RightRep EN · PT Open the app
BLOG

Can You Train With Tennis Elbow? The Overuse Audit: Where Smidt's Lancet Three-Arm Watched the Cortisone Shot Win 92% Success at Six Weeks and Then Lose the Year at 69% Against Physiotherapy's 91%, Where Coombes's JAMA Factorial Inverted the Four-Week Win Into 54% Recurrence Against 12% on a Placebo Injection, Where Bisset Counted 47 of 65 Steroid Successes Quietly Regressing, Where the Lancet Meta-Analysis Saw the Injection's Effect Sign Flip Itself From SMD 1.44 to −0.31 as the Calendar Turned, Where Krogh's Saline-Controlled Trial Found Platelet-Rich Plasma No Better Than Salt Water, Where the Brazilian Society of Shoulder and Elbow Surgery Read the Same Injector File in Portuguese, Where Tyler's Rubber-Bar Eccentric RCT Improved DASH 76% vs 13% and Pain 81% vs 22%, Where Cook and Purdam Retired the Word 'Inflammation' and Made Load the Dose and Rest the Wrong Tool, Where Haahr's 266 Fresh Cases Improved 83% Within a Year Whatever the Treatment While Physical Strain at Work Wrote the Odds at OR 8.5, and Where Shiri's Census Puts the Peak at 40–60 With Force Times Repetition Times Posture as the Engine: the Lateral Elbow Is the One Injury Lifting Itself Feeds — and the One It Treats

By RightRep AI Assistant · · 16 min read

There is a fence at the top of this post and it outranks everything below it: an elbow that won't extend, a joint that locked or popped with a sudden trauma and is now swelling fast, pain that arrived with a fall or a snap rather than with weeks of gradual aggravation, or elbow pain traveling with neck pain, numbness or tingling down the forearm and fingers is a clinician-this-week file — some of it is a same-day file — and none of it is a programming question. The second fence: this audit covers the overuse file — the lateral elbow tendinopathy that grip-heavy loading builds over weeks and months. A suspected partial tendon tear, an elbow that gives way mid-rep, or pain after an actual accident belongs in the examination room before it belongs in any block below [2]. The third fence: what follows is training-through evidence for diagnosed or unmistakable lateral elbow pain; if you have never been examined and the pain is not clearly on the outside of the elbow, get it named first — the same discipline the rotator-cuff audit demanded of the shoulder. With the fences respected, the question underneath is the one every lifter eventually asks, usually while holding a straight bar: the rest script ("it's inflammation, stop lifting until it calms down") versus the ignore script ("it's just tennis elbow, train through it"). For once in this injury family, the evidence contradicts the rest script far more than the ignore script — because this is the one injury in the family that training itself both causes and cures.

The short answer: train — re-dose the load, don't remove it

Strip the file to its spine. The treatment with the best six-week numbers in tennis elbow history is the one with the worst one-year numbers: in the Dutch Lancet three-arm trial, corticosteroid injections hit 92% success at six weeks against physiotherapy's 47% and wait-and-see's 32% — and by 52 weeks the same injection arm sat at 69% while physiotherapy reached 91% and even doing almost nothing reached 83% [1]. The recurrence behind that reversal is not subtle: the JAMA factorial trial counted 54% recurrence within a year in the corticosteroid arm against 12% on a placebo injection, with one-year complete-recovery-or-much-improvement of 83% versus 96% [2]. The injection's advantage is a mirage with a date on it, and the meta-analysis watched the sign flip: SMD 1.44 in favor of corticosteroid at short term, −0.40 at intermediate, −0.31 at one year — the shot is, statistically, worse than leaving it alone by the anniversary [3]. Meanwhile the cheapest tool in the file is an eccentric wrist-extensor exercise with a rubber bar: adding it to standard care improved DASH 76% versus 13% and pain 81% versus 22% in seven weeks [4]. And the prognosis file says the elbow mostly gets better anyway — 83% of 266 fresh cases improved within a year regardless of which treatment they received, while high physical strain at work multiplied the odds of a bad year 8.5-fold [5]. The verdict writes itself: keep training, cut the aggravating dose, add the eccentric work, decline the quick-fix injection unless a clinician argues for it — and let the calendar, not the cortisone, do the healing.

Who is asking: the scale file

This is not a tennis injury; the name is a nineteenth-century misnomer that stuck. The occupational census puts lateral epicondylitis at roughly 1.0–1.3% of men and 1.1–4.0% of women in the general population, peaking between 40 and 60 — and the engine, study after study, is forceful activity, especially high force combined with high repetition or awkward wrist posture [6]. General-practice consultations run 0.3–1.1 per 100 patients per year [6]. Look at that risk engine and then look at a barbell session: grip under load, wrist extension under load, hundreds of reps a week, straight-bar curls and heavy rows and pull-ups and the deadlift lockout — the gym recreates the exposure profile of an assembly line, at higher force. This library has circled the elbow without ever opening its most common file: the generic tendonitis audit asked whether exercise helps tendinopathy in general, the bench-and-shoulders audit counted lifters' injury rates, the lockout post even passed through the elbow on its way to an answer about joint extension — but the lateral elbow itself, the overuse injury a gripping hand feeds with every straight bar it holds, was never audited. In either language. The family needed its fifth door: after the tear, the tenant and the disc, the overuse file — the one where the injury and the treatment are the same verb, loaded differently.

What the name isn't telling you: the continuum file

Start by deleting the word the diagnosis is named after. Lateral "epicondyl-itis" implies inflamed tissue; when surgeons biopsied these tendons, they found degenerative change, not inflammatory armies — which is one reason anti-inflammatory everything (pills, ice, the NSAID question) keeps underperforming here. The model that replaced inflammation is the continuum: a reactive tendon (acute overload, swelling, cells waking up — young or newly-loaded tissue, still reversible), tendon dysrepair (failed healing architecture trying to adapt), and degenerative change (the chronic gym file: disorganized matrix, cell death, vessels and nerves growing where they don't help) [7]. The stage names sound academic until you see they are literally dose instructions: a reactive elbow wants load reduced, not removed, for days-to-weeks while it settles; a degenerative elbow wants load progressively increased — mechanical stimulation is the only signal that reorganizes the matrix, and rest merely preserves the mess [7]. This is why both internet scripts fail: the rest script treats a degenerative tendon like a reactive one and starves the only repair mechanism it has; the ignore script keeps feeding a reactive tendon the exact dose that created it. The audit's job is the middle path: identify where on the continuum the elbow sits, and prescribe the load, because — as the rotator-cuff and disc audits found in their own joints — for the chronic end of these files, loading is not the risk. It is the treatment.

The injection door: the six-week mirage

Now the door most elbow patients actually get walked through, and the most instructive bait-and-switch in sports medicine. The Dutch trial randomized 185 patients with at least six weeks of symptoms to corticosteroid injections, physiotherapy, or wait-and-see [1]. At six weeks the shot was untouchable: 92% success versus 47% and 32%, better on every measure [1]. At 52 weeks the podium had inverted — 69% versus 91% and 83% — with long-term differences significantly favoring physiotherapy and a high recurrence rate dragging the injection arm down [1]. The Australian trial repeated the experiment with 198 patients and kept score of the regression: the injection arm's early successes failed at a rate of 47 out of 65, the group did significantly worse than physiotherapy in the mid-to-long term, and by a year most participants in every arm reported success [8]. The meta-analysis of 41 trials and 2,672 tendons put the whole phenomenon in one line: short-term SMD 1.44 in favor of injection, intermediate-term −0.40, long-term −0.31 — with one tendon rupture among 991 injected patients, so the harm is mostly the recurrence and the delayed recovery, not the needle [3].

Then came the trial that closed the argument by making it cleaner: a 2×2 factorial in 165 patients — real injection versus placebo injection, each with and without physiotherapy [2]. At one year, the corticosteroid arm reported complete recovery or much improvement in 83% versus 96% on placebo and recurred in 54% versus 12%; at 26 weeks it was 55% versus 85% [2]. Read that comparison slowly, because it is the most placebo-controlled sentence in the tendinopathy literature: the injection did not merely fail to add benefit over doing nothing — it did worse than a sham injection, and the people it "cured" in month one had a coin-flip chance of being back where they started by month twelve [2]. The modern injectables do not rescue the door: in the saline-controlled three-arm trial, platelet-rich plasma beat salt water on nothing — pain at three months identical across PRP, saline and glucocorticoid arms (PRP versus saline difference −2.7 points, CI −8.8 to 3.5) — while glucocorticoid did briefly win at one month and even quieted the ultrasound (less Doppler flow, thinner tendon) without buying any three-month pain relief [9]. And the injector file keeps being read honestly in Brazil: the Brazilian Society of Shoulder and Elbow Surgery's evidence-based review of hyaluronic-acid infiltration found three usable studies of very low certainty, concluding corticosteroids for short-term control and hyaluronic acid possibly better from eight weeks to six months — a "maybe, at low confidence" that itself outranks most injection marketing [10]. The honest summary of the entire door: an injection can buy a quiet month, at the price of a more turbulent year [1][2][3].

The exercise door: the same verb, re-dosed

If the injection is the mirage, the exercise door is the slow, boring, winning door — and its best trial is gym-native. Tyler's team took 21 chronic patients and randomized them to standard care versus standard care plus one exercise: an isolated eccentric wrist-extensor loading pattern using an inexpensive rubber bar, progressed over roughly seven weeks [4]. Every outcome separated: DASH function improved 76% versus 13% (P=.01), pain 81% versus 22% (P=.002), tenderness 71% versus 5% (P=.003), wrist and finger-extension strength 79% versus 15% (P=.011) [4]. The mechanism is the continuum made practical: the tendon that grip-heavy concentric, high-volume, awkward-angle loading injured responds to eccentric, progressive, controlled loading by reorganizing [4][7]. The physiotherapy arms of the big trials add the rest of the story: the Australian eight-session package (manual therapy plus exercise) beat wait-and-see over the first six weeks and beat the injection over everything after, with the physiotherapy group also reaching less for extra treatments like NSAIDs [8]. And the honesty clause the anti-injection camp never quotes: in the factorial trial, physiotherapy added nothing significant at one year (91% versus 88% global improvement, recurrence 29% versus 38%) — because the natural history of this condition is already 83% improvement in a year, and the control arms were improving too [2][5]. Exercise is not magic against tennis elbow; it is the fastest and most reliable route through a condition that was going to improve anyway — minus the injection's recurrence tax — and it is the route that leaves the tendon loaded, adapted and ready for the bar [1][4][8].

The prognosis file: load exposure writes the odds

The quietest trial in this file may be the most useful one for a gym. Danish occupational specialists randomized 266 new lateral epicondylitis cases to minimal intervention — information, encouragement to stay active, graded self-performed exercises — versus whatever general practice usually does, and followed everyone for a year [5]. Two findings matter. First, 83% improved within the year, and the type of treatment given or chosen had no relation to the prognosis — the condition is substantially self-limiting, which reframes every aggressive quick-fix [5]. Second, what did predict a bad year was exposure: poor overall improvement tracked with manual jobs (OR 3.0), high physical strain at work (OR 8.5) and high baseline pain (OR 2.3), and failure to halve the pain tracked with manual work, high strain, high distress and the dominant elbow (OR 3.1) [5]. For a lifter, that file reads as a programming instruction, not a warning label: the dominant-side elbow that grips every straight bar, every session, at the same width and the same wrist angle is the occupational-repeat-strain patient of the gym — and the fix is dose design, not withdrawal. Rotate implements (straight bar to EZ-bar to dumbbells to neutral-grip handles), vary grip width and wrist position week to week, let the machines file carry volume while the barbell carries intensity, and treat straps and grip changes the way the gloves-versus-chalk audit treated them: as tools that re-route load, not crutches that weaken you. The elbow that keeps meeting the same force, the same repetition and the same posture is the elbow that keeps failing [5][6].

The gear door: the brace is a tool, not a treatment

The gym's reflex purchase — the counterforce strap below the elbow — has its own controlled file, and it is a fair one: 51 patients randomized to kinesiotape, corticosteroid injection, or counterforce brace all improved pain and tendon thickness over four weeks, none of the three beat the others, the brace and the injection (but not the tape) improved function scores, and nothing in the trial improved grip strength [11]. That is the correct reading of the brace: it works something like as well as the things that work modestly, by off-loading the extensor origin at the moment of peak force — a load-management tool for the weeks when you're re-dosing the bar, in the same spirit the knee-sleeve audit granted compression. It is not a cure, and wearing it to keep training through the exact aggravating pattern without changing anything else is the ignore script with a strap on it [11].

The five rules

One, name the file before you load it: lateral elbow pain in a gripping athlete is usually what it looks like, but the fences — trauma, locking, neurological symptoms, an elbow that won't extend — get examined first, and everything below assumes they're clear [2]. Two, re-dose, don't remove: cut volume on the aggravating patterns (usually wrist-extension-under-load: straight-bar curls, heavy rows, pull-ups) by half, swap implements and grips, and keep the pain-free arc of the body training hard — the disc audit's finding generalizes: the rest of the system does not wait for one tendon. Three, eccentric the tendon itself: the rubber-bar protocol or a managed wrist-extensor eccentric, three times a week, progressively loaded — it is the best-controlled exercise result in this file [4]. Four, decline the quick-fix injection unless a clinician makes the case: 54% versus 12% recurrence and 83% versus 96% one-year recovery is the entire argument in two numbers [2][3]. Five, judge the trend in months, not sessions: 83% improve within a year, load exposure writes the odds, and the elbow that is no better after a properly re-dosed, eccentric-loaded three months goes back to the clinician — that referral is the designed escalation, not a failure of the block [5].

The fences, one more time

The overuse file has soft edges, but its borders are hard: sudden post-traumatic elbow pain with swelling or loss of extension is an examination, this week; elbow pain with neck pain, numbness, tingling or weakness into the hand is a different structure's problem wearing a tennis-elbow mask; a joint that locks or gives way is not this diagnosis; and nothing in this post replaces the clinician who puts hands on the arm [2]. What this post does replace is the two scripts: rest is the wrong tool for a degenerative tendon, and the injection is a six-week loan at one-year interest. Load is the cause; load, re-dosed, is the cure.

The file on one screen

  • The mirage: 92% success at six weeks, 69% at a year — while physiotherapy ran 47% then 91% and wait-and-see 32% then 83% [1].
  • The recurrence tax: 54% versus 12% within a year, injection versus placebo; recovery-or-much-improvement 83% versus 96% [2].
  • The sign flip: corticosteroid SMD +1.44 short-term, −0.40 intermediate, −0.31 long-term — the shot is the meta-analytic underdog by the anniversary [3].
  • The regression file: 47 of 65 injection successes relapsed in the Australian trial [8].
  • The PRP null: no better than saline at three months on any measure, while the steroid's ultrasound 'improvement' bought no pain relief [9].
  • The exercise door: rubber-bar eccentric added to standard care — DASH 76% vs 13%, pain 81% vs 22%, strength 79% vs 15% in seven weeks [4].
  • The prognosis: 83% improve within a year regardless of treatment; high physical strain OR 8.5, dominant elbow OR 3.1 — dose design writes the odds [5].
  • The scale: 1.0–1.3% of men, 1.1–4.0% of women, peak 40–60, force × repetition × posture as the engine — the gym's own risk profile [6].
  • The Brazil anchor: the Brazilian Society of Shoulder and Elbow Surgery reads the injector file at very low certainty — steroid for weeks, maybe hyaluronic acid for months [10].
  • The gear door: brace, tape and injection all improved pain similarly at four weeks, none superior, nothing fixed grip strength [11].

Re-dose the load, eccentric the tendon, respect the fences — and let the camera keep the reps honest at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Smidt N, van der Windt DA, Assendelft WJ, Devillé WL, Korthals-de Bos IB, Bouter LM. Corticosteroid injections, physiotherapy, or a wait-and-see policy for lateral epicondylitis: a randomised controlled trial. The Lancet, 2002;359(9307):657–662. pubmed.ncbi.nlm.nih.gov
  2. Coombes BK, Bisset L, Brooks P, Khan A, Vicenzino B. Effect of corticosteroid injection, physiotherapy, or both on clinical outcomes in patients with unilateral lateral epicondylalgia: a randomized controlled trial. JAMA, 2013;309(5):461–469. pubmed.ncbi.nlm.nih.gov
  3. Coombes BK, Bisset L, Vicenzino B. Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials. The Lancet, 2010;376(9754):1751–1767. pubmed.ncbi.nlm.nih.gov
  4. Tyler TF, Thomas GC, Nicholas SJ, McHugh MP. Addition of isolated wrist extensor eccentric exercise to standard treatment for chronic lateral epicondylosis: a prospective randomized trial. Journal of Shoulder and Elbow Surgery, 2010;19(6):917–922. pubmed.ncbi.nlm.nih.gov
  5. Haahr JP, Andersen JH. Prognostic factors in lateral epicondylitis: a randomized trial with one-year follow-up in 266 new cases treated with minimal occupational intervention or the usual approach in general practice. Rheumatology, 2003;42(10):1216–1225. pubmed.ncbi.nlm.nih.gov
  6. Shiri R, Viikari-Juntura E. Lateral and medial epicondylitis: role of occupational factors. Best Practice & Research Clinical Rheumatology, 2011;25(1):43–57. pubmed.ncbi.nlm.nih.gov
  7. Cook JL, Purdam CR. Is tendon pathology a continuum? A pathology model to explain the clinical presentation of load-induced tendinopathy. British Journal of Sports Medicine, 2009;43(6):409–416. pubmed.ncbi.nlm.nih.gov
  8. Bisset L, Beller E, Jull G, Brooks P, Darnell R, Vicenzino B. Mobilisation with movement and exercise, corticosteroid injection, or wait and see for tennis elbow: randomised trial. BMJ, 2006;333(7575):939. pubmed.ncbi.nlm.nih.gov
  9. Krogh TP, Fredberg U, Stengaard-Pedersen K, Christensen R, Jensen P, Ellingsen T. Treatment of lateral epicondylitis with platelet-rich plasma, glucocorticoid, or saline: a randomized, double-blind, placebo-controlled trial. American Journal of Sports Medicine, 2013;41(3):625–635. pubmed.ncbi.nlm.nih.gov
  10. Pinto GMR, Zanesco L, Amaral MVG, Matsunaga FT, Bonadiman JA, de Medeiros JF, Campos MCO, Malavolta EA. Evidence-based reviews and recommendations from the Brazilian Society of Shoulder and Elbow Surgery: hyaluronic acid infiltration for elbow epicondylitis in adults. Acta Ortopédica Brasileira, 2026;34(Spe1):e303257. pubmed.ncbi.nlm.nih.gov
  11. Hoseini SM, Taghipour M, Jokar R, Mostafaloo M, Shirafkan H, Javanshir K. Comparison of kinesiotape, counterforce brace, and corticosteroid injection in patients with tennis elbow: a prospective, randomized, controlled study. PLoS One, 2025;20(7):e0328396. pubmed.ncbi.nlm.nih.gov