X-ray diagnosis of osteoarthritis of the knee joints (gonarthrosis)

Knee joints are among the most difficult joints to properly examine radiographically due to their structural complexity and wide range of motion. Gonarthrosis can be localized only in a certain section of the joint, which also complicates the diagnosis of joint changes in osteoarthritis of the knee joints (gonarthrosis).

The anatomical and biomechanical features of the knee joint initially suggest a significant frequency of damage not only to bone structures, but also to the ligament-meniscus complex (LMC). Therefore, the high percentage of primary diagnostic errors in the analysis of radiographs can be explained by the fact that the main attention is paid only to changes in bone structures. Numerous functional tests and positions allow us to analyze and, based on certain signs, assume with a high degree of probability the presence of damage to the LMC during radiography. Taking into account the identified changes, the X-ray examination can be supplemented with other visualization methods - ultrasound, MRI, etc.

The main rule for X-ray examination of the knee joint is polyposition.

Standard projections used in knee joint radiography include direct (anteroposterior) and lateral. If necessary, they are supplemented with right or left oblique, as well as axial and other projections.

The effectiveness of X-ray diagnostics of knee joint lesions largely depends on the quality of the X-ray images.

In the direct projection, the internal and external contours of the joint space have different curvature and orientation, due to which they cannot be obtained as an ideal single line on the same image. Its internal part is better seen when the central X-ray beam is perpendicular to the table surface, and the external part - with a caudocranial displacement of the beam by 5-7°. A compromise is achieved depending on the area of interest. The axis of rotation of the knee passes through the medial region of the joint, which is therefore more often subject to changes compared to the external one. Therefore, when taking an image of the knee in a direct projection, the preferred position is when the joint is in a state of maximum extension with a perpendicular direction of the central beam to the object of study and its centering on the midpoint of the knee, slightly shifted inward.

Quality criteria for radiographs

In direct projection	Symmetry of the axial sides of both condyles of the femur The location of the intercondylar tubercles in the center of the intercondylar fossa Partial masking of the head of the fibula by the metaepiphysis of the tibia (approximately 1/3 of its transverse size) Overlay of the patella contours on the central region of the femoral metaphysis
In lateral projection	Possibility of examination of the PFO joint and tibial tuberosity
In all projections	Location of the joint space in the center of the radiograph Clear image of the spongy bone structure

The image taken in the position of maximum knee extension is the standard anteroposterior projection. It allows examination of the anterior part of the radiographic joint space.

Direct images taken with the knee flexed at 30° (Schuss position) or 45° (Fick position) are taken to assess the condition of the posterior sections of the joint space, at the level of which damage to the subchondral sections of bones (osteonecrosis) and cartilaginous structures (osteochondritis) is most often detected.

These positions are convenient for studying the intercondylar space, which in this position is maximally accessible for viewing, and also allow the detection of free foreign bodies in the joint cavity, formed as a result of damage to the articular cartilage.

A direct projection image of the knee joint can be taken with the patient lying down or standing. When the pathology is mechanical in nature and damage to the ligamentous apparatus is suspected, it is preferable to take an X-ray standing both under load and in a relaxed state to examine the X-ray joint space and the joint axis.

An X-ray examination of the knee joint in a direct projection is necessarily supplemented by an image in a lateral projection.

In lateral radiography, the central beam passes along the joint space with a 10° slope in the caudocranial direction. In this case, the edges of the femoral condyles overlap each other, and their articular surfaces are displaced in their posterior lower part. This allows one to clearly distinguish their contours and assess the condition of the PFO of the articulation.

A lateral view of the knee joint is taken either with the patient lying on his side, with the joint completely relaxed, or standing, without load on the joint being examined. Slight knee flexion (30° or 15°) allows one to determine the condition of the PFO of the joint. Flexion is intended to visualize the patella at the moment of its introduction into the intercondylar region.

Carrying out radiography in the lateral projection allows us to identify transient instability (delay in the entry of the patella into the intercondylar fossa), which may disappear at 30° flexion or not be detected on an axial image when the minimum flexion is 30°, and also to assess the height of the patella and the condition of its articular surface.

The different areas of the articular surface of the knee on the lateral image have characteristic distinctive features. These differences are related to the functional features of each area. The shape of the femoral condyles is a mirror image of the anterior part of the corresponding tibial plateau, with which contact is established during extreme knee extension.

In the presence of transient patellar instability or suspected cruciate ligament injury, additional stress testing is necessary.

The lateral image is especially important for studying the PFO joint.

In assessing the topography of the patella, various measurement coefficients are used, of which the most commonly used is the Cato index. To measure this index, an image taken with the knee joint flexed at 30° is required.

The Cato index is the ratio of the distance from the lower edge of the patella to the anterior superior angle of the tibia (a) to the length of the articular surface of the patella (b). Normally, this ratio is usually equal to 1.0±0.3.

Too high position of the patella (patella alta) leads to its delayed introduction into the trochlear orifice, which can be the cause of patellofemoral instability. The patella index is used to diagnose such instability.

On the lateral image, the patella profile has two posterior lines, one of which corresponds to the patella crest, and the other, denser one, to its outer edge. The distance between these two lines (a-a) is the patellar index (normally 5 mm). Values <2 mm indicate instability, which, however, may be transient, disappearing with flexion at an angle of more than 15-30°.

The trochlear index is measured from the bottom of the intercondylar fossa to the articular surface of the patella, namely to its crest, and is determined at a distance of 1 cm from the upper edge of the intercondylar surface, which corresponds to the zone of introduction of the patella at the very beginning of flexion. Normally, it should be equal to 1 cm. Values < 1 cm indicate patellar dysplasia, which is often combined with underdevelopment of the articular surface of the patella. With high index values, one should think about excessive depth of the intercondylar fossa, which increases the risk of developing patellar chondropathy.

A certain role in the diagnosis of knee joint lesions is given to patellofemoral axial projections.

Radiography at 30° flexion is the most informative for studying the radiographic joint space of the PFO. At a smaller flexion, the thickness of the soft tissues through which the beam passes is large, which negatively affects the image quality. This axial projection differs from others with a large flexion angle in the visualization of the edges of the trochlear notch. The inner edge of the intercondylar fossa is very short, the inner and outer edges have an angular appearance, significantly sharper than in the lower and middle segments of the trochlea. The outer part of the PFO of the joint is subjected to greater loads than the inner one. Therefore, the subchondral bone is denser at the level of the outer section, and the bone trabeculae are oriented outward.

An axial image at 30° is most convenient for detecting patellar instability (external transient subluxations of the patella occur only at the very beginning of flexion) and early osteoarthrosis of the lateral PFO joint.

Traditionally, the classification of I. Kellgren and I. Lawrence (1957), improved by M. Lequesne in 1982, is used to determine the radiographic stage of osteoarthritis of the knee joints. It is based on an assessment of the degree of narrowing of the radiographic joint space, subchondral osteosclerosis and the size of marginal bone growths; it distinguishes 4 stages.

Stages of osteoarthritis (according to Kellgren I. and Lawrence L, 1957)

• 0 - No radiographic signs
• I - Doubtful
• II - Minimum
• III - Average
• IV - Expressed

Despite the certain conventionality of such division of osteoarthrosis into radiological stages, this method is successfully used in modern radiology subject to a number of conditions. In particular, for timely detection of gonarthrosis, it is necessary to examine the joint in three projections: anterior, lateral and axial, which allows assessing the medial, lateral, PFO and TFO of the joint.

For a more accurate assessment of radiographic changes in osteoarthritis, A. Larsen (1987) proposed a more complex technique that allows for a quantitative assessment of the severity of osteoarthritis.

Criteria for osteoarthritis (Larsen A., 1987)

• 0 - No radiographic signs
• I - Narrowing of the radiographic joint space by less than 50%
• II - Narrowing of the radiographic joint space by more than 50%
• III - Weak remodulation
• IV - Average remodulation
• V - Expressed remodulation

Early radiological signs (correspond to stages I-II of arthrosis according to Kellgren):

• stretching and sharpening the edges of the intercondylar eminence of the tibia (at the site of attachment of the cruciate ligament);
• slight narrowing of the joint space (usually in the medial part of the joint);
• sharpening of the edges of the articular surfaces of the condyles of the femur and tibia, more often in the medial part of the joint (associated with a greater load on this part of the joint), especially in the presence of varus deformity; less often - in the lateral part or simultaneously in both halves of the articular surface.

Radiological signs of progression of arthrosis of the knee joints (correspond to stage III-IV of arthrosis according to Kellgren):

• increase in narrowing of the radiographic joint space;
• development of subchondral osteosclerosis in the most loaded part of the joint;
• the appearance of multiple large osteophytes on the lateral, anterior and posterior edges of the articular surfaces;
• subchondral cysts (rarely found);
• secondary synovitis with the development of subpatellar or popliteal Baker's cyst;
• flattening and unevenness of the articular surfaces of the femur and tibia, loss of their anatomical and functional differentiation;
• polyhedral irregular shape of the sesamoid bone (fabella);
• it is possible to detect calcified chondromata;
• development of aseptic necrosis of bone condyles is possible (rare).

Quite often, osteoarthritis of the knee joints manifests itself in the form of arthrosis

PFO (almost always external, sometimes external and internal, rarely only internal).

External osteoarthrosis of the knee joint usually manifests itself at the beginning of its development at the level of the upper cartilaginous sector of the intercondylar groove and the lower cartilaginous sector of the patella, corresponding to the part of the knee joint that is visualized in this projection. The greatest load on the subchondral sections of the bones is noted at the very beginning of knee flexion, at the moment when the patella begins to enter the intercondylar fossa. Therefore, changes in the PFO of the joint are quite common, but, as a rule, are rarely diagnosed in time. The main reason for the untimely diagnosis is that in practice, radiographic axial projections are not used sufficiently. Therefore, direct radiography of the knee joints must be supplemented with a targeted image of the patella in the lateral or axial projection.

Radiological signs of osteoarthritis of the knee joint in lateral and axial projections include:

• narrowing of the radiographic space between the patella and the femur;
• OF on the posterior angles of the patella and femoral condyles;
• subchondral osteosclerosis of the patella;
• single subchondral cysts with a sclerotic rim. It should be noted that radiologically, three stages of osteoarthritis are distinguished

Subchondral osteocondensation and increased trabecular pattern of the outer edge of the patella, which experiences the greatest external loads ("hyperpressure syndrome"), correspond to stage I arthrosis. At stage II, there is infringement (local narrowing) of the joint space, even in the absence of signs of patellar subluxation. Stage III arthrosis of the knee joint is characterized by almost complete disappearance of the radiographic joint space, compaction of the subchondral cortical layer, in the thickness of which rarefaction areas are formed - cortical cysts, and the appearance of perichondral osteophyte beak-shaped formations. Detection of marginal osteophytes of the patella allows us to assume with a high degree of certainty damage to the articular cartilage. Their presence along the contours of the outer and inner condyles of the femur and tibia indicates damage to the meniscus of the corresponding side. Severe arthrosis most often occurs when the axis of the patella is displaced due to its external subluxation, which occurs as a result of dysplasia or disruption of the articular relationships of the PFO articulation.

Using an axial image at 30° also allows one to calculate the Bernageau index - the distance between the anterior tibial tuberosity and the intercondylar fossa, which normally ranges from 10 to 15 mm. A decrease or increase in this distance usually indicates dysplasia of the femoral condyles or patella, which is expressed in instability of the PFO joint.

Studying the X-ray joint space of the PFO with the knee flexed at 60 and 90° allows for a detailed study of the middle and lower parts of the intercondylar space and the upper part of the patella. Pathological changes in these areas are usually observed later than in the upper parts of the intercondylar fossa.

The standard assessment of joint radiographs according to Kellgren and Lawrence is mainly suitable for use in everyday clinical practice. More detailed classification of the severity of osteoarthritis is often required in clinical and epidemiological studies. For this purpose, the height of the joint space of the knee joint is measured with a thin plastic ruler graduated in 0.5 mm or with calipers. Such quantitative assessment will be more accurate if special computer programs for processing radiographs are used.

JC Buckland-Wright et al. (1995) proposed measuring the height of the radiographic joint space (in mm) on macroradiographs of the knee joints in the outer, middle and inner thirds of the TFO medially and laterally.

It is obvious that in assessing the radiographs of joints of patients with osteoarthrosis, it is impossible to limit oneself to the study of the height of the joint space, therefore, semi-quantitative assessment methods, which are widely used in large-scale clinical and epidemiological studies, are more preferable. All these methods have a common principle - the most important radiographic symptoms of osteoarthrosis (height of the joint space, osteophytosis, subchondral sclerosis, subchondral cysts) are assessed in points or degrees (usually from 0 to 3).

One of the first to propose a semi-quantitative assessment of knee joint radiographs was S. Аbаск (1968). According to this method, the four above-mentioned radiographic criteria of osteoarthritis are assessed in points from 0 to 3 in the PFO and TFO. The main disadvantages of this scale are: the lack of assessment of the PFO of the knee joint and the high probability of ambiguous interpretation of radiographic symptoms by different specialists. A similar system was developed by RD Altaian et al. (1987). Taking into account the main disadvantage of these two systems (assessment of only the TFO of the knee joint), TD. Spector et al. (1992) proposed a method for semi-quantitative assessment of knee joint radiographs in the "sunrise" projection, which allows for an optimal examination of the PFO. In the "Radiographic Atlas of Osteoarthritis" by S. Barnett et al. (1994), an assessment in the standard lateral projection was added to the assessment of the PFO of the joint in the "sunrise" projection.

We propose our own method for semi-quantitative assessment of gonarthrosis progression:

1. Reduction in the height of the joint space:

• 0 - absent,
• 1 - minor,
• 2 - moderate,
• 3 - complete obliteration of the interosseous space;

2. Osteophytes:

• 0 - absent,
• 1 - 1-2 small osteophytes,
• 2 - one large or 3 small osteophytes or more,
• 3 - 2 large osteophytes or more;

3. Subchondral cysts:

• 0 - absent,
• 1 - 1-2 small cysts,
• 2-1 large or 3 small cysts or more, 3-2 large cysts or more;

4. Subchondral sclerosis:

• 0 - absent,
• 1 - minor, local (in the medial or lateral part of the TFO or PFO joint),
• 2 - moderate,
• 3 - significantly pronounced, widespread.

RD Altman et al. (1995) combined a semi-quantitative assessment of both parts of the knee joint into a single system and published the "Atlas of Individual Radiographic Symptoms of Osteoarthritis", which was also called the "ORS Atlas". The advantages of this system also include the fact that it contains real radiographs of knee joints with osteoarthritis. Along with this, the "ORS Atlas" has a number of disadvantages. Among them, the following can be highlighted:

• the gradations of narrowing of the joint space and increase in the size of osteophytes have unequal intervals,
• Some knee radiographs show rare types of osteophytes,
• The quality of X-ray images varies, making comparison difficult,
• the presence of several radiographic symptoms (narrowing of the joint space, osteophytosis, etc.) on one X-ray image, which complicates working with the Atlas and can lead to a biased assessment of real X-ray images,
• The large volume of the Atlas makes it difficult to use.

Y Nagaosa et al. (2000) took into account the shortcomings of previous systems of semi-quantitative assessment of knee joint radiographs and developed their atlas, the illustrative material of which is a graphic image of the contours of the knee joint components in the direct projection (TFO joint) and in the "sunrise" projection (PFO joint). An important advantage of the system of Y Nagaosa et al. is not only that they separately consider the medial and lateral parts of the TFO and PFO of the knee joint, but also that the radiographic symptoms of osteoarthritis are presented separately for men and women.

In a study of 104 patients with confirmed osteoarthritis of the knee joints (according to the ACR criteria, 1990), we studied the size and direction of osteophyte growth and assessed possible relationships between their size and other radiographic data associated with osteophyte growth.

Standard radiographs of both knee joints were analyzed (except for patients who underwent patellectomy or arthroplasty). Radiologically, gonarthrosis was defined as the presence of uniform or uneven narrowing of the radio-articular space and marginal osteophytes (ACR criteria, 1990). Radiography of the knee joints was performed in standard projections: anteroposterior with full extension of the lower limbs and axial.

When evaluating radiographs, the knee joint was conventionally divided into sections in accordance with modern recommendations: lateral and medial TFO, lateral and medial PFO. Narrowing of the radioarticular space in each of these sections, as well as the sizes of osteophytes in each of the 6 areas: lateral and medial articular surfaces of the femur (LB and MB, respectively), tibia (LBB and MBB), patella (LN and MN), as well as osteophytes of the lateral and medial condyles of the femur (LM and MM) were assessed on a scale from 0 to 3 according to the Logically derived line drawing atlas for grading of knee osteoarthritis certification system. The direction of osteophyte growth was visually divided into 5 categories - upward (ascending growth), upward laterally, laterally, downward laterally or downward (descending growth).

Cortical bone deformity (local bone deformity or “wear and tear”) and chondrocalcinosis in the TFO and PFO were graded using a 2-point system (0 = absent, 1 = present). The tibiofemoral angle, an indicator of varus deformity, was graded in the anteroposterior projection. Patellar subluxation on axial knee images was graded 0-1 medially and 0-3 laterally. Joint space narrowing in each region studied and lateral patellar subluxation were also graded 0-3, respectively.

In 92 patients, a close correlation was found between the radiographic data of the right and left knee joints.

Osteophytes were found in all areas studied, and various forms and directions of their growth were noted.

Correlation coefficient (r) of some radiographic parameters between the right and left knee joints

The analyzed indicator	Correlation coefficient (r)
	Minimum	Maximum
Narrowing of the RSCh	0.64	0.78
Presence of osteophytes	0.50	0.72
Localized bone deformation	0.40	0.63
Chondrocalcinosis	0.79	0.88

Some relationships between the presence of osteophytes and their sizes with other radiographic data

Localization of OF	Total number of OF	Direction of OF growth (difference between 0-1 and 2-3 degrees of OF size)	Direction of growth of the OF (difference between 0-1 and 2-3 degrees of local narrowing of the RSH)
LB	42	P=0.011	P=0.006
LBB	48	P>0.1	P<0.001
MB	53	P=0.003	P=0.001
MBB	49	P<0.05	P<0.05
LN	28	P=0.002	P>0.1
LM	30	P>0.1	P<0.001
MN	28	P>0.1	P>0.1
MM	34	P=0.019	P>0.1

Similar patterns were observed when analyzing the direction of osteophyte growth depending on the degree of local narrowing of the joint space. In LB, MB, MBB, LM, the severity of local narrowing of the gap was associated with the direction of growth of large osteophytes. The direction of osteophyte growth in LBB was associated not with the size of osteophytes, but with local narrowing of the joint space of the lateral and medial TFO, and in MN it did not correlate with either the size of osteophytes or the degree of local narrowing.

A positive correlation between the size of osteophytes and the degree of local joint space narrowing was found in all regions except the medial PFO. In the latter, the sizes of osteophytes in the patella and MM positively correlated with the narrowing of the medial TFO space. The size of osteophytes in the LB and LBB of the lateral TFO positively correlated with the degree of narrowing of the lateral PFO.

To clarify the relationships between some radiographic and general clinical data with osteophyte size, the latter were analyzed using multivariate analysis.

Local space narrowing was associated with the presence of osteophytes in most of the analyzed sites. Osteophytes in the LBB were associated with medial TFO and lateral PFO space narrowing. Osteophytes in the LN and LM correlated more with lateral patellar subluxation than with local narrowing. Grades 2-3 medial PFO osteophytes were not associated with local narrowing, but were associated with varus deformity and medial TFO space narrowing. The degree of local TFO deformity was associated with the presence of grade 2-3 osteophytes in both lateral and medial TFOs.

Factors associated with the presence of osteophytes, depending on the size of the latter (above) both in the lateral TFO and (osteophytes of 2-3 degrees) in the lateral PFO. Chondrocalcinosis was caused by the growth of osteophytes in many areas. The presence of lateral patellar subluxation closely correlated with the growth of osteophytes in the lateral PFO, and varus deformity - with the presence of osteophytes of 2-3 degrees in the medial TFO. The total number of osteophytes correlated with the number of osteophytes in the MB and MM.

Region	Factor
	Osteophytes 0-1 degree	Osteophytes 2-3 degrees
LB	Local deformation of the PFO	Chondrocalcinosis
	Chondrocalcinosis	Local deformation of the TFO
	Narrowing of the joint space of the lateral TFO
LBB	Chondrocalcinosis	Female gender
	Local deformation of the PFO	Chondrocalcinosis
	Narrowing of the joint space of the lateral PFO	Local deformation of the TFO
	Narrowing of the joint space of the medial TFO
MB	Lateral subluxation of the patella	Local deformation of the TFO
	Narrowing of the joint space of the medial TFO	Total number of osteophytes
	Female gender	Female gender
		Varus deformity
MBB	Local deformation of the TFO	Chondrocalcinosis
	Narrowing of the joint space of the medial TFO	Age
		Varus deformity
LN	Local deformation of the PFO	Local deformation of the PFO
	Lateral subluxation of the patella	Lateral subluxation of the patella
	Chondrocalcinosis	BMI
	BMI
LM	Lateral subluxation of the patella	Lateral subluxation of the patella
	Localized chondromalacia of the PFO	Narrowing of the joint space of the lateral FO
	Chondrocalcinosis	Varus deformity
	Medial subluxation of the patella
MN	Narrowing of the joint space of the medial PFO	Varus deformity
MM	Narrowing of the joint space of the medial TFO	Narrowing of the joint space of the medial TFO
		Total number of OF
		BMI

The sizes of osteophytes growing towards each other in the same section correlated in all analyzed sections: the correlation coefficient r was 0.64 for the lateral TFO, 0.72 for the medial TFO, 0.49 for the lateral PFO, and 0.42 for the medial PFO.

Consequently, in all parts of the knee joint, except for the LBB and MN, the direction of osteophyte growth changes with an increase in the size of the latter and the degree of narrowing of the joint space. The discovered correlations support the hypothesis about the influence of both general and local biomechanical factors on the formation of osteophytes. The influence of the latter is evidenced by the correlation we discovered between such parameters as:

• the size of osteophytes in the medial PFO and narrowing of the medial TFO gap;
• the size of the LBB osteophytes and the narrowing of the gap of both the medial TFO and lateral PFO;
• size of osteophytes in the lateral PFO and lateral subluxation of the patella;
• the size of osteophytes of the medial TFO and PFO and the presence of varus deformity. On the contrary, when analyzing the relationships between chondrocalcinosis and the total number of osteophytes, multidirectional changes were found.

It can be assumed that local instability is an important triggering biomechanical mechanism for osteophyte formation. Experimental models of osteoarthrosis have demonstrated that osteophyte formation in unstable joints accelerates with movements in this joint and slows down with immobilization. As noted by LA Pottenger et al. (1990), surgical removal of osteophytes during knee arthroplasty in patients with osteoarthrosis leads to worsening of joint instability, which allows us to speak about the stabilizing role of osteophytes in this pathology. Our observation that lateral growth of osteophytes promotes an increase in the area of the loaded articular surface is confirmed by the data obtained by JM Williams and KD Brandt (1984). For small osteophytes, the predominant growth direction is lateral (with the exception of the LBB, where osteophytes grow predominantly upward, provided that the gap of the medial TFO is narrowed, and the lateral TFO is minimally involved in the process). LA. Pottenger et al. (1990) showed that even vertical osteophytes can stabilize the joint, apparently by creating a newly formed tibial surface and limiting excessive valgus movement. In contrast to the small osteophyte, the large osteophyte grows predominantly upward or downward. This phenomenon may reflect anatomical limitation of "lateral" growth by adjacent periarticular structures or compensatory processes of expansion and mechanical strengthening of the osteophyte base to prevent dislocation.

Among such compensatory changes, it is necessary to mention the so-called tide lines, which are calcification zones connecting the hyaline cartilage with the subchondral bone. Normally, they are wavy and therefore effectively counteract significant loads. In osteoarthrosis, due to the fact that the cartilage is destroyed, and new cartilage is formed in the form of osteophytes, this zone is rebuilt. Therefore, one of the manifestations of osteoarthrosis is the presence of multiple tide lines. Since the articular surface of the bone is exposed, the compensatory mechanism is the formation of dense sclerosis (eburnation), often combined with the formation of deep grooves (depressions). The latter are especially often found in the knee joint (PFO), where they can be considered a means of stabilizing the joint, providing it with "rails". These grooves were well visualized in axial images of the PFO in the patients we examined.

A close correlation was observed between the osteophyte size and local cartilage thinning, especially in the medial TFO and lateral PFO. However, the osteophyte size in the lateral TFO correlated more with the narrowing of the joint spaces of the medial TFO and lateral PFO, rather than its own joint space, and the osteophyte size in the medial PFO correlated not with local space narrowing, but with the narrowing in the medial TFO. Apparently, the osteophyte size can be influenced by both adjacent and local changes in the joint, which can be mediated by biochemical or mechanical growth factors. The latter can most likely explain the relationship between the osteophyte sizes of the medial TFO and PFO with varus deformity. G. I. van Osch et al. (1996) suggested that the processes of cartilage damage and osteophyte formation are not directly related, but are caused by the same factor and develop independently of each other. Such independent development is observed in the lateral PFO and medial TFO, and the size of the osteophytes is associated more with lateral patellar subluxation and varus deformity than with local narrowing of the joint space.

The association between the total number of osteophytes and their distribution at several sites supports the concept of a constitutional determinacy of osteophyte formation and a “hypertrophic” bone response. There may be individual differences in the response to some growth factors, such as TGF-beta or bone morphogenic protein-2, which is involved in osteophyte growth. An interesting observation is the association between chondrocalcinosis and the number of osteophytes: clinical studies suggest a specific relationship between calcium pyrophosphate crystals (a common cause of chondrocalcinosis) and the “hypertrophic” outcome of osteoarthritis. TGF-beta, in addition to stimulating osteophyte growth, increases the production of extracellular pyrophosphate by chondrocytes, and mechanical stimulation of chondrocytes increases the production of ATP, a potent source of extracellular pyrophosphate, thereby predisposing to the formation of crystals of the latter.

The data we obtained suggest the involvement of a number of factors in the pathogenesis of osteoarthritis, including local biomechanical, constitutional and others, which determine the size and direction of growth of osteophytes formed during the progression of the disease.