Researchers are hopeful that a different way to read knee X-rays could one day make it easier to predict osteoarthritis (OA).
Osteoarthritis, or degenerative arthritis, is the most common type of arthritis, and it is characterized by the breakdown and loss of cartilage, the tissue that cushions the bones of the joints.
Beyond pain relief, little can be done once you have the condition takes hold, but doctors say there are things that may help stave it off, including losing weight, exercising and taking precautions to avoid injury.
But now, in a new study, Lior Shamir, PhD, and a group of scientists at the National Institutes on Aging at the National Institutes of Health in Baltimore, say they’ve created a computer program that can analyze X-rays of knees and find perhaps the earliest warning signs of osteoarthritis.
Researchers digitized 200 X-rays taken during the mid-1980s of patients who had normal knees at the time but developed osteoarthritis 20 years later. The scientists analyzed pixels in the scanned knee images and built a computer algorithm from that information that measures the fine textures of the bone and subtle changes in the pixel intensity to determine who will develop osteoarthritis.
“The algorithm doesn’t look at what radiologists look at. It doesn’t look at OA or the narrowing of the space between the joints. What it does look at is the structure of the bone. The alteration of the bone tissues compared to healthy bone,” Shamir explains.
The study, published online in April 2009 in the journal Osteoarthritis and Cartilage, found that the algorithm was accurate 72 percent of the time and could make the prediction as many as 20 years before symptoms were reported.
Shamir says the discovery is important for the basic science of studying osteoarthritis since it is one of the most prevalent diseases in the industrialized world with an estimated 80 percent of people over the age of 65 suffering from it, but little is known about this painful condition.
Shamir says this finding is also significant because it shows that there are scientific indications of osteoarthritis long before patients feel pain or are diagnosed with the condition. “Practically, if we can diagnose OA before it becomes symptomatic, at a very early stage it can be prevented or at least delayed by physical therapy, nutrition and especially weight control. All of that can significantly delay the onset of the disease.”
Dr. Shamir says the there is a high margin of error with the algorithm, but he thinks scientists can decrease that over time.
David Felson, MD, MPH, a professor of medicine and epidemiology at Boston University School of Medicine says the study needs to be replicated in another sample before it is shown to work because many of these kinds of preliminary findings are not successfully replicated.
Even if the method is validated, Dr. Felson says he isn’t sure how valuable such early knowledge would be for patients or clinicians. In a study published in 2008 in The New England Journal of Medicine, Dr. Felson proved that as many as half of people who have osteoarthritis that is visible on X-rays don’t feel pain and probably never will.
“Knowing who gets radiographic osteoarthritis has little relation to knowing who has pain and discomfort from disease,” Dr. Felson says.
“Ultimately, I don’t care much to know who gets radiographic disease,” Dr. Felson says, adding that people who are told by a doctor that they have osteoarthritis that’s visible on an X-ray, but who don’t feel any pain or symptoms, shouldn’t feel that they need to do anything about it. “They are not necessarily at risk for pain or disability,” he says.
But Shamir disagrees, arguing that even when patients feel no pain despite radiographic evidence of osteoarthrits, it is still an indication that the disease is developing and the cartilage is degenerating. “So you might not feel the pain immediately, but it’s not a good sign,” he says.
“If we can identify those individuals who have the process already started or accelerated faster than others, than these individuals should be more careful about taking actions like weight control,” Shamir says.