Correlations of magnetic resonance imaging classifications with preoperative functions among patients with refractory lateral epicondylitis

BACKGROUND: To evaluate the correlations between three magnetic resonance imaging (MRI) classifications and preoperative function in patients with refractory lateral epicondylitis (LE). METHODS: We retrospectively reviewed patients with refractory LE who underwent arthroscopic treatment. Signal changes in the origin of the extensor carpi radialis brevis (ERCB) were evaluated based on three different MRI classification systems. Spearman’s rank correlation analysis was used to analyse the correlation between each MRI classification and the preoperative functional and visual analogue scale (VAS). The lateral collateral ligament complex (LCL) in all patients was evaluated using both MRI and arthroscopy. The Mann–Whitney U test was used for the comparison of preoperative VAS and all functional scores between patients with refractory LE combined with LCL lesions, and those without. RESULTS: There were 51 patients diagnosed with refractory LE between June 2014 to December 2020, all of whom were included in this study. The patients included 32 women and 19 men with a mean age of 49.1 ± 7.6 years (range, 39–60 years). The average duration of symptoms was 21.1 ± 21.2 months (range, 6–120 months). The intra-observer agreements for Steinborn et al.’s classification were 77.9%, 76.0%, and 76.7%, respectively. The inter-observer reliabilities of the three classifications were 0.734, 0.751, and 0.726, respectively. The average intra-observer agreement for the diagnosis of abnormal LCL signal was 89.9%, with an overall weighted kappa value of 0.904. The false-positive rate was 50%, and the false-negative rate was 48% for LCL evaluation on MRI. Spearman's rank correlation analysis did not find significant correlation between any of the three MRI classifications and preoperative VAS or any functional scores (all P > 0.05). There were no significant differences in the VAS and functional scores between patients with abnormal LCL signals on MRI and those without LCL lesions (all P > 0.05). CONCLUSIONS: Preoperative MRI findings in patients with refractory LE cannot reflect the severity of functional deficiency. Preoperative MRI grading of the origin of the ERCB and preoperative MRI for LCL signal change cannot assist the surgical plan for the treatment of patients with refractory LE.