LAS VEGAS — There was inaccurate interobserver reliability to support use of visual assessment alone to determine the presence and degree of patellar maltracking in patients with patellar instability, according to a presenter at the Arthroscopy Association of North American and American Orthopaedic Society for Sports Medicine Specialty Day at the American Academy of Orthopaedic Surgeons Annual Meeting.
“Overall, we showed inaccurate interobserver [reliability] to support the use of visual assessment in determining the presence or degree of patellar maltracking, and surgical experience does not correlate with the ability to determine maltracking,” Matthew Best, MD, said during his presentation. “These findings highlight the importance of objective radiographic measurements.”
Best and colleagues identified patients with patellar instability and video recorded active knee extension. They also obtained dynamic CT scans. Of 56 eligible practicing orthopedic surgeons from the International Patellofemoral Study Group, 32 members completed a survey to determine the presence or absence of patellar maltracking and also graded the J-sign. They used the maximum amount of lateral translation of the patella at any point during the knee flexion-extension cycle to determine the grade. Grade 0 was classified as up to one quadrant of lateral translation. Grade 1 was greater than or equal to two quadrants but less than three quadrants. Grade 2 was greater than or equal to two quadrants less than three quadrants. Grade 3 was greater than or equal to three quadrants.
During the qualitative analysis, 10 single knee videos were viewed by respondents and they determined whether the patella was tracking normal or abnormal. The J-sign for each patient was then rated during the quantitative analysis in which the respondents viewed 20 different single knee videos.
Results from the qualitative analysis demonstrated respondents were correct 68% of the time about determining the tracking in videos. Best noted the free-marginal kappa indicated that interobserver reliability was inaccurate. Respondents were able to correctly identify grade 3 53% of the time, grade 1 and 2 about half of the time and grade 0 68% of the time. The free-marginal kappa indicated that interobserver reliability was inaccurate.
No significant difference was seen between 6 to 10 years, 11 to 15 years and more than 15 years of surgeons’ practice experience and more accurate ability to identify maltracking.
“There was a small improvement in the ability to determine the grade on the videos, but this was not statically significant,” Best said. – by Monica Jaramillo
Best M. Accuracy and reliability of the visual assessment of patellar tracking. Presented at: Arthroscopy Association of North American and American Orthopaedic Society for Sports Medicine Specialty Day at the American Academy of Orthopaedic Surgeons Annual Meeting; March 16, 2019; Las Vegas.
Disclosure: Best reports no relevant financial disclosures.