Prevention of pregnancy in the dog with a combination of prostaglandin F₂α and bromocriptine

Abstract

Fifteen mated bitches were given prostaglandin F_2α (PGF_2α) [250 μg/kg body weight] and bromocriptine (10 μg/kg BW) twice daily from days 6 to 10 of diestrus. Progesterone concentrations declined during treatment. None of the bitches whelped. Daily treatment with PGF_2α and bromocriptine for 5 d appears to induce luteolysis and prevent early pregnancy.

The problem of canine overpopulation has long been recognized (1), with seemingly little progress being made to reduce the number of unwanted dogs that are destroyed each year. A major source of unwanted dogs is accidental matings of bitches whose owners are unwilling to pursue surgical sterilization. In the past, the earliest treatment for mismating consisted of the administration of natural or synthetic estrogen within 1 to 5 d of breeding in an attempt to disrupt embryonic implantation. This treatment has been associated with pyometra, nonreversible anemia due to bone marrow suppression, and even death (1,2,3). Currently, there is no method that is 100% reliable for terminating pregnancy safely in late estrus or early diestrus.

Maintenance of canine pregnancy is dependent on functional corpora lutea (CL) throughout gestation (2). Luteal function in the dog is thought to depend on pituitary support, rather than uterine or placental influences (4). The 2 hormones principally responsible for maintenance of the canine CL are luteinizing hormone (LH) and prolactin (PRL) (4).

The LH peak of estrus is sufficient to establish luteal function for approximately 4 wk. Prolactin becomes the dominant luteotropin during the 2nd half of gestation (3,5). The dopamine agonists bromocriptine and cabergoline have been used to suppress PRL (3,5,6) and, hence, progesterone in the pregnant bitch (4). Bromocriptine and cabergoline have been used to induce abortion during the 2nd half of gestation but not during the early luteal phase (3,5). Prolonged fetal expulsion due to insufficient smooth muscle contraction appeared to be a drawback to the use of these products (5).

Although prostaglandin F_2α (PGF_2α) can be used to terminate pregnancy in the 2nd half of gestation (3,7,8), it does not appear to cause luteolysis reliably during the 1st half of gestation. A protocol using 250 μg/kg body weight (BW) of PGF_2α, q12h, for 4 d beginning on day 5 of cytologic diestrus terminated pregnancy in 4 of 5 cases (7). A combination of a PGF_2α analog (cloprostenol) and cabergoline administered daily for 5 d was 100% effective in terminating pregnancy when initiated on day 25 after the LH peak (3), which corresponds to approximately day 17 of cytologic diestrus. We hypothesized that a combination of PGF_2α and bromocriptine, at sufficient dosage and duration, would be an effective abortifacient in early pregnancy in the dog.

Ultrasonography as early as 20 d may reliably detect pregnancy, but it is preferably done at 30 d after breeding so as to account for later conception dates (9). However, no test is available to determine if bitches are pregnant during the first 3rd of gestation. Termination of pregnancy in midgestation is undesirable, since there will be a more prominent vaginal discharge and expulsion of fetuses. Cytologic stains of vaginal smears should be used as part of the initial examination to determine if mismated bitches are in estrus and if sperm cells are present, to prevent unnecessary abortive treatment. Behavioral estrus is often not well correlated with the bitch's fertile period: fertile matings may occur from 5 to 7 d prior to conception (9). A lack of detectable sperm cells does not rule out a mating, but their presence provides confirmation (10). A complete understanding of canine vaginal cytology is necessary for implementing this protocol, so that the onset of diestrus and hence day 1 of treatment may be determined accurately. A discussion between the veterinarian and the owner outlining all of the possible treatment options and side effects is warranted.

Fifteen bitches were presented to the Veterinary Teaching Hospital of the Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, Saskatchewan. The bitches represented a variety of breeds and ranged in weight from 5 to 45 kg. All owners had planned to breed their bitches later, but in each case there had been an escape and suspected mating. The animals underwent a general physical examination, and a vaginal cytologic examination with Wright's Giemsa stain (Hema-Tek Slide Stainer; Miles Scientific, Napierville, Illinois, USA) was performed to determine the stage of the estrous cycle. A 2nd vaginal smear was stained with a sperm stain (eosin-nigrosin) to look for spermatozoa (10). Bitches were considered treatment candidates if estrus was confirmed and sperm cells were detected in the vaginal smears. Serum progesterone levels were not measured, as they are not well correlated with behavioral estrus.

The owners were given the following options: 1) ovariohysterectomy; 2) allowing the bitch to whelp the litter; 3) waiting until pregnancy could be diagnosed and, if it was, terminating the pregnancy; and 4) terminating the pregnancy in early diestrus. All owners opted to terminate the possible pregnancy in early diestrus.

Vaginal cytologic examinations were repeated once daily to determine the 1st day of diestrus (1st appearance of intermediate and parabasal vaginal epithelial cells with or without neutrophils).

Treatment was started on day 6 of diestrus. At 0800, 10 μg/kg BW of a solution of bromocriptine (Parlodel; Novartis Pharmaceuticals, Dorval, Quebec) was given, PO. The solution was prepared by crushing and then dissolving a 2.5-mg tablet of bromocriptine in 100 mL of water, for a final concentration of 25 μg/mL. At 0845, 250 μg/kg BW of PGF_2α (Lutalyse; Pharmacia Animal Health, Orangeville, Ontario) was administered, SC. The animals were fed at 1030. Bromocriptine and PGF_2α were administered again at 1700 and 1745, respectively, in the same doses. This daily routine was repeated for a total of 5 d. The treatments were spaced 45 min apart, because PGF_2α injections often result in vomiting, especially following the initial 2 or 3 treatments. Episodes of vomiting and retching usually occurred within 20 min of injection and lasted for 5 to 10 min.

Blood samples were collected for determination of the serum progesterone concentration once per day, immediately before the morning treatment. Collection was continued for 2 d following completion of the treatment protocol, and the owners were instructed to return on day 28 to 30 of diestrus, at which time pregnancy was looked for by ultrasonography or abdominal palpation. The owners were contacted at approximately day 65 of diestrus to confirm the diagnosis of nonpregnancy.

After 6 treatments, all 15 dogs had serum progesterone concentrations of < 2.0 ng/mL (6.36 nmol/L) (Table 1). It has been stated that serum progesterone concentrations must be sustained at less than 2.0 ng/mL for at least 2 d to result in termination of pregnancy (1,8). Although it was impossible to determine the pregnancy status of the dogs before treatment, the luteal suppression observed, coupled with the absence of whelping, supports the effectiveness of this treatment protocol. It appears that starting the treatment protocol on day 6 of diestrus, rather than on day 5, and continuing for a full 5 d is important, as we experienced 100% success versus the 80% reported previously (7). Speculation that luteal sensitivity to PGF_2α begins around day 5 of diestrus, unless higher doses are used, is likely correct (7); however, the suggested maximal dose is 250 μg/kg BW, after which the side effects become too severe (7,8).

Table 1.

The side effects of PGF_2α are mainly due to the effects on smooth muscle and include panting, hypersalivation, vomiting, defecation, and urination (1,2,3,7,8). The appearance and severity of side effects is dose dependent (1,2,8) and usually transient, subsiding within 60 min of treatment (1). One may note only defecation at a dose of 110 μg/kg BW, whereas hypersalivation, vomiting, and urination are observed at 250 μg/kg BW. Higher doses have been associated with neurologic signs, such as ataxia and depression (8). The median lethal dose in the dog is 5130 μg/kg BW (1,8). Reportedly, atropine (500 mg/kg BW, IM) or chlorpromazine (0.25 mg/kg BW, IM) has been administered concurrently with PGF_2α to lessen the frequency and severity of side effects (1); however, we are unable to report on the effectiveness of these drugs. Bromocriptine and cabergoline have also been associated with vomiting (1,2,5). Interestingly, this has not been our experience using the formulation described. Further studies in which PGF_2α is used alone for 5 d beginning on day 6 of diestrus are warranted, as are studies using lower doses of PGF_2α in combination with bromocriptine.

Neither bromocriptine nor PGF_2α is approved for use in the dog; therefore, owner permission must be obtained. As a result of the need to administer the 2 agents at appropriate times to minimize side effects, it is prudent to admit the bitches daily for treatment. In fact, all bitches presented for pregnancy termination after roaming free and unsupervised while in estrus should undergo at least a vaginal cytologic examination. Separate vaginal smears should be stained with Wright's Giemsa and eosin–nigrosin to determine if the dogs are in estrus and whether sperm cells are present, respectively. Serum progesterone may also be assayed to confirm ovulation.

The drug costs associated with this protocol are negligible; however, the preparation of vaginal smears and daily hospitalization may be cost prohibitive for some owners. We also recommend follow-up ultrasonography at midgestation to confirm the nonpregnant status. If pregnancy is detected, termination may be attempted or the bitch be allowed to proceed to term. We advise that the total cost of treatment not be less than that of ovariohysterectomy, so that this, or any other, method of pregnancy prevention (termination) will not become an alternative to permanent sterilization. The protocol outlined in this article is currently the safest and most effective method for preventing pregnancy in early diestrus in our hands. CVJ