Stratifying Minimal Versus Severe Pain in Knee Osteoarthritis Using a Musculoskeletal Ultrasound Protocol

作者：Mary Monica Bernardo‐Bueno, Consuelo Gonzalez-Suarez, Andrea Kristina Malvar, Ronald Cua, Daniel Feliciano, Bee Giok Tan‐Sales, Svetlana Maris O. Aycardo, Millicent Tan‐Ong, R. Chan, Francisco de los Reyes · 发表于：Journal of Ultrasound in Medicine · 年份：2018 · DOI：10.1002/jum.14819 · 被引用次数：15 · 研究领域：Musculoskeletal synovial abnormalities and treatments、Osteoarthritis Treatment and Mechanisms、Tendon Structure and Treatment

OBJECTIVE: The aim of this cross-sectional correlational study was to determine the association of pain with morphologic and inflammatory sonographic findings in patients with knee osteoarthritis. METHODS: A total of 113 participants with knee osteoarthritis were assessed using visual analog scale pain score and sonography. Ultrasound evaluation included morphologic changes (ie, articular cartilage degeneration, medial and lateral meniscal protrusion, and presence of osteophytes on the joint margins) and inflammatory changes (ie suprapatellar effusion and/or synovitis, Baker cyst, superficial and deep infrapatellar effusion, pes anserine tendinopathy, and Hoffa panniculitis). RESULTS: Cluster analysis via Ward's method grouped patients with minimal pain (visual analog scale score, 0-4) and with substantial pain (visual analog scale score, 5-10). Stepwise logistic regression yielded 5 variables that significantly explained the variation in the probability of perceived substantial pain at 10% level of significance: lateral cartilage clarity (LCC; P = .025), medial cartilage clarity (MCC; P = .20), medial cartilage thickness (MCT; P = .041), medial meniscus protrusion (MMP) (P = .029), and osteophytes at medial femoral margin (P = .082), with 63% overall prediction accuracy. When age and sex were added, 4 variables remained significant at a 10% level of significance: LCC, MCC, MCT, and MMP, with 65% overall prediction accuracy. The receiver operating characteristic curve of this...