Effects of low-intensity pulsed ultrasound on the infrapatellar fat pad in knee osteoarthritis: a randomized, double blind, placebo-controlled trial.

Small single-centre trial: 26 knees at one orthopaedic clinic in Japan, 13 in each group, which is the minimum the authors' own power calculation required. The two groups were not matched for age. The ultrasound group averaged 63.5 years and the exercise-only group 56.5 years, a difference the authors report as statistically significant. Outcomes were measured only at the end of the 10th session. There was no follow-up, so nothing is known about whether any difference lasted after treatment stopped. The results text contradicts itself: one sentence says VAS, knee range of motion and Kujala score showed no significant difference between groups, while later sentences say all three improved more with ultrasound. The table marks them as significant, but the reporting is inconsistent. The authors admit they cannot say what the change in the patellar tendon/tibia angle actually reflects. They suggest it could mean reduced fat pad swelling or forward shift of the tibia, but the study cannot distinguish these. Everyone had grade II osteoarthritis, medial knee pain and enough function to exercise, so the findings say nothing about more severe disease or people who cannot exercise.

The authors declare no conflict of interest. The paper does not state who funded the trial or who supplied the ultrasound device used.

Do not read this as evidence that ultrasound works on its own, or that the change in the fat pad is why the pain improved. Everyone in both groups did the same daily strengthening exercises, so the ultrasound cannot be separated from them. The trial had only 26 knees at a single clinic, the two groups differed in age at the start, and nobody was assessed after the last session.