Specific to gynecology, the introduction of high-resolution transvaginal ultrasonography has substantially changed the approach to diagnostic of gynecologic conditions. Today, it is an essential tool in patients suffering from reproductive disorders for assessment of follicular growth pattern, the structure of perifollicular vascular network and endometrium enable close to monitoring and prediction of success of medically assisted reproduction.
Ultrasound and Follicular Growth
Antral follicles of different sizes are present during all phase of menstrual cycle. Commonly, a diameter of 2 mm is considered as the lower limit at which antral follicles might be visualized although with improvement of ultrasound machines even smaller follicles can be visualized today. Visible characteristics of antral follicles that can be used to predict the final fate of the follicle: size (largest diameter), shape, echogenicity and quality of antral edge.
In natural cycle, throughout the follicular phase, antral follicle which will gain dominance is usually regularly shaped with regular antral edge and larger than follicles undergoing atresia. Dominant follicle has echogenicity in the middle range, while follicles destined to become atretic generally display higher echogenicity. Early angiographic studies in the natural cycle showed that the main characteristics of the blood flow in the perifollicular tissue of non-dominant growing antral follicles were lower speed and higher resistance compared to the perifollicular blood flow of the dominant follicle. Dominant follicle undergoes remarkable changes during the last 7 days of its development. There is a marked increase of number and size of granulosa cells, and increase in perifollicular blood flow in the vessels of the teca. These changes can be visualized by ultrasound as an increase in diameter and volume of the follicle.
At the time of ovulation, the dominant follicle which has reached its maximal size ruptures, and shortly after, luteogenesis begins. Several sonographic parameters have been investigated as potential markers of ovulation:
Disappearance of dominant follicle or sudden decrease of its size (the most frequent sign of ovulation with sensitivity of 84%),
Increase of intrafollicular echogenicity,
Loss of follicular wall regularity,
Accumulation of free fluid in the pouch of Douglas.
Doppler studies of dominant follicle in natural cycle have showed that there is a marked increase of blood flow velocity in perifollicular vessels around the time of ovulation.
Ultrasound and Medically Assisted Reproduction
Early follicular FSH is the most widely used marker of ovarian reserve although it is well-known that it has a relatively limited value in predicting ovarian response. On the other hand, markers such as Anti-Müllerian Hormone (AMH), and antral follicle count (AFC), are today well established as one of the most reliable ovarian response predictors to exogenous stimulation. AFC is the most verified sonographic marker, but there are also other sonographic markers that have been investigated, such as ovarian volume and ovarian stromal blood flow.
Antral Follicle Count
The number of antral follicles (AFC) is a better predictor of ovarian response to controlled ovarian stimulation compared to ovarian volume or age alone: if antral follicle count is less than three, there is a significantly higher chance of cycle cancellation, detection of lower estradiol (E2) levels, and use of higher doses of gonadotropins. Recently introduced sonography-based automated volume count (SonoAVC) algorithm allows for automatic follicle count on ovaries recorded by ultrasound; this approach enables lower intra- and interobserver variability, reduces the risk of measurement error, and consequently, could further improve the predictive capability of AFC in women undergoing ovarian stimulation.
Despite its early promising predictive importance, it is nowadays accepted that ovarian volume has a relatively limited predictive value of ovarian response to stimulation. Evidence of the literature and 3D ultrasound technology has shown that there was no statistically significant difference in ovarian volume between low responders and controls on day 3 of the cycle.
Ovarian Blood Flow Assessment
Stromal blood flow is another aspect of ovarian physiology that was made possible to investigate with the introduction of Doppler technique. Commonly used parameters were stromal peak systolic velocity (PSV), pulsatility index (PI), resistance index (RI), and, lately, vascularization index (VI), flow index (FI), and vascularization flow index (VFI). Early investigations have also suggested the value of ovarian stromal blood flow parameters in predicting the success of MAR, but other have failed to prove their predictive ability. Ovarian stromal vessels are thin and torturous, and it is impossible to obtain the angle between the ultrasound beam and the intraovarian vessels accurately.
The introduction of 3D, and power Doppler angiography, represented significant improvement in terms of ovarian blood flow assessment. In the study using 3D ultrasound, it was demonstrated that the mean ovarian stromal flow index (FI) was an important predictor of ovarian response to controlled stimulation. Similar findings of lower VI, FI, and VFI in poor responders were confirmed in another study. In contrary to these findings, it has been shown that all three indices of vascularity (VI, FI, VFI) were significantly increased during gonadotropin stimulation in the group of normal responders compared to the low ovarian reserve group. However, only AFC and none of the blood flow indices is identified as independent predictor of ovarian response, and IVF outcome.
Ultrasound Monitoring in Controlled Ovarian Stimulation
Sonographic evaluation of follicle growth, and the number of follicles, can influence the dose of gonadotropins as well as the timing of hCG administration for triggering final oocyte maturation.
Maturity of oocyte is closely associated with the follicle size and the serum estradiol levels. Common approaches to controlled ovarian stimulation monitoring involve baseline ultrasound and estradiol measurements. Afterwards, the approaches vary and usually combine ultrasound and estradiol monitoring, but the target remain the same: obtaining an adequate number of follicles with the estradiol levels that are consistent with the follicle cohort (serum E2 of at least 200 pg/mL per follicle measuring ≥14 mm).
Ultrasound Evaluation of the Endometrium
For any assisted procedure, optimally primed endometrium is essential for successful implantation. The development of the endometrium, leading to the susceptibility of the endometrium during window implantation, requires the subtle interaction of an extremely large number of factors. Morphological characteristics of endometrium depend on the circulating levels of estrogen and progesterone. Transvaginal ultrasonography can provide insight to the state and development of the endometrium and the capability to be receptive for embryos. Anterior and posterior myometrial-endometrial interfaces are easily recognized in all phases of the menstrual cycle, at the end of menstrual phase, the endometrium appears as a thin, hyperechoic line.
During proliferative phase, it becomes thicker (double layer endometrial thickness is normally 5-12 mm) and less echogenic. As early as day 6 and as late as 1 day before the LH peak, the sonographic picture of endometrium changes to the “triple stripe” pattern. At the time of ovulation, endometrium thickness is 10-16 mm. After ovulation, in the secretory phase, characteristic “triple strip” disappears as the consequence of progesterone influence. Endometrium becomes homogenous and hyperechoic. Thickness of endometrium increases only slightly. As a contrast to hyper-echogenicity of endometrium, observer could also identify a hypoechoic band at the interface with the myometrium. Endometrial thickness is defined as the distance between the anterior and posterior stratum basalis layers. Its measurement should be performed in the sagittal plane.
Endometrial thickness measurement is an essential part of evaluation of patients with fertility problems. Although multiple studies investigated the predictive value of endometrial thickness and sonographic appearance in IVF cycles, conclusions remain controversial. A recent review has shown that although frequently used cut-off of 7 mm is related to lower chance of pregnancy, the discriminatory capacity of this value in the prediction of pregnancy is virtually absent. There are also other echo-graphic aspects of endometrium, such as homogenous appearance on the day of hCG administration and uterine artery pulsatility index higher than 3.0 that were associated with poor results of stimulated IVF/ICSI cycles. While some investigators suggested that excessive endometrial growth (greater than 14 mm) was also shown to be poor prognostic indicator, others refused this conclusion.
The importance of Doppler angiographic indices as tools for evaluation of endometrium has also been studied: cycle-dependent changes in uterine blood flow (VI and PI of uterine artery) are evident, but their predictive value is limited due to diurnal variations and difference between the two uterine arteries (ipsilateral or contralateral to the dominant follicle).