Facet Arthropathy

Facet arthropathy, or zygapophyseal joint osteoarthritis, is degenerative disease of the posterior synovial joints of the spine, characterised by cartilage loss, subchondral sclerosis, osteophyte formation and capsular hypertrophy. It is a major contributor to axial mechanical low back pain and to spinal stenosis.

Prevalence. On CT it is near-universal with age. The Framingham sample found it in 59.6% of men and 66.7% of women, in 24% of those under 40 and in 89.2% of those aged 60 to 69. It is commonest at L4-L5 (45.1%), then L5-S1, then L3-L4.

Contribution to back pain. No reliable figure exists. Framingham found no association between facet OA at any level and low back pain. Estimates of facet-mediated pain in the literature range from roughly 10% to 40%, and the width of that range reflects differing diagnostic-block criteria rather than differing populations.

Progression. Obesity, heavy labour and previous spine surgery (adjacent segment disease) accelerate it.

Why L4/5. L4/5 is the most mobile lumbar segment and bears the highest biomechanical stress. It is the fulcrum for lumbar lordosis and takes the maximum extension and rotation moments, which is why facet arthropathy and degenerative spondylolisthesis are both most common at this level.

The joint. A synovial diarthrodial joint with hyaline cartilage, enclosed in a capsule richly innervated with mechanoreceptors and nociceptors. In the lumbar spine the joints are orientated sagittally, which resists rotation; in the thoracic spine they lie coronally, which allows it.

Innervation. Each joint has a dual supply, from the medial branches of the dorsal rami at the level above and the level below, so the L4/5 joint is supplied by the L3 and L4 medial branches. The medial branch courses over the SAP-TP junction, where the superior articular process meets the transverse process, and that point is the target for both diagnostic blocks and radiofrequency ablation. Because of the dual supply, two levels must be blocked or ablated for every joint.

View larger

View larger

View larger

Load sharing. The facets bear 0-16% of axial load in neutral, rising to 40% in extension. They are the primary resistors of rotation and excessive translation, and they resist shear by preventing anterior translation of the vertebra above.

The degenerative cascade. Facet arthropathy develops through a well-defined sequence:

1. Initiation - primary cartilage wear from ageing, repetitive loading or trauma triggers chondrocyte dysfunction
2. Synovial inflammation - cartilage breakdown products stimulate synovitis, causing pain and joint effusion
3. Subchondral change - bone sclerosis and subchondral cysts develop as cartilage protection is lost
4. Osteophyte formation - marginal osteophytes form as a compensatory response to increased stress
5. Capsular hypertrophy - the joint capsule thickens, contributing to lateral recess stenosis
6. Instability - advanced degeneration may lead to subluxation and dynamic instability, and subluxation to spondylolisthesis

The three-joint complex. The disc and the two facet joints at each level degenerate together, in a vicious cycle. Disc degeneration costs height and increases facet load, which accelerates facet degeneration. Facet degeneration, through synovial hypertrophy, causes posterior element instability and shifts load, which accelerates disc degeneration and adjacent-level disease in turn.

Stenosis. Facet hypertrophy narrows the canal at three sites:

• Central - bilateral facet hypertrophy with ligamentum flavum thickening creates a trefoil canal
• Lateral recess - superior articular process hypertrophy compresses the traversing nerve root
• Foraminal - subluxation and osteophytes narrow the neural foramen

History. Axial mechanical low back pain, worse with extension and rotation. Prolonged standing, walking and extension activities aggravate it; flexion, sitting and rest relieve it. Pain may radiate to the buttock or thigh but rarely below the knee, the pseudo-radicular pattern.

Stiffness and night pain. Morning stiffness is common and reflects an inflammatory component. Night pain is rare, and if present should make you consider other pathology.

Examination. Inspection shows loss of lumbar lordosis, a protective flexion posture. The facet joints are tender 2-3 cm lateral to the midline, extension and ipsilateral rotation are painful, and movement may bring audible or palpable crepitus. Neurology is usually normal unless there is concomitant stenosis or radiculopathy, and straight leg raise is negative unless there is a concurrent disc herniation.

The extension-rotation test. The patient stands, extends the lumbar spine, then rotates toward the painful side. Reproduction of the typical pain suggests facet-mediated pain, and sensitivity increases when it is combined with paraspinal tenderness over the facet joints. The 2020 multispecialty consensus found that no single history, examination or imaging finding reliably diagnoses facet pain in isolation, so a positive test suggests the diagnosis without confirming it.

Against discogenic pain. Disc pain is worse with flexion, sits in the midline or across both buttocks, and is axial, or in the leg if there is a herniation. Its imaging signs are disc desiccation, Modic changes and annular tears, and its diagnostic test, provocative discography, is controversial.

Radiographs come first. AP and lateral films assess alignment, disc height and osteophytes, and oblique views show the facet joint space and the pars interarticularis. Flexion-extension laterals look for dynamic instability: translation over 3 mm or angulation over 10 degrees. The findings are joint space narrowing, sclerosis, osteophytes and the vacuum phenomenon.

MRI is second line, and its key finding is facet hypertrophy with ligamentum flavum thickening producing the trefoil canal. Each sequence shows something different:

• T2 sagittal and axial - facet joint effusion (high signal) and cartilage loss
• T1 - subchondral sclerosis (low signal) and fatty infiltration
• STIR - bone marrow oedema, a sign of active inflammation

CT is the alternative when MRI is degraded by metal artefact or detailed bony anatomy is needed. It shows osteophytes, subchondral cysts and joint orientation excellently.

View larger

SPECT-CT is the functional study, for discordant imaging and clinical findings or suspected facet pain. Increased tracer uptake indicates active inflammation, and in multi-level disease it guides which levels to block.

Diagnostic blocks. Imaging facet OA does not confirm the pain source. Controlled diagnostic medial branch blocks are the gold standard, and a positive result is 80% pain relief with concordant provocation on examination. A single block has a false-positive rate of 30-40%; two separate blocks, with a short-acting and a long-acting anaesthetic, reduce it to under 10%. The 80% threshold is itself disputed (see Controversies).

The two techniques differ in lesion geometry, and the geometry dictates how the electrode must be positioned relative to the medial branch.

Conventional (thermal) RFA heats the active tip to roughly 80-90°C for 60-90 seconds. The lesion forms radially around the shaft of the active tip and extends only minimally beyond its end, so the electrode is placed parallel and tangential to the nerve as it crosses the SAP-TP junction, to capture the greatest length of medial branch. Lesion size grows with a larger-gauge cannula, a longer active tip, higher temperature and longer duration, and several overlapping lesions are often made to allow for anatomical variability in the course of the nerve.

Cooled (water-cooled) RFA circulates chilled water through the electrode to prevent the charring and impedance rise that limit conventional lesions. It deposits more energy, typically at a set tip temperature near 60°C, and creates a larger, more spherical lesion that projects beyond the tip. That permits a more perpendicular approach and, in theory, more reliable capture of a variably located nerve, at the cost of longer procedure time and higher consumable cost. Comparative trials are limited and have not established clear clinical superiority of either modality.

Why placement matters. The commonest technical cause of a "failed" RFA is a lesion that misses the nerve. Placement technique, active-tip length and multiple lesions drive success, not target temperature alone.

Before lesioning. Whatever the modality, sensory stimulation (50 Hz, seeking concordant paraesthesia at low voltage) and motor stimulation (2 Hz, confirming the absence of lower-limb muscle contraction that would signal proximity to the ventral ramus) confirm safe, accurate targeting.

Pulsed RFA uses bursts that keep tissue below roughly 42°C and is non-neurodestructive. It is generally considered insufficient for durable medial branch denervation and is not the standard for facet pain; continuous thermal or cooled lesioning is preferred.

Two targets. An intra-articular (IA) injection places local anaesthetic, with or without corticosteroid, into the facet joint capsule under fluoroscopic or CT guidance. The joint holds only about 1-2 mL and over-distension can rupture the capsule; the injection anaesthetises the joint itself, and the steroid is given with therapeutic intent, though evidence for durable benefit is weak. A medial branch block (MBB) anaesthetises the nerves that supply the joint, not the joint cavity.

Why MBB comes before ablation. The medial branch is the exact structure that radiofrequency ablation denervates, so a positive MBB maps directly onto the RFA target and a positive IA injection does not. The 2020 multispecialty consensus concluded that MBB is more predictive of RFA success than IA injection. In FACTS the two blocks behaved similarly, with positive-block rates of 54% (IA) and 55% (MBB) and post-RFA responder rates of 51% and 56%. Neither was therapeutic: both blocks are prognostic, not therapeutic, and their role is to select candidates.

After ablation. Medial branch RFA is generally safe. Besides neuritis and infection, skin burns occur but are rare with proper technique, and worsening pain, in 5-10%, may indicate the wrong diagnosis.

Conservative care. A figure of 40-60% achieving satisfactory relief is sometimes quoted, but no cited trial on this page reports a success rate for conservative care. Both randomised trials used a structured exercise programme as the comparator, and in MINT the exercise programme alone performed as well as exercise plus radiofrequency denervation, which is the strongest statement about conservative care the evidence supports.

Radiofrequency ablation. The randomised trials disagree with the observational literature, and the trials are better. MINT's facet arm found a mean difference of -0.18 (95% CI -0.76 to 0.40) at three months against a prespecified 2-point threshold, a confidence interval that excludes clinically important benefit. The 2015 Cochrane review, whose search closed before both trials, found moderate-quality evidence of a -1.47 point short-term advantage over placebo, itself below that threshold, and concluded there was no high-quality evidence of benefit.

The case for selection. In FACTS, responders after a positive block did better after ablation than after a saline block (56% vs 24% at three months), which is the strongest remaining argument for careful selection.

Fusion. No source cited on this page reports fusion outcomes for facet-mediated pain. Figures quoted in the fusion literature apply to structural indications, spondylolisthesis and stenosis, and should not be transferred; fusion for isolated facet pain without structural pathology is not supported by any evidence cited here.

Global Epidemiology

• Facet osteoarthritis is near-universal with age: CT prevalence rises from roughly 24% under 40 to 89% at 60-69 years (Framingham cohort), with the highest burden at L4-L5.
• Imaging facet OA correlates poorly with symptoms; facet-mediated pain is estimated to account for 10-40% of chronic low back pain, the wide range reflecting differing diagnostic-block criteria.
• Lumbar fusion and facet-intervention rates vary several-fold between countries and even between regions of the same country, reflecting practice variation rather than disease burden.

The RFA efficacy debate. Observational series and earlier reviews report 50% or greater relief in 60-70% of well-selected patients, but the two highest-quality pragmatic RCTs are sobering: MINT (2017) found no clinically important benefit of ablation over exercise, and FACTS (2018) found the blocks themselves no more therapeutic than saline. Critics argue MINT used a liberal 50% block threshold and possibly suboptimal lesioning; proponents of the RCTs argue prior series were confounded by selection bias and placebo response. The honest exam answer is that RFA may help carefully selected patients, but the strongest evidence is sobering and routine use is not supported.

Diagnostic block threshold, 50% or 80%. No consensus exists. An 80% dual-block standard maximises specificity but yields more false negatives and is costly; a single 50% block (used in MINT) increases sensitivity but raises false positives to 30-40%. Guidelines deliberately leave this to local protocol.

Is "isolated facet pain" a real entity? Facet OA almost always coexists with disc degeneration in the three-joint complex, and imaging facet OA does not correlate with pain. Whether a purely facet-mediated pain phenotype exists, and whether it ever justifies fusion, remains contested; fusion for axial pain alone is widely discouraged.