Study: Radiograph Interpretation Not So Black-And-White - Horse Racing News | Paulick Report

Study: Radiograph Interpretation Not So Black-And-White

It's no surprise to anyone who has bought or sold a horse at public auction that a veterinarian's interpretation of a sale entry's radiographs can make or break the bank. A consignor's veterinarian may have one interpretation of a set of x-rays, while a buyer's expert may see things differently.

But just how common is this disagreement?

Researchers in Australia and the U.S. attempted to answer this question in a study published in the Equine Veterinary Journal late last year. Sale radiographs for 167 yearlings were read twice by four veterinarians specializing in radiology, and the experts were instructed to look for the presence or absence of common issues. This yielded 1,336 total readings of x-rays (167 sets x four experts x two readings), and researchers found that the majority of the time, veterinarians in the study who believed an x-ray had no significant abnormalities drew the same conclusion upon reading the x-rays a second time, and tended to agree with each other about the absence of issues. (The lone exception to this seemed to be on the topic of linear defects in front and hind proximal sesamoid bones, which the researchers considered to be due to these issues occurring commonly in yearlings.)

When it came to identifying an issue however, the group was less consistent. Large, obvious findings like cyst lesions in the knee were generally agreed upon. Other issues, like sagittal ridge defects in front ankles (a somewhat subtle problem) produced particularly consistent agreement. There was little consensus on issues such as lucency in sesamoid bones or the presence of osteophytes in the knee/hock.

Other areas, like upright front feet, irregular vascular channels, and certain types of osteophytes, saw the experts agreeing with each other less than half the time; some disagreed on the presence of findings as much 60-70 percent of the time for certain defects.

The study also determined that experts tended to be more consistent with their own previous judgments of a set of x-rays (which were not made identifiable to the participants in the study) than with each other.

The study did not explore reasons behind the differences in reporting, although researchers believe experience is not likely a factor, since specialists were used. Instead, Dr. Melissa Jackson of Melbourne Polytechnic thinks the results could point to individual reading styles of the radiologists and the standardization or criteria for reporting findings. Further research is required to explore the reasons for these differences, but Jackson isn't convinced that the quality of the radiographs is to blame.

“Previous studies have shown that the quality of radiographic examinations can affect reporting agreement; however provision of good quality radiographs whilst improving reporting agreement still resulted in overall agreement levels that were unacceptable,” she said.

What's the takeaway for sellers? Ensure that the veterinarian taking images for the repository is providing the best images possible, Jackson said. And, sellers and buyers alike should be aware that a difference in interpretation of pre-sale x-rays is just that.

There's no absolute right or wrong in this endeavor, of course. Jackson said the best defense is to hire a veterinarian who is up on their reading.

“Buying a horse is risky – buyers can reduce some of that risk by undertaking a pre-purchase exam, including radiographic examination, but it does not eliminate the risk, and buyers need to consider what risk they are prepared to take,” she said.

“Veterinarians examining repository radiographs should consider the published evidence, such as the studies by Kane et al (Equine Veterinary Journal, 2003), and Spike-Pierce and Bramlage (Equine Veterinary Journal, 2003). These studies have aimed to examine association between the presence of orthopedic findings and race performance.

“While some radiographic findings have been associated with reduced performance, for many findings the occurrence in the yearling population is very low and as such it has not been possible to assess the effect that they have on future performance.”

Paulick Report Icon

Receive daily headlines, breaking news alerts, promotions, and much more!

Become An Insider

Support our journalism and access bonus content on our Patreon stream

Learn More