Clinical and genetic characteristics of Legg-Calve-Perthes disease

INTRODUCTION

Legg-Calve-Perthes disease (LCPD) or coxa plana is an idiopathic avascular necrosis (AVN) of the femoral head epiphysis that occurs during childhood. It was first described independently in 1910 by Georg Perthes’, Arthur Legg, and Jacques Calvé.^[1] LCPD is characterized by hip pain, restricted mobility, reduced physical activity, and disability typically leading to gait disturbance.^[2]

It usually occurs in 3–12 years old children. The highest occurrence is at 5–7 years.^[3] This condition is 4 times more common in boys.^[4] LCPD is bilateral in 10–20% of affected cases. The incidence of LCPD varies in different geographical regions. Overall, analysis of data from 16 countries has shown that the incidence of the disease ranges from 0.2 per 100,000 to 19.1 per 100,000.^[2,5] A single study has reported a highest incidence of 29.0 per 100,000 individuals in the Faroe Islands of Denmark.^[6] It has also been observed that children from families of low socioeconomic status may have high incidence and are, therefore, disproportionately affected.^[2,7]

Key events in the development of LCPD can be divided into four distinct phases, including (a) subchondral fracture, (b) fragmentation, (c) re-ossification, and (d) healing with residual deformity. The first stage of this condition is characterized by a temporary disruption of the femoral capital epiphysis blood supply leading to infarction of subchondral cortical bone and capital epiphysis. As a result, growth of the ossific nucleus ceases and the infracted bone dies. This stage is termed necrosis.^[8] Over time fragmentation takes place. In this process, the dead bone is reabsorbed, and new bone is formed.^[9] In the re-ossification phase, the femoral head eventually heals, and the femoral epiphysis re-establishes due to the osteoblastic activity. However, during this process, the re-ossified femoral head may get enlarged or flattened, and deformity can develop due to reshaping because of weight bearing pressure on the weakened epiphysis during growth. This stage is called remodeling. Remodeling typically leads to disturbance in gait, mobility restriction, pain, and reduction in physical activity.^[3,10,11]

The aim of this review article is to discuss the etiological factors underlying LCPD with special emphasis on the role of coagulation factors and mutations in the genes encoding those coagulation factors.

ETIOLOGY OF LCPD

The exact cause of LCPD is not known. A large nationwide Swedish study found an association between suboptimal birth characteristics and breech presentation during delivery with the development of LCPD.^[12] A strong correlation has also been described with maternal smoking as well as exposure to second-hand smoke.^[13-15] Moreover, high rates of obesity and hypertension were also identified in a group of children having LCPD.^[16]

The cause of the LCPD is largely unknown, though, many experimental and clinical studies provide support to the idea that the temporary disruption of the blood supply to the femoral head is a key event in the pathogenesis of the disease. Various diagnostic tools including selective angiography,^[17] bone scintigraphy,^[18] perfusion magnetic resonance imaging (MRI),^[19] and the biopsy studies^[20] from the early stages of the disease show clear indication of disruption of perfusion and bone damage consistent with AVN. The underlying cause may be disruption of supply of blood to the femoral epiphysis due to trauma, coagulopathy, or the use of steroids. Disruption of blood supply might be due to either thrombophilia (an increased tendency for thrombus formation) or hypofibrinolysis (a reduced ability for thrombolysis). The two processes have been suggested to play an essential role in the pathogenesis of osteonecrosis.^[21-24] This is supported by the fact that thrombophilia and some forms of coagulopathy are present in approximately 50% and 75% of the LCPD patients, respectively.^[25]

Some authors refer to the injury and initial bone collapse as the cause of LCPD. Others have proposed intravascular thrombosis as a causative mechanism.^[26] Recently, congenital growth hormone deficiency has also been associated with LCPD.^[27]

PATHOBIOLOGY OF THE LCPD

The difficulty in getting clinical specimens from patients with LCPD is the key issue in understanding the pathophysiology of this condition. The current knowledge of the LCP disease process is based on the review of a histopathological findings of six femoral heads, a few reports of isolated necropsy, and findings from some surgical biopsies.^[20,28-32] Briefly, the pathological process affects mainly the bony epiphysis, physis, and metaphysis with the covering articular cartilage. The articular cartilage changes include increased thickness in its deep layer with necrosis of the chondrocytes, cessation of the endochondral bone formation, a clear separation of articular cartilage from the subchondral bone, invasion of the cartilage by blood vessels, and new accessory ossification. In the epiphyseal bone, the necrosis of the marrow space and the trabecular bone, compression fracture of trabeculae, invasion of the fibro-vascular granulation tissue, necrotic bone resorption, and thickened trabeculae due to new bone formation have been reported. The physeal changes can often be seen in the anterior part of the femoral head, with focal areas of growth cartilage extending into the metaphysis. Premature growth arrest of the growth plate is seen only in 30% of LCPD patients, suggesting that the growth plate continues to function in the majority of the patients. Changes in metaphysis are usually observed during the initial stages of the disease.^[32]

DIAGNOSIS OF LCPD

Imaging (radiographs and MRI) is used to diagnose LCPD. Recently, non-invasive magnetic resonance angiography has been shown to provide clear images of the epiphyseal blood supply of the femoral head in LCPD patients.^[33] However, early radiographs can be normal. Imaging studies of the initial stages of the disease show epiphyseal cartilage hypertrophy, changes in the epiphysis, and subchondral fracture. In advanced stages of LCPD, flattening of the femoral head, fragmentation, and healing are evident in radiographs. In addition, MRI shows decreased femoral head perfusion and bone marrow changes.^[3]

MANAGEMENT OF LCPD

The main goal of the management for LCPD is to manage pain and symptoms.^[34] Moreover, the aim is to promote self-healing of the femoral head with minimal deformity and restoration of the hip range of motion.^[34] This is achieved by maintaining the optimum local environment in and around the hip joint.^[35,36] Repair and resorption take place, leading to remodeling and femoral head deformation.^[9] Once children start to recover with the femoral head remodeling, symptomatic improvement may be observed in the hip joints.

Non-surgical, as well as surgical intervention, is required to manage the LCPD. Non-surgical management is preferred over surgical options. This type of management typically suffices in children with bone age below 6 years.^[37] Almost 90% of physicians refer LCPD children for physical therapy^[38] with the aim to maintain hip joint mobility. The evidence base for this type of management is limited.^[39] Non-surgical management options include orthosis, physical therapy such as stretching and strengthening regimes, walking aids, activity modification, or watchful waiting.^[40,41] The selection of the type of non-surgical interventions is totally dependent on the treating physician. However, robust evidence regarding which kind of intervention is most effective for treating children with LCPD is lacking.^[37,39] In a few cases, the hip deformity is severe and surgical containment may be necessary. Operative treatment is performed based on age, severity, and type of deformity. It can be femoral or pelvic osteotomies, varus or valgus femoral osteotomies, shelf procedure, hip arthroscopy, or hip arthrodiastasis.^[42] In many cases, surgical intervention is usually considered after skeletal maturity in late adolescence.^[43] Clinical guidelines for the management of LCPD are absent and currently, no standardized approach is there for selecting the right treatment.

GENETICS OF LCPD

LCPD is a complex disease, and the interaction of multiple environmental and genetic factors may play a role in the development of the disease. LCPD-associated factors include deprivation, the abnormal clotting mechanism, smoke exposure, and genetic predisposition.^[4,44,45] Although most LCPD cases are isolated, multiple cases in families have also been reported, with some families having more than two affected individuals.^[46-48] Several reports of siblings with similar features of disease have also been published.^[49,50] Moreover, a study has shown the incidence of 2.5% in first-degree relatives of LCPD index cases.^[51] This incidence of LCPD in relatives is much higher (35 times) than that in the general population. These studies raise the possibility of a strong genetic component in the etiology of LCPD. Furthermore, the occurrence of LCPD-like hip changes in a few inherited dysplasias of the skeleton, such as trichorhinophalangeal syndrome^[52-54] and Floating-Harbor syndrome,^[55] also suggests the presence of major genetic defects underlying LCPD. In the succeeding section, genetic factors associated with the pathogenesis of LCPD will be discussed.

COAGULATION FACTORS AND LCPD

AVN develops, at least in part, as the end result of partially and temporarily disrupted blood flow to the femoral head.^{[22-24,45,56,57]} Defects in coagulation factors have been considered as one of the possible causes of the impaired supply of blood to the femoral epiphysis.^[58] The thrombophilia-based hypothesis (thrombophilia followed by thrombotic venous occlusion of the femoral head) of LCPD is based on the notion that thrombosis selectively blocks the femoral head venous outflow, leading to increased intraosseous pressure, and subsequent AVN. Several studies have investigated coagulation abnormalities in LCPD patients. In a case–control study on the correlation between LCPD and coagulation defects, 19 patients out of 44 patients were found to have a protein-C deficiency and four patients out of 44 patients were found to have a protein-S deficiency.^[56] Three fourths of the study patients had coagulation abnormalities. A high risk of LCPD is statistically significant with decreasing levels of protein C and a high risk of LCPD is nearly significant with decreasing levels of protein S was found.^[22] Therefore, it was concluded that the thrombophilic state due to abnormalities of the coagulation system plays a role in the LCPD.^[22]

While some studies have found an increased rate of abnormalities of coagulation in the patients with LCPD,^{[21,22,56,59,60]} others have not found any association at all.^[61-64] The discrepancy might be due to some confounding factors, including a small sample size in some studies, the retrospective study design, use of suboptimal controls, and non-standardized range of laboratory values for coagulation factor level. As far as prospective studies are concerned, a random series of 50 consecutive LCPD patients did not show a difference in the prevalence of antithrombin III, protein-S, or protein-C deficiencies between the group of LCPD patients and the estimated frequency of disease in the population.^[65] Recently a Cincinnati case–control study with 72 non-selected, consecutive LCPD patients and 197 healthy controls failed to find an increase in the prevalence of antithrombin-III, protein-S, and protein-C deficiencies.^[66] Nevertheless, the Cincinnati study did find an increased prevalence of factor-V Leiden and anticardiolipin antibodies (IgG and IgM) in LCPD patients.

Mutations in genes involved in coagulation, including factor V Leiden (F5; OMIM 612309), prothrombin II (F2; OMIM 176930), and methylenetetrahydrofolate reductase (MTHFR; OMIM 607093) have the tendency to cause vascular occlusion. Below, studies describing mutations in F5, F2, and COL2A1 (OMIM 120140) genes and their association with the LCPD will be discussed.

CONCLUSION

LCPD is a rare hip condition of unknown etiology and is considered to be caused by temporary cessation of the blood flow to the femoral head resulting in venous occlusion and necrosis of the femoral head. The susceptibility to developing LCPD is determined by various environmental and genetic factors, while the nature and level of interaction between them are still not clear. Occurrence of multiple cases of LCPD in families and variability in the phenotype is well documented. Variations in genes that code for coagulation factors (F5 and F3) and structural components of bones (COL2A1) might have a role in the etiology of LCPD but, so far, no strong candidate genes underlying increased risk have been identified. In conclusion, there is still much to learn about the etiology and the pathogenesis of LCPD. The overall pathogenesis of the deformity of the femoral head is complex, with multiple underlying factors. Therefore, an effective treatment should take into account the mechanical as well as biological factors involved in the pathogenesis of LCPD.

AUTHORS’ CONTRIBUTIONS

Both authors have contributed to the conception and design of the work, critically reviewed and approved the final draft and are responsible for the manuscript’s content and similarity index.

ETHICAL APPROVAL

The authors confirm that this review had been prepared in accordance with COPE roles and regulations. Given the nature of the review, the IRB review was not required.