PHILADELPHIA - Drawing upon an international database of some 16,000 dogs, researchers at the University of Pennsylvania have pinpointed what believed to be the first solid predictor, in any species, of future arthritis. The scientists have found that laxity in the hip joint - several millimetersworth of excessive play between the ball of the femur and the hip socket - correlates strongly with the advent of hip arthritis later in a dog life.

"The relationship between hip laxity and arthritis in dogs is akin to the relationship between high cholesterol and heart disease in humans," said lead author Gail K. Smith, professor of orthopedic surgery and chair of the Philadelphia Department of Clinical Studies at Penn School of Veterinary Medicine. "Hip laxity is no guarantee of arthritis later in life, but it is a very solid risk factor."

The finding, reported in the Dec. 15 issue of the Journal of the American Veterinary Medical Association, could lead to new ways of averting or minimizing the occurrence of canine arthritis, which afflicts an estimated 70 to 80 percent of dogs in certain breeds. Since a canine generation is just 30 to 36 months, Smith said selective breeding to avoid high-laxity individuals could slash the incidence of canine arthritis within 10 years.

Smith, who began collecting data on arthritis in dogs in 1983, says the physiological similarities between dogs and humans make it very likely that joint laxity could similarly signal the likelihood of arthritis in people, whose laxity could be remedied in humans with medications. There is currently no such risk factor used to predict the onset of arthritis among hu-mans.

In dogs, several larger breeds are most prone to arthritis: golden retrievers, Labrador retrievers, German shepherds, Newfoundlands and St. Bernards. Conversely, certain breeds that have long been bred for speed or athletic prowess, such as performance borzois and racing greyhounds, almost never develop arthritis.

"This research gives dog breeders an additional tool they may use in their efforts to decrease the incidence of hip dysplasia," said Mary B. Mahaffey of the University of Georgia College of Veterinary Medicine, who was not involved in the work. "The authors give good recommendations for breeders striving to decrease the incidence of hip dysplasia in their kennels, and should allow breeders to make reasonably good progress in reducing the incidence of hip dysplasia."

As in humans, canine arthritis becomes more symptomatic with age. More than half of 2-year-old golden retrievers show radiographic signs of degenerative joint disease, and more than 90 percent of susceptible dogs show signs by old age. Treatment generally involves painkillers. Although some dog owners elect to perform hip replacements on severely affected pets, the expense of this surgery causes others to elect euthanasia.

The current study grew out of Smith development of a now-licensed system called the Penn Hip Improvement Program, or PennHIP. Some 1,400 veterinarians worldwide have been trained to use PennHIP to measure hip laxity among dogs; it from these clinicians that Smith gathered data on the 15,742 dogs included in the JAVMA paper.

PennHIP positions dogs differently than traditional radiography of the hip, which images dogs with rear legs extended. With PennHIP, the veterinarian takes one image of a sedated dog hip in the conventional position. Then, with the hips in a more neutral position, this image is supplemented with two others: one with the femoral head pushed in toward the hip socket and one with it pulled away from the socket. Comparing the latter images lets clinicians determine how many millimeters of play exist between femur and socket.

Penn remains the central repository for images collected using PennHIP, allowing for population studies far beyond the fewer than 100 animals involved in most veterinary studies. The number of dogs profiled in the database is growing by roughly 3,000 a year.

Smith co-authors on the JAVMA paper include Philipp D. Mayhew, Amy S. Kapatkin, Frances S. Shofer and Thomas P. Gregor, all of the Philadelphia Department of Clinical Studies at the Penn School of Veterinary Medicine. Their work was partially supported by the Synbiotics Corporation, which has licensed PennHIP.