A recent study led by Miho J. Tanaka, MD, PhD, director of Women’s Sports Medicine at Massachusetts General Hospital and associate professor of orthopedic surgery at Harvard Medical School, has identified key MRI measurements that can help predict the risk of patellar instability.
Patellar instability is a complex condition influenced by several anatomical factors, and consistent measurement has long been a challenge for clinicians. Dr. Tanaka and her team set out to clarify which MRI-based indicators are most predictive.
“When assessing morphologic risk factors for patellar instability—including patella alta, trochlear dysplasia, or malalignment—it is important that we are consistent about the measurements we are using,” said Dr. Tanaka.
Key Findings
The study analyzed MRI scans from 128 patients with confirmed patellar instability and compared them to 128 age- and sex-matched controls. Using both traditional regression methods and advanced machine learning models, the researchers identified the most important predictors of patellar instability:
-
Insall-Salvati ratio (a measure of patella alta)
-
Tibial tubercle–trochlear groove (TT-TG) distance
-
Trochlear depth
Among various models tested, the random forest machine learning model achieved the highest performance, with strong accuracy in identifying at-risk patients.
Why This Matters
Traditionally, many clinicians have relied on the Caton-Deschamps index to characterize patella alta. However, the findings from this study suggest that the Insall-Salvati ratio may be a more predictive measure.
“Perhaps this should influence the measurements that we choose to report in our studies, so that we can be consistent in using the measurements that we now know to be the most predictive,” Dr. Tanaka explained.
Looking Ahead
Dr. Tanaka emphasized the promise of machine learning in advancing patellofemoral research:
“The ability to process large amounts of information using machine learning algorithms is going to be a great application to the study of patellofemoral instability. The more we are able to build out these algorithms, the more precisely we will be able to predict or identify patients who are at risk for failure.”
This research highlights the importance of both consistent measurement practices and new technologies in improving outcomes for patients with patellar instability.