Eccentric Loading for Tendinopathy: The Best-Validated Rehab Protocol

Educational journalism, not medical advice. Every claim here is checked against its cited sources by editor Tim Bunce — a health writer, not a physician. It isn’t specific to your situation: for health decisions, talk to your own clinician. How we work →

Why eccentric specifically

Tendinopathy involves disorganised collagen and impaired tendon remodelling. The Alfredson 1998 paper (Achilles tendinopathy) pioneered the eccentric loading approach, showing dramatic improvements where other conservative treatments had failed. The follow-up biomechanical work documented that eccentric contractions produce a specific tendon-loading pattern — the slow lengthening under load generates the mechanotransduction signal that drives tendon repair and remodelling Alfredson 1998.

The effect appears specific to eccentric loading. Concentric-only training, isometric holds, and stretching alone don’t produce the same outcomes. Heavy slow resistance training (a variation that includes both concentric and eccentric at slow tempos) performs comparably to pure eccentric in some trials Beyer 2015.

What the trial evidence shows

• Achilles mid-substance tendinopathy: 60-80% symptom improvement at 12 weeks of structured eccentric heel drops Alfredson 1998.
• Patellar tendinopathy: Eccentric single-leg squats on decline board produce 60-75% improvement at 12 weeks.
• Lateral epicondyle (tennis elbow): Eccentric wrist-extension protocols produce 60-80% improvement at 6-8 weeks.
• Rotator cuff tendinopathy: Eccentric external-rotation work effective; heavy slow resistance often equivalent.
• Insertional Achilles tendinopathy responds less reliably to standard eccentric protocols than mid-substance. Modified protocols (avoiding dorsiflexion past neutral) work better.

“Eccentric loading protocols produce clinically meaningful improvement in chronic tendinopathy outcomes across multiple tendon sites, with response rates of 60-80% at 12 weeks. The loading must be sustained at the prescribed intensity despite discomfort, which is the dominant barrier to patient adherence.”

— Alfredson et al., Am J Sports Med, 1998 view source

The Achilles protocol (Alfredson)

The original validated protocol:

• Stand on a step with the ball of the foot on the edge, heel hanging off.
• Use both legs to rise up to full plantarflexion (heels up).
• Transfer weight to affected leg only.
• Slowly lower the affected heel below the step level over 3-5 seconds. This is the eccentric phase.
• Use the other leg to return to the top. Don’t do the concentric portion on the affected side.
• 3 sets of 15 reps with knee straight + 3 sets of 15 with knee bent (targets soleus separately from gastrocnemius).
• Twice daily, 7 days per week, for 12 weeks.
• Add load when pain at the planned reps drops below 5/10: backpack with weights, weighted vest.
• Expected pain: 3-5/10 during the loading phase. Pain that returns to baseline by next morning is fine; worsening pain across days suggests the dose is too high.

The patellar tendon protocol

• 25° decline board for the affected leg. Stand on it with the affected leg only.
• Slow single-leg squat down to ~70° knee flexion over 3-5 seconds. The decline angle increases patellar tendon loading.
• Use the other leg or hands to return up.
• 3 sets of 15 reps, twice daily, 12 weeks.
• Add load progressively with weighted vest or dumbbells as pain at planned reps drops.

The lateral epicondyle protocol

• Sit with forearm on knee, hand off the edge, palm down. Hold a 1-2 kg dumbbell.
• Use the opposite hand to lift the wrist up to extension.
• Slowly lower the wrist into flexion over 3-5 seconds.
• 3 sets of 15 reps, daily, 6-8 weeks.
• Progress load as pain at planned reps drops.

Common pitfalls

• Stopping because it hurts. The expected pain (3-5/10) is part of the protocol. Patients who stop at the first discomfort don’t complete the dose and don’t get the outcome.
• Going too fast. The slow eccentric (3-5 seconds) is critical. Fast lowering doesn’t produce the same effect.
• Not loading progressively. Adults often stop adding weight when symptoms improve. The progression matters — finishing the protocol at bodyweight only doesn’t produce the same tendon adaptation as adding 15-25% bodyweight over the 12 weeks.
• Skipping days. The published trials use daily or twice-daily dosing. Three sessions weekly produces inferior outcomes.
• Expecting fast results. Most adults notice meaningful improvement at 6-8 weeks. Faster results suggest the underlying pathology wasn’t tendinopathy.

Practical takeaways

• Eccentric loading is the best-validated rehabilitation approach for chronic tendinopathy: 60-80% symptom improvement at 12 weeks across Achilles, patellar, and lateral epicondyle protocols.
• The protocols are uncomfortable by design. 3-5/10 pain during loading is expected, not a contraindication.
• Critical details: slow tempo (3-5 seconds), high frequency (daily or twice daily), progressive load.
• Expect 6-12 weeks before meaningful improvement.
• Insertional Achilles tendinopathy needs modified protocol (avoid dorsiflexion past neutral); see a physiotherapist for individualised adjustment.

Isometric holds for pain relief: what the evidence actually shows

You will see "isometrics first, eccentrics later" repeated all over rehab social media. An isometric contraction is one where the muscle works hard but the joint does not move — think of pushing against an immovable wall, or holding a leg-extension machine still at a fixed angle. The enthusiasm traces to a small, much-shared 2015 crossover study by Rio and colleagues, who tested single sessions of isometric versus moving (isotonic) contractions in volleyball players with patellar tendinopathy. The isometric session produced a striking immediate drop in tendon pain that persisted for at least 45 minutes, alongside a reduction in a brain-based measure called cortical inhibition Rio 2015. That single result launched the idea that an isometric hold is a reliable on-the-spot painkiller for cranky tendons.

The larger picture is more sober. When Clifford and colleagues pooled ten randomised trials in a 2020 systematic review and meta-analysis covering patellar, rotator-cuff, lateral-elbow, Achilles and gluteal tendinopathies, they found that isometric exercise did help — but it was not superior to ordinary moving (isotonic) loading for either pain or function, and in rotator-cuff cases it was no better than ice in the short term. Most of the included studies were also rated low quality Clifford 2020. The honest reading is that isometric holds are a legitimate, low-risk option you can fold into a loading programme — especially useful on a flare-up day or for an athlete who needs to settle pain before competing — but they are not a magic switch, and the dramatic single-session relief seen in six volleyball players does not reliably reproduce across tendons or people. They are a tool in the box, not a mandatory first step before you are "allowed" to load the tendon properly.

Is eccentric loading uniquely best — or is the real ingredient just loading the tendon?

The classic story, anchored by Alfredson's 1998 heel-drop trial, is that the slow lengthening (eccentric) phase carries a special healing signal that other contraction types cannot match Alfredson 1998. That story is now contested by better head-to-head evidence. The Beyer trial randomised people with Achilles tendinopathy to eccentric training or heavy slow resistance (slow lifts using both the lifting and lowering phases) and found both produced equally good, lasting results at one year — with heavy slow resistance actually winning on patient satisfaction and time efficiency Beyer 2015. For the patellar tendon, a 2024 network meta-analysis went further: when it ranked eccentric, isometric, heavy slow and moderate slow resistance against each other for improvement on the standard VISA-P symptom score, eccentric-only training ranked lowest, while moderate slow resistance ranked best — though the authors stressed that no two modes were directly compared head-to-head, so the ranking is suggestive rather than definitive Li 2024.

A 2023 BJSM systematic review with meta-analysis of resistance-exercise dose by Pavlova and colleagues helps explain why the "magic of eccentric" framing keeps weakening: across studies, the variable that tracked most consistently with better outcomes was higher load (intensity), not which contraction type was used. Training volume showed minimal, inconsistent effects, and — notably — the lowest training frequencies were associated with the largest effect sizes, not the highest Pavlova 2023. The practical takeaway is freeing rather than discouraging: the active ingredient appears to be progressively loading the painful tendon with meaningful resistance over months. Eccentric heel drops remain the most-studied, best-validated entry point and a perfectly good choice, but if a person cannot tolerate the twice-daily heel-drop grind, a heavy slow resistance routine done two to three times a week is a legitimate, evidence-supported alternative rather than a compromise.

How much pain is safe, and do you have to stop your sport?

The single most common reason loading programmes fail is that people quit the moment the tendon talks back. Yet some discomfort during rehab is expected and, within limits, acceptable. The most useful framework here is the "pain-monitoring model" tested by Silbernagel and colleagues, who randomised people with Achilles tendinopathy either to rest from running and jumping for six weeks or to keep doing those activities while following pain rules. Both groups recovered, and the continued-activity group showed no worse outcomes Silbernagel 2007. In other words, for many people, completely stopping their sport is not required for the tendon to settle.

The rules that made continued activity safe in that trial are concrete and worth memorising: pain during the activity should stay at or below roughly 5 out of 10; it should not climb sharply afterwards; and it should settle back to your normal baseline by the next morning. Stiffness or pain that is clearly worse the following day — or that keeps creeping up week to week — is the signal to scale back the load, not to push through Silbernagel 2007. This pain-guided approach pairs naturally with the dose evidence above: because higher load matters more than punishing frequency, you generally have room to lift heavier on fewer days while staying inside the pain ceiling, rather than chasing the old twice-daily, seven-days-a-week prescription Pavlova 2023. If pain is sharp, swells visibly, or is severe enough that you are limping, that is outside this model — stop and get it assessed rather than self-managing.

Who needs to be careful: medications, age, and the injection trap

Loading is remarkably safe, but a few situations warrant a clinician's eye before you start a programme on your own. The clearest is the fluoroquinolone class of antibiotics — drugs such as ciprofloxacin and levofloxacin. The US Food and Drug Administration requires a boxed warning on these drugs for tendinitis and tendon rupture, which most often strikes the Achilles and can appear within hours of the first dose or as long as several months after finishing the course. The risk is higher in people over 60, in those taking corticosteroids, and in organ-transplant recipients, and patients are advised to stop the drug and avoid loading the area at the first sign of tendon pain FDA 2016. If your tendon pain began around a recent course of one of these antibiotics, that is a reason to see a doctor first — a fluoroquinolone-affected tendon can progress to a full rupture, and aggressive heel drops are exactly the wrong move until it has been assessed.

A second cautionary tale is the corticosteroid injection, still sometimes offered for stubborn tennis elbow or other tendons. The temptation is understandable because the short-term relief is real and often dramatic. But Coombes and colleagues' large 2010 systematic review in The Lancet found that for lateral epicondylalgia the benefit reverses over time: injected patients did markedly better than no treatment at first, yet did worse than the no-intervention group at intermediate and long-term follow-up Coombes 2010. That pattern — short-term gain, longer-term penalty — is a strong argument for treating a progressive loading programme as the first-line plan and viewing a steroid injection as, at most, a short bridge to get you loading, not a cure. None of this means you should avoid medical care; it means the opposite. Anyone who is pregnant, has diabetes (which independently impairs tendon healing), is over 60, takes corticosteroids, or whose pain followed a fluoroquinolone prescription should map out a loading plan with a clinician rather than relying on a generic protocol — the heel-drop and decline-squat doses validated in trials like Alfredson's were tested in otherwise healthy adults, not in these higher-risk groups Alfredson 1998.

References