Minimal Stimulation and Natural Cycle In Vitro Fertilization, 1st ed. 2015

7. Ovarian Stimulation in Patients with Ovarian Insufficiency

Markus Nitzschke 


Department of Assisted Reproduction, Obstetrics and Gynecology, ICI – Instituto Canario de Infertilidad, Calle León y Castillo, 294, Las Palmas de Gran Canaria, C. P. 35005, Spain

Markus Nitzschke



In general, patients with low ovarian reserve are difficult to manage and have a relatively poor prognosis. Hormonal changes due to ovarian insufficiency can influence and change the individual menstrual cycle pattern of each patient over time, which may result in difficulties to conceive naturally. Observing the menstrual cycle pattern of patients with low ovarian reserve, it is possible to distinguish different stages of ovarian insufficiency. All clinical stages can be treated by a simple medical treatment, based on oral contraceptive pills, ethinyl estradiol, and Clomiphene citrate, with or without the combination with modified natural cycle in vitro fertilization (IVF). Experience shows that ovulation can successfully be controlled by the use of Clomiphene citrate and does not necessarily require gonadotropin-releasing hormone (GnRH) analogs for pituitary suppression. This knowledge opens a new space for development of alternative protocols respecting the patients’ own physiology with no need for heavy stimulation. Patients with ovarian insufficiency may benefit from this approach, which can be offered before referring them to egg donation.


Low ovarian reserveClomiphene citrateNatural cycle IVF


Treating infertility patients with low ovarian reserve is one of the most frustrating and challenging tasks that infertility specialists can encounter in their practice. These patients usually show strange menstrual cycle patterns. They do not respond normally to ovarian stimulation regimens, and they often develop treatment-resistant ovarian cysts. The percentage of infertility patients with low ovarian reserve is growing in all infertility practices in the developed world, due to a growing older population, and the decision of many women to start reproduction in their mid-30s. In some areas, especially in bigger cities, the percentage of patients with low ovarian reserve exceeds 50 % in most infertility practices. Despite the urgent need for new treatment regimens for this patient population, very few groups are undertaking serious research in this field. For lack of better options, many infertility specialists simply refer those patients to egg donation.

Inspired by a publication from Japan (Teramoto and Kato 2007), our group started to look for alternative treatment options for this particular patient group. We used the most empirical approach possible. We went back to pure observation of our patients’ cycles, trying to understand their individual hormonal pattern. After observation of hundreds of menstrual cycles, we were able to classify four different types of ovulation in women with low ovarian reserve. We called them “compensated stage,” “stage of desynchronized menstrual cycle,” “stage of pathologic premature LH rise,” and “stage of suppressed follicle growth.” In a second step, we used an adaptive treatment approach on each ovulation group. The different treatments were based on our personal experience, and different approaches described in the literature, mainly those of Teramoto (Teramoto and Kato 2007).

The stimulation protocol for IVF, which has been described by Teramoto and Kato in their publication from 2007, involves the use of 50 mg Clomiphene citrate from cycle day 3 onwards. This new protocol is designated as “minimal ovarian stimulation.” It is not designed for women with low ovarian reserve in particular. Clomiphene is administered in this method for a relatively long period of time, i.e., 10–12 days until the day before maturation is triggered by administration of a GnRH agonist. Oocytes are then retrieved 32–35 h later. By this method, Teramoto makes use of the antagonistic action of Clomiphene citrate to the estradiol receptor on the hypothalamus level, inhibiting both positive and negative feedback and resulting in the induction of the ovarian stimulation and suppression of ovulation.

The other main inspirations for a new treatment approach for patients with low ovarian reserve came from a group in New Jersey, USA (Check et al. 1990). They used ethinyl estradiol to bring high FSH levels down to normal levels. As we will explain later in this chapter, very high FSH and LH levels tend to downregulate the FSH receptors on the antral follicles and prevent follicle growth. If serum FSH levels can be reduced to physiological levels, using ethinyl estradiol, the receptors can recover and follicle growth and ovulation can be restored.

The new treatment approach that we have developed was mainly based on the natural cycle of the patients. The usual medication we used in these cases was Clomiphene citrate only to control ovulation, the way it was already described in the publication from Teramoto and Kato. In other patients with ovulation disorders, we successfully applied ethinyl estradiol, combined oral contraceptive pills, or Clomiphene citrate depending on the pathology. The following observations and description of protocols are not evidence-based and rely only on the empiric experience of our group. Nevertheless, using our approach, we managed to have pregnancies and live births in nearly menopausal women and other “hopeless” cases, which were originally referred to egg donation.

In 2011, we performed a proof of concept study with 10 patients at the Instituto Mexicano de Infertilidad in Guadalajara (Mexico), a private IVF clinic. Patients were informed about off-label use of the medication. The local Ethics Review Committee approved the study protocol, and a written informed consent was obtained from all patients. Ten patients with AMH <1.0 nmol/L and severe cycle disorders due to ovarian insufficiency were observed over an 8-month period. Patients were 22–42 years old (average 39.3). Blood samples were drawn to determine FSH, LH, E2, and transvaginal ultrasound scans were performed on different days of the cycle. Depending on the cycle pattern of each patient, we offered individualized treatment approaches based on natural cycle IVF using Clomiphene to control ovulation as described by Teramoto (Teramoto and Kato 2007), GnRH agonists to induce ovulation and either ethinyl estradiol or combined oral contraceptive pills to regulate the cycle. Embryos were vitrified in day 2 stages and transferred later in artificial cycles.

In this proof of concept study, we used the following protocol:

The patients started with combined oral contraceptive pills (Desogestrel 0.15 mg, ethinyl-estradiol 0.03 mg) for 10 days between days 16–26 of their menstrual cycle. After discontinuation of the contraceptive pills, we waited for the onset of withdrawal bleeding and performed a hormonal assay of FSH and Estradiol (E2) on the second day of bleeding. If we found FSH >25 IU/L, we started a treatment with ethinyl-estradiol 25 μg per day for 7 days. If FSH was <25 IU/L, there was no additional treatment. Then we started one pill of Clomiphene citrate (50 mg) from day 7 of withdrawal bleeding, which was continued daily till ovulation induction trigger. Another hormonal assay with LH and E2 was performed on day 10. If E2 was still under the value of 300 pg/mL on day 10, E2 and LH were repeated every morning until E2 was higher than 300 pg/mL. We instructed the patient to induce ovulation with a GnRH analog injection 0.1 mg. Egg retrieval was performed 36 h after ovulation induction. All embryos were vitrified 2 days after egg retrieval. In the following cycle, we treated the patient again with combined oral contraceptive pills for 15 days. These could be started anywhere in the menstrual cycle. We performed a GnRH analog injection of 3.75 mg together with pill number 10, then 5 more days of oral contraceptive pills. Then we started estradiol valerate 4 mg/d with the first day of withdrawal bleeding. We performed a vaginal ultrasound scan between days 12–14 of estrogen treatment. If the endometrial lining was found <6 mm, we increased the dosage of estradiol valerate to 6 mg/d for 5 more days; then we performed another vaginal ultrasound scan. If the endometrium was >6 mm, we instructed the patient to start vaginal progesterone 400 mg/d. Then we thawed and transferred 1 embryo 56 hours after initiation of vaginal progesterone.

We were able to perform oocyte retrievals and embryo transfers in all 10 patients. A total of 33 natural cycles were initiated. Premature ovulation occurred in 3 cycles (9.0 %), and no retrieval was attempted. Among the attempted 30 oocyte retrievals, 21 (70.0 %) were successful. Out of those 21 oocytes 11 (36.6 % per retrieval) were mature, and 10 (33.3 % per retrieval) were immature. Intracytoplasmic sperm injection (ICSI) resulted in 8 fertilizations (72.7 % per mature oocyte). Out of 8 transfers, 3 (37.5 %) resulted in a biochemical pregnancy. Two patients delivered (25.0 %); one patient had a miscarriage at 8 weeks of pregnancy.

As it was already mentioned in the introduction of this chapter, independently of the study, based on observation of hundreds of menstrual cycles with the help of vaginal ultrasounds and hormonal assays, in women with low ovarian reserve, we were able to classify four different stages of ovarian insufficiency:

We would like to call the first stage “compensated stage.” In this stage the endocrine compensation mechanisms are still working, and the women still have long and regular cycles of 26–33 days. Nevertheless, the anti-Mullerian hormone (AMH) or the antral follicle count (AFC) can be low, and in some cycles the early follicular phase FSH can be elevated. But as long as the menstrual cycle stays long and regular, the low ovarian reserve probably does not reduce the fertility. In most of the cases, there is another explanation for the infertility of the couple, if they are seeking for treatment. These patients are the typical low responder, if they are treated with a classical hyperstimulation protocol, such as the long agonist protocol or antagonist protocol.

The second stage of ovarian insufficiency we call “stage of desynchronized menstrual cycle.” Probably, due to a lower number of antral follicles in the ovaries, serum inhibin levels and other inhibiting hormones fall under a certain limit and as a consequence, basal FSH levels rise (Welt et al. 2005). Without the stabilizing effect of these hormones, the estrogen and progesterone producing corpus luteum is not strong enough to completely block FSH production in the second half of the cycle. Due to these higher basal FSH levels in the luteal phase, follicular development starts too early, already in the middle of the luteal phase. On the onset of menses, a dominant follicle has already been recruited, which, under normal circumstances, would only happen on day six of menstrual cycle. Because of the early follicular recruitment during the luteal phase of the previous cycle, ovulation will occur earlier, and the whole cycle becomes shorter. But a shorter follicular phase may have negative effects on endometrium maturation and implantation of the embryo. Often, in patients with short menstrual cycles less than 26 days, a present dominant follicle of 12–14 mm or more can be detected at vaginal ultrasound at the beginning of the cycle. These patients often have a normal early follicular FSH but high early follicular estradiol levels over 60 pg/mL. In order to resynchronize the cycle again, we treated the patient with combined oral contraceptive pills for at least 10 days during the luteal phase, for example, from day 16 to day 26 of menstrual cycle. The additional effect of the combined oral contraceptive pills together with the hormones produced by the corpus luteum suppresses FSH and avoids follicle formation during the luteal phase. After discontinuation of combined oral contraceptive pills, the withdrawal bleeding marks the beginning of the new, resynchronized cycle.

The third stage of ovarian insufficiency we called “stage of pathologic premature LH rise.” The progression of antral follicle loss and the lack of inhibiting hormones produced by the ovaries seem to destabilize the hypothalamus – pituitary system even further in this stage. In addition to higher baseline FSH levels, the system seems to be more sensitive to estrogen when it comes to the triggering of LH for ovulation. Under normal circumstances, the LH rise is triggered by estrogen on the hypothalamus level, once the follicle is mature. In this stage of ovarian insufficiency, LH rise is triggered too early, when estrogen levels are still low and the follicle is still small and immature. The eggs ovulated from these small and immature follicles are of a bad quality and may not lead to a pregnancy. The typical pattern in this stage is short menstrual cycles of less than 26 days, high early follicular phase FSH up to 25–30 IU/L, and normal or low early follicular estradiol levels. Normally there is no dominant follicle at the beginning of the cycle, but nevertheless, ovulation occurs too early, due to the premature LH rise, which leads to a shorter follicular phase. The small and immature follicles seem to produce a weak corpus luteum, as we measured low progesterone during the luteal phase. The resulting luteal phase insufficiency also leads to a shorter second half of the cycle. Our way to treat the pathologic premature LH rise is to use the antiestrogenic effect of Clomiphene citrate around ovulation. As a competitive inhibitor of estrogen receptors on the hypothalamus level, the system is tricked to “think” that the follicle is still not mature enough to trigger the LH surge. Depending on the dosage, Clomiphene citrate is able to delay LH surge or to block it completely (Teramoto and Kato 2007). In this case, it is necessary to trigger LH surge and ovulation artificially with GnRH analogs, once the follicle has reached maturity. It is important to understand that Clomiphene citrate can be used in two different ways: either for ovarian stimulation, if given before day 6 of menstrual cycle, but in this case, it may have a negative impact on the endometrium and implantation (Wallace et al. 2011). If Clomiphene citrate is given after day 6 of menstrual cycle, it has only an inhibiting effect on LH surge and ovulation but probably no negative impact on the endometrium. In patients in the “stage of pathologic premature LH rise,” the use of Clomiphene citrate given around ovulation can improve follicle maturation and oocyte quality. The standard protocol in this case was 50 mg Clomiphene citrate per day from day seven of menstrual cycle including the evening of ovulation induction by GnRH analogs. We did not use human chorionic gonadotropin (hCG) to trigger ovulation, if Clomiphene citrate was used after day 6 of the menstrual cycle. Clomiphene citrate has a relatively long half-life, and hCG somehow boosts its stimulating effect on the ovaries. If both drugs are used together, there is a higher risk for functional ovarian cysts than if GnRH analogs are used for triggering. Some people think that using GnRH analogs for triggering ovulation may cause a luteal phase defect and decrease implantation rates, but this is only the case after ovarian stimulation with more than four mature follicles, not in a natural cycle. If the ovulation has been from a mature follicle, due to the use of Clomiphene citrate, the resulting corpus luteum should be strong enough for a good luteal phase.

The fourth stage of ovarian insufficiency we called “stage of suppressed follicle growth.” In this stage, either very high FSH levels of over 30 IU/L or high LH levels seem to block follicle growth, probably due to the downregulation of the respective receptors on the follicle itself. These patients have long and irregular cycles. Most of the time the follicular growth is inhibited by the high serum FSH or LH levels, but from time to time, the ovaries seem to recover, FSH and LH levels fall to a certain limit, and a follicle may grow again. Once a progesterone-producing corpus luteum has developed, the ovary may enter into a dynamic phase with several ovulations during the luteal phase, as it was already described in the second stage of ovarian insufficiency. This is possible, because progesterone has a negative feedback on the hypothalamus, so FSH and LH are kept relatively low. But, this phenomenon of luteal phase ovulations may also lead to longer periods of amenorrhea. Every time a corpus luteum becomes weaker and is about to vanish, another one appears and progesterone production continues without withdrawal bleeding. To break the cycle of ovulations during luteal phase, we treated the patients with combined OCPs for at least 10 days, which can be started at any time during amenorrhea. The pre-condition here is that serum progesterone levels should be higher than 1.0 ng/mL. If no progesterone is present and FSH and LH levels are very high, there is no use of COCP treatment. In this case, a treatment with ethinyl estradiol (EE) alone can be tried. In most of the cases, especially when at least one antral follicle can be seen on vaginal ultrasound, a treatment with 25–50 μg EE over 7–10 days is sufficient to bring FSH and LH levels down to nearly physiological levels and allow the antral follicle to grow (Check et al. 1990). In order to prevent a pathologic premature LH rise, as described in the previous stage, it is useful to treat the patient additionally with 25–50 mg Clomiphene citrate per day, once the follicle reaches 12–14 mm and continue this treatment till ovulation induction with GnRH analogs.


Our experience shows that ovulation can successfully be controlled by the use of Clomiphene citrate, as described by Teramoto (Teramoto and Kato 2007) and does not necessarily require GnRH analogs for pituitary suppression. This knowledge opens a new space for development of alternative protocols respecting the patients’ own physiology with no need for heavy stimulation. Patients with ovarian insufficiency may benefit from this approach, which can be offered before referring them to egg donation. In order to achieve high pregnancy rates in patients with low ovarian reserves, it is important to evaluate the patient’s cycle pattern or to diagnose a possible cycle disorder. Then, the treatment can be individually fixed depending on the age of the patient, her ovarian reserve, and her cycle pattern. Using this empiric approach, we defined four different stages of ovarian insufficiency with the related cycle disorders. We believe that this classification may become a helpful tool for clinicians to evaluate the ovarian reserve of their patients clinically, hoping that this new approach will open a new field of investigation.


Check JH, Nowroozi K, Chase JS, Nazari A, Shapse D, Vaze M. Ovulation induction and pregnancies in 100 consecutive women with hypergonadotropic amenorrhea. Fertil Steril. 1990;53(5):811–6.PubMed

Teramoto S, Kato O. Minimal ovarian stimulation with clomiphene citrate: a large-scale retrospective study. Reprod Biomed Online. 2007;15(2):134–48.CrossRefPubMed

Wallace KL, Johnson V, Sopelak V, Hines R. Clomiphene citrate versus letrozole: molecular analysis of the endometrium in women with polycystic ovary syndrome. Fertil Steril. 2011;96(4):1051–6.CrossRefPubMed

Welt CK, Hall JE, Adams JM, Taylor AE. Relationship of estradiol and inhibin to the follicle-stimulating hormone variability in hypergonadotropic hypogonadism or premature ovarian failure. J Clin Endocrinol Metab. 2005;90(2):826–30. Epub 2004 Nov 23.CrossRefPubMed