BY ANDREA S. KLAUSER, M.D., TOBIAS DE ZORDO, M.D., AND RALPH FASCHINGBAUER, M.D.
DR. KLAUSER is an associate professor of radiology at the University Hospital of Innsbruck in Austria. DR. DE ZORDO is a radiology fellow at the same institution and also a research fellow at the London Health Science Centre, University Hospital, in London, Ontario, Canada. DR. FASCHINGBAUER is a radiology fellow at the University Hospital of Innsbruck.
Our assessment of lateral epicondylitis comprised 38 elbows from patients with clinically suspected tennis elbow and 44 asymptomatic elbows. Once again, the healthy volunteers' elbows revealed a hard tendon structure, whereas a statistically significant softening could be observed in the patients' elbows (Figures 6 and 7).17 RTSE also detected lateral collateral ligament and overlaying fascia involvement. Comparison with clinical examination as the reference standard showed RTSE to have a sensitivity of 100%, specificity of 89%, and accuracy of 94%. Correlation with ultrasound findings was good once again (r = 0.900).
These results show that RTSE is of value in the diagnosis of tendinopathy, at least in Achilles tendon and common extensor origins at the elbow. Although these are preliminary results, significant differences between asymptomatic and symptomatic tendons were found in both cases.
The sensitivity and specificity of RTSE were almost identical to ultrasound in diagnosing Achilles tendinopathy. Slightly better results for RTSE were obtained when diagnosing tennis elbow. Both studies uncovered more tendon alterations when RTSE was used instead of conventional ultrasound, perhaps indicating that RTSE is better at detecting tendinopathy at an early stage. This suggestion might be further supported by the frequent detection on RTSE of contralateral Achilles tendon involvement, which was found only in patients and not in healthy volunteers.
Our study population was relatively small. Further longitudinal studies should now be performed to verify our preliminary findings. One possible weakness in our study was the use of clinical examination as the gold standard. No other accurate, noninvasive reference standard is available, however. Results were correlated with ultrasound, which is considered to be one of the methods of choice for diagnosing tendinopathy. Correlation with MRI will be examined in future studies.
Another limitation is the operator dependency of this freehand technique. Care was taken to obtain reproducible images, but we did not evaluate reproducibility in any detail. Neither interobserver nor intraobserver variability was calculated.
