Cartilage Thickness Mismatches in Patellar Osteochondral Allograft Transplants Affect Local Cartilage Stresses

Ryan Rosario; Ellen M Arruda; John A Grant; Rhima M Coleman

doi:10.1002/jor.25569

. Author manuscript; available in PMC: 2024 Nov 1.

Published in final edited form as: J Orthop Res. 2023 Apr 24;41(11):2372–2383. doi: 10.1002/jor.25569

Cartilage Thickness Mismatches in Patellar Osteochondral Allograft Transplants Affect Local Cartilage Stresses

Ryan Rosario ^1,², Ellen M Arruda ^1,^3,⁴, John A Grant ⁵, Rhima M Coleman ^1,^3,^*

PMCID: PMC10560315 NIHMSID: NIHMS1890358 PMID: 37031360

Abstract

Osteochondral allograft implantation is a form of cartilage transplant in which a cylindrical graft of cartilage and subchondral bone from a donor is implanted into a patient’s prepared articular defect site. No standard exists for matching the cartilage thickness of the donor and recipient. The goal of this study was to use finite element (FE) analysis to identify the effect of cartilage thickness mismatches between donor and recipient cartilage on cartilage stresses in patellar transplants. Two types of FE models were used: patient-specific 3D models and simplified 2D models. 3D models highlighted which geometric features produced high-stress regions in the patellar cartilage and provided ranges for the parameter sweeps that were conducted with 2D models. 2D models revealed that larger thickness mismatches, thicker recipient cartilage, and a donor-to-recipient cartilage thickness ratio (DRCR) <1 led to higher stresses at the interface between the donor and recipient cartilage. A surface angle between the donor-recipient cartilage interface and cartilage surface normal near the graft boundary increased stresses when DRCR>1, with the largest increase observed for an angle of 15°. A surface angle decreased stresses when DRCR<1. Clinical Significance: This study highlights a potential mechanism to explain the high rates of failure of patellar OCAs. Additionally, the relationship between geometric features and stresses explored in this study led to a hypothetical scoring system that indicates which transplanted patellar grafts may have a higher risk of failure.

Keywords: Cartilage, allograft, patella, finite elements, bone

1. Introduction:

Articular cartilage is a connective tissue found on the surface of diarthrodial joints that facilitates locomotion by providing a low friction articulating surface. Articular cartilage injuries are common in the knee, with lesions found in 60% of patients who underwent routine knee arthroscopies¹. These injuries are particularly common for the patella, which accounted for over 20% of observed lesions¹. If left untreated, these lesions cause joint pain and can lead to the initiation and progression of osteoarthritis^2,3. Because cartilage is avascular and thus has limited ability to self-repair, cartilage lesions typically undergo surgical restoration if non-operative management does not successfully alleviate the patient’s symptoms.

Osteochondral allograft (OCA) transplantation is one method to restore the smooth articulating cartilage surface. In OCA transplantation, a cylindrical graft of cartilage and subchondral bone from a donor is transplanted into a recipient’s prepared articular defect site. Recipient defect sites are prepared such that the donor cartilage surface is as congruent as possible with the surrounding recipient cartilage surface. This intervention is the primary repair for large, full-thickness lesions, especially when the subchondral bone is damaged. OCA transplantation has multiple documented benefits. Unlike autologous chondrocyte implantation, OCA can be completed in a single operation with a faster rehabilitation protocol than other interventions for severe defects⁴. Following the procedure, patients indicate large increases in functional outcomes based on IKDC scores⁵.

While these benefits are encouraging, OCA transplantation has multiple limitations. OCAs may not peripherally integrate with the surrounding cartilage and thus create a discontinuous cartilage surface⁶. Furthermore, OCA transplantation has a relatively high level of reoperation. In a meta-analysis of 17 studies that examined patellar OCA throughout the knee joint, Familiari et al. found that 13% of OCAs fail at five years, 21% at 10 years, and 33% at 20 years⁵. In the same study, 30% of patients that underwent OCA transplantation required reoperation, although mostly for smaller procedures such as scar resection⁵. The long-term failure and reoperation rates are higher for patellar OCA transplantation, with 44% of grafts failing at 15 years and 52% of patients requiring reoperation⁶. It is unclear what causes these high failure and reoperation rates. A better understanding of the underlying causes of OCA transplantation failure would lead to improved OCA transplantation protocols and thus improved outcomes for patients.

One potential cause for OCA transplantation failure may be the mismatch between donor and recipient cartilage thickness. Cartilage thickness is not part of current graft selection protocols⁴. Limited radiographic evidence examining cartilage thickness suggests that such mismatches are common following OCA transplantation⁷. According to linear elasticity, these abrupt changes in cartilage thickness cause local increases in stress at the subchondral bone surface when a load is applied to the cartilage surface⁸.

The goal of this study was to identify the effect of mismatches between donor and recipient cartilage thickness on cartilage stresses in a human patellar OCA model. We hypothesized that larger cartilage thickness differences would lead to increased stresses in the patellar cartilage, potentially causing local cartilage damage and initiating failure of the graft construct.

2. Methods:

2.1. Patient-specific 3D model generation

Ten patient-specific 3D models were constructed from nano-CT scans of 16 mm diameter patellar OCA transplantations conducted on cadaveric patellae by a fellowship-trained sports medicine orthopedic surgeon⁹. Individual patellae had been scanned using a Phoenix Nantom S with Phoenix Datos∣x 2 Acquisition, version 2.3.2 (Phoenix X-Ray, GE Inspection Technologies; Wunstorf, Germany). Scan settings of 90 kV, 250 μA, mode 0, 3 frames averaged, 1 skip, 1000 ms exposure time, 1000 images per scan, a diamond-coated tungsten target, and a 0.5 mm aluminum filter were used. Scans were constructed at 40 μm resolution using Phoenix Datos∣x 2 Reconstruction v 2.2.1-RTM (Phoenix X-Ray, GE Inspection Technologies; Wunstorf, Germany). These ten scans were selected from a set of 23 scans and were chosen to be representative of the sample based on age, tibial width, and patellar geometric markers.

3D models were generated from these scans of OCA-transplanted patellae to identify potential geometric features of interest using coarse finite element models and to provide physiologic ranges for those geometric parameters. To generate these models, tissue surfaces were segmented using Dragonfly (Object Research Systems, Montreal, Canada). Donor and recipient cartilage and bone were assumed to be laterally integrated because this lateral integration is a hallmark of a successful cartilage repair. Solid 10-node quadratic C3D10HS tetrahedral meshes were constructed from the segmented surfaces using HyperMesh (Altair HyperWorks, Troy, MI, USA). The average element edge length was 100 μm near the boundary between the donor and recipient cartilage and 600 μm far from the boundary. Element sizes were chosen to balance identification of high-stress regions and computation time. The meshed geometry was then imported into Abaqus 2019 (SIMULIA, Johnston, RI, USA). For these exploratory 3D models, cartilage was modeled as linear elastic with $E = 7 MPa$ and an Poisson’s ratio of $v = 0.46$ ^10,11 (Figure 1). A 1 MPa surface pressure was applied to the articulating surface. However, it should be noted that increasing the applied load leads to a linear increase in the locally observed stresses, so all stresses reported hereafter are shown as multiples of the applied load. The bone surface was fixed in all translations.

OCA-transplanted patellae (a) were scanned using nano-CT (b). The cartilage geometries were segmented from the nano-CT scans and used to generate patient-specific 3D finite element models of patellar cartilage (c). Cartilage was modeled as linear elastic with $E = 7 MPa$ MPa and $v = 0.46$ .

2.2. Simplified 2D model generation

Simplified 2D models were constructed to isolate the impact of geometric features that were observed in the patient-specific 3D models. These models were axisymmetric with a donor cartilage radius of 8 mm and recipient cartilage radius of 28 mm. The donor and recipient cartilage surfaces were flush. The subchondral surface was fixed and a pressure of 0.1 MPa was applied to the surface to enable model convergence by preventing excessive deformation of the compliant surface layer of articular cartilage.

To generate more physiologically relevant predictions, the simplified 2D models of OCA-transplanted patellae incorporate depth-dependent compressive properties. Schinagl et al. conducted confined compression experiments to measure the depth-dependent aggregate modulus of the cartilage matrix after full relaxation¹². Compared to the study by Schinagl et al., this study sought to understand the patellar cartilage stress response at strain rates on the order of 10%/s, which are closer to the activities of daily living¹³. Thus, the depth-dependent material properties measured by Schinagl et al. were scaled, such that the apparent aggregate modulus of the tissue as measured by Schinagl et al. matched the aggregate modulus at higher strain rates as calculated from the Young’s modulus and Poisson’s ratio using the equation:

H_{A} = \frac{E (1 - v)}{(1 + v) (1 - 2 v)}

where $H_{A}$ is the aggregate modulus (Figure 2).

Compressive stiffness as a function of normalized depth. On the x-axis, a normalized depth of 0 refers to the articulating surface, and a normalized depth of 1 refers to the subchondral bone surface. The depth-dependent aggregate modulus after complete relaxation were scaled to reflect the dynamic response of the tissue. This scaling was conducted such that the apparent aggregate modulus of the entire tissue as measured by Schinagl *et al*. matched the apparent aggregate modulus calculated of the tissue during dynamic material characterization tests¹².

Three geometric features that were identified from the 3D model explorations were analyzed using the simplified 2D models: the donor-to-recipient cartilage thickness ratio ( $D R C R = \frac{D o n o r c a r t i l a g e t h i c k n e s s}{R e c i p i e n t c a r t i l a g e t h i c k n e s s}$ ), the recipient cartilage thickness, and the angle between the donor and recipient cartilage (Figure 3). DRCR was varied from 0.33 – 3. Recipient cartilage thickness was varied from 1 – 6 mm. The surface angle deviation between the donor-recipient cartilage interface and the recipient surface normal was varied from 0 – 60°. These ranges were chosen to encompass the variation in each parameter that was observed in the patient-specific 3D models. Geometries were meshed using quadratic eight-node axisymmetric CAX8 elements. Meshes were locally refined at the graft boundary for each model based on the size and desired resolution of the stress concentration. The minimum element size was 1 μm. Additional models were created to understand the impact of the assumptions of linear elasticity, lateral integration, no shear forces, and a rigid subchondral bone surface (Supplemental Material).

Geometry for the simplified 2D models. The dimensions for this geometry were chosen based on the dimensions of CT scans of transplanted patellae. The red square identifies the sharp corner which all subsequent stress vs distance curves reference. The angle between the donor-recipient cartilage interface and the recipient cartilage surface normal is represented by α. Cartilage was modeled with depth-dependent compressive properties, adapted from Schinagl *et al*.

2.3. Data analysis

Maps of cartilage thickness and the angle between the donor-recipient cartilage interface and local surface normal (surface angle deviation) were created for patient-specific 3D models using a custom MATLAB script. The thickness and surface angle deviation maps were divided into twelve 30° wedge-shaped regions. Wedges were defined using an axis that was parallel to the donor-recipient cartilage interface and that passed through the center point of the donor subchondral bone surface (Figure 4a). Thicknesses were averaged for 2 mm on either side of the graft boundary to determine cartilage thicknesses. Surface angle deviations were averaged within 1 mm of the graft boundary. Compressive stress (minimum principal stress) contours were extracted for all models and normalized by the applied pressure. For the simplified 2D models, scatterplots of compressive stress versus distance from the sharp corner along the subchondral bone surface and along the donor-recipient cartilage interface were extracted.

(a) Representative thickness map of the cartilage from one OCA-transplanted patella including boundaries for 12 wedges; (b) histogram of the donor-to-recipient cartilage thickness ratio (DRCR); and (c) histogram of recipient cartilage thickness near the graft boundary. Histograms were created using thickness values within the highlighted region of interest (ROI) from twelve 30° wedge-shaped regions of each patella. DRCR and recipient thicknesses were heterogeneous within and across samples.

The scatterplots of compressive stress versus distance that were extracted from each simplified 2D model were used to propose a hypothetical scoring method for evaluating the impact of cartilage thickness mismatches in OCA-transplanted patellae. For each scatterplot, the distance at which compressive stresses along the donor-recipient cartilage interface and along the subchondral bone surface fell below a given threshold was extracted. Threshold values of 1.5x and 2x the applied pressure were chosen to ensure the findings were insensitive to the choice of threshold value. The size of the high-stress region defined by a given threshold was then determined for each model by calculating the area of the quarter ellipse defined by the distance along the subchondral surface and donor-recipient cartilage interface. A scoring system was iteratively developed, using the key geometric features that were previously identified. The calculated score was then correlated with the size of the high-stress region. The final scoring system was chosen that balanced feasibility of measurement, ease of use, and correlation with the size of the high-strain region.

3. Results

Patient-specific 3D and simplified 2D models both indicated that thickness mismatches between donor and recipient cartilage cause stress concentrations that would otherwise be absent with a thickness-matched graft. Compressive stress contours of the 3D patient-specific models revealed three key geometric features affected the stress concentrations in OCA-transplanted patellae: the DRCR, recipient cartilage thickness, and cartilage surface angle deviations. Because of the assumptions of linear elasticity and no shear forces, the stresses shown here slightly underestimate the stresses likely to occur in a successful in vivo OCA transplant. Modeling cartilage as linear elastic as opposed to hyperelastic slightly underestimates the stresses that occur along the donor-recipient interface (Supplemental Materials, Figure A1). Including a shear force at the cartilage surface can increase compressive stresses at the graft boundary (Supplemental Materials, Figure A2). Modeling the bone layer as rigid instead of deformable had minimal impact on observed stresses (Supplemental Materials, Figure A4).

3.1. Impact of cartilage thickness differences and recipient cartilage thickness

Thickness maps of patient-specific 3D models revealed that large thickness differences between donor and recipient cartilage were common and spatially heterogeneous (Figure 4a). The DRCR for each cartilage wedge ranged from 0.7 to 5.5 with mean 1.5±0.8 (Figure 4b). The average recipient cartilage thickness for each wedge ranged from 0.8 mm to 4.7 mm with mean 2.6±0.8 mm (Figure 4c).

3D finite element analysis revealed that high-stress regions occur in the cartilage of OCA-transplanted patella that have cartilage thickness mismatches. These high-stress regions occurred at the graft boundary near the subchondral bone surface. When the DRCR<1, high stresses were present in the donor cartilage, and stress shielding occurred in the recipient cartilage. When the DRCR>1, high stresses were present in the recipient cartilage, and stress shielding occurred in the donor cartilage (Figure 5a).

(a) Slice of patient-specific 3D model, and simplified 2D models with donor cartilage that is thicker (b) and thinner (c) than recipient cartilage. 3D and 2D models show that stress concentrations caused by cartilage thickness mismatches are localized to the region with thinner cartilage. Additionally, the incorporation of depth-dependent material properties in the simplified 2D models caused the stress concentration to grow in size and to be extend further along the interface than along the subchondral bone surface. This and all subsequent line graphs show compressive stress normalized by the applied surface pressure.

Simplified 2D models reproduced key features that were observed in the patient-specific 3D models while also enabling closer analysis of the underlying physics. As in the 3D models, the location of the high-stress region in the 2D models depended on the DRCR. With the more refined mesh that was possible with 2D models, the high-stress region at the sharp corner between the donor and recipient cartilage could be resolved more clearly (Figure 5b-c). The shape of the high-stress region was ellipsoidal, extending further along the donor-recipient cartilage interface than along the subchondral bone surface (Figure 5b-c, Figure 6). This asymmetric shape was caused the incorporation of depth-dependent compressive properties, which created a discontinuity in compressive stiffness across the donor-recipient cartilage interface in addition to the discontinuity in cartilage thickness.

Compressive stress vs. distance from the sharp corner along the interface for DRCR>1 (a) and DRCR<1 (b). Compressive stress vs. distance from the sharp corner along the subchondral bone surface for DRCR>1 (c) and DRCR<1 (d). Larger thickness and having a DRCR>1 result in higher stresses. Additionally, the stresses along the interface were higher than stresses along the subchondral bone surface.

To compare the effect of simultaneously varying DRCR and recipient cartilage thickness, the distance at which stresses fell below a given threshold along the interface and along the subchondral bone surface was extracted from each simplified 2D model. Local stresses exceeded 1.5x the applied pressure within 9-238 μm of the sharp edge along the subchondral bone surface (Figure 7a) and within 12-453 μm along the donor-recipient cartilage interface (Figure 7b). Local stresses exceeded 2x the applied pressure within 2-71 μm of the sharp edge along the subchondral bone surface (Figure 7c) and within 3-73 μm along the donor-recipient cartilage interface (Figure 7d).

Heat maps of the distance away from the sharp corner along the interface (a) and the subchondral bone surface (b) that stresses exceed 1.5x the applied pressure. Similar heat maps were created using a threshold value of 2x the applied pressure (c, d). The coloration corresponds to the distance in microns. Greater thickness mismatches, thicker recipient cartilage, and a DRCR>1 resulted in larger high-stress regions. This trend held as the threshold value was changed.

The size of the high-stress region depends on the DRCR and the recipient cartilage thickness. If we define the thickness mismatch as

T h i c k n e s s m i s m a t c h = {\begin{matrix} {}^{D o n o r t h i c k n e s s}∕_{R e c i p i e n t t h i c k n e s s} & i f D R C R > 1 \\ {}^{R e c i p i e n t t h i c k n e s s}∕_{D o n o r t h i c k n e s s} & i f D R C R < 1 \end{matrix}

then greater thickness mismatches tended to lead to larger high-stress regions. Additionally, given a constant thickness mismatch, thicker recipient cartilage led to larger high-stress regions. However, the increases in the size of the high-stress region caused by greater thickness mismatches and thicker recipient cartilage depended on the DRCR. Given equivalent thickness mismatches, having a DRCR>1 led to larger high-stress regions. For example, given 2 mm thick recipient cartilage and a high-stress threshold of 1.5x, donor cartilage with a DRCR=3 had a high-stress region 5.8x larger than donor cartilage with a DRCR=0.33=1/3 (Figure 7a,b).

3.2. Impact of surface angle deviations

Surface angle deviation maps of patient-specific 3D models revealed that large surface angle deviations were common and spatially heterogeneous (Figure 8a). The average surface angle deviation for each wedge ranged from 6° to 51° with mean 25±11° (Figure 8b).

(a) Contour map of the surface angle deviation for one patella including the boundaries for the twelve wedges; (b) histogram of the average surface angle deviation of the wedges using the region highlighted in the ROI in (a). Surface angle deviation is defined as the angle between the donor-recipient cartilage interface and the local surface normal.

Simplified 2D models demonstrated that the impact of surface angle deviations depended on the DRCR. When DRCR>1, small surface angle deviations caused stresses to increase along the interface and subchondral bone surfaces. Stresses fell back down to the value associated with a surface angle deviation of 0° as the surface angle deviation was increased to 60° (Figure 9a,c). When DRCR<1, stresses along the interface and the subchondral bone surface decreased as the surface angle deviation increased, with the stress concentration vanishing when the surface angle deviation was 60° (Figure 9b,d)

Impact of surface angle variations on stress vs distance away from the sharp corner. Figures (a) and (c) used a model with a recipient cartilage thickness of 2 mm, and figures (b) and (d) had a recipient cartilage thickness of 6 mm. For DRCR>1, small surface angles increased stresses close to the sharp corner (a,c). For DRCR<1, greater surface angles caused greater decreases in stresses (b,d). Surface angles caused a greater change in stresses along the interface (a,b) than along the subchondral bone surface (c,d).

The effects of surface angle deviations on stresses varied with the thickness mismatch and the recipient cartilage thickness. When DRCR>1, stress increases along the interface caused by surface angle deviations were greater when the recipient cartilage was thinner (Figure 10a). Stress increases along the subchondral bone surface were greater when the recipient cartilage was thicker and thickness mismatch was greater (Figure 10c). When DRCR<1, stress decreases along the interface were greater when the thickness mismatch was greater (Figure 10b). Stress decreases along the subchondral bone surface were greater when the recipient cartilage was thinner and the thickness mismatch was greater (Figure 10d).

Difference between stresses observed for models with a 30° and 0° surface angle. Along the interface, for DRCR>1, surface angles caused greater stress increases along the interface when the recipient cartilage was thinner (a). For DRCR<1, surface angles caused greater decreases when the thickness mismatch was greater (b). Along the subchondral bone surface, the magnitude of the differences in stress caused by surface angles increased as recipient cartilage thickness and the thickness mismatch increased (c,d).

3.3. Hypothetical scoring method

The relationships among recipient cartilage thickness, donor cartilage thickness, and surface angle deviation that were identified using the simplified 2D models were used to develop a hypothetical scoring method (Table 2). When using a threshold value of 1.5x the applied surface pressure to define the high-stress region, the relationship between this proposed score and the log of the size of the high-stress region had an $R^{2} = 0.92$ (Figure 11). When the threshold was increased to 2x the applied surface pressure, the relationship between this proposed score and the log of the size of the high-stress region had an $R^{2} = 0.89$ .

Table 2.

A hypothetical scoring method for determining the size of the high-stress region caused by thickness mismatches after OCA transplantation in the patella.

Criterion	Score	Condition
A. Recipient cartilage thickness	Recipient cartilage thickness [mm]
B. Surface angulation	abs(Angle [°] – 60) / 15	DRCR > 1
B. Surface angulation	– Angle [°] / 15	DRCR < 1
C. Thickness mismatch	2 × (Donor thickness) / (Recipient thickness)	DRCR > 1
C. Thickness mismatch	2 × (Recipient thickness) / (Donor thickness)	DRCR < 1
D. Donor Thicker?	2	DRCR > 1
D. Donor Thicker?	0	DRCR < 1
Score	(Score A) + (Score B) + (Score C) + (Score D)

Open in a new tab

Relationship between the hypothetical scoring method and the size of the high-stress region for all simplified 2D models. Here the high-stress region is defined using a threshold value of 1.5x the applied pressure. Each data point represents a unique combination of donor cartilage thickness, recipient cartilage thickness, and surface angle deviation.

4. Discussion

This study used patient-specific 3D and simplified 2D models to examine the impact of cartilage thickness mismatches caused by OCA-transplantation in the patella on patellar cartilage stresses. In both models, stress concentrations occurred at the sharp subchondral bone corner between the donor and recipient regions. This stress concentration would not have been present had the donor cartilage thickness matched the recipient cartilage thickness.

The patient-specific 3D models highlighted which geometric features produced high-stress regions in the patellar cartilage and where those stresses occurred. Furthermore, the 3D models provided ranges for the geometric parameter sweeps that were conducted with simplified 2D models. The 2D models, which included depth-dependent cartilage stiffness, described the relevant physics and clearly identified the relationship between donor cartilage thickness, recipient cartilage thickness, and surface angle deviation. Generally, larger thickness mismatches with thicker recipient cartilage caused higher stresses. Additionally, surface angle deviations, defined as the angle between the vector describing the graft interface and the local cartilage surface normal, increased stresses when DRCR>1 and decreased stresses when DRCR<1.

The results from this study support our hypothesis that greater thickness mismatches would lead to higher stresses, while also revealing unexpected physics with important clinical implications. It was expected that stress would increase as the thickness mismatch increased and as the distance from the graft boundary decreased based on analyses of similar problems involving diameter discontinuities in a cylindrical shaft⁸. However, it was unexpected that stresses would increase as recipient cartilage thickness increased and that stresses for a given thickness mismatch would vary depending on whether the DRCR > 1 or DRCR < 1. These unexpected relationships were caused by the finite dimensions used in this model, which differed from many problems analyzing stress concentrations where it is assumed that the loaded material is very long in one or more dimensions. Lastly, this study revealed that surface angle deviations can have large impacts on stress. Surface angle deviations were not considered in our initial hypothesis because, to our knowledge, there was no prior research quantifying this parameter for OCA transplanted patellae. It should be noted that findings from this study do not confirm our hypothesis – such confirmation would require additional experiments.

The local stress increases found in this study suggest a mechanism to explain the lack of lateral integration and overall graft failure for some patellar OCA transplants. Multiple studies have found that cyclically loading osteochondral cores can lead to cell death and extracellular matrix damage in the cartilage deep zone with an applied surface load as low as 6-8 MPa^14-16. Furthermore, single load-unload experiments on cartilage explants with strain rates on the order of 0.5 s⁻¹, which is similar to strain rates observed during daily activities, can lead to cell death when the applied surface load exceeds 20 MPa^17,18.

Daily activity coupled with the thickness mismatch caused by patellar OCA could lead to stresses that would exceed the thresholds needed to cause matrix damage and chondrocyte depth in the deep layer of the tissue. Daily activities such as level walking, stair ascent, and running can produce contact pressures in patellar cartilage up to 5 MPa, 7 MPa, and 10-15 MPa, respectively^19-24. This study found that thickness mismatches caused by patellar OCA can lead to stresses in the deep layer 1.5-2x the applied pressure. Coupled together, the local stresses in the deep zone of cartilage would exceed even the most conservative threshold of 20 MPa required to induce local matrix damage and cell death. This localized cell death around the graft boundary could prevent lateral integration and lead to the commonly observed clefts between donor and recipient cartilage^25,26. Additionally, local cell death could serve as an initiation point for further cartilage degradation, potentially explaining the failure of some OCA transplants.

Stress increases caused by cartilage thickness mismatches could explain differences in failure rates observed across cartilage repair procedures. In a small study of 10 knees with OCA transplants on the femoral condyle, Davidson et al. reported <0.1 mm average difference between donor and recipient cartilage thickness, even when cartilage thickness was not used as a graft matching criterion²⁷. This better thickness match could help to explain why OCA transplantation in the femoral condyles tends to have better long-term outcomes compared to the patellae^5,6. In addition to explaining better outcomes for femoral OCA, thickness mismatches could explain why large OCAs have lower failure rates. In a study of 149 knees with large defects that had an average size of 7.2 cm², Sadr et al. found a 7% 10-year failure rate, which is much lower than the average 10-year failure rate of 21% for patellar OCAs that was found by Familiari et al.^5,28. For a large defect, the graft boundary will be peripherally located. Applied contact pressures at the periphery will be lower. This will lower stresses near the graft boundary, which would decrease the size of the high-stress region, thus potentially reducing the likelihood that these grafts would fail from increased stresses caused thickness mismatches^13,29,30. While currently speculative because of limited available data, this study highlights a proposed mechanism to explain differences in repair failure rates and to guide future clinical practice.

This study informs future OCA monitoring following transplantation, highlighting which OCA-transplanted patellae could be at a higher risk of failure because of cartilage thickness mismatches. To determine the risk of failure due to cartilage thickness mismatches, the donor cartilage thickness, recipient cartilage thickness, and surface angle deviation would all need to be measured. Donor and recipient cartilage thicknesses could be measured at time of implantation or with follow-up imaging³¹. Surface angle deviation could be measured at the superior, inferior, medial, and lateral graft boundaries of the OCA-transplanted patella using post-operation CT or MRI imaging. Once these variables are quantified, a scoring method, such as that proposed in Table 1, could then be used to estimate the size of the high-stress region that could serve as an initiation point for graft failure in an OCA-transplanted patella for each region where cartilage thicknesses and surface angle deviations have been measured. With knowledge of the risk of failure, surgeons and patients can then make informed decisions regarding appropriate levels of activity and joint health monitoring after surgery.

Table 1.

Characteristics of the chosen scans compared to the population of available scans. With the exception of gender, scans were selected such that the range of each characteristic was represented in the analyzed population and that there was no statistically significant difference in mean values.

Characteristic	Sample Donor Min/Max/Average	Sample Recipient Min/Max/Average	Population Donor Min/Max/Average	Population Recipient Min/Max/Average
Gender [is female?]	− / − / 0.60	− / − / 0.70	− / − / 0.39	− / − / 0.35
Age [years]	13 / 28 / 20.8±3.9	14 / 26 / 18.9±4.0	12 / 29 / 19.9±5.0	13 / 27 / 18.7±4.4
Tibial width [cm]	6.6 / 8.3 / 7.3±0.6	6.5 / 8.1 / 7.3±0.6	6.6 / 8.3 / 7.3±0.5	6.5 / 8.1 / 7.3±0.5
Patellar width [mm]	34 / 50 / 42±5	33 / 46 / 41±4	34 / 50 / 42±4	33 / 46 / 41±3
Patellar height [mm]	25 / 37 / 31±3	26 / 36 / 30±3	25 / 37 / 30±3	26 / 36 / 31±3
Ridge distance [mm]	20 / 32 / 24±4	20 / 28 / 23±3	20 / 32 / 24±3	20 / 28 / 23±3
Ridge distance [% of width]	51 / 73 / 57±7	52 / 64 / 57±4	51 / 73 / 57±6	51 / 64 / 56±4

Open in a new tab

This work had several limitations. First, a simplified loading regime was used. This was done to enable identification of the underlying physics and because of a lack of images of the femurs matched to each patella. Future work could use realistic loading to describe in vivo mechanics more accurately. Second, this study analyzed ten transplanted patellae. While this is a large sample for a finite element-based study, it may be insufficient to capture the full range of geometric heterogeneity. This limitation was partially addressed by the simplified 2D models which enabled the identification of broadly applicable underlying physics. Third, this study did not account for local tissue remodeling. It is expected that tissue properties will change in response to the local biochemical environment after a surgical intervention and the mechanical environment described in this study. Future work could model this process and predict long-term tissue outcomes. Finally, the results used to develop the hypothetical scoring system in this study have not been clinically validated. Further clinical studies of OCA-transplanted patellae are needed to refine the scoring system and better predict real-world outcomes.

In summary, cartilage thickness mismatches caused by OCA transplantation in the patella can lead to increased stresses in the cartilage near the graft boundary. These stresses are generally higher with greater thickness mismatches and thicker recipient cartilage. Surface angle deviations had an impact that depended on recipient cartilage thickness and the DRCR. These high stresses caused by thickness mismatches provide a potential mechanism to explain the failure of some OCA transplants. Clinically, this study found that donor cartilage thickness, recipient cartilage thickness, and surface angle deviation can be used to estimate the size of the high-stress region caused by thickness mismatches in OCA-transplanted patella and thus predict where failure may occur.

Supplementary Material

supinfo

NIHMS1890358-supplement-supinfo.docx^{(23.5KB, docx)}

fA1

Figure A1 (a) Hyperelastic stress contour; (b) Linear elastic stress contour; (c) hyperelastic strain contour; (d) linear elastic strain contour; line graph of compressive stress vs distance for linear elastic (LE) and hyperelastic (HE) models (e) along the subchondral bone and (f) along the interface between the donor and recipient. All contours came from models with 6 mm thick donor cartilage and 2 mm thick recipient cartilage.

NIHMS1890358-supplement-fA1.tiff^{(73.9MB, tiff)}

fA2

Figure A2 Donor cartilage that is 3x recipient cartilage thickness without (a) and with (b) peripheral integration; donor cartilage that is 0.33x recipient cartilage thickness without (c) and with (d) peripheral integration. All models had 2 mm thick recipient cartilage.

NIHMS1890358-supplement-fA2.tiff^{(36.6MB, tiff)}

fA3

Figure A3 (a) Geometry for plane strain shear simulations; contour of shear into the graft (b) and out of the graft (c) with donor cartilage that is 3x recipient cartilage thickness; contour of shear into the graft (d) and out of the graft (e) with donor cartilage that is 0.33x recipient cartilage thickness. All models had 2 mm thick recipient cartilage.

NIHMS1890358-supplement-fA3.tiff^{(17.9MB, tiff)}

fA4

Figure A4 Stress contours for a model with DRCR=3 and deformable bone (a), DRCR=3 and deformable bone within region of interest highlighted in red (b), DRCR=3 and rigid bone (c), DRCR=0.33 and deformable bone (d), and DRCR=0.33 and rigid bone (e). Comparison of stress vs distance along the subchondral bone surface (f) and along the interface (g) for rigid and deformable bone models. All models had a recipient cartilage thickness of 2 mm.

NIHMS1890358-supplement-fA4.tiff^{(976.1KB, tiff)}

Acknowledgements:

The authors would like to thank Dr. David Nordsletten, Dr. Jonathan Estrada, Peter Kuetzing, Nathaly Villacis, and Charlotte Andreasen for their feedback on model analysis and results. We would also like to acknowledge our funding sources: National Science Foundation Graduate Research Fellowship Award No. 1841052, NSF Grant No. 1537711, ANRF Award No. 047420, Orthopaedic Research and Education Foundation Award No. 651100, NIAMS Award No. 1R21AR07401101, NIAMS Award No. P30AR069620, and JRF Ortho.

References:

1.Curl WW, Krome J, Gordon ES, et al. 1997. Cartilage injuries: A review of 31,516 knee arthroscopies. Arthroscopy 13(4):456–460. [DOI] [PubMed] [Google Scholar]
2.Ding C, Garnero P, Cicuttini F, et al. 2005. Knee cartilage defects: Association with early radiographic osteoarthritis, decreased cartilage volume, increased joint surface area and type II collagen breakdown. Osteoarthritis and Cartilage 13(3):198–205. [DOI] [PubMed] [Google Scholar]
3.Wluka AE, Ding C, Jones G, Cicuttini FM. 2005. The clinical correlates of articular cartilage defects in symptomatic knee osteoarthritis: A prospective study. Rheumatology 44(10):1311–1316. [DOI] [PubMed] [Google Scholar]
4.Cavendish PA, Everhart JS, Peters NJ, et al. 2019. Osteochondral Allograft Transplantation for Knee Cartilage and Osteochondral Defects. JBJS Reviews 7(6):1–12. [DOI] [PubMed] [Google Scholar]
5.Familiari F, Cinque ME, Chahla J, et al. 2018. Clinical Outcomes and Failure Rates of Osteochondral Allograft Transplantation in the Knee A Systematic Review. The American Journal of Sports Medicine 46(14):3541–3549. [DOI] [PubMed] [Google Scholar]
6.Chahla J, Sweet MC, Okoroha KR, et al. 2017. Osteochondral Allograft Transplantation M in the Patellofemoral Joint A Systematic Review. The American Journal of Sports Medicine 42(12):3009–3018. [DOI] [PubMed] [Google Scholar]
7.Ackermann J, Merkely G, Shah N, Gomoll AH. 2019. Decreased Graft Thickness Is Associated With Subchondral Cyst Formation After Osteochondral Allograft Transplantation in the Knee. American Journal of Sports Medicine 47(9):2123–2129. [DOI] [PubMed] [Google Scholar]
8.Pilkey WD, Pilkey DF. 2008. Peterson’s Stress Concentration Factors, 3rd ed. Hoboken, NJ: John Wiley and Sons Inc. 1–56 p. [Google Scholar]
9.Patel K, Salka NS, Ramme A, et al. 2021. Does the Symmetry of Patellar Morphology Matter When Matching Osteochondral Allografts for Osteochondral Defects Involving the Central Ridge of the Patella? Cartilage 13(Suppl 2):703S–712S. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Laasanen MS, Töyräs J, Korhonen RK, et al. 2002. Biomechanical properties of knee articular cartilage. Biorheology 40(1–3):133–140. [PubMed] [Google Scholar]
11.Shepherd DET, Seedhom BB. 1999. Thickness of human articular cartilage in joints of the lower limb. Annals of the Rheumatic Diseases 58(1):27–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Schinagl RM, Gurskis D, Chen AC, Sah RL. 1997. Depth-dependent confined compression modulus of full-thickness bovine articular cartilage. Journal of Orthopaedic Research 15(4):499–506. [DOI] [PubMed] [Google Scholar]
13.Adouni M, Shirazi-Adl A. 2014. Evaluation of Knee Joint Muscle Forces and Tissue Stresses-Strains During Gait in Severe OA Versus Normal Subjects M. Journal of Orthopaedic Research 32:69–78. [DOI] [PubMed] [Google Scholar]
14.Clements KM, Bee ZC, Crossingham GV., et al. 2001. How severe must repetitive loading be to kill chondrocytes in articular cartilage? Osteoarthritis and Cartilage 9(5):499–507. [DOI] [PubMed] [Google Scholar]
15.Kang RW, Friel NA, Williams JM, et al. 2010. Effect of Impaction Sequence on Osteochondral Graft Damage: The Role of Repeated and Varying Loads. Am J Sports Med 38(1):105–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Thibault M, Robin Poole A, Buschmann MD. 2002. Cyclic compression of cartilage/bone explants in vitro leads to physical weakening, mechanical breakdown of collagen and release of matrix fragments. J. Orthop. Res 20(6):1265–1273. [DOI] [PubMed] [Google Scholar]
17.Torzilli PA, Grigiene R, Borrelli J, Helfet DL. 1999. Effect of Impact Load on Articular Cartilage: Cell Metabolism and Viability, and Matrix Water Content. Journal of Biomechanical Engineering 121(5):433–441. [DOI] [PubMed] [Google Scholar]
18.Huang W, Warner M, Sasaki H, et al. 2020. Layer dependence in strain distribution and chondrocyte damage in porcine articular cartilage exposed to excessive compressive stress loading. Journal of the Mechanical Behavior of Biomedical Materials 112(May):104088. [DOI] [PubMed] [Google Scholar]
19.Lenhart RL, Smith CR, Vignos MF, et al. 2015. Influence of step rate and quadriceps load distribution on patellofemoral cartilage contact pressures during running. Journal of Biomechanics 48(11):2871–2878. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Huberti HH, Hayes WC. 1988. Contact pressures in chondromalacia patellae and the effects of capsular reconstructive procedures. Journal of Orthopaedic Research 6(4):499–508. [DOI] [PubMed] [Google Scholar]
21.Brechter JH, Powers CM. 2002. Patellofemoral joint stress during stair ascent and descent in persons with and without patellofemoral pain. Gait & Posture 16(2):115–123. [DOI] [PubMed] [Google Scholar]
22.Hart HF, Patterson BE, Crossley KM, et al. 2022. May the force be with you: understanding how patellofemoral joint reaction force compares across different activities and physical interventions—a systematic review and meta-analysis. Br J Sports Med 56(9):521–530. [DOI] [PubMed] [Google Scholar]
23.Lenhart RL, Smith CR, Vignos MF, et al. 2015. Influence of step rate and quadriceps load distribution on patellofemoral cartilage contact pressures during running. Journal of Biomechanics 48(11):2871–2878. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Liao T-C, Keyak JH, Powers CM. 2018. Runners With Patellofemoral Pain Exhibit Greater Peak Patella Cartilage Stress Compared With Pain-Free Runners. Journal of Applied Biomechanics 34(4):298–305. [DOI] [PubMed] [Google Scholar]
25.Horas U, Pelinkovic D, Herr G, et al. 2003. Autologous Chondrocyte Implantation and Osteochondral Cylinder Transplantation in Cartilage Repair of the Knee Joint [7] (multiple letters). Journal of Bone and Joint Surgery 85-A(2):185–192. [DOI] [PubMed] [Google Scholar]
26.Jamali AA, Emmerson BC, Chung C, et al. 2005. Fresh osteochondral allografts: Results in the patellofemoral joint. Clinical Orthopaedics and Related Research (437):176–185. [PubMed] [Google Scholar]
27.Davidson PA, Rivenburgh DW, Dawson PE, Rozin R. 2007. Clinical, histologic, and radiographic outcomes of distal femoral resurfacing with hypothermically stored osteoarticular allografts. American Journal of Sports Medicine 35(7):1082–1090. [DOI] [PubMed] [Google Scholar]
28.Sadr KN, Pulido PA, McCauley JC, Bugbee WD. 2016. Osteochondral Allograft Transplantation in Patients with Osteochondritis Dissecans of the Knee. American Journal of Sports Medicine 44(11):2870–2875. [DOI] [PubMed] [Google Scholar]
29.Guess TM, Liu H, Bhashyam S, Thiagarajan G. 2013. A multibody knee model with discrete cartilage prediction of tibio-femoral contact mechanics. Computer Methods in Biomechanics and Biomedical Engineering 16(3):256–270. [DOI] [PubMed] [Google Scholar]
30.Smith CR, Won Choi K, Negrut D, Thelen DG. 2018. Efficient computation of cartilage contact pressures within dynamic simulations of movement. Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization 6(5):491–498. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Determann JR, Fleischli JE, D’Alessandro DF, Piasecki DP. 2017. Patellofemoral Osteochondral Allografts: Can We Improve the Matching Process? Journal of Knee Surgery 30(8):835–841. [DOI] [PubMed] [Google Scholar]
32.Deneweth JM, Arruda EM, McLean SG. 2015. Hyperelastic modeling of location-dependent human distal femoral cartilage mechanics. International Journal of Non-Linear Mechanics 68:146–156. [Google Scholar]
33.Sennett ML, Meloni GR, Farran AJE, et al. 2018. Sprifermin treatment enhances cartilage integration in an in vitro repair model: SPRIFERMIN TREATMENT ENHANCES CARTILAGE INTEGRATION. J. Orthop. Res 36(10):2648–2656. [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Gelse K, Riedel D, Pachowsky M, et al. 2015. Limited integrative repair capacity of native cartilage autografts within cartilage defects in a sheep model: INTEGRATIVE REPAIR BY CARTILAGE AUTOGRAFTS. J. Orthop. Res 33(3):390–397. [DOI] [PubMed] [Google Scholar]
35.Khan I, Gilbert S, Singhrao S, et al. 2008. Evaluation of the reasons for failure of integration during cartilage repair. A review. eCM 16:26–39. [DOI] [PubMed] [Google Scholar]
36.Trengove A, Di Bella C, O’Connor AJ. 2022. The Challenge of Cartilage Integration: Understanding a Major Barrier to Chondral Repair. Tissue Engineering Part B: Reviews 28(1):114–128. [DOI] [PubMed] [Google Scholar]
37.S hua Qian, S rong Ge, Q liang Wang. 2006. The Frictional Coefficient of Bovine Knee Articular Cartilage. Journal of Bionic Engineering 3(2):79–85. [Google Scholar]
38.Wang J, Zhou B, Liu XS, et al. 2015. Trabecular plates and rods determine elastic modulus and yield strength of human trabecular bone. Bone 72:71–80. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supinfo

NIHMS1890358-supplement-supinfo.docx^{(23.5KB, docx)}

fA1

NIHMS1890358-supplement-fA1.tiff^{(73.9MB, tiff)}

fA2

NIHMS1890358-supplement-fA2.tiff^{(36.6MB, tiff)}

fA3

NIHMS1890358-supplement-fA3.tiff^{(17.9MB, tiff)}

fA4

NIHMS1890358-supplement-fA4.tiff^{(976.1KB, tiff)}

[R1] 1.Curl WW, Krome J, Gordon ES, et al. 1997. Cartilage injuries: A review of 31,516 knee arthroscopies. Arthroscopy 13(4):456–460. [DOI] [PubMed] [Google Scholar]

[R2] 2.Ding C, Garnero P, Cicuttini F, et al. 2005. Knee cartilage defects: Association with early radiographic osteoarthritis, decreased cartilage volume, increased joint surface area and type II collagen breakdown. Osteoarthritis and Cartilage 13(3):198–205. [DOI] [PubMed] [Google Scholar]

[R3] 3.Wluka AE, Ding C, Jones G, Cicuttini FM. 2005. The clinical correlates of articular cartilage defects in symptomatic knee osteoarthritis: A prospective study. Rheumatology 44(10):1311–1316. [DOI] [PubMed] [Google Scholar]

[R4] 4.Cavendish PA, Everhart JS, Peters NJ, et al. 2019. Osteochondral Allograft Transplantation for Knee Cartilage and Osteochondral Defects. JBJS Reviews 7(6):1–12. [DOI] [PubMed] [Google Scholar]

[R5] 5.Familiari F, Cinque ME, Chahla J, et al. 2018. Clinical Outcomes and Failure Rates of Osteochondral Allograft Transplantation in the Knee A Systematic Review. The American Journal of Sports Medicine 46(14):3541–3549. [DOI] [PubMed] [Google Scholar]

[R6] 6.Chahla J, Sweet MC, Okoroha KR, et al. 2017. Osteochondral Allograft Transplantation M in the Patellofemoral Joint A Systematic Review. The American Journal of Sports Medicine 42(12):3009–3018. [DOI] [PubMed] [Google Scholar]

[R7] 7.Ackermann J, Merkely G, Shah N, Gomoll AH. 2019. Decreased Graft Thickness Is Associated With Subchondral Cyst Formation After Osteochondral Allograft Transplantation in the Knee. American Journal of Sports Medicine 47(9):2123–2129. [DOI] [PubMed] [Google Scholar]

[R8] 8.Pilkey WD, Pilkey DF. 2008. Peterson’s Stress Concentration Factors, 3rd ed. Hoboken, NJ: John Wiley and Sons Inc. 1–56 p. [Google Scholar]

[R9] 9.Patel K, Salka NS, Ramme A, et al. 2021. Does the Symmetry of Patellar Morphology Matter When Matching Osteochondral Allografts for Osteochondral Defects Involving the Central Ridge of the Patella? Cartilage 13(Suppl 2):703S–712S. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] 10.Laasanen MS, Töyräs J, Korhonen RK, et al. 2002. Biomechanical properties of knee articular cartilage. Biorheology 40(1–3):133–140. [PubMed] [Google Scholar]

[R11] 11.Shepherd DET, Seedhom BB. 1999. Thickness of human articular cartilage in joints of the lower limb. Annals of the Rheumatic Diseases 58(1):27–34. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] 12.Schinagl RM, Gurskis D, Chen AC, Sah RL. 1997. Depth-dependent confined compression modulus of full-thickness bovine articular cartilage. Journal of Orthopaedic Research 15(4):499–506. [DOI] [PubMed] [Google Scholar]

[R13] 13.Adouni M, Shirazi-Adl A. 2014. Evaluation of Knee Joint Muscle Forces and Tissue Stresses-Strains During Gait in Severe OA Versus Normal Subjects M. Journal of Orthopaedic Research 32:69–78. [DOI] [PubMed] [Google Scholar]

[R14] 14.Clements KM, Bee ZC, Crossingham GV., et al. 2001. How severe must repetitive loading be to kill chondrocytes in articular cartilage? Osteoarthritis and Cartilage 9(5):499–507. [DOI] [PubMed] [Google Scholar]

[R15] 15.Kang RW, Friel NA, Williams JM, et al. 2010. Effect of Impaction Sequence on Osteochondral Graft Damage: The Role of Repeated and Varying Loads. Am J Sports Med 38(1):105–113. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] 16.Thibault M, Robin Poole A, Buschmann MD. 2002. Cyclic compression of cartilage/bone explants in vitro leads to physical weakening, mechanical breakdown of collagen and release of matrix fragments. J. Orthop. Res 20(6):1265–1273. [DOI] [PubMed] [Google Scholar]

[R17] 17.Torzilli PA, Grigiene R, Borrelli J, Helfet DL. 1999. Effect of Impact Load on Articular Cartilage: Cell Metabolism and Viability, and Matrix Water Content. Journal of Biomechanical Engineering 121(5):433–441. [DOI] [PubMed] [Google Scholar]

[R18] 18.Huang W, Warner M, Sasaki H, et al. 2020. Layer dependence in strain distribution and chondrocyte damage in porcine articular cartilage exposed to excessive compressive stress loading. Journal of the Mechanical Behavior of Biomedical Materials 112(May):104088. [DOI] [PubMed] [Google Scholar]

[R19] 19.Lenhart RL, Smith CR, Vignos MF, et al. 2015. Influence of step rate and quadriceps load distribution on patellofemoral cartilage contact pressures during running. Journal of Biomechanics 48(11):2871–2878. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Huberti HH, Hayes WC. 1988. Contact pressures in chondromalacia patellae and the effects of capsular reconstructive procedures. Journal of Orthopaedic Research 6(4):499–508. [DOI] [PubMed] [Google Scholar]

[R21] 21.Brechter JH, Powers CM. 2002. Patellofemoral joint stress during stair ascent and descent in persons with and without patellofemoral pain. Gait & Posture 16(2):115–123. [DOI] [PubMed] [Google Scholar]

[R22] 22.Hart HF, Patterson BE, Crossley KM, et al. 2022. May the force be with you: understanding how patellofemoral joint reaction force compares across different activities and physical interventions—a systematic review and meta-analysis. Br J Sports Med 56(9):521–530. [DOI] [PubMed] [Google Scholar]

[R23] 23.Lenhart RL, Smith CR, Vignos MF, et al. 2015. Influence of step rate and quadriceps load distribution on patellofemoral cartilage contact pressures during running. Journal of Biomechanics 48(11):2871–2878. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] 24.Liao T-C, Keyak JH, Powers CM. 2018. Runners With Patellofemoral Pain Exhibit Greater Peak Patella Cartilage Stress Compared With Pain-Free Runners. Journal of Applied Biomechanics 34(4):298–305. [DOI] [PubMed] [Google Scholar]

[R25] 25.Horas U, Pelinkovic D, Herr G, et al. 2003. Autologous Chondrocyte Implantation and Osteochondral Cylinder Transplantation in Cartilage Repair of the Knee Joint [7] (multiple letters). Journal of Bone and Joint Surgery 85-A(2):185–192. [DOI] [PubMed] [Google Scholar]

[R26] 26.Jamali AA, Emmerson BC, Chung C, et al. 2005. Fresh osteochondral allografts: Results in the patellofemoral joint. Clinical Orthopaedics and Related Research (437):176–185. [PubMed] [Google Scholar]

[R27] 27.Davidson PA, Rivenburgh DW, Dawson PE, Rozin R. 2007. Clinical, histologic, and radiographic outcomes of distal femoral resurfacing with hypothermically stored osteoarticular allografts. American Journal of Sports Medicine 35(7):1082–1090. [DOI] [PubMed] [Google Scholar]

[R28] 28.Sadr KN, Pulido PA, McCauley JC, Bugbee WD. 2016. Osteochondral Allograft Transplantation in Patients with Osteochondritis Dissecans of the Knee. American Journal of Sports Medicine 44(11):2870–2875. [DOI] [PubMed] [Google Scholar]

[R29] 29.Guess TM, Liu H, Bhashyam S, Thiagarajan G. 2013. A multibody knee model with discrete cartilage prediction of tibio-femoral contact mechanics. Computer Methods in Biomechanics and Biomedical Engineering 16(3):256–270. [DOI] [PubMed] [Google Scholar]

[R30] 30.Smith CR, Won Choi K, Negrut D, Thelen DG. 2018. Efficient computation of cartilage contact pressures within dynamic simulations of movement. Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization 6(5):491–498. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] 31.Determann JR, Fleischli JE, D’Alessandro DF, Piasecki DP. 2017. Patellofemoral Osteochondral Allografts: Can We Improve the Matching Process? Journal of Knee Surgery 30(8):835–841. [DOI] [PubMed] [Google Scholar]

[R32] 32.Deneweth JM, Arruda EM, McLean SG. 2015. Hyperelastic modeling of location-dependent human distal femoral cartilage mechanics. International Journal of Non-Linear Mechanics 68:146–156. [Google Scholar]

[R33] 33.Sennett ML, Meloni GR, Farran AJE, et al. 2018. Sprifermin treatment enhances cartilage integration in an in vitro repair model: SPRIFERMIN TREATMENT ENHANCES CARTILAGE INTEGRATION. J. Orthop. Res 36(10):2648–2656. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Gelse K, Riedel D, Pachowsky M, et al. 2015. Limited integrative repair capacity of native cartilage autografts within cartilage defects in a sheep model: INTEGRATIVE REPAIR BY CARTILAGE AUTOGRAFTS. J. Orthop. Res 33(3):390–397. [DOI] [PubMed] [Google Scholar]

[R35] 35.Khan I, Gilbert S, Singhrao S, et al. 2008. Evaluation of the reasons for failure of integration during cartilage repair. A review. eCM 16:26–39. [DOI] [PubMed] [Google Scholar]

[R36] 36.Trengove A, Di Bella C, O’Connor AJ. 2022. The Challenge of Cartilage Integration: Understanding a Major Barrier to Chondral Repair. Tissue Engineering Part B: Reviews 28(1):114–128. [DOI] [PubMed] [Google Scholar]

[R37] 37.S hua Qian, S rong Ge, Q liang Wang. 2006. The Frictional Coefficient of Bovine Knee Articular Cartilage. Journal of Bionic Engineering 3(2):79–85. [Google Scholar]

[R38] 38.Wang J, Zhou B, Liu XS, et al. 2015. Trabecular plates and rods determine elastic modulus and yield strength of human trabecular bone. Bone 72:71–80. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Cartilage Thickness Mismatches in Patellar Osteochondral Allograft Transplants Affect Local Cartilage Stresses

Ryan Rosario

Ellen M Arruda

John A Grant

Rhima M Coleman

Abstract

1. Introduction:

2. Methods: