Why No News is Good News in Imaging Knee OA

The results of two new studies in osteoarthritis (OA) imaging may come as a surprise. In terms of both cartilage thickness and bone marrow lesions, the best news from progressive imaging of osteoarthritic knees is not a reduction in signal, but a lack of change.

The best indicator of  non-progression in knee OA is not the thickness of knee cartilage, but rather a relative lack of activity in the joint over time, according to research at the University of Copenhagen described at the Osteoarthritis Research Society International meeting in Philadelphia. The study used publicly available imaging data from two substudies of the Osteoarthritis Initiative (OAI) and a separate set of 267 images of osteoarthritic knees from a Copenhagen population.

Researchers found that a measure of change in the activity of cartilage between baseline and later readings (one year for the OAI cohorts and 21 months for the Copenhagen subjects) was significantly associated with changes in Kellgren & Lawrence (K&L) scores from 0 to 3. Simple changes in cartilage thickness or joint narrowing did not correspond significantly with K&L scores in these populations.

"From our point of view, this is a novel, simple efficacy marker," said Erik Dam PhD, an imaging expert with the contract research firm Biomediq, which collaborated in the study.

There's no need to examine subregions, he added: It's sufficient to assess changes across the entire joint. But analyzing data from subregions in patients with pre-osteoarthritis of the knee (defined as K&L measures between 0 and 1) in the OAI cohorts, the team found that all but one of seven activity measures correlated with progression at the p < 0.001 level or better, while only 3 of the thickness measures were significantly associated with progression, and only at p < 0.05. The activity measure also correlated significantly with later knee replacement, while thickness and joint narrowing did not.

Thickening and joint space narrowing are ambiguous, Dam pointed out, because changes on imaging may represent growth or may reflect hypertrophy or swelling. "Think what your choice of cartilage efficacy marker is saying," Dam said. "If you're looking at changes in thickness, your marker is saying that thinning is bad and thickening is good. If you look at activity, change is bad. Homeostasis is good."

Separate analysis of OAI data by a multicenter team from the US and Australia yielded new information about the significance of bone marrow lesions seen on imaging that appear to correlate with cartilage loss and progression. The exact relationship has been unclear, said Jeffrey Driban MD, assistant professor of rheumatology at Tufts University Medical Center. From the results he reported, it appears that any change in the density of bone marrow lesions, including regression of signal, is probably a bad sign.

Analyzing weight-bearing knee radiographs, MRI images, and  dual energy x-ray absorptiometry (DXA) results for 404 subjects in the OAI, the research team found that knees with either regression or progression in periarticular bone marrow density were more likely to have sclerotic progression in the medial tibia compartment, defined as either joint space narrowing or cartilage thickness changes, than those that have no change in lesion density.