Abstract

This study aimed to perform a systematic reviewof the diagnostic test accuracy of physical examination tests, including crepitus of the knee, pain during functional activities, and manual tests of the patella in suspected patellofemoral osteoarthritis (PF OA) cases. The searched languages were English, Chinese, Korean, and Japanese. The PubMed, Cumulative Index to Nursing and Allied Health Literature,Web of Science, Korea Studies Information Service System, China National Knowledge Infrastructure, and Ichushi databases were searched electronically. The inclusion criteria of this systematic review were: (1) original articles; (2) prospective cohort or cross-sectional studies with isolated PF OA; (3) magnetic resonance imaging and/or orthopedic procedures used as the reference standard; and (4) odds ratio and/or test accuracy reported. The risk of bias was evaluated using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS)-2 tool. Seven citations were included in the systematic review. Overall, the risk of bias was favorable. Among the included studies, four reported the odds ratio of the physical examination tests, while three reported the sensitivity and specificity for test accuracy. The odds ratio of crepitus to identify cartilage lesions of the patella was between 1.74 and 5.49. Additionally, the odds ratio of pain during activities including walking and descending stairs was between 1.01 and 1.6 (original data = 0.60). The odds ratio of the manual tests of the patella was between 1.9 (original data = 0.52) and 2.7. Thus, crepitus was possibly the least efficient parameter for diagnosing PF OA among the three physical examination tests.