Author's address (for correspondence): A Günzel-Apel, Unit for Reproductive Medicine of Clinics University of Veterinary Medicine, Hannover, Bünteweg 15, 30559 Hannover, Germany. E-mail: firstname.lastname@example.org
Progesterone profiles of individual bitches may vary considerably both between and within individuals during pregnancy and non-pregnancy. Suspected luteal deficiency is commonly purported but is difficult to evaluate in clinical cases when progesterone is supplemented because this masks the underlying hormone changes. Therefore, in this study, suspected cases of luteal deficiency (six pregnancies from five bitches) were supplemented with oral medroxyprogesterone acetate (MPA), allowing measurement of endogenous progesterone using conventional assay. MPA (0.1 mg/kg) treatment commenced between days 30 and 36 after estimated ovulation and was continued for 18–28 days. Endogenous progesterone was measured throughout treatment, and blood was additionally analysed for prolactin (PRL) and relaxin (RLN) as well as MPA. The latter revealed delayed MPA clearance in two bitches, in which Caesarean operation had to be performed because of a low foetal heart rate. In two cases with confirmed basal concentrations of both P4 and MPA at term, spontaneous parturition occurred. Low endogenous progesterone during pregnancy was not apparent in three bitches that had previously had a short inter-oestrous interval of which two had previously had confirmed short luteal phase. However, in the remaining two cases, there had been previous pregnancy failure, but in only one of these, a premature decrease in endogenous progesterone to <2 ng/ml was detected. The latter had also low concentrations of PRL and RLN. The results of this preliminary clinical study suggest that abnormal progesterone profiles in pregnancy may be uncommon in bitches even when there has been previously documented short inter-oestrous interval. However, luteal deficiency may be suspected in bitches with a history of repeated pregnancy failure or abortion. MPA supplementation appears to be efficacious for management of suspected luteal deficiency and verification of the ovarian dysfunction, but care should be taken regarding the timing of MPA withdrawal and prolongation of pregnancy because of delayed elimination of MPA from blood circulation.
The 55- to 80-day luteal phase of the bitch involves progesterone values above 1.5 ng/ml (Concannon 2011). Pregnancy lasts 63 ± 1 days if measured from the day of ovulation (ov), and corpora lutea are the only source of progesterone (P4). Individual P4 profiles may vary considerably including the first day of P4 declining <2 ng/ml, often accepted as the cut-off value regarding impending parturition or pregnancy loss (Verstegen-Onclin and Verstegen 2008; Concannon 2011). Enhanced P4 secretion in pregnancy may involve increases in prolactin (PRL) possibly in response to increased relaxin (RLN; Concannon 2011). In German Shepherd (GSD) bitches, with a short inter-oestrous interval (<5 months) and reduced fertility, P4 concentrations measured from day 5 to day 45 after ov were significantly lower and decreased to <2 ng/ml earlier and more often than in GSD bitches with an inter-oestrous interval ≥6 months (p < 0.05; Günzel-Apel et al. 2006). Pregnancies in bitches with short inter-oestrous intervals had significantly lower concentrations of PRL and RLN between days 20 and 60, which suggest a secondary luteal deficiency because of reduced PRL and perhaps RLN. Comparing normocyclic pregnant GSD bitches with beagles, there were significantly lower P4 values in GSD only in the late luteal phase (days 50–60 after ov; p < 0.05), whereas PRL was significantly (p < 0.05) lower, and RLN tended to be lower in the GSD throughout pregnancy (Günzel-Apel et al. 2009).
The incidence of hypoluteoidism in the population of breeding bitches is still unknown, which is partly due to an insufficient diagnostic approach and lack of clinical data. According to Root Kustritz (2001), P4 concentrations <5 ng/ml in week 4 or 5 may indicate hypoluteoidism and may be considered as a prerequisite for supplemental treatment. A subsequent recommendation was to only start supplementation if P4 concentrations decreased to <5 ng/ml before the last week of gestation, and avoid early administration of progestins as some may be teratogenic if given early in pregnancy (Root Kustritz 2005).
Verification of suspected hypoluteoidism may give an idea about its incidence within the population of breeding bitches, breeds or genetic lines and allow for removal of affected dogs from breeding if hereditability of this ovarian dysfunction is assumed. However, monitoring the ovarian P4 secretion in pregnant bitches is not possible if exogenous P4 is administered (Günzel-Apel et al. 2006, 2009; Verstegen et al. 2008). In our study, we administered the synthetic progestin MPA that allowed monitoring of P4 and luteal function in pregnant bitches suspected of hypoluteoidism.
Materials and Methods
A total of five bitches with different histories suggesting a potential for hypoluteoidism in a future pregnancy were included in the study (Table 1).
Table 1. Data and histories as well as treatment periods and pregnancy outcome of bitches treated with medroxyprogesterone acetate (MPA) for evaluation of luteal function during pregnancy
Period of MPA treatment (days after ov)
Beagle, 3 year
Inter-oestrous interval of 4 months with short luteal phase.
As the history of case 1 and case 2 is identical.
No previous breeding
Spontaneous parturition day 60, 10 vital puppies
Bernese Mountain dog, 4 year
Caesarean operation day 62, four vital puppies
German Shepherd dog I, 5 year
Reduction of inter-oestrous interval from 5.5 to 4 months.
One normal previous pregnancy with seven puppies
Spontaneous parturition day 64, three vital puppies plus one per Caesarean operation
German Shepherd dog II, 3.5 year
Inter-oestrous interval 5 months.
In each of three pregnancies, confirmed loss of all conceptuses within 5 weeks
Caesarean operation day 62, one vital puppy, two dead puppies, one mummified foetus
German Shepherd dog II, 4.5 year
Spontaneous parturition day 58, three vital puppies plus one large dead puppy by Caesarean operation
Rottweiler, 6 year
Inter-oestrous interval 6 months. Three normal pregnancies with 8–14 puppies; three further breedings with suspected pregnancy loss
Caesarean operation day 64, five live and four dead puppies
Prior to breeding, the bitches were submitted to a breeding soundness examination, including inspection of the external genitalia, vaginoscopy, vaginal cytology and bacterial culture, ultrasound of uterus and ovaries, and were found to be healthy. Additional serological testing of bitches 4 (cases 4a and 4b) and 5 for canine herpes virus antibodies gave negative results. In two bitches, cases 1 and 2, progesterone was also monitored in the luteal phase preceding their first breeding.
Ovulation was estimated on the basis of a P4 concentration of 5–8 ng/ml. All bitches were bred once or twice within the period days 1–3 after ov by a fertile male of the corresponding breed. The first sonographic pregnancy diagnosis was performed 23–25 days after ov and repeated at mean intervals of 4 (1–8) days until the time of anticipated parturition. Concurrent blood serum samples were immediately analysed for P4 for monitoring luteal function and assessing the pre-partum P4 decline. Additional serum aliquots were stored at −20°C for later assays.
During pregnancy, medroxyprogesterone acetate (MPA, Perlutex® tablets; Selectavet Dr. Otto Fischer GmbH, Weyarn-Holzolling, Germany) was administered orally (0.1 mg/kg body weight) once daily according to Görlinger et al. (2005), starting days 30–36 after ov and was discontinued on days 51–58 based on concentrations of endogenous P4 (Table 1). After MPA withdrawal, ultrasound was used to assess foetal heart rate. Caesarean operation was performed when low foetal heart rates were detected.
Progesterone concentrations were determined by Immulite® (Siemens Healthcare Diagnostics GmbH, Eschborn, Germany). PRL concentrations were determined using a validated RIA (Okkens et al. 1985) and RLN concentrations as described by Einspanier et al. (2002). Serum samples were additionally analysed for MPA concentrations using a commercially available MPA H3 RIA (Immunometrics, London, UK) to verify clearance of blood circulation from MPA after the end of treatment.
In case 1, progesterone concentrations during the non-pregnant luteal phase were above 5 ng/ml in weeks 4 and 5, but decreased to 2 ng/ml on day 50. In case 2, the P4 concentration in week 5 was almost 5.0 and <1 ng/ml on day 53 after ov. During pregnancy, in both bitches, P4 concentrations showed a normal pattern, with values above 5 ng/ml until day 51 (case 1) and day 46 (case 2), decreasing to <2 ng/ml between days 58 and 60 in case 1 and between days 60 and 62 in case 2 (Fig. 1).
In case 3, P4 declined rapidly between day 33 (13.2 ng/ml) and day 39 (6.1 ng/ml) followed by a slow further decrease to 5 ng/ml in the 7th week (day 46) and to 3.1 ng/ml on day 61, 3 days before spontaneous parturition (Fig. 1). Spontaneous parturition also occurred in case 1 (day 60 after ov), whereas Caesarean operation had to be performed on day 62 in case 2 (Table 1).
The three cases with a history of confirmed (cases 4a and 4b) and suspected previous pregnancy failure (case 5; Fig. 2) also showed P4 concentrations clearly above 5 ng/ml in week 4, but thereafter, a sharp decrease occurred within 6 days between day 31 (18.4 ng/ml) and day 37 (7.4 ng/ml) in case 4b and between day 30 (12 ng/ml) and day 36 (5.5 ng/ml) in case 5. Values <2 ng/ml were reached between days 55 and 62 in case 4a and between days 55 and 58 in case 4b. Despite an identical period of MPA supplementation (days 31/32–57 after ov) in the two consecutive pregnancies of the same bitch (cases 4a and 4b), the first pregnancy had to be delivered by Caesarean operation on day 62, whereas the second pregnancy ended on day 58 with a spontaneous parturition (Table 1).
In case 5, P4 concentrations dropped to <5 ng/ml between days 36 and 43 after ov, and a premature decrease to <2 ng/ml was seen on day 51 after ov clearly indicating hypoluteoidism. After discontinuation of MPA on day 58, no signs of impending parturition were observed until decreasing foetal heart rates observed on day 63. Caesarean operation on day 64 resulted in five live and four dead pups (Table 1). No foetal abnormalities were observed.
Of the six cases, the beagle bitch (case 1) showed the highest level of PRL serum concentrations increasing continuously from mid-to-late pregnancy (8.0–42.4 ng/ml; Fig. 3a). From the two other bitches with a history of short inter-oestrous interval, case 3 had markedly lower PRL values in the corresponding period (3.7–6.3 ng/ml). From the three cases with a history of previous pregnancy losses, the highest PRL concentrations were measured during the second pregnancy of the GSD bitch (case 4b, 5.3–21.9 ng/ml). During the first pregnancy of the same bitch (case 4a), PRL values between days 41–55 were markedly lower (4.9–9.5 ng/ml). In case 5, PRL concentrations were at a similar low level (3.3–8.2 ng/ml) as in case 3 (Fig. 3a).
Regarding RLN concentrations, highest values were found in the pregnancies of the three bitches with a history of a short inter-oestrous interval (case 1: 1.5–3.8 ng/ml, case 2: 1.9–2.3 ng/ml, case 3: 0.9–2.0 ng/ml; Fig. 3b), with the beagle bitch (case 1) showing a high level in the last 3 weeks of pregnancy. From the bitches with a history of previous pregnancy failure, RLN concentrations of <1 ng/ml were measured throughout pregnancy in case 4a (0.6–0.9 ng/ml) and in case 5 (0.3–0.9 ng/ml). Slightly higher values were analysed in the second pregnancy of bitch 4 (case 4b: 1.0–1.3 ng/ml).
In cases 1 and 4b in which pregnancies ended with spontaneous parturition, serum MPA concentrations declined to <1 nmol/l within 3 days and 1 day after MPA withdrawal, respectively (Table 2). In cases 2 and 5, MPA concentrations were still above 1 nmol/l on days 5 and 4 after MPA withdrawal, respectively, thus probably prolonging gestation especially in case 5, in which P4 values were below 2 ng/ml since at least day 51 (Table 2). Delivery of four dead puppies of nine may additionally indicate a delayed surgical intervention by 24 h.
Table 2. Blood serum concentrations of progesterone (P4 ng/ml) and medroxyprogesterone acetate (MPA; nmol/l) in the pre-partum period of six pregnancies starting the day before discontinuation of oral MPA treatment (0.1 mg/kg body weight; day 0)
Case no. (day of MPA withdrawal related to ovulation)
Days from MPA withdrawal (day 0)
Spontaneous parturition (P)
1 (day 52)
3 (day 56)
No further values available
4b (day 57)
Caesarean operation (C)
2 (day 57)
4a (day 57)
5 (day 58)
In this study, bitches with suspected luteal deficiency were supplemented with oral MPA (0.1 mg/kg), allowing measurement of endogenous progesterone throughout pregnancy. This treatment regime has been shown to prevent abortion in a bitch suffering from low progesterone concentrations between days 42 and 47 after mating (Görlinger et al. 2005). Furthermore, MPA (10 mg/dog/day i.e. approximately 0.7 mg/kg) given orally in the mid-luteal phase did not affect the secretion of either P4 or PRL in normocyclic non-pregnant beagle bitches (Günzel-Apel et al. 2009). In the present study, treatment was not started before day 30 after ov to ensure the completion of embryonic organogenesis. Discontinuation of MPA treatment, based on endogenous P4 concentrations, was the earliest on day 52 (case 1) and the latest on day 58 (case 5). Spontaneous parturition took place 7 days (case 1), 6 days (case 3) and 1 day (case 4b) after the last MPA administration, the latter being similar to the case described by Görlinger et al. (2005); however, Caesarean operation had to be performed in the previous pregnancy of the same bitch (case 4a) 6 days after MPA discontinuation. As in cases 2 and 5 MPA concentrations were still at an elevated level on days 5 and 4 after the end of treatment, respectively, MPA related prolongation of gestation appears to be likely. No congenital abnormalities were noted in any puppies despite the concern that progestin administration during pregnancy may induce foetal deformities (Root Kustritz 2005), probably because MPA has a low propensity for producing androgenic side effects. Many factors, among them body weight and age but not the reproductive status (luteal phase vs pregnancy), both in mid- (days 20–40 after ov) and late luteal phase (days 41–55 after ov), have been shown to influence characteristics and progesterone production of the corpora lutea (Marinelli et al. 2009). However, the luteal phase P4 patterns of the two bitches with an inter-oestrous interval of 4 months (cases 1 and 2) clearly indicate that shortening of the inter-oestrous interval can at least partly be related to a relatively short luteal phase. This was also demonstrated in short cycling GSD bitches (Günzel-Apel et al. 2006). However, luteal function in the non-pregnant state does not necessarily represent the competence of corpora lutea for maintenance of pregnancy. Normal luteal function was verified during pregnancies in cases 1, 2 and 3.
None of the bitches with a history of suspected previous pregnancy loss (cases 4a, 4b and 5) had P4 concentrations <5 ng/ml in weeks 4 and 5 of pregnancy, which would indicate hypoluteoidism according to Root Kustritz (2001). However, P4 concentration in case 5 was 5.5 ng/ml on day 36 after ov and decreased continuously to <2 ng/ml before day 51 indicating hypoluteoidism. In case 4a (GSD bitch with a history of previous pregnancy loss within the first 5 weeks, diagnosed by ultrasound examination), P4 concentrations were >2 ng/ml until day 55 (4.2 ng/ml) and normal luteal function during the whole pregnancy can be assumed from the value measured on day 62 (1.8 ng/ml). Luteal function seems borderline in the second pregnancy of the same bitch (case 4b), as P4 concentrations decreased sharply to 0.9 ng/ml on day 58, when parturition occurred.
Prolactin concentrations are mainly reported for pregnant beagle bitches (Onclin and Verstegen 1997; Günzel-Apel et al. 2009) and are in accordance with those found in case 1 of our study. None of the other bitches showed an equally high PRL concentration. Significantly lower PRL values were found throughout pregnancy in GSD dogs compared to pregnant beagle bitches, both groups with a normal cycle length (Günzel-Apel et al. 2009). In GSD bitch I (case 3) and in the first pregnancy of GSD bitch II (case 4a), PRL values were at markedly lower level than in the second pregnancy of GSD bitch II (case 4b), in which also a more pronounced PRL increase was observed, indicating intra-individual differences both in luteal and pituitary hormone secretion. Lowest PRL concentrations were detected in case 5, in which hypoluteoidism was verified. As this was also true for the RLN concentrations, insufficient luteotropic support should be considered as causal background.
Highest RLN values were found in the beagle bitch (case 1), fitting well to those measured in a previous study (Günzel-Apel et al. 2009). In the same study, RLN concentrations in normocyclic pregnant GSD dogs (n=11) were markedly lower, similar to the RLN values of GSD I (case 3) and GSD II (case 4b). As in bitch 4 (cases 4a and 4b) the first three previous pregnancies ended with foetal loss, a congenital ovarian dysfunction cannot be excluded. In the Rottweiler bitch with three normal previous pregnancies (case 5), acquired hypoluteoidism is assumed.
Despite the small number of cases included in this preliminary clinical study, the results indicate that the incidence of hypoluteoidism may be low in bitches with short inter-oestrous intervals. However, dogs with a history of recurrent pregnancy loss may represent recurring instances of hypoluteoidism. MPA supplementation in cases of suspected pregnancy loss enables maintenance of pregnancy and verification of luteal dysfunction. However, clearance of MPA from blood circulation appears to differ between bitches resulting in a potential risk of pregnancy prolongation. Thus, careful prediction of parturition and monitoring of the foetal heart rate in the prepartum period is required to prevent intrauterine foetal death.
Conflicts of interest
None of the authors have any conflicts of interest to declare.
AGA supervised the clinical cases, evaluated the results and wrote the manuscript. CU was responsibly involved in the clinical care of the pregnant bitches, in controlling MPA treatment and in blood sampling. KW assisted in the clinical care of the bitches and in blood sampling. AE conducted the relaxin analysis. CO conducted the prolactin analysis. MP conducted the progesterone analysis.