The need for non-invasive tools to accurately quantify the in vivo mechanical properties of tendon is well known. Studies published in the literature have shown that mechanical properties of tendon, such as strain and stiffness are good predictors of the amount of mechanical damage the tendon has sustained [1-4] Preliminary studies in our lab have shown that applying an acoustoelastic analysis to ultrasound images of tendon during loading has potential to estimate load-dependent stiffness changes.[5] This thesis investigates whether acoustoelastic (AE) ultrasound might be used to identify regional changes in tendon mechanics due to tearing, with potential clinical application for evaluating rotator cuff tendon damage.

To understand how tearing affects the ultrasound signal we mechanically characterized a sheep infraspinatus tendon ex vivo tear model. We then used the model to examine the average grayscale ultrasound echo intensity changes during loading for both partial- and full-thickness tendon tears. The ultrasound echo change becomes less systematic and the magnitude of the change decreased when a tear was introduced, but these changes were only significant when the defect was through the full thickness of the tendon. We performed complimentary tissue strain analysis using optical markers, which showed that there may be a stiffness difference between the bursal and articular regions of the sheep 1ST. The average grayscale echo intensity results suggest that acoustoelastic analysis may be able to detect this difference.

In order to verify that the AE effect can be detected in vivo we performed acoustoelastic analysis on cine ultrasound sequences collected during a single artificially induced muscle twitch contraction. We found that the tendon image intensity increased proportionally with the muscle contraction force in four of the five subjects. Estimates of muscle twitch contraction times were comparable to values obtained using traditional methods (muscle-tendon-junction tracking). From this study, we concluded that the AE effect is discernible during in vivo tendon loading, but that reliably observing the phenomenon in individual subjects can be challenging.

Finally we applied AE analysis to cine ultrasound sequences collected on patients who had a history of rotator cuff tears. From this study we learned that using AE techniques for examining shoulder tendon pathologies will be challenging. Based on our prior work and AE theory, it is desirable to align the transducer with the tendon fiber direction and to apply well controlled stretch on to the tissue. Because patients with rotator cuff pathology often have a limited range of motion getting them into a position where the tendon can be both imaged and stretched simultaneously was difficult. Going forward, it will be important to devise simple techniques to control both the stretch and the rate of stretch applied to the tendon. Finally because of the way the supraspinatus tendon wraps around the humeral head it will be important to better understand the effects of both tears and tendon anisotropy on image intensity.