The Validity of the McMurray Test and Joint Line Tenderness for Diagnosing Meniscal Lesions of the Knee: a Systematic Review

Article type
Year
Authors
Scholten R, Deville W, Biji D, Opstelten W, van DPC, Bouter L
Abstract
Introduction:

Objectives: To summarize the available evidence of the diagnostic validity of the McMurray test and joint line tenderness (JLT) for diagnosing meniscal lesions of the knee.

Methods: Computerized literature search of MEDLINE and EMBASE and additional reference tracking. Studies had to address the validity of at least one physical diagnostic test for the assessment of meniscal lesions of the knee with arthrotomy, arthroscopy or MRI as gold standard. Study selection, assessment of the methodological quality and data extraction were performed by two reviewers independently of each other. Meta-analysis was performed according to a protocol based on Irwig et al. (J Clin Epidemiol 1995;48:119-30)

Results: Of 374 eligible studies, 10 met the inclusion criteria. Three additional studies were found by reference tracking. Twelve studies addressed the McMurray test or JLT or both. The most prevalent methodological shortcomings were: no independent (blind) assessment of the results of the index test and reference test (11 studies), existence of verification bias (all studies), missing values not addressed properly (7). No study had been performed in a primary care setting and in 7 studies no inclusion criteria were stated. The index test was described in sufficient detail in 6 studies and in 6 studies a definition of a positive test result was given. Study results were highly heterogeneous. The sensitivity of the McMurray test ranged from 0.20 to 0.66 and the specificity ranged from 0.57 to 0.96. The sensitivity of JLT ranged from 0.58 to 0.95 and the specificity ranged from 0.05 to 0.74. The summary ROC-curves indicated little discriminative power of both tests.

Discussion: The methodological quality of studies addressing the diagnostic validity of the McMurray test and JLT was poor and the results were highly heterogeneous. The poor test characteristics of both tests indicate that they are of little value for clinical practice. If the existence of a meniscal lesion has significant treatment consequences, more advanced diagnostic methods seem to be warranted.