Abstract
Assisted reproductive treatment (ART) has become the only hope for biologically own children for numerous infertile couples. It is estimated that 1.7–4.0 per cent of children born in developed countries are the result of assisted conception . As a minimally invasive diagnostic tool, ultrasound is used readily throughout ART – starting with the pre-treatment assessment of pelvic organs, through cycle monitoring, oocyte collection and embryo replacement, to diagnosis of complications and outcome monitoring.
Introduction
Assisted reproductive treatment (ART) has become the only hope for biologically own children for numerous infertile couples. It is estimated that 1.7–4.0 per cent of children born in developed countries are the result of assisted conception [1,2]. As a minimally invasive diagnostic tool, ultrasound is used readily throughout ART – starting with the pre-treatment assessment of pelvic organs, through cycle monitoring, oocyte collection and embryo replacement, to diagnosis of complications and outcome monitoring.
In this chapter, we will cover the application of ultrasound in assisted reproduction.
The Pre-treatment Scan
This scan, carried out before embarking on ART, serves multiple purposes. First, it serves to identify any pathology that might have contributed to infertility, or which may affect the outcome of the treatment (hydrosalpinx or uterine anomaly). Findings of the pre-treatment scan help to predict the response to ovarian stimulation by assessing the antral follicle count (AFC) and help guide dosing of gonadotrophins, provide a guide to endometrial receptivity and inform on the ease of access to the ovaries during oocyte collection.
Antral follicle count and anti-Mullerian hormone levels (AMH) are the best predictors of oocyte yield and stimulation response in ART, with an ongoing debate as to the superiority of one test over another [3]. Ultrasound-based AFC measurement can be performed with ease using 2D and 3D ultrasound modalities (the exact method is described in Chapter 7) (Figure 11.1). Typically, AFC assessment is carried out at the beginning of the menstrual cycle, as the follicles of interest are 2–10 mm in diameter [4–6]. Scanning later in the menstrual cycle could underestimate the value due to exclusion of the dominant follicle(s) as these are >10 mm in diameter; however, some evidence indicates that accuracy of AFC is not affected when scanned at any stage of the menstrual cycle [7]. Antral follicle count produces instant results and, as such, has the advantage over AMH. There are, however, limitations of the test, and these are related to variations between centres, differences in ultrasound equipment, training and the timing of scanning [3]. Three-dimensional ultrasound scanning may be able to overcome the inconsistencies when assessing AFC; however, the technology is not in routine use.
Figure 11.1 SonoAVC of an ovary during controlled ovarian hyperstimulation, with follicles represented by different colours. Use of SonoAVC allows for automated measurement of volume and the average diameter of each follicle on the output seen to the right of the SonoAVC image. It shows the total number of follicles, lists each follicle number and its relaxed sphere diameter (d(V)), three orthogonal diameters (dx, dy and dz), mean diameter of dx, dy and dz (m-d) and follicle volume (V) in millilitres.
In juvenile women, a transvaginal ultrasound scan may be inappropriate, limiting the use of AFC as a predictor of ovarian reserve [8]. Obesity may also affect the AFC due to the poor image quality when carrying out the scan, as well as due to reported significant inter-cycle variability of AFC in this population [9–11].
In order to obtain the most reliable information from the AFC scan, the process has to be standardized and consistent. It is recommended that the procedure be carried out between days 2 and 4 of a natural menstrual cycle or of a contraceptive pill cycle; only follicles measuring 2–10 mm should be included and appropriate training should be provided with the use of standard ultrasound settings and equipment [4,12]. Identification and characterization of ovarian cysts must be carried out during the scan. Details of ovarian assessment are described in Chapters 2 and 8.
Assessment of the uterus and endometrium, as well as of the adnexa, is a component of the pre-treatment scan. The endometrial and myometrial assessment for the presence of endometrial polyps, submucosal and intramural fibroids, and adenomyosis are described in Chapters 4 and 6.
In order to exclude the presence of endometrial polyps, the scan ideally should be carried out in the late follicular phase or mid-cycle. Endometrial polyps distort the triple-layer appearance and are easily distinguishable as hyperechoic structures distorting the midline echo with generally a single feeding blood vessel (Figure 11.2). Scanning early in the cycle may cause confusion as the endometrial cavity may be distended by sloughed cells and blood clots produced during menstrual shedding. Doppler imaging may help to differentiate these structures, with Doppler signal present in polyps and fibroids and absent in the menstrual material (Figure 11.3). Adenomyosis within the myometrium is most clearly visible in the luteal phase of the menstrual cycle due to decidual reaction of the ectopic endometrium. The adenomyotic islands appear hyperechogenic compared to the myometrium and are similar in appearance to the endometrium (Figure 11.4). Hyperechoic and thick endometrium, as seen in the luteal phase, allows for better outline of the endometrial cavity, thus increasing the chances and accuracy of congenital uterine anomaly diagnoses. This is detailed in Chapter 5.
Adnexal lesions, mainly hydrosalpinges, may be present at the pre-treatment scan, and may also develop during controlled ovarian stimulation. A detailed scan of the adnexal region is essential in every gynaecological scan (see Chapter 8), and even more so in the ART setting, as the presence of a distended fallopian tube significantly decreases the chances of success [13]. Ultrasound-guided aspiration of hydrosalpinx may be considered (see Chapter 12); however, salpingectomy offers the best chance of favourable ART outcome [13].
Cycle Monitoring
Ultrasound assessment throughout the natural menstrual cycle or controlled ovarian hyperstimulation (COH) serves the purpose of assessment of follicular maturity and endometrial receptivity [14]. Endometrial assessment comprises description of the appearance and thickness, as detailed in Chapters 2 and 6. Follicular assessment aims to measure the response to treatment in the form of linear measurements of follicular growth, often combined with measurements of serum estradiol, luteinizing hormone (LH) and progesterone [15].
In a natural menstrual cycle, the follicle grows at a rate of 2 mm per day until approximately 20–25 mm, when ovulation is expected to occur (Figure 11.5) [14]. It is occasionally possible to visualize the cumulus oophorus prior to ovulation, represented as an irregularity on the internal wall of the leading follicle. Shortly prior to ovulation, the uniform walls of the follicle loose cohesion and appear blurred on ultrasound scan (Figure 11.6), which is followed by the presence of free fluid in the pouch of Douglas. After ovulation occurs, the follicle undergoes rapid changes and transforms into a corpus luteum. Sonographically, the appearance of this structure varies greatly, but most commonly has an irregular, thick-walled structure with heterogeneous content and very intense blood flow on the periphery, as demonstrated by Doppler ultrasound (Figure 11.7) [15]. Rarely, rupture does not occur and the follicle still undergoes luteinization in a rare condition known as luteinized unruptured follicle syndrome (LUFS) (Figure 11.8).
In the in vitro fertilization (IVF) setting, ultrasound scan is used to monitor the ovarian response to gonadotrophins in order to obtain a large number of mature oocytes safely, minimizing the risk of ovarian hyperstimulation syndrome (OHSS) and to time administration of the final oocyte maturation trigger. Most IVF centres aim to achieve at least three follicles measuring 17–18 mm in diameter before the maturation trigger is administered. The likelihood of obtaining a mature oocyte increases with the size and volume of the follicle. An oocyte recovery rate of 83.5 per cent may be achieved when aspirating follicles of 18–20 mm in diameter (equivalent to 3–4 ml of fluid), but a high cleavage rate of 92 per cent may be achieved when oocytes are obtained from follicles measuring 23–24 mm (6–7 ml) [16]. Follicle tracking may be carried out using 2D and 3D ultrasound modalities, with a live or offline analysis. There is no uniform standard recommending follicular diameter measurements [17] with some measuring the single largest diameter and others measuring the mean of two or three diameters obtained from one or two planes, respectively. Automated measurement of follicular size using SonoAVC, while speeding up the analysis, does not translate to an improved ART outcome [18] (Figure 11.9). Three-dimensional assessment of follicular size does, however, correlate closely with actual volume of aspirated follicular fluid [19,20] and number of mature oocytes [20,21](Figure 11.1).
(a,b) The left and right stimulated ovaries by 3D transvaginal ultrasound (3DTVU) scanning plus SonoAVC software one day before oocyte pickup, showing colour encoded 3D reconstruction of follicles.
(c) Detailed SonoAVC report showing measured diameters (dx, dy, dz), automatically calculated mean diameter and volumes, with the colour-coded corresponding follicle.
Endometrial assessment forms an integral component of US cycle monitoring as it allows for a minimally invasive and non-disruptive indirect assessment of the organ. In the most basic and commonly used form, the assessment comprises endometrial thickness measurement and description of the pattern (see Chapters 2 and 6). Endometrial thickness below 7 mm is associated with a low chance of conception [22]; however, pregnancies have been reported with a thickness of 4 mm [23]. A large study of 2896 IVF/intra-cytoplasmic sperm injection (ICSI) cycles by Chen et al., investigating endometrial thickness and pattern on the day of hCG administration, demonstrated that a thin (≤7 mm) and non-triple-layer endometrium was associated with very poor outcomes (clinical pregnancy rate of 14.3 versus 24.4 per cent in women with thin endometrium and a triple-layer appearance; number of patients = 52, p > 0.05). In the same study, increasing endometrial thickness was associated with increasing clinical pregnancy rates, with no differences in miscarriage rates. The endometrial pattern did not discriminate between the conception and non-conception cycles; however, miscarriage was much higher in the non-triple-layer appearance of the endometrium (15.6 versus 7.9 per cent; p < 0.05) [24]. Similar findings have been reported by other authors [25,26]. It is believed that the premature luteinization of the endometrium, as demonstrated by uniformly hyperechoic appearance with loss of midline echo, may be out of phase with the transferred embryo, and as such, leads to lower implantation rates [27]. In a study by Friedler et al., the authors demonstrated that a homogeneous endometrium on the day of hCG administration has a negative predictive value for conception of 87.5 per cent versus a positive predictive value for conception of 33.1 per cent (specificity 13.7 per cent) for a tri-laminar pattern [28].
Studies of endometrial blood flow assessed using 2D colour or power Doppler, or pulse wave Doppler, report discrepant results, with some suggesting low resistance spiral artery waveforms associated with pregnancy [29], and others finding no such correlation [30,31]. The measurement of Doppler indices on endometrial spiral arteries is a difficult and time-consuming process, and is dependent on the distance of the artery from the transducer and the angle of insonation. As such, it is not used in routine clinical practice. Similarly, 3D power Doppler assessment of endometrial and subendometrial vascularity does not clarify the usefulness of this modality as a marker of endometrial receptivity and predictor of ART outcome [32,33]. Finally, endometrial contractions of five or more per minute have been associated with a reduced chance of conception, and can be considered as a potential marker of endometrial receptivity [34,35].
Transvaginal Ultrasound Guided Oocyte Retrieval
Successful oocyte retrieval needs to be fast and precise, aiming to retrieve the maximal number of undamaged oocytes from the ovarian follicles without complications. It is a critical process, as increasing the number of mature oocytes suitable for IVF procedures improves the likelihood of generating good-quality embryos and achieving a successful pregnancy. The evolution and chain of events in the developments in oocyte retrieval techniques (Figure 11.10) have led to increased safety and better pregnancy rates in ART treatment.
Commonly, the final oocyte maturation trigger is administered when ≥3 follicles of ≥17 mm size are documented on ultrasound [36,37]. The trigger can also be given when ≥3 follicles have a diameter of 18 mm [38], or when ≥1 follicle is of ≥18 mm in size and 3 follicles of ≥15 mm [39,40]. Higher pregnancy, implantation and live birth rates were seen in patients when the 17 mm to 10 mm follicle ratio on the day of hCG administration reached 60 per cent and the peak estradiol per oocyte level was within 100–399 pg/ml [41]. Although flexibility in the administration of the oocyte maturation trigger provides convenience for the physician and the patient [42], there is a risk of early progesterone rise causing premature closing of the implantation window [43]. Oocyte retrieval is usually scheduled for 36 hours post-administration of the final oocyte maturation trigger [44], although it could be performed between 32 and 36 hours, with a chance of obtaining mature oocytes reported at 39 hours [45–47].
Appropriate equipment selection, adequate anaesthesia and patient preparation are essential for the procedure to be carried out safely and with optimal results. The transvaginal transducer should be covered with a sterile latex-free probe cover with ultrasound gel to assure optimal pelvic organ visualization. A sterile, well-fitting needle guide is attached to the transducer and its patency is tested by passing the oocyte retrieval needle through it before inserting the transducer into the patient. Following introduction of the transducer into the vagina, a scan should be carried out to evaluate the pelvis and the presence of any free fluid, and to assess the ovaries and the uterus (Figure 11.11). At the same time, assessment of the possible entry points is carried out after studying the vaginal wall vascularity by power Doppler and the relationship of the ovary with the neighbouring vessels (Figure 11.12). Rotation of the probe helps to distinguish a uniformly spherical follicle from a vessel that in one plane may be circular but becomes a tubular structure in another plane. Addition of Doppler imaging may also help to differentiate these structures (Figure 11.13); however, Doppler signal may be absent or very weak when the angle between the beam and the blood flow is 90 degrees. When choosing a point of entry of the oocyte collection needle, areas where the bladder (Figure 11.14), bowel, cervix, ureter or other pelvic structures could be inadvertently injured should be avoided. If no obvious point of safe entry can be identified, abdominal pressure by the assistant may move structures and create a safe window for passage of the needle (Figure 11.15). When access to the ovary is possible only through the uterus, transuterine puncture is a safe alternative. In this instance, the needle is passed through the myometrium, avoiding the endometrium and uterine vessels [48]. Due to a small increase in miscarriage rates with transuterine oocyte collection, a freeze-all strategy may be applied [49].
Figure 11.11 Assessment before oocyte collection. Following introduction of the vaginal transducer into the posterior vaginal fornix, both ovaries are visualized in relationship to surrounding structures and the aspiration strategy is planned.
