Aglepristone and cloprostenol combination in the termination of late-term pregnancy in queens

Abstract

Objectives

Termination rates for the highly recommended aglepristone (AGL) treatment are low in late-term pregnancy in queens. We studied the effects of an AGL and cloprostenol (CLO) combination in the termination of late-term pregnancy.

Methods

Pregnant queens were assigned to two groups. Queens in the AGL group (n = 10) received AGL 10 mg/kg, twice, 24 h apart. Queens in the AGL-CLO group (n = 9) were additionally injected with a single dose of CLO (5 μg/kg) 24 h after the second dose of AGL. Progesterone, 17beta(β)-oestradiol, cortisol, oxytocin and prostaglandin F2alpha (PGF2α) metabolite were measured in sera obtained at days 0, 1 and 2, and on the day of abortion.

Results

Average gestational age in both groups was similar (AGL 38.61 ± 0.91 days vs AGL-CLO 39.39 ± 1.35 days; P >0.05). Termination rates were 80% and 100% in the AGL and AGL-CLO groups, respectively (P <0.05). Fetal expulsion time was significantly longer (P <0.001) in the AGL group (96.9 ± 6 h) compared with the AGL-CLO group (69.8 ± 3.3 h). Duration of abortion was 19.8 ± 2.6 h and 12.6 ± 1.4 h in the AGL and AGL-CLO groups, respectively (P <0.05). Both treatments were well tolerated. Significantly (P <0.05) lower serum progesterone concentrations were observed in both groups at the day of abortion and concentrations in the AGL-CLO group (4.19 ± 0.80 ng/ml) were lower than in the AGL group (9.89 ± 2.21 ng/ml; P <0.05).

Conclusions and relevance

AGL and CLO combination increases pregnancy termination rate in late-term pregnant queens. In addition, CLO contributes to a decrease in luteal function in AGL-treated late-term pregnant queens.

Introduction

Termination of pregnancy is a common practice in the control of reproduction in cats. Pregnancies after unwanted or unobserved matings are usually detected in the advanced stages of gestation, when the abdomen is enlarged.^1–3 Medical methods of mid- and late-term pregnancy termination in queens include repeated administrations of prostaglandins (PGs), dopamine agonists and combinations thereof. Nevertheless, these applications have been reported to be disadvantageous due to the necessity of multiple applications and unsatisfactory efficacy.^4–8 Prostaglandin F2alpha (PGF_2α) has been reported to induce abortion in queens in the last trimester by causing luteolysis. For this purpose, cloprostenol (CLO), a synthetic PGF_2α, can be used. However, this hormone has significant side effects (salivation, vomiting, diarrhoea) and the success rate (25–75%) varies. ³ Progesterone, produced by corpora lutea (CL), is responsible for the maintenance of pregnancy in queens. ⁹ However, progesterone is produced from the placenta after 40–45 days of pregnancy. Even after 45–50 days of pregnancy, CL is not necessary to maintain pregnancy. ¹⁰ From this point, pregnancy can be terminated by blocking progesterone receptors. Recently, aglepristone (AGL), a progesterone receptor antagonist, has been recommended for medical termination of feline pregnancy.^3,11 AGL, licensed for pregnancy termination in bitches in Europe from day 0 to 45 after mating, interferes with progesterone metabolism not only in the uterus, but in all target organs and at all stages of pregnancy, as confirmed by the fact that it can also be used in the induction of parturition. ¹²

The safety and efficacy of AGL treatment has been demonstrated in pregnancy termination in queens, with no negative effects on subsequent fertility.^2,13 Two injections of AGL shortly after mating prevents implantation in 100% of cases. ¹⁴ Reported success rates in mid-term pregnancy, however, are lower than in mid-term pregnancy termination in dogs. ¹⁰ On days 25–26 after mating, AGL (10 mg/kg SC 24 h apart) resulted in abortion in 87% of ultrasonographically confirmed feline pregnancies. ¹⁵ A higher dose of AGL does not alter the outcome of treatment; when two doses of 15 mg/kg AGL were injected subcutaneously 24 h apart to 61 queens at mid-gestation the termination rate was 88.5%. ² AGL’s effectiveness for termination appears to decrease as the pregnancy advances. AGL injected in pregnant queens on days 45 and 46 after mating terminated pregnancy in only 66.7% of cases. ¹⁶

Pregnancy termination is achieved in around 4 days after the second injection of AGL and fetal expulsion is generally completed within 8–10 days. Treatment failure results in continuation of pregnancy, fetal retention, partial abortion or retained fetal membranes.^2,15 Exogenous PGF_2α, acting on both luteal function and myometrial activity, has potential effects in the prevention of such complications. The effectiveness of PGs on the induction of abortion is known to begin on day 40 of pregnancy. ¹ Applications before this period may also be insufficient to stimulate abortion. ⁴ Although the effectiveness of PGs in these animals cannot be fully explained, they have been reported either to induce direct abortion on the uterus or to induce luteolysis of gravid CL. ¹⁷ Fieni et al ² have reported placental retention and consequent uterine infection after mid-term feline pregnancy termination with AGL that disappeared after PG treatment. In addition, fetal expulsion is accompanied by an increase in oestrogen, oxytocin and PG concentrations. A significant peak of plasma PGF_2α metabolite (PGFM) concentrations, the main metabolite of PGF_2α, is observed 42 h before fetal expulsion. We therefore studied the effects of the addition of a single PG (CLO) injection to the common AGL protocol in the termination of late-term pregnancy in queens.

Materials and methods

The respective local authority approved the animal experimentation (Local Ethics Committee on Animal Experiments, Ankara University, Turkey; approval no: 122-474).

A total of 19 queens (aged 1–4 years) of different breeds (Table 1), introduced to the Department of Obstetrics and Gynecology Small Animal Clinic, Faculty of Veterinary Medicine, Ankara University, because of unwanted late-term pregnancy, were included. Pregnancy diagnosis and determination of gestational age were performed by ultrasonography (6.5 MHz convex transducer; Mindray DC-N3) according to Zambelli et al. ¹⁸ At the first ultrasonographic examination, we attempted to determine the number of fetuses. Then, blood samples were taken from the animals to evaluate serum biochemical parameters (blood urea nitrogen, urea, creatinine, alkaline phosphatase, aspartate aminotransferase, gamma-glutamyl transpeptidase, creatine kinase, albumin) and a complete blood count was performed in order to determine the general health status of the queens. Those with abnormal values were considered unhealthy and were not included further in the study.

Table 1

Descriptive data of the queens in both groups

Group	ID	Breed	Age (months)	Weight (kg)	Volume of AGL (ml)	Pregnancy day	Fetal expulsion time (h)	Duration of abortion (h)	Litter size
AGL	1	Tabby	15	2.85	1.0	36.3	94	18	5
	2	Mix	20	4.15	1.4	38.6	88	12	3
	3	Angora	10	3.20	1.1	44.8	122	11	4
	4	Mix	13	3.40	1.1	38.5	91	26	6
	5	Bombay	27	3.80	1.3	35.9	105	30	2
	6	Tabby	14	3.05	1.0	37.8	–	–	3
	7	Persian	38	4.10	1.4	36.8	67	14	7
	8	Mix	12	3.01	1.0	35.9	92	28	5
	9	Siamese	19	3.30	1.1	42.2	116	19	4
	10	Mix	13	3.03	1.0	38.9	–	–	3
AGL-CLO	1	Angora	21	4.25	1.4	41.2	70	7	4
	2	Tabby	12	3.28	1.1	33.0	55	8	6
	3	Bombay	14	2.95	1.0	43.6	63	13	7
	4	Van	12	3.24	1.1	38.7	68	12	3
	5	Mix	18	3.90	1.3	44.3	82	18	5
	6	Mix	15	4.58	1.5	35.6	65	16	4
	7	Tabby	21	3.27	1.1	36.5	79	15	4
	8	British	23	3.50	1.2	43.7	84	16	5
	9	Mix	16	3.10	1.0	37.4	62	8	4

AGL = aglepristone; CLO = cloprostenol

Queens were randomly assigned to two groups (AGL and AGL-CLO groups). In the AGL group (n = 10), animals were injected with AGL 10 mg/kg (as a single bolus) subcutaneously twice, 24 h apart. Animals in the AGL-CLO group (n = 9) were additionally injected with a single subcutaneous dose of CLO (5 μg/kg) 24 h after the second AGL injection. All animals were hospitalised and pregnancy termination was monitored by daily ultrasonography examinations. Queens were monitored 24 h a day using a video camera for the detection of side effects, such as vomiting, diarrhoea and anorexia/depression. In addition, after AGL administration, the cats were palpated to detect local pain. Fetal expulsion time (defined as the mean interval between injection of the first dose of AGL and expulsion of the last fetus), duration of abortion, number of fetuses (Table 1) and side effects observed were recorded. The number of fetuses were determined following the abortion. Moreover, all queens were checked ultrasonographically to determine whether there were any fetuses left in the uterus.

Peripheral blood samples were collected from the cephalic vein at days 0 (first AGL injection), 1 (second AGL injection) and 2 (24 h after second AGL), prior to administration of abortifacient agents, and on the day of abortion immediately after expulsion of the last fetus. Sera obtained after blood centrifugation (3000 g for 15 mins) were frozen at −20°C until hormone measurements were performed. Serum progesterone, 17beta(β)-oestradiol and cortisol concentrations were determined by an electrochemiluminescence immunoassay using an auto-analyser (Cobas6000 C601; Roche Diagnostics). 17β-oestradiol was measured with intra- and inter-assay coefficients of variation (CV) of 2.4% and 3.8%, respectively. The sensitivity of the assay was 5 pg/ml. For progesterone, maximum intra- and inter-assay CVs and the minimal assay sensitivity were 1.1%, 3.3% and 0.030 ng/ml, respectively. Intra- and inter-assay CVs were 1.6% and 1.9%, respectively, for cortisol, with an assay sensitivity of 0.054 µg/dl.

For oxytocin measurement, a feline-specific commercial ELISA kit (Mybiosource) was used and run in accordance with the manufacturer’s instructions. The sensitivity of the assay was 0.8 pmol/l. The detection range of the kit was 5.0–160.0 pmol/l. Both the intra- and inter-assay CVs were <15%. A commercial 3, 14-dihydro-15-keto Prostaglandin F_2α EIA kit (DetectX; Arbor Assays) was used and run according to the manufacturer’s instructions for PGFM measurements. The sensitivity was 20.8 pg/ml. The cross-reactivity for 13, 14-dihydro-15-keto PGF_2α and the PGE₂ metabolite were 100% and 1.5%, respectively. The intra- and inter-assay precision variation was 13.2 % and 12.6%, respectively.

Data are presented as mean ± SEM. All statistical comparisons were performed using SPSS (Version 14 for Windows; IBM). The assumptions of normality and the homogeneity of variances were tested by the Shapiro–Wilk’s test and Levene’s test, respectively. Continuous variables that were normally distributed were compared between the groups by an unpaired Student’s t-test. Continuous variables that were not normally distributed were compared between groups by the Mann–Whitney U-test. The frequency of side effects were compared by χ² test. The significance level was set at P <0.05.

Results

The clinical results are illustrated in Table 2. The average gestational age in both groups was similar (P >0.05). Pregnancy was successfully terminated in 8/10 queens (80%) in the AGL group. No clinical signs associated with pregnancy termination were observed in two queens and fetal viability was detectable on ultrasonography on day 8. Ovariohysterectomy was performed with the owner’s consent to terminate the pregnancy. Abortion was successfully induced in all nine queens in the AGL-CLO group (P <0.05). Average fetal expulsion time was significantly longer (P <0.001) in the AGL group compared with the AGL-CLO group. Duration of abortion was 19.8 ± 2.6 h and 12.6 ± 1.4 h in the AGL and AGL-CLO groups, respectively (P <0.05). Litter size was not different (P >0.05) between the groups.

Table 2

Clinical findings in aglepristone (AGL) and AGL plus cloprostenol (CLO) groups, and side effects observed during the study

Parameters	AGL (n = 10)	AGL-CLO (n = 9)	P value
Gestational age (days)	38.61 ± 0.91	39.39 ± 1.35	0.629
Pregnancy termination rate (n [%])	8 (80)	9 (100)	<0.05
Fetal expulsion time (h)*	96.9 ± 6.1	69.8 ± 3.3	0.001
Duration of abortion (h)*	19.8 ± 2.6	12.6 ± 1.4	0.024
Litter size	4.4 ± 0.5	4.7 ± 0.4	0.480
Side effects (n)
Pain at the injection site	3	4	0.650
Anorexia/depression	5	4	0.988
Vomiting	0	3	0.087
Diarrhoea	0	1	0.474

Data are mean ± SEM unless otherwise indicated

*

Two queens in the AGL group that failed to abort were excluded from the analysis

Both treatments were well tolerated. Side effects recorded were pain at the injection site, anorexia/depression, vomiting and diarrhoea (Table 3). Side effects were self-limiting and resolved spontaneously. Local reaction was observed at the injection site in three queens in the AGL group and in four queens in the AGL-CLO group after the first injection (P = 0.650). Transient anorexia/depression was observed in four and five queens in the AGL and AGL-CLO groups, respectively (P = 0.988). Self-limiting vomiting (P = 0.087) and diarrhoea (P = 0.474) were observed after CLO injection in three and one queens, respectively.

Table 3

Observed side effects during the study in both groups

Group	ID	Vomiting	Duration of vomiting (h)/no. of episodes	Diarrhoea	Durations of diarrhoea (days)/no. of episodes	Anorexia/depression	Pain at the injection site
AGL	1	No	–	No	–	No	No
	2	No	–	No	–	Yes	Yes
	3	No	–	No	–	No	No
	4	No	–	No	–	No	No
	5	No	–	No	–	Yes	Yes
	6	No	–	No	–	No	No
	7	No	–	No	–	Yes	Yes
	8	No	–	No	–	Yes	No
	9	No	–	No	–	Yes	No
	10	No	–	No	–	No	No
AGL-CLO	1	No	–	No	–	Yes	Yes
	2	Yes	2/5	Yes	1/4	Yes	No
	3	No	–	No	–	No	Yes
	4	No	–	No	–	No	No
	5	Yes	1/3	No	–	Yes	Yes
	6	No	–	No	–	Yes	Yes
	7	No	–	No	–	No	No
	8	No	–	No	–	No	No
	9	Yes	1/2	No	–	No	No

AGL = aglepristone; CLO = cloprostenol

Mean serum progesterone concentrations in the AGL and AGL-CLO groups at days 0, 1 and 2, and the day of abortion are shown in Figure 1. Concentrations were stable and did not differ between study groups (P = 0.621) at day 0 (AGL 18.05 ± 2.99 ng/ml vs AGL-CLO 21.15 ± 3.67 ng/ml), day 1 (AGL 19.24 ± 3.10 ng/ml vs AGL-CLO 19.22 ± 2.75 ng/ml) and day 2 (AGL 19.78 ± 3.23 ng/ml vs AGL-CLO 19.37 ± 3.22 ng/ml). Significantly (P = 0.041) lower mean progesterone concentrations were observed in both groups at the day of abortion compared with the mean concentrations at days 0, 1 and 2. In addition, serum progesterone concentrations were lower in AGL-CLO-treated queens (4.19 ± 0.80 ng/ml) than in AGL-treated queens (9.89 ± 2.21 ng/ml; P <0.014).

Figure 1

Mean serum progesterone concentrations in aglepristone (AGL) and aglepristone + cloprostenol (AGL-CLO) groups at days 0, 1 and 2, and the day of abortion. A statistical difference (P <0.05) between time points is indicated by different letters

Mean serum 17β-oestradiol, cortisol, oxytocin and PGFM concentrations are illustrated in Figure 2. Serum 17β-oestradiol concentrations did not differ between the AGL and AGL-CLO groups (P = 0.848). Mean 17β-oestradiol concentrations were stable over time (P = 0.942), ranging from 15.50 ± 2.56 to 13.50 ± 2.16 and from 15.22 ± 2.41 to 13.33 ± 1.76 pg/ml in the AGL and AGL-CLO groups, respectively. Serum cortisol concentrations were not different between the AGL and AGL-CLO groups at days 0, 1 and 2, and at the day of abortions (P = 0.750). Concentrations increased over time in both groups (P = 0.001) and the highest values were observed on the day of abortion.

Figure 2

Mean serum (a) 17β-oestradiol, (b) cortisol, (c) oxytocin and (d) prostaglandin F_2α metabolite (PG) concentrations in aglepristone (AGL) and aglepristone + cloprostenol (AGL-CLO) groups at days 0, 1 and 2, and the day of abortion. A statistical difference (P <0.05) between time points is indicated by different letters

Similar concentrations of oxytocin were observed in the study groups at any sampling time (P = 0.758). Oxytocin concentrations significantly increased at the day of abortion (AGL 24.78 ± 1.35 vs AGL-CLO 25.70 ± 1.27 pmol/l; P = 0.012). Likewise, serum PGFM concentrations did not differ between the AGL and AGL-CLO groups (P = 0.995). PGFM concentrations before administration of the first dose of AGL were 4.67 ± 0.17 and 4.73 ± 0.16 ng/ml in the AGL and AGL-CLO groups, respectively. Concentrations were stable in the following two days. A significant (P <0.001) increase was observed on the day of abortion in both groups (AGL 5.24 ± 0.21 ng/ml vs AGL-CLO 5.22 ± 0.20 ng/ml).

Discussion

The AGL efficacy rate of 80% seen in the present study was slightly lower than success rates reported for AGL treatment on days 25–26 ¹⁴ and on days 33.3 ± 4.2.² This result supports the suggestion that AGL-induced termination is less effective in late-term feline pregnancy than in mid-term termination. ¹¹ Even lower termination rates have been reported for progesterone receptor antagonists at later pregnancy. In an early study, a lower potent anti-progestin mifepristone terminated pregnancy in only 1/4 queens between days 40 and 46 post-mating, ¹⁹ and AGL, administered to pregnant queens on days 45 and 46 after mating, terminated two-thirds of pregnancies. ¹⁶ The reasons for treatment failure are unclear. The luteotropic action of prolactin and placental progesterone production has been suggested to interfere with the action of AGL in late-term pregnancy. ¹⁶ The present study demonstrated that the addition of a single CLO administration to the common AGL protocol increased success rate in the termination of feline pregnancy and pregnancy termination was achieved in all queens studied.

Treatment failure in AGL-induced pregnancy termination may be partial or complete with fetal retention, subsequent birth of normal viable kittens and/or expulsion of macerated fetuses. Repetition of treatment with the same protocol or with PGs has been suggested. ¹⁵ Surgical intervention is an alternative in such cases. Ovariohysterectomy performed in two queens after detecting no sign of abortion on day 8 of the present study highlights the importance of ultrasonography examination on days 7–10 in AGL-induced pregnancy termination in the queen.^3,15 Continuation of late-term pregnancy with live and growing fetuses may pose ethical problems, preventing further medical and surgical interventions.

Shorter fetal expulsion time and mean duration of abortion achieved by the combination of AGL and CLO has importance from a practical point of view. Hospitalisation is routine during pregnancy termination in companion animals in our clinic, and a reduction in fetal expulsion time offers shorter hospital stays. Furthermore, close monitoring of queens is needed during abortion. In another study, the effectiveness of AGL on the induction of abortion in queens on days 25–26 after mating was assessed, and the mean duration of AGL-induced abortion, defined as the time from first occurrence of vaginal bleeding to expulsion of all fetuses, was 1 day in nine cats and 2 days in five cats. ¹⁵ The mean duration of abortion observed in the present study is in accordance with the average parturition length in queens. Most queens deliver kittens within a period of several hours. In a large questionnaire-based study, the time from the birth of the first to the last kitten was < 6 h in nearly 86% of cases. ²⁰ Occasionally, queens have intervals between births of kittens of up to 1 or 2 days. ²¹ Root et al ²² reported an average parturition length of 16 h (range 4–42 h) in seven research colony cats.

Induction of abortion with AGL is considered safe for queens. ² The negligible side effects observed in the present study are in agreement with previous studies that used AGL for pregnancy termination in queens.^2,11,15,23 Pain at the subcutaneous injection site or a mild local reaction is a common adverse effect of AGL in cats. Pain at the injection site has been reported to affect up to 10% of AGL-treated queens.^2,15 In a previous study, AGL treatment induced a transient decline in general demeanour in 9.3% of queens. ² In another study including six queens, two queens had decreased feed intake and three had increased vocalisation. ²³ Twenty-four-hour monitoring of queens using a video camera is an effective method for identifying anorexia/depression, vomiting and diarrhoea. However, increased frequency of pain at the injection site may cause anorexia/depression. The AGL injection site in cats and dogs is the area between the shoulders. This region is easily accessible in both species. However, as AGL is an oily preparation, it may cause pain at the application area. This affects the comfort and welfare of the animal, and may cause anorexia. ² At the same time, the amount of drug given may cause increased pain as it was determined that the incidence of pain was higher in animals treated with large amounts of drugs in this study.

Self-limiting vomiting and diarrhoea observed after CLO injection are among the classical side effects of PG administration in cats and they are due to the stimulatory effect of PGF_2α on smooth muscles. Synthetic analogues that are administered at lower doses and have a longer duration of action are advocated over natural PGs to avoid side effects in companion animals. ²⁴ However, the synthetic analogue CLO has more side effects in cats because it is more potent. ²⁵ In a previous study, 5 µg/kg CLO on three consecutive days induced vomiting in all of the 13 pregnant queens. ⁸ In the present study, a single injection of CLO induced vomiting in one-third of the queens. Gastrointestinal side effects can be prevented by combining CLO with antiemetic drugs. ¹

AGL treatment in pregnant queens leads to an initial increase in progesterone concentrations and then to a continuous decrease. A significant increase in progesterone has been reported 60 h after AGL treatment. ² In another study, progesterone concentrations started to increase 28 h after AGL injection and peaked at 77 h. ²³ Early blood samplings in the present study are considered to be the reason for the lack of progesterone increase at days 1 and 2. It was also suggested that this protective increased progesterone release is only observed in some queens. ¹⁵

Progesterone concentrations typically decrease after day 30 of normal pregnancy in the cat.^26,27 Although the placenta is an additional source of progesterone in pregnant queens, ²⁸ placental progesterone does not affect serum progesterone concentration. ²⁹ The markedly low serum progesterone concentrations observed in both groups on the day of abortion is a well-described phenomenon after application of progesterone antagonists in dogs ¹² and cats, ² supposedly caused by fetal death. In addition, AGL has an incomplete luteolytic effect in pregnant dogs and it initiates an endocrine cascade similar to that observed during normal luteolysis. Consequently, uteroplacental PGF_2α output increases and leads to preterm luteolysis and/or abortion. ³⁰ Contrary to this, AGL has a luteolytic effect when used in diestrus bitches at days 29–30 following the estimated day of LH surge with complete luteolysis occurring in about 8–10 days. ³¹

In the regulation of luteal function, there is a difference between the first and second parts of the luteal phase in cyclic and pregnant bitches. The function of CL is independent of pituitary support between ovulation and 24–28 days of the luteal phase. ³² During the second part of the luteal phase, luteotropic factors such as prolactin and luteinising hormone are necessary for the maintenance of luteal function.^32–34 Therefore, progesterone receptor antagonists show a different endocrine effect in the early luteal phase in non-pregnant cyclic bitches than in early or mid-gestation bitches. ³⁵ The short duration of the luteal phase after AGL administration in mid-term pregnancies is associated with an increase in plasma PGFM concentration. The PGFM induces fetal expulsion; ³⁶ however, it has been reported that PG has no effect on progesterone secretion in non-pregnant bitches. ³⁵ Nevertheless, it should be kept in mind that progesterone secretion differs, as cats were used in the present study. Further studies are needed to clarify whether a similar luteolytic effect of AGL exists in queens.

Possible mechanisms for the increased pregnancy termination rate and significantly shorter fetal expulsion time and mean duration of abortion observed in the AGL-CLO group are faster reduction of luteal progesterone production and/or an increase of uterine motility. Contraction of the myometrium is one of the most significant physiological effect of PGF_2α in the female reproductive system. To a lesser extent, it is involved in relaxation of the cervix in the queen. ³⁷

The role of PGF_2α in the regulation of gravid CL in the domestic cat, however, is not fully understood. In contrast to early studies reporting that PGF_2α or its analogues given to pregnant queens induces luteolysis and consequent abortion after day 33 ⁹ or day 40 of gestation, ⁴ most studies have demonstrated only a partial decrease in luteal function after exogenous PGF_2α administration.^17,38 Considering synthetic analogues, three injections of CLO were ineffective in inducing abortion in queens at 21–22 or 35–38 days of pregnancy. ⁸ Regarding the day of pregnancy chosen in the present study, Baldwin et al ³⁹ reported that 40 days appeared to be the critical time for initiation of PG effectiveness. Furthermore, days 38 and 39 of pregnancy have been identified as a period of early histomorphological CL regression in the domestic cat, which is characterised by a decrease in intraluteal progesterone concentrations. ⁴⁰

Luteal regression is steadily progressive in the domestic cat, ⁴¹ starting at the time when PGFM concentrations increase in urine and feces. ⁴² The lower serum progesterone concentrations on the day of abortion in the AGL-CLO group indicates that CLO administration contrib-uted to a decrease in luteal function in AGL-treated late-term pregnant queens. Yet, higher than baseline concentrations of progesterone at the day of abortion demonstrates the failure in inducing complete luteolysis by a single CLO administration. In the pregnant queen, progesterone concentrations decline to 10 ng/ml by day 60 of pregnancy, to 5 ng/ml immediately before parturition and to <1 ng/ml after parturition. ²⁶

Stable plasma oestradiol concentrations were reported for queens and bitches after AGL treatment. In the present study, average serum 17β-oestradiol concentrations did not differ between study groups and levels observed corresponded to normal pregnancy values. ²⁶ In AGL-induced feline pregnancy termination, an increase 12 h before the time of expulsion and subsequent return to normal value at the time of expulsion were reported. ² In the present study, blood samples were collected after completion of last fetal expulsion and no change in 17β-oestradiol concentration was observed. In physiological parturition, mean serum oestradiol concentrations peak about 1 week before fetal expulsion and then drops by the day of abortion. ²⁷

In the present study, serum cortisol concentrations increased in a similar fashion in both groups; the highest values were observed after completion of fetal expulsions. Progesterone and glucocorticoid receptors share regions of high homology and AGL has anti-glucocorticoid effects because of a possible blocking of glucocorticoid receptors. Blocking of pituitary glucocorticoid receptors results in an increased adrenocorticotropic hormone release and a subsequent elevation of cortisol secretion. ⁴³ Little is known about the effects of AGL treatment on adrenal function in cats. Elevated plasma cortisol concentrations in AGL-treated queens has been reported previously. ² It is anecdotally believed that there is a stronger association between AGL treatment and plasma cortisol concentrations in queens vs bitches. One of our concerns during the study was that CLO administration might have induced a dramatic increase in cortisol concentrations. Only one blood sampling after CLO administration in the present study limits the interpretation. Taking a few extra millilitres of blood to get some serum on which to assay cortisol would have created no extra stress to the queens. However, serum cortisol concentrations were not different between the study groups and no stress response requiring intervention was observed.

Taking the role of oxytocin in abortion and parturition into consideration, increased serum oxytocin concentrations observed at the day of abortion in the present study was expected. Increased plasma oxytocin concentrations reaching a peak at the time of fetal expulsion after AGL-induced feline abortion was reported previously. ² Fetal expulsion was further characterised by an increased PGFM, which was confirmed in the present study. PGFM concentrations markedly increase in the plasma of pregnant queens and levels are mirrored by PGF_2α concentrations in placental tissue. ⁴⁴ Owing to the very short half-life of PGF_2α, CLO injection on day 2 of the study did not affect PGFM concentrations on the day of abortion.

Conclusions

The addition of a single CLO administration to the common AGL protocol in the termination of late-term pregnancy in queens increases the pregnancy termination rate and significantly shortens fetal expulsion time and mean duration of abortion. Also, CLO administration contributes to a decrease in luteal function in AGL-treated late-term pregnant queens. However, the serum progesterone concentration on the day of abortion was lower in the AGL-CLO group than in the AGL group. On the day of abortion, PGFM, oxytocin and cortisol levels were found to be at their highest levels. Moreover, the side effects in both treatment groups were considered to be self-limiting and tolerable as they resolved spontaneously.