What Your Tendon Actually Goes Through

Strength and ConditioningRehabPerformance Coaching
Written By Cat Hoh

Coaches drill eccentrics.. Clinicians reach for isometrics… But, how do both muscle contractions play apart in performance and rehab?

A tendon does two jobs every stride. It passes the muscle's force to the bone, and it stores then returns elastic energy so the muscle spends less of its own. How well it does the second job comes down to its stiffness, its size and how its collagen is organised, and all three answer to how the tendon is loaded.

Load builds tendons and load breaks them, sometimes in the same athlete in the same month. The difference is dose, recovery, and the state the tendon is already in. What follows is the mechanics of that, then the pathology when it goes wrong, then how to load it back to health.

What a tendon is, in one breath

Tendon is dense connective tissue built in a hierarchy: type I collagen assembled into fibrils, then fibres, then fascicles, then the tendon. Tenocytes sit between the fibres and do the maintenance, producing and remodelling the matrix around them. Two features shape everything else. Tendon is relatively hypocellular, with few cells across a lot of matrix, and it is poorly vascularised, leaning on diffusion for nutrition. Both make it slow to repair once it is damaged (Wang and Zhao, 2022).

One session versus months of training

The response to a single session is fast, small and mostly temporary. Give the tendon weeks instead, and it remodels into something larger and stiffer that tolerates more load. Tap between the two timescales below.

Figure 01 / Interactive

Same tendon, two timescales

Timescale / tap to change

During one session After months of training
Hoursone session

Stiffness drops after maximal contractions, and barely moves after running or hopping. Thickness thins briefly after eccentric work, then recovers. Collagen synthesis rises, peaks near 24 hours, stays up 2 to 3 days. Injury risk nudges up while stiffness is briefly down.

Weeks+of training

Stiffness and cross-sectional area increase; frequent exercisers carry thicker tendons. Collagen remodels into a larger, stiffer tendon that returns energy more efficiently. Injury risk falls as load tolerance climbs.

Mostly Achilles and patellar data. Source: Wang and Zhao (2022). Synthesis staying elevated 2 to 3 days is why heavy tendon sessions want a day or two between them.

What actually drives the change: load, not the kind of contraction

Here is where the coaching shorthand tends to be wrong. The size of the load matters more than whether the contraction shortens, lengthens or holds. A meta-analysis of tendon adaptation concluded that loading magnitude drives the change, and it takes high intensity sustained past 12 weeks to shift tendon properties. Quinlan and colleagues (2021) tested it head to head: at a moderate load, concentric and eccentric training produced the same change in the tendon, and differed only in what they did to the muscle.

Eccentric loading was never the active ingredient for the tendon. The load was, and eccentric exercise was mostly a convenient way to deliver it.

What contraction type does decide is where the adaptation lands in the muscle-tendon unit. Tap through the four modes.

Figure 02 / Interactive

Where the adaptation lands

Loading type / tap to change

Isometric Plyometric Concentric Eccentric

Tendon stiffness rises

Active muscle stiffness flat

Isometric holds stiffen the tendon and leave the muscle much as it was. A better spring for transmitting force.

Tendon stiffness flat, more extensible

Active muscle stiffness rises

Plyometric jumps leave the tendon's stiffness alone and stiffen the muscle underneath. A better spring for storing and returning energy.

Tendon modulus rises

Muscle architecture pennation angle

Concentric loading gives the same tendon result as eccentric, and builds the muscle at a steeper pennation angle.

Tendon modulus rises

Muscle architecture fascicle length

Eccentric loading gives the same tendon result as concentric, lengthens the muscle fascicles, and is often better tolerated in rehab.

Bar lengths are illustrative, not measured. Isometric and plyometric from Kubo et al. (2017), 11 untrained men, one mode per leg. Concentric and eccentric from Quinlan et al. (2021), 60% of one-rep max over 8 weeks. So you pick the contraction to shape the muscle, not to stiffen the tendon.

When loading tips into injury

Tendinopathy is the umbrella term for load-related tendon pain, lost function and reduced load tolerance. The old "tendinitis" label has largely gone, because the change in the tissue is not a classic inflammatory response. That said, the strict "non-inflammatory" line has softened: inflammatory cells and cytokines do show up in overuse tendinopathy (Millar and colleagues, 2010; Rees and colleagues, 2014, both via Cook et al., 2016). The current reading is that these signals are real but do not look like traditional inflammation, and inflammation is not established as the primary driver.

The most useful framework for the tissue is Cook and Purdam's continuum: three stages, with movement in both directions, driven mainly by adding or removing load. Reversible early, barely reversible at the end. Tap a stage to see it.

Figure 03 / Interactive

The tendon continuum

Stage / tap to change

Reactive Dysrepair Degenerative
Normal healthy Reactive swells Dysrepair disorganises Degenerative cell death reversible with load management (early) limited to no reversibility (late) Adding load pushes right (worse); reducing load moves left (recovery), mainly in the early stages.

Reactive

In the tissue
Proliferative response to sudden overload; proteoglycans and bound water up, tendon thickens; collagen mostly intact
Imaging
Fusiform swelling, diffuse hypoechoic on ultrasound, little MRI change
Typical athlete
Younger; a load spike or a direct blow; detrained athlete returning
Reversible?
Yes, if load is reduced
Loading
Reduce load; avoid aggressive eccentric and energy-storage work

Dysrepair

In the tissue
Failed healing; more cells, matrix disorganising; vascular and nerve ingrowth can begin
Imaging
Swollen, focal hypoechoic areas, some Doppler vascularity
Typical athlete
Any age, chronically overloaded
Reversible?
Partly
Loading
Manage load, load to stimulate the matrix, progress carefully

Degenerative

In the tissue
Cell death and acellular areas, disordered matrix, little collagen, vessels; can rupture
Imaging
Hypoechoic regions, few collagen reflections, large vessels, focal
Typical athlete
Middle-aged recreational athlete, recurrent pain
Reversible?
Little to none
Loading
Progressive heavy loading to build the healthy tissue, not the lesion

Acute and chronic are a different axis. They describe how long symptoms have run (chronic is conventionally over three months); reactive and degenerative describe the tissue. Source: Cook and Purdam (2009); degenerative loading updated in Cook et al. (2016).

Two-thirds of tendons degenerate enough to rupture were pain-free right up until they did.Cook and Purdam, 2009

That is the trap. Pain does not track the stage. Tendons that look normal on imaging can hurt, and a pain-free tendon is not always a healthy one, so you cannot read the pathology off the pain level alone.

Rehab: match the load to the stage

The one idea that matters most: the right intervention depends on where the tendon sits on the continuum. A reactive tendon wants less load, so you cut the frequency and intensity of high-tension, energy-storing work and let it settle. A degenerative tendon wants the opposite, progressive heavy loading. The mechanism has been reframed since 2009: loading does not reliably restructure the degenerate core, so the aim is to build the load capacity of the healthy aligned tissue around it, what Cook calls treating the doughnut, not the hole (Cook et al., 2016).

On which loading protocol, the honest reading is that several work and none is clearly best.

Figure 04 / Table

Loading protocols, head to head

ComparisonWhat the trial found
Eccentric vs heavy slow resistanceEqual at 52 weeks; heavy slow resistance had a short-term edge in satisfaction at 12 weeks onlyBeyer 2015, 58 patients, chronic midportion Achilles
Isometric vs isotonicIsometric not superior on any outcome; evidence weak, 7 of 10 trials poor qualityClifford 2020, meta-analysis of 10 RCTs
Progressive tendon-loading vs eccentricProgressive loading gave better outcomes at 24 weeksBreda 2021, patellar, via Wang and Zhao
Concentric vs eccentricSimilar change in tendon modulusQuinlan 2021, moderate load, 8 weeks

Mostly Achilles and patellar, quality varies. Sources: Beyer (2015), Clifford (2020), Wang and Zhao (2022), Quinlan (2021).

Isometrics get oversold. They do build tendon stiffness, and a hold can take the edge off pain in a session, but as a treatment they are one tool inside a progressive plan, not a switch that turns pain off. Where the problem sits on the tendon also changes the exercise: the Alfredson eccentric protocol was built for mid-portion Achilles, and its full-dorsiflexion position aggravates an insertional tendon by compressing it against the heel bone. A floor-level version that avoids dorsiflexion lifted satisfaction from about a third to two-thirds (Jonsson et al., 2008). Do not overcorrect and ban dorsiflexion entirely, since tendon force is highest there; limit it while painful and reintroduce it as symptoms allow.

The pain rule. Some pain during loading is fine, within limits. Silbernagel's model allows activity pain up to 5 out of 10, provided it settles by the next morning and does not climb from week to week. Athletes who kept running and jumping under that rule did no worse than those who rested (Silbernagel et al., 2007).

Sequencing follows from all of it. Since the stretch-shortening cycle, the running and jumping, is what provokes tendinopathy, energy-storage load comes last.

Figure 05 / Diagram

Return-to-sport loading sequence

Settle / manage load Heavy slow strength Faster, heavier lifts Plyometric & energy-storage introduce last progression over weeks to months

Energy-storage load provokes tendinopathy, so it goes last, once earlier stages are handled within the pain rule above. Sources: Cook and Purdam (2009); Silbernagel et al. (2007). Reintroducing plyometrics too early is a common way to send a settling tendon backwards.

The honest edges

Hold the picture loosely where the evidence is thin. Most of it comes from the Achilles and patellar tendons, and whether it transfers to the shoulder, elbow or hip is assumed more than shown. Several strong-sounding claims rest on modest foundations: the stiffness study is 11 people, the isometric verdict rests on ten trials with seven of poor quality, and much of the load-beats-type consensus comes from a small, connected set of labs. Women are close to absent from all of it, even though their tendons synthesise collagen differently after exercise.

So the working rule is duller than the folklore and easier to defend. Pick the load and the time under tension first, because that is what the tendon is listening to. Give it 12 weeks and more, not three. Treat the day after a hard session as seriously as the session. Reach for contraction type to shape the muscle or stay under someone's pain ceiling, not as your lever on the tendon. Match the loading to the stage. And read every number here as an opening bet on the athlete in front of you, not a rule they owe you.

References

Cat Hoh
Previous

Same Engine, Different Miler
Next

Eight Hours Is the Floor: Sleep and the Athlete