Long‐term clinical outcomes and retrieval analysis of a cementless total knee replacement in a dog

Abstract

Objective

The aim of the study was to describe long‐term outcomes and report post‐retrieval implant analysis following cementless total knee replacement (TKR) in a dog.

Animal

A seven‐year‐old, male neutered, Labrador retriever.

Methods

The dog presented for evaluation of chronic left pelvic limb lameness. Orthopedic examination identified bilateral cranial drawer and medial buttress. Radiographs revealed severe bilateral osteoarthritis with moderate joint effusion/synovial hypertrophy. Given the end‐stage osteoarthritis, TKR was considered more appropriate than tibial plateau leveling osteotomy (TPLO). The dog underwent a left cementless TKR.

Results

Immediate postoperative radiographs showed appropriate implant positioning. Moderate left pelvic limb lameness with full, pain free and stable stifle range of motion (ROM) was documented at 2 weeks. Increased left hindlimb weight‐bearing with a peak vertical force of 70% bodyweight (BW) versus 50% BW on the contralateral leg was reported at 6 weeks. Radiographs showed good implant osseointegration. Left stifle ROM was 50°/170°, a 30° increase compared to preoperative values. Additional rechecks at 14 and 30 weeks showed gradual improvement in stifle ROM and weight‐bearing. The dog was euthanized 6 years after surgery for reasons unrelated to TKR. Radiographs demonstrated static implant position without signs of osteolysis and gross examination revealed mild polyethylene wear on the caudal aspects of the tibial insert. Histological evaluation of the implant‐bone interface showed extensive and robust osseointegration.

Conclusion

This case demonstrates that cementless TKR can be associated with excellent clinical function over the course of at least 6 years and suggests that early surgical intervention could be considered.

1. INTRODUCTION

Total knee replacement (TKR) is considered a reliable procedure that can significantly enhance quality of life and improve clinical outcomes in dogs with end‐stage osteoarthritis (OA) when performed with optimal implant positioning. ¹ , ² Total knee replacement is indicated in cases of marked or progressive deformity of the stifle, severe OA or trauma. ³ , ⁴

Postoperative outcomes after TKR depend on adequate positioning and sizing of the surgical implants, meticulous operative technique and restoration of appropriate collateral ligament balancing ⁵ Inaccurate sizing or improper alignment can adversely affect load distribution at the bone‐implant interface and alter the tension in the collateral ligaments ⁶ Subsequent disruption of normal joint mechanics may lead to premature implant loosening, joint stiffness, collateral ligament failure and severe instability. ⁷ , ⁸

While short‐term outcomes have been associated with effective pain control and restoration of joint function, there is a dearth of information about mid‐ and long‐term clinical outcomes after canine TKR. Indeed, only one study followed patients for 12 months after cemented TKR. That study reported improvement in limb function despite marked preoperative OA. ¹ In veterinary medicine, reports on the long‐term survival of canine TKR components are lacking. Although osteolysis appears to be less important in canine TKR than in THR, implant wear remains an ongoing concern due to the semi‐constrained design of the prosthetic stifle joint. ⁹ While a number of factors such as surgical technique, implant alignment and design can affect the durability of a total knee prosthesis in human medicine, studies suggest that polyethylene wear has the most significant impact on the longevity of the prosthesis. ¹⁰ , ¹¹ Loosening and excessive wear are frequent reasons for revision surgery in human orthopedics. ¹²

To date, there are no data on implant survival/durability and long‐term clinical outcomes following TKR in dogs. Were these data available, they could be used to support rational decision‐making in terms of both the timing of surgery and the choice between cementless versus cemented TKR for a particular dog. The aim of this case report was to describe long‐term postoperative outcomes and implant survival in a dog implanted with a cementless TKR for 6 years.

2. MATERIALS AND METHODS

2.1. Case description

A 7‐year‐old, male neutered, 34‐kg Labrador retriever presented to our hospital for evaluation of a chronic left pelvic limb lameness of several years' duration. The severity of the lameness varied between non‐ to mild weight‐bearing (grade 2 to 4 out of 5 ¹³ ). A left cranial cruciate ligament rupture had been diagnosed several years earlier, but surgery had been declined. Physical examination was unremarkable, but orthopedic examination showed mild‐to‐moderate, toe touching, left pelvic limb lameness (grade 2 to 3 out of 5 ¹³ ) with crepitus as well as bilateral positive cranial drawer tests and overt medial buttresses. The overall stifle range of motion (ROM) was 90° (flexion 70° – extension 160°). Routine blood work and urinalysis were within normal limits. Orthogonal radiographic views revealed severe bilateral stifle OA, with moderate joint effusion/synovial hypertrophy and dystrophic mineralization secondary to cranial cruciate ligament disease (Figure 1). Considering the severity of the OA, a left TKR was recommended rather than a tibial plateau leveling osteotomy (TPLO).

FIGURE 1.

Preoperative radiographs showing the extent of stifle osteoarthritis.

2.2. Surgery

A left cementless TKR (BioMedtrix, Whipanny, New Jersey) was performed as previously described. ² Briefly, following a medial stifle arthrotomy the joint was fully exposed, including eversion of the patella (Figure 2A), then, using a dedicated external tibial alignment guide (Figure 2B), a tibial plateau ostectomy was completed to reduce the tibial plateau angle from 28° to 6°. A custom spacer, inserted between the femoral condyles and the resected tibial plateau, was used to balance tension in the collateral ligaments through the full range of joint motion (Figure 2C,D). To accommodate the keel of the prosthesis tibial component, a pilot hole was drilled in the tibial plateau, using a dedicated drill guide. Femoral ostectomies (cranial, caudal, distal and chamfer) were sequentially performed using a dedicated femoral cutting block secured to the condyle (Figure 2E). Using a dedicated drill guide, a pilot hole was drilled through the distal femoral ostectomy to accommodate the keel of the femoral component of the prosthesis. Trial implants were then applied to verify the adequacy of bone bed preparations and soft tissue tension. The final implants, a 34‐mm tibial component with a 7‐mm ultra‐high molecular weight polyethylene (UHMWPE) liner, and a 34‐mm femoral component were then sequentially press fitted (Figure 2F). Collateral ligament integrity and appropriate tensioning, as well as joint stability, were confirmed throughout ROM before closure. Postoperative radiographs confirmed adequate alignment and excellent component positioning (Figure 3 – top row).

FIGURE 2.

Intraoperative photographs. (A) Shows articular damage to the tibial plateau, femoral condyle and trochlea. Both cruciate ligaments were absent; note the presence of vestigial meniscal tissues. (B) Intraoperative photograh showing the lateral aspect of the tibia with the external tibial alignment guide used to achieve a tibial plateau angle to 6° while performing the tibial plateau ostectomy. (C) Due to lateral condyle hypoplasia, a custom stepwise insert was used to balance tension between the collateral ligaments during range of motion. (D) Femoral cutting guide secured to the femoral condyle. The yellow lines show the parallel relationship between the cut tibial surface and the femoral caudal cutting slot. (E) Femoral condyle and tibial plateau after completion of all ostectomies. (F) Intraoperative photograph showing the medial aspect of the left stifle with the prosthesis in situ prior to closure.

FIGURE 3.

Immediate (top row) and 30 weeks (bottom row) postoperative radiographs of left stifle showing the proper position of the cementless prosthesis.

3. RESULTS

3.1. Postoperative follow‐up

In the absence of complications, the dog was discharged 4 days postoperatively with analgesics and antibiotics. While enrolment in a formal physical rehabilitation program was recommended, it is unclear if this was implemented. The 2‐week recheck revealed moderate left (operated) pelvic limb lameness at the walk, with toe‐touching at rest (grade 2 to 3 out of 5 ¹³ ). Pain‐free and stable stifle ROM in flexion/extension, internal/external rotation as well as valgus/varus was documented. At 6‐weeks, near normal left pelvic limb function and moderate right pelvic limb lameness were evident at the walk and trot. Left and right peak vertical force (PVF) at the trot were 70% and 50% BW, respectively. Full, stable and pain‐free left stifle ROM was also documented. The overall ROM was 120 (50° in flexion to 170° in extension) an increase of 30° compared to preoperative values. Radiographs showed robust osseointegration of the implant components with no evidence of osteolysis. The 14‐week recheck yielded similar clinical, ROM and radiographic findings to those seen after 6 weeks, except for an improvement in right pelvic limb PVF from 50% to 68% BW. At the last recheck (30 weeks), while the clinical, radiographic (Figure 3 – bottom row), ROM and force plate data from the left pelvic limb were virtually unchanged, the PVF on the right pelvic limb had decreased to 45% BW, confirming worsening of the right pelvic limb lameness. At the owner's request, the dog was managed conservatively. No further follow‐up was available until 6 years after surgery when the dog was euthanized for reasons unrelated to its surgery or ongoing orthopedic problems.

3.2. Post‐mortem evaluation

Post‐mortem radiographs showed static femoral and tibial components, without evidence of periprosthetic osteolysis or aseptic loosening (Figure 4 – top row). Additionally, more severe right stifle OA compared to radiographs performed 6 years prior was also documented (Figure 4 – bottom row). Following soft tissue dissection of the left femur and tibia, while preserving the collateral ligaments, the implants were grossly evaluated. Both cementless components were stable on manual palpation, without overt evidence of structural damage. There was no evidence of gouging or scratching of the femoral component and no signs of metallosis or polyethylene wear debris in adjacent soft tissues (Figure 5).

FIGURE 4.

Top row: Retrieval radiographs obtained 6 years after surgery. Note the static position of the prosthetic components. There was no evidence of metallosis, aseptic loosening or any complications. Bottom row: Radiographs of the right stifle showing severe osteoarthritis and partial mineralization of the menisci. Total knee replacement surgery had been declined at the 6‐month postoperative recheck of the left total knee replacement.

FIGURE 5.

Photographs of the post‐mortem specimens. (A) Prosthesis in situ. Note the absence of metallosis or overt structural damages to the prosthetic components. (B) Articular surface of the tibial insert. Note the presence of mild abrasion wear at the gliding interface between the tibial and femoral components (red arrow) as well as cold‐flow structural damage at the caudal‐most medial edge of the insert (blue arrow). (C) Femoral and tibial components showing the locations and orientations of the histological sections on the femur (yellow dashline) and the tibia (blue dashline). Mirror sections were obtained on the opposite sides. An additional central section was obtained on the tibia (green dashline).

The specimen was initially fixed in 10% neutral buffered formalin, then stored in 70% alcohol until processing. Following fixation, the UHMWPE tibial insert was removed from its baseplate, stained with India ink and photographed. Compared to an un‐implanted UHMPWE tibial insert (Figure 6A,C), the only findings of damage were (1) mild plastic flow along the abaxial edges of the insert (lateral>medial – Figure 6B), (2) more severe plastic flow and delamination along the caudal edges of the implanted insert (medial>lateral – Figure 6D), and (3) loss of normal machining marks (Figure 6B) and occasional irregular scratching in the deepest part of the medial and lateral articulations.

FIGURE 6.

Analysis of surface damage/wear of the UHMPWE tibial insert. A non‐implanted insert is to the left (A, C) and an as‐retrieved insert to the right (B, D). Note the normal, fine machining lines (A, C) on the polyethylene surface. Damage, in the form of cold‐flow and delamination, is evident along the caudal edges of the medial and lateral articulations (yellow oval – lateral; pink oval – medial). Note the disappearance of the original machining lines (B, D). There is also mild cold‐flow along the abaxial border of the lateral articulation, leading to a slight distortion of this edge (convex shape – green arrows [B], compared to the straight edge of a non‐implanted tibial insert [A]).

In order to document long‐term osseointegration, the distal femur and proximal tibial were retrieved en bloc, serially dehydrated through a graded series of 70%, 80%, 95% and 100% alcohol, then embedded without decalcification in epoxy resin. The plastic‐embedded implants were sectioned in the sagittal plane using a water‐cooled diamond saw (Isomet 5000; Buehler UK Ltd) to generate sections through the medial and lateral femoral condyle and through the medial, central and lateral regions of the tibial baseplate (Figure 5C). Sections were manually ground to a final thickness of 150–200 microns on a grinder‐polisher (Ecomet 4; Buehler UK Ltd). Sections were microradiographed at 22 kV, 6 mAs (Faxitron MX‐20; Faxitron Corporation, Wheeling, Illinois). Good to excellent osseointegration was noted on both the femoral and tibial components (Figure 7). Sections were then stained with a combination of Stevenel's blue and van Gieson picrofuchsin and examined under transmitted white light microscopy. These images confirmed the presence of viable new bone (pink on these stained images) apposing and growing into the beaded porous ingrowth of the surface of both components (Figures 8, 9, 10). The cross‐sectional view through the medial region of the femur showed the implant in direct contact with bone. There was minimal fibrous tissue formation (yellow arrows) adjacent to the top and bottom part of the femoral component (Figure 8). Sections through central, medial and lateral regions of the tibia revealed fibrous tissue formation (yellow arrows, Figures 9 and 10) cranially and caudally, in areas devoid of the porous coating. In contrast, in areas with the porous coating, there was robust new bone formation along and between the beads (green arrows, Figures 9 and 10).

FIGURE 7.

Microradiographs of ground sections through the lateral (A) and medial (B) femoral condyle as well as through the medial (B – top), central (B – center) and lateral (B – bottom) aspects of the tibial plateau. These sections demonstrate good‐to‐excellent bone apposition against the undersurface of the press‐fit implant interfaces with the femur (green arrows) and tibia (yellow arrows).

FIGURE 8.

Histology through the medial condylar region of the femoral component after 6 years of in vivo service. There is direct apposition of bone (pink stained tissues) against the undersurface of the implant in all four zones. Note the presence of amorphous fibrous tissue (yellow arrows) immediately proximal to the implant in Zone 1 and immediately caudal to the implant in Zone 4.

FIGURE 9.

Histology of the lateral tibial implant‐bone interface after 6 years of in vivo service. There is a direct apposition of bone (pink stained tissues) against the cobalt‐chrome beads on the porous fixation surface (green arrows). In areas where there is no porous surface, especially cranially and caudally in Zones 1 and 2, amorphous fibrous tissue (yellow arrows) was present adjacent to and under the tibial component.

FIGURE 10.

Histology of the medial tibial implant‐bone interface after 6 years of in vivo service. There is direct apposition of bone (pink stained tissues) against the undersurface of the implant. In the central region, Zone 2, pink bone is visible between the cobalt‐chrome beads, indicating bone ingrowth into the porous structure (green arrows). In areas where there is no porous surface, especially cranially and caudally in Zones 1 and 3, amorphous fibrous tissue was present under the tibial component (yellow arrows).

4. DISCUSSION

This case report documents successful clinical and functional outcomes in a dog implanted with a cementless TKR for over 6 years. Analysis of the implant‐bone interface and the articular surface demonstrated adequate fixation and minimal wear over this implantation period.

Clinical outcomes TKR can be evaluated in different ways, including orthopedic examination, owner reported outcome measures (OROMs) ¹⁴ subjective (visual) lameness assessment, objective gait analysis, goniometry and thigh circumference measurements as estimates of joint ROM and muscle mass, respectively. These examinations are most often supplemented by diagnostic imaging to document implant position and stability. To date, there has only been one publication of long‐term clinical outcomes following cemented TKR with a maximum follow‐up of just 12 months. ¹

In this clinical patient with a cementless TKR, limb function and implant performance were assessed subjectively and objectively using a combination of orthopedic examination, plain and high resolution radiography, goniometry, force plate analysis, wear analysis and undecalcified histology.

Subjective lameness assessments demonstrated that the dog was moderately lame ¹³ on the operated left pelvic limb, with toe‐touching at rest, 2 weeks after surgery. Weight‐bearing improved by 6 weeks postoperatively and there was no visible lameness at 14 weeks postoperatively. Serial force plate analysis was performed to evaluate the extent of functional recovery following surgery. Force plate analysis was not performed prior to surgery because the dog was only occasionally toe‐touching on the left limb at the time of presentation. Postoperative gait analysis showed improvement in the left limb, with peak vertical forces of 70% bodyweight (BW) at 6, 14 and 30 weeks. These findings suggest satisfactory stability of the TKR implants. In contrast, the PVF of the right pelvic limb increased from 50% BW at 6 weeks to 68% BW at 14 weeks, then decreased to 45% BW at 30 weeks. These changes likely reflected off‐loading of the right pelvic limb in response to rapid progression of OA secondary to untreated cranial cruciate ligament rupture in the right stifle. Left stifle joint ROM increased over 30° over the first 6 weeks after surgery, reaching values that were similar to those of a normal, healthy stifle joint. ¹⁵ Taken as a whole, these findings demonstrate that the patient regained satisfactory limb function following cementless TKR and that the prosthetic joint was able to support the transfer of additional loading away from the diseased contralateral stifle joint.

The stability and performance of TKR components are determined by several factors, including implant design, the accuracy of implant positioning and joint alignment, the restoration of soft tissue balance ¹⁶ and the activity level of the patient in the postoperative period. While the design of the implant is largely beyond the surgeon's control, optimal implantation alignment and collateral ligament balance depend on meticulous execution of bone cuts, accurate positioning of the implants and appropriate tensioning of the surrounding soft tissues. Studies have demonstrated that implant malalignment increases load transfer and contact pressures on the UHMWPE tibial insert ¹⁷ accelerating implant wear and increasing the risk of implant failure over time. ⁸

In our case, radiography and gross evaluation of the stifle joint and implants 6 years after surgery did not identify any signs of implant failure, infection or osteolysis. The femoral and tibial components were stable and well‐positioned, without evidence of metallosis or joint instability. Normal patellar tracking was also observed.

Implant wear due to mechanical stress is one of the common causes of implant failure in human TKR. ¹⁶ Debris from the implant surfaces may detach and enter joint tissues, triggering inflammatory responses that lead to bone resorption and periprosthetic inflammation. ¹⁸ The resulting osteolysis leads to loss of bone at either the cement‐bone or the implant‐bone interface. ¹⁹ Over time, this biological phenomenon inevitably leads to mechanical loosening of the implant. Minimal implant wear, in the form of cold‐flow and mild delamination at the caudal edge of the implant, was evident on the tibial insert from this case. We suspect that this damage most likely resulted from minor cranio‐caudal as well as rotational tibial motion over 6 years, which we surmise could be mitigated by reducing the TPA or increasing the level of interfacial constraint via iteration of the prosthetic profile and thus congruity. This may improve stability and reduce UHMWPE wear. In an attempt to enhance wear resistance, current tibial inserts are manufactured from highly crossedlinked and vitamin E‐stabilized UHMWPE, which was not available at the time of surgery. Studies showed that achieving proper soft tissue balancing and an adequate alignment of the prosthetic components can facilitate long‐term implant survivorship, even in the presence of implant wear that may occur over time. ²⁰

While primary stability of cementless TKR relies on initial press fit, bone ingrowth at the implant/bone interface determines long‐term stability. Although we did not use mechanical testing to document implant fixation in this dog, manual manipulation of the femoral and tibial components post‐mortem did not reveal any gross instability. This observation was supported by our radiographic and histological findings which indicated robust bone formation on and within the porous fixation surfaces on both femoral and tibial components. While it has been suggested that variability of bone ingrowth may be caused by relative motion between implants and bone ²¹ the findings from both microradiography and histology, in this case, were qualitatively similar to those reported previously from a series of preclinical TKR cases with 14‐month follow‐up ² and provide indirect evidence that the implants were mechanically stable. Additionally, retrieval studies from human TKR have suggested that fibrous tissue ingrowth occurs in the area where bone ingrowth is missing. ²² , ²³ This is consistent with our observation of fibrous tissue only in the cranial and caudal regions of the tibial baseplate, in areas where there is no porous surface available for bone ingrowth.

As is the case with cementless total hip replacement, cementless TKR continues to gain popularity in human and veterinary medicine due to the numerous advantages it offers over cemented implants. These advantages include enhanced osseointegration with bone stock preservation. In an attempt to hasten bone ingrowth, the original cobalt chrome beaded tibial trays, used in this case, have been replaced with titanium components featuring an EBM preparation of the osteointegration surface. Other benefit of cementless TKR include shorter operative times, avoidance of exposure to potentially dangerous methyl methacrylate vapor in the operating room, elimination of the intraoperative risks of embolic showering and systemic hypotension due to cement pressurization, and a reduced risk of osteolysis due to the removal of bone cement as a source of wear debris. ²⁴ Although the reported risk of revision in people is higher with cementless TKR compared to cemented implants (most commonly as a result of aseptic loosening of the tibial component), encouraging results have been seen at 10‐year follow‐up for cementless femoral components. ²⁵ , ²⁶ , ²⁷

In veterinary medicine, data on the survivorship of cementless TKR in dogs remains scarce, but the limited number of reports that are available have demonstrated successful outcomes with cementless tibial fixation. ²¹ , ²⁸ The results from this clinical case lend further support to this earlier work, as well as to encouraging early results with cemented canine TKR published by Liska and Doyle ¹ and by Eskelinen et al. ²⁹ Although larger, multicenter randomized clinical trials are needed to directly compare outcomes in cemented and cementless TKR in dogs, the long‐term survival of these implants opens up the potential for their use in younger dogs with severe stifle joint disease.

AUTHOR CONTRIBUTIONS

All authors equally contributed to conception of the study, study design, data acquisition and interpretation. Fracka AB, DVM, MSc, PgCertSAS, PgCertECC, AFHEA, MRCVS: Drafted, revised and approved the submitted manuscript. Dejardin LM, DVM, MS, DACVS, DECVS: Performed surgery with the assistance of MJA. Allen MJ, Vet MB, PhD, MRCVS: Performed post‐mortem evaluation of the TKR specimen and implants. Both LMD and MJA produced images, revised and approved the submitted manuscript.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest related to this study.

Fracka AB, Allen MJ, Dejardin LM. Long‐term clinical outcomes and retrieval analysis of a cementless total knee replacement in a dog. Veterinary Surgery. 2025;54(3):621‐631. doi: 10.1111/vsu.14175