The importance of luteal phase support (LPS) is widely recognized in assisted reproductive technology (ART) and an adequate luteal phase is mandatory for embryo implantation. Under physiological conditions, the luteal phase is mainly sustained by a luteinizing hormone (LH) and a follicle-stimulating hormone (FSH) surge during ovulation, which induces the “luteinization” of the granulosa cells, thereby promoting the formation of the corpus luteum (1). In particular, LH is important, not only during the late phase of folliculogenesis (2–6) and for the support of the corpus luteum (7), but also in the early stage of embryo implantation (8–10). During LP, the corpus luteum sustains the endometrium, preparing it for implantation that usually occurs 6 days after fertilization (2). In medically assisted reproduction, a luteal phase defect is observed. Indeed, the supraphysiological production of sexual steroids during ovarian stimulation (OS) significantly suppresses LH levels, thereby causing impairment of LPS (11). Clinically, the latest Cochrane meta-analysis which included 94 randomized controlled trials (RCTs) (26,198 infertile women undergoing ART) confirmed that LPS is associated with significantly higher live births and ongoing pregnancy compared with placebo or no treatment (12). LPS is carried out mainly with progesterone (Pg) and/or other hormones that promote its production, such as human chorionic gonadotropin (hCG) and the gonadotropin-releasing hormone agonist (GnRHa). Pg supplementation is by far the most common strategy adopted worldwide (11, 13). Pg can be administrated vaginally, rectally, intramuscularly, and orally (12). Oral Pg seems to be more effective, using synthetic Pg rather than natural micronized Pg (11, 14, 15). Notably, among oral products, dydrogesterone seems to offer better results in terms of live births and pregnancy rates in fresh cycles compared with vaginal micronized Pg (16); however, vaginal administration of Pg is still the most preferred route among clinicians (17, 18). Conversely, <6% of practitioners opted for an intramuscular or oral route (13, 17).

Recently, a water-soluble formulation of Pg for subcutaneous administration has been introduced (19, 20). A preliminary study demonstrated that this route can induce endometrial decidualization and present bioavailability similar to oil-based products (19–21). In the present review, we summarize the clinical studies that have adopted subcutaneous Pg for LPS in an ART setting.

A total of 100 items were identified through database research. Fourteen full-text papers were assessed for eligibility. Five of them did not report data concerning clinical outcomes or compliance of patients and were excluded (17, 19–21, 24). Five RCTs, two retrospectives, and two cohort studies were selected (Figure 1). In two RCTs and one non-RCT study, data were extracted from conference abstracts (25–27). All characteristics of the studies included are illustrated in Table 1 (fresh cycle) and Table 2 (frozen cycle). The risk of bias in non-RCTs is stated in Table 3, while the risk of bias in RCTs is illustrated in Figure 2.

Subcutaneous Pg represents one of the most recent innovations proposed for LPS in IVF. This formulation was obtained by combining Pg with hydroxypropyl-b-cyclodextrin, which in turn increases the solubility of Pg (21). A preliminary RCT demonstrated that 25 mg/day or 50 mg/day of subcutaneous Pg can induce decidualization in 24 volunteers, whose ovarian function was suppressed with GnRH agonist and subsequently estrogenized using transdermal route (20). Sator et al. demonstrated that subcutaneous formulation was bioequivalent to intramuscular formulation in terms of the extent of exposure (19). In detail, the subcutaneous formulation is promptly absorbed and could achieve progesterone peak serum levels at an earlier time than the intramuscular route (19). In 24 donors undergoing OS and triggered with GnRHa, randomized to subcutaneous or intramuscular Pg, endometrial histology and endometrial receptivity array (ERA) transcriptomic expression showed non-significant differences in the secretory dating (24). Subcutaneous administration can be self-administered and is associated with lower injection site reactions, such as pain and irritation, compared with the intramuscular route (28). In addition, Mele et al. reported a significantly reduced level of stress-related hormones, such as Prolactin and Cortisol (30).

As an alternative to the vaginal route, subcutaneous Pg could be proposed to women who are reluctant to use vaginal medication for cultural or religious reasons. In addition, a vaginal route could be associated with vaginal irritation and discomfort (29), and the insertion of a vaginal device might be potentially associated with an increased risk of genital tract infection. Despite this belief, clinical data concerning genital (herpes virus, vaginal bacterial infection, vulvovaginal mycotic infection) and urinary tract infections seem to be similar when comparing vaginal and subcutaneous products (29). The potentially better compliance of subcutaneous vs. vaginal administration was investigated in several trials with mixed results. The most robust RCT trials (28, 29) conducted during fresh cycles did not demonstrate any relevant differences in terms of tolerability or adverse side-effects when comparing vaginal and subcutaneous administration. Conversely, two unpublished RCTs conducted in FET cycles reported higher satisfaction using the subcutaneous route. Similarly, Venturella et al. (33) observed better compliance with the subcutaneous route in a prospective study involving women with a previous experience of vaginal Pg; however, in FET cycles higher miscarriage and lower clinical pregnancy rates of the subcutaneous vs. the vaginal approach were reported in one RCT, involving oocytes donor recipients (25). This effect may be linked to the fact that the appropriate dosage in these populations is mainly unknown. So far, only one study compared intramuscular and subcutaneous Pg with no differences in terms of pregnancy rate between groups, but significantly lower prolactin and cortisol levels in the women who had received the subcutaneous route (30).

However, more evidence is necessary to better understand the application of subcutaneous Pg in ART. To the best of our knowledge, no RCT compared a subcutaneous Pg vs. oral dydrogesterone which could be a valuable approach for LPS in women undergoing ART (16, 39). In addition, so far no studies have been developed in women with a low prognosis to ART (40–46) or in women at risk of a hyper-response, such as those affected by PCOS (47–49). As yet, no cost-effective analysis has been conducted.

To conclude, a fair amount of evidence exists to support the hypothesis that in women with a good prognosis undergoing fresh IVF cycles, subcutaneous Pg is not inferior to vaginal products. In the FET cycle, data concerning efficacy are mixed with one RCT conducted in oocyte donor recipients which observed reduced clinical pregnancy rates and increased miscarriage rates in women undergoing a subcutaneous route. This data should be interpreted with caution considering that there is still too much uncertainty about the dosages to be used in women undergoing the FET cycle.

Data concerning the acceptance of a subcutaneous vs. vaginal route are encouraging despite the different scales and questionnaires used to test the acceptance of women among trials. Therefore, subcutaneous Pg could be proposed to women who are against vaginal administration and, in contrast with the intramuscular route, may be associated with better tolerability and reduced injection site reaction.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

AC idealized the article and wrote the first draft. All authors participated in the literature research and paper editing. All authors listed have made an intellectual contribution to the work and approved the final version.

AC and CA decleare fees from Merck Serono Italia outside the submitted work. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.