Tennis Elbow: The Evidence on What It Is and How to Treat It

Tennis elbow — or lateral epicondylalgia — is one of the most common upper limb conditions seen in clinical practice, affecting an estimated 1–3% of the adult population at any given time. It has nothing to do with playing tennis. Most people who present with it have never picked up a racquet. It occurs in office workers, tradespeople, manual labourers, musicians, and recreational gym-goers — anyone whose work or sport places repetitive load on the forearm extensors.

And for most of them, the first thing they're told to do is get a cortisone injection. The research is unambiguous on what happens next: short-term relief, followed by worse outcomes at 12 months than if they'd done nothing at all. The reason that cycle persists is a fundamental misunderstanding of what tennis elbow actually is at the tissue level — and that misunderstanding starts with the name.

It's Not Tendinitis — It's Tendinopathy

The old name for this condition was lateral epicondylitis. The suffix -itis means inflammation. The problem is that the histological evidence — tissue biopsy studies — does not support active inflammation as the primary process in chronic lateral epicondylalgia. Nirschl and Ashman, who have studied elbow tendinopathy extensively, described the tissue changes seen on biopsy as angiofibroblastic dysplasia: a disorganised proliferation of fibroblasts and immature collagen, accompanied by abnormal vascular ingrowth, but conspicuously lacking the inflammatory cell infiltrate (neutrophils, macrophages) that characterises true -itis pathology.¹

This distinction is not semantic — it has direct treatment implications. Anti-inflammatory strategies (NSAIDs, cortisone) target a process that is not the primary driver of the condition. They may modulate pain in the short term, but they do not address the disorganised tendon tissue that is the underlying mechanical problem. Understanding this is why the management has shifted dramatically over the past two decades.

Cook and Purdam's influential 2009 paper in the British Journal of Sports Medicine presented a tendon pathology continuum model that reframed how clinicians think about this: tendon pathology progresses from reactive tendinopathy (true short-term load-reactive changes) through failed healing to degenerative tendinopathy — each stage requiring a different management emphasis, but all stages sharing the requirement for appropriate mechanical loading to drive recovery.²

The Anatomy: Why This Tendon Is Vulnerable

The common extensor tendon originates at the lateral epicondyle of the humerus. It is the shared proximal attachment of the extensor carpi radialis brevis (ECRB), extensor digitorum communis, extensor digiti minimi, and extensor carpi ulnaris. The ECRB is the most commonly involved — it sits deep to the extensor digitorum communis and is subject to the highest mechanical stress during wrist extension and forearm rotation.

Coombes, Bisset, and Vicenzino's integrative model (2009, British Journal of Sports Medicine) describes three contributing systems: local tendon pathology at the common extensor origin; motor system changes — reduced grip strength, altered neuromuscular control of the forearm musculature; and pain system changes — altered central processing that amplifies the pain experience beyond what the local tissue state would predict.³ This three-system model explains why the condition doesn't always respond to locally focused treatment: you need to address all three contributors.

What the Evidence Says About Cortisone

The cortisone story in lateral epicondylalgia is one of the clearest examples in musculoskeletal medicine of a treatment that feels effective and isn't. Smidt and colleagues' 2002 landmark trial in the Lancet randomised patients to corticosteroid injection, physiotherapy, or a wait-and-see approach. At 6 weeks, cortisone was clearly the most effective — large, statistically significant reductions in pain and disability. At 52 weeks, the injection group had the worst outcomes of the three groups: higher recurrence rates, lower success rates, and no advantage in pain or function over the group that received no treatment at all.⁴

Bisset and colleagues' rigorous RCT in the BMJ in 2006 replicated this finding, comparing corticosteroid injection, physiotherapy (mobilisation with movement plus exercise), and a wait-and-see approach in 198 participants. Cortisone injection produced the best short-term results. At 12 months, physiotherapy produced significantly better outcomes than both injection and wait-and-see — with the injection group showing the highest recurrence rates.⁵

Mobilisation with Movement: The Manual Therapy Evidence

One of the most consistent manual therapy findings in lateral epicondylalgia is the efficacy of Mulligan's mobilisation with movement (MWM) — a technique involving sustained lateral glide of the head of the radius while the patient performs a pain-free gripping or wrist extension task. The proposed mechanism is a positional fault correction that alters the mechanics of the radiohumeral joint, immediately reducing pain during provocation.

Bisset and colleagues' 2006 RCT included MWM as part of the physiotherapy intervention and found it contributed meaningfully to the superior outcomes in that group. Vicenzino, Cleland, and Bisset's 2007 clinical commentary in the Journal of Manual & Manipulative Therapy reviewed the available evidence and concluded that manual therapy techniques — particularly MWM — are an evidence-supported adjunct to exercise in lateral epicondylalgia management, producing both immediate and short-term pain reduction that facilitates the patient's ability to perform progressive loading exercises.⁷

Exercise: The Foundation of Recovery

Progressive loading is the best-supported long-term treatment for lateral epicondylalgia. The rationale is mechanobiological: tendon tissue remodels in response to mechanical stress, and appropriate loading drives the production of organised type I collagen — the structural protein that restores normal tendon architecture and load-bearing capacity.

Karanasios and colleagues' 2021 systematic review and meta-analysis in the Scandinavian Journal of Medicine & Science in Sports, pooling data from 31 RCTs, found that exercise interventions produced statistically significant improvements in pain and function, with eccentric and progressive resistance protocols demonstrating the most consistent effects. Importantly, exercise outcomes at 12 months were superior to both cortisone injection and a wait-and-see approach.⁸

Bisset and Vicenzino's 2015 review in the Journal of Physiotherapy provides the most practical treatment hierarchy based on current evidence: multimodal physiotherapy combining exercise and manual therapy is the recommended first-line approach; corticosteroid injection is appropriate only for short-term pain management in severely limited cases when rehabilitation cannot otherwise be commenced; and surgery is reserved for the small proportion (estimated at <10%) of people who do not respond after 6–12 months of optimal conservative management.⁹

Treatment	Short-term (6 wks)	Long-term (12 months)
Cortisone injection	Strong pain reduction	Worse than physio; high recurrence4,5
Wait and see	Slow improvement	Comparable to injection; better than injection for recurrence4
Physiotherapy (MWM + exercise)	Good improvement	Best outcomes; lowest recurrence5,8
NSAIDS alone	Modest short-term effect	No effect on tendon remodelling; not recommended as primary treatment6

What Happens in a Session

Assessment identifies the primary pain source and any contributing factors — cervical or thoracic joint stiffness that refers to the elbow, altered scapular control affecting forearm mechanics, or specific grip and wrist mechanics that are overloading the tendon. Treatment then addresses the local tendon through graduated loading, while manual therapy techniques (including MWM and targeted soft tissue work to the forearm musculature) reduce sensitivity and restore normal motor patterns. Home exercise is an essential component: the session initiates the process, but the progressive loading between sessions is where the tendon actually adapts.

Recovery timelines vary. Most people with acute or sub-acute lateral epicondylalgia — symptoms less than 3 months — respond well within 6–8 weeks of active management. Those with chronic, long-standing pathology (especially those who have had multiple cortisone injections) often have more extensive tendon disorganisation and require a longer, more progressive rehabilitation course. But the principle is the same: the tendon needs to be loaded, not rested into recovery.