Follicular Growth

div class=”ChapterContextInformation”>

© Springer Nature Switzerland AG 2020
A. Malvasi, D. Baldini (eds.)Pick Up and Oocyte

8. Monitoring Follicular Growth

Maria Elisabetta Coccia1  , Francesca Rizzello2   and Eleonora Ralli3  

Department of Obstetrics and Gynecology, University of Firenze, Florence, Italy

Assisted Reproduction Center, Careggi University Hospital, University of Florence, Florence, Italy

Department of Clinical and Experimental Biomedical Sciences, University of Florence, Florence, Italy



Maria Elisabetta Coccia


Francesca Rizzello (Corresponding author)


Eleonora Ralli


Ovarian stimulationOvarian reserveIVF/ICSIFollicleUltrasound monitoring

Monitoring follicular growth usually involves a combination of hormonal assays and ultrasonic measurements of follicle size (Fig. 8.1). Nowadays, Transvaginal Ultrasound (TVUS) is the “gold standard” for monitoring ovarian ovulation in both spontaneous and stimulated cycles.


Fig. 8.1

Monitoring follicular growth usually involves a combination of hormonal assays and ultrasonic measurements of follicle size

Accurate follicular monitoring is particularly important in In Vitro fertilization (IVF) plus Intracytoplasmic Sperm Injection (ICSI) program in order to:

  • predict ovarian responsiveness to exogenous gonadotropins (it allows to individualize treatment protocols optimizing results and reducing complications)

  • evaluate the number, size of developing follicles and their growth pattern before oocyte retrieval (predictive of oocytes competence)

  • estimate the appropriate time to trigger the final oocyte maturation before ovum pickup

  • assess the risk of ovarian hyperstimulation syndrome (OHSS)

  • make decisions on early cancellation of cycles without proper ovarian response, thus avoiding unnecessary waste of resources

During a cycle of IVF/ICSI, women receive daily doses of gonadotropins (FSH, LH, hMG) to induce multifollicular development in the ovaries. Controlled ovarian hyperstimulation (COH) induces the growth of heterogeneous cohorts of follicles containing oocytes at different stages of maturity and competence and only few oocytes will be able to undergo meiosis, fertilization, and embryo development.

The proportion of competent oocytes is directly related to follicular size. Developmental competence refers to the oocyte’s ability to mature, fertilize, and finally yield viable offspring [1].

In order to predict ovarian response, clinicians individualize the most suitable starting dose of gonadotropins on the basis of patient characteristics predictive of ovarian response such as age, clinical history, and other ovarian reserve tests (ORTs).

Nowadays, antral follicular count (AFC) and Anti-Müllerian hormone (AMH) appear to be the most useful markers of ovarian reserve in addition to chronological age. The retrieval of 5–15 oocytes is often considered a normal response to stimulation [2]. Poor and hyper-response to COH are significantly correlated with risk of cycle cancellation. Moreover, hyper-response is associated with increased risk of OHSS.

In this chapter, monitoring follicular growth during IVF/ICSI cycle will be presented.

8.1 Controlled Ovarian Stimulation: The Rational Basis

According to the model proposed by Baird (1987), during the menstrual phase the concentration of FSH rises to a level high enough to activate a single small antral follicle (2–4 mm) that secretes higher quantities of Estradiol (E32). The production of E2 and inhibin by the developing follicle is able to suppress the concentration of FSH below this threshold level (negative feedback). For a critical period, the dominant follicle becomes increasingly sensitive to FSH and consequently is less dependent on the level of plasma FSH. These conditions led to the growing of the dominant follicles and the inhibition of other follicles [3].

During COH, treatments with Clomiphene or gonadotropins maintain the period during which the level of FSH remains above this threshold allowing multiple follicular development. Under this condition, the follicles continue to grow at approximately the same rate, thus they are never completely synchronous (Fig. 8.2).


Fig. 8.2

FSH threshold concept in a stimulated cycle. During COH, levels of circulating FSH are elevated above threshold. The number of preovulatory follicles increases in an asynchronous manner

It has been demonstrated that FSH stimulates granulosa cells of early antral follicles and induces LH receptor formation on preovulatory follicle [4]. LH activates adenyl cyclase with the resulting production of cAMP that represents an amplified response to FSH [5]. Therefore, the maturing follicle reduces its dependency on FSH by acquiring LH receptors. The granulosa cells from early antral follicles are only responsive to FSH; granulosa cells from FSH-stimulated follicles are responsive to both FSH and LH [6].

The revised two-cells—two gonadotropins theory suggests a key role for LH not only in theca cells steroidogenesis stimulation, but also in follicle growth and maturation in the mid-luteal phase [7] (Fig. 8.3).


Fig. 8.3

The two-cells–two gonadotropins theory

Ovarian stimulation protocols combine the use of human menopausal gonadotropin (hMG), urinary or recombinant FSH, recombinant LH with GnRH agonists or antagonists in order to increase oocyte number and to avoid premature LH surge. Given the physiological role of FSH and LH, the rational basis of controlled ovarian stimulation is to increase the duration that serum FSH concentrations are maintained above the threshold by direct administration of FSH.

8.2 Prediction of Ovarian Response

During an IVF/ICSI cycle, daily doses of gonadotropins (FSH, LH, hMG, Corifollitropin alfa, Follitropin delta) are used to induce multifollicular response in the ovaries. Although the number of eggs retrieved seems to depend on the starting/total doses of gonadotropins, individual woman’s response differs [8, 9].

A low or poor ovarian response has been defined as the retrieval of three or fewer oocytes [10]. A hyper-response (or high response) is often described as the retrieval of 15–20 or more oocytes and is associated with an increase in the risk of OHSS [11, 12].

During fresh IVF cycles, some authors observed that the best chance of live birth was associated with the number of 5–15 eggs retrieved during oocyte pickup. Whereas a decline was observed after the retrieval of 20 or more oocytes [9].

In an effort to predict the response and outcome in couples prior to IVF/ICSI and counsel them, the estimation of ovarian reserve is routinely performed through various ovarian reserve tests or multifactorial algorithms [13].

Basal FSH (b-FSH), measured in serum in the early follicular phase of a menstrual cycle, is a simple and reliable test. It is less expensive than the other tests, easily applicable and is, therefore, the most widely used screening test in infertility programs. Previous studies demonstrated that a combination of early follicular FSH and age seems to be better than age alone in predicting outcome in women undergoing IVF [14]. B-FSH was later integrated with the AFC. AFC is measured by ultrasound and is a count of the number of antral follicles measuring about 2–10 mm (according to standard criteria) that are available in both ovaries [15].

More recently, AMH has emerged as a robust marker of ovarian function. It is a dimeric glycoprotein produced by granulosa cells of preantral (primary and secondary) and small antral follicles in the ovary.

The production of AMH starts following follicular transition from the primordial to the primary stage, and it continues until the follicles reach the antral stages, with diameters of 2–6 mm [16] (Fig. 8.4).


Fig. 8.4

Follicular growth sensitive to FSH and AMH

AMH represents a more direct and independent measure of the growing preantral and antral follicular pool and is not influenced by the menstrual cycle or pregnancy [17, 18]. The current evidence shows that AFC and AMH appear the most useful markers of ovarian reserve in addition to chronological age [19, 20].

According to the above-mentioned criteria, patients might be classified as predicted hyper-, normal, or hyporesponders:

8.3 Predicted Low Responders

According to Bologna criteria, two of these three criteria are required to define a patient as poor ovarian responder:

  1. 1.

    advanced maternal age (≥40 years) or any other poor ovarian response risk factor (genetic or acquired conditions, pelvic infection, ovarian endometriomas, and patients who have undergone ovarian surgery for ovarian cyst, chemotherapy, shortening of the menstrual cycle)


  2. 2.

    a previous cycle with poor ovarian response (≤3 oocytes with a conventional stimulation protocol)


  3. 3.

    a low ovarian reserve test in terms of AMH (<0.5–1.1 ng/mL) (<3.6–7.8 pmol/L) and AFC (<5–7 follicles)


Moreover, two cycles with poor ovarian response after maximal stimulation might classify a patient as a poor responder even in the absence of the other criteria [10]. More recently, the Poseidon Group (2016) suggested a more specific definition of “low prognosis” patients that introduces two new categories of compromised response: “suboptimal response” and “hyporesponse.” “Suboptimal responders” include women with retrieval of 4–9 oocytes which is associated to a significantly lower live birth if compared to patients normoresponders (10–15 oocytes), regardless of age. On the other hand, when higher dose of gonadotropins and more prolonged stimulation are required to obtain more than three oocytes, patients are defined “hyporesponders” [21, 22] (Fig. 8.5). Lensen et al. (2018) in the last Cochrane reviews on individualized gonadotropin dose selection used the following cutoffs to categorize women as predicted low responders: AMH < 7 pmoL/L (0.98 ng/mL), AFC < 7, bFSH > 10 IU/L [2].


Fig. 8.5

Differentiation into groups of low responders according to Poseidon criteria

8.4 Predicted Hyper-Responders

  1. 1.

    age < 35 years old


  2. 2.

    low Body Mass Index (BMI) (<18.5 Kg/m2)


  3. 3.

    higher AMH levels (cutoff value 3.4 ng/mL)


  4. 4.

    AFC ≥ 24


  5. 5.

    polycystic ovarian syndrome (PCOS) (Fig. 8.6a, b)


  6. 6.

    history of OHSS



Fig. 8.6

(a, b) Schematic representation of micropolycystic ovary and its echographic appearance

Lensen et al. (2018) adopted only AMH > 21 pmol/L (2.9 ng/mL) and AFC > 15 as predictors for high responders [2].

8.5 Predicted Normal Responders

Categorization of normal responders is frequently based on the exclusion of a poor or a hyper-response to ovarian stimulation, rather than by using specific inclusion criteria. Unfortunately, these criteria will include also patients with different prognosis (optimal and suboptimal) [21].

  1. 1.

    AFC 8–15


  2. 2.

    age 35–39 years


  3. 3.

    AMH >1.2 and <3.75 ng/mL


Lensens (2018) adopted the following inclusion criteria: AMH 7–21 pmol/L, AFC 7–15 categorized as predicted normal responders (bFSH was not included as it is not a reliable predictor for normal response) (Lensens 2018) (Table 8.1).

Table 8.1

Criteria to categorize women into poor, hyper-, and normoresponders

Predicted response



Poor responders

Bologna Criteria (Ferraretti 2014)

1. Advanced maternal age (≥40 years)

Or any other poor ovarian response risk factor

 – Genetic or acquired conditions

 – Pelvic infection

 – Ovarian endometriomas

 – Ovarian surgery for ovarian cyst

 – Chemotherapy

 – Shortening of the menstrual cycle

2. A previous cycle with poor ovarian response

– ≤3 oocytes with a conventional stimulation protocol

3. A low ovarian reserve test

– AMH (<0.5–1.1 ng/mL) (<3.6–7.8 pmol/L)

– AFC (<5–7 follicles)

– Two previous cycles with poor ovarian response after a maximal stimulation (even in the absence of the other criteria)

Poseidon Group (2016)

– AMH (<1.2 ng/mL) (<8.6 pmol/L)

– AFC (<5 follicles)

Or unexpected poor or suboptimal ovarian response

Poor response: retrieval of <4 oocytes

Suboptimal response: retrieval of 4–9 oocytes

Hyporesponse: higher dose of gonadotropins and more prolonged stimulation to obtain more than 3 oocytes

Lensen et al. (2018)

AMH < 7 pmol/L, AFC < 7, bFSH > 10 IU/L, it would be better numbering the list…)


ASRM (2016); Kwee J (2007)

1. Age < 35 years old

2. Low Body Mass Index (BMI) (BMI < 18.5 kg/m2)

3. Higher AMH levels (cutoff value 3.36 ng/mL, 24 pmol/L)

4. AFC ≥ 24

5. Polycystic ovarian syndrome (PCOS)

6. History of OHSS

Lensen et al. (2018)

1. AMH > 2.9 ng/ml (>21 pmol/L)

2. AFC > 15

Normal responders

Based on the exclusion of a poor or an hyper-response to ovarian stimulation

1. AFC 8–15

2. Age 35–39 years

3. AMH > 1.2 ng/mL (8.6 pmol/L) and <3.75 ng/mL (26.8 pmol/L)

Lensens (2018)

1. AMH 0.98–2.9 ng/mL, (7–21 pmol/L)

2. AFC 7–15

8.6 Protocols of Ovarian Stimulation

The main objective of a correct prediction of ovarian response is the personalization of ovarian stimulation to induce multifollicular development in the ovaries. Personalized treatment offers every single patient the best treatment tailored to her own unique characteristics, thus maximizing the chances of pregnancy, eliminating the iatrogenic and avoidable risks, such as OHSS, minimizing the risk of cycle cancellation and enabling clinicians to give women more accurate information on their prognosis [20].

Finally, and not least importantly, personalized therapy showed an important reduction of costs probably due to a reduced incidence of OHSS, drug consumption, and cancellation of cycles [23].

Dose increases during ovarian stimulation do not seem to affect the number of retrieved oocytes [24, 25]; therefore, the choice of an appropriate starting dose appears crucial for the final ovarian response.

Diverse therapeutic protocols have been suggested for the each category of patients. Usually, the gonadotropin starting dose is 150 IU for expected high responders, 200–225 for normal responders, and 300–450 IU for expected poor responders [2].

In recent times, clinical practice of infertility treatment is moving from standardized to individualized FSH dosing. Actually, new FSH preparations, as follitropin delta, integrate individualized dosing as part of the clinical managing. The dosing algorithm for follitropin delta aims to individualize the starting dose to each woman based on serum level of AMH and body weight, staying on the same dose throughout the whole stimulation cycle [26].

8.7 Prediction of Oocyte Competence Through Monitoring Follicular Growth

TVUS follow-up of follicular growth is crucial for ART treatment. Counting and measuring growing follicles allows optimal dosage of hormones and correct timing of administration of the ovulation trigger. Early detection of ovarian hyperstimulation allows to freeze embryos and transfer frozen-thawed embryos later.

The evaluation of growing follicles by ultrasound includes (Fig. 8.7): follicle diameter, follicle growth pattern, follicular wall thickness, perifollicular vascularity, and perifollicular blood flow. A follicle that is >10 mm in diameter at the first scan, grows at a rate of 2–3 mm per day, has no internal echogenicity and has thin (pencil line) wall, is more likely to become a leading follicles [27].


Fig. 8.7

The evaluation of growing follicles by ultrasound includes: follicle diameter, follicle growth pattern, follicular wall thickness, perifollicular vascularity, and perifollicular blood flow

Whether the follicles that are visualized represent potentially healthy follicles with competent oocytes has not been established. In addition, there is considerable variability in the clinical definition and technical methodology used to count and measure follicles in both published studies and clinical practice. As a further complication, more than one individual may scan a patient during one cycle.

8.8 Monitoring Follicular Growth

COH is monitored by serial TVUS plus serum E2 measurements performed every second day from stimulation day 5–6. The frequency of monitoring will depend on the biophysical parameters of follicular growth and hormonal parameters, principally E2 levels. The wall of mature ovarian follicles is a complex structure with endocrine function. It is constituted by two cell layers. The outer layer is comprised of two types of cells and forms the theca: the theca interna contains cuboidal cells, synthesizing the androstenedione and is well vascularized; the theca externa mainly consists of connective tissue. The inner layer of the follicular wall is termed the granulosa and is composed of stratified cells which convert androstenedione into E2 through the action of the aromatase. The limit between the thecal layers and the granulosa is well defined histologically by a thick basal lamina [28]. The follicle wall is defined as the combined thicknesses of the stratum granulosa and theca. In order to measure the follicle size by the 2D-transvaginal scanning, the vaginal probe is placed as close as possible to the follicle to make follicle borders visualize clearly. The mean follicular diameter is calculated based on the mean of the two longest diameters. The two orthogonal diameters (d1 and d2) should be determined at the scanning plane corresponding to the largest follicle diameter, by placing the calipers on the fluid–follicle interface [29] (Fig. 8.8). Data on preferable placement of calipers on follicle borders (inner, outer, interface) are scarce. Some authors measure follicular size using the internal diameters of the area [30]. The more accurate way of measurement will be probably clarified through studies on 3D ultrasound [31].


Fig. 8.8

Follicle size by the 2D-TVUS. The two orthogonal diameters (d1 and d2) should be measured at the scanning plane of the largest follicle diameter

8.9 First Scan Under COH

Ovarian response can be identified early in the cycle. The first scan under COH for IVF-ICSI is performed on stimulation day 6–7.

Hodgen et al. classified ovarian response in the following way: “non-responders” as patients whose E2 levels did not reach 300 pg/mL by day 8 of stimulation; “slow responders” E2 levels were <300 pg/mL by day 5, but >300 pg/mL by day 8 of stimulation; “fast responders” with E2 levels >300 pg/mL by day 5 of stimulation [32] (Fig. 8.9).


Fig. 8.9

Patients classification on day 8 of stimulation

Although debated, the practice of adjusting the dose on day 6–7 of stimulation, when cycle monitoring indicates that the ovarian response has been unsatisfactory, still represents a common practice. On the basis of both TVUS findings and E2 levels, gonadotropin dosages are adjusted up or down with 450 IU as the maximum daily dose allowed. How effective such dose increase or reduction is, is not well established [33]. Both very slow and very rapid estrogen growth rates, as calculated from the 4 days preceding oocyte retrieval were associated with a reduced pregnancy rate [34].

According to some authors, the dose of gonadotropin should not be altered as long as serial E2 levels increase between 50 and 100% every other day [35]. Suggested steps of dose adjustments ranges from 25–50 IU/day [36] to 75–100 IU/day. The dosage should be reduced by 75 IU if two consecutive E2 levels rise by >100% (Silverberg et al., 1991). From stimulation day 6–7 onward, COH continues until at least one dominant follicle reaches 18 mm diameter, with appropriate E2 levels (Fig. 8.10a, b).


Fig. 8.10

Stimulation day 6, follicles measuring less than 14 mm: (a) POR patient; (b) woman with PCO

8.10 Follicle Diameter and Anatomopathological Considerations

During follicular monitoring, clinicians should keep in mind that the size of follicles is related to the stage of oocytes. Prior to the widespread use of transvaginal sonography, morphometric data derived from studies of the entire follicular population of normal human ovaries (ovariectomy for carcinoma of the breasts or cervix, hysterectomy for fibroids) obtained at various stages of the menstrual period, allowed the development of a theoretical model of progression of follicles from the preantral to the ovulatory stage [37]. Follicular dimensions were measured with an ocular micrometer.

Histological examination of the human ovary revealed a great variety in the size of follicles from the preantral (0.1 mm) to the ovulatory stage (20 mm). Based on the number of granulosa cells, the entire follicular population was divided into classes representing the stages of follicular development. At each follicle diameter corresponded a number of granulosa cells (GC) (Figs. 8.11 and 8.12).


Fig. 8.11

Stages of folliculogenesis in the adult human ovary


Fig. 8.12

Stage of follicles maturation

The numbers of preantral follicles throughout the cycle is more or less constant. Thus, the entry of follicles into this class is continuous during the cycle on the basis of a succession of waves, or cohorts, of growing follicles. Therefore, if an ovary is observed at any given point during the cycle, a large number of follicles of different diameter would be predictable, especially in young women (Table 8.2). In the early luteal phase (days 15–19), immediately after high levels of 17β-estradiol, the gonadotropins start the growth of follicles by the differentiation of the epitheloid cells in the theca interna (preantral follicles, 0.1–0.2 mm diameter).

Table 8.2

Follicular size and visualization by ultrasound


Follicular size









Early antral






Ovarian follicles larger than 2 mm in diameter can be observed by TVUS

The entry of the follicles into class 2 (0.2–0.4 mm diameter with the appearance of an antrum) is observed 25 days later simultaneously in both ovaries. Afterwards follicular growth becomes more complex with an asymmetry between the two ovaries.

Ovarian follicles larger than 2 mm in diameter can be observed by TVUS and are highly sensitive to gonadotropins. Follicles of 2–10 mm are related to the amount of the primordial follicle pool. Unfortunately, the same range 2–10 mm include also follicles at early stages of atresia and thus poorly responsive to gonadotropin.

As a consequence, the AFC tends to overestimate the number of gonadotropins responsive follicles. Healthy antral follicles tend to be 4–6 mm in diameter and might best represent the age-dependent proportion of the antral follicle pool. However, measuring the diameter of every follicle would require very long time.

The largest healthy follicles measure 2 and 5 mm in diameter at the end of the luteal phase and show a higher mitotic activity than the mid-luteal phase. The follicle destined to ovulate in the next cycle will be recruited among these follicles in the early follicular phase (days 1–5).

Its diameter is between 5.5 and 8.2 mm and increases significantly through cellular proliferation and accumulation of fluid in the antrum [38, 39]. During 2 weeks, it grows from 6.9 ± 0.5 mm (days 1–5) to 13.3 ± 1.2 mm (days 6–10) and then to 18.8 ± 0.5 mm (days 11–14).

Hypertrophy of the theca interna and its vascularization begins in the mid-follicular phase and continues until the ovulatory LH surge [38]. The granulosa shows a characteristic organization at the time of the plasma peak of 17β-estradiol and then becomes disorganized during the pre-luteinization of the follicle [40].

8.11 Follicle Sizes on the Day of Trigger

Since 1973, Edwards highlighted a strong association between follicle size and oocyte recovery [41]. In spontaneous cycles, preovulatory follicles reach the diameter of 17–25 mm in diameter [42]. During COH, the relationship between follicle size and oocyte maturity might be different.

It is more likely that a large follicle contains a free-floating cumulus-oocyte complex, compared to a small one [4346]. In literature, data about ideal follicle size at the time of hCG are controversial and demonstrate the difficulty in defining a universally accepted threshold of follicle dimension to predict the presence of good competent oocytes. Generally, the incidence of mature oocytes increases with follicular size [47]; however, both too small and too large follicles are associated with low oocyte recovery, fertilization, and cleavage rates.

Table 8.2 summarizes data on oocyte competence and follicular size (Table 8.2). Oocytes are frequently found in the metaphase II (MII) when retrieved from follicles larger than 16 mm (approximately corresponding to a 2 mL volume). Some authors consider 18 mm as a better cutoff to discriminate follicles with higher chances of mature oocytes [48, 49].

In follicles below 12 mm a higher proportion of immature, germinal vesicle (GV) or metaphase I (MI) oocytes is found and some authors recommend not aspirate this pool of follicles [48, 50, 51].

Another study, reported that follicles from 11 to 15 mm sometimes can harbor mature oocytes [52]. Both in normal and PCO ovaries, follicles smaller than 14 mm diameter occasionally might contain MII oocytes [4951, 5356].

Discordant findings on the relationship between follicular size and embryo cleavage rates were reported [48, 51, 57, 58] (Table 8.3).

Table 8.3

Human studies on follicle size and IVF outcomes


Study design

Age (years)

COH protocol


Number of oocytes

Mean diameter at HCG

Mean diameter at oocyte retrieval


Fertilization rate

Cleavage rate

% good quality embryos

Wittmaack, 1994




Only IVF


From 16 to 18 mm

21 mm

Worst <12 mm and >24 mm

Increasing from 12 mm onwards

Increasing from 12 mm onwards


Inaudi, 1995




Only IVF


At least 3 foll > 16 mm


None in the recovery rate




Dubey, 1995




Only IVF


At least 2 foll > 20 mm



Increasing from 16 mm onwards



Ectors, 1997








IVF: n.e.

IVF: increasing from 16 to 23, later decreasing

IVF: increasing from 16 to 23, later decreasing

IVF: increasing from 16 to 23, later decreasing

Bergh, 1998


33 mean







IVF: increasing

IVF: none

n.e. but increasing pregnancy rate only in IVF

Teissier, 2000


28 mean


Only IVF








Triwitayakorn, 2003






At least 2 foll of 18 mm


Always increasing together with oocyte recovery rate




Rosen, 2008






At least 2 foll of 18 mm


IVF: n.e./ICSI: always increasing

IVF and ICSI: always increasing


IVF and ICSI: always increasing

Lee, 2010






When 2–3 foll >18 mm


Increasing from 18 mm onwards




Mehri, 2013






At least 2 foll of 20 mm


Increasing from 18 mm onwards

Increasing from 18 mm onwards



Wirleitner, 2018




Only IVF


At least 2 foll of 18 mm

13–23 mm


Similar results

Similar results

Similar results

Modified from Revelli (2014)

Only gold members can continue reading. Log In or Register to continue

Mar 28, 2021 | Posted by in OBSTETRICS | Comments Off on Follicular Growth
Premium Wordpress Themes by UFO Themes