Novel prosthetic medial meniscus restores contact pressures to native values in a meniscectomized canine stifle

Abstract

Objective

The aim of this study was to ascertain whether a prosthetic meniscus adapts pressures to native tibial contact pressures in a canine stifle joint after total medial meniscectomy.

Study design

Ex vivo biomechanical experimental study.

Sample population

Seven cadaveric hindlimbs of seven large‐breed dogs.

Methods

Cadaveric canine stifles were loaded vertically at 135° of flexion. The peak pressure (PP), mean pressure (MP), and contact area (CA) on the medial and lateral tibial plateaus were measured for the intact stifle, after medial meniscectomy and after implantation of the prosthetic meniscus. Comparisons were made between same parameters in varying conditions, but not in varying compartments.

Results

Mean PP and MP increased significantly after medial meniscectomy (6.1 and 2.3 MPa, respectively) relative to those of the native stifle joint (3.5 and 1.0 MPa, respectively), whereas meniscectomy resulted in significantly lower mean CA on the medial tibial plateau (65 mm²) compared to the native stifle joint (170 mm²). The prosthetic meniscus significantly decreased PP to 3.3 MPa and MP to 1.0 MPa and significantly increased CA to 96 mm² relative to the meniscectomized stifle. PP and MP were not statistically different between the native stifle and prosthetic meniscus.

Conclusion

The prosthetic meniscus significantly restored tibial contact pressures to normal (native) values in the meniscectomized canine stifle.

Clinical significance

A prosthetic meniscus has the potential to restore biomechanical function in the canine stifle with an impaired medial meniscal function.

1. INTRODUCTION

Cranial cruciate ligament (CCL) disease causes stifle pain, lameness, and osteoarthritis (OA) and is the most common orthopedic disease affecting a dog's stifle. ¹ , ² Along with rupture of the CCL, concurrent meniscal injury has been reported in 33–70% of cases. ³ , ⁴ , ⁵ , ⁶

The main functions of the meniscus, including force distribution and stifle stabilization, are impaired following damaged CCL or (isolated) meniscal injury, possibly leading to craniocaudal instability, probable lameness and pain with the progression of OA. ² , ⁷

Hemi‐meniscectomy and total meniscectomy are among the surgical procedures performed for canine stifles with cruciate and meniscal injuries. ⁴ , ⁸ , ⁹ CCL and meniscal injury results in decreased exercise, inactivity stiffness, incapacity, and reduced general function in 3–5 years. ¹⁰ Significant articular cartilage damage or degeneration is associated with meniscal surgery, ¹¹ moreover, OA is more severe in canines when CCL injury is accompanied by meniscus injury than when the meniscus is intact, ¹² , ¹³ which underscores the importance of normal meniscal function. Load transmission is an important function of the meniscus. Canine cadaveric studies have reported that meniscal injuries and partial meniscectomy lead to an increase in the peak pressure (PP) and a decrease in the contact area (CA) on the tibial plateau, ⁸ , ¹⁴ the latter supposedly causing OA ¹⁵ and increased stress. ¹⁶ , ¹⁷ Reduced cartilage tensile stiffness has also been reported in impaired functioning menisci. ¹⁸

Alternative methods of preventing meniscal lesions, such as meniscal release, have also been investigated. ⁹ This procedure is used in canines undergoing tibial plateau leveling osteotomy (TPLO) to reduce the risk of meniscal injury after femorotibial impingement of the medial meniscus. However, meniscal release causes a disturbance in pressure distribution and load transmission, ⁹ and cartilage damage and degeneration of the stifle may occur. ¹⁹ The biology and mechanics of stifle joints subjected to meniscal release are negatively impacted, contributing to the controversy surrounding the procedure.

Meniscal allografts in dogs yield outcomes similar to those in humans related to alleviating pain and limb function, although no comparison was made between the native and treatment group and complications such as bone marrow lesions and synovitis were observed. ²⁰ Further experiments with allograft in dogs showed promising results by increasing the meniscus porosity, but cartilage wear was still present. ²¹

Suture techniques are more frequently used in humans than in veterinary medicine, but recent research has demonstrated their advantages in veterinary surgery. ²² , ²³ However, reports of long‐term outcomes and studies with larger samples remain limited.

Scaffolds with different compositions were used in two different canine studies ²⁴ , ²⁵ ; bone marrow mesenchymal stem cells yielded cell compositions similar to those of a native meniscus, and a collagen scaffold generated substantial tissue integration. In dogs, porous implants, such as lateral meniscus implants, have shown satisfactory clinical results for lameness, but various degrees of cartilage damage in the knee have been reported. ²⁶ An artificial prosthetic meniscus made of polycarbonate urethane (PCU) was developed as an alternative to meniscal allografts in humans. The first trial of the synthetic prosthetic meniscus in goats showed only minor wear and limited implant deformation as a result of physiological loading. ²⁷ In a cadaveric study, the human prosthetic meniscus did not restore PP and mean pressure (MP) to the levels observed in the native stifle; however, it reduced the pressures by approximately two‐thirds of the increase caused by the meniscectomy. ²⁸ In a clinical trial, implantation of the prosthetic meniscus resulted in clinically relevant improvement of pain scores. ²⁹

To the best of our knowledge, no study has investigated the tibial contact biomechanics of a synthetic prosthetic meniscus in the canine stifle. This cadaveric study aimed to examine the pressure distribution of a prosthetic meniscus after medial meniscectomy of the canine stifle. Partial medial meniscectomy results in elevated peak pressures of 58.3–88.9% and a reduced CA of 47.2% on the tibial plateau. ⁸ , ¹⁷ We hypothesized that implantation of a prosthetic medial meniscus significantly restores the disturbed pressure‐distribution after meniscectomy to native values, potentially restoring medial meniscal function.

2. MATERIALS AND METHODS

This study was approved by the Animal Experiment Committee and Bioethics Committee of Utrecht University (no. AVD115002016531).

2.1. Prosthetic meniscus

The prosthetic meniscus was three‐dimensional (3D)‐printed from PCU using Samaplast AG (St. Margrethen, Switzerland). The two meniscal horns consisted of stiff PCU (Figure 1), and the meniscal body consisted of more flexible PCU. For horn fixation, 2.4‐mm cortical screws with mushroom tips were used. The prosthesis was based on the medial meniscus of the right intact healthy stifle of a Labrador Retriever with a bodyweight of 32 kg. The meniscal geometry was retrieved from a computed tomography (CT) scan of the stifle performed with a resolution of 16.4 μm and X‐ray voltage of 70 kVp using a μCT 100 scanner (SCANCO Medical AG, Brüttiselen, Switzerland). The tips of the screws protruded from the cortex of the tibial plateau, and the horns were connected to the screw tip in a press‐fit manner. The same prosthesis size was used for all cadaveric stifles (Figure 2).

FIGURE 1.

Dorsal (A) and caudal (B) views of the native medial meniscus (left) and printed prosthetic meniscus (right).

FIGURE 2.

Image of a three‐dimensional model rendering of a canine cadaveric medial meniscus retrieved by a μCT scan. Note the hole in the horns for fixation over the mushroom screw tip.

2.2. Animals

Left and right cadaveric stifles (n = 7), including the complete femur, tibia, and fibula, were harvested from euthanized dogs unrelated to the study at the veterinary clinic of Utrecht University. Dog cadavers were collected in an established donor consent program for education and research purposes, that is, a formal donation system, where owners can authorize university in advance to utilize the animal's body for education and research purposes. Owner consent was obtained, and a form was signed for each cadaver. The limbs were disarticulated at the level of the hip joint post‐mortem.

2.3. Surgical technique

The skin, subcutaneous tissue, major vessels, nerves, and surrounding muscle bellies were removed from the stifles to increase their visibility and accessibility. However, the collateral ligaments, cruciate ligaments, and menisci remained intact to maintain joint stability. The femur and tibia were cut with an oscillating saw at the proximal (femur) and distal (tibia) thirds, and the soft tissues were cleaned. The tibial and femoral ends were fixed with cement in two bone pots.

A total meniscectomy of the medial meniscus was performed by removing its attachment to the tibia and capsule after evaluating the native stifle joint. The prosthesis was fixed in the correct position using an aiming device that pinpointed the cranial and caudal horn attachments after evaluating the meniscectomized stifle joint. Holes in the proximal tibia were drilled with a 1.8 mm Steinmann pin, and the trajectories were aimed in mediolateral and distal‐proximal directions. Following full screw insertion, the tips of the screws exited the anatomical cranial and caudal horn attachment points on the tibial plateau (Figure 3), whereas the screw heads rested on the cranial and caudal tibial facies medialis (Figure 4). The prosthetic meniscus horns were securely fastened to the stainless‐steel screw tip endings by pressing the hole in the horn over the mushroom tip of the screw.

FIGURE 3.

Proximal view of the lateral and medial (right) tibia plateau after implantation of the medial prosthetic meniscus and fixation with two screws that exit with their mushroom tips through the holes in the horns.

FIGURE 4.

Cranial view of the test set up with pressure‐sensitive sensors leading to the medial and lateral stifle compartments. (A) Mediocaudal view. (B) Prosthetic meniscus (asterisk) fixated with rigid screws to the cranial and caudal horn. (C) Tibial bone pot with screws allowing the valgus or varus of the tibia in relation to the femur.

2.4. Test setup

The joint was placed at 135° flexion, which corresponds to the stance phase angle of the canine hind leg, ³⁰ and the bone pots were mounted on a hydraulic testing rig (MTS Systems, Eden Prairie, Minnesota). To allow physiological varus or valgus alignment in all meniscal conditions, the tibial bone pot was mounted on its cranial and caudal screws, allowing rotation around the craniocaudal axis. A pressure‐sensitive sensor (Tekscan 4041; Tekscan, South Boston, Massachusetts), calibrated before and after the measurements, was placed on the medial and lateral tibial plateaus underneath the meniscus or prosthesis.

2.5. Testing protocol

Axial forces of 400 N were applied to the femur, and all measurements were performed after 20 s of constant load to allow the stifle joint to find its natural alignment and stable position. Every stifle was measured once in its native state, followed by measurements of the meniscectomized stifle and prosthesis in situ. To assess any loss of sensor sensitivity, meniscectomy measurements were repeated after prosthesis removal.

2.6. Statistical analysis

PP, MP, and CA were calculated separately for the medial and lateral tibia plateau, using the pressure measurements and the outcomes of the initial calibration. In case a difference of >10% was detected between the pre‐ and post‐calibration measurements, the post‐calibration was applied to the second part of the measurements instead.

Each parameter (PP, MP, and CA) was compared between different conditions (native, meniscectomy, and prosthesis), but not between different compartments (medial and lateral).

A sample size of seven stifles is sufficient to detect pressure differences of 0.7 MPa, assuming a standard deviation (SD) of 0.5 MPa, using a power of 80%, and a two‐sided alpha of 5% and a significance level of 0.05.

The assumed difference and SD were based on a study involving a human meniscus prosthesis, ²⁸ where peak and mean pressures were reduced from 5.5 to 3.7 MPa and from 1.6 to 0.9 MPa, respectively. For the power analysis, the smallest observed difference (0.7 MPa) was used to ensure a conservative estimate. Preliminary results from a clinical investigation using this device indicate that such pressure differences are sufficient to achieve a clinically relevant reduction in pain scores. ²⁸

For the CA analysis specifically, a separate power calculation was performed. Given the size difference between the human and canine meniscus, a scaling factor of 28% was applied to the SD of CA reported in the human cadaveric studies. ²⁸ , ³¹ This resulted in an assumed SD of 30 mm². With this SD, the defined sample size allows for detection of CA differences of 40 mm² with a power of 80% (and a two‐sided alpha of 5%).

Linear mixed models were used to study the effects of the meniscal condition on MP, PP, and CA on the medial and lateral tibia plateau. Stifle ID was implemented as a random factor, and the meniscal condition was included as a fixed factor. The effects on the different outcome measures were evaluated separately, and p‐values of < .05 were considered statistically significant. Assumptions regarding normal distribution of residuals and random effects were evaluated and not violated. A Pearson correlation matrix was performed to evaluate the correlation of MP, PP, and CA on the medial and lateral tibia plateau (Figure 5). All statistical analyses were performed using R (version 4.4.2; R Foundation for Statistical Computing, Vienna, Austria).

FIGURE 5.

Heat map with Pearson correlation matrix. All meniscal conditions are included. The upper triangle contains correlations with 95% confidence interval (CI), the diagonal row contains density plots, and the lower triangle contains scatter plots.

3. RESULTS

Four left and three right cadaveric stifles, including the complete femur, tibia, and fibula, were harvested from dogs weighing 25–40 kg.

The mean (± SD) PP was significantly increased in the medial tibial compartment of the meniscectomized stifles (6.1 ± 2.3 MPa) relative to the native stifles (3.5 ± 1.4 MPa) (Table 1, Figure 6). The mean (± SD) MP in the meniscectomized stifles (2.3 ± 0.7 MPa) was significantly increased in comparison with the native stifles (1.0 ± 0.2 MPa). PP and MP returned to native levels after prosthetic implantation (3.3 ± 1.5 MPa and 1.0 ± 0.4 MPa, respectively) and pressures were not significantly different from the intact native stifle condition (Figure 7, Table 2).

TABLE 1.

Mean and standard deviation of the biomechanical outcome measures.

Compartment	Condition	Peak pressure (MPa)	Mean pressure (MPa)	Contact area (mm2)
Medial	Native	3.5 ± 1.4	1.0 ± 0.2	170 ± 48
	Meniscectomy	6.1 ± 2.3	2.3 ± 0.7	65 ± 9
	Prosthesis	3.3 ± 1.5	1.0 ± 0.4	96 ± 24
Lateral	Native	1.7 ± 0.8	0.7 ± 0.3	134 ± 25
	Meniscectomy	2.4 ± 1.3	0.9 ± 0.4	104 ± 19
	Prosthesis	2.0 ± 0.9	0.8 ± 0.4	112 ± 23

FIGURE 6.

Representative example of pressure mapping results in the native stifle (A), after total medial meniscectomy (B), and following implantation of a prosthetic meniscus (C). In each image, the medial tibial plateau is on the left and the lateral is on the right. Cranial is on top and caudal is at the bottom.

FIGURE 7.

Mean and standard deviation of the biomechanical outcome measures. (A) mean tibial contact pressure, (B) peak tibial contact pressure, and (C) tibial contact area. All were measured using an axial load of 400 N.

TABLE 2.

Difference in outcome measures between the different meniscal conditions.

Outcome	Condition 1	Condition 2	Difference a	95% CI	p‐value
Medial peak pressure (MPa)	Native	Meniscectomy	2.7	(1.9, 3.4)	<0.001
	Meniscectomy	Prosthesis	−2.8	(−2.1, −3.5)	<0.001
	Native	Prosthesis	−0.1	(−0.9, 1.6)	0.68
Medial mean pressure (MPa)	Native	Meniscectomy	1.4	(1.2, 1.6)	<0.001
	Meniscectomy	Prosthesis	−1.4	(−1.6, −1.2)	<0.001
	Native	Prosthesis	0.0	(−0.2, 0.2)	0.99
Medial contact area (mm2)	Native	Meniscectomy	−105	(−132, −79)	<0.001
	Meniscectomy	Prosthesis	31	(5, 58)	0.02
	Native	Prosthesis	−74	(−100, −48)	<0.001

Abbreviation: CI, confidence interval.

^a Mean difference between conditions 1 and 2: A positive difference indicates an increase from condition 1 to condition 2.

The mean (± SD) CA in the medial tibial plateau in the native stifle (170 ± 48 mm ² ) was significantly larger than after medial meniscectomy (65 ± 9 mm ² ) and was significantly increased after implantation of the prosthesis (96 ± 24 mm ² ) but remained significantly smaller than the CA in the intact native stifle (Figure 7, Table 1).

Smaller effects of the medial meniscal conditions (native, meniscectomy, and prosthesis) were observed on the lateral tibial plateau. Mean values of PP, MP, and CA after prosthesis implantation were all between the values obtained for the native and meniscectomized stifle. Lateral contact pressures were not significantly increased after implantation of the prosthetic meniscus in the medial meniscectomized stifle relative to the meniscectomized stifle (Figures 6 and 7, Table 1).

Medial and lateral MP showed a strong positive correlation (Figure 5).

4. DISCUSSION

Current treatment standards for meniscal injury in dogs remain insufficiently developed. Given the potential long‐term functional impairments associated with meniscal injury, ¹⁰ alternative treatment options should be provided. The PCU‐based implant could improve outcomes after meniscal injuries in dogs. There are several indications for a prosthetic meniscus, such as in patients with isolated meniscal injuries in whom (hemi‐) meniscectomy is considered. Given that partial removal of the meniscus cannot maintain meniscal function, ⁸ replacement with a prosthetic meniscus may improve the mechanical environment of the stifle joint. In addition, the use of a prosthesis can be indicated for a CCL‐deficient stifle joint with a meniscal lesion, as several studies have shown a significant increase in PP. ³² , ³³ Patients who have undergone (hemi‐)meniscectomies and have persistent lameness could benefit from pain reduction from a prosthetic meniscus by improving contact pressures and meniscal function. Chondroprotection may also be a feature of this implant, as other biological meniscal implants have already demonstrated this capability. ³⁴

The present study aimed to demonstrate the tibial contact pressures and areas in a normal intact stifle following total medial meniscectomy and after implantation of a prosthetic medial meniscus. This was to ascertain whether a prosthetic meniscus can restore the disturbed pressure‐distribution after meniscectomy to native values. The medial and lateral compartments were not compared because this was beyond our research question. Meniscectomy resulted in increased PP and MP, as previously demonstrated in other studies. ⁸ , ²⁸ One study showed a decrease in PP in the lateral compartment after medial hemi‐meniscectomy. ⁸ The latter was also observed in a study investigating contact mechanics in TPLO. ³⁵ Parts of the meniscus remained intact in both experiments, which may have accounted for the discrepancy. In this study, no significant differences were found between the prosthetic meniscus and the native stifle in terms of PP and MP of the medial compartment (Table 2), increases in pressures resulting from meniscectomy were restored to native values. The correlations of MP and PP with pain are unclear; however, cartilage damage and OA progresses after meniscal injury, both potentially causing pain. ³⁶ A meniscectomized stifle treated with a prosthesis results in decreased PP which can potentially lead to less pain.

This study confirmed that the medial tibial CA in the meniscectomized stifles was smaller than that in the native stifles. Implantation of the prosthetic meniscus resulted in increased CA relative to the meniscectomized stifle. A decrease in CA has also been observed in studies excising the meniscotibial ligament, ³⁷ serial meniscectomies, ⁸ and the CCL. ³⁵ The reason for the relatively small CA measured in the stifle joint with the prosthetic meniscus may be the limitations of placing the Tekscan sensor. It was not possible to place the sensor completely beneath the prosthetic meniscus due to the presence of fixation screws. Therefore, the entire contact surface was not covered by the sensor, potentially leading to a lower CA in the prosthesis group than the actual CA. Other studies reported more precise sensor placement, but damage to the capsule and attachment eventually changed the kinematics. ³² , ³³ , ³⁷

In our study, screws were used to fix a prosthetic meniscus. It is reasonable to assume that rigid fixation of the prosthesis decreases load sharing relative to a more flexible method, which is a greater load on direct contact, because the meniscus converts compressive loads into hoop stresses. In a native stifle, both menisci move according to the tibiofemoral position. ³⁸ A change in fixation methods, such as flexible bands mimicking the native meniscal ligaments, may lead to a more equal distribution of tibiofemoral pressures. This may lead to decreased PP and MP and increased CA. A decrease in load sharing due to rigid fixation was also suspected in a study of humans with a similar prosthesis. ²⁸ In turn, excessive flexible fixation can extrude without converting a load into hoop stress; therefore, load sharing depends on the flexibility of the entire construct.

This study had limitations. The degree of decline in pressure that is clinically relevant is not known. In a human experiment, PP and MP decreased by 38% and 44%, respectively, from meniscectomy to implantation of the prosthetic meniscus. ²⁸ This resulted in pain relief. ²⁹ In this study, implanting the prosthetic meniscus decreased PP and MP by 46% and 57% relative to meniscectomy, respectively, thus potentially reducing pain in a clinical setting. A stifle with an intact CCL was used, and follow‐up studies may include a stifle with a sectioned CCL. The joint laxity of the soft tissue in cadavers (as observed in other studies), ³⁹ removal of most of the soft tissue, and removal of the CCL would have resulted in even less stability relative to the otherwise intact CCL‐deficient knee joint. This would have made it more difficult to perform the tests. Additionally, our goal was to examine the effects of removal of the meniscus and replacement with a prosthesis. Further alterations in contact mechanisms are likely to occur when the CCL is cut, ⁹ , ³² and the effect of the prosthetic meniscus can be affected. Furthermore, this was not a dynamic study; we did not study how the prosthesis will function during movements such as flexion and extension. In pressure mapping, it is important to determine how the pressure is distributed at certain forces. Axial forces of 400 N were applied to the femur, as a 2009 study showed that 30% of the bodyweight load of dogs weighing 28–35 kg on the hind leg resulted in a contact force of 395 ± 72 N in a normal stifle. ³² Therefore, we assume that a dynamic study on pressure mapping is not needed, even though the effects of flexion angle and translation are not considered and contact mechanics are significantly influenced by the flexion angle. ³⁹ The size and shape of our prosthesis did not vary. The prosthesis was oversized for certain joints; otherwise, the shape of the prosthesis seemed appropriate for all stifles. A more adapted size may improve the contact mechanics of the prosthesis with the femur and tibia and likely increase the CA while decreasing the MP. This issue can be solved when manufacturing a patient‐specific prosthesis according to the imaging of the contralateral stifle or when several off‐the‐shelf sizes are available. The human prosthesis was presoaked in water for at least 3 weeks before testing. This resulted in higher flexibility and, therefore, less load through the prosthesis. The prosthesis carries a higher load if it is not pre‐soaked (meaning it is a stiffer material), and this may lead to poor pressure distribution. Future in vivo safety and feasibility experiments in canines are required to evaluate the clinical performance and biological properties of prosthetic menisci. This prosthesis can potentially be used as an alternative treatment option for meniscal lesions.

This study demonstrated that a prosthetic meniscus significantly reduces the contact pressures in a meniscectomized stifle and restores the pressure distribution to native values. The stifle with the prosthesis showed no statistically significant difference in tibial contact pressure from that of the native stifle. Improvement in limb function could potentially be expected in dogs with meniscal injuries treated with prosthetic menisci.

AUTHOR CONTRIBUTIONS

Fohrmann F.M.C, DVM: Performed the surgeries and drafted the manuscript. Van Minnen BS, MSc: Performed the mechanical tests, worked on the statistics, and helped draft the manuscript. Krauss MW, DVM, DECVS: Performed the surgeries. Van der Veen AJ, PhD: Performed the mechanical tests and prepared the specimens. Tryfonidou MA, DVM, PhD, DECVS: Ensured ethical approval and advised on the statistics and the outlines of the manuscript. Meij BP, DVM, PhD, DECVS: Performed the surgeries and helped draft the manuscript. Van Tienen TG, MD, PhD: Performed the surgeries and participated in the study design. All authors have read, reviewed, and approved the final manuscript.

FUNDING INFORMATION

This study was funded by the MATMED program of Interreg North‐West Europe.

CONFLICT OF INTEREST STATEMENT

Van Tienen TG receives a management fee from ATRO Medical BV. Van Minnen BS and van der Veen AJ are employees of ATRO Medical BV. Van Minnen BS, van der Veen AJ, and van Tienen TG own stocks in ATRO Medical BV and are inventors of applicable patents but have no ownership. The patent publication numbers are as follows: WO2015057056 (2014), WO2015135907 (2015), WO2019125167 (2019), WO2022060218 (2022), WO2023/126435 (2023). The remaining authors have no conflicts of interest to declare.

Fohrmann FMC, van Minnen BS, Krauss MW, et al. Novel prosthetic medial meniscus restores contact pressures to native values in a meniscectomized canine stifle. Veterinary Surgery. 2026;55(2):335‐344. doi: 10.1111/vsu.70009