Chapter Outline
Incidence 666
Etiology 666
Clinical Features 666
Radiographic Findings 667
Pathophysiology 668
Biomechanics 669
Treatment 669
Congenital coxa vara is a developmental abnormality characterized by a primary cartilaginous defect in the femoral neck with an abnormal decrease in the femoral neck-shaft angle, shortening of the femoral neck, relative overgrowth of the greater trochanter, and shortening of the affected lower limb. Typically the deformity either is not present at birth or is sufficiently subtle not to be recognized at that time. Affected patients almost invariably present after walking age, and sometimes as late as adolescence, with a limp (Trendelenburg or short-leg gait) and, in unilateral cases, relatively mild limb length inequality.
This disorder has specific radiographic characteristics and a unique clinical presentation and should be distinguished from acquired causes of coxa vara deformity and coxa vara associated with congenital femoral deficiency. * In the literature, however, confusion and controversy exist regarding terminology and classification of this disorder. Coxa vara has been variously referred to as “congenital,” “developmental,” “cervical,” and “infantile.” † To add to the confusion, some authors, in discussing developmental or congenital coxa vara, do not distinguish between the disorder described in this chapter and congenital coxa vara with short femur ; others include acquired causes of coxa vara. Furthermore, some cases of coxa vara are associated with skeletal dysplasias, especially cleidocranial dysostosis, metaphyseal dysostosis, and some types of spondylometaphyseal dysplasia. ‡ Some authors exclude, and others include, this skeletal dysplasia–associated type of coxa vara in series describing developmental coxa vara. We believe that the term developmental coxa vara should be used to describe the clinical entity characterized by postnatal presentation of coxa vara without other known cause, with typically mild limb shortening and characteristic radiographic features; the deformity may or may not be associated with a generalized skeletal dysplasia. Acquired forms of coxa vara and those associated with significant femoral deficiency should be considered separate entities ( Box 19-1 ).
Acquired Coxa Vara
Slipped capital femoral epiphysis
Sequela of avascular necrosis of the femoral epiphysis
Legg-Calvé-Perthes disease
Traumatic coxa vara
Femoral neck fracture
Traumatic hip dislocation
Sequela of reduction for developmental dysplasia of the hip
Septic necrosis
Other causes of avascular necrosis of the immature femoral head
Coxa vara associated with pathologic bone disorders
Osteogenesis imperfecta
Fibrous dysplasia
Renal osteodystrophy
Osteopetrosis
Other bone-softening conditions affecting the femoral neck
Congenital Femoral Deficiency With Coxa Vara
Developmental Coxa Vara
Isolated (may be bilateral)
Associated with a skeletal dysplasia
Cleidocranial dysostosis
Metaphyseal dysostosis
Other skeletal dysplasias
* References .
† References .
‡ References .
Fiorani in 1881 was the first to publish a clinical description of a case of bending of the neck of the femur. The term coxa vara was coined by Hofmeister in 1894. The association of coxa vara with other malformations was noted by Kredel in 1896. Amstutz in 1970 described two patients with coxa vara who had previously had negative findings on radiographs of the hips. The term developmental coxa vara was first used by Hoffa in 1905 and later by Duncan. As mentioned previously, subsequent authors have used not only “developmental” but “congenital,” “infantile,” or “cervical” coxa vara to describe this condition.Incidence
Developmental coxa vara is rare; its incidence was estimated by Johanning to be 1 in 25,000 live births in the Scandinavian population. It has no racial predilection. The disorder appears to be equally common in boys and girls. Various series report the ratio of unilateral to bilateral cases to be between 1 : 2 and 3 : 1. Bilateral cases may be more likely to be associated with generalized skeletal dysplasia, so the examiner should seek further evidence of such dysplasia during the physical and radiographic examination of patients with bilateral coxa vara.
Etiology
The disorder is frequently present in skeletal dysplasias with known genetic causes, such as the following: cleidocranial dysostosis; metaphyseal dysostosis, Jansen type; and spondylometaphyseal dysplasia, especially Kozlowski type. All these skeletal dysplasias are autosomal dominant disorders. There is also a presumed genetic cause in “isolated” developmental coxa vara (not associated with a generalized skeletal dysplasia). Reports have noted the condition in families and in both homozygous and heterozygous twins.
Clinical Features
The deformity does not manifest until after birth and usually not until walking age. Clinically, the child presents with a painless limp resulting from a combination of true Trendelenburg gait and relatively minor limb length inequality in unilateral cases. Easy fatigability or aching pain around the gluteal muscles may be a complaint. In patients with bilateral involvement, the complaint is usually of a waddling gait, similar to that seen in bilateral developmental dysplasia of the hips, with or without fatigue or muscular pain.
On physical examination, abduction and internal rotation of the affected hip are limited. With increasing coxa vara, the tip of the greater trochanter translates proximally relative to the center of the femoral head, and the origin and insertion of the hip abductors approach each other. The Trendelenburg test result is positive. In contradistinction to developmental dysplasia of the hip, the patient has no telescoping of the hip or other signs of instability, such as the Ortolani sign. Shortening is present in unilateral cases but seldom exceeds 3 cm at skeletal maturity, even in untreated patients.
Evidence of generalized skeletal dysplasia should be sought, especially if the family history is positive for similar deformity or short stature, the affected patient is of short stature, or involvement is bilateral. None of these features specifically implies the presence of an identifiable skeletal dysplasia, however, because family history, short stature, and bilateral involvement can all be present in patients with presumed isolated developmental coxa vara. Some authors in their reviews of this condition have specifically excluded patients with coxa vara associated with skeletal dysplasias ; we do not make this distinction when the clinical and radiographic features of the coxa vara deformity are otherwise identical to those of isolated developmental coxa vara deformity.
Radiographic Findings
Plain anteroposterior (AP) radiographs demonstrate a decreased neck-shaft angle of the affected hip, a widened radiolucent line corresponding to the proximal femoral physis, and characteristically, but not universally, a triangular piece of bone in the medial femoral neck abutting the physis and bounded by two radiolucent bands traversing the neck and forming an inverted V ( Figs. 19-1 and 19-2 ). The superior, more horizontal radiolucent line is the capital femoral physis; the inferior, more vertical line is an abnormal area of faulty maturation of cartilage and irregular ossification. Acetabular dysplasia is often seen with coxa vara, and the more severe the varus of the femoral neck, the greater is the slope of the acetabulum and sourcil.
If generalized skeletal dysplasia is suspected on the basis of the family history or short stature in the affected individual, a skeletal survey will reveal features characteristic of those conditions. These features include deficiencies of the clavicle and medial portions of the pubic rami in cleidocranial dysostosis, as well as generalized physeal widening and angular deformity in metaphyseal dysostosis. The radiographic differential diagnosis should also include coxa vara produced by avascular necrosis (from trauma or infection, or associated with developmental dysplasia of the hip or Legg-Calvé-Perthes disease), and pathologic bone-associated conditions such as osteogenesis imperfecta, fibrous dysplasia, osteopetrosis, or renal osteodystrophy.
The amount of varus deformity of an affected hip may be quantified on AP radiographs by measuring the neck-shaft angle, the head-shaft angle, or the Hilgenreiner-epiphyseal angle (H-E angle), as described by Weinstein and colleagues ( Fig. 19-3 ). Because the pathophysiology and rationale for operative treatment imply sagging or slippage of the femoral epiphysis (and the triangular neck fragment characteristic of this condition), some authors have found that traditional measurement of the neck-shaft angle does not adequately reflect the amount of deformity, anticipated surgical correction, or prognosis for spontaneous progression or postoperative recurrence in this disorder. Measurement of the angle between the axis of the femoral shaft and a line perpendicular to the base of the femoral epiphysis, the head-shaft angle, more accurately reflects the severity of the deformity and its likely progression or correction. Weinstein and co-workers identified the prognostic value of the H-E angle, the angle between the Hilgenreiner line and a line drawn parallel to the physis of the proximal femur. On an AP radiograph of the pelvis with the hips in neutral position, this angle is usually between 0 and 25 degrees (average of 16 degrees, in Weinstein and colleagues’ series of 100 normal hips). Weinstein and associates, in a study of 22 patients with coxa vara, found that in hips with an H-E angle greater than 60 degrees the deformity invariably progressed and merited surgical correction; hips with H-E angles less than 45 degrees remained stable or improved and thus could be treated expectantly, and those with angles between 45 and 59 degrees had an indeterminate prognosis and had to be monitored for progression by serial radiographs. The value of this measurement with respect to prognosis for recurrence of deformity was confirmed in other series of surgically treated patients.
Pathophysiology
The precise cause of developmental coxa vara is unknown, but the condition probably results from a primary defect in enchondral ossification of the medial part of the femoral neck. Early in fetal development, the proximal femoral physis extends across the upper end of the femur as a crescentic line of cartilage columns that soon differentiate into cervical epiphyseal and trochanteric apophyseal portions. The medial cervical portion matures early, thus elongating the femoral neck, and the ossification center of the capital femoral epiphysis appears within the first 3 to 6 months of postnatal life. The lateral portion of the crescentic physis matures into the greater trochanteric apophysis, and the trochanteric secondary center of ossification begins to ossify at 4 years of age. The neck-shaft angle and the length of the upper end of the femur are determined by the relative amount of growth at these two sites. According to Von Lanz and Wachsmuth, the mean angle of the femoral neck and shaft is 148 degrees at 1 year of age, and it gradually decreases to 120 degrees in the adult ( Fig. 19-4 ).
Anatomic descriptions of coxa vara were first published by Hoffa in 1905, by Helbing in 1906, and by Schwarz in 1913. Later reports were published by Barr, Camitz, Zimmerman, and Burckhardt. In investigations of fetal femoral head specimens, large amounts of fibrous tissue rather than cancellous bone were found in the medial part of the metaphysis of the femoral neck. Thus the mechanically weak femoral neck could be passively deformed into a varus angulation under the stress of muscle forces and body weight, and the capital physis could migrate inferiorly through this weakened portion of the femoral neck. Creation of support of this weakened portion of femoral neck is part of the rationale for the Y valgus-displacement osteotomy described by Pauwels ( Plate 19-1 ). In a description of four cases, Blockey suggested that the cause of the condition was antecedent unrecognized or unreported trauma, but no other published investigation appears to support this premise.
Biopsy specimens of the capital femoral physis in patients have generally revealed disorganized islands of cartilage cells in relatively reduced numbers; the normal alignment into columns typical of a healthy physis is absent, and no calcification is evident in the physis. Magnetic resonance imaging in two patients revealed no evidence of true slippage of the capital epiphysis on the femoral neck.
Chung and Riser described the postmortem gross, microscopic, and vascular findings of the hip in a 5-year-old child who died 2 years after valgus osteotomy for developmental coxa vara. These investigators noted that the acetabular volume and femoral head were smaller, the femoral neck was shorter, and the physis was wider on the affected side than on the normal contralateral side. Both the number of blood vessels on the metaphyseal side of the physis and the number of medial ascending cervical arteries were decreased. The bony trabecular network providing support for the medial neck was absent in the epiphysis and metaphysis of both hips in this report, a difference the investigators found striking compared with specimens of normal hips from children of similar age.
Biomechanics
According to Pauwels, in the normal hip the compressive force (R) is perpendicular to the center of the hip joint ( Fig. 19-5, A ). As a result the physeal cartilage and hyaline cartilage of the acetabulum are under compressive force (D), which is evenly distributed throughout. Normally, the physis is perpendicular to the resultant compressive force R ( Fig. 19-5, B ). Stresses on the medial side of the femoral neck are compressive (D), whereas those on the lateral side are tensile (Z). In Figure 19-5, A , S represents the shearing force.
In coxa vara, with a progressive decrease in the femoral neck-shaft angle, the physis changes its position from horizontal to vertical and thereby becomes increasingly inclined relative to force R ( Fig. 19-5, C to E ). The shearing force (S) across the physis gradually increases. The upper femoral epiphysis tends to tilt and become displaced medially, and the tensile stresses (Z) increase. Growth of the femoral neck is less on the medial side than on the lateral side.
The femoral neck-shaft angle affects the direction, position, and magnitude of loading and stressing of the proximal end of the femur ( Fig. 19-6 ). In coxa vara the femoral neck-shaft angle is decreased; consequently the tip of the greater trochanter is elevated, and the position and direction of muscular force (M) are altered. The point of intersection (X) of muscular force M with the line of action of the partial body weight (K) is lowered. The resultant compressive force R (which connects point X with the center of the femoral head) diverges more than normal in coxa vara, and the lever arm (h) of the abductor muscles is lengthened.
The length of the femoral neck also affects the magnitude of its mechanical stressing. The length of the lever arm (h) of the muscular force (M) is diminished by shortening of the femoral neck ( Fig. 19-7 ). In response to efforts to preserve equilibrium, the muscle forces and resultant compressive forces (R) increase. Therefore shortening of the femoral neck increases bending stress.
