care of the Equipment

All probes should be cleaned after each patient examination. Transvaginal probes should be covered by a protective sheath prior to insertion. Patients should be questioned about latex allergy prior to use of a latex sheath. Following each examination, the sheath should be disposed, and the probe washed, dried, and appropriately disinfected. The type of antimicrobial solution and the methodology for disinfection depend on manufacturer and infectious disease recommendations.

Documentation

Adequate documentation is essential for high-quality patient care. A permanent record of the ultrasound examination and its interpretation should be kept by the facility performing the study. Images of all appropriate areas, both normal and abnormal, should be recorded. Variations from normal size should be accompanied by measurements. Images are to be appropriately labeled with the examination date, facility name, patient identification, and image orientation and/or organ imaged when appropriate. A report of the ultrasound findings should be included in the patient's medical record. Urgent or clinically important unexpected results should be communicated verbally to any referring and/or treating physician and this communication should be documented in the report. Retention of the permanent record of the ultrasound examination should be consistent with both clinical needs and the relevant legal and local health care facility requirements.

PART II: GUIDELINES FOR PERFORMANCE OF THE ULTRASOUND EXAMINATION OF THE FEMALE PELVIS FOR INFERTILITY AND REPRODUCTIVE MEDICINE

The following guidelines describe the examination to be performed for each organ and anatomic region in the female pelvis. Whenever possible, all relevant structures should be identified by the vaginal approach. When a transvaginal scan fails to image all areas needed for diagnosis, a transabdominal scan should be performed. In some cases, both a transabdominal and a transvaginal scan may be needed.

General Pelvic Preparation

For a pelvic sonogram performed transabdominally, the patient's urinary bladder should, in general, be distended adequately to displace the small bowel and its contained gas from the field of view: Occasionally, overdistension of the bladder may compromise evaluation. When this occurs, imaging may be repeated after the patient partially empties her bladder.

For a transvaginal sonogram, the urinary bladder is preferably empty. The patient, the sonographer, or the physician may introduce the transvaginal transducer, preferably under real-time monitoring. Transvaginal sonography is a specialized form of a pelvic examination. Therefore, policies applied locally regarding chaperone or patient privacy issues during a pelvic examination should also be applied during a transvaginal ultrasound examination.

Uterus

The vagina and the uterus provide anatomic landmarks that can be used as reference points when evaluating the pelvic structures. In evaluating the uterus, the following should be documented: (1) uterine size, shape, and orientation; (2) the endometrium; (3) the myometrium; and (4) the cervix. The vagina may be imaged as a landmark for the cervix and lower uterine segment. Uterine length is evaluated on a long-axis view as the distance from the fundus to the cervix. The anteroposterior (AP) diameter of the uterus is measured in the same long-axis view from its anterior to posterior walls, perpendicular to its long axis. The transverse diameter is measured from the transaxial or coronal view, perpendicular to the long axis of the uterus. If volume measurements of the uterine corpus are performed, the cervical component should be excluded from the uterine measurement.

Abnormalities of the uterus should be documented. The endometrium should be analyzed for thickness, focal abnormalities, and the presence of fluid or masses in the endometrial cavity. Assessment of the endometrium should allow for variations expected with phases of the menstrual cycle and with hormonal supplementation. The endometrial thickness measurement should include both layers, measured anterior to posterior, in the sagittal plane. Any fluid within the endometrial cavity should be excluded from this measurement. If the endometrial echo is difficult to image or ill defined, a comment should be added to the report.

The myometrium and cervix should be evaluated for contour changes, echogenicity, and masses. Masses, if identified, should be measured in at least 2 dimensions and their locations recorded.

Adnexa (Ovaries and Fallopian Tubes)

When evaluating the adnexa, an attempt should be made to identify the ovaries first since they can serve as a major point of reference for assessing the presence of adnexal pathology. Although their location is variable, the ovaries are most often situated anterior to the internal iliac (hypogastric) vessels, lateral to the uterus, and superficial to the obturator internus muscle. The ovaries should be measured, and ovarian abnormalities should be documented. Ovarian size can be determined by measuring the ovary in 3 dimensions on views obtained in 2 orthogonal planes. It is recognized that the ovaries may not be identifiable in some women. This occurs most frequently after menopause or in patients with a large leiomyomatous uterus.

The normal fallopian tubes are not commonly identified. This region should be surveyed for abnormalities, particularly dilated tubular structures.

If an adnexal mass is noted, its relationship to the uterus and the ovaries, if separately visualized, should be documented. Its size, echogenicity, and internal characteristics (cystic, solid, or complex) should be determined. Doppler ultrasound may be useful in select cases to identify the vascular nature of pelvic structures.

Cul-de-Sac

The cul-de-sac and bowel posterior to the uterus may not be clearly defined. This area should be evaluated for the presence of free fluid or masses. When free fluid is detected, its echogenicity should be assessed. If a mass is detected, its size, position, shape, echogenicity, internal characteristics (cystic, solid, or complex), and relationship to the ovaries and uterus should be documented. Identification of peristalsis can be helpful in distinguishing a loop of bowel from a pelvic mass. In the absence of peristalsis, differentiation of normal loops of bowel from a mass may be difficult. A transvaginal examination may be helpful to distinguish a suspected mass from fluid and feces within the normal rectosigmoid. An ultrasound water enema study or a repeat examination after a cleansing enema may also help to distinguish a suspected mass from bowel.

Limited Examination

In some circumstances, a limited pelvic ultrasound examination is appropriate, especially when monitoring ovarian stimulation (eg, an ovarian folliculogram study or determining endometrial qualities prior to cryopreserved embryo transfer). A comprehensive exam should have previously been performed in the preceding 4 to 6 months to rule out other gynecologic pathology. The limited exam can be restricted to the organ or measurements of interest. In the case of an ovarian folliculogram, the following should be documented: ovarian follicle number in each ovary, endometrial thickness, and endometrial morphologic appearance. In addition, follicular diameters in 2 dimensions for each follicle above 10 mm should be recorded. A single recorded value representing the mean of 2 diameter measurements performed at right angles is also acceptable. Given that these patients will have had a full pelvic exam at the appropriate interval prior to initiating therapy, for infertility patients undergoing limited folliculogram studies, permanent recorded images should be obtained as indicated. Pertinent clinical information should be recorded in the patient record.

Ultrasound-Guided Procedures

1. Follicle puncture: Ultrasound-assisted (transvaginal or transabdominal) follicle puncture for retrieving eggs for in vitro fertilization (IVF) is appropriate in the following circumstances:

   1. The patient has undergone comprehensive sonographic evaluation of the pelvis within 4 to 6 months prior to the start of hormonal stimulation of the ovaries.

   2. Real-time continuous guidance is available, and the image demonstrates a safe approach for the needle path.

   3. The ovaries can be brought in close proximity to the ultrasound transducer, thus avoiding the puncture of vital structures (eg, bowel and blood vessels).

2. Cyst aspiration: Ultrasound-assisted (transvaginal or transabdominal) ovarian cyst puncture and aspiration is appropriate in patients who have been diagnosed with a persistent ovarian cyst and who meet the following criteria:

   1. Failed resolution of the cyst following observation and/or hormonal manipulation.

   2. The cyst is unilocular and thin walled without internal excrescences or septations.

   3. Real-time continuous guidance is available, and the image demonstrates a safe approach for the needle path.

   4. The cyst can be brought in close proximity to the ultrasound transducer, thus avoiding the puncture of vital structures (eg, bowel and blood vessels).

3. Embryo transfer: Ultrasound-assisted embryo transfer is appropriate in patients undergoing a "fresh" IVF cycle or following embryo cryopreservation or embryo/egg donation. If an abdominal ultrasound examination is performed, the bladder should be full to facilitate visualization of the endometrium and the transfer catheter.

Qualifications and Responsibilities of the Physician

Physicians who perform or supervise ultrasound-guided follicular aspiration or embryo transfer should be skilled in pelvic ultrasonography and appropriate placement of catheters and ultrasound-guided needle placement. They should understand the indications, limitations, and possible complications of the procedure. Physicians should have training, experience, and demonstrated competence in gynecologic ultrasonography and treatment procedures. Physicians are responsible for the documentation of the examination, quality control, and patient safety. Urgent or clinically important unexpected results should be communicated verbally to any referring and/or treating physician and this communication should be documented in the report.