Shining a light on pregnancy loss risk factors

In January, TTR spoke to Equine Specialist Consulting’s Dr Joan Carrick about the circumstances and causes of equine pregnancy loss, a subject Dr Carrick is going in-depth on as part of a multi-stage project funded by the AgriFutures Thoroughbred Horses Program.

Phase Three of the project has now concluded and on Wednesday, July 29, Dr Carrick presented some of the project’s insights to the industry at the AgriFutures Thoroughbred Horses Program Research Update, hosted at the Hunter Thoroughbred Breeders’ Association headquarters.

What the project has uncovered so far has been illuminating, but it also highlights the need for further investigation on the subject.

Identifying risk factors

Phase Three of the project revolved around a case-controlled study into mid to late term pregnancy loss in New South Wales and Queensland. Data from 1487 mares in the Hunter Valley and Queensland who were pregnant in the 2024/25 breeding season was collected and analysed to look for risk factors that could be leading to pregnancy loss.

“A big part of this project is that we wanted to identify risk factors for loss that we could then provide breeders with identifiable actions they could take to reduce pregnancy losses,” Dr Carrick said. The incidence of losses in Queensland was higher than in New South Wales, which confirmed the need for region-specific management procedures to reduce losses in future.

“It was really important to us that part of this project takes place in Queensland, where the industry is very different to how it’s set up in New South Wales. It’s much more extensive and their margins are slimmer.

“It’s important to not ignore those breeders or assume their challenges will be the same as the Hunter Valley. Improving successful foaling rates improves productivity, and therefore profitability, at every farm, but it must be recognised that there are regional differences.”

The primary mare factors that correlated with pregnancy loss were age, number of pregnancies, and number of previous losses, but as Dr Carrick pointed out, these factors are all interconnected. The older a mare gets, the more pregnancies she is likely to have and therefore the more she has had to lose.

The difference in average age between mares with healthy foals and mares who lost pregnancies was also small, and Dr Carrick found that the number of losses varied from farm to farm, year to year, in each region - even when farms were relatively close together.

“That’s something that we have to keep monitoring, and if we start to see the averages rising, that is something the industry will need to address,” she said.

“This isn’t to say that any one factor is a definite reason to remove a mare from your herd. This study wasn’t designed to separate out those three factors either to find out if one or the other is more crucial, that needs a different designed study.

“But it is worth considering all three when assessing the composition of your herds.”

When and why do losses occur?

A particularly important part of the project mapped out when losses were most likely to occur. In both the Hunter Valley and Queensland, more than half of losses occurred before 270 days, with the biggest spike in losses - both physically observed and identified by scan - happening between 150 and 270 days of gestation.

The challenging information that came with these discoveries was in whether or not losses were investigated. In observed losses occurring between 150 and 270 days, 75% of fetal matter was both examined and sampled for further analysis, and between 270 and 320 days, that number moved to 80%.

However, the statistic fell dramatically when losses occurred past the 320-day mark in gestation. During the study, all losses after 320 days underwent some kind of examination, but in only 15% of cases were samples taken. Dr Carrick stressed that it’s vital to fully investigate losses at all stages of pregnancy and remain vigilant for communicable diseases.

“It is really encouraging to see that losses tend to be fully investigated when they are observed and occur between 150 and 320 days of gestation,” she said. “But this was not the case at other points of gestation.

“This study reinforces the findings in the previous study. Virtually none of the losses that took place once the mare was in the foaling unit were fully investigated. In nearly every case, a vet examined the mare, the foal, and the placenta, but in very few did they take samples.”

Dr Carrick highlighted that this limits our understanding of what causes pregnancy loss and also limits what we can do to improve the odds next time.

“The feedback I receive is that it’s too expensive, which I can understand, but the next pregnancy will be worth much more than what you spend on the post-mortem,” she said. “The findings from that post-mortem may be vital to increasing your chances of a successful pregnancy next time.”

Dr Carrick’s suggestion was to test every loss for abortive diseases like Equine Herpes Virus and Chlamydia psittaci as a bare minimum, especially when they occur in the critical end stage of pregnancy. Understanding the causes is both part of mitigating future losses and looking after your staff.

“Many of these diseases, like Chlamydia psittaci, leptospirosis, and Q fever, can not only cause problems in your foaling unit, but they can pose a risk to your staff,” she said. “It is really important to consider it from a workplace health and safety point of view as well.”

Proper investigation leads to prevention

A full investigation of each loss incorporates a proper examination of the mare and the placenta.

“If you see ascending placentitis, the infection has entered through the mare’s cervix and that means you need to take a closer look at her conformation,” Dr Carrick said. “Does the mare have conformational issues or damage that predispose her to disease? Does she need a Caslick procedure, or does an existing Caslick require revision?

“All of these factors need to be considered in the mare. If a mare has underlying physical issues that can't be corrected, she may require closer monitoring during late gestation.”

The project also examined the role of management practices in reducing pregnancy loss and found that interventions are most effective when they are applied to the mares that genuinely need them. Unnecessary treatments and monitoring can add cost and labour without delivering meaningful benefits.

“For example, mares that had undergone a Caslick procedure were no more likely to lose a pregnancy than mares without a Caslick,” she said. “We aren't challenging the existing research on the benefit of Caslick procedures if a mare requires one. What our findings suggest is that the reverse may also be true: if a mare doesn't need a Caslick, there is little benefit in performing one.

“The same principle applies to other management practices. The mares most likely to benefit are those with a clear clinical indication for that particular intervention.”

The same applies to mares under additional monitoring, and the use of antibiotics and pregnancy maintenance treatments like Regumate.

“We don’t need to apply blanket treatments, we need to identify high risk mares and target our interventions accordingly,” Dr Carrick said. “Developing better ways to identify the mares that are most likely to benefit from each intervention is an area where further research could deliver real value for the industry.”

Practical management techniques

The research showed that paddock management was one area where improvements appeared to benefit all mares.

“Some of the environmental management research is quite exciting, and it identifies practical changes breeders can make on farm to help improve pregnancy outcomes,” Dr Carrick said. “The prior project in 2021 highlighted a lot of this information and some of it was not what you would expect, so to have the exact same data be important in the second study really strengthens it. These are consistent and important findings.”

Paddocks where mares suffered less losses tended to be irrigated, have shade structures, and fenced water sources. If they had trees, these were fenced off and maintained, and paddocks were separated by laneways from neighbouring paddocks.

One interesting finding is that paddocks beside roads, with either double fencing or tree lines separating them from the road, were also less likely to have pregnancy losses.

“Obviously, these techniques aren’t going to prevent all abortions, but the data suggests that paddocks that are more managed are where mares lose less pregnancies.”

Many of these land management techniques are already in place on stud farms across the country, but the data now gives these techniques a scientific backing.

The next steps

The conclusions of Phase Three made several recommendations for further study. Among them was the need for further, deeper research into Queensland’s regional differences, the mechanism behind an association between Salmonella vaccination and reduced incidences of pregnancy loss, and what specific factors affect losses at various stages of gestation.

But Dr Carrick would also like to see the industry take the valuable information already uncovered and work together to better understand pregnancy losses.

“I am very proud of what we have achieved with this project, and we have gained a lot of very important information for the industry,” she said. “The overarching aim of this now is to get the industry to value this information. There are future pregnancy losses we can prevent if we are investigating each loss properly.”

Phase Three’s findings back up what Phase One and Two of the project concluded; the benefit of a full investigation into every loss extends beyond the individual mare owner. The latter phases had determined that an effective regional surveillance system investigating and communicating causes for pregnancy losses could gain broodmare owners in the Hunter Valley between $850,000 and $1.7 million annually.

“Knowing what caused a loss benefits more than just the individual mare owner, it benefits every mare in that paddock and every farm in that region,” Dr Carrick said. “It might seem like more money upfront, but preventative systems will save the industry both time and money in the long run.”

What happens next is up to the industry itself.