You are here

No need for luteal phase support in IVF cycles after mild stimulation: proof-of-concept study

Reproductive BioMedicine Online

Editorial Comment

This small-scale study in a private IVF clinic investigated whether luteal-phase support can be avoided if patients receive mild IVF treatment. Luteal-phase support is usually required with conventional IVF methods as the corpus luteum is thought to be unable to produce sufficient progesterone. The authors suggest that mild/minimal stimulation IVF may avoid defects in corpus luteum function and therefore bypass the requirement for luteal-phase support. 


This is a pilot study performed in a private IVF unit. The objective of the study was to investigate whether luteal support is required in IVF cycles after mild stimulation with clomiphene citrate and low FSH doses. The study included 15 patients with good prognosis (defined as ≤38 years old with normal ovarian reserve and normovulatory cycles, body mass index <29 kg/m2, no previous cycles, no severe endometriosis, no history of recurrent miscarriage, no endocrine/autoimmune diseases and no surgical semen extraction from the partner) undergoing IVF with mild stimulation. Patients were monitored during the luteal phase by serum progesterone and LH. The luteal support was started only when necessary. No patient needed luteal phase support because the resultant steroid environment was different from that associated with conventional stimulation techniques. The live birth rate was 40% (6/15) and the implantation rate 30% (6/20). There are several benefits to mild stimulation, including low cost, less patient distress and improved endometrial receptivity. Our study supports the concept that mild stimulation may have an additional benefit during the luteal phase, by obviating the need for luteal phase support.

Keywords: IVF, luteal phase, mild stimulation.


Evidence in favour of mild stimulation for IVF has been accumulating in recent years (Verberg et al., 2009). The benefits include a reduction in cost, treatment complexity and patient discomfort, enabling patients to continue treatment to increase their chances of parenthood. More recently the concept that use of mild ovarian stimulation may increase oocyte quality and improve endometrial receptivity has gained support (Arce et al, 2014 and Fatemi, Popovic-Todorovic, 2013).

When minimal stimulation for IVF was initially proposed, the concept that luteal phase supplementation may not be necessary was postulated by Macklon and Fauser (2000) but no data are available in the literature on this topic.

It is known that the physiology of the luteal phase is altered in IVF cycles, regardless of the type of stimulation protocol used (Fatemi et al., 2007). Thus, luteal phase support is mandatory in IVF cycles, as it significantly improves implantation, pregnancy and delivery rates (Pritts and Atwood, 2002). Several reasons have been proposed for the deficient luteal phase. The current theory is that the supraphysiological concentrations of sex steroid hormones observed in both the follicular and luteal phases cause a pituitary downregulation of gonadotrophin secretion (Bourgain and Devroey, 2003). The conventional human chorionic gonadotrophin (HCG) bolus to trigger ovulation supports stimulation of the corpora lutea for at least 7 days, but later, the low endogenous LH levels are insufficient to sustain adequate corpus luteal activity. Without any exogenous support, the length of the luteal phase may be <12 days (Beckers et al., 2003). Moreover, the luteal phase is even more defective after triggering ovulation with gonadotrophin-releasing hormone (GnRH) agonist because of the short duration of the induced LH peak. The duration of the non-supplemented luteal phase may be as short as 4 days (Beckers et al., 2003).

In mild stimulations, the endocrine luteal environment after the recruitment of fewer follicles may differ from conventional stimulations (Verberg et al., 2009), avoiding the pituitary suppression and then obviating the need for luteal support. However, this concept has never been investigated.

Despite widespread interest, mild ovarian stimulation regimens still face resistance in clinical practice, because of fears of negatively influencing the results (Fauser et al, 2010 and Gleicher et al, 2012). By contrast, in our private IVF clinic, the mild stimulation approach has been used since 2009 as the first option for patients with good prognosis. The ovarian stimulation chosen for this approach is a fixed protocol of clomiphene citrate (CC) and low doses of FSH. The results and the benefits obtained in 193 couples were published and discussed in a recent paper (Ferraretti et al., 2015). In this previously published study, all patients received conventional luteal support with progesterone, starting 3 days after egg retrieval. In the present study, the endocrine profiles of the luteal phase were monitored in order to investigate whether luteal phase support can be avoided in mild stimulations. To the best of our knowledge, no study has been published thus far that has investigated this concept.

Materials and methods

The present study was conducted from 1 November 2014 to 31 March 2015. The study included 15 consecutive patients who fulfilled the criteria for undergoing mild ovarian stimulation (age ≤38 years, normal ovarian reserve and normovulatory cycles, body mass index [BMI] < 29, no previous IVF cycles and no partner with surgical semen extraction). The mean age of the female partner was 33.4 ± 2.8 years, and the indications for treatment were tubal factors in two cases, male factors in six cases, and both male and female factors in seven cases. The mean duration of infertility was 3.1 years. The study was approved by the ethical review board of the S.I.S.Me.R. clinic on 20 October 2014.

As previously described (Ferraretti et al., 2015), the mild stimulation procedure consisted of a fixed protocol of CC (100 mg/day on cycle days 3−7) and 150 IU of recombinant FSH (rFSH) on cycle days 5, 7 and 9. The first ultrasound examination and oestradiol assay were performed on cycle day 9. GnRH antagonist treatment was initiated on cycle day 9, with one or two more monitoring visits (ultrasound and oestradiol) planned prior to HCG administration, depending on the ovarian response. An additional 150 IU rFSH was administered on cycle day 11, if needed. Egg retrieval was performed 34 h after HCG administration (5000 IU), and oocytes were inseminated by IVF or intracytoplasmic sperm injection. In accordance with our current policy to reduce the incidence of twins, the number of embryos transferred was decided upon based on the ‘in-vitro’ performance of each couple. In cases of normal development of more than two embryos, an elective single embryo transfer (eSET) was performed on day +4 or +5 post egg retrieval and the surplus embryos were cryopreserved by vitrification. Two embryos were transferred on day +3 or +4 only when they were the only embryos available.

In the conventional approach, the luteal phase support with exogenous progesterone is always started 3 days after egg retrieval. In this study, patients were monitored during the luteal phase and exogenous progesterone support would have been started only in case of altered endocrine profiles.

The hormonal investigation included assays of serum progesterone and LH on the day of HCG administration, on the day of egg retrieval, on the day of embryo transfer and on day +9 from egg retrieval.

The occurrence of a pregnancy was investigated by measuring β-HCG and progesterone 14 days after egg retrieval. Only clinical pregnancies (gestational sacs with fetal heartbeat) were considered in the results. Implantation rate is defined as the number of gestational sacs on the total embryos transferred.


All patients underwent egg retrieval and embryo transfer. As presented in Table 1, the mean amount of FSH administered was 450 ± 75 IU. On the day of HCG administration, the mean levels of oestradiol were 980 ± 250 pg/ml. The mean number of oocytes recovered was 5.4 ± 2.9, and the mean number of embryos transferred was 1.2 ± 0.3. In ten cycles, one eSET was performed and 21 surplus embryos were cryopreserved. The remaining five patients received two embryos and had no embryos available for cryopreservation. The transfers were performed on day 3 in four cases, on day 4 in six cases, and on day 5 in five cases.

Table 1

Clinical data and outcomes.


Patients (n) 15
Age (mean ± SD) 33.5±2.6
FSH IU (mean ± SD) 450±75
Oestradiol at HCG (pg/ml) (mean ± SD) 980±250
Total oocytes collected n (mean ± SD) 82(5.4±2.9)
Inseminated eggs (n) 73
2 PN n (fertilization rate %) 62(85)
Embryos on day 2 n (cleavage rate %) 61(98)
Grade 1 embryos on the total number of embryos on days 3–5 (%) 36/55(65)
Transferred embryos n (mean ± SD) 20(1.2±0.3)
Cryopreserved embryos (n) 21(in10cycles)
Clinical pregnancies (n) 6
Implantation rate % (n) 30%(6/20)
Miscarriages (n) 0
Delivery rate % (n) 40(6/15)

Data are presented as total number, mean ± standard deviation (SD or number [%]).

HCG = human chorionic gonadotrophin.

The progesterone profile is presented in the upper panel of Figure 1. The mean serum concentration on the day of HCG administration was 0.57 ng/ml (range 0.2−1.0 ng/ml). An initial increase was observed on the day of egg retrieval (mean 3.8 ± 1.7 ng/ml), and all patients had values of >35 ng/ml on the day of transfer (mean 49.6 ± 12.5 ng/ml). On day 9 the mean levels showed a decline (25.8 ± 6.5 ng/ml), but the values were still over the physiological range in all patients.

Figure 1

Figure 1

Serum concentrations of progesterone (ng/ml) (upper panel) and LH (mIU/ml) (lower panel) from the day of HCG administration to day 9. Day 0: day of egg retrieval. Values are mean ± SD. HCG = human chorionic gonadotrophin.


The LH profile is presented in the lower panel of Figure 1. Mean serum concentrations were 4.2 mIU/ml at the time of HCG administration and 6.9 mIU/ml at egg retrieval. Between days +3 and +5, LH serum levels were maintained at 0.2−1.8 mIU/ml, but an increase of up to >3.5 mIU/ml was observed on day +9 (mean 4.3 ± 0.6 mIU/ml) as a sign of a normal pituitary function able to maintain corpus luteal function.

Based on their endocrine profiles, no patients received exogenous progesterone in the luteal phase.

A total of six clinical pregnancies out of 15 embryo transfers were obtained (Table 1). No miscarriages were observed and all pregnancies resulted in a live birth. The implantation rate was 30% (6/20) and the live birth rate 40% (6/15). No twins occurred.

No differences were observed in the LH and progesterone profiles of the early and middle luteal phases between pregnant and non-pregnant patients. On the day of the pregnancy test (day 14), the mean progesterone serum levels were 43.2 ng/ml in conception cycles and 3.4 ng/ml in non-conception cycles. The luteal phase length in non-conception cycles was normal (13.7 days, range 12−17 days) despite the absence of luteal support. To date, four cryopreserved embryo transfers have been performed in the non-pregnant patients resulting in one additional pregnancy.


The aim of the present study was to evaluate the endocrine profiles of the luteal phase after minimal stimulation in order to investigate whether luteal support may be obviated. The number of patients studied was small but the concept had not previously been investigated. No patients required luteal phase support. The live birth rate (40%) and implantation rate (30%) were comparable to data obtained in a previous study in which 299 fresh cycles were stimulated with the same mild stimulation protocol and supported by progesterone in the luteal phase (implantation rate 23.5% and ongoing pregnancy rate 35%) (Ferraretti et al., 2015). These preliminary data support the concept that mild/minimal stimulation (with CC plus a low dosage of FSH) may avoid defects in the corpus lutea function and, consequently, the need for exogenous luteal support.

According to current theories (Bourgain, Devroey, 2003 and Fauser, Devroey, 2003), the luteal phase dysfunction after conventional stimulation is due to the negative feedback to the hypothalamic/pituitary axis produced by high levels of steroids in both the follicular and luteal phases. Milder regimens, able to maintain lower levels of steroids, may therefore reduce the side effects associated with high steroid production. Thus, the endocrine environment of the luteal phase may differ with respect to the ability of the pituitary to sustain the activity of the corpus luteum for its normal lifespan. Smitz et al. (1988) have shown that the endocrine environment after human menopausal gonadotrophin (HMG) cycles is different from that resulting from CC-HMG stimulation, with very low LH levels in gonadotrophin cycles and normal levels of LH in CC cycles. An alternative hypothesis relates to the use of CC. CC is a selective oestrogen receptor modulator with a half-life of 2 weeks (Young et al., 1999). In the luteal phase, some endogenous LH may be released in response to the long-lasting oestrogen receptor antagonism from residual CC activity.

The endocrine profiles of the luteal phases observed in the studied patients may provide support for the previous hypotheses. Similar to conventional stimulations (and in contrast to natural cycles), the early luteal phases were characterized by very high concentrations of progesterone (as a consequence of the massive stimulation of corpora lutea by HCG) and low levels of LH. It is not possible to evaluate whether the low levels of LH present during the early luteal phase may still have a biological role, but a potential pituitary deficit is overridden for several days by the actions of exogenous HCG on the corpus luteum. No luteal phase support was therefore necessary during this period. Nevertheless, in the conventional approach, progesterone supplementation is usually initiated before this time (Connell et al., 2015) to overcome the LH deficit and to maintain unaltered the progesterone levels in the middle/late luteal phase. Without any luteal support, the levels of progesterone in the studied patients started to decrease in the middle luteal phase (Figure 1). However, the presence of progesterone levels still in the normal range, and the concomitant increase in the serum LH levels observed on day 9, were the events considered key to obviating the need for luteal phase support. In contrast to conventional protocols, the endogenous LH was actually sufficient to maintain corpus luteal function until the time at which sufficient endogenous HCG was being produced by the implanted embryo.

In conclusion, these preliminary data support the hypothesis that minimal stimulations may produce less luteal dysfunction than conventional stimulations. Larger, prospective controlled trials are needed to investigate the concept before claiming that luteal support may not be mandatory in these cases. Mild stimulation for IVF has several benefits, including reduced global costs and patient distress, and may improve egg quality and endometrial receptivity compared with conventional stimulations (Baart et al, 2007, Bourgain, Devroey, 2003, Fauser et al, 2010, and Verberg et al, 2009). If the hypothesis tested in this study is confirmed, an additional step will be reached in considering the benefits of mild stimulations. This will help to lower the cost and stress associated with IVF, which is important from a global perspective (Fauser and Serour, 2013).


  • Arce et al, 2014 J.C. Arce, A.N. Andersen, M. Fernandez-Sanchez, H. Visnova, E. Bosch, J.A. Garcia Velasco, P. Barri, P. De Sutter, B.M. Klein, B.C. Fauser. Ovarian response to recombinant human follicle-stimulating hormone: a randomized, antimullerian hormone-stratified, dose-response trial in women undergoing in vitro fertilization/intracytoplasmic sperm injection. Fertil. Steril. 2014;102:16331-16640
  • Baart et al, 2007 E.B. Baart, E. Martini, M.J. Eijkemans, D. Van Opstal, N.G. Beckers, A. Verhoeff. Milder ovarian stimulation for in-vitro fertilization reduces aneuploidy in the human preimplantation embryos. Hum. Reprod. 2007;22:9809-9888
  • Beckers et al, 2003 N.G. Beckers, N.S. Macklon, M.J. Eijkemans, M. Ludwig, R.E. Felberbaum, K. Diedrich, S. Bustion, E. Loumaye, B.C. Fauser. Nonsupplemented luteal phase characteristics after the administration of recombinant human chorionic gonadotropin, recombinant luteinizing hormone, or gonadotropin-releasing hormone (GnRH) agonist to induce final oocyte maturation in in vitro fertilization patients after ovarian stimulation with recombinant follicle-stimulating hormone and GnRH antagonist cotreatment. J. Clin. Endocrinol. Metab. 2003;88:4186-4192 Crossref
  • Bourgain, Devroey, 2003 C. Bourgain, P. Devroey. The endometrium in stimulated cycles for IVF. Hum. Reprod. Update. 2003;9:515-522 Crossref
  • Connell et al, 2015 M.T. Connell, J.F. Szatkowski, N. Teery, A.H. DeCherney, A.M. Propst, M.J. Hill. Timing luteal support in assisted reproductive technology: a systematic review. Fertil. Steril. 2015;103:939-946
  • Fatemi, Popovic-Todorovic, 2013 H.M. Fatemi, B. Popovic-Todorovic. Implantation in assisted reproduction: a look at endometrial receptivity. Reprod. Biomed. Online. 2013;27:5305-5338
  • Fatemi et al, 2007 H.M. Fatemi, B. Popovic-Todorovic, E. Papanikolau, P. Donoso, P. Devroey. An update of luteal phase support in stimulated IVF cycles. Hum. Reprod. Update. 2007;13:5815-5890
  • Fauser, Devroey, 2003 B.C. Fauser, P. Devroey. Reproductive biology and IVF: ovarian stimulation and luteal phase consequences. Trends Endocrinol. Metab. 2003;14:236-242 Crossref
  • Fauser, Serour, 2013 B.C. Fauser, G.I. Serour. Introduction: optimal in vitro fertilization in 2020: the global perspective. Fertil. Steril. 2013;100:297-298 Crossref
  • Fauser et al, 2010 B.C. Fauser, G. Nargund, A.N. Andersen, R. Norman, B. Tarlatzis, J. Boivin, W. Ledger. Mild ovarian stimulation for IVF: 10 years later. Hum. Reprod. 2010;25:2678-2684 Crossref
  • Ferraretti et al, 2015 A.P. Ferraretti, L. Gianaroli, M.C. Magli, P. Devroey. Mild ovarian stimulation with clomiphene citrate launch is a realistic option for in vitro fertilization. Fertil. Steril. 2015;104:333-338 Crossref
  • Gleicher et al, 2012 N. Gleicher, A. Weghofer, D.H. Barad. A case-control pilot study of low-intensity IVF in good-prognosis patients. Reprod. Biomed. Online. 2012;24:396-402 Crossref
  • Macklon, Fauser, 2000 N.S. Macklon, B.C. Fauser. Impact of ovarian hyperstimulation on the luteal phase. J. Reprod. Fertil. Suppl. 2000;55:101-108
  • Pritts, Atwood, 2002 E.A. Pritts, A.K. Atwood. Luteal phase support in infertility treatment: a meta-analysis of the randomized trials. Hum. Reprod. 2002;17:2287-2299 Crossref
  • Smitz et al, 1988 J. Smitz, P. Devroey, M. Camus, J. Deschacht, I. Kahn, C. Staessen, L. Van Waesberghe, A. Wisanto, A.C. Van Steirteghem. The luteal phase and early pregnancy after cmbined GnRH-agonist/HMG treatment for superovulation in IVF or GIFT. Hum. Reprod. 1988;3:585-590 Crossref
  • Verberg et al, 2009 M.F.G. Verberg, N.S. Macklon, G. Nargund, R. Frydman, P. Devroey, F.J. Broekmans, B.C. Fauser. Mild ovarian stimulation for IVF. Hum. Reprod. Update. 2009;15:921-929
  • Young et al, 1999 S.L. Young, M.S. Opsahl, M.A. Fritz. Serum concentrations of enclomiphene and zuclomiphene across consecutive cycles of clomiphene citrate therapy in anovulatory infertile women. Fertil. Steril. 1999;71:639-644 Crossref
Unlabelled box

Anna Pia Ferraretti has been a specialist in reproductive medicine since the late 1970s and has a degree in obstetrics and gynaecology plus a PhD in endocrinology. She is currently Coordinator of Gynaecological Endocrinology in the IVF Unit, S.I.S.Me.R., Bologna, Italy and a memberof the ESHRE Special Interest Group on Global and Socio-cultural Aspects of Infertility.

Key Message

This study shows that IVF can be performed without luteal-phase support. Stimulation was done using clomifene citrate, gonadotrophins and GnRH antagonist treatment. The LH and progesterone profiles of the luteal phase and its length were normal, as were the pregnancy and implantation rates. These results show that an adequate, low-cost, simplified, efficient and patient-friendly approach is possible.

Unlabelled box


Reproductive Medicine Unit, S.I.S.Me.R., Via Mazzini 12, 40138 Bologna, Italy

* Corresponding author.