Most tendon advice is "do eccentrics" with no numbers attached. The trials specify sets, reps, loads, angles and a rule for when to progress. Those parameters are below, each with the trial it came from.
Findings are grouped by what kind of evidence each one is. That ordering is the point: a sentence from an FDA label and a sentence from a forum are not the same claim, and this page will not present them as though they were.
Patellar tendinopathy isometric stage, fully specified: 5 repetitions of a 45-second single-leg isometric hold at 70% of maximum voluntary contraction, performed at 60 degrees of knee flexion on a leg press or leg extension, daily. This is the most completely reported isometric prescription in a positive randomised trial. [1]
Dose or parameter studied: 5 reps x 45 s hold; 60 degrees knee flexion; 70% MVC; single-leg leg press or leg extension; daily
Studied in: 76 athletes aged 18-35 with ultrasound-confirmed patellar tendinopathy, VISA-P <80, playing sport at least 3x/week; mean age 24, 76% male, median symptom duration 2 years. Predominantly young male jumping athletes — not a general adult population.
The isotonic stage of the same protocol progressed from 4 sets of 15 reps in a restricted mid-range (10-60 degrees knee flexion) to 4 sets of 6 reps at higher load through a fuller range (near-full extension to 90 degrees), performed every second day, alternating with the isometric day. [1]
Dose or parameter studied: 4 sets x 15 reps at 10-60 degrees, progressing to 4 sets x 6 reps at near-full extension to 90 degrees; every second day
Studied in: Same 38 patients randomised to progressive tendon-loading exercise; young athletes with chronic patellar tendinopathy. Range-of-motion restriction early is a deliberate load-limiting device, not an arbitrary detail.
THE NUMERIC PAIN-MONITORING RULE, with its source: progression was gated on a single-leg squat used as a standardised pain provocation test. If that squat produced VAS 3 or less on 0-10, and the current stage had been performed for at least one week, the athlete advanced. This is the ≤3/10 rule that most modern tendon protocols cite. [1]
Dose or parameter studied: Provocation test = one single-leg squat; threshold = VAS ≤3/10; minimum 1 week per stage before advancing
Studied in: Patellar tendinopathy, 18-35, ultrasound-confirmed. The rule was applied to exercise progression AND to return to sport and general athletic activity in the same trial.
The staged, pain-limited protocol beat the traditional pain-provoking eccentric decline squat by 9 VISA-P points at 24 weeks. Notably, the comparator arm was deliberately instructed to work AT VAS 5 or above — so this is a direct test of loading into pain versus loading under a pain ceiling, and the ceiling won. [1]
Dose or parameter studied: Comparator: eccentric decline squat on 25-degree board, twice daily, 12 weeks, instructed to reach VAS ≥5/10 during exercise
Studied in: 76 athletes with mostly chronic patellar tendinopathy (median 2 years), 82% of whom had already failed prior treatment.
The second, older numeric rule (the Silbernagel pain-monitoring model) uses a HIGHER ceiling than the ≤3/10 rule: load is added when pain does not exceed 5/10, and tendon-loading sport is permitted provided pain stays at or below 50/100 mm AND subsides within 24 hours. The 24-hour resolution criterion is the part most often dropped when people quote this rule. [2]
Dose or parameter studied: Add load (5 kg increments) when pain ≤5 on 0-10 NRS; resume loading sport if pain ≤50 mm on 0-100 mm VAS and it subsides within 24 h
Studied in: Protocol description for recreational athletes aged 18-60 with chronic (≥3 months) unilateral midportion Achilles tendinopathy. The rule originates in Achilles work and is applied to patellar and elbow tendons by convention, not by trial.
The evidence that you can keep training on a symptomatic tendon: an RCT randomised patients to continue running and jumping under the pain-monitoring model versus stopping those activities for 6 weeks. Continuing was not worse. This is the trial behind 'you do not have to rest it'. [3]
Dose or parameter studied: Continued running/jumping under pain-monitoring model vs 6 weeks cessation; identical rehab programme in both arms
Studied in: 38 patients with Achilles tendinopathy (19 per arm). Achilles, not patellar or elbow — the transfer to elbow tendons is an assumption, not a finding.
HONEST CAVEAT ON THE PAIN RULE ITSELF: a systematic review of 30 RCTs found pain-based progression criteria dominate the literature but are not backed by strong evidence, and no trial has yet compared the same exercise programme run under different pain thresholds. The ≤3/10 and ≤5/10 numbers are conventions that happened to be used in trials that worked — they are not validated cut-points. [4]
Studied in: 30 RCTs of midportion Achilles, patellar or gluteal tendinopathy with progressive exercise as monotherapy. Lower limb only; elbow tendons were not reviewed.
Heavy slow resistance for the patellar tendon, fully specified: 3 sessions per week; three bilateral movements (squat, leg press, Bulgarian split squat); load stepped 12RM (weeks 1-2) to 10RM (3-4) to 8RM (5-6) to 6RM (7-8); tempo enforced by metronome at 3 s eccentric plus 3 s concentric, 6 s per repetition; 4 sets per movement with 2 minutes rest. [5]
Dose or parameter studied: 3x/week; 12RM→10RM→8RM→6RM over 8 weeks; 3 s eccentric + 3 s concentric (6 s/rep); 4 sets; 2 min rest; >70% 1RM, no more than 12 reps/set
Studied in: 51 college students aged 18-24 with chronic (>3 months) patellar tendinopathy at a Chinese sports university, VISA-P <80, ultrasound-confirmed. Almost entirely male (46/51). The 4 sets and 2 min rest are stated in the combined-therapy paragraph, which explicitly states that group followed the same regimen as the HSR group.
That same HSR trial used a slightly different numeric rule from Breda's: pain under VAS 3 during a session was acceptable provided it did not INCREASE session to session. The between-session trend, not the within-session number, was the stop signal. [5]
Dose or parameter studied: Continue if VAS <3 and pain does not increase in subsequent sessions
Studied in: College-age athletes with chronic patellar tendinopathy under direct therapist supervision every session — a much more supervised setting than self-management.
The foundational HSR trial: 12 weeks of HSR matched eccentric decline squats and beat corticosteroid injection in the long run. Corticosteroid looked good short-term then deteriorated; HSR held its gains at 6 months and produced measurable tendon change (reduced swelling and vascularisation, elevated collagen turnover). [6]
Dose or parameter studied: 12 weeks; HSR vs eccentric decline squat vs peritendinous corticosteroid injection
Studied in: 39 MALE patients with patellar tendinopathy. All male — no female data in this trial at all.
ISOMETRIC ANALGESIA, THE ORIGINAL POSITIVE RESULT: a single bout dropped decline-squat pain from 7.0/10 to 0.17/10 and held the reduction for 45 minutes, with a parallel drop in cortical inhibition. This near-total abolition of pain is the number that made isometrics famous — and it came from six volleyball players. [7]
Dose or parameter studied: Single bout of isometric quadriceps contractions vs isotonic; outcomes at 0 and 45 min
Studied in: SIX volleyball players with patellar tendinopathy, single-blinded crossover. n=6. This is the entire evidence base for the headline effect size.
The in-season isometric protocol with its parameters: leg-extension isometric holds at 60 degrees knee flexion at 80% MVIC, 4 sessions per week for 4 weeks, versus isotonic leg extension at 80% of 8RM with time-under-load matched. Isometrics produced greater immediate analgesia across the 4 weeks. Note the intensity here is 80% MVIC, not the 70% used by Breda. [8]
Dose or parameter studied: Isometric leg extension holds, 60 degrees knee flexion, 80% MVIC, 4x/week for 4 weeks (time under load and inter-set rest matched to isotonic comparator at 80% of 8RM)
Studied in: 20 in-season jumping athletes (18 men, mean 22.5 years) with clinically diagnosed patellar tendinopathy, training/competing 3x/week. Baseline median pain only 5/10 — a milder group than Rio 2015.
REPLICATION FAILURE 1: an independent, pre-registered crossover trial in 21 patients found the pain reduction after isometric exercise was 0.9 points on 0-10 — not 6.8 — was gone by 45 minutes, and was no different from dynamic exercise. The authors concluded contraction mode may not be what matters. [9]
Dose or parameter studied: Single session high-load isometric vs dynamic resistance exercise, 7-day washout; outcomes pre, immediately post, 45 min post
Studied in: 21 participants with patellar tendinopathy, Denmark. Pre-registered, participants blinded to hypothesis — methodologically stronger than the original.
REPLICATION FAILURE 2: from the same research group as the original, a 4-week in-season trial found isometric and isotonic programmes both reduced pain, with no significant difference between them. Both worked; the isometric advantage did not appear. [10]
Dose or parameter studied: 4 isometric or isotonic sessions per week for 4 weeks
Studied in: 29 jumping athletes with patellar tendinopathy playing at least 3x/week, with no modification of training or competition load.
REPLICATION FAILURE 3, the systematic review: across 13 studies and 346 participants with local musculoskeletal pain, there was no consistent evidence that isometric exercise produces hypoalgesia, and no consistent superiority over comparison interventions. This contrasts with healthy people, where the effect IS consistent. Isometrics are well tolerated — that is the defensible claim, not that they reliably kill pain. [11]
Studied in: 346 participants across 13 randomised and crossover trials in people with localised musculoskeletal symptoms, upper and lower limb.
ELBOW, AND THE MOST ACTIONABLE FINDING HERE: isometric wrist extension at 10 x 15 s was tested at an intensity 20% BELOW versus 20% ABOVE each person's own pain threshold. Above threshold, pain was HIGHER immediately after and still higher 30 minutes later. Below threshold, it was no different from doing nothing. Pressure pain threshold and pain-free grip did not improve in either condition. At the elbow, pushing an isometric into pain makes it worse. [12]
Dose or parameter studied: 10 x 15 s isometric wrist extension; load set 20% below or 20% above individual pain threshold
Studied in: 24 individuals with unilateral chronic lateral epicondylalgia, median 3-month duration, middle-aged. Randomised order, blinded examiner.
Eight weeks of daily unsupervised progressive isometric exercise for lateral elbow tendinopathy beat wait-and-see on only one of three primary outcomes. Pain-free grip strength and global rating of change were no better than doing nothing. The authors' own conclusion is that isometrics as a sole treatment are of doubtful efficacy at the elbow. [13]
Dose or parameter studied: Single supervised instruction session, then 8 weeks of daily unsupervised progressive isometric exercise
Studied in: 40 adults with unilateral lateral elbow tendinopathy of at least 6 weeks duration (21 exercise, 19 wait-and-see); 98% completed. General clinical population, not athletes.
For lateral elbow tendinopathy, 12 weeks of home-based heavy slow resistance with elastic bands improved symptoms short and long term, and adding either a corticosteroid injection or tendon needling added nothing. Corticosteroid actually made patient-reported outcomes WORSE at 52 weeks than loading alone. [14]
Dose or parameter studied: 12 weeks home-based HSR with elastic band exercises; follow-up at 12, 26 and 52 weeks (primary endpoint 52 weeks)
Studied in: 60 patients with chronic unilateral lateral elbow tendinopathy, randomised double-blind placebo-controlled. Improvements were ~20 DASH points and ~2.5 NRS points across all arms.
TOLERABILITY WARNING FOR HEAVY ELBOW LOADING: in a feasibility RCT, a 12-week HSR programme (36 sessions; a wrist flexion/extension exercise plus a forearm pronation/supination exercise with a dumbbell) was completed by only 6 of 19 participants — 32% — mainly because it aggravated pain. The investigators concluded HSR was not suitable for this population. Heavy loading that the patellar tendon tolerates, the common extensor origin often does not. [15]
Dose or parameter studied: 12-week home programme, 36 sessions, 2 exercises: wrist extension/flexion, and forearm pronation/supination with a dumbbell; plus extensor stretching 3x/day, 3 sets of 30 s
Studied in: Secondary-care tennis elbow patients at Oslo University Hospital (n=19 in the HSR arm of 60 randomised). Secondary care means a more refractory group than typical gym-goers.
THE HONEST BASELINE FOR DISTAL BICEPS: distal biceps tendinopathy is very rare and there is limited published evidence on it. Everything above about elbow-flexor loading is extrapolation — there is no loading protocol trial for this tendon. The review also flags that tendinopathy here can progress to rupture, which is why it should not be self-managed by simply training through it. [21]
Studied in: Narrative review of medial elbow, distal biceps and distal triceps tendinopathy. Explicitly a scarcity-of-evidence review, not a trial.
DISTAL BICEPS — WHICH MOVEMENTS LOAD ELBOW FLEXORS MOST: in a direct EMG comparison of curl grips at 8RM, biceps brachii excitation was 19% higher with a supinated grip than pronated and 12% higher than neutral; brachioradialis was also highest supinated. So supinated curls load the elbow flexors most; pronated (reverse) and neutral (hammer) curls load biceps brachii least while shifting work to anterior deltoid. [16]
Dose or parameter studied: Bilateral biceps curl, non-exhaustive 6-rep sets at 8RM, supinated vs pronated vs neutral forearm
Studied in: TEN competitive bodybuilders, healthy, surface EMG. No tendinopathy patients; EMG amplitude is a proxy for muscle drive, NOT a measurement of distal biceps tendon force.
COUNTERINTUITIVE FINDING FOR PULL-UPS: across supinated (chin-up), pronated, neutral and rope grips, arm and forearm muscle activation over a complete repetition was SIMILAR — switching to a pronated grip does not meaningfully unload the biceps in a pull-up. What did differ was phase: biceps brachii and brachioradialis activation was significantly HIGHER in the concentric than the eccentric phase, so the lowering portion is where elbow-flexor drive is lowest. [17]
Dose or parameter studied: Bodyweight pull-ups, four grip variations, surface EMG of eight shoulder-arm-forearm muscles normalised to %MVIC
Studied in: 19 strength-trained males, mean age 24.9, healthy. Not tendinopathy patients. Again, EMG amplitude is not tendon load.
Same pattern in the lat pulldown: pronating the grip increased latissimus dorsi activity but had NO effect on biceps brachii, at either wide or narrow grip. Switching from an underhand to an overhand pulldown will not spare the elbow flexors. [18]
Dose or parameter studied: Anterior lat pulldown at 70% 1RM, 5 reps, 2 s concentric and 2 s eccentric cadence; wide (carrying width) and narrow (biacromial) grips, pronated and supinated
Studied in: 12 healthy men. Healthy, no elbow symptoms; single session EMG.
GRIP AIDS, PART 1 — THICK GRIPS DO CHANGE RECRUITMENT, AND IN THE DIRECTION THAT MATTERS FOR A DISTAL BICEPS: adding fat-grip attachments significantly INCREASED forearm muscle activation and significantly DECREASED upper-arm (biceps and triceps) activation during deadlift, bent-over row and pull-ups, while cutting 1RM in every exercise and reducing pull-up reps. Upright row and concentration curl showed no change. [19]
Dose or parameter studied: Fat Gripz attached to Olympic barbell vs bare bar; 1RM testing in deadlift, bent-over row, upright row, concentration curl; max pull-ups
Studied in: 15 resistance-trained men, mean age 22.4, healthy. Acute single-session EMG and 1RM; no tendinopathy patients and no tendon-load measurement.
GRIP AIDS, PART 2 — LIFTING STRAPS: in the only strap EMG trial retrieved, straps changed nothing in the lat pulldown — not 1RM, not reps to failure at 70% 1RM, not latissimus dorsi activation. IMPORTANT LIMITATION: this study measured latissimus dorsi only. It did NOT measure biceps brachii or forearm flexors, so it cannot answer whether straps offload the elbow flexors. That question is untested. [20]
Dose or parameter studied: Lat pulldown, 3 sets to concentric failure at 70% 1RM, 60 s rest, with vs without lifting straps; plus 1RM test
Studied in: 12 resistance-trained men, mean 6.6 years training experience, healthy. Crossover. Only latissimus dorsi EMG reported.
Where a drug on this page has a US label, this is what it was approved for and at what dose. Anything else is off-label — which is not the same as unsafe, but does mean no regulator has reviewed it for that use.
| Drug | Approved for | Approved dose | On-label for this use? |
|---|---|---|---|
| Methylprednisolone acetate (Depo-Medrol) — injected corticosteroid, the only prescription drug appearing in the tendinopathy loading trials retrieved here (used as comparator/adjunct in Kongsgaard 2009 and Couppé 2022) | The Indications section retrieved from DailyMed covers Section A, intramuscular administration, and lists allergic states, dermatologic diseases, endocrine disorders, gastrointestinal diseases and haematologic disorders. Tendinopathy, epico | Verbatim from label: 'The initial dosage of parenterally administered DEPO-MEDROL will vary from 4 to 120 mg, depending on the specific disease entity being treated.' No tendon-specific dose was retri | No |
These are the questions the literature does not answer. They are listed because on this subject the gaps are load-bearing: most of what circulates as settled practice sits in one of them.
Every identifier below was checked to resolve to a real record before publication. Citations retrieved from PubMed and DailyMed.
/* First touch: store whatever brought them here before anything overwrites it. */ (function () { try { if (localStorage.getItem('tl_attr')) return; var p = new URLSearchParams(location.search); var fields = ['utm_source','utm_medium','utm_campaign','utm_content','utm_term','ref','rdt_cid','fbclid','gclid']; var found = {}; fields.forEach(function (f) { var v = p.get(f); if (v) found[f] = String(v).slice(0, 80); }); if (Object.keys(found).length) { found.landedAt = new Date().toISOString(); found.landingPage = location.pathname; localStorage.setItem('tl_attr', JSON.stringify(found)); } } catch (e) {} })();
This calculator does the arithmetic you typed and nothing else. It does not know what is actually in your vial, whether the label is accurate, or anything about you. Confirm the vial strength and the diluent volume on your own label before you draw, and take dosing decisions to a qualified provider.