These human and animal studies highlight the importance of considering the patellofemoral compartment in studies of knee osteoarthritis. Patellofemoral osteoarthritis is common in both men and women, associated with disability, and frequently independent of tibiofemoral disease. Furthermore, the disparity of patellofemoral osteoarthritis across the articulating surfaces of the joint is intriguing. In both human and animal subjects, softening, fibrillation, proteoglycan loss, and chondrocyte clustering occur earlier in the disease process, are more severe, and cover a larger portion of the cartilage surface of the patella compared with the femoral groove. Careful consideration of the mechanics of the patellofemoral joint and differences between patellar and femoral groove cartilages may hold crucial insights into the etiology of osteoarthritis. In the following discussion, we will consider three hypotheses that may influence the disparate progression of patellofemoral osteoarthritis (Fig. 4). These hypotheses include the duration of the load experienced by the articulating surfaces; the histological, material and compositional properties of the opposing cartilages; and the anabolic/catabolic metabolism of patellar and femoral groove chondrocytes. We propose that these factors interact and perhaps confound one another in the disparate etiology of patellofemoral osteoarthritis (Fig. 4).
HYPOTHESIS 1: LOAD DURATION
The directly articulating cartilages of the patellofemoral joint experience equal and opposite contact stresses (Newton's third law of action and reaction). Despite an equal magnitude of load, the duration of the loads on a particular area of each surface may differ considerably. For example, one might expect areas of the patellar articular cartilage to be more consistently loaded throughout the range of motion of the knee compared with the intermittent loading of the femoral groove as the patella slides along its length. Intermittent loading may enable fluid flow to flush out waste products and imbibe nutrients that are essential to the maintenance and regeneration of articular cartilage. Static load may reduce this effect, leaving the avascular tissue devoid of nourishment. It is interesting to note that in the 5-yr post-ACLT feline knees exhibiting disparate disease in the patellofemoral compartment, a similar phenomenon was evident in the experimental tibiofemoral compartments (8). The more consistently loaded tibial plateaus had more severe cartilage damage (full- or partial-thickness erosions) compared with the intermittently loaded femoral condyles (texture changes and only one partial thickness erosion).
In parallel to the work previously described, there is a portion of literature comparing static and dynamic loading of explants of articular cartilage and the effects on the biological response of the tissue. Palmoski and Brandt (26) applied static or dynamic stress to articular cartilage taken from healthy adult dogs. Glycosaminoglycan synthesis was suppressed to 30%-60% of that in controls, and protein synthesis also decreased with static stress. In contrast, when the load was cycled, glycosaminoglycan synthesis was increased by 34%, and there was no effect on protein synthesis. Sah et al. (27) examined the biosynthetic response of calf articular cartilage to dynamic unconfined compression. At high frequencies (0.01-1 Hz), strains of only 1.5% increased 3H-proline and 35S-sulfate incorporation by 20%-40%. In contrast, at low frequencies (0.0001-0.001 Hz), strains less than 5% had no effect. The results from these studies support the notion that static load suppresses the biological response of cartilage, whereas dynamic cyclical load stimulates synthesis. The more consistent loading of the patella compared with the femoral groove may therefore predispose it to impaired biological activity and thus accelerated progression of patellofemoral osteoarthritis.
HYPOTHESIS 2: HISTOLOGICAL, MATERIAL, AND COMPOSITIONAL PROPERTIES
In our study of the feline patellofemoral joint, we began with a thorough histological analysis of the cartilage thickness, chondrocyte shape, and chondrocyte volumetric fraction distribution throughout the depth of healthy patellar and femoral groove tissues (5). We measured these parameters from ruthenium hexamine trichloride/glutaraldehyde-fixed sections taken from the center of the patella or femoral groove. These samples were fixed in situ, still fully intact and attached to their complete native bone under control or loaded (9 MPa) conditions. Nine megapascal was chosen as the average patellofemoral contact pressure measured in situ using Fuji film when 170 N of force was applied across the joint (6). The peak patellofemoral contact force measured in vivo during normal cat gait is 170 N (14).
Cartilage Thickness and Strain
We found that patellar cartilage was approximately twice as thick as femoral groove tissue (Figs. 2, 5). The extra thickness seemed to be due to an extended deep zone of patellar articular cartilage (Figs. 2, 5). For a given load, the femoral groove articular cartilage experienced less strain than the patellar tissue. Furthermore, total tissue strain seemed to be distributed more evenly throughout the layers of the femoral groove compared with the patellar cartilage (Fig. 5). The patellar articular cartilage seemed to experience the greatest strains in the middle zone and the smallest strains in the deep zone (Fig. 5).
Chondrocyte Aspect Ratio
In both patellar and femoral groove feline cartilages, the cells were organized in a typical zonal arrangement (Figs. 2, 6). At the surface, chondrocytes were elliptical in shape and horizontally orientated; the cells in the middle layer were more rounded in shape, and in the deep layer, the cells were elliptical in shape, vertically oriented, and often arranged in columns (P < 0.001). In unloaded samples, the chondrocytes in the superficial zone were more rounded and found within a smaller relative depth of tissue in the patella compared with those in the femur (P < 0.001) (Figs. 2, 6). The deep zone columns of cells were also more numerous in the patellar cartilage compared with those in the femoral cartilage. When load was applied to the cartilage, chondrocytes flattened throughout the depth of both femoral and patellar tissues (P < 0.001) (Fig. 6). Consistent with total tissue strain, the femoral groove chondrocytes were compressed more uniformly throughout the cartilage depth compared with the patella (Fig. 6). Patellar chondrocytes seemed to be compressed more in the upper 40% of the cartilage depth than the femoral chondrocytes. However, in the bottom 40% of the tissue, the patellar chondrocytes seemed to experience similar or less change in aspect ratio than those of the femoral groove.
Chondrocyte Volumetric Fraction
The volumetric fraction of feline chondrocytes was greater in the patellar than in the femoral groove articular cartilage (P < 0.05) and decreased from the superficial zone to the deep layer in both tissues (P < 0.05) (Fig. 7). The difference between patellar and femoral cartilage was particularly noticeable in the middle layer where the femoral chondrocyte volumetric fraction was markedly smaller than the corresponding patellar value (Fig. 7). Semiquantitative analysis involving measurements of chondrocyte cross-sectional area and visual inspection of morphological sections suggested that this contrast in volumetric fraction was due to both larger and more numerous chondrocytes in the patellar cartilage than in the femoral groove cartilage. Chondrocyte volumetric fraction decreased with load in all layers of both tissues (P < 0.05) (Fig. 7). The magnitude of this change was greater in all layers of the patellar cartilage than in those of the femoral cartilage (P < 0.05).
Material Properties and Biochemical Composition
Other authors have measured the material properties and biochemical composition of the patellofemoral cartilages. Froimson et al. (13) used a needle probe to measure cartilage thickness in the patellofemoral compartment of fresh-frozen healthy human knees. Compressive aggregate modulus, permeability, and Poisson's ratio were also determined for these samples by biphasic indentation testing. Full-thickness cartilage samples adjacent to the indentation sites were taken to measure wet weight, sulfated glycosaminoglycan content, and hydroxyproline content. Numerous significant differences were found between the juxtapose patellofemoral cartilages (Table 1). Patellar cartilage was thicker (P < 0.05) and had a lower compressive aggregate modulus (P < 0.001) and larger permeability to fluid flow (P < 0.001) than the femoral cartilage. The water content of the patella was higher (P < 0.05), and the proteoglycan content was lower (P < 0.05), than that of the femur. No differences were found between the Poisson's ratio and the collagen contents of the tissues (Table 1).
Together, these studies demonstrate structural and compositional differences between the patellar and femoral groove cartilages. Differences exist between unloaded patellar and femoral groove articular cartilage in parameters such as cartilage thickness, chondrocyte shape, and chondrocyte volumetric fraction in both magnitude and depth distribution. Furthermore, under identical applied loads, changes to all of these parameters differ in magnitude and depth distribution between patellar and femoral groove articular cartilage. The patellar is more easily compressed, with a larger water content that is better able to flow out of the tissue during compression, compared with the more proteoglycan-rich femoral groove. These structural and compositional differences between the patellar and femoral groove cartilages may indicate the function of these two surfaces within the patellofemoral joint. The patellar cartilage may play the dominant role in maximizing patellofemoral joint congruence through cartilage deformation during in vivo loading. We hypothesize that the more readily deformed patellar cartilage and chondrocytes may result in predisposing them to altered structural damage and/or biosynthetic activity compared with the femoral groove. One might expect that a larger deformation may create higher localized stresses that could lead to fissuring at the surface of the patellar cartilage close to the edges of contact with the femoral groove. Furthermore, the larger changes in chondrocyte aspect ratio and volumetric fraction may stimulate a more robust metabolic response in the patellar than in the femoral groove chondrocytes.
HYPOTHESIS 3: CHONDROCYTE ANABOLIC/CATABOLIC METABOLISM
In situ Messenger Ribonucleic Acid Response to Muscle-Induced Cyclical Load
We recently developed a novel experimental technique to cyclically load intact lapine patellofemoral joints using quadriceps muscle stimulation (9). The cartilages were cyclically loaded for 2 s every 30 s for 1 h, resulting in an average contact pressure of 4.6 MPa. Cartilage was harvested from central and peripheral regions of the patellar and femoral groove surface, either immediately after loading or after 3 h of recovery. Biological response was assessed on the messenger ribonucleic acid (mRNA) level using reverse transcriptase-polymerase chain reaction (PCR).
RNA concentration (micrograms of RNA per milligram of tissue wet weight) was not significantly influenced by load, location (patella or femoral groove), site (central or peripheral), or harvest time (immediate or 3-h postload). Reverse transcriptase-PCR analysis of RNA from the control and experimental animals revealed that mRNA levels for tissue inhibitor of metalloprotease-1 (TIMP-1), matrix metalloprotease-3 (MMP-3), and basic fibroblast growth factor (bFGF) from the immediate sacrifice group were all significantly affected by the loading protocol. On average, mRNA levels for TIMP-1, MMP-3, and aggrecan were larger in femoral groove compared with those in patellar tissue and vice versa for bFGF and biglycan. Furthermore, tissue from the peripheral harvest sites had significantly greater mRNA levels for decorin compared with that from the central site. In complete contrast, after a 3-h recovery period, neither load nor location had a significant effect on mRNA levels for any of the genes assessed, although harvest site influenced bFGF mRNA levels. Therefore, the rapid disparate changes in mRNA levels immediately after load seemed to be transient. These results support the notion that the opposing cartilage surfaces of the patellofemoral joint respond metabolically to load and are heterogeneous at the cell metabolism level in an in vivo setting.
Proportion of Superficial and Deep Zone Chondrocytes
One of the interesting findings from our histological analyses of the healthy feline patellofemoral joint outlined previously was that the extra thickness of the patellar groove cartilage, compared with femoral groove cartilage, seemed to be the result of an additional depth of deep zone (Figs. 2, 5). There is further evidence in the literature supporting differences in the metabolic capacities of chondrocytes isolated from the superficial zone as compared with the deep zone of cartilage (1,2,19). Deep zone chondrocytes synthesize significantly more proteoglycans and collagens than those from the superficial zone in both explant and isolated chondrocyte culture. In isolated chondrocytes, these differences seem to be retained even after 100-fold expansion. Synthesis and secretion of the superficial-zone protein from superficial zone but not from middle or deep zone chondrocytes has also been reported in both explant and enzymatically digested agarose or alginate cultures. Additionally, superficial and deep zone chondrocytes have been shown to differ in their response to interleukin-1, interleukin-1 receptor antagonist protein, and bone morphogenetic protein (BMP) 2 or BMP-7 overexpression (4,21). These observations together may suggest that the differing proportions of superficial and deep zone chondrocytes in the patellar and femoral groove cartilages may influence the anabolic and/or catabolic capacity of the tissues, thus predisposing the patellar cartilage to osteoarthritic degeneration as opposed to the femoral groove cartilage.
SUMMARY AND CONCLUSIONS
Osteoarthritis is a painful and debilitating disease affecting millions of people worldwide. There is presently no cure for this complex disease that can manifest itself in many joints of the body. Studies of knee osteoarthritis have historically focused on the tibiofemoral compartments, with little or no attention being given to the patellofemoral compartment. In this article, patellofemoral osteoarthritis has been highlighted. Patellofemoral osteoarthritis is a common clinical diagnosis associated with disability and often independent of tibiofemoral disease. Furthermore, the manifestation of patellofemoral osteoarthritis across the articulating surfaces of the joint is disparate, the patella demonstrating more severe signs of degeneration at a younger age or shorter time after injury compared with the juxtaposed femoral groove. Investigation into this disparity may hold crucial insights into the etiology of osteoarthritis.
In this article, we have discussed three main hypotheses that may influence this disparity both independently and collectively: load duration; histological, material, and compositional properties; and chondrocyte anabolic/catabolic metabolism (Fig. 4). The differences between patellar and femoral groove cartilages and their corresponding chondrocytes are numerous (Table 2). Further consideration of the potential interactions between these properties multiplies the complexity of this "simple" joint. For example, the longer load duration experienced by the patellar cartilage may decrease chondrocyte metabolism relative to the femoral groove. This may be countered, however, by the softer patellar cartilage containing a higher percentage of more metabolically active deep zone chondrocytes. The patellar cartilage will undergo greater deformation than the femoral groove cartilage under a given load, thus resulting in larger aspect ratio and volume changes to patellar chondrocytes and a potentially greater influence on their metabolism. It is clear that even in the simple patellofemoral joint, where cartilage interacts directly with cartilage, the mechanical and metabolic interactions are abundant.
It is interesting to look at these data alongside the body of work by Cole and Kuettner (10) and Kuettner and Cole (21) comparing human cartilages of the ankle and knee. Ankle cartilage is more resistant to progressive degeneration and osteoarthritis than knee cartilage. It is also thinner, stiffer, and less permeable and has a higher proteoglycan content compared with knee cartilage. In addition, ankle chondrocytes synthesize more glycosaminoglycan and protein than knee chondrocytes and up-regulate matrix synthesis as opposed to collagen degradation in response to degeneration. It is interesting that a number of these differences between ankle and knee cartilage match those between femoral groove and patellar cartilage (Table 2). The properties of the more osteoarthritis-resistant ankle and femoral groove cartilages have similar relationships to their counterpart knee and patellar cartilages, respectively.
In conclusion, the patellofemoral compartment of the knee should be considered in addition to the tibiofemoral compartments when diagnosing and investigating knee osteoarthritis. Patellofemoral osteoarthritis is disparate across the joint: the patellar side demonstrating more severe degeneration earlier in the disease process compared with the juxtaposed femoral groove. Duration of the load experienced by the articulating surfaces; the histological, material, and compositional properties of the opposing cartilages; and the anabolic/catabolic metabolism of the chondrocytes from the patellar and femoral groove may all influence the disparate etiology of patellofemoral osteoarthritis.
This study was supported by a postdoctoral fellowship from the Arthritis Foundation.
1. Aydelotte, M.B., R.R. Greenhill, and K.E. Kuettner. Differences between sub-populations of cultured bovine articular chondrocytes. II. Proteoglycan metabolism
. Connect. Tissue Res
. 18:223-234, 1988.
2. Aydelotte, M.B., and K.E. Kuettner. Differences between sub-populations of cultured bovine articular chondrocytes. I. Morphology and cartilage
matrix production. Connect. Tissue Res
. 18:205-222, 1988.
3. Brandt, K.D. Animal models of osteoarthritis. Biorheology.
4. Cheng, C., E. Conte, N. Pleshko-Camacho, and C. Hidaka. Differences in matrix accumulation and hypertrophy in superficial and deep zone chondrocytes are controlled by bone morphogenetic protein. Matrix Biol
. 26:541-553, 2007.
5. Clark, A.L., L.D. Barclay, J.R. Matyas, and W. Herzog. In situ chondrocyte
deformation with physiological compression of the feline patellofemoral joint. J. Biomech
. 36:553-568, 2003.
6. Clark, A.L., W. Herzog, and T.R. Leonard. Contact area and pressure distribution in the feline patellofemoral joint under physiologically meaningful loading conditions. J. Biomech
. 35:53-60, 2002.
7. Clark, A.L., T.R. Leonard, L.D. Barclay, J.R. Matyas, and W. Herzog. Heterogeneity in patellofemoral cartilage
adaptation to anterior cruciate ligament transaction: chondrocyte
shape and deformation with compression. Osteoarthritis Cartilage.
8. Clark, A.L., T.R. Leonard, L.D. Barclay, J.R. Matyas, and W. Herzog. Opposing cartilages in the patellofemoral joint adapt differently to long-term cruciate deficiency: chondrocyte
deformation and reorientation with compression. Osteoarthritis Cartilage.
9. Clark, A.L., L. Mills, D.A. Hart, and W. Herzog. Muscle-induced patellofemoral joint loading rapidly affects cartilage
mRNA levels in a site specific manner. J. Musculo. Res
. 8:1-12, 2004.
10. Cole, A.A., and K.E. Kuettner. Molecular basis for differences between human joints. Cell. Mol. Life Sci
. 59:19-26, 2002.
11. Emery, I.H., and G. Meachim. Surface morphology and topography of patello-femoral cartilage
fibrillation in Liverpool necropsies. J. Anat
. 116:103-120, 1973.
12. Felson, D.T., R.C. Lawrence, P.A. Dieppe, R. Hirsch, C.G. Helmick, J.M. Jordan, R.S. Kington, N.E. Lane, M.C. Nevitt, Y. Zhang, M. Sowers, T. McAlindon, T.D. Spector, A.R. Poole, S.Z. Yanovski, G. Ateshian, L. Sharma, J.A. Buckwalter, K.D. Brandt, and J.F. Fries. Osteoarthritis: new insights. Part 1: the disease and its risk factors. Ann. Intern. Med
. 133:635-646, 2000.
13. Froimson, M.I., A. Ratcliffe, T.R. Gardner, and V.C. Mow. Differences in patellofemoral joint cartilage material properties
and their significance to the etiology of cartilage
surface fibrillation. Osteoarthritis Cartilage.
14. Hasler, E.M., and W. Herzog. Quantification of in vivo
patellofemoral contact forces before and after ACL transection. J. Biomech
. 31:37-44, 1998.
15. Hasler, E.M., W. Herzog, T.R. Leonard, A. Stano, and H. Nguyen. In vivo knee
joint loading and kinematics before and after ACL transection in an animal model. J. Biomech
. 31:253-262, 1998.
16. Herzog, W., M.E. Adams, J.R. Matyas, and J.G. Brooks. Hindlimb loading, morphology and biochemistry of articular cartilage
in the ACL-deficient cat knee
. Osteoarthritis Cartilage.
17. Herzog, W., S. Diet, E. Suter, P. Mayzus, T.R. Leonard, C. Muller, J.Z. Wu, and M. Epstein. Material and functional properties of articular cartilage
and patellofemoral contact mechanics in an experimental model of osteoarthritis. J. Biomech
. 31:1137-1145, 1998.
18. Hjelle, K., E. Solheim, T. Strand, R. Muri, and M. Brittberg. Articular cartilage
defects in 1,000 knee
19. Hu, J.C., and K.A. Athanasiou. Chondrocytes from different zones exhibit characteristic differences in high density culture. Connect. Tissue Res
. 47:133-140, 2006.
20. Kellgren, J.H., and J.S. Lawrence. Radiological assessment of osteo-arthrosis. Ann. Rheum. Dis
. 16:494-502, 1957.
21. Kuettner, K.E., and A.A. Cole. Cartilage
degeneration in different human joints. Osteoarthritis Cartilage.
22. Mankin, H.J., H. Dorfman, L. Lippiello, and A. Zarins. Biochemical and metabolic abnormalities in articular cartilage
from osteo-arthritic human hips. II. Correlation of morphology with biochemical and metabolic data. J. Bone Joint Surg. Am
. 53:523-537, 1971.
23. McAlindon, T.E., S. Snow, C. Cooper, and P.A. Dieppe. Radiographic patterns of osteoarthritis of the knee
joint in the community: the importance of the patellofemoral joint. Ann. Rheum. Dis
. 51:844-849, 1992.
24. Meachim, G., and I.H. Emery. Quantitative aspects of patello-femoral cartilage
fibrillation in Liverpool necropsies. Ann. Rheum. Dis
. 33:39-47, 1974.
25. Outerbridge, R.E. The etiology of chondromalacia patellae. J. Bone Joint Surg. Br
. 43:752-757, 1961.
26. Palmoski, M.J., and K.D. Brandt. Effects of static and cyclic compressive loading on articular cartilage
plugs in vitro
. Arthritis Rheum
. 27:675-681, 1984.
27. Sah, R.L., Y.J. Kim, J.Y. Doong, A.J. Grodzinsky, A.H. Plaas, and J.D. Sandy. Biosynthetic response of cartilage
explants to dynamic compression. J. Orthop. Res
. 7:619-636, 1989.
28. Seedholm, B.B., T. Takeda, M. Tsubuku, and V. Wright. Mechanical factors and patellofemoral osteoarthrosis. Ann. Rheum. Dis
. 38:307-316, 1979.
29. Suter, E., W. Herzog, T.R. Leonard, and H. Nguyen. One-year changes in hind limb kinematics, ground reaction forces and knee
stability in an experimental model of osteoarthritis. J. Biomech
. 31:511-517, 1998.
30. Wieland, H.A., M. Michaelis, B.J. Kirschbaum, and K.A. Rudolphi. Osteoarthritis - an untreatable disease? Nat. Rev. Drug Discov
. 4:331-344, 2005.
Keywords:©2008 The American College of Sports Medicine
cartilage; chondrocyte; knee; histology; material properties; metabolism