Detecting Meniscal Tears in Primary Care: Reproducibility and Accuracy of 2 Weight-Bearing Tests and 1 Non-Weight-Bearing Test

STUDY DESIGN: Diagnostic accuracy study using a cross-sectional design. OBJECTIVES: To determine the interexaminer reliability and the diagnostic accuracy in primary care of 1 existing weight-bearing meniscal test, the Thessaly test, 1 new weight-bearing test, the deep squat test, and 1 non-weight-bearing test, the joint-line tenderness test. BACKGROUND: Meniscal tears are difficult to detect in primary care. Although valuable in secondary care, weight-bearing physical examination tests require validation in primary care in unselected patients. METHODS: Between October 2009 and December 2013, 121 patients (age range, 18-65 years) seen in primary care and suspected of having internal derangement of the knee of less than 6 months in duration were included in the study. Diagnostic accuracy of the 3 meniscal tests was determined based on assessment with magnetic resonance imaging. The meniscal tests were performed by 3 trained physical therapists, who were not informed about the patient history and magnetic resonance imaging results. Each test was performed independently by 2 of the 3 trained physical therapists in alternating pairs. RESULTS: The Thessaly test and the deep squat test had a moderate level of interexaminer reliability, with kappas of 0.54 and 0.46, respectively. The joint-line tenderness test had poor interexaminer reliability and was therefore not assessed for diagnostic accuracy. The following results are reported separately for both examiners. The Thessaly test had a sensitivity of 66.7% (95% confidence interval [Cl]: 53.0%, 78.0%) and 51.2% (95% Cl: 36.8%, 65.4%), a specificity of 37.9% (95% Cl: 27.2%, 50.0%) and 43.5% (95% Cl: 30.2%, 57.8%), a positive likelihood ratio of 1.07 (95% Cl: 0.82, 1.41) and 0.91(95% Cl: 0.62, 1.33), and a negative likelihood ratio of 0.88 (95% Cl: 0.54, 1.45) and 1.12 (95% Cl: 0.72, 1.76). Similarly, the deep squat test had a sensitivity of 74.5% (95% Cl: 61.1%, 84.5%) and 76.7% (95% Cl: 62.3%, 86.9%), a specificity of 42.4% (95% Cl: 31.2%, 54.4%) and 36.2% (95% Cl: 24.0%, 50.5%), a positive likelihood ratio of 1.29 (95% Cl: 0.97, 1.68) and 1.20 (95% Cl: 0.92, 1.58), and a negative likelihood ratio of 0.60 (95% Cl: 0.35, 1.04) and 0.64(95% Cl: 0.33, 1.25). CONCLUSION: Although the Thessaly and deep squat tests have a moderate level of reliability, neither test is sufficiently accurate to help in the diagnosis of meniscal tears in primary care. Future research should focus on other relevant patient variables instead of on physical examination tests in the detection of meniscal tears. LEVEL OF EVIDENCE: Diagnosis, level 3b