Purpose: To determine the sensitivity and specificity of ultrasonography (US) in the detection of lateral epicondylitis and identify the US findings that are most strongly associated with symptoms.
Materials and methods: Internal review board approval was obtained for retrospective review of the patient images, and the need for informed consent was waived. Internal review board approval was also obtained for scanning the 10 volunteers, all of whom gave informed consent. The study was compliant with the Health Insurance Portability and Accountability Act. US of the common extensor tendon was performed in 20 elbows in 10 asymptomatic volunteers (six men, four women; age range, 22-38 years; mean age, 29.6 years) and 37 elbows in 22 patients with symptoms of lateral epicondylitis (10 men, 12 women; age range, 30-59 years; mean age, 46 years). Fifty-seven representative images, one from each elbow, were randomly assorted and interpreted by three independent readers who rated each common extensor tendon as normal or abnormal. Abnormal images were further classified as demonstrating one or more of eight US findings. Readers interpreted each image at two separate sessions to determine intrareader variability. The authors calculated the sensitivity and specificity of US in the diagnosis of lateral epicondylitis and the odds ratio for each US finding. Odds ratios were considered statistically significant at P < .05 when 95% confidence intervals did not include one.
Results: Sensitivities of US in the detection of symptomatic lateral epicondylitis ranged from 72% to 88% and specificities from 36% to 48.5%. Odds ratios for the following findings were statistically significant (P < .05) for both reading sessions: calcification of common extensor tendon, tendon thickening, adjacent bone irregularity, focal hypoechoic regions, and diffuse heterogeneity. Odds ratios for lateral epicondyle enthesophytes were statistically significant (P < .05) for the first reading session only. Odds ratios for linear intrasubstance tears and peritendinous fluid were not statistically significant.
Conclusion: US of the common extensor tendon had high sensitivity but low specificity in the detection of symptomatic lateral epicondylitis. The relationship between symptoms and intratendinous calcification, tendon thickening, adjacent bone irregularity, focal hypoechoic regions, and diffuse heterogeneity was statistically significant.