Bones of the Stifle Joint¹¹⁸

The quadrangular distal end of the femur protrudes caudally and contains three major articular areas, one each on the medial and lateral femoral condyles and the third within the femoral trochlea on the cranial surface. The medial and lateral femoral condyles are thick, roller-like surfaces that are convex in both the sagittal and frontal planes and are separated by the intercondyloid fossa. The fossa is slightly oblique in orientation, such that the caudal part is located more laterally than the cranial part. The caudal articular surface of the condyles is continuous with small articular facets on adjacent epicondyles that articulate with the fabellae, small sesamoid bones in the tendons of origin of the gastrocnemius muscle. The femoral trochlear (patellar) surface is the smooth, wide, grooved articular surface of the cranial femur that is continuous with the articular surfaces of the femoral condyles. The crests of the trochlea, the medial and lateral trochlear ridges, diverge proximally. The patella articulates with the femoral trochlea.

The proximal end of the tibia is slightly convex in the sagittal plane and triangular in cross-section, with its apex cranial. The proximal articular surface is divided into medial and lateral condyles, and each has approximately the same surface area; however, the medial condyle is oval and the lateral is nearly circular. A sagittal, nonarticular area known as the intercondylar eminence separates these two articular areas. The medial and lateral intercondylar tubercles are triangular projections atop the intercondylar eminence; these tubercles articulate with the femur on their abaxial surfaces. The cranial intercondyloid area is an oval depression cranial to the intercondylar eminence that serves as the attachment site for the cranial cruciate ligament and the cranial meniscal ligaments (Figure 62-1). The caudal intercondyloid area is a smaller depression caudal to the intercondylar eminence that serves as an attachment site for the caudal meniscal ligaments. At the caudal-most aspect, the broad popliteal notch separates the tibial condyles; the caudal cruciate ligament attaches to the lateral edge of the popliteal notch. The extensor (muscular) groove is a small notch at the cranial margin of the lateral tibial condyle that extends to the margin of the articular surface of the lateral condyle. The long digital extensor tendon runs through this groove. In humans, the angular, cranial margin of the extensor groove is known as the tubercle of Gerdy.³⁹ The tibial tuberosity is the large, quadrangular cranioproximal process that serves as the attachment for the patellar ligament, as well as for parts of the biceps femoris and sartorius muscles. The cranial border of the tibia (formerly called the tibial crest) extends distally from the tibial tuberosity. The obliquely placed facet for articulation of the fibular head, the facies articularis fibularis, is on the caudolateral surface of the tibial condyle. A small tubercle on the medial aspect of the head of the fibula articulates with the tibia at this facet.

Sesamoid Bones of the Stifle Joint¹¹⁸

The patella, an ossification in the tendon of insertion of the quadriceps muscle group, is the largest sesamoid bone in the body. It is ovate in shape and slightly curved to articulate with the femoral trochlea. The blunt, proximal portion is termed the base, and the slightly pointed distal surface is termed the apex. The articular surface is smooth and convex in all directions. The part of the tendon of the quadriceps femoris muscle between the patella and its insertion on the tibial tuberosity is called the patellar ligament; the patellar ligament is sometimes referred to metonymically as the patellar tendon. The patella alters the direction of pull of the tendon of the quadriceps femoris muscle as it travels over the femoral trochlea (pulley), provides a greater bearing surface for the tendon on the trochlea, and protects the tendon. The cranial articular surface of the stifle joint is further increased by the presence of two or three parapatellar fibrocartilages. These grooved cartilages are embedded within the lateral and medial femoral fasciae, ride on the trochlear ridges of the femur, and tend to prevent patellar luxation. Proximally, these two cartilages curve toward each other and meet, or a third cartilage may be located between them.

Three other small sesamoid bones, or fabellae, are situated in the stifle region. The lateral and medial fabellae are located in the lateral and medial heads of origin of the gastrocnemius muscle, respectively. The lateral fabella is larger and nearly spherical, except for a truncated end distally and a flat articular surface that articulates with the facet on the caudal aspect of the lateral femoral condyle. The medial fabella is smaller and more angular in shape. It articulates with the small facet on the caudal aspect of the medial femoral condyle, but this articulation is inconsistent. The popliteal sesamoid bone is the smallest sesamoid; it is located within the tendon of origin of the popliteus muscle, adjacent to its muscle fibers, and articulates with the lateral condyle of the tibia.

Articulations of the Stifle Joint¹¹⁸

The main articulation of the stifle is the femorotibial joint, which comprises the articulation between the thick, roller-like condyles of the femur and the flattened condyles of the tibia; this joint is the primary weight-bearing articulation. The congruity between the articular surfaces of the femoral and tibial condyles is enhanced by the interposition of the medial and lateral menisci. The femoropatellar joint communicates freely with the femorotibial articulation, and the two joints are interdependent, because the patella is firmly attached to the tibia by the patellar ligament. Thus, any movement between the femur and the tibia results in motion between the patella and the femur. The role of the femoropatellar joint is to improve the efficiency of the extensor mechanism by increasing the moment arm of the quadriceps muscles. The proximal tibiofibular joint is the final component of the stifle joint.

The stifle joint capsule forms three freely communicating sacs, one each at the medial and lateral femorotibial articulations, and the third between the patella and the femur. The femorotibial sacs are divided into femoromeniscal and tibiomeniscal parts by the menisci. Distal to the patella, the synovial and fibrous layers of the joint capsule are separated by the infrapatellar fat pad (body); thus the fat pad is extrasynovial. Small recesses of the condyloid joint space are found caudally and proximally as the joint capsule extends between the caudal, proximal parts of the femoral condyles and the fabellae that articulate with them. Three other subpouches exist in the lateral femorotibial joint capsule. A small extension traverses laterally between the head of the fibula and the lateral condyle of the tibia. Another extends distally through the extensor groove, and the last partially surrounds the tendon of origin of the popliteal muscle at the level of the lateral femoral condyle and serves as a synovial bursa. A small synovial bursa is frequently located between the patellar ligament and the tibial tuberosity, just proximal to the attachment of the ligament.

Ligaments of the Stifle Joint

The menisci are attached to the tibia by a series of paired ligaments. The cranial meniscotibial ligament of the medial meniscus runs from the cranial axial border of the medial meniscus to attach to the tibia at the cranial intercondyloid area immediately cranial to the intermeniscal ligament, the attachment of the cranial meniscotibial ligament of the lateral meniscus, and the tibial attachment of the cranial cruciate ligament (Figure 62-2). The cranial meniscotibial ligament of the lateral meniscus attaches to the cranial intercondyloid area of the tibia just caudal to the attachment of the cranial meniscotibial ligament of the medial meniscus and the intermeniscal ligament (see Figure 62-2). The caudal meniscotibial ligament of the medial meniscus runs from the caudal axial border of the medial meniscus to attach to the caudal intercondyloid area of the tibia (see Figure 62-1). The caudal meniscotibial ligament of the lateral meniscus runs from the caudal axial border of the lateral meniscus to attach in the popliteal notch, just caudal to the caudal intercondyloid area (see Figure 62-2). These ligaments are also known as the cranial and caudal tibial ligaments of the respective menisci.¹¹⁸ The meniscofemoral ligament of the lateral meniscus, also known as the femoral ligament,¹¹⁸ runs dorsally from the caudal axial border of the lateral meniscus to attach within the intercondylar fossa of the femur as it blends into the medial femoral condyle. The intermeniscal ligament is a transverse fibrous band that extends from the caudal side of the cranial tibial ligament of the medial meniscus to the cranial side of the cranial tibial ligament of the lateral meniscus (see Figure 62-2).¹¹⁸

Four femorotibial ligaments in the stifle joint provide primary ligamentous support: two collateral ligaments and two cruciate ligaments. The cruciate ligaments are located within the stifle joint (intra-articular), but because they are covered by synovium, they are considered to be extrasynovial. These ligaments are designated cranial and caudal based on their tibial attachment. In addition, they decussate, or cross, each other—hence their name. The cranial (lateral) cruciate ligament attaches to the caudomedial aspect of the lateral femoral condyle and the caudolateral part of the intercondyloid fossa of the femur and runs diagonally in a cranial, medial, and distal direction across the intercondyloid fossa to attach to the cranial intercondyloid area of the tibia.20,99,156 This ligament is grossly divided into a larger caudolateral part that attaches at the caudolateral aspect of the tibial attachment site and a smaller craniomedial band that attaches at the craniomedial aspect of the tibial attachment site. The craniomedial fibers spiral outward axially approximately 90 degrees as the ligament courses from proximal to distal.²⁰ The caudal (medial) cruciate ligament attaches to the lateral surface of the medial femoral condyle and runs caudodistally to attach to the medial edge of the popliteal notch of the tibia, medial to the caudal meniscotibial ligament of the lateral meniscus. The caudolateral fibers spiral inward abaxially about 90 degrees as the ligament courses from proximal to distal.²⁰

The caudal cruciate ligament is larger and longer than the cranial cruciate ligament. The cruciate ligaments comprise a core region of fascicles containing collagen fibrils and fibroblasts, and are covered by an epiligamentous region composed of synovial intima and underlying loose connective tissue. The epiligamentous region is absent only where the cranial cruciate ligament wraps around the caudal cruciate ligament.¹⁵²

Abundant mechanoreceptors and proprioreceptors are located within the center of the ligament; innervation arises from afferent axons that penetrate from the synovium that envelops the ligament.18,388 The lateral (fibular) collateral ligament attaches proximally to an oval area on the lateral epicondyle of the femur and passes superficial to the tendon of origin of the popliteus muscle. The distal attachment is found primarily on the head of the fibula, with a few fibers attaching to the adjacent lateral condyle of the tibia.^118,361 The medial (tibial) collateral ligament attaches proximally to an oval area on the medial epicondyle of the femur.³⁶¹ As it courses distally, it blends with the joint capsule and has a strong attachment to the medial meniscus.³⁶¹ It passes superficial to the tibial insertion of the semimembranosus muscle and over a bursa at the medial tibial condyle to attach distally to a long rectangular area at the medial border of the tibial metaphysis.³⁶¹ The medial collateral ligament is fused with the joint capsule and the medial meniscus, while the lateral collateral ligament is attached only loosely to the joint capsule and is separated from the lateral meniscus by the tendon of origin of the popliteus muscle.³⁶¹ The thin medial and lateral femoropatellar ligaments are continuations of the femoral fascia that originate from the sides of the patella. The lateral femoropatellar ligament is better defined and attaches to the lateral fabella. The poorly defined medial femoropatellar ligament blends with the periosteum at the medial epicondyle of the femur.

Meniscus

Meniscal Function

The menisci provide several vital functions in the stifle joint, including load bearing, load distribution, shock absorption, and joint stability. Their wedge shape is ideal to act as a spacer between the femoral condyle and the tibial plateau, and when no compressive weight-bearing loads are present across the joint, it limits contact between articular surfaces. Under static-loading conditions, the menisci assume a significant load-bearing function in the stifle joint. As the joint is loaded, contact between the femoral condyle and the meniscus increases, and the larger contact area created by the meniscal-articular interface lowers the stress of the articular cartilage of the femur and tibia, protecting against mechanical damage to both the chondrocytes and extracellular matrix. Together, the menisci bear between 40% and 70% of the load across the stifle joint.²⁰¹ The role of the meniscus as a weight-bearing structure was demonstrated in a cadaveric study that evaluated the effect of segmental hemimeniscectomy and meniscal release on contact pressures in the medial compartment of the stifle joint. Removal of the caudal horn of the medial meniscus causes a focal area of high pressure in the caudal region of the medial tibial condyle, corresponding to pressures higher than 10 MPa.²⁸⁶ This alteration of articular cartilage contact pressures is one of the factors contributing to articular cartilage degeneration following meniscectomy.^88,119,387

The response of the menisci to loads applied to the stifle joint results from their macrogeometry, their well-organized structure, and the nature of their proximal and distal attachments.¹²⁶ Different models demonstrate that the meniscus absorbs energy by undergoing elongation as a load is applied to the knee. As the joint compresses, the wedge-shaped meniscus extrudes peripherally, and its circumferentially oriented collagen fibers elongate (hoop stress). The force required to restrain the radial extrusion of the meniscus is derived from the large tensile hoop stress developed within the strong circumferential collagen fiber bundles caused by the extrusive effect. These hoop forces are transmitted to the tibia through the strong cranial and caudal attachments of the menisci. The importance of an intact functional unit (meniscal ligaments, meniscal body, and peripheral rim) is evident when the effect of meniscal release on load transmission across the joint is considered. Transection of the caudal meniscotibial ligament of the medial meniscus causes a 140% increase in peak contact pressure and a 50% decrease in contact area.²⁸⁴ A midbody (central) meniscal release and a hemimeniscectomy cause similar changes²⁶ (Figure 62-6), further demonstrating that the meniscus should be considered as a whole functional unit that includes intact ligaments, the peripheral rim, and most of the meniscal body. The concept of the meniscus as a functional unit should serve as the basis for decision making in meniscal treatment.

The menisci contribute to joint stability by increasing the congruity of the femorotibial joint.²⁸⁵ The menisci provide concavity to the relatively convex tibial plateau. Their effect is evident when it is considered that the meniscus functionally decreases the tibial plateau slope (Figure 62-7), as the prominent caudal horn effectively raises the caudal aspect of the tibial plateau. The role of the menisci is different between the cranial cruciate ligament–deficient stifle joint and the intact stifle joint.²⁸⁵ In the cranial cruciate ligament–deficient stifle joint, the caudal pole of the meniscus acts as a wedge, preventing the tibia from further subluxation and thereby playing a primary role in joint stability. This wedge effect increases the risk of meniscal tear in the untreated cranial cruciate ligament–deficient joint. In the normal stifle joint, loss of the meniscus causes minimal translation, suggesting a secondary role in joint stability of the meniscus when the cranial cruciate ligament is intact. The contribution of the meniscus to joint stability and kinematics in the cranial cruciate ligament–deficient stifle joint depends on the magnitude of instability. In a cadaveric study investigating the contribution of the meniscus to craniocaudal stability in cranial cruciate ligament–deficient stifles stabilized by tibial plateau leveling osteotomy, tibial plateau leveling osteotomy eliminated the wedge effect of the meniscus, suggesting a protective effect of tibial plateau leveling osteotomy against postliminary (occurring after the index [initial] surgical procedure) meniscal tears.²⁸⁵ However, as suggested by the incidence of postliminary meniscal tears,^74,274,349 tibial plateau leveling osteotomy may be less likely to protect the meniscus during activities other than straight-line walking. In more provocative gaits, internal-external rotational instability may be more relevant than simple craniocaudal translation, as reported during walking on a treadmill.³⁴⁴

The biomechanics of meniscal treatments in dogs has been extensively studied in an attempt to understand its effect on meniscal and joint function.188,201,284–287 As noted through the work of Fairbank, total meniscectomy is not a benign operation.¹¹⁹ Cox and others found that meniscectomies in stifle joints lead to gross and microscopic degenerative changes in the articular cartilage. They noted that partial meniscectomies lead to less severe degenerative changes.⁸⁸ They believed there was a direct relationship between the degree of degenerative change and the amount of meniscus removed. This supposition was recently confirmed in a cadaveric study evaluating the effect of serial meniscectomies on femorotibial contact pressures and on meniscal strain.²⁸⁷ The authors reported that smaller (30% radial width) partial meniscectomies had minimal effects on the biomechanics of meniscal function, whereas larger partial (75% radial width) and hemimeniscectomies resulted in significant changes in meniscal and femorotibial contact mechanics (Figure 62-8).²⁸⁷ The authors of this study speculated that loss of peripheral meniscal tissue eliminates the spacer effect of the meniscus, which is necessary for hoop tension to develop. The authors suggested that to act as an effective functional unit, the meniscus needs more than 25% of the radial width of the peripheral meniscal tissue. Using a similar ex vivo model, 75% partial meniscectomy was compared with meniscal repair to evaluate whether repairing a vertical longitudinal tear located in the red-red zone could be advantageous for the load-bearing function of the meniscus. Meniscal repair was effective in decreasing the peak contact pressure and increasing the contact area caused by a bucket-handle tear.³⁴⁸ In contrast, partial meniscectomy did not change the contact pressure pattern noted under the torn meniscus. Although clinical recommendations should not be based on cadaveric studies, interpretation of the large body of literature investigating the in vivo effect of meniscectomy on progression of osteoarthritis^88,387 strongly indicates that a prudent approach to meniscal resection is to preserve the greatest amount of functional meniscal tissue. A conservative approach may also be valid for subclinical meniscal tears because a torn meniscus may continue to facilitate load transmission, as shown in the case of stable vertical longitudinal tears in a laboratory study.³⁴⁷

Effect of Meniscal Release on Meniscal Function

Meniscal release is complete transection of the medial meniscus at the meniscotibial ligament or through the midbody of the meniscus (Figure 62-9), with the goal of eliminating the wedge effect of the caudal horn of the medial meniscus during femorotibial subluxation. Meniscal release has been investigated in both in vivo and ex vivo experiments, to determine its biomechanical and biologic effects on joint function.^{188,222,284,286} Transection of the caudal meniscotibial ligament of the medial meniscus caused a 50% decrease in contact area and a 140% increase in the magnitude of pressure on the medial compartment of a cranial cruciate ligament–deficient stifle joint treated with a tibial plateau leveling osteotomy.²⁸⁴ In addition, a significant caudal shift of load transmission was noted after release. The abnormal contact mechanics following transection of the meniscotibial ligament was similar to that observed after a midbody meniscal release; based on these findings, a caudal release would not appear to confer a biomechanical advantage compared with a midbody meniscal release.^284,286 As a result of these biomechanical alterations, supraphysiologic loading of articular cartilage may induce an upregulation in synthesis and degradation of cartilage matrix, which ultimately may lead to osteoarthritis. It is interesting to note that the focal high-pressure area observed in cadaveric experiments after meniscal release^284,286 is consistent with the focal cartilage defect that Luther et al. reported in an in vivo experimental model of caudal release.²²² It is likely that a combination of inflammatory and degradative mediators originating from the transected meniscus and biomechanical abnormalities from the loss of hoop tension play key roles in the development of cartilage degeneration following caudal release.²²²

Meniscal release is equivalent to caudal hemimeniscectomy with regard to meniscal function, further supporting the importance of an intact functional unit formed by the meniscal ligaments and the peripheral rim.²⁸⁷ The effects of medial caudal hemimeniscectomy and total meniscectomy were investigated in vivo in stifle joints of clinically normal dogs to evaluate their effect with respect to development of osteoarthritis (Figure 62-10).¹⁸³ The results of this study indicate that the severity of secondary osteoarthritis induced in stifle joints at 16 weeks after both types of meniscectomies is similar in terms of pathologic changes in the joint, but the less radical meniscal excision is associated with less disruption of chondrocyte metabolism.¹⁸³ Additionally, this study confirmed that incomplete meniscal regeneration, a semilunar-shape regrowth structure derived from the synovial membrane, is formed in 3 to 6 months.¹⁸³ It is important to understand that this regenerated meniscus is not as functional as the original meniscus and does not prevent the development of osteoarthritis.^101,183

Theoretical Models of Stifle Joint Stability and Instability

In the dog, the dynamic nature of instability in the cranial cruciate ligament–deficient stifle joint was first clinically described by Henderson and Milton in 1978, when they described the tibial compression test as a diagnostic test for evaluation of the integrity of the cranial cruciate ligament.¹⁵⁷ In this report, they recognized that two major factors account for the joint compressive force between the tibia and the femur: direct forces of weight bearing, and contraction of the gastrocnemius muscle.¹⁵⁷ In 1984, Slocum described the cranial tibial thrust force as a shear force generated in the stifle during weight bearing that acts to thrust the tibia cranially.³²² The authors concluded that the cranial tibial thrust force is a result of tibial compression and the slope of the tibial plateau.³²²

Slocum proposed the Active Model of the Stifle in 1993 (Figure 62-11).³²⁵ In this model, stifle joint stability is maintained by a synergism between the muscle forces responsible for stifle joint flexion and extension, the cranial tibial thrust force, the pull of the stifle flexor muscles of the thigh, and the passive restraints of the stifle joint including the cranial cruciate ligament and the caudal pole of the medial meniscus (Figure 62-11, A).³²⁵ Based on this model, the magnitude of the cranial tibial thrust force is dependent on the magnitude of the joint compressive force and the slope of the tibial plateau. In the normal stifle joint, the cranial tibial thrust force is counteracted by active elements such as the caudal thigh muscles and by passive elements including the cranial cruciate ligament and the caudal pole of the medial meniscus. According to this model, if the cranial tibial thrust force is not neutralized by the pull of the stifle flexors of the thigh, the cranial cruciate ligament begins to rupture or ruptures, and cranial tibial subluxation occurs during the stance phase of the gait. Leveling the tibial plateau (Figure 62-11, B) reduces the magnitude of the cranial tibial thrust force to regain stifle joint stability during the stance phase of the gait.^322,325

An alternative biomechanical model of the stifle was proposed by Tepic in 2002 (Figure 62-12), based on the work of Nisell.²⁶² The Tepic model suggests that the total joint force (joint reaction force) is not parallel to the functional axis of the tibia as proposed by Slocum, but instead is parallel to the patellar ligament (Figure 62-12, A). Tepic theorized that in Slocum’s model, which was developed based on analysis of the tibial compression test, the total joint force was presumed to be parallel to the Achilles tendon, which is essentially parallel to the functional axis of the tibia as defined by Slocum. However, in contrast to Slocum, it was theorized that under weight-bearing conditions, the force applied to the paw is not similar to the moment applied to the paw during the tibial compression test. Instead, Tepic suggests that the force applied to the paw is parallel to the patellar ligament. Thus, based on this model, stabilization procedures should be aimed at leveling the tibial plateau such that it is perpendicular to the patellar ligament, or altering the angle of the patellar ligament such that it is perpendicular to the tibial plateau (Figure 62-12, B).³⁴⁵ An alternative method of patellar tendon (ligament) angle (PTA) measurement based on the common tangent at the tibiofemoral contact point was a further refinement of this model.⁹⁶

Avulsion of the Cranial Cruciate Ligament

In skeletally immature animals, the attachment of ligament to bone by Sharpey’s fibers may be materially stronger than the bone itself. Thus, an acute overload of the ligament, such as a fall from a height, may result in avulsion of the ligament and a small piece of bone from the femoral (Figure 62-13, A) or, more commonly, the tibial (Figure 62-13, B) attachment site, rather than a midsubstance ligamentous tear. Because this injury more closely resembles an avulsion fracture, rather than ligamentous injury, it may be amenable to primary repair by reattachment of the avulsed fragment to bone (Figure 62-14).

Treatment Considerations

Proximal tibial epiphysiodesis (i.e., surgically induced premature union of the epiphysis with the diaphysis) has been described as a sole treatment to reduce the tibial plateau angle in young dogs with cranial cruciate ligament–deficient stifle joints, or to augment primary repair of a ligamentous avulsion (Figure 62-15, A).³⁶⁶ A Kirschner wire is inserted into the center of the cranial intercondyloid area of the tibia and is oriented parallel to the tibial shaft. Correct placement of the Kirschner wire is confirmed by intraoperative fluoroscopy or radiography, and a cannulated bone screw is advanced over the Kirschner wire (Figure 62-15, B–C). Epiphysiodesis occurs in the cranial aspect of the physis while the caudal aspect of the physis continues to grow, resulting in a reduction of the tibial plateau angle as long as residual growth remains at the time of surgery (Figure 62-15, D–E). In a case series of 22 joints, a reduction in the tibial plateau angle of 4 degrees to 24 degrees was observed, and an improved or normal gait was evident at follow-up examination in 18 of 22 joints.³⁶⁶ Valgus deformity occurred in 3 of 22 joints as the result of eccentric insertion and/or angulation of the screw (see Figure 62-14, C and E),underscoring the importance of precise screw positioning. The authors suggest intraoperative verification of Kirschner wire and screw position by fluoroscopy or radiography.³⁶⁶

Acute Traumatic Rupture of the Cranial Cruciate Ligament

Excessive limb loading, traumatic hyperextension, and/or excessive internal rotation of the tibia may overload the cranial cruciate ligament, resulting in acute rupture. This is a rare injury; however, when it occurs, dramatic pain, joint effusion, severe lameness, and stifle joint instability are present. This injury usually results in a midsubstance “mop end” tear (Figure 62-16, A), and radiographic findings include a dramatic effacement of the infrapatellar fat pad by a soft tissue opacity in the lateral view, consistent with edema of the fat pad and/or joint effusion, and the absence of osteophytosis (Figure 62-16, B).

Progressive Degeneration of the Cranial Cruciate Ligament

Despite years of clinical and basic science investigation since the description of cruciate ligament disease in the dog in 1926, the cause and pathogenesis of this extremely common disorder remain elusive. Biomechanical evaluations of intact cranial cruciate ligament preparations have demonstrated a decrease in material properties (modulus of elasticity, maximum stress, strain energy) with aging. These changes were more pronounced and occurred at an earlier age in dogs weighing more than 15 kg as compared with dogs weighing less than 15 kg.³⁶³ Histologic evaluation of these ligaments revealed evidence of degenerative changes, such as loss and metaplasia of ligamentocytes and failure to maintain collagen fibers.³⁶³ These findings are consistent with the observation that cruciate disease tends to occur at a younger age in large-breed dogs.⁴⁴

The exact cause of cranial cruciate ligament rupture is poorly understood. Factors that have been implicated include abnormal conformation and gait,²⁶ increased tibial plateau angle,^{223,256,293,326} obesity,^209,363 and lack of fitness⁴⁴; however, none of these has proved causative. For instance, the mean tibial plateau angle was reported to be significantly greater in dogs with cranial cruciate ligament rupture than in dogs with an intact cranial cruciate ligament.²⁵⁶ However, additional investigations failed to substantiate this finding.^297,381 A retrospective study revealed that neutering increases the prevalence of cranial cruciate ligament injury.³²⁰ It has been shown that the cranial cruciate ligament is rich in mechanoreceptors and proprioceptors,^18,388 and that joint loads that cause increased strain in the cranial cruciate ligament result in simultaneous contraction of the caudal thigh muscles and relaxation of the quadriceps muscle group,^242,329 which is protective of the cranial cruciate ligament. Thus, it is possible that obesity and/or poor physical condition may mitigate these protective mechanisms, resulting in repetitive strain injury of the cranial cruciate ligament, leading to mechanical failure.^300,329

Histologic examination of ruptured cranial cruciate ligaments revealed lack of collagen fiber maintenance³⁶³ and loss of fibroblasts from the core region, despite a normal cell number density in the epiligamentous (thin layer of connective tissue covering the ligament) region.¹⁵² The decline in fibroblast numbers was further characterized by a decrease in the numbers of typical fusiform and ovoid fibroblasts, an increased number of cells undergoing chondroid metaplasia, and extensive disruption of the ligamentous matrix.^152,363 Decreased birefringence and elongation of crimping in the remaining collagen fibrils were identified, which implicates progressive mechanical overload as a cause of failure.¹⁵² A proliferative epiligamentous repair response was identified that resulted in eventual covering of the torn ends of the ligament. However, no bridging scar was identified between the ends.¹⁵² Factors that may contribute to this degeneration include immune-mediated degeneration,^137,261 acquired loss of blood supply in the midportion of the ligament,³⁶³ and a smaller intercondylar notch width.⁶ In simplistic terms, the ligament is too weak to withstand the forces applied to it, or the forces applied are greater than the strength of the ligament. In all likelihood, various interrelated factors contribute to rupture of the cranial cruciate ligament in the dog.

The extracellular matrix of ruptured cranial cruciate ligaments has been shown to have an increased matrix turnover, demonstrated by increased collagen and glycosaminoglycan synthesis, compared with intact cranial cruciate ligaments.⁸⁰ Antibodies to type I and II collagen have been found in the sera and synovial fluid of dogs with spontaneous cranial cruciate ligament rupture, leading to speculation that immunologic reactivity may play a role in cruciate ligament disease.²⁶¹ However, in a more recent study comparing the prevalence of antibodies to collagen types II and I in the synovial fluid of dogs with cranial cruciate ligament disease, and dogs with osteoarthritis secondary to other pathologic processes, the authors concluded that the increase in synovial fluid antibodies was not specific for the type of joint disorder. In addition, the investigators suggested that it is unlikely that the anticollagen antibodies play an active role in the initiation of weakening of the cranial cruciate ligament.⁹⁸

Breed variation in the material properties of the cranial cruciate ligament has been reported. One in vitro investigation comparing the cranial cruciate ligament from the Rottweiler with that of the racing Greyhound revealed that Rottweiler ligaments had a significantly greater cross-sectional area at their tibial attachment. Mechanical testing of these ligaments showed that the ultimate failure load, normalized to body mass, was significantly greater in the extended racing Greyhound stifle during loading in cranial tibial subluxation. The conclusion from this work was that the Rottweiler ligament is more vulnerable to damage because it requires half the load per unit body weight to rupture compared with the Greyhound ligament.³⁸⁵

Cranial cruciate ligament disease affects a wide variety of dog breeds; in one study, the highest prevalence was seen in the Rottweiler, Newfoundland, and Staffordshire Terrier, and the lowest prevalence in affected breeds was noted in the Dachshund, Basset Hound, and Old English Sheepdog.³⁷⁷ In another study of dogs that sustained cranial cruciate ligament rupture before 2 years of age, predisposed breeds included the Neapolitan Mastiff, Akita, Saint Bernard, Rottweiler, Mastiff, Newfoundland, Chesapeake Bay Retriever, Labrador Retriever, and American Staffordshire Terrier.¹¹² Female dogs have an increased prevalence of cruciate ligament disease compared with male dogs,³⁷⁷ and in both studies, neutered dogs had a higher prevalence than sexually intact dogs.^112,377 However, age at the time of ovariohysterectomy was not associated with the prevalence of cruciate disease.³⁷⁷ Several studies have documented that smaller dogs—those weighing less than 22 kg—tend to be affected later in life than larger dogs.^{37,44,365,377} This epidemiologic finding is consistent with histologic analyses revealing that the cranial cruciate ligaments of dogs weighing less than 15 kg generally showed less severe degeneration of the ligament than those of larger dogs, and onset of the degenerative process was delayed by several years in the smaller dogs.³⁶³

Rupture of the contralateral cranial cruciate ligament occurred in 37% of dogs at a mean of 17 months following initial diagnosis of unilateral cranial cruciate ligament rupture; the remaining 63% had not ruptured their contralateral cranial cruciate ligament at a mean follow-up period of 31 months in one study.¹⁰⁶ In another study, 48% of Labrador Retrievers ruptured the contralateral cranial cruciate ligament at a median time of 5.5 months.⁶¹ Tibial plateau angle was not found to be a useful predictor of contralateral rupture in dogs with unilateral cranial cruciate ligament disease,⁶⁸ and in the Labrador Retriever, age, tibial plateau angle, sex, and weight did not appear to be risk factors for contralateral rupture.⁶¹

Historical findings include hindlimb lameness that is worse following exercise or periods of rest. Gait evaluation is consistent with hindlimb lameness referable to the stifle. The severity of lameness depends on the severity of ligament disruption. In cases of relatively stable partial tears, lameness may be subtle and noted only following periods of strenuous activity. In cases of complete rupture, lameness can be severe or non–weight bearing. In most cases of complete rupture, non-–weight-bearing lameness persists for several days; following this, moderate to severe weight-bearing lameness occurs. Stiffness after rest, particularly following periods of exercise, is often noted.

Physical Examination

Physical examination reveals stifle pain with flexion and extension, variable crepitus, and possible clicking associated with a meniscal tear (meniscal click). In cases of partial tear of the cranial cruciate ligament, a pain response is commonly elicited when the stifle joint is placed in full extension. In chronic cases, quadriceps muscle atrophy is notable, and periarticular fibrosis is most evident on the medial side of the stifle. This medial periarticular hypertrophy has been referred to as medial buttress formation. Stifle joint effusion is often noted by the absence of clearly defined, sharp borders on the medial and lateral aspects of the patellar ligament. Dogs will often sit with the affected leg projecting out to the side, rather than sitting squarely (fully flexing the stifle joint, with the hock, stifle, and hip in the same plane [such as a sitting Sphinx] and the tuber calcanei in close proximity to the ischiatic tuberosity); this has been termed an abnormal “sit test.”³²³ Disorders of the hock may also result in an abnormal sit test; thus careful physical examination is warranted to ensure that abnormalities of the hock do not exist or coexist.

The cranial drawer test is the mainstay of diagnosis of stifle instability; it is a test in which the operator creates craniocaudal tibial translation by applying a force to the tibia, while holding the femur stable. This test is usually performed with the operator positioned caudal or caudal and lateral to the patient. The thumb of one hand is placed behind the lateral fabella and the index finger is placed on the patella. The thumb of the other hand is placed behind the fibular head and the index finger is placed on the tibial tuberosity. It is important to ensure that the thumbs and fingers are securely positioned on these bony landmarks; failure to do so may allow motion of soft tissues to be misinterpreted as stifle joint instability. The tibia is firmly manipulated in a caudal and then cranial direction, and motion in the sagittal plane is monitored. In general, any motion in an adult animal is considered abnormal. In young dogs, a small degree of physiologic stifle joint instability may be present; this is termed puppy drawer. If present in a juvenile patient, it can be differentiated from pathologic instability by the sudden endpoint or stop after 1 to 3 millimeters of motion. In some adult animals with severe muscle wasting, a small amount of cranial drawer may be present, but an abrupt stop is typically evident if the cranial cruciate ligament is intact. It is important to assess for instability in a range of motion from nearly full extension to flexion. Because the collateral ligaments become taut in full extension, cranial drawer is diminished or absent in this position.

In cases of partial tear of the cranial cruciate ligament, cranial drawer may be present or absent.311,342 If the craniomedial band is torn and the caudolateral part is intact, cranial drawer is present in flexion only because the intact caudolateral part is taut in extension. If the caudolateral part is torn and the craniomedial band is intact, no cranial drawer is present because the craniomedial band is taut in both flexion and extension. Mild effusion of the stifle joint, pain when the stifle is held in extension, and discomfort noted during the cranial drawer test are consistent with partial cranial cruciate ligament rupture; these findings are useful in confirming that the cause of the lameness is related to the stifle joint, even in the absence of discernible cranial drawer. In cases of stable partial tears of the cranial cruciate ligament, alternative diagnostic methods such as radiography, magnetic resonance imaging (MRI,), and arthroscopy can be used to confirm the presence of stifle pathology and partial tear of the ligament.

The tibial compression test is another test of stifle joint stability in the sagittal plane; in this test, the operator creates stifle joint compression that results in a cranial tibial thrust force. The intact cranial cruciate ligament counters cranial tibial thrust; however, if the ligament is ruptured, cranial tibial translation occurs. To perform this test, the examiner is usually located caudal or caudal and lateral to the patient. The patient can be standing or in lateral recumbency with the affected limb uppermost. The index finger of one hand is placed on the tibial tuberosity, and the thumb, palm, and remaining fingers of that hand are used to grasp the femoral condyles and to maintain stifle joint extension. The index finger is used to press caudally on the tibial tuberosity, reducing the tibia into a neutral position if it was cranially subluxated. The other hand is used to grasp the metatarsus, and the tarsocrural joint is alternately flexed and extended, simulating contraction of the gastrocnemius muscle and thus the tibial compression mechanism. The tibial tuberosity is monitored with the index finger for cranial-to-caudal motion in the sagittal plane.¹⁵⁷ The presence of motion suggests compromise of the cranial cruciate ligament.