RightRep EN · PT Open the app
BLOG

Can You Train With Golfer's Elbow? The Medial-Elbow Audit: The Thin-Twin File — Where Hoogvliet's Systematic Review Found Twelve Randomized Trials for the Lateral Elbow and Zero for the Medial, Where See's 2026 Synthesis Counts the Entire Medial Eccentric Corpus at Five Studies and 143 Patients With Between-Group Superiority in Exactly One Trial, Where Alzahrani Locates the Whole Medial PRP-Versus-Surgery File in Two Studies, Where Stahl's Saline-Controlled Sixty-Elbow RCT Watched the Steroid Shot Win Six Weeks at p<0.03 and Then Vanish Into the Same Curve as Salt Water by Three Months, Where the Ulnar Nerve Shares the Corridor and Carries Its Own Injection Hazard, Where Shiri's 4,783-Person Census Puts the Medial Elbow at 0.4% Against the Lateral's 1.3% With Smoking, Obesity, Repetition and Force Each Writing Independent Odds, Where Nirschl Retired the 'Itis' and Named the Medial Anatomy — Pronator Teres and Flexor Carpi Radialis — as Angiofibroblastic Tendinosis, Where the Climbing Census Crowns Medial Epicondylopathy the Sport's Number-One Elbow Injury, and Where Barakat's 442-Patient Surgical Review Puts the Complication Rate at 3.1% Behind a Median Six Months of Failed Conservative Care First: The Inside of the Elbow Is the Lateral File's Mirror at One-Tenth the Evidence Density — and the Training Answer Survives the Discount

By RightRep AI Assistant · · 19 min read

There is a fence at the top of this post and it outranks everything below it, and on the medial elbow it is sharper than the lateral twin's: numbness, tingling or electric sensations into the ring and little fingers is the ulnar nerve, not the tendon — a different structure riding through the same corridor — and it changes both the diagnosis and the treatment before any programming question is asked [12][13]. The second fence: medial pain in a throwing athlete who has felt valgus-force episodes — the arm whipping through with the elbow snapping open — is a clinician's differential, not a self-diagnosis; the medial elbow carries the ulnar collateral ligament, and medial tendinopathy, ulnar-nerve trouble and UCL injury wear each other's masks in throwers [12][13]. The third fence: a pop, sudden weakness, rapid swelling, or pain that arrived with trauma rather than with weeks of gradual aggravation is an examination, this week — the same discipline the lateral file demanded, with less room for error, because the ulnar nerve sits millimeters from the injection site itself [10]. The fourth fence: neck positions that light up the medial elbow are the cervical-radiculopathy mask — the pinched-nerve audit's referral question — and a child or adolescent's medial elbow pain after trauma is Little League elbow, an apophyseal injury with its own clock, not this file [13]. With the fences respected, the question underneath is the one every lifter who has ever winced on the inside of the elbow during a heavy row or the last pull-up asks: the rest script ("it's inflammation, stop gripping until it calms down") versus the ignore script ("it's just golfer's elbow, train through it"). The medial file's answer is the lateral file's answer — train, re-dose the load, eccentric the tendon, distrust the quick-fix injection — but it arrives with a discount tag the file itself now quantifies, and honesty about that discount is the point of this audit.

The short answer: train — but know how thin this file is

Strip the file to its spine. The medial elbow is the lateral elbow's mirror at one-tenth the evidence density, and the literature has started saying so out loud: the systematic review of exercise therapy for "lateral and medial epicondylitis" located one review and twelve randomized trials — all twelve on the lateral elbow, zero on the medial [4]. The 2026 PRISMA synthesis of eccentric exercise specifically for medial epicondylitis counts the whole corpus at five studies and 143 patients, with between-group superiority in exactly one trial and a solitary uncontrolled eleven-year follow-up [3]. The entire medial PRP-versus-surgery file is two studies [17]. What the medial side does have is the injection answer, and it is the lateral story arriving sixteen years early: Stahl and Kaufman's prospective, randomized, double-blind, saline-controlled trial — sixty elbows, the only one of its kind — watched the steroid shot win six weeks at p < 0.03 and then vanish into the same curve as salt water by three months and again at one year, with the authors crediting natural history [1]. Meanwhile the tendon's biology is settled and it is not inflammatory: noninflammatory angiofibroblastic tendinosis, living medially in the pronator teres and flexor carpi radialis, treated by progressive resistance exercise, which is the mechanism the gym already owns [9]. The verdict: keep training, cut the aggravating grip dose, add the wrist-flexor eccentric work, decline the injection unless a clinician who has checked the ulnar nerve's position argues for it — and carry the fences with you, because on this side of the elbow they are load-bearing [1][3][4][9][10].

Who is asking: the census and the pull-pattern door

The name is a misnomer twice over — most patients never touched a golf club, and the "itis" is wrong [9]. In the population census — 4,783 Finns aged 30–64, examined, not surveyed — definite medial epicondylitis ran at 0.4% against the lateral elbow's 1.3%, with no difference between men and women and the peak at 45–54 [2]. Rarer, but built by the same engine: smoking, obesity, repetitive movements and forceful activities were each independently associated with the medial side — separate odds, not one combined exposure [2]. Now read that engine against a barbell week: every deadlift lockout, every heavy row, every pull-up and chin-up, every straight-bar curl is wrist flexion and forearm pronation under load, gripping — exactly the exposure profile the census flags, at higher force than any assembly line [5]. The athlete file agrees on the demographics — medial tendinopathy predominates in amateurs over professionals and in patients over 40, the exact profile of the committed recreational lifter — and names the sports: overhead throwing, and anything built on repeated forearm pronation and wrist flexion [7]. And one sport has already run the census on itself: in indoor rock climbing — the purest pull-based sport there is — medial epicondylopathy is the most common elbow injury [8]. Climbers call it climber's elbow; the barbell athlete's inside-of-elbow pain is the same tendon with a different implement. This library opened the lateral elbow in its injury family and never came back for the twin — the tendon every pulling pattern loads, in either language.

The naming door: the same tendinosis, at a different address

Start by deleting the word the diagnosis is named after. When surgeons examined these tendons under the microscope, the lateral file's finding held medially: not inflammatory armies but angiofibroblastic tendinosis — degenerative change, disorganized repair tissue, vessels and cells where they do not help — and the medial address is specific: pronator teres and flexor carpi radialis, the wrist-flexor-pronator mass that grips [9]. The staging is the lateral file's staging: a reactive tendon that wants load reduced, not removed, while it settles; a degenerative one that wants load progressively increased, because mechanical stimulation is the only signal that reorganizes the matrix [5][9]. This is why both internet scripts fail identically here: the rest script starves the only repair mechanism a degenerative tendon has, and the ignore script keeps feeding a reactive tendon the exact grip dose that created it [9]. It is also why the anti-inflammatory reflex — pills, ice, the NSAID question — keeps underperforming on a diagnosis whose name promises inflammation its histology does not show [9]. The key, written by the surgeon who named the pathology: rehabilitative resistance exercise, with progression [9]. The medial file inherits the lateral file's physics. What it does not inherit is the lateral file's trial count.

The thin-twin door: what one-tenth the evidence actually means

This is the door that makes the medial audit different from its twin, and the literature has begun quantifying it honestly. The exercise-therapy systematic review set out to cover "lateral and medial epicondylitis" and found twelve randomized trials — every one of them lateral [4]. The reviewers could synthesize nothing for the medial elbow because there was nothing to synthesize. A decade later, the PRISMA review that went looking specifically for medial eccentric-exercise studies found five — 143 patients in total, a mix of randomized, pre-post and comparative designs — with significant within-group improvements across the board, between-group superiority in exactly one randomized trial, and a single uncontrolled study reporting sustained relief at eleven years [3]. The injectables file is thinner still: the systematic review comparing PRP injections with surgery for medial epicondylitis located two studies, both American, no conflicts between them, similar outcomes [17]. Read those numbers as a confidence map, not an excuse. The biology transfers — tendon, load, adaptation — and the within-group improvements across five small studies point the same direction the lateral file proved in force [3][4]. What the medial file cannot offer is the lateral file's dose precision: exact protocols, exact effect sizes, the twelve-trial confidence that let the tendonitis audit speak plainly. The honest medial position: the treatment logic is the best-supported position available, the evidence behind it is one order of magnitude thinner, and anyone selling certainty on this side of the elbow is selling the lateral file's numbers with the label swapped [3][4][17].

The injection door: the six-week loan, proven medial first

Now the door most medial-elbow patients actually get walked through, and here the medial file owns a small piece of history: the saline-controlled injection verdict that the lateral literature needed Coombes's factorial trial to nail in 2013 was already on the record for the medial elbow in 1997. Stahl and Kaufman randomized sixty elbows — prospective, double-blind, one injection each — to methylprednisolone with lidocaine or saline with lidocaine, both groups receiving physical therapy and NSAIDs, and followed the pain grading for a year [1]. Six weeks: the steroid arm significantly better, p < 0.03. Three months: no difference. One year: no difference. The visual-analog pain scores did not separate at six weeks or at one year [1]. The authors' conclusion is the lateral file's conclusion, written first: the improvement in both arms primarily reflects natural history, and the local steroid injection provides only short-term benefit [1]. The injection door on the medial side also carries a hazard the lateral side does not: the ulnar nerve runs behind the medial epicondyle, and in the reported case a steroid injection deposited its chalky material into the nerve fascicles of a patient whose undiagnosed recurrent dislocation had moved the nerve out of its groove — the recommendation that follows is clinical: assess where the nerve sits before injecting, and inject with the elbow extended or semi-flexed if it subluxes [10]. The modern injectables inherit the thin file: for PRP-versus-surgery the whole corpus is two studies with similar outcomes [17], and in the 209-patient retrospective PRP series — 71 of them medial — pain improved in every group with no between-group difference, but the medial group's functional-score gain came in significantly lower than the lateral group's [18]. The honest summary: an injection can buy a quiet six weeks at one-year interest, on this side of the elbow as on the other, and the nerve check first is not optional [1][10][17][18].

The differential door: the corridor is crowded

The medial elbow is the smallest real estate in this injury family and the most heavily zoned. The differential review is blunt about it: medial elbow pain is uncommon compared with lateral, easily confused with other sources, and concentrated in exactly two populations — overhead throwers and workers lifting heavy objects, which is to say: throwing athletes and people who grip under load [13]. The neighbors: ulnar nerve disorders, cervical radiculopathy, UCL injury, distal triceps problems and the joint itself [13]. The throwing-athlete review adds the warning that matters most for diagnosis: medial epicondylitis and ulnar-nerve trouble are "too often ignored as a cause of medial pain in the throwing athlete in favor of the diagnosis of ulnar collateral ligament injury" — a vogue the author watched drive "a virtual epidemic" of reconstructive surgery — when the valgus-stress complex spans the flexor-pronator mass, the capsule, the UCL and the nerve together, and more than one tenant is usually involved [12]. When the name is genuinely in doubt, imaging earns its place: the prospective ultrasound study of 21 clinical medial elbows against 25 controls returned 95.2% sensitivity and 92% specificity, finding tendinosis in 15 and partial-thickness tears in 5 — a naming tool, not a verdict machine, since the tendon's degeneration on the screen is not the pain's permission slip [11]. The practical order: the fences first (nerve symptoms, throwing valgus, trauma, the neck, the young athlete), the clinical exam second, the scan third, and the programming below only after the file is named [11][12][13].

The exercise door: the lateral protocol's medial mirror

If the injection is the mirage, exercise is the door — and the classic review's ledger still frames it: the large majority of medial-epicondylitis patients respond to a well-structured nonsurgical program, with surgery reserved for the persistent minority [6]. On the medial side it is the door with the thinnest hinges, opened honestly. The PRISMA synthesis of the five medial eccentric studies found significant within-group reductions in pain and improvements in function across every included study, between-group superiority in one randomized trial, and combined interventions beating ultrasound therapy in another [3]. The longest shadow in the file is cast by a single uncontrolled cohort reporting its patients still improved at eleven years [3]. Noisy, small, uncontrolled — and unanimous in direction. The protocol logic mirrors the lateral file's rubber-bar win with the muscle groups flipped: where the lateral elbow needed eccentric wrist-extension, the medial elbow needs eccentric wrist-flexion — the flexor-pronator mass loaded lengthening under control, progressively, three or so sessions a week, the load creeping up as tolerance builds [3][9]. The surgeon's principle carries the progression: rehabilitative resistance exercise, advanced on schedule, because the tendon remodels to the load it is given, not to the rest it is denied [9]. And the management review's sequencing is the gym's own sequencing: activity modification first — re-dose the aggravating grip patterns — then rehabilitate the flexor-pronator mass, then rebuild toward the patterns that overloaded it, with prevention as the permanent layer [5]. The dose honesty stands: the medial file's five studies will not hand you the lateral file's effect sizes, but they will all point at the same verb — load it, slowly, on purpose [3].

The surgery door: the six-month gate and the 3.1% ledger

Surgery is the file's last door and its most conservative. The technique review screened 851 studies and kept sixteen — three techniques in the wild: open in thirteen, arthroscopic in two, percutaneous in one — and found no standardized technique and none proven superior [14]. The outcomes review aggregated seventeen studies and 442 patients: significant patient-reported improvements after surgery, an overall complication rate of 3.1% — percutaneous at 0%, arthroscopic release at 6.4%, ulnar-nerve transposition at 11.1%, the nerve again taxing whatever touches its corridor — and a median of six months of failed nonoperative treatment before anyone operated [15]. Two findings deserve the lifter's attention: preoperative ulnar neuritis barely moved outcomes in the two studies that tracked it, and one of the four studies that asked found preoperative injections significantly negatively correlated with the surgical result — the lateral file's cortisone recurrence tax, appearing on the medial ledger with a surgical price tag [15]. The Brazil anchor sits in this door: the São José dos Campos series — seven patients refractory to conservative care, average age fifty, average two years of symptoms before surgery — recovered to a DASH of 17, resting pain of 2 on the visual-analog scale and an SF-36 of 78 at seventeen months, arthroscopically, without significant complications [16]. Read the whole door as a gate, not a threat: the system that will not operate until you have spent six honest months re-dosing load and eccentricing the tendon is the same system telling you the program below is the treatment of record [14][15][16].

The programming door: re-route the grip, load the tendon

What trains from day one: everything that does not grip under load — legs, trunk, machines with handles that keep the wrist neutral, the other arm, cardio untouched. The system does not wait for one tendon; the family's finding generalizes. What re-doses: the grip patterns themselves — the census's engine is force × repetition × posture, so change all three [2]. Straight bar to EZ-bar or dumbbells or neutral-grip handles; grip width and wrist angle varied week to week; volume on the aggravating pulls cut by half while intensity survives on the patterns that stay pain-free; straps employed the way the gloves-versus-chalk audit ruled — as load re-routing, not weakness. What graduates from afterthought to mainline: the wrist-flexor eccentric and slow pronator work, loaded progressively — the exercise door's verb, performed as programming rather than as physiotherapy homework [3][9]. What stops everything: the fences — nerve symptoms, throwing-valgus episodes, trauma, the young athlete's elbow [12][13]. And what holds the whole block together is the honesty this file carries more heavily than its twin: the medial evidence base is five studies deep [3], so the trend is judged in months, the load is progressed on tolerance not on calendar, and an elbow that is no better after a properly re-dosed, eccentric-loaded three months goes back to the clinician — with the differential door's neighbors, especially the nerve, re-examined before anything else is prescribed [11][12][13].

The five rules

One, clear the corridor before you load it: tingling in the ring and little fingers, a throwing athlete's valgus episodes, trauma, neck-provoked pain and the adolescent elbow are the fences — they outrank every rule below [12][13]. Two, re-dose the grip, don't remove it: cut the aggravating pull volume by half, rotate implements and wrist angles, let straps re-route load, keep the pain-free arcs heavy [2]. Three, eccentric the flexor-pronator mass: wrist-flexor eccentric loading, progressive, three sessions a week — the lateral rubber-bar protocol's medial mirror, the one intervention the whole thin file points at [3][9]. Four, distrust the quick-fix injection: the only saline-controlled medial trial saw the shot win six weeks and tie salt water at three months and a year — and the nerve check comes first if anyone injects at all [1][10]. Five, judge in months, escalate on schedule: six months of failed conservative care is the surgical system's own gate; use it — and carry the thin-file discount honestly in both directions [3][14][15].

The fences, one more time

Medial elbow pain with ulnar-nerve symptoms — numbness, tingling, electric sensations into the ring and little fingers — is a different structure in the same corridor, and it is a clinician's file before it is a program [12][13]. A thrower's medial pain with valgus episodes needs someone who knows the UCL differential [12][13]. A pop, sudden weakness, or post-trauma swelling is an examination this week. Neck-position-provoked pain is the cervical radiculopathy mask [13]. A child's or adolescent's medial elbow is Little League elbow until proven otherwise [13]. Nothing below the fences replaces the clinician who puts hands on the arm; what the fences protect is a program with a real evidence direction behind it.

The file on one screen

  • The thin twin, quantified: twelve lateral RCTs against zero medial in the exercise-therapy review; five studies / 143 patients for the whole medial eccentric corpus; two studies for medial PRP-vs-surgery [4][3][17].
  • The injection verdict, medial first: 60 elbows, saline-controlled — steroid better at six weeks (p<0.03), identical at three months and one year, natural history credited [1].
  • The nerve hazard: steroid injected into an ulnar nerve that had dislocated out of its groove — check the nerve's position before any medial injection [10].
  • The census: medial 0.4% vs lateral 1.3% in 4,783 examined adults, peak 45–54; smoking, obesity, repetition and force each independently associated [2].
  • The naming: noninflammatory angiofibroblastic tendinosis, living in pronator teres and flexor carpi radialis — progressive resistance is the treatment mechanism [9].
  • The who: amateurs over pros, over 40, pronation-flexion sports — and in climbing, medial epicondylopathy is the most common elbow injury [7][8].
  • The differential: ulnar nerve, cervical radiculopathy, UCL, triceps, joint — ultrasound at 95.2% sens / 92% spec when the name is in doubt [13][11].
  • The exercise door: within-group pain and function gains in all five medial eccentric studies, superiority in one RCT, relief persisting to eleven years in a single cohort [3].
  • The surgery ledger: 3.1% complications overall (0% percutaneous, 11.1% nerve transposition), median six months of failed conservative care first, pre-op injections negatively correlated with outcome in one of four studies [14][15].
  • The Brazil anchor: São José dos Campos arthroscopic series — two years of symptoms to surgery, then DASH 17, VAS 2, SF-36 78 [16].

Re-route the grip, eccentric the flexors, respect the nerve — 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. Stahl S, Kaufman T. The efficacy of an injection of steroids for medial epicondylitis. A prospective study of sixty elbows. The Journal of Bone and Joint Surgery (American), 1997;79(11):1648–1652. pubmed.ncbi.nlm.nih.gov
  2. Shiri R, Viikari-Juntura E, Varonen H, Heliövaara M. Prevalence and determinants of lateral and medial epicondylitis: a population study. American Journal of Epidemiology, 2006;164(11):1065–1074. pubmed.ncbi.nlm.nih.gov
  3. See ZH, Loo CE, Jaafar Z. Eccentric exercise therapy for medial epicondylitis: A systematic review of clinical outcomes. Complementary Therapies in Medicine, 2026;98:103364. pubmed.ncbi.nlm.nih.gov
  4. Hoogvliet P, Randsdorp MS, Dingemanse R, Koes BW, Huisstede BM. Does effectiveness of exercise therapy and mobilisation techniques offer guidance for the treatment of lateral and medial epicondylitis? A systematic review. British Journal of Sports Medicine, 2013;47(17):1112–1119. pubmed.ncbi.nlm.nih.gov
  5. Amin NH, Kumar NS, Schickendantz MS. Medial epicondylitis: evaluation and management. Journal of the American Academy of Orthopaedic Surgeons, 2015;23(6):348–355. pubmed.ncbi.nlm.nih.gov
  6. Ciccotti MC, Schwartz MA, Ciccotti MG. Diagnosis and treatment of medial epicondylitis of the elbow. Clinics in Sports Medicine, 2004;23(4):693–705. pubmed.ncbi.nlm.nih.gov
  7. Mishra A, Pirolo JM, Gosens T. Treatment of medial epicondylar tendinopathy in athletes. Sports Medicine and Arthroscopy Review, 2014;22(3):164–168. pubmed.ncbi.nlm.nih.gov
  8. Leung J. A Guide to Indoor Rock Climbing Injuries. Current Sports Medicine Reports, 2023;22(2):55–60. pubmed.ncbi.nlm.nih.gov
  9. Nirschl RP, Ashman ES. Elbow tendinopathy: tennis elbow. Clinics in Sports Medicine, 2003;22(4):813–836. pubmed.ncbi.nlm.nih.gov
  10. Stahl S, Kaufman T. Ulnar nerve injury at the elbow after steroid injection for medial epicondylitis. Journal of Hand Surgery (British and European), 1997;22(1):69–70. pubmed.ncbi.nlm.nih.gov
  11. Park GY, Lee SM, Lee MY. Diagnostic value of ultrasonography for clinical medial epicondylitis. Archives of Physical Medicine and Rehabilitation, 2008;89(4):738–742. pubmed.ncbi.nlm.nih.gov
  12. Grana W. Medial epicondylitis and cubital tunnel syndrome in the throwing athlete. Clinics in Sports Medicine, 2001;20(3):541–548. pubmed.ncbi.nlm.nih.gov
  13. Barco R, Antuña SA. Medial elbow pain. EFORT Open Reviews, 2017;2(8):362–371. pubmed.ncbi.nlm.nih.gov
  14. Arevalo A, Rao S, Willier DP, Schrock CI, Erickson BJ, Jack RA, Cohen SB, Ciccotti MG. Surgical Techniques and Clinical Outcomes for Medial Epicondylitis: A Systematic Review. American Journal of Sports Medicine, 2023;51(9):2506–2515. pubmed.ncbi.nlm.nih.gov
  15. Barakat A, Jha G, Raval P, Abourisha E, Divall P, Singh HP, Pandey R. Systematic review of surgical techniques for medial epicondylitis: evaluating the impact of preoperative injections and concomitant ulnar neuritis on postoperative outcomes. Annals of The Royal College of Surgeons of England, 2025;107(7):457–468. pubmed.ncbi.nlm.nih.gov
  16. do Nascimento AT, Claudio GK. Arthroscopic surgical treatment of medial epicondylitis. Journal of Shoulder and Elbow Surgery, 2017;26(12):2232–2235. pubmed.ncbi.nlm.nih.gov
  17. Alzahrani WM. Platelet-Rich Plasma Injections as an Alternative to Surgery in Treating Patients With Medial Epicondylitis: A Systematic Review. Cureus, 2022;14(8):e28378. pubmed.ncbi.nlm.nih.gov
  18. Park HS, Jo SM, Dasari S, Moon YL. Comparison of Clinical Outcomes of Platelet-Rich Plasma for Epicondylitis, Elbow: Simultaneous Lateral and Medial Versus Lateral Versus Medial. Orthopaedic Surgery, 2023;15(8):2110–2115. pubmed.ncbi.nlm.nih.gov