What the Research Says About Synovial Inflammation in Osteoarthritis

We only cite studies published in peer-reviewed journals. We summarize findings without overstating conclusions.

This article summarizes a state-of-the-art narrative review published on February 2, 2017, in the open-access journal Arthritis Research and Therapy (BioMed Central/Springer Nature), titled “Synovitis in Osteoarthritis: Current Understanding with Therapeutic Implications.” The review was authored by Alexander Mathiessen (Department of Rheumatology, Diakonhjemmet Hospital, Oslo, Norway) and Philip G. Conaghan (Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, and the National Institute for Health Research Leeds Musculoskeletal Biomedical Research Unit, Leeds, UK). Philip Conaghan is one of the most influential figures in osteoarthritis research internationally, with extensive prior work on the role of synovial inflammation in OA; Alexander Mathiessen was conducting his own research on hand OA and synovitis detection at the time of this review. The paper synthesizes evidence from tissue histology, immunohistochemistry, contrast-enhanced MRI, ultrasound imaging, longitudinal cohort studies, and clinical trials to address a question that had been building for decades in OA science: is the synovium merely an innocent bystander in this disease, or an active driver of both pain and structural damage? The full text is freely available at PubMed Central (PMC5289060).

The authors declare no conflicts of interest directly relevant to this review. The paper was funded through Conaghan’s affiliation with the NIHR Leeds Musculoskeletal Biomedical Research Unit.

Background: Why OA Was Once Called “Wear and Tear” – and Why That Is No Longer Adequate

Osteoarthritis has long been the disease that medicine explained most dismissively: wear and tear, the inevitable consequence of aging joints, a matter of cartilage grinding away under decades of use. This framing carried real consequences for patients – if OA was simply mechanical deterioration, there was no point looking for inflammatory processes to treat, no reason to expect drugs targeting inflammation to help, and no pathological target more promising than pain management and joint replacement when function became intolerable.

This understanding was already being revised in the late 1980s and 1990s, when arthroscopic studies and tissue biopsies began documenting clear signs of inflammatory activity in the synovium – the inner lining of the joint – of OA patients. It was challenged more dramatically by the emergence of high-resolution MRI and advanced ultrasound in the 2000s, which allowed researchers to visualize joint tissues in living patients in ways that X-ray never could. What these imaging studies revealed was not the relatively quiescent, degenerating joint of the “wear and tear” model but a joint in active biological turmoil: inflamed synovium, swollen and thickened, often filled with fluid, accompanied in many patients by bone marrow lesions and meniscal damage detectable on MRI years before joint space narrowing became apparent on conventional X-ray.

Mathiessen and Conaghan’s 2017 review represents a landmark synthesis of this evidence. Its central thesis – that synovitis is not a peripheral feature of OA but a central one, present across all disease stages, measurable by multiple methods, demonstrably linked to both pain and structural progression, and potentially modifiable by anti-inflammatory treatment – had significant implications for how OA is conceptualized and, ultimately, treated.

The Synovium: Anatomy and Normal Function

To understand synovitis, it is necessary first to understand what the synovium normally is and does. The synovium is a specialized connective tissue membrane that lines the inner surface of all diarthrodial joints – the movable joints of the body, including the knee, hip, shoulder, hands, wrists, and ankles. It wraps around the joint cavity but does not cover the articular cartilage surfaces themselves.

The authors describe two distinct layers of the synovium. The outer layer, called the subintima, is typically up to 5 millimeters thick and consists of connective tissue in fibrous, adipose, or areolar (loose) form depending on the region of the joint. It is richly supplied with blood vessels, lymphatic vessels, and nerve fibers but is relatively sparse in cells. This outer layer provides structural support and serves as the conduit for the vascular supply that will ultimately feed the inner synovial layer.

The inner layer – the intima, which faces the joint cavity – is extremely thin: just one to four cells deep, typically only 20 to 40 micrometers in total thickness. The cells making up this lining are called synoviocytes, and they exist in two functionally distinct populations. Type A synoviocytes are macrophages – immune cells whose primary roles are phagocytosis (engulfing and destroying debris and pathogens) and the production of cytokines (signaling proteins that orchestrate immune responses). Type B synoviocytes are fibroblast-like cells that synthesize and secrete the components of synovial fluid, particularly hyaluronic acid, which gives synovial fluid its viscous, lubricating character, and also produce lubricin (proteoglycan 4), the molecule that reduces friction between the cartilage surfaces during movement. In a healthy joint, the Type B fibroblasts dominate, and the thin intimal lining maintains the synovial fluid composition necessary for cartilage nutrition and joint lubrication.

The synovial fluid itself – the small quantity of viscous fluid filling the joint cavity – serves two essential functions: it lubricates the cartilage surfaces, dramatically reducing the friction of movement, and it provides the sole nutrient supply to articular cartilage, which has no direct blood supply of its own. Cartilage is an avascular tissue that relies on diffusion of nutrients from synovial fluid for survival. When the synovium is inflamed and its composition and function disturbed, both the quality of lubrication and the nutritional support of cartilage are compromised.

What Synovitis Looks Like: The Histological Evidence

Synovitis is defined as inflammation of the synovial membrane. When pathologists examine synovial tissue from OA joints under the microscope – typically obtained at the time of joint replacement surgery or by synovial biopsy – they find a characteristic pattern quite different from the thin, relatively acellular lining of a healthy joint.

The histological hallmarks of OA synovitis, as summarized by Mathiessen and Conaghan, are: hyperplasia and hypertrophy of the synovial lining layer (the intima thickens from its normal 1–4 cells to many more cell layers); infiltration of the sublining tissue by mononuclear inflammatory cells, primarily monocytes and macrophages, along with T lymphocytes, B lymphocytes, mast cells, and plasma cells; increased vascularity (the growth of new blood vessels, a process called neovascularization or angiogenesis); fibrosis of the sublining tissue (deposition of excess collagen); and thickening of the overall synovial membrane, which may become visible to the naked eye and detectable on imaging as a hypertrophied, irregular tissue mass.

While this pattern shares some features with the severe synovitis of rheumatoid arthritis – which also involves mononuclear infiltrates, lining hyperplasia, and neovascularization – OA synovitis is generally less intense and less architecturally organized than the pannus tissue (the aggressive, invasive inflammatory membrane) that characterizes RA. OA synovitis also lacks the prominent germinal center formation (organized clusters of B cells producing autoantibodies) seen in established RA. These differences reflect the underlying distinction between OA (primarily a degenerative disease with secondary inflammatory features) and RA (primarily an autoimmune disease with the synovium as its principal target tissue).

Critically, synovitis in OA is not rare. Studies cited by the authors document synovitis histologically in the majority of OA patients examined, and often at rates approaching 50 to 80 percent of those with established OA at the knee, hip, or hand. What is particularly striking is that synovial changes can be detected even in the early stages of OA, including in patients who have joint pain and arthroscopic evidence of early cartilage damage (chondropathy) but who do not yet show visible joint space narrowing on conventional X-ray – the standard radiographic benchmark for OA diagnosis. This early presence implies that synovitis is not simply a late-stage consequence of severe cartilage loss releasing enormous quantities of debris into the joint, but may be initiating or at minimum amplifying the disease process from very early on.

The Molecular Mechanism: How Cartilage Breakdown and Synovial Inflammation Amplify Each Other

One of the most important conceptual contributions of the Mathiessen and Conaghan review is its articulation of the vicious cycle by which OA synovitis and cartilage degradation mutually sustain and amplify each other. Understanding this loop is essential to understanding why OA, once established, tends to progress – and why interrupting the synovial component of this loop represents a potentially disease-modifying therapeutic strategy.

Cartilage Debris as the Initiating Signal

In OA, articular cartilage begins to break down – whether from mechanical overload, aging, prior injury, obesity-related metabolic changes, or other factors. As cartilage matrix is degraded, fragments of its structural components are released into the synovial fluid. These fragments include pieces of collagen, fibronectin, hyaluronic acid, aggrecan (the major cartilage proteoglycan), and crystals of calcium pyrophosphate and basic calcium phosphate, which are commonly found in OA joints.

These released fragments are not immunologically inert. They function as what immunologists call damage-associated molecular patterns (DAMPs) – endogenous alarm signals generated when tissues are injured or stressed, which alert the innate immune system that something is wrong. They are recognized by pattern recognition receptors on the surface of synovial macrophages (the Type A synoviocytes) and other immune cells in the synovium. Crucially, the synovial macrophages attempt to phagocytose this cartilage debris – swallowing and breaking down the fragments – but the act of phagocytosis and the ongoing DAMP signaling activates the macrophages, causing them to release a cascade of pro-inflammatory cytokines.

The Pro-Inflammatory Cytokine Cascade

The activated synovial macrophages become the primary source of the pro-inflammatory cytokines that drive OA synovitis. The most important of these in OA are interleukin-1β (IL-1β), tumor necrosis factor-alpha (TNF-α), and interleukin-6 (IL-6). These cytokines act in multiple directions simultaneously.

They act on the synoviocytes themselves, stimulating further proliferation and the production of more inflammatory mediators, creating a self-reinforcing local inflammatory loop. They act on blood vessel endothelium, promoting increased vascular permeability (which allows fluid to leak into the joint, contributing to the joint effusion or swelling that patients feel) and the expression of adhesion molecules that recruit additional circulating inflammatory cells – more macrophages, T cells, and neutrophils – from the bloodstream into the synovium, amplifying the inflammatory infiltrate.

They act on chondrocytes – the cells within cartilage responsible for maintaining the cartilage matrix – in ways that actively accelerate cartilage destruction. IL-1β and TNF-α in particular stimulate chondrocytes to shift from a matrix-building state to a matrix-degrading one: they upregulate the production of matrix metalloproteinases (MMPs), the enzymes that cleave collagen and other matrix proteins, while simultaneously suppressing the synthesis of new collagen and aggrecan. The net result is accelerated cartilage loss.

They also sensitize the nerves within the synovium to painful stimuli, lowering the threshold at which nociceptors (pain-sensing nerve endings) fire. The synovium is richly innervated, and inflammatory mediators including prostaglandin E2 (PGE2), substance P, and nerve growth factor (NGF) – all produced or stimulated by the cytokine environment – directly sensitize these nociceptors, contributing to the pain experience of OA. This is one mechanism by which synovitis contributes to OA pain beyond the mechanical pain of bone-on-bone contact: it creates a chemically sensitized, hypersensitive pain environment in the joint even when structural damage is not yet severe.

Completing the Vicious Cycle

The accelerated cartilage degradation driven by synovial cytokines releases more cartilage debris into the joint fluid, which is phagocytosed by more synovial macrophages, which are further activated and release more cytokines – and the cycle continues. Additional contributors include the subchondral bone, which can release inflammatory mediators through breaches in the tidemark separating it from the cartilage, and meniscal damage in the knee, which releases tissue debris that joins the DAMP signal pool in the joint cavity. The infrapatellar fat pad of the knee – a fat-filled tissue that sits behind the patellar tendon – also becomes hypertrophic and begins secreting adipokines and inflammatory cytokines, adding another local source of pro-inflammatory signaling that acts on the joint tissue.

What emerges from this picture is not a disease of simple mechanical wear – it is a disease of failed tissue homeostasis in which the inflammatory response to injury, rather than resolving and allowing repair, becomes self-sustaining and destructive. The synovium is not an innocent bystander: it is an active amplifier and perpetuator of the damage.

Detecting Synovitis: Imaging Tools and Their Findings

For most of medical history, OA was diagnosed and monitored through conventional X-ray radiography, which shows only the bony structures of the joint – specifically, the narrowing of the joint space (as cartilage is lost), the formation of osteophytes (bony spurs), and changes in the subchondral bone. X-ray is entirely blind to synovial tissue. The emergence of MRI and high-resolution musculoskeletal ultrasound changed this fundamentally.

MRI: The Gold Standard for Whole-Joint Assessment

Magnetic resonance imaging has become the gold standard for visualizing the full spectrum of OA joint pathology. It can image cartilage, bone marrow, menisci, ligaments, synovial fluid, and – with the use of intravenous gadolinium contrast agent – the vascular proliferation within thickened synovial tissue that is the hallmark of active synovitis. Contrast-enhanced MRI (CE-MRI) is the most sensitive imaging method for synovitis detection, because it directly visualizes the increased blood flow and vascular permeability of inflamed synovium through gadolinium enhancement.

The MRI findings in OA synovitis are striking in their prevalence. A landmark MRI study cited by the authors – examining 87 patients meeting clinical criteria for knee OA using 1.5-Tesla MRI with gadolinium contrast – found that synovitis was present at multiple intraarticular sites in 86 percent of subjects, with the majority having synovitis at six or more of nine assessed anatomical sites within the joint. This near-ubiquitous distribution of synovitis in established knee OA challenged the assumption that inflammation was a feature of only a “subset” of OA patients.

MRI also identifies two specific synovitis patterns that have proven clinically important: effusion-synovitis (the accumulation of excess fluid in the joint cavity, which can be measured volumetrically) and Hoffa synovitis (inflammation in the infrapatellar fat pad, visible as increased signal intensity on specific MRI sequences). Both patterns have been associated with pain and progression in longitudinal studies. The non-CE-MRI sequences – which do not require gadolinium injection – can detect joint effusion and synovial thickening but may miss smaller areas of active synovitis that are only revealed with contrast enhancement, as shown in illustrative MRI comparisons discussed in the paper.

MRI-detected bone marrow lesions (BMLs) – areas of abnormal signal within subchondral bone, representing edema, microtrauma, or necrosis – are also closely linked to synovitis. BMLs are strongly associated with knee pain and represent areas of stressed subchondral bone where the boundary between bone and cartilage is disrupted. The co-occurrence of synovitis and BMLs in many OA patients reflects the integrated nature of the whole-joint pathology, rather than isolated compartmentalized damage.

Ultrasound: Accessible, Dynamic, and Power Doppler-Capable

High-resolution musculoskeletal ultrasound offers a different set of advantages compared to MRI: it is less expensive, widely available in rheumatology outpatient clinics, performed in real time during the clinical examination, and capable of assessing dynamic joint properties during movement. Ultrasound can detect synovial hypertrophy (thickened synovial tissue on grey-scale imaging) and joint effusion. Critically, it can assess vascular flow within synovial tissue using power Doppler imaging – a technique that detects blood flow by measuring the Doppler shift of ultrasound waves reflected from moving red blood cells. Increased power Doppler signal in synovial tissue indicates active neovascularization and increased inflammatory vascularity, providing a direct functional measure of synovitis activity.

Ultrasound has been used extensively in hand OA research, where the small joint size makes MRI imaging more challenging. Studies by Alexander Mathiessen and colleagues using ultrasound in hand OA patients found that both grey-scale synovitis and power Doppler signals strongly predicted radiographic progression of hand OA after five years – and that persistent inflammation at follow-up was an even stronger predictor than baseline inflammation alone. These findings have important implications: they suggest that measuring and monitoring synovitis activity over time, rather than just at a single assessment point, may be necessary to capture its full prognostic significance.

Clinical detection methods – palpating for joint swelling or warmth, or detecting effusion by physical examination – are less sensitive than imaging for synovitis but remain important in clinical practice. The disconnect between clinical and imaging assessment is itself informative: imaging regularly detects synovitis that would not be apparent on physical examination, particularly in deeper joints like the hip or in early-stage disease where volumes of inflamed tissue are smaller.

Synovitis, Pain, and Structural Progression: The Evidence

Synovitis and Pain

The relationship between synovitis and pain in OA is supported by multiple lines of evidence reviewed by the authors, though it is not the sole cause of OA pain and the relationship has some important nuances.

Cross-sectional studies consistently show associations between imaging-detected synovitis and patient-reported pain levels. Studies using contrast-enhanced MRI in knee OA found that synovitis scores were associated with pain even after adjustment for cartilage loss and bone marrow lesions – suggesting that synovitis contributes to pain independently of the structural damage it may be causing. Studies using ultrasound in hand OA found similar cross-sectional associations between synovial thickening, power Doppler signal, and tenderness or pain in affected finger joints.

The mechanistic basis for this pain contribution is multi-layered: the synovium is richly innervated with nociceptive (pain-sensing) nerve fibers, the inflammatory cytokine environment chemically sensitizes these nerves, joint distension from effusion activates mechanosensory nociceptors in the joint capsule, and some inflammatory mediators (substance P, NGF, PGE2) directly lower the threshold for pain signaling. Pain from synovitis is therefore both a structural-mechanical phenomenon (distension) and a neurochemical one (peripheral sensitization).

An important nuance is that the association between synovitis and pain is not universal or perfectly linear. Some OA patients with imaging-confirmed synovitis report little pain, while others with severe pain have modest synovitis findings. This reflects the multi-factorial nature of OA pain, which also involves central sensitization (changes in how the central nervous system processes pain signals), bone marrow lesions, psychological factors, and structural damage independent of synovitis. The authors are appropriately measured in framing synovitis as an important contributor to OA pain, not its sole cause.

Synovitis and Structural Progression

The evidence that synovitis predicts structural worsening of OA is particularly compelling. A longitudinal analysis of 531 knee OA patients found that ultrasound-detected joint effusion at baseline predicted not only subsequent pain and radiographic progression but also joint replacement – meaning that patients with visible joint effusion at baseline were significantly more likely to progress to the point of needing surgery. A large multicentre European prospective study (over 500 knee OA subjects followed for three years) found that baseline joint effusion detected by ultrasound was a significant predictor of knee joint replacement at three-year follow-up, with a hazard ratio of 2.63 – meaning patients with baseline effusion were more than twice as likely to proceed to joint replacement as those without.

MRI-based longitudinal studies showed that knee joints in which synovitis scores increased over time had higher rates of radiographic progression and cartilage deterioration at two years than those in which synovitis scores declined. In hand OA, the Mathiessen et al. ultrasound study found that grey-scale synovitis and power Doppler signals predicted radiographic progression after five years, with persistent inflammation being an even stronger predictor. Additional hand OA studies found that the combination of Hoffa synovitis, effusion-synovitis, medial bone marrow lesions, and meniscal damage on MRI elevated the risk of incident radiographic OA two years before the radiographic changes became apparent – meaning synovitis was detectable before the disease crossed the threshold of conventional radiographic diagnosis.

Together, these findings support a causal role for synovitis in structural OA progression – not merely an association. If synovitis were simply a response to cartilage loss rather than a driver of further loss, it should not consistently predict future worsening independently of existing structural damage. The data suggests it does.

Therapeutic Implications: What Happens When You Target the Synovium

The therapeutic section of the review asks a direct question: if synovitis drives OA pain and structural progression, do treatments that reduce synovial inflammation produce measurable benefit? The answer, examined across several drug classes, is encouraging but incomplete – and reveals both the promise of this therapeutic approach and the methodological challenges of testing it.

Intra-Articular Corticosteroids

Intra-articular (directly into the joint) corticosteroid injections are the most widely used anti-inflammatory treatment in OA and have been administered clinically for decades. Corticosteroids are potent broad-spectrum suppressors of inflammation, reducing cytokine production, immune cell infiltration, and vascular permeability throughout the inflamed synovium.

The evidence for short-term symptom benefit is well-established. In knee OA, intra-articular corticosteroids consistently reduce pain for weeks to several months, though the effect diminishes with time. In hip OA, a study of 120 patients receiving either 40 mg or 80 mg of intra-articular corticosteroid found that both doses produced statistically significant benefits in pain, stiffness, and disability at six weeks, while the higher 80 mg dose maintained significant benefit at twelve weeks. Intramuscular depot corticosteroid has also shown short-term pain benefit in knee OA, though without measurable reduction in ultrasound-detected synovial inflammation on the same assessments – a puzzling finding that may reflect the temporal gap between symptom relief and structural inflammatory resolution, or the limitations of ultrasound for capturing all aspects of synovitis response.

The picture is less clear for structural modification. Repeated corticosteroid injections have not convincingly demonstrated disease-modifying effects in large RCTs – and in one notable trial by McAlindon et al., intra-articular triamcinolone (a corticosteroid) every three months over two years was associated with significantly greater cartilage volume loss than intra-articular saline, despite equal pain outcomes, raising concerns that repeated administration might accelerate the very damage it was intended to slow. This finding complicated the clinical landscape considerably and highlighted that synovitis suppression and cartilage protection may not automatically go together with broad anti-inflammatory approaches. Newer extended-release formulations of corticosteroids – designed to maintain therapeutic concentrations in the joint for longer periods with fewer injections – are under investigation as potentially safer alternatives, with early studies showing sustained analgesic benefit.

Low-dose oral corticosteroids (prednisolone 5 mg) were found to have no significant analgesic effect in a short-term (four-week) randomized controlled trial of 70 hand OA patients, suggesting that the systemic route at low doses does not adequately concentrate anti-inflammatory drug at the joint target. This underscores the importance of intra-articular delivery for achieving local therapeutic concentrations.

Hydroxychloroquine

Hydroxychloroquine is an anti-malarial drug that has been used for decades in rheumatoid arthritis and systemic lupus erythematosus for its modest but reliable anti-inflammatory and immunomodulatory properties. It suppresses toll-like receptor signaling, reduces cytokine production, and may inhibit the inflammatory activation of macrophages – the synovial cell population most central to OA synovitis. Its potential in OA, particularly in hand OA with prominent inflammatory features (so-called “inflammatory OA” or erosive OA), has attracted clinical interest.

The authors review early evidence for hydroxychloroquine in OA but note that the data remain limited and mixed. Pilot studies suggested possible symptom benefit in hand OA, and mechanistic rationale is plausible, but adequately powered RCTs were not available at the time of this review. Subsequent trials have produced variable results, and hydroxychloroquine is not currently a standard recommended treatment for OA. The drug remains an area of ongoing investigation, particularly for inflammatory OA subtypes.

Methotrexate

Methotrexate – the cornerstone of rheumatoid arthritis pharmacotherapy – has also been explored in OA, again particularly in hand and knee OA with prominent synovitis. Its mechanisms include folate pathway interference that suppresses rapidly proliferating immune cells, and modulation of adenosine signaling that has anti-inflammatory effects.

An open-label pilot study reviewed by the authors, using methotrexate for knee OA pain, reported both a reduction in pain and a reduction in ultrasound-detected synovitis. A separate study used 10 mg weekly methotrexate for two months in painful hand OA and found significant improvement in pain at two months. These results were suggestive but the open-label design (patients and clinicians both knew what treatment was being given, creating placebo response risk) limits their interpretability. The authors call for properly conducted, blinded RCTs to determine whether methotrexate’s anti-synovial effects in OA translate into genuine symptom and structural benefit. This remains an active research question.

Biologic Agents: Targeting Specific Cytokines

The identification of IL-1β, TNF-α, and IL-6 as key drivers of OA synovitis has naturally led researchers to ask whether the biologics developed for RA – which target exactly these molecules – might benefit OA patients. The answer so far has been largely disappointing, though the story is incomplete.

Intra-articular anakinra (a recombinant IL-1 receptor antagonist that blocks IL-1 signaling) was tested in a study of 160 patients with symptomatic OA. A single injection showed limited benefit, and larger trials have not firmly established anakinra as effective for OA. The biology is plausible – IL-1β drives both cartilage degradation and synovial inflammation in OA – but the drug’s short half-life in the joint space may mean that single-dose intra-articular delivery does not maintain adequate IL-1 blockade for clinically meaningful duration. Extended-release intra-articular formulations are being explored.

Anti-TNF biologics (such as adalimumab and etanercept), which are transformative in RA and ankylosing spondylitis, have shown modest to no benefit in OA trials. A possible explanation relates to the different character of synovitis in OA versus RA: in RA, TNF-α is a dominant and relatively early driver of the synovial inflammatory cascade; in OA, IL-1β may be more central, or the inflammatory milieu may be more heterogeneous, with different pathways dominant in different patients or disease stages. There is also emerging evidence that OA synovitis exists in biological subtypes – some more macrophage-driven, some more fibroblast-driven – which may respond differently to specific cytokine-targeting strategies. Matching biologic treatment to the specific inflammatory subtype of each patient’s OA is a direction the field is beginning to explore.

IL-6 inhibitors (tocilizumab) and other emerging biologics have been studied in small OA trials. The evidence at the time of this review was insufficient to draw firm conclusions. Synerkine, a fusion protein combining IL-4 and IL-10 (both anti-inflammatory cytokines) with the goal of promoting both anti-inflammatory and cartilage-protective effects, showed early preclinical promise and was entering early clinical investigation.

The Measurement Problem: Connecting Synovitis Reduction to Clinical Benefit

A recurring methodological challenge in OA synovitis research – acknowledged throughout the review – is that demonstrating an imaging change in synovitis (reduced MRI synovitis score, reduced power Doppler signal) does not automatically establish that this change drives the clinical benefit a patient experiences. The correlation between change in synovitis on imaging and change in pain in response to anti-synovial therapy has been variable in existing trials. Some studies show parallel improvement in both; others show pain improvement without measurable synovitis reduction, or synovitis reduction without pain improvement. Validating imaging measures of synovitis as true treatment response biomarkers – reliable surrogates for clinical benefit – is an important priority for future OA trials, and the authors flag this as an area needing standardization and validation work.

The Limitations: What This Review Does Not Settle

As a narrative review rather than a systematic review with formal risk-of-bias assessment, this paper synthesizes evidence selectively and qualitatively. No formal quality weighting of individual cited studies is performed, meaning that results from small observational studies and larger, better-powered trials are discussed in the same narrative without always distinguishing their relative evidentiary weight.

The causal relationship between synovitis and OA progression – while supported by the longitudinal evidence cited – has not been definitively established. Association studies, even longitudinal ones, cannot fully exclude the possibility that a third factor causes both synovitis and structural worsening independently. Randomized intervention evidence demonstrating that reducing synovitis translates to measurable structural protection (not just pain relief) would provide the most direct causal evidence, and this remains limited and somewhat contradictory (particularly in light of the cartilage loss finding with repeated corticosteroid injections).

The heterogeneity of OA synovitis across different patients and different joint sites means that general conclusions about “synovitis in OA” may mask important biological subgroups with different pathways, different natural histories, and different responses to treatment. The optimal patient selection – identifying which OA patients with synovitis would benefit most from anti-synovial treatment – is not yet established.

Finally, the review was published in 2017 and the OA therapeutic landscape has continued to evolve. Subsequent trials – including larger methotrexate studies and newer biological agent trials – have since been published with results that in some cases confirmed and in others complicated the picture at the time of this writing.

Summary of Key Takeaways

• Osteoarthritis is no longer accurately described as a “wear and tear” disease of cartilage alone. Modern imaging and histological evidence firmly establish it as a whole-joint disease in which synovial inflammation – synovitis – is a common and biologically active component, present in all stages of OA and across all major joint sites including knee, hip, and hand.
• The normal synovium is a thin (1–4 cell) lining composed of macrophage-type (Type A) and fibroblast-type (Type B) synoviocytes; it maintains synovial fluid composition for joint lubrication and cartilage nutrition. In OA synovitis, this layer thickens dramatically, inflammatory cells infiltrate the sublining tissue, neovascularization occurs, and the cytokine environment shifts to a pro-inflammatory state dominated by IL-1β, TNF-α, and IL-6.
• The mechanism connecting cartilage damage and synovitis is a self-amplifying loop: cartilage breakdown fragments are released as damage-associated molecular patterns (DAMPs), phagocytosed by synovial macrophages, which activate and release pro-inflammatory cytokines; these cytokines accelerate further cartilage degradation, nerve sensitization, and vascular changes, releasing more debris and sustaining the cycle. The infrapatellar fat pad and subchondral bone are also active participants in this inflammatory circuit.
• MRI with gadolinium contrast is the most sensitive imaging tool for synovitis detection and found synovitis at six or more intraarticular sites in 86 percent of knee OA patients studied. Ultrasound with power Doppler imaging is widely available and clinically practical, detecting synovial hypertrophy and inflammatory vascular flow that predicts radiographic progression – with persistent inflammation over time being a stronger predictor than a single baseline assessment.
• A longitudinal study of 531 knee OA patients found that ultrasound-detected effusion predicted pain, radiographic progression, and joint replacement. The combination of Hoffa synovitis, effusion-synovitis, bone marrow lesions, and meniscal damage on MRI predicted incident radiographic OA two years before it became radiographically apparent. In hand OA, ultrasound-detected inflammation predicted five-year radiographic progression.
• Intra-articular corticosteroids produce well-established short-term pain benefit in knee and hip OA, but the evidence for structural protection is absent and possibly counterproductive with repeated administration. Extended-release intra-articular formulations are under investigation as a safer alternative.
• Biologic agents targeting IL-1, TNF-α, and IL-6 – which transformed RA treatment – have not produced consistent or strong clinical benefit in OA trials to date. The different biology of OA synovitis compared to RA, and the likely biological heterogeneity of OA synovitis across patients, may explain these disappointing results and point toward the need for more precise patient selection.
• Methotrexate showed promising results in open-label studies of OA with prominent synovitis, reducing both pain and ultrasound-detected inflammation, but properly blinded RCTs were lacking at the time of the review. Properly designed trials of anti-inflammatory drugs in synovitis-enriched OA patient populations represent a priority for future research.
• The authors conclude that synovitis in OA is both a source of pain and an independent driver of structural progression – making it a compelling therapeutic target for both analgesia and disease modification. Identifying valid imaging biomarkers of synovitis response to treatment, and defining patient subgroups most likely to benefit from anti-synovial strategies, are key next steps for translating these mechanistic insights into clinical practice.

Mathiessen, Alexander, and Philip G. Conaghan. “Synovitis in Osteoarthritis: Current Understanding with Therapeutic Implications.” Arthritis Research and Therapy, vol. 19, no. 1, 2017, article 18. https://doi.org/10.1186/s13075-017-1229-9. Full text available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC5289060/.