Canine Breeding Management and Artificial Insemination: Techniques and Caveats
World Small Animal Veterinary Association World Congress Proceedings, 2004
Patrick W. Concannon, MS, PhD, DACT (Hon)
Dept. Biomedical Science, College of Veterinary Medicine, Cornell University
Ithaca, NY, USA

An understanding of the processes and timing of ovulation, oocyte maturation, and period of peak fertility is critical to clinical management of natural breeding or artificial insemination using fresh, fresh-chilled or frozen-thawed semen. The follow aspects of breeding management require consideration.


An exam can reveal potential infectious processes, genital anomalies that might compromise mating or parturition, inheritable defects that ethically preclude breeding, or a serious medical condition. Brucella canis testing is an absolute requirement; testing of both bitch and stud can prevent transmission of the disease between facilities; failure to test the bitch can result in ethical and legal concerns. Canine brucellosis has been reported in the United States, Mexico, Japan, Germany, Eastern Europe, South America and Madagascar, and may be even more widespread. Abortion is a typical consequence of B. canis infection. Serology antibody tests may utilize antibodies that cross-react strongly with antibodies to several nonpathogenic bacterial species, and an immunofluorescent antibody test or rapid slide agglutination test can result in a false positive result. Initial positive results should be confirmed via improved, more specific serology tests and/or culture of B canis. Dogs can be infected with B. abortus from cattle and suffer effects seen with B. canis, including abortion. Herpes virus testing should be considered for large kennels with questionable hygiene. Testing pre-breeding is important, as it takes several weeks for inoculations to produce titers sufficient to protect the fetuses and newborn.


It is important to determine any previous failures to conceive, abnormal periods of proestrus or estrus, hormone administrations, potential mistiming of prior matings, and any prior suspected or confirmed abortions or resorptions, as well as vaccination and health history.

LH surge and ovulation

The ovulation-inducing surge in circulating luteinizing hormone (LH) occurs in late proestrus or early estrus, with LH levels increasing for 1-2 days and being elevated for 1-3 days. Best estimates for ovulation time are 48-60 hrs after a pronounced surge in LH and 36 to 48 hr after the LH peak. The surge in LH is concomitant with the initial rapid phase of the pre-ovulatory rise in progesterone that occurs in the dog.

Oocyte Maturation

Oocytes are ovulated as primary oocytes 2 days after the LH surge and do not undergo maturation to secondary oocytes and to a state where fertilization can be completed until at least 2 and more likely 3 days after ovulation, i.e., at 4 or 5 days after the LH surge.

Normal Fertility

Pregnancy rates in dogs of "normal" fertility appear to be higher than of any other domestic species. Some research and commercial colonies regularly enjoy pregnancy rates of greater than 95%. This relates to a prolonged period of potential fertility during the proestrus/estrus period. It also suggests that bitches which fail to become pregnant readily should be considered subfertile. However, there are often circumstances that do not favor success, including restriction to a single mating or breeding, scheduled transport of bitches long distances to a specific stud, and/or exposure to sub-fertile males, questionable ejaculates or processed semen.

Period of potential Fertility

The period during which a mating might result in pregnancy ranges from 5-6 days before ovulation until 7-8 days after ovulation--an apparent duration of 12 to 14 days. Although uncommon, pregnancies can result from single breedings or inseminations as early as 3 or 4 days before the LH surge (at 5-6 days before ovulation and 7-8 days before oocytes are fully mature). One factor involved is that the canine sperm can persist in the female tract for several days, up to 8 days in the extreme, and have in some cases been shown retain fertility for up to 5 or 6 days before ovulation. Pregnancy can sometimes also occur from a mating as late 9 or 10 days after the LH surge (at 7-8 days after ovulation and 4-6 days after oocyte maturation). Thus, some matured oocytes can remain capable of responding normally to fertilization after 5 or more additional days in the uterine tubes.

Period of peak fertility

In colonies of dogs with high fertility, pregnancy rates following single matings are unaltered with mating occurring as early as the day of the LH surge and as late as day 5 or 6, but fall off rapidly with earlier and later matings.

Thus, Day 0 to Day 5 (relative to the LH surge) is considered the period of peak fertility. Robust canine sperm routinely survive with high fertility in the female tract for the 2-day period between LH surge and ovulation.

Fertilization period

Matings which can result in rapid syngamy and zygote formation, without a reduction in fecundity or success rate, are limited to the period from Day 4 to Day 5 or 6 (after the LH surge). After that time, bitches can be fertile but the pregnancy rate declines presumably because matured oocytes then begin undergoing degeneration, and because cervical closure also occurs shortly thereafter, as early as Day 7.

Suboptimal fertility

In contrast to research colonies with high-fertility and high fecundity, the clinical situation often deals with bitches of unknown fertility, problem bitches which previously failed to conceive or maintain a pregnancy, and/or bitches for which there will be the constraint of limited access to stud dog, transportation issues, or insemination with questionable or compromised semen.


In most species fertilization refers to the penetration of mature secondary oocyte, and the rapidly ensuing events leading to zygosity including extrusion of second polar body, male pronucleus formation, and pronuclear fusion. In dogs, there is conflicting evidence as to whether or not sperm may penetrate immature oocytes in vivo just as they do in vitro. Interestingly, sperm penetration of the canine oocyte in vitro speeds up the process of oocyte maturation and formation of the female pronucleus. In contrast, evidence of in vivo sperm penetration of recently-ovulated immature oocytes was not observed in one study.

Periovulatory events

Serum estradiol from developing follicles increases throughout most of proestrus, with behavior transitioning slowly from anti-sexual to asexual, to teasing but non-receptive. Serum LH and FSH are low due to ovarian hormone negative feedback. Near the end of proestrus, there is a decline in serum estrogen and a surge in LH and FSH, for 1-2 days in the case of LH, and a simultaneous rise in serum progesterone. The initial surge in LH and rise in progesterone occur 2 days before the resulting ovulations, during and following which progesterone continues to increase. The transition from "non-receptive proestrus" to "receptive estrus" behavior occurs on average one day after the LH surge, but can range from 3 days before to 5 or more days after the LH surge, or nor occur despite normal endocrine changes. Vulval and vaginal edema and vaginal mucosal hyperplasia and cornification mirror the rise and fall in estrogen with variable lag periods.

Determination or estimation of the time of LH surge and of ovulations 2 days later

These can be estimated prospectively, in real time, or retrospectively based on the following relationships. In bitches with an obvious and rapid transition from proestrus to estrus behavior, it usually occurs on Day 1 (i.e., one day after the LH surge and one day before ovulation) with a normal range of ± 2 days. The vulval and vaginal edema and turgor caused by the proestrus rise in estrogen is diminished, and often rapidly, during the decline in estrogen that occurs immediately before and during the LH surge, and during ovulation. Exfoliate vaginal cytology reveals continuously increasing numbers of superficial cells until there is an absence of non-superficial cells and leukocytes during normal proestrus; this state is maintained at the maximum superficial cell index (SCI, % superficials) until after ovulation; ovulation typically occurs between 1 and 10 days after maximal SCI is first noted; and, an abrupt decrease in the SCI (metestrus or diestrus "shift") occurs at 4-9 days after ovulation, i.e., 6-11 days after the LH surge.

Generic or physiological proestrus and estrus

In breeding management programs it has become common to consider these stages of the cycle independent of the underlying behavioral definitions, and to refer to a physiological or endocrinological estrus as beginning at the time of the preovulatory LH surge and ending with the metestrus/diestrus shift in vaginal cytology. These, in fact, closely approximate the mean onset and termination times for estrus behavior in bitches


Some bitches may have a slow, intermittent transition from proestrus to estrus instead of a rapid and obvious transition, or they may display it to some males and not others, and in the extreme may not accept a male until late in what would otherwise be estrus, around 7-8 days after the LH surge, and often too late for a fertile mating.

Vaginal cytology

The collection as well as the handling and staining of smears should be standardized, and begin early; the first 1 or 2 smears may be confusing because of accumulated cellular and other debris. Progressive increases in cellularity and SCI confirm a normal proestrus and a rise in estrogen. Without special staining there are no cytological changes that are indicative of ovulation. In late estrus, the abrupt decline in SCI occurs on average 8 days after the LH surge but varies from Day6 to 11, and by convention is referred to as the "end-of-estrus", "metestrus", or "diestrus" smear, and the day is often designated as Diestrus-1 regardless of the behavioral status of the bitch. The techniques and stains have been reviewed elsewhere1,2. Leukocytes and bacteria in vaginal smears are often normal findings, especially in proestrus.

Vulval edema

Monitoring of this is easy and should be part of every breeding management. Gentle palpation can confirm a progressive increase in size and turgor of the vulva and perineum throughout most of proestrus, as well as the expected decrease in turgor that is first encountered at or shortly before ovulation and is typically very obvious after ovulation. The vulva remains enlarged albeit softened or flaccid well beyond the end of estrus and the fertile period.


Changes in the gross morphology of the vaginal mucosa can be observed and appreciated using either direct viewing via pediatric proctoscope or macroscopic viewing via fiber optic or digital endoscope. During the proestrus rise in estrogen the longitudinal mucosal folds become progressively enlarged, rounded and turgid and the color transitions from pink to white due to thickening and cornification of the surface epithelium. Segmentation of the mucosal folds also appears more pronounced, with the surface of each segment being turgid and smooth. The periovulatory decline in estrogen and rise in progesterone is accompanied by increasingly pronounced wrinkling of all mucosal surface; this crenation begins 1-3 days before ovulation. It is grossly obvious by the time of ovulation or one day later. It becomes more pronounced for 1-2 days, and then less so as the size of the folds diminishes and the surface epithelium is no longer uniformly covered by a thick layer of superficial cells and becomes mottled red-white in early metestrus.

False proestrus or false estrus

A normal periovulatory-like rise and subsequent decline in plasma estrogen can occur without ovulation, with or without overt estrus behavior, with normal-like changes in vulval and vaginal morphology and edema, and with normal or nearly normal changes in vaginal cytology including an 80-100% SCI and a subsequent metestrus smear. The cause is a failure of surge release of LH and FSH or an inadequate release, and it can be accompanied by a small but transient increase in plasma progesterone or there may be no rise in plasma progesterone.

Progesterone assays

The assay of serum or plasma progesterone every 1-3 days has become the gold standard for confirming ovulation and estimating the time of its occurrence, sampling frequency varying with the status of the bitch or the degree of accuracy required. Daily measurement by accurate and quantitative assays beginning in mid-late proestrus can detect the initial abrupt rise in progesterone and thus determine the day of the LH surge with an accuracy of ± 0.5 to ± 1 day depending on the bitch and the sensitivity and accuracy of the assay. On the day of the LH surge, serum progesterone increases abruptly, i.e., at a rate greater than over the previous 2-3 days. The increase may be from 0.4 to 0.8 ng/ml, from 0.7 to 1.2 ng/ml, or from 0.8 to 2, 3 or more ng/ml. Neither the magnitude of the increase, nor level attained, is predicable, consistent or diagnostic. However, an increase in progesterone of some magnitude occurs consistently, and is almost always followed by a further increase on one or both of the next two days. With such assays the initial increase in progesterone, if confirmed by a subsequent increase, should be considered to have occurred on the day of the ovulatory LH surge, and a single critical insemination scheduled for 4 to 5 days later. Possible breed differences, or differences in "problem" bitches, have not been subjected to detailed study. Some clinics have also been successful inseminating 1-2 days after progesterone is determined to have reached 5 ng/ml (mean progesterone at 1-2 days after ovulation), whereas others wait until 10 ng/ml is reached (mean progesterone at 3-4 days after ovulation).

Service laboratories and quantitative progesterone, LH and FSH assays

Many service laboratories can now provide results for accurate progesterone assay by RIA or CIA in 24 h or less. There is also the possibility of service laboratories providing FSH assay (by ELISA) and LH (by RIA) using commercial reagents (BiocodeHycel, Belgium). LH assay would require daily measurement, as the surge may only last one day. The FSH assay could be more useful since the FSH surge typically begins coincident with, but lasts one day longer than, the LH surge.

Qualitative or semi-quantitaive assays

In-office semi-quantitative ELISA assay kits for progesterone and LH are available. Anecdotally, the LH kits have yielded success in for some clinics and not others. The progesterone kits are usually reliable, but anecdotally some may vary control problems. Their level of sensitivity for the detection of a fist-rise in progesterone varies, and often they do not detect increases until progesterone is above 2 ng/ml. The initially detected rise may be actually at 0, 1, or 2 days after the LH surge, and such variation must be appreciated and incorporated in planned breedings. In such cases, it is critical to observe the bitch for cytological or vaginoscopic evidence of impending metestrus/diestrus and inseminate or breed immediately if such changes are noted.

Progesterone results vary with sample handling

Even with accurate RIA and CIA, the results can depend on the sample (serum versus plasma), the anticoagulant if plasma, and the durations held at room and/or refrigerator temperature before separation and freezing. Thus, these should be kept consistent. One study found that that quantitative retention of assayable progesterone in serum requires clotting for 2 or more hrs at room temperature and then separation immediately or after refrigeration storage.


Ovulation can be detected in real-time using ultrasound. Previously-obvious anechoic antral follicles rapidly appear much more echogenic and less distinct. Whether the timing of the change is consistent in relation to the actual time of ovulation among bitches has not been determined. It has been reported to be more accurate than the use of progesterone assays, but considerable equipment expense, training, and experience are required to utilize the technique. Whether the change in appearance involves a change in follicular fluid echogenicity, collapse of follicles, bleeding into follicles, or rapid proliferation of luteal tissue into follicles is not known. However, this echogenic appearance is only transient as the corpora lutea rapidly cavitate and again appear to be fluid filled and have an anechoic center 24 to 72 h later.

Semen deposition

Natural mating. With most bitches and males of reasonable fertility, mating anytime between 2 and 5 days after the LH surge (1-4 d post ovulation) should be satisfactory, especially if there is a repeat mating 1-2 days later. With problem bitches, many specialists will use intrauterine deposition for at least one service, to confirm both semen quality and deposition, preferably at 4-5 days post-LH. AI of fresh or chilled semen. Many cases can be handled with vaginal deposition, but with problem bitches, or lower quality semen, intrauterine deposition at 4-5 days post-LH may be preferred. Frozen-thawed semen. Sperm are sufficiently compromised that intra-uterine insemination is considered necessary and the timing critical, optimally at 5 days after the LH surge.

Insemination devices and techniques

Rigid, stainless steel, "Scandinavian" canine-insemination pipettes operated without visualization of the cervix and by trans-abdominally palpating, stabilizing, and maneuvering the cervix are used routinely by several academic and specialty clinics with staff trained and experienced in the technique. They are sized for small, medium and large dogs. Requiring less extensive training but more expensive equipment is the technique of catheterization under direct visualization of the cervical os using fiber optic or digital macro endoscopy. This has been done successfully and reasonably routinely using rigid devices, semi-rigid urinary catheters or balloon-ended angio-catheters. However, the time involved varies greatly among cases, ranging from 5 minutes to 35 minutes. It can become tedious if the bitch is not tractable or if the catheter, scope, tract, or bitch cannot be maneuvered so as to allow the catheter tip to be "turned up" and thus inserted into and passed up the cervical canal which is usually oriented with the cervical os facing ventrally. Regardless of method, volumes should be minimized by using concentrated or centrifuged semen samples to reduce loss from cervical backflow. Because of the experience required and potential tedium involved, some clinics prefer to utilize a surgical exposure and insemination by direct injection through the uterine wall, and do so on a routine basis when dealing with problem bitches, questionable semen or frozen-thawed semen. For deep anterior vaginal insemination of fresh or fresh-chilled semen, many facilities use the "pistolet osiris" canine insemination device with an inflatable "bulbus" which causes retention of the semen in the anterior vagina and presumably facilitates semen movement through the cervix.

Surgical insemination

For many this raises a question of veterinary ethics. Further, it cannot be repeated the next day vs. transcervical cannulation which can. Advocates note that the procedure can be accomplished rapidly via a small incision with minimal trauma to the patient, and can be more cost-effective in terms using and scheduling a predictable amount of clinician's time.

Late estrus intra-uterine fresh-semen ai

Uterine insemination can result in higher pregnancy rates (30-60%) than natural mating or vaginal insemination (0-15%) when the bitch is 7-10 days after the LH surge, immediately before and shortly after the metestrus/diestrus shift in cytology and cervical closure has occurred.

Additional considerations

Client education, counseling, prophylactic antibiotics, and use of estrus-induction protocols are other potential services offered by specialty practices in breeding management of problem bitches.


Other references are available upon request

1.  Concannon P.W., England G., Verstegen J. and Linde-Forsberg C. (Eds.) Recent Advances in Small Animal Reproduction. International Veterinary Information Service, Ithaca NY (, 2003; Document A1203.0703

2.  Concannon, P, England, G, Farstad, W, Linde-Forsberg, C, Verstegen, J, and Doberska, C. Advances In Reproduction In Dogs, Cats And Exotic Carnivores. J. Reprod. Fertil. 57 (Suppl.), Cambridge, 2001.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Patrick W. Concannon, MS, PhD, DACT (Hon)
Dept. Biomedical Science, College of Veterinary Medicine
Cornell University
Ithaca, NY

MAIN : Reproduction : Artificial Insemination
Powered By VIN