Discussion

Lesions of flexor tendons all too often threaten the athletic career as these tendons play a vital role in locomotor efficiency, acting as energy saving springs. This biomechanical function is based on the unique architecture with a collagenous matrix that is 3-dimensionally organised into tendon bundles. Symptoms of tendinopathy are frequently only the tip of the iceberg; their development may be due to single overloading, but more often it is a gradual matrix deterioration due to repetitive overstraining that can remain unobserved for months or even years. Early detection of exercise effects might play a crucial role in injury-prevention.
Ultrasonography (US) has, like no other imaging technique, the potential to provide an inward view into the tendon's architecture. However, US measurements of the tendon's cross- sectional area (Avella et al. 2009) or mean echogenicity (Gillis et al. 1993) has up until this point shown little benefits for the monitoring of exercise effects and for the prediction of tendon injury. This is caused by the fact that US assessment is essentially subjective and/or poorly reproducible due to instrumental variables and transducer handling (van Schie et al. 1999). Another confounding factor is that, as a consequence of limits of resolution, every US image is a mixture of structural reflections and interfering echoes: only relatively large structures, like secondary tendon bundles (fasciculi), generate reflections, while smaller entities, such as fibrils and cells, will result in interference, each with their specific dynamics in real-time US which is not captured in still 2-dimensional US images (van Schie et al. 2001).
Therefore, a method for 'computerised ultrasound tissue characterisation' (UTC) was created for objective evaluation of the integrity of the collagenous matrix. UTC is based on standardised data-collection by means of an ultrasound probe that moves automatically along a tendon's long axis, collecting transverse images at even distances of 0.2 mm. In this way, a 3D ultrasound data-block is created and dedicated UTC-algorithms can quantify the dynamics of echo-patterns in contiguous images which allows the discrimination of 4 different echo-types, related to size and stage of integrity of structures in the matrix:
- Echo-type I, generated by reflections at intact and aligned
fasciculi
- Echo-type II, generated by reflections at discontinuous or waving fasciculi
- Echo-type III, generated by interfering echoes from mainly
fibrillar components
- Echo-type IV, generated by mainly cellular components and fluid in amorphous tissue.

Fundamental research revealed that the ratios of these 4 echo-types are highly correlated with histo-morphological characteristics of tendon tissue, showing the discriminative power of UTC for tissue characterisation (van Schie et al. 2001, 2009). Furthermore, UTC appeared to be highly reproducible with excellent intra- and inter-observer reliability (ICC over 0.90), which facilitates objective monitoring.

Some observations:
A. Normal superficial digital flexor tendons in young-mature horses (2 - 5 years of age) are characterised by 85 - 90% type I, 10 - 15% type II and less than 5% type III plus IV echoes,
B. In racehorses increasing percentages of echo-types II and III can be found 24 - 48 h post race, indicative for a swollen ground substance and separation of tendon fibrils with no free fluid. These changes usually return to baseline within
72 h post race. This reaction isn't necessarily the result of changes in collagen structure or content but may also be caused by short-duration up-regulation of larger proteoglycans and aggregation of water resulting in swelling of the ground substance (Docking et al. 2012).
C. Exercise effects may also become persistent with increased percentages of echo-types III and IV, indicative for a disintegrated collagenous matrix with increased amounts of fibrillar and cellular components and free fluid. This stage of
'dysrepair' is the result of cumulative effects of repetitive overstraining that may ultimately lead to degeneration.
These observations confirm the concept that clinical tendinopathy may have an insidious onset, being the result of a continuum inflicted by repetitive overstraining (Cook and Purdam 2009).



The following schedule for UTC-check-up's of high-performance horses, aiming at early detection of matrix deterioration, is recommended:
- measurement of individual baseline values of echo-types I, II, III and IV at 3 timepoints
- prior to major event
- 4 - 7 days after major event,
- if normalised: re-check at 8-12 w.
- if still abnormal: re-check at 4 w.

It is concluded that exercise effects can be monitored by means of UTC and the detection of 'reactive' and 'dysrepair' stages may be used as an objective tool for adjusted exercise protocols, aiming at injury-prevention.