Medial Epicondylitis (Golfer's Elbow)

Medial epicondylitis, golfer's elbow, is a degenerative tendinopathy of the flexor-pronator origin at the medial epicondyle of the humerus. It is 3-5 times less common than lateral epicondylitis, and despite the name it affects a wide range of people beyond golfers.

At work. Manual labourers such as carpenters and plumbers are at risk, as is anyone whose job involves repetitive gripping and lifting, forceful wrist flexion or prolonged typing at a computer.

In sport. Golf gives the condition its name, but it is not the only culprit:

• Golf, with improper swing mechanics
• Baseball, in pitchers, from valgus overload
• Tennis with a heavy topspin forehand, and other racquet sports with wrist-heavy strokes
• Weightlifting, with heavy deadlifts and rows

Natural history. Most cases, 90%, resolve with conservative treatment within 6-12 months. Cases that have run for more than 12 months are more likely to need intervention. Involvement is bilateral in 10-20%.

The flexor-pronator origin. The common flexor origin arises from the anterior aspect of the medial epicondyle, and its muscles are described in two layers:

• Superficial layer: pronator teres (the origin of its medial head), flexor carpi radialis, palmaris longus (absent in 15% of the population) and flexor carpi ulnaris, which has a dual origin from humeral and ulnar heads
• Deep layer: flexor digitorum superficialis, flexor digitorum profundus, flexor pollicis longus and pronator quadratus

The two that fail. The medial head of pronator teres and flexor carpi radialis are the primary structures involved and the most commonly affected. Pronator teres is the most medial and superficial muscle, which leaves it vulnerable to overuse. FCR has the most robust tendinous origin and bears significant loads during wrist flexion.

The ulnar collateral ligament. The UCL lies deep to the flexor-pronator mass. Its anterior bundle is the primary valgus stabiliser, running from the anteroinferior aspect of the medial epicondyle to the sublime tubercle of the ulna. In overhead athletes, distinguishing a UCL injury from medial epicondylitis is critical.

The ulnar nerve. The nerve runs through the cubital tunnel behind the medial epicondyle and can be compressed at several points on its way past the elbow.

A failed healing response. Medial epicondylitis is a tendinosis, a failed healing response to repetitive microtrauma at the flexor-pronator origin. The dominant histological finding is angiofibroblastic hyperplasia rather than an inflammatory infiltrate.

How the lesion develops. The process runs in three phases:

1. Microtrauma. Repetitive eccentric loading during wrist flexion causes microtears in the tendon substance, and pronation adds torsional stress to the common flexor origin.
2. Failed healing. Angiofibroblastic degeneration replaces normal type I collagen with disorganised type III. Neovascularisation and nerve ingrowth follow, with substance P and CGRP upregulated in the degenerate tissue; the neoinnervation is the source of chronic pain.
3. Chronic tendinosis. Mucoid degeneration leaves grey, friable tissue. The tendon thickens and weakens structurally, loses its normal hierarchical collagen structure, and may tear partially or completely.

Histopathology. Alongside the angiofibroblastic change, the degenerate origin shows:

• Hypercellular fibroblast proliferation
• Ground substance accumulation (mucoid change)
• Disorganised collagen fibres without the normal crimp pattern
• Focal necrosis in severe cases

Nirschl's phases stage the pain, the clinical severity grade adds strength and duration, and the ulnar nerve is graded separately, because the nerve rather than the tendon drives the surgical result.

Gabel-Morrey. Gabel and Morrey (1995) found that the single strongest predictor of the result of flexor-pronator debridement was not the tendinosis but the severity of the coexisting ulnar neuropathy, so they graded medial epicondylitis by its nerve involvement:

• Type I, no or only mild ulnar neuropathy, subdivided into IA (no nerve symptoms) and IB (mild, intermittent symptoms with no motor deficit)
• Type II, moderate-to-severe ulnar neuropathy, with persistent symptoms, a motor deficit and abnormal nerve conduction studies

Why the grade matters. Types IA and IB do very well with debridement, 24 of 25 good or excellent in the original series, whereas type II does poorly, only 2 of 5, a statistically significant difference. A moderate-to-severe ulnar neuropathy must therefore be identified, graded and formally addressed by decompression or transposition, and the patient counselled that the nerve, not the tendon, drives the outcome.

History. Medial elbow pain radiates into the forearm flexors. It aches or burns and is worse with use, particularly gripping, lifting and wrist flexion. The pain first follows activity, then accompanies it, and finally arrives at rest; night pain suggests severe or chronic tendinosis.

What the patient cannot do. Grip weakens, so opening jars, shaking hands and turning doorknobs become difficult. The golf swing, throwing and racquet sports are painful, and at work the keyboard, tools and heavy lifting become difficult.

Examination. Work through it in order:

1. Look. Swelling is rare. Assess the carrying angle and look for muscle atrophy, which appears in chronic cases or with ulnar neuropathy.
2. Feel. Maximum tenderness lies 5-10mm distal and anterior to the medial epicondyle, over the flexor-pronator origin. UCL tenderness is more posterior and distal.
3. Move. Active and passive motion is usually full, with pain at terminal extension as the flexors are stretched. Compare pronation and supination with the other side.
4. Strength. Grip is reduced against the contralateral side, measured objectively with a Jamar dynamometer. Resisted wrist flexion reproduces the pain, and so does resisted pronation (pronator teres).
5. Special tests. The golfer's elbow test is positive. Valgus stress at 30 degrees of flexion should be negative unless the UCL is injured as well. The milking manoeuvre, pulling the thumb with the elbow flexed, applies valgus stress to assess the UCL in overhead athletes.
6. Neurovascular. The ulnar nerve in every patient (below); the median nerve for pronator syndrome, a rare association; and the radial pulse and capillary refill.

Differential diagnosis. The table covers the main alternatives; a flexor-pronator muscle strain should also be considered.

A clinical diagnosis. History and examination, with a positive golfer's elbow test and point tenderness at the flexor origin, are usually sufficient. Imaging is not routinely required, and MRI is not needed for typical medial epicondylitis that is responding to conservative care.

Radiographs. AP and lateral elbow films, taken if the diagnosis is uncertain, are usually normal but may show calcification at the flexor-pronator origin in chronic cases. They rule out bony pathology such as arthritis, loose bodies or fracture, and stress views are added if a UCL injury is suspected.

View larger

MRI is the gold standard, performed with and without contrast. Order it for:

• Diagnostic uncertainty, when medial epicondylitis cannot be distinguished from a UCL injury
• The overhead athlete with medial elbow pain, to assess the UCL
• Failure of 3-6 months of conservative treatment when surgery is being considered, for surgical planning
• Suspected complete tendon rupture, which is rare but needs repair

What MRI shows. On T2 and STIR the flexor-pronator origin has increased signal from oedema and degeneration, with fluid signal if there is a partial tear, peritendinous oedema in acute exacerbations and bone marrow oedema (stress reaction) in chronic cases. On T1 the signal is normal to slightly low, the tendon is thickened and has lost its normal striated appearance, and a partial tear appears as a focal signal abnormality. The same study assesses the integrity of the UCL.

View larger

Ultrasound. Dynamic ultrasound is operator-dependent. It can show tendon thickening, hypoechoic regions of degeneration and neovascularisation on Doppler, and it is useful for guided injection, but it gives less detail of the UCL than MRI.

View larger

Nerve conduction studies. NCS and EMG of the ulnar nerve are indicated by a positive Tinel's sign, a positive elbow flexion test or intrinsic weakness. They show reduced conduction velocity across the cubital tunnel, and denervation in ulnar-innervated muscles if the neuropathy is chronic, and they guide the need for cubital tunnel release.

Conservative treatment comes first and succeeds in most patients. Injection is an adjunct for persistent symptoms, and surgery is for genuine failure after a structured trial.

Algorithm

View larger

Consent. Cover infection, ulnar nerve injury, stiffness, recurrence from incomplete debridement and CRPS (rates under Complications), and a prolonged recovery of 6-12 months.

Equipment. Beyond a standard orthopaedic tray, have ready:

• Upper arm tourniquet (sterile field)
• Rongeur and curette
• Non-absorbable sutures for the tendon repair (FiberWire, Ethibond)
• Suture anchors if bone quality is poor (elderly, osteoporotic)
• Nerve retractors: blunt Hohmann, small Richardson
• Loupe magnification, helpful for the ulnar nerve

Positioning. The patient lies supine with the arm board extended at 90 degrees, the affected arm fully accessible and the head turned away from the operative side. An upper arm tourniquet goes on over padding at 250-300mmHg for adults; the arm may be exsanguinated with an Esmarch or by gravity elevation (avoid if ulnar nerve symptoms). Prep circumferentially from mid-humerus to fingertips and free-drape the arm so the elbow can be flexed and extended during the operation. Confirm that the ulnar nerve is palpable before the incision.

Recurrence. The commonest cause is incomplete debridement, with grey tissue left behind. Next comes early return to activity with reinjury, and third, poor rehabilitation compliance. If conservative treatment fails again, revision surgery involves:

• MRI to assess the extent of the recurrent pathology
• More extensive debridement, which may need a larger exposure
• Ulnar nerve transposition, if it was not done at the index operation
• Counselling on realistic expectations, since the success rate is lower, 70-80%

The balance. Rehabilitation weighs early motion, which prevents elbow contracture, against protection of the repair while the tendon heals to bone. Gentle motion therefore starts early, if the repair is stable, but resisted strengthening waits until 6 weeks. The eccentric exercises that begin then are key to tendon remodelling and to preventing recurrence.

Conservative treatment. Improvement accumulates over a year, and the table shows what drives it at each stage.

Surgical treatment. Good or excellent results are reported in 80-90%, and depend on complete debridement, proper rehabilitation and a conservative trial of 6 months or more. Return to sport or work is achieved in 85-90%, influenced by pre-injury level, a 6-12 month timeline and realistic expectations. Ulnar nerve symptoms improve in 80-90% if the nerve is transposed, depending on preoperative severity, the duration of symptoms and the transposition technique.

Conservative against surgical. Return to activity takes 6-12 months with conservative care, and 6-12 months after surgery. Conservative care is low-cost and its complications minimal, limited to the risks of injection; surgery costs moderate to high and carries complications in 5-10% (nerve, infection, stiffness). Recurrence is 5-10% after successful conservative treatment and 10-15% after surgery.

Where the evidence runs out. Beyond the injection questions under Management, three areas are unsettled:

• Surgical technique. There are no high-quality comparative trials of open, arthroscopic and percutaneous debridement; open remains the reference standard.
• The ulnar nerve. When to decompress rather than transpose in mild or intermittent neuritis is not standardised.
• Eccentric loading. It is the cornerstone of rehabilitation, but the optimal dose, frequency and pain-monitoring threshold remain debated.

Global epidemiology:

Society guidance, side by side:

Registry and outcome notes:

• Medial epicondylitis involves no implant, so it is not tracked by arthroplasty registries (NJR, AJRR, AOANJRR). Outcome evidence therefore rests on case series (Gabel & Morrey; Vangsness & Jobe) and a small number of RCTs (Stahl & Kaufman; epicondylar PRP trials).
• Reported good/excellent surgical outcomes cluster around 85-90%, consistent across decades and health systems.
• The most robust prognostic signal across series is concomitant ulnar neuropathy severity, not geography.

High- versus limited-resource practice variation: