Legg-Calve-Perthes Disease

History

In 1910, Legg, Calve, and Perthes independently described a condition of the hip in children [1-3]. Calve thought that the condition was due to rickets [2]. Perthes thought that it was related to degenerative arthritis, probably of an infectious nature [3]. Legg, however, hypothesized that impairment of blood supply to the femoral epiphysis was the cause of the condition; a hypothesis that parallels our current understanding of the pathogenesis [1]. Of historical interest, Waldenstrom described a case of tuberculosis of the hip in 1909 that may have actually been a case of Legg-Calve-Perthes (LCP) disease [4]. In 1935, LCP disease was first described in the veterinary literature by Tutt [5]. Later, Spicer (1936), Schnelle (1937), and Moltzen-Nielsen (1938) described the condition, using some form of a synonym of LCP disease, in the veterinary literature [6]. Common synonyms for this condition are avascular necrosis of the femoral neck, aseptic necrosis, osteonecrosis, coxa plana, osteochondritis deformans juvenilis, and osteochondrosis.

Pathogenesis

The main histologic feature of LCP disease in both dogs and humans has been described as ischemic necrosis of the center of ossification of the femoral head [7]. Ponseti et al., described histological, histochemical, and ultrastructural observations of biopsy specimens from the lateral aspect of the femoral head and neck of children with LCP disease [7]. They reported that beneath normal cartilage was thickened epiphyseal cartilage that contained areas of hypercellular and fibrillated cartilage with prominent blood vessels [7]. Ultrastructural examination of these areas revealed irregularly oriented collagen fibers and variable amounts of proteoglycan granules. Hypercellular areas suffered from a decrease in proteoglycans, glycoproteins, and collagen compared with that of normal epiphyseal cartilage. These findings suggest that the disease could be a localized expression of a generalized transient disorder of epiphyseal cartilage that is responsible for delayed skeletal maturation. The collapse of the femoral head likely results from a mechanical failure of this necrotic area that heals as a disorganized epiphyseal cartilage matrix with abnormal ossification. The severe deformity of the femoral head is a consequence of collapse of this mechanically inferior repairing cartilage. What remains unclear is whether the abnormalities of the epiphyseal cartilage are primary or secondary to ischemic events [8].

In the dog, three histologically distinct stages of the disease have been described: ischemic, early repair, and advanced repair [9].

The ischemic (necrosis) stage is characterized by empty osteocyte lacunae and the absence of viable marrow. The chondrocytes of the articular and physeal cartilages are histologically and ultrastructurally normal during this stage. The metaphyseal trabeculae may be thickened, but the process of endochondral ossification is generally uninterrupted.

In the early repair stage, the articular cartilage develops clefts and fissures as the subchondral bone begins to collapse beneath it, and the overall shape of the femoral head appears flattened (coxa plana). The repair process begins with revascularization at the periphery of the epiphysis. Fibrovascular tissue composed of capillaries, macrophages, fibroblasts, and histiocytes advances toward the center, resorbing the necrotic marrow debris and dead trabeculae. The articular cartilage appears thickened, especially in the zone of calcification, and the physeal cartilage becomes invaded with fibrovascular repair tissue.

In the advanced repair stage, the articular cartilage becomes markedly thickened with clefts and infolding, and the entire femoral head appears enlarged, with eventual loss of its normal spherical shape. Areas of osteoclastic absorption and extensive new bone formation can be seen histologically at this stage. The disease is considered irreversible at this point because the collapse and repair permanently change the contour of the femoral head. Ultimately, progressive osteoarthrosis develops. The etiology for the ischemia, subsequent fragmentation, and protracted reformation of the femoral head observed in this disorder remain unknown [7].

Etiology

Many theories as to the etiology of LCP disease have been proposed and disproved since the disease was first described in the dog in 1935. Because the disease affects primarily small and toy breeds, heritability or anatomic variations in small breeds would seem to be factors contributing to the development of the clinical disease. Indeed, Vasseur et al. [9] demonstrated that LCP disease was a heritable condition in the Manchester terrier. To understand the exact pattern of inheritance for LCP disease, additional investigation of affected pedigrees must be performed. Intuitively, this condition is likely heritable in similar dog breeds. Supporting information is found in a study comparing the vascular anatomy of the hip in a miniature dog with that in a normal-sized mongrel [10]. This study demonstrated a distinct difference between the two groups in the channel of the superior retinacular vessels. In the miniature dog, the vessels coursed through a shallow neck and appeared as a suspended bridge as compared with a deep fossa of the femoral neck in the mongrel dog [10]. The consequence of the difference in vascular supply may be that the femoral heads of miniature dogs are more susceptible to vascular compromise or insults from trauma, synovitis, or vascular abnormalities.

A more recent suggestion is that LCP disease arises after ischemic infarction, venous or arterial, of the capital femoral epiphysis. Vascular compromise leading to LCP disease has been experimentally demonstrated in animal models by several authors [11-13]. Coagulation disorders as a source of infarction were examined in both humans and dogs. Evidence of thrombocytosis and hypofibrinolysis secondary to a protein C and S deficiency was reported in 1997 by Glueck et al. [14]. However, other investigators have failed to reproduce their findings [13]. The largest study that evaluated coagulation abnormalities involved a population of 139 children in Northern Ireland with LCP disease. Of the children afflicted with the condition, 38% had a prolonged activated partial thromboplastin time compared with 5.9% in a control group of 220 children [13]. However, no significant differences existed in antithrombotic factors protein C, protein S, or antithrombin (AT) III, or resistance to activated protein C. An association between the prolonged partial thromboplastin time and a clotting factor deficiency was not demonstrated [13]. Likewise, Brenig et al., were unable to demonstrate any alterations in protein C, protein S, activated protein C, factor II, factor V, factor VIII:C, or AT III activities in plasma samples of 18 dogs with histopathologically confirmed LCP disease [15]. The cause for prolonged activated partial thromboplastin time in some of these patients remains unknown. Genomic DNA from 15 dogs affected with LCP disease has also been evaluated for mutations in the protein C gene, however no mutations were found [16].

Diagnosis

Patients generally present in the first year of life with the owner complaining that the dog has a limp. High-risk breeds include terriers, toy poodles, Chihuahuas, Lhasa Apsos, miniature Pinschers, Pugs, and other toy breed dogs [17]. Although LCP disease in people affects males more often than females, no sex predilection has been identified in the dog. Depending on the stage of the disease, the severity of lameness can range from subtle to non-weight bearing. Clinical signs can also vary if the disease is bilateral, which reportedly occurs in 12 to 16% of cases [18,19]. Physical examination findings generally include lameness, muscle atrophy, and pain on extension and abduction of the affected hip joint(s). Given the breeds affected with LCP disease, medial patellar luxation is also a common incidental finding.

Confirmation of examination findings can be performed using radiography. Radiographic findings depend on disease stage (Fig. 103-1). Early findings can include evidence of an increase in radiopacity as new bone is laid down on empty lacunae. Progression of the disease includes resorption of necrotic bone. Loss of bone is faster than the production of new bone in this pathologic process and radiographic evidence of osteolysis will be present as the disease progresses. As the femoral head begins to collapse, it will lose its spherical shape and an increase in the joint space may be present. Following collapse, osteoarthritis will rapidly form and osteophytes will be present on radiographs. If both LCP disease and a medial patellar luxation are present and one needs to identify which is most likely causing the clinical signs, nuclear scintigraphy can be performed. Empirically, LCP disease is almost always a bigger contributor to clinical lameness than is a medial patellar luxation. If the etiology of the radiographic findings is in question, arthrocentesis with cytology and a culture should be performed. However, given the infrequency of idiopathic septic arthritis in the dog, this is rarely needed.

Figure 103-1. Hip-extended radiographs of a 10-month-old Cairn terrier (left) with evidence of sclerosis in the femoral neck and collapse of the femoral head. Radiographic progression of the disease is evident when the dog is 14 months old (right) with complete collapse of the femoral head and severe osteoarthritis.

Treatment

Nonsurgical management consists of strict exercise restriction. In an effort to enforce exercise restriction of unilateral LCP disease a non-weight-bearing sling can be applied. The authors would suggest the use of a Robinson sling, however, the use of an Ehmer sling has been reported to be successful in at least one case report [20]. The duration of exercise restriction varies, but it has been reported that if nonsurgical management is going to be successful it generally takes longer than 2 months [18]. Monthly radiographs should be taken to follow the progression of the disease, and immobilization of the limb continued until complete resolution of the radiolucent areas. In a retrospective report of dogs diagnosed with LCP disease that were treated nonsurgically, only 25% had resolution of their lameness [18]. It is important to note that, even though the prognosis with nonsurgical management is guarded, it remains a reasonable first option (when the radiographic progression of the disease shows no loss of the spherical nature of the femoral head) because surgical management is via a salvage procedure.

A patient that presents with collapse of the femoral head and incongruency of the coxofemoral joint should be treated surgically by femoral head and neck excision. This salvage procedure can be performed simply, is relatively [18] inexpensive, and provides for an improved prognosis. Junggren reported that 30 of 36 dogs that had LCP disease treated with femoral head and neck excision had complete resolution of their clinical signs. In addition, the vast majority of these dogs (80%) had fully recovered within 2 months of surgery. Following femoral head and neck excision, the patient should have exercise restriction until suture removal (7-14 days). After this period, the dog can return to its regular activities. Postoperative rehabilitation including cold and hot packing of the area, passive range of motion, and swimming may further improve the number of dogs that respond favorably to treatment. Several commercially available total hip replacement systems now offer miniature implants. This is no longer true. Patients that have bilateral LCP disease can be treated with bilateral, simultaneous femoral head and neck excision. A ventral approach to the hip for bilateral surgery may reduce morbidity if the surgeon is familiar with the technique.