4.1 Introduction

In ovarian stimulation cycles for IVF/ICSI, a defective luteal phase occurs in almost all patients as a result of the multifollicular development and supraphysiological hormonal levels, which inhibit the LH secretion by the pituitary via negative feedback actions at the level of the hypothalamic-pituitary axis. Progesterone is crucial to induce secretory transformation of the endometrium after previous estrogenization, a process which will finally result in a receptive endometrium. Consequently, luteal phase support (LPS) represents an essential part of ART treatment in case of a planned fresh embryo transfer, as it is crucial to counterbalance the luteal phase insufficiency.

The endocrine profile of the luteal phase is influenced substantially from the medication, used for final oocyte maturation. Depending on the stimulation protocol and the ovarian response of the patient, either hCG, a GnRH agonist, or the combination of both are used for final oocyte maturation. The task of the reproductive medicine specialist is to tailor the LPS according to the patient´s specific characteristics and the stimulation protocol. Therefore this chapter aims to describe the physiology of the luteal phase in the natural cycle, the different endocrine profiles in the luteal phase of ovarian stimulation cycles, the existing approaches for LPS and the necessity, as well as the possibilities, to individualize LPS.

4.2 Physiology of the Luteal Phase in a Natural Cycle

The luteal phase of a menstrual cycle is the time between ovulation and, in case of conception, the establishment of a pregnancy or otherwise the onset of menses.

In a natural cycle, estradiol synthesis increases progressively from the dominant follicle and initiates LH surge. Even before the LH surge, a small increase in progesterone levels is seen, which reflects the increasing LH pulse amplitude and frequency leading up to the surge. A LH surge of 24–36 hours is sufficient to initiate the resumption of oocyte meiosis, luteinization of granulosa cells, ovulation, and the initial phase of corpora lutea development. Progesterone and 17α-hydroxyprogesterone (17α-OHP) plasma concentrations increase rapidly after the LH surge, indicating the beginning of granulosa and theca cell luteinization. The hormonal activity of the corpus luteum depends on the pulsatile LH secretion resulting in a production of up to 40 mg of progesterone per day and additional a significant amount of androgens and estradiol. LH pulses during the luteal phase are of a reduced frequency and of a greater amplitude than in the follicular phase and during the course of the luteal phase, the mean LH pulse frequency declines from 15.2 pulses/24 h in the early to 8.4/24 h in the late luteal phase (1). Also a reduction in the amplitude of LH pulses was also observed in the late luteal phase.

In case of a conception, the embryo will start to produce human chorionic gonadotropin (hCG) from day 7 of fertilization. hCG has structural similarities with LH and activates the same receptor. Therefore the hCG, produced by the embryo, will counterbalance the subsiding LH pulses and maintain the hormonal activity of the corpus luteum until the progesterone production shifts from the corpus luteum to the placenta at around 9 weeks of pregnancy. Without a conception, luteolysis will occur as a physiological process of corpus luteum regression. Functional luteolysis with reduced secretion of progesterone is followed by structural luteolysis.

4.3 Physiology of the Luteal Phase in Ovarian Stimulation Cycles for IVF/ICSI

In stimulated cycles for IVF, a defective luteal phase occurs in almost all patients (2). Initially it was thought that the defective luteal phase was due to the aspiration of a large number of granulosa cells during oocyte retrieval procedure. Later on it was demonstrated that the reason for the luteal phase insufficiency, seen after controlled ovarian stimulation, seems to be the multifollicular development achieved during the follicular phase. In IVF/ICSI treatment daily dosages of exogenous gonadotropins are administered to support multifollicular growth with the aim of maximizing the number of available oocytes for fertilization, resulting in a large number of growing follicles and supraphysiological hormonal levels. These supraphysiological levels inhibit the LH secretion by the pituitary via negative feedback actions at the level of the hypothalamic-pituitary axis (3).

Final oocyte maturation is the crucial step in ovarian stimulation cycles for IVF in order to retrieve mature oocytes for further processing in the IVF laboratory. Whereas in GnRH (Gonadotropin-releasing-hormone)-agonist cycles, only hCG can be administered for final oocyte maturation, in GnRH antagonist cycles, hCG as well as GnRH-agonists can be used. However, due to the different ways of action of hCG and GnRH agonist, the endocrine profile of the luteal phase in stimulated cycles depends substantially from the medication used for final oocyte maturation.

Table 4.1 summarizes the differences between the natural and the stimulated cycle.

4.4 Endocrine Profile of the Luteal Phase after HCG Trigger

As described above, under physiological conditions ovulation is induced by a rise in LH levels. In the early years of IVF treatment, purified LH was not commercially available at adequate doses to support final follicular maturation, therefore hCG was used instead routinely for final oocyte maturation. HCG is a glycoprotein hormone with the typical heterodimeric structure also exhibited by LH, FSH, and TSH. These hormones share a common α-subunit but have distinctly different ß-subunits. hCG and LH have 85 percent of the amino acid sequence of their ß-subunit in common. The main differences between them lie within the N-linked oligosaccharides and the C-terminal sequence. The latter, and especially the O-linked oligosaccharides in this peptide, are responsible for the longer half-life of hCG compared with LH. Nowadays hCG is available in urinary and recombinant form, both with a comparable half-life time of approximately 30 hours (4). HCG attaches to and activates the same receptor as LH and is therefore capable to induce final oocyte maturation. The most important difference between LH and hCG is the difference in half-life, which is 6–8 times longer for hCG compared to LH. The longer half-life of hCG is the crucial factor for the higher risk of ovarian hyperstimulation syndrome (OHSS), especially in the high-responder patient. When the hCG stimulus on the corpora lutea subsides, luteolysis will occur due to a lack of endogenous LH.

4.5 Endocrine Profile of the Luteal Phase after GnRH Agonist Trigger

Contrary to hCG, the GnRH-agonist acts by dislocating the GnRH antagonist from the GnRH receptors in the pituitary and its administration results in a surge of LH and FSH (so-called flare-up). This surge is sufficient to induce final oocyte maturation and ovulation (5). Despite the fact that the application of a GnRH agonist trigger in a GnRH antagonist protocol imitates the natural mid-cycle surge of LH and FSH and the magnitude of natural and GnRH agonist–induced LH surges are comparable, distinct differences between both are found. The spontaneous LH surge in a natural cycle is characterized by an ascending phase of approximately 4 hours, a peak plateau of 20 hours, and a descending phase of 20 hours whereas the GnRHa-induced LH/FSH surge has a significantly shorter ascending phase. The differences between natural and artificial induced LH surge cannot be overcome by an increase in the GnRH agonist dosage or by repeat GnRH agonist administration. As the duration of the LH/FSH surge is critical for a normal luteal function and an LH increment of too short a duration prevents the granulosa cells from completing luteinization, further on impairs the secretory function of the corpus luteum is commonly impaired and the life-span is shortened. At the same time, the supraphysiological levels of estradiol and progesterone after ovarian stimulation alter LH secretion from the pituitary via negative feedback mechanisms. Pulsatile LH secretion from the pituitary continues, however the mean LH concentration and LH pulse amplitude are lower compared to the natural cycle. Hence, the process of luteolysis starts very early in the luteal phase which is demonstrated by the decline of progesterone and estradiol levels two days after ovulation (6). Previously it was believed that the luteal phase is always characterized by severe luteolysis and luteal phase insufficiency as a result of the short duration of the induced LH/FSH peak after GnRH agonist administration (7). Interestingly, luteolysis after GnRH-agonist trigger is not always complete and may vary, indicating individual differences among patients. It seems that longer stimulation duration as well as a higher level of progesterone on the day of final oocyte maturation and more retrieved oocytes will result in higher levels of progesterone 48 hours after oocyte retrieval, pointing toward a sustained secretory capability of the corpora lutea even after GnRHa trigger (8).

In patients, considered to be at risk for OHSS development, final oocyte maturation by administration of a GnRH agonist in a GnRH antagonist protocol is meanwhile considered as the “gold standard” as it leads to a significant reduction in the incidence of OHSS (9). Hence it has to be noted, that even GnRH agonist trigger does not completely avoided development of OHSS. Table 4.2 presents the different approaches for final oocyte maturation and their implications for the luteal phase.

4.6 Progesterone and Endometrial Receptivity

Progesterone induces the secretory transformation of the endometrium after previous estrogenization, a process which is dependent from the time of progesterone exposure and which will finally result in a receptive endometrium.

Endometrial receptivity, a good quality, euploid embryo at blastocyst developmental stage and the synchrony between both are mandatory for successful implantation (10). The implantation process is characterized by the apposition of a blastocyst-stage embryo to a receptive endometrium, the attachment to the maternal endometrial epithelium and finally the invasion into the endometrial stroma. It is well known that an embryo has the ability to implant in different tissues outside the uterine cavity, nevertheless, implantation in the endometrium can only occur when the endometrium is receptive. This phase is nowadays referred to as the “window of implantation” (WOI). It is interesting to note that this nowadays well recognized and often used phrase was first described long before the first successful IVF treatment (11).

Due to ethical considerations, the exact duration of the WOI is difficult to define in humans. In an idealized 28 day cycle, it is assumed that the WOI takes place between day 19 and day 21 of the menstrual cycle, however different durations are described, from 48 hours up to possibly 4 days (12, 13). The methods taken to identify endometrial receptivity in the human endometrium have changed over the last decades. Changes of the endometrium, occurring under the influence of progesterone, have been described histologically already in 1950. Later on, it was demonstrated that the endometrium at the time of receptivity expresses proteins, thought to facilitate and promote implantation and a family of cell surface glycoproteins, so-called integrins, were identified. Their expression in the endometrium appeared to be cyclical during the menstrual cycle and particularly “up-regulated” during the WOI. Currently endometrial receptivity is investigated by gene expression patterns (14).

4.7 Luteal Phase Support in ART Cycles

To counterbalance the luteal phase insufficiency after ovarian stimulation and to maintain the secretory endometrium, an adequate luteal phase support is crucial and a must after ovarian stimulation in case of a planned fresh embryo transfer.

In a natural cycle, luteal phase deficiency (LPD) was defined as having mid-luteal progesterone levels below 10 ng/mL (31.8 nmol/L) or a sum of three random serum P measurements < 30 ng/mL (95.4 nmol/L) (15). The correction of a dysfunctional corpus luteum with the symptoms shortened luteal phase and premenstrual spotting, by administration of progesterone, was first described in 1949 (16) and a defective luteal phase was defined, if the serum mid-luteal progesterone levels are less than 10 ng/mL. Prevalence of luteal phase defect (LPD) in natural cycles in normo-ovulatory patients with primary or secondary infertility was demonstrated to be about 8.1 percent. However in 2012, the ASRM stated that there is no reproducible, physiologically relevant, and clinically practical standard to diagnose LPD or distinguish fertile from infertile women (17). The minimum threshold of progesterone level that is essential for the maintenance of a pregnancy is unknown and successful pregnancies have been reported even when the concentration of progesterone was never above 15 nmol/L for the first 14 days.

In IVF treatments, the lower limit of progesterone levels to achieve and maintain a pregnancy is not defined yet. It seems that a progesterone level of more than 30 ng/mL and an estradiol level of more than 100 pg/mL at the day of implantation are more likely to have a viable and ongoing pregnancy compared to patients with hormone levels below these thresholds. Other publications report successful pregnancies with mid-luteal progesterone levels above 17 ng/mL or even as low as a progesterone level above 15 ng/mL two days after GnRH-agonist trigger, provided that adequate luteal phase support was applied.

In the last years, there is an increasing trend to tailor treatment approach according to the patient´s characteristics as a consequence of the increasing knowledge on interpatient variability. This knowledge challenges the old concept of one type of luteal phase support for all IVF patients and advocate to implement a personalized approach into daily clinical routine, which depends on the type of final oocyte maturation and the ovarian response.