Congenital Dislocation of the Knee in the Delivery Room

Introduction

The estimated incidence of the congenital dislocation of the knee (CDK), also known as genu recurvatum, is reported in 1 per 100 000 live births, which is approximately 1% the incidence of developmental dysplasia of the hip [1]. It is twice as frequent in females than males and can occur as an isolated condition, but more often in association with other intrinsic/genetic abnormalities such as Larsen syndrome, arthrogryposis multiplex congenita, and Ehler’s-Danlos syndrome [2,3]. In addition, CDK can be associated with developmental dysplasia of the hip and talipes equinovarus [4–6].

Case Report

A 23-year-old woman (gravida 4 para 2 abortus 2 living 1) gave birth by normal vaginal delivery to a 3100-g male neonate at 39 weeks of gestation. The pregnancy was uncomplicated. The prenatal ultrasound was normal and did not detect any breech presentation or oligohydramnios. Apgar scores were 8 and 9 at 1 and 5 minutes, respectively. While examining the newborn, there was an apparent extension deformity of the left knee (reported as “backwards”) with approximately 45° of hyperextension (Figure 1). The left knee could passively flex to approximately 30°. The infant did not exhibit any signs of distress or pain during gentle manipulation of the left knee. The skin was warm and well-perfused with brisk capillary refill. Left knee anterior skin folds were present and there was no sign of talipes equinovarus (clubfoot). Barlow and Ortolani maneuvers do not demonstrate a “clunk.” The right extremity and the remainder of the physical examination were otherwise unremarkable.

A lower-extremity radiograph demonstrated subluxation/dislocation of the left knee with anterior translation of the tibia/fibula relative to the femur, and an abnormal anterior angulation at the knee with no fracture (Figure 2). Normal newborn care was ordered, and an orthopedic surgery consult was placed. Reduction of the knee was not attempted at initial evaluation. The neonate was referred to the pediatric orthopedic service in another institution, where the knee was non-surgically reduced, and serial casting was performed for 3 months. Upon conclusion of the treatment, the baby was examined by us and an excellent clinical outcome was detected with almost complete normal flexion and extension of the knee without any deformity. Final radiographs were not obtained. A hip ultra-sound did not detect any developmental dysplasia.

Discussion

Mechanical anomalies that can lead to CDK include fetal mal-position such as entrapment of the leg under the mandible, oligohydramnios, prolonged breech position, primary contracture of the quadriceps tendon due to fibrosis, and anterior cruciate ligament malformation [7]. An intrinsic or genetic anomaly often associated with CDK is Larsen syndrome, which has a frequency of approximately 1 per 100 000 live births and is due to mutations within filamin B. Associated symptoms include facial dysmorphism, talipes equinovarus, joint hyper-elasticity, and multiple dislocations [8,9]. Other genetic syndromes associated with CDK include Desbuquois dysplasia (osteochondrodysplasia that presents with short stature/extremities, severe joint laxity with dislocation, osteopenia, kyphoscoliosis, and characteristic facies), Collins Pope syndrome (bilateral congenital dislocation of the hip, congenital heart defects, characteristic facies), Marfan syndrome, Ehler’s-Danlos syndrome, and Down syndrome (muscular hypotonia and ligamentous hyper-laxity) [2,10]. CDK is associated with congenital hip dislocations, but may also be seen with arthrogryposis, myelomeningocele, talipes equinovarus, cleft palate, and scoliosis [1,2]. A hip ultrasound is recommended if CDK is present. Of all CDK cases, hip dislocation is involved in 45%, talipes equinovarus and other feet deformities in 3%, and elbow dislocation in 10% [6]. A study of 51 infants with varying degrees of CDK found isolated hip instability in 11 limbs and isolated foot abnormality in 7 limbs [2]. The findings from our case demonstrated no associated facial dysmorphisms, other joint dislocations, or syndromes evident on the initial physical exam.

Incidental prenatal diagnosis of CDK by abdominal radiograph performed for another reason has been reported [11]. Prenatal diagnosis of CDK is challenging, as it is often undetected by ultrasound. In a recent review of 20 cases of CDK detected by prenatal ultrasound, the median gestational age at diagnosis was 20 weeks (range, 14–38 weeks) [12]. A 3-dimensional (3-D) with a real-time 3-D (4-dimensional) ultrasonographic evaluation seems advantageous, as it offers a more dynamic interpretation of the limb position and mobility [13]. During initial evaluation after birth, hyperextension of the knee is an obvious deformity to assess for, while evaluating anterior skin grooves is important for prognosis. Absence of anterior skin grooves typically demonstrates a long-standing dislocation of the knee and severe deformity, whereas a greater number skin folds correlates with a more recent in utero dislocation and more favorable prognosis [14,15]. In our case, the patient presented with anterior skin grooves.

Currently there is no consensus on the classification of CDK, but specific criteria are typically used to classify CDK [2,7,8]. Mehrafshan et al proposed a classification system based on reduction and stability of the knee [2]. Type I CDK is easily reducible, providing a reduction snap as the femoral condyles pass in flexion, while remaining stable in flexion. Type II provides a recalcitrant dislocation, reducible but unstable. If posterior anterior pressure is relaxed on the femoral condyles, iterative dislocation occurs. Lastly, type III is an irreducible dislocation. Other classifications such as that of Abdelaziz et al [7] classify CDK based on knee flexion range of motion (ROM) and femoro-tibial relationship through radiography. Grade I CDK is when ROM is >90° with simple recurvatum on radiography, while grade II demonstrates ROM between 30° and 90° with subluxation/dislocation evident on imaging. Grade III has <30° with dislocation on radiography. Among these 2 classification systems, there appears to be some overlap. Mehrafshan et al [2] evaluated ROM and found significant differences in flexion among type I, II, and III. Individuals classified under type I at initial evaluation demonstrated knee flexion ROM at 85° (10°–140°), type II at 34° (10°–110°), and type III CDK at −5° (−35°−18°). Our patient was likely grade II according to the Abdelaziz et al [7] classification since knee flexion at initial examination was approximately 30° and radiography showed subluxation/dislocation.

Common pathologic findings are a shortening of the quadriceps femoris tendon, a tight anterior articular capsule, and hypoplasia of the suprapatellar bursa [16]. Most authors recommend initiation of intervention as soon as possible to prevent contractures and to promote development of the anterior distal femoral epiphysis [16,17]. Among treatment options, the literature reports that conservative treatment with serial casting/bracing or surgery as potential options, depending on the classification. Rampal et al [8] suggest that conservative treatment for type I and II include therapy with quadriceps stretching and application of an anterior full-leg splint that will progressively change to increase knee flexion and maintain reduction over time. They achieved successful reduction of all type I cases and 11 out of 16 type II cases with this non-operative treatment. The remaining type II knees underwent progressive cutaneous limb traction until reduction, and 90° flexion of the knee was demonstrated on lateral X-ray. Then, cast immobilization was applied for 3 weeks, and an anterior splint for 4–6 weeks yielded stability. Limb traction and casts were used for type III as well, but less favorable results were obtained. Only 43% of type III achieved reduction and stability with traction and casts and the remaining knees required surgery. Surgical options commonly included percutaneous quadriceps recession and V or Y quadreps-plasty [7,8,18,19]. Complications of treatment include recurrence, greenstick fracture of the tibia during manipulation and casting, and development of valgus knee deformity after surgery [1,16,17].

Prognostic factors that correlate with better prognosis and lower rates of complications include reducibility and stability on primary neonatal examination and presence of anterior skin grooves [8,17]. Poor prognostic factors include an association with a genetic syndrome, knee flexion <50° (related to severe quadriceps retraction), and the absence of anterior skin grooves [8,16,17]. Our patient with CDK had anterior skin grooves and reducibility with stability yielding a more favorable outcome, although knee flexion was evaluated at approximately 30°, providing some less favorable factors.

Conclusions

Congenital dislocation of the knee is a rare finding. The diagnosis is primarily clinical and radiographs are used to confirm and assess the degree of dislocation, which is important for management and prognosis. Intervention, ranging from serial casting to surgery, is required as soon as possible. It may be associated with genetic syndromes but also with developmental hip dysplasia and clubfoot, and screening is warranted for these conditions. Further research should be done to improve prenatal detection of this deformity.