Diagnostic value of self-reported mechanical symptoms for diagnosing large meniscal tears in patients aged 40 years or older with meniscal tears

Purpose: Meniscal tears, which are longitudinal-vertical tears (e.g. bucket-handle tears) and/or involve all three meniscal sub regions (anterior horn, body and posterior horn), are typically large and assumed to cause patient perceived mechanical knee symptoms (knee grinding or clicking and knee catching or locking). However, whether the presence of such mechanical symptoms is useful for diagnosing these forms of meniscus tears is unknown. Therefore, we investigated the diagnostic values of having mechanical symptoms in diagnosing a bucket-handle tear and/or a tear involving all three meniscal sub regions in middle-aged and older patients undergoing arthroscopic meniscal surgery.
Methods: This study is a secondary analysis of Pihl et al. OARSI 2017. The study included patients aged 40 years or older undergoing surgery for a meniscal tear from Knee Arthroscopy Cohort Southern Denmark (KACS) consecutively recruited from February 2013 to January 2014. Within 2 weeks prior to surgery, patients completed online questionnaires self-reporting presence of mechanical symptoms (i.e. knee grinding or clicking and knee catching or locking) assessed by two single items from the Knee Injury and Osteoarthritis Outcome score (KOOS). At arthroscopy, the operating surgeon recorded information about specific meniscal pathology (i.e. tear type and radial location) using a modified version of the International Society of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine (ISAKOS) classification of meniscal tears. Sensitivity, specificity, positive predictive value and negative predictive value were calculated separately for experiencing knee grinding and/or clicking, knee catching and/or locking and the two combined in diagnosing a bucket-handle tear and/or a tear involving all three meniscal sub regions (i.e. the anterior horn, mid-body and posterior horn).
Results: In total, 398 patients of the 491 KACS cohort patients were aged 40 years or older. Of these, 397 (mean age 54.4 (40-76) years, 54% men) with complete baseline data for the analyses in this study were included. Most patients, 341 (86%), experienced knee grinding and/or clicking, 202 (51%) experienced knee catching and/or locking, and 190 (48%) patients experienced both types of symptoms. The prevalence of having a bucket-handle tear and/or a tear involving all three meniscal sub regions was 15% (n=61). Self-reported knee grinding and/or clicking showed the highest sensitivity, but also the lowest specificity (Table 1), while knee catching and/or locking and the combination of all mechanical symptoms showed very similar values for sensitivity and specificity. The predictive values were virtually the same for all mechanical knee symptoms with low positive predictive values and fairly large negative predictive values (Table 1).
Conclusions: In this exploratory study we found a low diagnostic value of experiencing mechanical symptoms for diagnosing large meniscal tears in a population of patients with meniscal tears. The presence of mechanical symptoms only marginally changed the baseline probability of having a bucket-handle tear and/or tear involving all three meniscal sub regions in patients aged 40 years or older with meniscal tears. As predictive diagnostic values are largely driven by prevalence of the outcome, the diagnostic value of mechanical symptoms can be expected to be of even less utility in patients with a suspected but not yet verified meniscal tear. Based on this and the results in this study, presence of patient-reported mechanical symptoms is of questionable value in the diagnosis of patients having a bucket-handle tear and/or a tear involving all three meniscal sub regions in middle-aged and older patients.