Osteoarthritis---What is new? Debunking pervasive dogmatic myths

                             Osteoarthritis---What is new?
                          Debunking pervasive dogmatic myths

                                              Dr K S Dhillon. FRCS.

Abstract

Evidence based medicine dictates the need for a paradigm shift in the way we approach clinical problems. A critical enquiry is essential and reliance on clinical experience, textbooks and our local expert alone will not be sufficient. Over the last two decades there has been a paradigm shift in the way we look at osteoarthritis as a disease. This has been possible due to renewed interest in the disease which has led to more research in this area of medicine. Most of this research is published in the medical (non-orthopaedic) and rheumatology literature leaving the orthopaedic community to rely on pervasive intuitive unsystematic clinical expertise of the local expert. A review of the literature was done to find out what is new and to debunk many of the pervasive dogmatic myths that exit in our orthopaedic community.

Osteoarthritis---What is new?

Introduction

Evidence based medicine dictates that our clinical decision making is based on evidence from clinical research rather than on intuitive unsystematic clinical experience. Unfortunately such clinical experience forms the basis of many of our training programs which impart many of the pervasive dogmatic myths that exist even today.
In 1990, when a retired member of the judiciary posed questions about post-traumatic osteoarthritis to the members of the Arthritis and Rheumatism Council of UK, Wright (1) conducted a survey to obtain the views of 120 of his rheumatology colleagues. The questions posed were;

•  What in the light of present knowledge is the likelihood of a patient developing post-traumatic osteoarthritis after an injury? Can the degree of certainty be expressed as a percentage?
• How soon will the osteoarthritis (OA) develop after an injury?
• Has age of the subject any bearing on development of OA since age is one of the factors taken into consideration when compensation is assessed? 
• Is there a varying degree of likelihood of OA for injury of different parts of the body?

Wright at the end of the survey noted that most of the views expressed ‘were anecdotal, were based on retrospective surveys, or were extrapolated from animal experiments’. He also noted that this is why most of the medico-legal reports usually do not have references to published work. Have we made any progress over the last two decades would an interesting question to answer.
Certainly the legal fraternity will be as interested now, as it was then, to know the facts which would help them when they are faced with conflicting professional medical opinions. Is it the paucity of medical evidence or the lack of knowledge among medical professionals that is the cause of anecdotal statements that are often made in many medico-legal reports even today?
Unfortunately the definition of truth for law and science is different. For law it is what survives at the end of an adversarial polemic but for science it is the hypothesis that remains after others hypotheses are excluded. The questions above are difficult to answer for they remain as hypotheses which have not been fully tested yet and any answer would be an opinion. How close are we to answering these questions and what is the reliability of these opinions?

What do we know about Osteoarthritis?

Osteoarthritis (OA) was believed to be a progressive degenerative disease of diarthrodial (synovial) joints characterised by pain, limitation of joint movements and eventually deformity of joints. Radiological examination of joints reveal the presences of one or more of the following characteristics, osteophytes, decreased joint space, subchondral bone sclerosis and subchondral cysts. It occurs due to abnormal excessive stresses on the joint or a defect or abnormality of the biomaterial that constitute the joint. It used to be broadly classified into two types -- primary or idiopathic OA where was no obvious cause could be found and secondary OA, where a contributing factor is present such as trauma, systemic metabolic disease (Gout/Ochronosis), autoimmune (Rheumatoid Arthritis) disease or genetic disease (Chondrodysplasia). It was believed to be a progressive disease of the articular cartilage.

What is new?

Now we know that osteoarthritis is not a discrete disease entity affecting the articular cartilage of the joints (2) but is an organ failure akin to heart or kidney failure (3). This organ or joint failure can result from problems with any of its components such as subchondral bone, cartilage, synovium, ligaments, periarticular muscles and or nerves (2). We have in the past known that ligament laxity places abnormal stresses on the joints and is a risk factor for OA (as seen in the knee, ankle and the trapezio-metacarpal joints). It is also well known that synovial diseases such as rheumatoid disease which cause cartilage destruction also predispose a joint to OA. What was less known is that periarticular muscle weakness (e.g. quadriceps muscle weakness) can place abnormal mechanical stresses on the articular cartilage leading to OA. It is also now known that impairment of proprioception, due to nerve damage, can through impaired coordination of muscular activity lead to joint damage (2).
Subchondral bone sclerosis was believed to cause increased stiffness of the bone leading to abnormal stresses on the adjoining articular cartilage, thereby leading to OA. Now some investigators believe that subchondral sclerosis results from reactivation of the secondary centre of ossification due to abnormal stresses on the joint. Ossification of the cartilage results in a decrease in the thickness of the articular cartilage which makes the cartilage more susceptible to damage which in turn can lead to OA. This has been borne out by bone scan studies which show increased activity in subchondral bone and this activity precedes radiographic evidence of OA by months to years (2). Bone micro-injury in the subchondral bone results in new bone formation and remodelling which leads to subchondral sclerosis and osteophyte formation.
Decrease in joint space on radiographic films is usually believed to be an indicator of cartilage loss and is used in many clinical trials to show progress of OA. A decrease in joint space in the knee can occur due to subluxation of the meniscus in patients with OA of the knee and a need for caution in interpretation of radiographs is essential (2).
OA is not a disease affecting the cartilage alone. It has been shown by bone scans and MRI studies to involve all joint tissues including the subchondral bone, capsule, synovium, blood vessels, nerves, muscles, bursae, and menisci (knee)(4). OA is not a purely destructive or degenerative disease but it is an active reparative process (5). Abnormal remodelling of joint tissues occur after insult to the joint and this remodelling is carried out by inflammatory mediators. The cascade of events that lead to ‘joint failure’ is the same irrespective of the underlying cause (6). A very complex interaction between the various parts of the joint maintains the hematosis of the cartilage and in OA this hemostasis fails (7). The failure of hemostasis starts with activation of surface chondrocytes. The matrix remodelling begins and the matrix stimulates certain receptors and these receptors produce inflammatory cytokines and chemokines, which leads to an inflammatory response in the joint. There appears to be some evidence that cartilage degradation products activate an innate immune response (7). Chemokines, alarmins and adipokines which play a role in inflammation and immunity are responsible for the cartilage damage. Hence OA is an inflammatory immune disease somewhat like rheumatoid arthritis and is not simply a degenerative disease due to wear of the articular cartilage (8).
It is well known that not all patients with OA have symptoms (9), many are asymptomatic and others have minimal symptoms and do not seek treatment. A significant proportion of patients with OA are treated by general practitioners and do not come to hospitals. Most studies are based on hospital population data and this does not give a true picture of the disease prevalence. More recently general population based studies have been conducted which show that OA is not necessarily a progressive disease (10). The damage and repair process can go on for a long time and the condition stabilizes in most cases and may reactivate years later (3). The radiographic changes remain the same when the disease is inactive or has stabilized. Bone scan studies have shown that the process can activate and switch off by itself (11). Most people do stabilize, so OA by itself is not as disabling as previously thought. Only in some patients, who do not stabilize and who have severe OA, does the pain become disabling. We have known for years that there are people in the community with severe radiographic OA who have no pain.
Pain in OA and other chronic diseases is a complex phenomenon and has been poorly understood in the past. However with more studies in recent years the picture is becoming clearer. The OA process is known to set off nocioceptive pain which can result in ‘pain sensitization and perpetuation of pain even if the nocioceptive drive (OA process) stops’ (3). Such pain is mediated by hyper-excitation of the central neurons in the spinal cord and or the brain and pain can continue even in the absences of peripheral nocioceptor stimuli in or around the joint. Anxiety, fear and stress perpetuates this sort of pain. Careful evaluation of the pain in OA is important to select suitable treatment. Pharmacological treatment for peripherally mediated pain (primary hyperalgesia) and that for centrally mediated pain (central hyperalgesia) often differs. Furthermore it is important to remember that a surgical option for treatment of OA in patients with hyperalgesia is unlikely to bear any fruits.

Classification of OA

Osteoarthritis is generally classified based on the aetiology and is sometimes classified by the joints involved and the specific features seen in the disease.

Classification by etiology (12)

1.Primary or Idiopathic-- Where the cause cannot be identified.

2.Secondary – Where identifiable cause is present.
   A. Traumatic
        Intra-articular fractures
        Ligamentous injury (Meniscal Injury)
        Major joint trauma
        Chronic repetitive injury (occupational)
   B. Metabolic
       Acromegaly
       Hemochromatosis
       Crystal deposition diseases
       Ochronosis
   C. Anatomic
        Slipped femoral epiphysis
        Blount’s Disease
        Perthes Disease
        Congenital hip dislocation
         Limb length discrepancy
         Hypermobility syndromes
    D. Inflammatory
        Autoimmune diseases
        Infective arthritis

Classification by Joints and specific features 

OA can be classified by the number of joints involved- monoarticular, oligoarticular or polyarticular and also by the main joint involved- Hip, Knee, Spine, Ankle, Hand and others. Descriptive terms are also often used – erosive, destructive, atrophic, inflammatory and others.
The distinction between primary and secondary OA is often blurred. It is now well known that many patients who develop secondary OA have inherent predisposition to generalised OA (12). There is evidence from population studies that patients who have OA of one joint also have OA of other joints that cannot be accounted for by chance or age alone (12). A strong association between hand and knee OA has been established (13). It is well known since 1952, when Kellgren and Moore defined generalised OA, that Heberden’s nodes were associated with polyarticular OA (14).

Diagnosis of OA

Radiographic criteria

The radiographic features commonly used to define OA include osteophytes, joint space narrowing, subchondral sclerosis, cyst formation and bony contour abnormalities (e.g. femoral head). The most widely used grading system is the one introduced by Kellgren and Lawrence (15). The grading system has 4 grades of OA based on radiological examination;

Grade 0: No OA; when there are no features of OA

Grade 1: Doubtful OA; when there are minimal osteophytes of doubtful significance                                                                   
Grade 2: Minimal OA; when there are definite osteophytes but the joint space is normal

Grade 3:  Moderate OA; when there is moderate decrease in joint space

Grade 4:  Severe OA; when joint space is markedly reduced with subchondral sclerosis

In the knee, scoring of the osteophytes is closely associated with pain and for the hip the narrowing of joint space is closely associated with pain (12). However, radiographic features do not relate consistently with pain, since it is well known that patient with minimal radiographic features can have severe pain and patients with severe radiographic features can have minimal or no pain.

Clinical diagnostic criteria

The most widely used criteria for diagnosis of OA is that developed by the American College of Rheumatology (ACR).
           
American College of Rheumatology (ACR) criteria for OA of the hand, hip and knee.

Hand	Clinical		OA is present if the items present are
	1	Hand pain, aching or stiffness for most days of prior month	1, 2, 3, 4 or 1, 2, 3, 5
	2	Hard tissue enlargement of two or more of ten selected hand joints a
	3	MCP swelling in two or more joints
	4	Hard tissue enlargement of two or more DIP joints
	5	Deformity of one or more of ten selected hand joints
Hip	Clinical and radiographic
	1	Hip pain for most days of the prior month	1, 2, 3 or 1, 2, 4 or 1, 3, 4
	2	ESR≤20 mm/h (laboratory)
	3	Radiograph femoral and/or acetabular osteophytes
	4	Radiograph hip joint-space narrowing
Knee	Clinical
	1	Knee pain for most days of prior month	1, 2, 3, 4 or 1, 2, 5 or 1, 4, 5
	2	Crepitus on active joint motion
	3	Morning stiffness ≤30 minutes in duration
	4	Age≥38 years
	5	Bony enlargement of the knee on examination
	Clinical and radiographic
	1	Knee pain for most days of prior month	1, 2 or 1, 3, 5, 6 or 1, 4, 5, 6
	2	Osteophytes at joint margins (radiograph)
	3	Synovial fluid typical of OA (laboratory)
	4	Age≥40 years
	5	Morning stiffness ≤30 minutes
	6	Crepitus on active joint motion

MCP, metacarpophalangeal joint; DIP, distal interphalangeal joint; ESR, erythrocyte sedimentation rate; PIP, proximal interphalangeal joint; CMC, carpometacarpal joint. (a) Ten selected joints include bilateral second and third PIP joints, second and third DIP joints, and first CMC joints.

Prevalence and Incidence of OA

Heine in 1926 showed through autopsy studies that there was an almost universal presence of pathological features of OA in people over the age of 65 (12). Most present day studies of epidemiology of OA are based on population based radiological surveys. A large such study from Netherland showed that 75% of women aged 60-70 years had OA of the DIP joints and by the age of 80 years about 40% of women had evidence of OA of the knee (16). The prevalence of symptomatic OA of the knee in the US is estimated to be 33% (15.7% moderate to severe OA) in adults above 63 years of age and that of the hip in adults above 55 years of age is estimated to be 3.2% (1.5% moderate to severe) (17). Studies in North England showed that the prevalence of symptomatic OA of the knee is about 6.2% for men and 12.5% for women, in adults older than 35 years. The studies also showed that the prevalence of symptomatic OA of the hip in adults over the age of 55 years was 5.5% in men and 3.6% in women (18).
The prevalence symptomatic OA of the knee in Asia varied with different studies, it ranged from 46% in urban population over 50 years of age in Korea to 14% in adults over the age of 40 years in urban Shanghai (19). In Malaysia a small population based study involving 348 adults above the age of 15 years showed a prevalence rate of symptomatic knee OA of 16% (20). As in Caucasians the prevalence was higher in females as compared with males in these studies. Hip pain and radiographic OA of the hip was rare in COPCORD studies conducted in Asia (19).

Risk Factors for OA

OA is not primarily a cartilage disease as was believed in the past; with more evidence becoming available we now know that OA of diarthrodial joint is an organ failure akin to heart or kidney failure (3). Components of the joint, cartilage, ligaments, menisci, muscles, bone, nerves and other tissues work in synergy to maintain proper function and prevent overloading of the organ. Systemic and local biomechanical factors make the joint susceptible to degeneration and development of OA. The degenerative process is accompanied by a reparative process and when the reparative process is overwhelmed by the degenerative process the OA progresses. The radiographic features of OA rarely improve with time but may remain the same over many years (21,22,23,24). Similarly symptoms of OA may remain the same, get worse or improve with time (3).
There are systemic as well as local  risk factors which influence the development and progress of osteoarthritis.

Systemic risk factors

Age and gender

Many studies have shown that that the prevalence and incidence of OA increases with age (18,25). This is due to age related deterioration of neuromuscular protective function, increased joint loading from obesity, increased joint instability (ligament instability) and loss of reparative capacity of the cartilage (3).
Females are more prone than males to age related progression of multiple joint OA (generalised OA) especially that involving the hand and the knee (14,26). However the frequency of hip OA does not appear to be influenced by gender although the hip OA progresses more rapidly among females (27,28). There is presently no scientific evidence of gender effect on progression of knee OA (22,29). Although there is evidence that females are more prone to generalised OA, there is no consistent evidence to link sex hormones to systemic predisposition to OA (3).

Bone density and osteoporosis

Population studies show that women with OA of the hip and knee have higher bone mineral density at distant sites as well as close to the joints involved as compared to those with no OA and that a high bone mineral density is strongly related to the presences of osteophytes (30,31). This relationship between bone density and the development of OA may be due to a genetic link (32). It has been established that subchondral bone is abnormal in OA, especially in the hip and the knee. In established OA the subchondral bone has lower mineral content, is less stiff, more porous and is biomechanically less competent (33).
The subchondral bone in patients with radiographic OA is metabolically active in some but not in others. Metabolic activity as seen on bone scintigraphy is a powerful predictor of progression of OA with further decrease in joint space in knee OA. Dieppe et al in a study showed that, in patients with radiographic knee OA followed up for 5 years, 88% of patients with severe scan activity initially, showed progression of the OA while none of the patients with no scan activity progressed (11). This shows that turnover and remodelling of periarticular bone plays an important structural role in progression of the disease. The study also found that age, sex, duration of symptoms and obesity had no predictive value with regards to progression of the OA.
Paradoxically, though people with high bone density are more prone to OA, studies show that progressive structural OA with cartilage loss is associated with low hip bone density. Progression of knee OA is also associated with faster bone loss at the hip (34) and worsening OA of the hand is associated increased bone loss in the metacarpals (35).

Genetics

There is plenty of evidence that in susceptibility to OA of the hand, knee, and hip in women, and hip OA in men, there is a significant genetic contribution.  Multiply unspecified genes are involved and environmental factors have a great influence on expression of the disease (3).

Nutrition

Articular cartilage is susceptible to oxidative damage and this damage could theoretically be averted by increased intake of antioxidants such as vitamin C and E. High dietary intake of vitamin C has been found, to slow radiographic progression of OA knee in women and also reduce the frequency of knee pain (36). Vitamin D which plays an important role in bone metabolism has also been found to have protective effect against knee OA progression in older men and women (37).

Local risk factors

While systemic risk factors play a role in the susceptibility of multiple joints to OA, local mechanical factors play an important role in the susceptibility of individual joints to OA.

Obesity

Obesity is probably the most established risk factor for susceptibility to OA of the knee (and to some extent the hip). It is also a risk factor for progression of OA of the knee (38). The mechanism underlying the pathogenesis is overloading of the joints leading to damage to the cartilage, ligaments and other supporting structures. Obesity linked metabolic factors such as adipocytokines, adiposity linked glucose, lipid abnormalities and chronic inflammation have also been implicated in the pathogenesis of OA (3).

Joint injury

There is a lot of evidence that acute joint injury such as fractures, dislocations, ligament injury as well as meniscal injury predisposes the joint to subsequent development of OA (39). The presence OA in another joint makes the injured knee more prone to OA (40). Occupations and elite/professional sports which subject the joints to excessive loads predispose joints to subsequent development of OA (41,42).
A 21 year follow up of 107 patients with open meniscectomy of knee showed a 48% prevalence of Kellgren- Lawrence (K-L) grade 2 (or more) OA of the knee compared to 5% for controls (43). Another 20 years follow up study of 170 post-meniscectomy patients showed a prevalence of grade 2 (or more) OA of 55% in the operated knee and 28% in the non-operated knee. Of these 170 patients only one patient (0.58%) had severe symptomatic OA which resulted in a total knee replacement (44). In a prospective median 36 years follow up of 1321 medical students, 141 sustained injuries to the knee, hip or both and of those injured, 96 developed OA of the affected joints. The overall prevalence rate was 68%  and it was 45% for the knee alone and 19% for the hip alone. The cumulative incidence of knee OA by 65 years of age was 13.9% in those injured and for those with no injury 6% (relative risk of 2.95) (45).

Joint deformity

Abnormal joint loading is a well-known risk factor for later development of OA. Congenital and developmental disorders of hip such as acetabular dysplasia, slipped femoral epiphysis, Perthes disease and various epiphyseal dysplasias are well known to lead to secondary OA of the hip. In fact Harris suggests that more than 90% of the hips with so called primary or idiopathic hip OA are due to unrecognised demonstrable abnormalities of the hip (46). A case control study by Lane et al showed that subclinical acetabular dysplasia (CE angle of < 30 degrees) is risk factor for hip OA in elderly white females (47).
The mechanical alignment of the knee (hip/knee/ankle angle) determines the load distribution in the knee during ambulation. Sixty to seventy per cent of the load goes through the medial compartment (48) which explains the more common occurrence of OA in the medial compartment of the knee (49). We also know that patients with lateral compartment OA tend to have a valgus malalignment while those with medial compartment OA tend to have a varus malalignment. This could be due to the loss of cartilage and subchondral bone that occurs with progression of the disease. Though mild knee malalignment could precede and contribute to the development of knee OA, this however has not been confirmed by studies (3).
The role of knee alignment in disease progression has been studied. Sharma et al in a longitudinal cohort study involving 237 patients with primary OA of the knee found that varus alignment was associated with a 4-fold increase in the odds of medial progression and valgus alignment was associated with a 5-fold increase in the odds of lateral progression. These effects could be detected as early as after 18 months of observation. The study also showed that having a more than 5 degrees of malalignment in either direction at baseline was associated with significantly greater functional deterioration (50).

Muscle Strength

Biomechanical studies have demonstrated that muscles attenuate joint loading (51). Consistent with these biomechanical findings, cross-sectional studies show that those patients with OA of the hip and knee have weaker quadriceps and hip abductor muscles as compared to those without OA (52,53).
On the other hand Chaisson et al in the Framingham cohort population based study with a follow up of 24 years showed that individuals with higher maximal grip strength are at an increased risk of developing OA of the metacarpophalangeal and the proximal interphalangeal joints of the hand (54).
Whether stronger quadriceps can protect the knee against OA remains uncertain. One study showed that stronger quadriceps did not reduce progression of OA of the knee (55). In patients with altered biomechanics of the knee due malalignment or ligament laxity, greater quadriceps strength was associated with greater progression of the OA (56).
This goes to show that there is complex interplay of systemic and local factors that determines the susceptibility to and progression of the disease. This is borne out by the finding that high bone density protects the knee joint against joint space narrowing but at the same time it predisposes the knee to development of early OA as has been highlighted above.

Natural History and Prognosis

Knee OA

It is a commonly held belief that osteoarthritis is invariably a progressive disease. However a review of medical literature shows that it is a myth perpetuated by the local expert with unsystematic clinical experience.
The progression of knee OA is usually slow and it can take many years for the disease to progress. It is also known that it can remain stable for many years (3). In patients with osteophytes alone on radiographic examination, only one third will show radiographic progression (57).
However in patients with osteophytes and subchondral sclerosis at baseline there was progression in majority of the cases and it was more in women than men (58). Dougados et al in a study of 353 patients with a mean age of 67 years, with a disease duration of 7 years showed that radiographic progression varied between 26.8% to 38.1% and improvement occurred in between 8.2% and 15.8% of the individuals. The study also showed that no significant decrease in joint space occurred over a 1 year period. Obesity and involvement of multiple joints was associated with worsening of joint space narrowing (59). Spector et al, in an analysis of 63 patients with a mean age of 60 years and a follow up of 11 years showed that one third of the knees showed radiographic progression. Thirty-eight per cent of knees with grade 2 (K-L grade) and 16% with grade 3 radiographic changes progressed. Improvement was seen in 10% of the knees. The symptoms did not correlate with radiographic progression (60).
Several studies have shown that higher body mass index, older age, presences of Heberden’s nodes and diagnosis of generalised OA are associated progression of radiographic OA (61).Dieppe et al in a study involving 75 patients, with a mean age of 64 years, who were having symptoms for 9 years and followed up for 5 years, found that in 60% of patients with positive radioactive bone scan at baseline the OA progressed while none of the patients with negative scans at baseline progressed. Hence the negative predictive value of a normal bone scan for radiographic progression of OA was 100% (11). Sharif et al found that baseline serum hyaluronic levels were significantly higher in patients with progressive OA of the knee compared with non-progressors (62).
Cartilage loss is the central pathological feature of OA, therefore it should be used as an outcome measurement of disease progression and this can be measure by scoring the joint space difference between serial radiographs. Schouten et al did a population survey of 239 patients with grade 2 or more OA in a Dutch town. The patients were followed up for 12 years. They measured cartilage loss (decreased joint space) and found that 34% of the subjects had cartilage loss after a 12 year follow up. They also found that age, body mass index, height, Heberden’s nodes and a clinical diagnosis of generalised OA were prognostic factors for cartilage loss. They did not find any statistically significant relationship between cartilage loss and gender, meniscectomy, injury, uric acid concentrations, chondrocalcinosis, smoking and occupation related factors (63).
Belo et al reviewed the medical literature up to December 2003 to study the prognostic factors for progression of knee OA. They found 1,004 studies of which 37 met their relevant inclusion criteria. Their review found that strong evidence existed to show that hyaluronic acid serum levels and generalised OA are predictive of progression of knee OA. There is only limited evidence linking progression with varus/valgus alignment and there was limited evidence of no association between progression and meniscectomy. They concluded that knee pain, radiological severity at baseline, sex, quadriceps strength, knee injury, and regular sporting activity does not seem to be related to progression of OA of the knee (64).
Bastick et al (65), in 2015, carried out an updated systematic review of evidence regarding prognostic factors for radiographic knee OA progression. They found that baseline knee pain, presence of Heberden's nodes and  varus malalignment were predictive clinical features for progression of knee OA. Similarly high levels of hyaluronic acid and tumor necrosis factor-α were predictive of knee OA progression. They also found as in previous studies that sex, knee injury, and quadriceps strength did not predict knee OA progression.
Leyland et al in a 14 year population-based cohort study of 1,122 knees found that the percentage of knees with radiographic OA that were replaced (TKR) at year 15 was 1.1% for grade 0 knees, 4.9% for grade 1 knees, 5.3% for grade 2 knees and 6.7% for grade 3 knees. The grade 0 knees at baseline had the highest number of knee replacements at year 15 (10 knees out of 1,122) and grade 3 the lowest number of replacements (2 knees). Hence the majority of individuals (68.4%) who underwent knee replacement did not have radiological evidence at baseline which is suggestive of the fact that radiographs are not the optimal tool for predicting TKR as a long-term outcome in younger individuals(mean age was 53 years at baseline) (66).
Despite so much that has been written and published on the subject, the relationship of radiographic progression of knee OA, to pain, disability and the need for a total knee replacement, remains unanswered.

Hip OA

Unlike OA of the knee where disease evolution is slow, OA of the hip has much more variable course. A minority of patients can have spontaneous radiological and symptomatic recovery and this is often seen in patients with osteophytosis and concentric disease. On the other hand many patients who go for total hip replacement have a short history of severe symptoms and the disease can progress rapidly within 3 months and 3 years to advanced stage of the disease (3). There are fewer studies of natural history of hip OA as compared to the knee. Three clinical studies have reported a wide ranging incidence of clinical deterioration of the disease ranging from 19% to 83% and three radiological studies have reported an incidence of radiological deterioration ranging from 29% to 65% (3).

Hand OA

OA of the hand is less of a clinical problem compared to OA of the lower limb weight bearing joints. The progression of hand OA is usually self-limiting and its evolution is complete after a few years. The condition usually enters a stable phase in the 7th and 8th decade of life (3).

Post-traumatic OA

Post-traumatic OA represents about 12% of the global OA burden (67). The risk of OA following significant joint trauma has been reported to range from 20% to 74% (68). Demographics of lower limb arthritis of the hip, knee, and ankle in patients presenting to a tertiary orthopaedic centre showed that 54% of ankle arthritis, 12.5% of knee arthritis, and 8% of hip arthritis are post-traumatic in origin (69). However this will not reflect true prevalence in the population because many patients with post-traumatic OA may not go to hospital for treatment. About 16% of patients with uncomplicated hip dislocation can develop OA and the figure can rise to 88% in some patients having a dislocation with severe complicated acetabular fractures (70). The average time to clinically apparent OA in young adults with history of joint injury was 22 years in a cohort of medical students (71). However in severe joint trauma arthritis may be evident within a year.
The pathogenetic process involved in post-traumatic OA is not fully understood. The known contributing factors include, acute mechanical cartilage injury, biological response to injury (bleeding and inflammation) and chronic cartilage overload from instability, incongruity and mal-alignment. Severity of the injury and age are contributing factors (67). The pathogenetic process can be divided in two, an acute posttraumatic phase and a chronic phase. The acute phase of injury to articular surface can be subdivided into three, each of which has a different repair process and different prognosis,
Damage to cells and matrix of the cartilage and subchondral bone without visible disruption of the joint surface.
Visible disruption of the articular surface in the form of chondral fissure, flaps or chondral defects.
Visible cartilage and bone disruption where haemorrhage, fibrin clot formation and activation of an inflammatory response follows (72).
Type 2 and 3 injuries will always have type 1 injuries and type 3 injuries will have both type 1 and 2 injuries. Intra-articular fracture will be an example of type 3 injury.
In the acute phase structural tissue damage occurs with death of chondrocytes. The accompanying haemorrhage dilutes the synovial fluid and lowers the hyaluronic acid levels which disrupt the lubrication of the joint and probably the nutrition of the cartilage. Suppression of collagen and proteoglycan synthesis occurs and at the same time various degrading enzymes and inflammatory mediators are released by the remaining viable cells. Initial cell necrosis is followed by spread of cell death by apoptotic mechanisms to the surrounding uninjured regions (73) . Chondrocyte death continues over 48 hours after the injury. Chondrocyte death and cartilage metabolism dysfunction presumably triggers a cascade of whole joint degeneration (67).
In the chronic phase metabolic changes in the cartilage, the subchondral bone and other joint tissues continue, albeit associated with reparative attempts, often over a long clinically asymptomatic period till the patient presents with pain.
 A commonly held belief by the orthopaedic community that intra-articular fragments must be reduced to within 2 mm of anatomic reduction has not been consistently substantiated by evidence based data. There are multiple studies which show that injuries with substantially greater incongruities are clinically well tolerated (72). Anatomical reduction is not always associated with a perfect clinical outcome and several long term studies have reported good clinical outcome with conservative treatment of intra-articular fracture despite imperfect reduction of the fractures (72). The outcome is often joint specific, for example, fractures of the distal radius with articular gaps and step-offs can have a high incidence OA but the long term clinical outcome can be good (74). In acetabular fractures restoration of superior weight bearing dome is crucial to good radiologic and clinical outcome but involvement of the posterior wall is a negative prognostic factor. In tibial plateau fractures, particularly the lateral plateau injuries, articular incongruity is well tolerated and the degree of incongruity has little effect in determining management outcome (72). However malalignment in the presence of an intra-articular fracture of the proximal tibia is associated with poor outcome. The reason for the tolerance of tibial plateau fractures to incongruity is probably due to greater thickness of the articular cartilage. The clinical outcome of distal femoral incongruity is not known, however animal studies show that step-offs of greater than the average thickness of articular cartilage cannot remodel successfully while those as large as the full thickness of the articular cartilage can remodel (72).
Correlating the magnitude and type of articular fracture as well as the post-treatment congruity with the development OA is difficult. Other factors such as joint stability, age of the patient and the presence of generalised OA play an important role. In short the association of accuracy of the reduction and the development of post-traumatic OA remains a major unresolved question (72).
Ligament injuries predispose individual to post traumatic OA. Porat et al in a study of soccer players with ACL (including meniscal) injuries found a 41% prevalence of Grade 2 or more (K-L) OA at 14 year follow-up. Twenty two per cent of the players had no OA. The rest had grade 1 radiographic evidence of possible OA. There was no difference in the prevalence of OA between those who had or not had surgery (75). In fact Daniel et al in a prospective study showed that those who had an ACL reconstruction had a higher prevalence of OA than those who did not (76). Porat et al (75) also showed that 55% of the players were involved in high level recreational activities and subjects with no OA had the same workload and recreational activities as those with OA.
Shelbourne et al (77) in a prospective study of the natural history of PCL injuries in young (average age 25.2 years) athletically active patients with isolated tears of the PCL treated conservatively and followed up for 2.3 to 11.4 years found no correlation between radiographic joint space narrowing and the grade of laxity. The mean Noyes score was 84.2 and Lysholm score was 83.4. Patients with greater laxity did not have worse subjective scores. Regardless of the amount of laxity, majority of the patients were able to return to the same sport. Patients with isolated tears of the PCL treated conservatively achieved a level of objective and subject knee function that was independent of the grade of laxity.
Dejour et al (78) in a long follow-up (mean 15 years) study of patients with tear of the PCL found a 17% prevalence of significant OA and lesser changes in 69% of the patients. Patients with isolated tears of the PCL did remarkably well functionally.
Malunion of diaphyseal fractures is often believed to predispose joints to OA. Studies involving tibial malunions have showed that there is no irrefutable evidence that tibial malunion leads to knee or ankle OA (79,80). Even union of the lateral malleolus with displacement will not lead to arthrosis (81). The ankle appears to be more resistant to OA than is often believed.

Treatment of OA

There are pharmacological, non-pharmacological and surgical approaches to the treatment of OA. Treatment is essentially symptomatic. Presently, there are no effective disease modifying drugs available nor is a cure in sight. There is a widely held view, though erroneously, that OA is inevitably a progressive disease. Studies have shown that 12 to 17% of patients can show improvement over years, 22.5 to 27% can remain the same and 56 to 64% can get worse (82,83). However radiographic changes, symptoms and function must be seen as independent outcome measures since they do not always correlate.

Non Pharmacological Treatment

Exercise

All guidelines on treatment of OA recommend exercise for the treatment of OA, especially for the knee. It has not been established which is the best form of exercise but aerobic and strengthening exercises have shown modest effects on pain relief. Exercises for patients with OA have to be individualised and patient centric. It is most likely to be sustained if it is part of the patient’s daily routine or when it is done in groups. There is no evidence that exercises will accelerate the OA provided injury is prevented (84).

Weight loss

All guidelines advocate weight loss for those patients who are overweight. Studies have shown that obesity is a risk factor for developing OA, as well as a risk factor for progression of OA. Random controlled trials (RCTs) have also shown that weight loss reduces pain and improves function in patients with knee OA.  A 5% weight reduction for those who are overweight is recommended for significant benefit. Calorie reduction should go hand in hand with exercise. However there is no evidence that weight reduction will slow progression of OA and there is also no evidence that it will benefit patients with hip OA (84).

Footwear and orthotics

Most RCTs show that there is no benefit of footwear and orthotics in the management of OA of the knee and there are no studies for hip OA. There have been two studies espousing the benefits of lateral and medial insoles for knee OA. However most believe that the symptomatic pain relief may be due to a placebo response. If the orthotic is not too expensive it may be offered as an adjunct to other therapy (84).

Knee braces, patellar bracing and walking aids
There is no firm evidence that these devices are beneficial in the treatment of OA. At best there is weak evidence of some benefit of these modalities of treatment. They could be used as an adjunct in the treatment with other forms of treatment (84).

Pharmacological Treatment

Simple analgesics and paracetamol

All guidelines recommend paracetamol or its equivalent for the treatment of OA. Its efficacy has been well established for knee OA but there is less evidence for its efficacy for OA of other sites. Though it is less effective as compared to nonsteroidal anti-inflammatories (NSAIDs) for symptomatic relief, the cost and lesser side effects gives it an edge over NSAIDs. However the lack of knowledge and belief in its efficacy among professionals and patients need to be overcome (84).

Topical NSAIDs and Capsaicin

Although some studies have shown the effectiveness of topical NSAIDs and Capsaicin for short term use in the treatment of OA, many believe that there is considerable chance of placebo effect as well as publication bias. There is weak (Grade D) evidence for use of such treatment modalities in OA of the knee. These modalities can be used as an adjunct for short term treatment (84).
NSAIDS and COXIBS
There is a large body of evidence on the effectiveness of NSAIDs and Coxibs in the treatment of symptoms of OA. However their effectiveness is small and is of short duration. These drugs due to their potential side effects should be used for the shortest duration whenever possible. NICE (National Institute for Health and Clinical Excellence, UK) having conducted some cost effectiveness studies has recommended the use of these drugs with proton pump inhibitors (PPIs) when indicated. According to NICE guidelines, Celebrex is the most cost effective of these drugs in the UK. For patients with gastrointestinal and cardiovascular risks, NSAIDS were not a cost effective alternative to paracetamol, with risk outweighing the benefits. The guidelines also recommend that in patients with OA who are on Aspirin should consider other alternatives before using this class of drugs. All patients should be informed of the possible side effects of the use of these drugs. Despite evidence that paracetamol maybe inferior to NSAIDs for pain relief in OA, the risk-harm trade off and the cost places paracetamol ahead of NSAIDs (84).

Opioids

Evidence for use of opioids in the treatment of arthritis is weak. Their use is recommended for moderate to severe OA. The benefits for pain is moderate and for function small. Their usefulness is limited by their well-known side effects. There is a lack of long term data on efficacy and safety of their use in OA (84).

Glucosamine and Chondroitin

The AAOS (American Academy of Orthopaedic Surgery) clinical practice guidelines (18th May 2013) (85) does not recommend the use of glucosamine and chondroitin for patients with symptomatic OA of the knee and the strength of their recommendation is strong. Despite availability of extensive literature on the subject there is no evidence that clinically important outcome has been achieved compared to placebo.

Hyaluronic Acid

The AAOS guidelines also strongly recommend against the use of hyaluronic acid for treatment of patients with symptomatic osteoarthritis of the knee. This recommendation is based on lack of efficacy of intra-articular hyaluronic.

Intra-articular Corticosteroids

The evidence for the use of intra-articular corticosteroids in treatment of knee OA is inconclusive. There is a need for the use clinical judgement for use of intra-articular corticosteroids. Patients with persistent synovitis of the knee may benefit.

Surgical Treatment

Arthroscopy, partial meniscectomy and debridement of the Knee

The AAOS guidelines strongly recommend against the use of arthroscopy with lavage and/or debridement of the knee in patients with OA. Randomised control trials have shown no benefits of such a procedure as compared to physical therapy and medical treatment (86,87). This recommendation does not apply to patients with a primary diagnosis of mechanical derangement of the knee who have concomitant OA of the knee. There is level I evidence that partial meniscectomy in patients with OA of the knee provides no benefit to the patients and that there is no scientific basis for continuing such a practice (88).

Osteotomies around the Knee

The AAOS guidelines recommend that a practitioner may do a valgus high tibial osteotomy for symptomatic medial compartment OA of the knee. However the strength of the recommendation for this procedure is limited because the quality of supporting evidence is unconvincing. Low-strength case series have showed decreased pain on VAS after high tibial osteotomies. For distal femoral varus osteotomies, evaluation for recommendation was not done due to lack of appropriate studies.

Joint Replacement Surgery

Joint replacement surgery is a well-known effective and cost-effective intervention for treatment of OA of the hip and the knee joints. Owing to its irreversible nature and its limited lifespan, it is usually reserved for patients with severe disabling OA which is not amenable to other forms of surgical or conservative treatment. The decision as to when in the course of the disease a joint replacement should be done has not been resolved. The indications for joint replacement remain unclear. After knee replacement 10-20% of people are unhappy with the outcome (89). Besides surgical technique and implant factors, much of the cause of pain and disability remains unexplained. Though socio-demographic factors such as older age, female gender, and low socioeconomic status have been associated with poorer outcome, physiological aspects do play an important role, possibly related to central sensitization (the dysfunction of pain modulation in the CNS) (90).
The long term survival of prosthesis in total joint arthroplasty of the hip and knee has been extensively studied. More than ninety per cent of hip prostheses do not need revision at 10 years and 80% of the total knee prostheses do not need revision at 15 years. However the quality of life (QOL) SF-36 scores for patients with joint arthroplasty are not as encouraging. Rat et al, in study of 3 and 10 year follow up of a patient cohort with hip and knee arthroplasty, showed that at 3 years the QOL scores remained limited as compared to age matched general population and at 10 years the scores were lower than the reference population (91).
The risk factors for requirement of joint replacement in patients with osteoarthritis have not been fully elucidated. However a dose-response association between body mass index and subsequent need for hip and knee replacement has been established. A higher body mass index and obesity significantly contributes to overall risk of undergoing a hip or knee replacement (92,93). Total physical activity level was found to have a dose-response relationship to risk of primary knee replacement but not to hip replacement (94). The Ontario Joint Replacement Registry 2004 report showed that 84% of patients receiving knee replacement were overweight or obese with a BMI of more than 25 (95).
It is often erroneously believed that a large proportion of patients with post-traumatic OA will need a joint replacement of the hip or the knee. An analysis of the patients who had knee replacement in Canada, by the National Canadian Joint Replacement Registry in 2004, showed that the primary diagnosis for total knee replacement was degenerative OA in 93%, inflammatory arthritis in 5%, post-traumatic OA in 2% and avascular necrosis in 1% of the patients (95). Post-traumatic OA constitutes 12% of the global burden of OA but only 2% of the total knee replacements done are for post-traumatic OA. Hence, primary degenerative OA is a bigger risk factor for knee replacement than post-traumatic OA.
Rademakers et al in an analysis of 109 patients with surgically treated fractures of the tibial plateau with an average follow up 14 years showed a 5% incidence of secondary OA of knee which required reconstructive surgery (knee replacement, arthrodesis or osteotomy) (96). Mehin et al in study involving 286 patients with tibial plateau fractures followed up for 10 years, found a 3% incidence of end-stage OA requiring reconstructive surgery (97). Two per cent of the 286 patients had a knee replacement.
Ankle joint replacement has been touted as a viable option for treatment of end stage ankle arthritis. However, some early reports showed failure rates as high as 72% (98). More recent studies have reported an 89% survivorship at 10 years but the quality of evidence in support of ankle replacement is weak and fraught with bias. High quality randomised control trials comparing ankle replacements with other forms of treatment for ankle arthritis are lacking (99).

Arthrodesis of Joints

Arthrodesis of major weight bearing joints of the lower limb such as the hip and knee was widely used for end-stage arthritis before the advent of successful joint replacement arthroplasty. Presently it is widely used for end-stage arthritis of the ankle and not for the hip and the knee where the results of hip and knee replacement are excellent. Arthrodesis provides excellent permanent pain relief but at the expense of loss of motion of the joint. Furthermore it can cause excessive stresses on adjoining joints leading to degeneration of these joints. Hip and knee arthrodesis is still a viable option in a selected group of patients. Limited lifespan of joint arthroplasty dictates that patients who are expected to live more than 30 years may be candidates for arthrodesis of the hip or knee for end-stage arthritis of the hip and knee joint. This is especially true for those patients who are involved in heavy manual labour.
Coester et al in a study of 23 patients with ankle arthrodesis for isolated post-traumatic OA of the ankle, followed up for a mean of 22 years found that there was accelerated degeneration of the subtalar, calcaneocuboid, naviculocuneiform, tarsometatarsal and the first metatarsophalangeal joints of the same foot as compared to the opposite foot. However the knee joint was spared of degenerative changes in all the patients (100).
Schafroth et al (101) in a retrospective study of 30 patients with an arthrodesis of the hip at an average follow up of 18.2 years found that the VAS (visual analogue scale) for pain in the fused hip was an average 1.9 (0-8) , the contralateral hip 2, ipsilateral knee 2.0, contralateral knee 1.8 and low back 3.6. The average walking distance was 111 minutes (range 10 to unlimited). The average SMFA (short musculoskeletal function assessment) was 31.2 (range 9-70). They concluded that if the arthrodesis is done with optimal alignment of the limb than complaints from adjoining joints is minimal even in the long term and an acceptable quality of life is possible. Seven of the hips were eventually successfully converted to a total hip arthroplasty (101).
There is a dearth of literature on the long term outcome of arthrodesis of the knee, which is partly due to the success of knee replacement since the 1970’s. Presently arthrodesis is done mainly for failed joint replacements, sepsis, Charcot joint, flail knee, tumours around the knee and end-stage post-traumatic arthritis in young individuals who are not suitable for a knee replacement. An arthrodesis of the knee can provide a stable and painless limb especially in patients in whom a lot of walking is required for their daily activities. An arthrodesis from a functional perspective will always be superior to an above knee amputation.

Conclusion

There has been much progress since the1990’s in the understanding of osteoarthritis, a disease which was previously believed to be a discrete entity affecting the articular cartilage. Bone scans and magnetic resonance imaging (MRI) has revealed that OA affects all structural components of the joint and not the articular cartilage alone and it is therefore an organ failure akin to heart or renal failure. New evidence shows that OA is an inflammatory immune disease somewhat like rheumatoid arthritis and is not simply a degenerative disease due to wear of the articular cartilage. The damage and repair process can go on for a long time and the condition stabilizes in most cases and may reactivate years later. Bone scan studies confirm that that the disease process can activate and switch off by itself.
 In contrast to previous studies based on hospital population data, now more general population based studies are available to provide a more realistic picture of the disease entity. General population studies show that OA is not necessarily a progressive disease as was previously believed, when studies were based on hospital population data. Furthermore such population based studies now provide more realistic data of the incidence and prevalence rates of the disease. These studies have elucidated the risk and prognostic factors of the disease which were not known previously.
Most epidemiology studies of OA are based on radiographic surveys which show a prevalence rate of 40% for knee OA in women by the age of 80 and hand OA, 75% by the age of 60-70 years. The prevalence of symptomatic OA is definitely lower for the hip as compared to the knee and there is wide variance in prevalence rate depending on age of the population studied and the part of the world where the study is conducted.
Systemic and local risk factors for OA have been quite extensively studied. Of the factors studied obesity is probably the most established factor for susceptibility to OA. Without doubt age, gender, joint injury and joint deformity are important risk factors. Some of the factors for progression of the disease are now known which include, age and body mass, clinical diagnosis of OA, elevated serum hyaluronic acid levels and severe periarticular bone scan activity.
Post-traumatic OA represents about 12% of global OA burden. The demographics of post-traumatic arthritis vary with different joints, with 54% of ankle arthritis, 12.5% of knee and 8% of the hip arthritis being post-traumatic in origin. The pathogenetic process involved though not fully understood is believed to start at the time of the injury. The latent period before the patient presents with symptoms is highly variable depending on the joint involved and the severity of the injury. It may take up to 22 years in young adults or 1- 2 years in patients with severe joint trauma which disrupts the joint.
A commonly held belief by the orthopaedic community that intra-articular fractures must be reduced to within 2mm of anatomical reduction is not evidence based. The outcome of treatment of intra-articular fractures is often joint specific. Intra-articular fractures of the distal radius and the tibial plateau especially the lateral plateau for example have a high tolerance to incongruity with little effect on determining management outcome. The outcome of distal femoral incongruity is not known but animal studies show that distal femoral step-offs are well tolerated. Dome incongruity and posterior wall involvement of the acetabulum is associated with poorer outcome. The association of accuracy of fracture reduction and development of OA is influenced by other factors such as joint instability, age and presence of generalised OA. This association between incongruity and the risk of OA remains a major unresolved question which needs further study.
Malunion of diaphyseal fractures is often believed to predispose adjoining joints to secondary OA. However studies involving tibial malunions have showed that there is no irrefutable evidence that these mal-unions lead to OA of the knee or the ankle.
Ligament injuries of the knee are less of a problem as compared to intra-articular fractures. About 41% of patients with ACL tears (including meniscal injuries) develop significant OA at 14 years follow-up and for the PCL tears 17% have significant OA at 15 years follow-up. Up to 55% of players with ACL tears and majority of patients with isolated PCL tears are able to return to high level recreational activities.
There is a widely held view, though erroneously, that OA is inevitable a progressive disease. Studies show that 12 to 17% of patients can show improvement over years, 22.5 to 27% remain the same, while 56 to 64% can get worse. Radiographic changes, symptoms and function are independent outcomes and the do not correlate.
Contrary to a commonly held view, studies reveal that there is no evidence that glucosamine, chondroitin, and hyaluronic acid have any clinical outcome benefits in the management of OA of the knee. There is strong clinical evidence (RCTs) that arthroscopic lavage, partial meniscectomy and /or debridement of the knee for OA is of no benefit to the patient.
It is often quite erroneously believed that a large proportion of patients with post-traumatic OA will require a joint replacement, however demographics of patients undergoing knee replacement for OA show that in 93% of patients the diagnosis was degenerative OA, 5% inflammatory arthritis, 2% posttraumatic arthritis and in 1% avascular necrosis. Post-traumatic OA forms 12% of the global burden of OA but only 2% of patients undergoing knee replacement have post-traumatic OA. In surgically treated fractures of the tibial plateau the incidence of post-traumatic OA requiring reconstructive (knee replacement, arthrodesis or osteotomy) surgery was only 5%. Hence a large proportion of patients do not develop end-stage post-traumatic OA which would require reconstructive surgery.
Now to answer the questions posed by the retired member of the judiciary that was raised earlier. In light of present knowledge there is sufficient evidence to express in percentages the likelihood of a person developing OA, though it cannot be expressed in absolute numbers. It would vary with the joint involved and the severity of the trauma. The degenerative process probably starts immediately at the time of injury and clinical manifestation occurs when the reparative process is overwhelmed by the degenerative process. The latent period may vary from a year to more than 20 years. We however need to remember that not all patients after an injury will develop OA. Age is a very important factor in both the risk for developing OA as well as for progression of the OA. Older age is a risk factor for both. Finally the likelihood for developing OA and the chance of needing future surgery is very much dependent of the joint involved.
Research in recent years has debunked many of the long held pervasive dogmatic myths perpetuated by intuitive and unsystematic clinical experience.

References

1. Wright V. Post-Traumatic Osteoarthritis – A medico-legal Minefield. British Journal of Rheumatology 1990; 29: 474-478.
2. Brandt KD, Radin EL, Dieppe PA, Van de Putte L. Yet more evidence that osteoarthritis is not a cartilage disease. Ann Rheum Dis 2006; 65:1261.
3. Dieppe P. Developments in osteoarthritis. Rheumatology 2011; 50: 245.
4. Hunter DJ. Imaging insights on the etiology and pathophysiology of osteoarthritis.  Rheum Dis Clin N Am 2009; 35:447.
5. Brandt KD, Dieppe P, Radin EL. The etiopathogenesis of osteoarthritis. Rheum Dis Clin N Am. 2008;34:531.
6. Loeser RF, Goldring SR, Scanzello CR, Goldring MB. Osteoarthritis: a disease of the joint as an organ. Arthritis Rheum. 2012; 64: 1697–707.
7. Houard X, Goldring MB, Berenbaum F. Homeostasis mechanisms in articular cartilage and role of inflammation in osteoarthritis. Curr Rheumatol Rep. 2013 Nov; 15 (11):375.
8. Ru Liu-Bryan. Synovium and the Innate Inflammatory Network in Osteoarthritis Progression. Curr Rheumatol Rep. 2013 May; 15(5): 323.
9. Lotz MK. New developments in osteoarthritis. Posttraumatic osteoarthritis pathogenesis and pharmacological options. Arthritis Research & Therapy 2010; 12: 408.
10. Felson DT, Nevitt MC. Epidemiological studies for osteoarthritis: new versus conventional study design approaches. Rheum Dis Clin N Am 2004; 30:783.
11. Dieppe P, Cushnaghan J, Young P, Kirwan J.  Prediction of progression of joint space narrowing in osteoarthritis of the knee by bone scintigraphy.  Ann Rheum Dis 1993; 52:557.
12. Arden N, Nevitt MC. Osteoarthritis: Epidemiology. Best Practice & Research Clinical Rheumatology 2006; Vol 20, No 1: 3-25.
13. Cushnaghan J and Dieppe P. Study of 500 patients limb joint osteoarthritis. Analysis by age, sex and distribution of symptomatic joint sites. Annals of the Rheumatic Diseases 1991; 50(1): 8-13.
14. Kellgren JH and Moore R. Generalized osteoarthritis and Heberden’s nodes. British Medical Journal 1952;1(475): 181-187.
15. Kellgren JH and Lawrence JS, Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957 Dec; 16(4): 494–502.  
16. Van Saase JL, van Romunde LK, Cats A, Vandenbroucke JP, Valkenburg HA. Epidemiology of osteoarthritis: Zoetermeer survey. Comparison of radiological osteoarthritis in a Dutch population with that in 10 other populations. Annals of the Rheumatic Diseases. 1989;48(4):271-280.
17. Lawrence RC, Helmick CG, Arnett FC, Deyo RA, Felson DT, Giannini EH, et al. Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the US. Arthritis & Rheumatism. 1998; 41(5): 778-799.
18. Lawrence JS, Bremner JM, Bier F. Osteo-arthrosis: prevalence in the population and relationship between symptoms and x-ray changes. Ann Rheum Dis 1966;25:1-24.
19. Fransen M, Bridgett L, March L, Hoy D, Penserga E, Brooks P. The epidemiology of OA in Asia. International Journal of Rheumatic Diseases 2011; 14: 113-121.
20. Veerapan K, Wigley RD, Valkenburg H. Musculoskeletal pain in Malaysia: a COPCORD survey. J Rheumatol. 2007; 34: 207-13.
21. Hernborg JS. The natural course of untreated osteoarthritis of the knee, Clinical Orthopaedics. 1977; 123:130–137. 
22.  Felson DT, Zhang Y, Hannan MT, Naimark A, Weissman BN, Aliabadi P et al. The incidence and natural history of knee osteoarthritis in the elderly, Arthritis and Rheumatism, 1995; 38:1500–1505. 
23. Cooper C, Snow S, TE, McAlindon TE, Kellingray S, Stuart B, Coggon D et al. Risk factors for the incidence and progression of radiographic knee osteoarthritis, Arthritis and Rheumatism 2000;43 (5): 995–1000. 
24. Lachance L, Sowers MF,  Jamadar D, Hochberg M. The natural history of emergent osteoarthritis of the knee in women. Osteoarthritis and Cartilage. 2002;10: 849–854.
25. Hernborg JS. The natural course of untreated osteoarthritis of the knee. Clinical Orthopaedics,1977; 123:130–137; 
26. Oliveria SA, Felson DT,  Reed J,  Cirillo PA, Walker AM. Incidence of symptomatic hand, hip, and knee osteoarthritis among patients in a health maintenance organization. Arthritis and Rheumatism, 1995;38: 1134–1141.  
27. Dougados M, Gueguen A, Nguyen M, Berdah L, Lequesne M, Mazieres B, Vignon E. Radiological progression of hip osteoarthritis: definition, risk factors and correlations with clinical status. Ann Rheum Dis. 1996 Jun;55(6):356-62. 
28. Ledingham J, Dawson S, Preston B, Milligan G, Doherty M. Radiographic progression of hospital referred osteoarthritis of the hip. Annals of the Rheumatic Diseases, 1993; 52: 263–267.
29. Lawrence RC, Helmick CG, Arnett FC, Deyo RA, Felson, David T, Giannini EH, et al. Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis and Rheumatism. 1998;41: 778–799.
30. Hannan M T, Anderson J J, Zhang Y, Levy D, Felson D T. Bone mineral density and knee osteoarthritis in elderly men and women: the Framingham Study. Arthritis and Rheumatism 1993; 36: 1671–1680.  
31. Nevitt M C, Lane N E, Scott J C, Hochberg M C, Pressman A R, Genant H K. et al. Radiographic osteoarthritis of the hip and bone mineral density. Arthritis and Rheumatism 1995;38: 907–916.
32. Antoniades L, MacGregor AJ, Matson M & Spector TD. A cotwin control study of the relationship between hip osteoarthritis and bone mineral density. Arthritis and Rheumatism 2000; 43: 1450–1455.
33. Ding M, Odgaard A & Hvid I. Changes in the three-dimensional microstructure of human tibial cancellous bone in early osteoarthritis.J Bone Joint Surg Br. 2003 Aug;85(6):906-12.
34. Zhang Y,Hannan M T, Chaisson C E, McAlindon T E, Evans S R, Aliabadi P. et al. Bone mineral density and risk of incident and progressive radiographic knee osteoarthritis in women: the Framingham Study. The Journal of Rheumatology 2000; 27:1032–1037.
35. Sowers MF,Zobel D, Weissfeld L, Hawthorne VM, Carman W. Progression of osteoarthritis of the hand and metacarpal bone loss, Arthritis and Rheumatism, 1991;34: 36–42.
36. McAlindon TE, Jacques P, Zhang Y, Hannan MT, Aliabadi P, Weissman B,et al. Do antioxidant micronutrients protect against the development and progression of knee osteoarthritis? Arthritis and Rheumatism 1996; 39: 648–656.
37. Felson DT, Zhang Y, Hannan MT, Naimark A, Weissman B, Aliabadi P, et al. Relation of dietary intake and serum levels of vitamin D to progression of osteoarthritis of the knee among participants in the Framingham Study. Annals of Internal Medicine 1996; 125: 353–359.
38. Cooper C, Snow S, TE, McAlindon TE, Kellingray S, Stuart B, Coggon D et al. Risk factors for the incidence and progression of radiographic knee osteoarthritis, Arthritis and Rheumatism 2000;43 (5): 995–1000.
39. Gelber AC, Hochberg MC, Mead LA, Wang NY, Wigley FM, Klag MJ. Joint injury in young adults and risk for subsequent knee and hip osteoarthritis , Annals of Internal Medicine, 2000;133: 321–328.
40. Doherty M, Watt I, Dieppe P. Influence of primary generalized osteoarthritis on development of secondary osteoarthritis. Lancet 1983; 2(8340): pp. 8–11.
41. Felson DT, Hannan MT, Naimark A, Berkeley J, Gordon G, Wilson PW. Occupational physical demands, knee bending and knee osteoarthritis: results from the Framingham Study, The Journal of Rheumatology, 1991;18:1587–1592.
42.   Buckwalter JA, Lane NE. Athletics and osteoarthritis The American Journal of Sports Medicine 1997;25: 873–881.
43. Roos H, Lauren M, Adalberth T, Roos EM, Jonsson K, Lohmander LS. Knee osteoarthritis after meniscectomy: prevalence of radiographic changes after twenty-one years, compared with matched controls. Arthritis and Rheumatism 1998;41: 687–693.
44. Englund M, Paradowski PT, Lohmander LS. Association of radiographic hand osteoarthritis with radiographic knee osteoarthritis after meniscectomy. Arthritis and Rheumatism 2004;50: 469–475.
45. Gelber AC, Hochberg MC, Mead LA, Wang NY, Wigley FM, Klag MJ. Joint injury in young adults and risk for subsequent knee and hip osteoarthritis. Annals of Internal Medicine 2000;133: 321–328.
46. Harris WH. Etiology of osteoarthritis of the hip, Clinical Orthopaedics and Related Research 1986;213: 20–33.
47. Lane NE, Lin p, Christiansen L, Gore LR, Williams EN, Hochberg MC, Nevitt C. Association of mild acetabular dysplasia with an increased risk of incident hip osteoarthritis in elderly white women: the study of osteoporotic fractures, Arthritis and Rheumatism 2000;43: 400–404.
48. Andriacchi TP. Dynamics of knee malalignment. The Orthopaedic Clinics of North America. 1994; 25:395–403. 
49. Felson DT, Nevitt MC, Zhang W, Aliabadi P, Baumer B, Gale D, et al. High prevalence of lateral knee osteoarthritis in Beijing Chinese compared with Framingham Caucasian subjects. Arthritis and Rheumatism 2002; 46:1217–1222.
50. Sharma L, Song J, Felson DT, Cahue S, Shamiyeh E, Dunlop DD. et al. The role of knee alignment in disease progression and functional decline in knee osteoarthritis .The Journal of the American Medical Association 2001; 286:188–195.
51. Nordin M, Frankel VH, editors. Basic biomechanics of the musculoskeletal system. Philadelphia: Lea & Febiger; 1989.
52. Ensrud KE, Nevitt MC, Yunis C, Cauley JA, Seeley DG, Fox KM,et al. Correlates of impaired function in older women. J Am Geriatr Soc 1994;42:481–9. 
53.  Slemenda C, Brandt KD, Heilman DK, Mazzuca S, Braunstein EM, Katz BP, et al. Quadriceps weakness and osteoarthritis of the knee. Ann Intern Med 1997; 127:97–104.
54. Chaisson, C. E., Zhang, Y., Sharma, L., Kannel, W. and Felson, D. T. Grip strength and the risk of developing radiographic hand osteoarthritis: results from the Framingham Study. Arthritis and Rheumatism 1999;42:33–38.
55. Brandt, KD, Heilman DK, Slemenda C, Katz BP, Mazzuca S, Braunstein EM, Byrd D. Quadriceps strength in women with radiographically progressive osteoarthritis of the knee and those with stable radiographic changes. J Rheumatol. 1999 Nov;26(11):2431-7.
56. Sharma L, Dunlop DD, Cahue S, Song J, Hayes KW. Quadriceps strength and osteoarthritis progression in malaligned and lax knees. Annals of Internal Medicine. 2003;138: 613–619.
57. Danielsson L, Hernborg J. Clinical and roentgenologic study of knee joints with osteophytes. Clin Orthop Rel Res 1970;69:224-6.
58. Hernborg JS, Nilsson BE. The natural course of untreated osteoarthritis of the knee. Clin Orthop Rel Res 1977; 123:130-7.
59. Dougados M, Gueguen A, Nguyen M, Thiesce A, Listrat V, Jacob L, et al. Longitudinal radiologic evaluation of osteoarthritis of the knee. J Rheumatol 1992; 19:378-84.
60. Spector TD, Dacre JE, Harris PA, Huskisson EC. Radiological progression of osteoarthritis: an 11 year follow up study of the knee. Ann Rheum Dis 1992; 51:1107-10.
61. Hochberg MC. Prognosis of osteoarthritis. Annals of the Rheumatic Diseases 1996;55:685-688.
62. Sharif M, George E, Shepstone L, Knudson W, Thonar EJMA, Cushnaghan J, et al. Serum hyaluronic acid level as a predictor of disease progression in osteoarthritis of the knee. Arthritis Rheum 1995; 38:760-7.
63. Schouten J, Ouweland F, Valkenburg H. A 12 year follow up study in the general population on prognostic factors of cartilage loss in osteoarthritis of the knee. Annals of Rheumatic Diseases 1992; 51: 932-937.
64. Belo JN, Berger MY, Reijman M, Koes BW, Bierma-Zeinstra SMA. Prognostic factors of progression of osteoarthritis of the knee: a systematic review of observational studies. Arthritis & Rheumatism (Arthritis Care & Research) 2007; 57(1): 13-26.
65. Bastick AN, Belo JN, Runhaar J, Bierma-Zeinstra SM. What Are the Prognostic Factors for Radiographic Progression of Knee Osteoarthritis? A Meta-analysis. Clin Orthop Relat Res. 2015 Sep;473(9):2969-89.
66. Leyland KM, Hart DJ, Javaid MK, Judge A, Kiran A, Soni A et al. The natural history of radiographic knee osteoarthritis: A fourteen-year population-based cohort study. Arthritis & Rheumatism 2012; 64(7): 2243-2251.
67. Brown TD, Johnston RC, Saltzman CL, Marsh JL, Buckwalter JA. Posttraumatic osteoarthritis: a first estimate of incidence, prevalence, and burden of disease. J Orthop Trauma,  2006; 20(10): 739-744.
68. Schenker ML, Mauck RL, Ahn J, Mehta S. Post-Traumatic arthritis following intra-articular fractures: First hit or chronic overload. University of Pennsylvania Orthopaedic Journal, 2012;22: 26-29.
69. Saltzman CL, Salamon ML, Blanchard GM, Huff T, Hayes A, Buckwalter A et al. Epidemiology of ankle arthritis. Iowa Orthop J, 2005; 25: 44-46.
70. Rodriguez-Merchan EC. Coxarthrosis after traumatic hip dislocation in the adult. Clin Orthop Relat Res 2000; 377: 92-8.
71. Gelber AC, Hochberg MC, Mead LA, Wang NY, Wigley FM, Klag MJ. Joint Injury in Young Adults and Risk for Subsequent Knee and Hip Osteoarthritis. Ann Intern Med 2000;133: 321-328.
72. Giannoudis PV, Tzioupis C, Papathanassopoulos A, Obakponovwe O, Roberts C. Articular step-off and risk of post-traumatic osteoarthritis. Evidence today. Injury, Int. Care Injured 2010;41: 986-995.
73. Kramer WC, Hendricks KJ, Wang J. Pathogenetic mechanisms of post-traumatic osteoarthritis: opportunities for early intervention. Int J Clin Exp Med 2011; 4(4):285-298.
74. Catalano LW, Cole RJ, Gelberman RH, Evanoff BA, Gilula LA, Borrelli J. Displaced intra-articular fractures of the distal aspect of the radius, long-term results in young adults after open reduction and internal fixation. The Journal of Bone & Joint Surgery 1997; 79(9): 1290-1302.
75. Porat AV, Roos EM and Roos H. High prevalence of osteoarthritis 14 years after an anterior cruciate ligament tear in male soccer players: a study of radiographic and patient relevant outcomes. Ann Rheum Dis 2004; 63: 269-273.
76. Daniel DM, Stone ML, Dobson BE, Fithian DC, Rossman DJ, Kaufman KR. Fate of the ACL injured patient: a prospective outcome study. Am J Sports Med 1994;22:632–44.
77. Shelbourne KD, Davies TJ, Patel DV. The natural history of acute, isolated, nonoperatively treated posterior cruciate ligament injuries. A prospective study. Am J Med 1999; 27(3): 276-83.
78. Dejour H, Walch G, Peyrot J, Eberhard P. The natural history of rupture of the posterior cruciate ligament. Rev Chir Orthop 1988; 74: 35-43.
79. Kristiansen KD, Kiar T, Blicher J.  No arthrosis of the ankle after 20 years after tibial shaft fracture. Acta Orthop Scand 1989; 60: 208-209.
80.   Merchant TC, Dietz FR. Long term follow-up after fractures of the tibia and fibular shafts. J Bone Joint Surg AM 1989; 71: 599-606.
81. Kristensen K D, Hansen T. Closed treatment of ankle fractures. Stage 11 supination eversion fractures followed for 20 years. Acta Orthop Scand 1985; 56(2):107-9.
82. Hernborg J S, Nilsson B E. The natural course of untreated osteoarthritis of the knee. Clin Orthop 1977; 123:130-7. 
83.  Massardo L, Watt I, Cushnaghan J, Dieppe P. Osteoarthritis of the knee joint: an eight year prospective study. Ann Rheum Dis.1989; 48:893–7.
84. March L, Amatya B, Osborne RH, Brand C. Developing a minimum standard of care for treating people with osteoarthritis of the hip and knee. Best Practice & Research Clinical Rheumatology 2010;24: 121–145.
85. Treatment of osteoarthritis of the knee, 2nd edition http://www.aaos.org/research/guidelines/TreatmentofOsteoarthritisoftheKneeGuideline.pdf.
86. Moseley JB, O'Malley K, Petersen NJ, et al. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med 2002;347:81-88.
87. Kirkley A, Birmingham TB, Litchfield RB, Giffin JR, Willits KR, Wong CJ, Feagan BG, Donner A, Griffin SH, D'Ascanio LM, Pope JE, Fowler PJ. A randomized trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med. 2008;359((11)):1097–107.
88. Sihvonen R, Paavola M, Malmivaara A, Itälä A, Joukainen A, Nurmi H, Kalske J, Järvinen TL, Finnish Degenerative Meniscal Lesion Study (FIDELITY) Group. Arthroscopic partial meniscectomy versus sham surgery for a degenerative meniscal tear. N Engl J Med. 2013 Dec 26; 369(26):2515-24.
89. Dieppe P, Lim K, Lohmander S. Who should have knee joint replacement surgery for osteoarthritis? International Journal of Rheumatic Diseases 2011; 14: 175–180.
90. Wylde V, Dieppe P, Hewlett S, Learmonth ID. Total knee replacement: is it really an effective procedure for all? Knee 2007;14(6):417-23.
91. Rat AC, Guillemin F, Osnowycz G, Delagoutte JP, Cuny D, Bamann C. Total hip or knee replacement for osteoarthritis: Mid- to long- term quality of life. Arthritis Care & Research. 2010;62(1): 54 – 62.
92. Liu B, Balkwill A, Banks E, Cooper C, Green J, Beral V. Relationship of height, weight and body mass index to the risk of hip and knee replacement in middle-aged women.Rheumatology (Oxford) 2007 ;4. 
93. Flugsrud GB, Nordsletten L, Espehaug B, Havelin LI, Engeland A, Meyer HE.The impact of body mass index on later total hip arthroplasty for primary osteoarthritis: a cohort study of 1.2 million persons. Arthritis Rheum. 2006; 54(3): 802-7.
94. Wang Y, Simpson JA, Wluka A, Teichtahl A, English D, Giles G et al. Is physical activity a risk factor for primary knee or hip replacement due to osteoarthritis? A prospective cohort study. The Journal of Rheumatology 2011; 38(2): 350-357.
95. Medical Advisory Secretariat. Total knee replacement: an evidence-based analysis. Ontario Health Technology Assessment Series 2005; 5(9) at http://www.health.gov.on.ca/english/providers/program/mas/tech/reviews/pdf/rev_tkr_061705.pdf.
96. Rademakers MV, Kerkhoffs GM, Sierevelt IN, Raaymakers EL, Marti RK. Operative treatment of tibial plateau fractures: five to 27 year follow-up results. J Orthop Trauma 2007; 21 : 5-10.
97. Mehin R, O’Brien P, Broekhuyse H, Blachut P, Guy P. Endstage arthritis following tibial plateau fractures: average 10 years follow-up. Can J Surg, 2012; 55(2):87-94.
98. Gougoulias, N., A. Khanna, and N. Maffulli.  How successful are current ankle replacements?: a systematic review of the  literature. Clin Orthop Relat Res 2010; 468(1):199-208.
99. Zaidi R, Cro S, Gurusamy K, Siva N, Macgregor A, Henricson A, Goldberg A. The outcome of total ankle replacement: A systematic review and meta-analysis. Bone Joint J 2013;95-B:1500–7.
100. Coester LM, Saltzman CL, Leupold J, Pontarelli W. Long term results following ankle arthrodesis for post-traumatic arthritis. J Bone Joint Surg Am, 2001 Feb 01; 83(2): 219-219.
101. Schafroth MU, Blokzijl RJ, Haverkamp D, Mass M, Marti RK. The long-term fate of the hip arthrodesis: does it remain a valid procedure for selected cases in the 21st century? Int Orthop. 2010 August; 34(6): 805–810.