Assessing the treatment response of lateral elbow tendinopathy using time-dependent ultrasonography, Doppler imaging, and elastography

Tendinopathies commonly exhibit the following histopathological changes: collagen disorganization, cellular density variations, increased fibroblast activity, neovascularity, and extracellular matrix myxoid and hyaline changes [22]. These lead to structural (thickening, hypoechogenicity, neovascularity, tears, calcifications, and enthesophytes) and mechanical (softening) abnormalities that can be detected with imaging techniques to inform care management [23].

Our study revealed that fenestration and surgery equally reduced tendon neovascularity in patients with LET, showing significant changes at 6–12 months postintervention. These results align with prior research on percutaneous therapies and surgery for tendinopathies [7, 8, 24].

However, our study found that fenestration was superior to surgery for maintaining or improving tendon structure. Surgery thickened tendons without enhancing echostructure, while fenestration maintained the tendon’s size and improved echostructure at 6–12 months. This discrepancy may stem from the different techniques used. Surgery ablates immature reparative tissue and repairs the tendon interface through a skin incision [25], while fenestration induces an acute inflammatory response and promotes healing and regeneration percutaneously [26].

Recently, Riggin et al [27] conducted research supporting the proposed healing mechanism of the fenestration procedure. Using US-guided fenestration on normal rat supraspinatus tendons, they compared mild needling (three passes), moderate needling (nine passes), and no needling in a control group. Assessments at 1- and 6-week intervals included Doppler imaging, histological and immunohistochemical analysis, cross-sectional measurement, and mechanical testing (percent relaxation; maximum load, stiffness, and stress; elasticity modulus). Results indicated no permanent tendon damage from needling. Both needling groups initially showed increased blood flow, type III collagen deposition, and inflammation, returning to baseline values by the 6-week. The moderate group exhibited increased tendon size at one week, returning to baseline at 6 weeks, while mild needling did not affect size. Both groups demonstrated decreased stiffness and modulus at one week, with only the mild group fully recovering at 6 weeks. Modulus did not return to baseline in the moderate group. These findings raise questions about the optimal fenestration extent and required time for tendon recovery. The moderate needling group may have fully recovered beyond the 6 weeks covered by the study.

Similarly, our study observed that fenestration elevated tendon SWV in patients with LET over the medium and long term, indicating increased tendon stiffness. In contrast, surgery tended to reduce tendon SWV in the medium term and did not show improvement compared to baseline in the long term. These findings concur with those from animal studies on surgically injured tendons, showing an initial decline and partial recovery of mechanical properties over time [28, 29].

We found a moderate to weak correlation between changes in SWV and clinical outcome, underscoring the need for cautious result interpretation. Increased SWV correlated with decreased PRTEE score, signifying improved clinical outcome. Long-term SWV changes (6–12 months postintervention) exhibited a stronger association with clinical outcome than changes within the first 6 months, suggesting that enhanced tendon stiffness becomes noticeable in the long term. This aligns with the lengthy physiological tendon healing process involving inflammation, repair, and remodeling spanning several months to a year. In the remodeling phase, starting 1–2 months after injury and extending over a year, collagen I synthesis occurs and contributes significantly to increased tendon stiffness [30]. Despite these observations, the absence of statistically significant differences in SWV between successful and failed treatments suggests that factors beyond elasticity influence clinical recovery.

We acknowledge some limitations. First, the sample size was relatively small, which may limit the generalizability of the findings. However, this randomized trial employed rigorous methods and statistical analyses to ensure the reliability and validity of the results. While generalizing to a larger population may be limited, the findings provide valuable evidence for future research. Second, blinded to the patient’s clinical outcome but not to the intervention groups, one radiologist performed the US examinations and imaging parameter assessments. However, the US exams were conducted over three years at 6-month intervals. This extended duration reduced recall bias and provided a representative approach to clinical practice. Third, the Acuson S3000 scanner had a maximum SWV of 10 m/s. Presently, some scanners can assess speeds over 16 m/s. However, only 0.2% (10/477) of our SWV measurements exceeded the 10 m/s limit, unlikely to have significantly impacted our results. A higher-frequency elastography-capable transducer could have been employed for SWV measurement in the CET-RCL complex. However, it is important to note that the spatial resolution of 2D-SWE is inversely related to transducer frequency, and as transducer frequency increases, shear wave propagation diminishes in deeper tissues due to US absorption [31, 32]. Furthermore, Dillman et al’s research [11] demonstrated the reliability and comparability of SWV measurements using the Acuson S3000/9L4 transducer/VTIQ elastography method (the method used in this study) to the Aixplorer (Supersonic Imaging)/SL 10-2 and SL15-4 transducer method at shallow depths (1.0, 2.5, 4.0 cm). Therefore, our selected approach was suitable for accurately evaluating SWV in the CET-RCL complex. We recognize that reflection artifacts off the lateral epicondyle may have affected SWV assessments within the CET-RCL complex. In this longitudinal study, we prioritized a standardized SWV assessment method to ensure reproducible measurements. Additionally, analyzing the delta change of SWV measurements between follow-ups minimized the potential impact of reflection artifacts on our results. We did not assess the inter-reader agreement for SWV, US, and power Doppler imaging parameters. However, in a previous study, we assessed inter-reader agreement for US and power Doppler imaging features of the CET-RCL using the same grading schemes [21]. Our findings revealed an almost perfect inter-reader agreement for maximum tendon thickness scores (intraclass correlation coefficient, 0.84), and perfect (k = 1.00) and moderate (k = 0.47) inter-reader agreement for power Doppler and echostructure scores, respectively.

In conclusion, while fenestration and surgery equally alleviated clinical symptoms and reduced tendon neovascularity in LET, fenestration exhibited additional benefits by improving tendon structural and mechanical properties. Shear-wave velocity may provide measurements to monitor tendon elasticity during healing. Even though no definitive correlation was observed between SWV changes and clinical outcome, this study underscores the importance of long-term imaging follow-up when monitoring tendon healing response.