Serum progesterone levels have been used to make prognoses about the continued development of pregnancy and even to diagnose pregnancy loss. The lowest progesterone level to be associated with a viable pregnancy was 5.1 ng/ml in Stovall et al.’s [12] series. A single progesterone level ≥25 ng/ml was associated with a 97 % likelihood of viable pregnancy [12]. Al-Sebai et al. [5] summarised 358 threatened abortions <18 weeks, a single progesterone level ≤ 45 nmol/l (14 ng/ml) was reported to differentiate between aborting and ongoing pregnancies (sensitivity 87.6 %, specificity 87.5 %). Serum progesterone levels of less than ≤12 ng/ml. were associated with an increased risk of miscarriage in Arck et al.’s [13] series.

The mechanisms by which luteal phase deficiency occur are unclear but could be associated with decreased FSH levels in the follicular phase (it has been suggested that luteal phase deficiency originates as a preovulatory event). Alternatively there may be an abnormal pattern of LH secretion. Luteal phase deficiency may be secondary to abnormal follicle formation, associated with poor oocyte quality, or a decreased response to progesterone by the endometrium [14]. However, there are pitfalls to using serum progesterone levels as a predictive marker of miscarriage, or for determining the need for progesterone supplementation. Progesterone secretion is pulsatile. Blood may be drawn at a pulse peak or nadir. These may vary tenfold [15]. Hormone levels may be normal, but histology abnormal due to deficiency of progesterone receptors. As with other presumptive causes of abortion, low hormone levels may be a result of abortion rather than its cause. In the blighted ovum or after embryonic death, there is no villous circulation. Trophoblastic failure after villous circulatory failure results in low hCG levels. If hCG does not stimulate the corpus luteum, progesterone levels will fall explaining the mechanism of expulsion, but not necessarily that of embryonic death, or the cause of abortion. Therefore in threatened miscarriage, diagnosis and treatment should be empirical rather than based on progesterone levels.

The results of progestogen treatment may be confounded as threatened miscarriage, may be due to separation of the placenta in a normal embryo, or a defence mechanism to prevent the continued development of an abnormal embryo, including abnormalities which are incompatible with life. The most important confounding factors are embryonic structural malformations, or chromosomal aberrations. In missed abortions 200 of 233 embryos have been reported to be structurally abnormal on embryoscopy [16]. These defects included: anencephaly, encephalocele, spina bifida, syndactyly, pseudo-syndactyly, polydactyly, cleft hand and cleft lip. Without embryoscopy these embryos would not have been diagnosed, and the patient might have been treated empirically with progesterone. If included in a trial, the results would be skewed in favour of a negative effect. However, embryoscopy is advanced technique, which is not usually available. Additionally, 70 % of miscarriages are blighted ova. Therefore, it is impossible to tell if a rudimentary embryo may have been structurally abnormal. At present, ultrasound is not sufficiently sensitive to diagnose these very early anomalies.

Up to 60 % of sporadic miscarriages [17–19] may be caused by chromosomal aberrations in the embryo. The most common aberrations which are incompatible with life include 16 trisomy and polyploidy. Additionally, with the introduction of comparative genomic hybridization (CGH) a further 15 % of so called normal embryos on conventional karyotyping, have been diagnosed with genetic aberrations [20, 21].

Progesterone cannot correct chromosomal aberrations. Unfortunately, the abortus is not usually karyotyped, even less are subjected to CGH microarray analysis. Hence, it is unclear whether miscarriage after supplemental progesterone is due to failure of treatment or confounding of the results by fetal chromosomal aberrations.

Before a trial of progesterone can be said to be conclusive, other predictive factors should be taken into account.

Progestogens have been used since the 1950s to prevent threatened miscarriage terminating as miscarriage. It has been difficult to prove whether progestogens are effective, due to the generally good natural history, and the effect of confounding factors. Recently, two systematic reviews have been performed. Wahibi et al. [22] carried out an analysis of two trials of oral dydrogesterone compared to placebo [23, 24] and two trials of vaginal progesterone [25, 26] (See Fig. 4.1). The overall figures showed a statistically significant benefit, OR = 0.53 (CI 0.35–0.79) in favour of progestogen supplementation. However, the analysis consisted of only four studies. The authors concluded, “The analysis was limited by the small number and the poor methodological quality of eligible studies, and the small number of the participants, which limit the power of the meta-analysis and hence of the conclusion”. It is interesting to note that in the women who were treated with vaginal progesterone the treatment was not statistically effective in reducing miscarriage when compared to placebo (RR = 0.47 95 % CI, 0.17–1.30) whereas oral progestogen (dydrogesterone) was effective (RR = 0.54 CI 0.35–0.84). In the light of Wahibi et al.’s [22] metaanalysis, two new guidelines were released regarding threatened miscarriage. In the U.K. the National Institute for Health and Care Excellence (NICE) [27] produced a guideline in 2012 which stated, that “data from a meta-analysis of several small studies suggest that progestogens are better than placebo”. The Royal Australian and New Zealand College of Obstetricians and/gynaecologists guideline [28] stated that, “Miscarriage was significantly less likely to occur on progestins than placebo or no treatment (risk ratio 0.53; 95 % CI 0.35–0.79)”. Both guidelines stated, that further research was necessary to confirm the results on a larger number of patients.

The author tried to enlarge the number of patients available for analysis by relaxing the strict methodological rules required by the Cochrane database for inclusion in metaanalyses. A literature search was performed for in September 2010 for all papers available at that time in EMBASE and Ovid MEDLINE^® that fulfilled the following criteria: original contributions with product name ‘Progestogen, Progestin, Progesterone, Micronised progesterone Duphaston’ or ‘dydrogesterone’, reports limited to clinical human data excluding reviews, case reports and editorials in any language. All articles considered were investigator initiated trials and published in the scientific literature.

No additional relevant trials were found for micronized or intramuscular progesterone, to those assessed in Wahibi et al.’s [22] metaanalysis. However, three additional trials of oral dydrogesterone were considered eligible for inclusion in a systematic review [29–31]. Carp’s [32] metaanalysis included five randomized studies. 660 patients were eligible for analyses of pregnancy maintenance. The results (Fig. 4.2) showed that the effect of dydrogesterone on the risk of miscarriage in women with threatened miscarriage appeared to be substantial. There was a statistically significant reduction in the odds ratio for miscarriage after dydrogesterone compared to standard care of 0.47 (CI 0.31–0.7). The 24 % miscarriage rate in control women (78/325) was reduced to 13 % (44/335) after dydrogesterone administration (11 % absolute reduction in the miscarriage rate). There has since been a further trial of vaginal micronised progesterone. Yassaee et al [33] have reported on the results of 60 patients treated with threatened miscarriage who were treated with vaginal micronised progesterone. Although there was a trend towards an improved prognosis, the results were not significantly better than in the control group.

Safety and side effects are always a major worry when drugs are used in pregnancy. The major concerns about safety have centred on the effect of progestogens on the androgen receptor. The original progestogens were derivatives of 19 nor testosterone. These compounds led to varying degrees of masculinization of female fetuses. The masculinization included clitoral enlargement, and varying degrees of labial fusion. However, these side effects are not seen with other progestins. On the contrary, progesterone itself has anti-androgenic effects and has been reported to lead to hypospadias. In a case–control study, a relationship was reported between the use of progesterone and isolated hypospadias in two uncontrolled observational studies [42, 43]. Carmichael et al. [44], studied the risk of hypospadias and periconceptional progestin intake. Progestin-related hypospadia was reported by 42 (8.4 %) case mothers and 31 (2.4 %) control mothers, for intakes from 4 weeks before conception to 14 weeks of pregnancy. The crude odds ratio for progestin use at any time was 3.7 (95 % CI 2.3–6.0). Additionally, Rock et al. [45] reported one case of undescended testis and one case of meningomyelocele in 93 women treated with progesterone in the first trimester. Check et al. [46] found two cardiovascular malformations, omphalocele, hydrocephalus and club foot (with cleft palate) in 382 women exposed to either progesterone or 17-α hydroxy-progesterone. These studies had no control group.