The Fetal Musculoskeletal System

Development

The high level of intrinsic contrast of the fetal extremities places them among the earliest structures that can be evaluated by ultrasound. By the end of the embryonic period, the differentiation of bones, joints, and musculature is similar to that of an adult and is associated with increased fetal movements. Transvaginal ultrasound can demonstrate the limb buds by 7 weeks' gestation, and the foot and hand plates are visible by 8 weeks. Osteogenesis begins in the clavicle and mandible by 8 weeks as well. By 11 or 12 weeks, the primary ossification centers of the long bones (e.g., scapula, ileum), as well as the limb articulations and phalanges, can be identified. The ischium, metacarpals, and metatarsals ossify during the fourth month of gestation. The pubis, calcaneus, and talus ossify during the fifth and sixth months. Ossification of the other tarsal and carpal bones occurs postnatally. The direction of growth in the long bones is from proximal to distal, and the lower extremities lag slightly behind the upper extremities.

Of the secondary ossification centers in the long bones, only the distal femoral epiphysis, the proximal tibial epiphysis, and occasionally the proximal humeral epiphysis ossify prenatally ( Fig. 40.1 ). The unossified epiphysis appears hypoechoic, with a variably, mildly echogenic center. Ossification begins centrally. The distal femoral epiphysis can ossify as early as 29 weeks' menstrual age and as late as 34 weeks. When it measures greater than 7 mm, the menstrual age is generally later than 37 weeks. The proximal tibial epiphysis begins to ossify by 35 menstrual weeks. In uncomplicated pregnancies the combination of a distal femoral epiphysis of 3 mm or greater and the presence of a proximal tibial epiphysis is considered a reliable marker of pulmonary maturity. Intrauterine growth restriction (IUGR) is associated with a delay in ossification of the distal femoral epiphysis and proximal tibial epiphysis. The earliest secondary epiphysis to ossify is the calcaneus, at approximately 20 weeks' gestation, thus marking the earliest point that assessment of delayed ossification of the secondary epiphyseal centers can be attempted.

The fascia within the muscle is highly echogenic compared with the relatively hypoechoic cartilage. The fetal musculature is slightly more echogenic than the relatively hypoechoic cartilage. The fetal joint spaces, in particular the knee, appear echogenic because of the combination of synovium, fat, and microvasculature. The normal development and ultimate function of the fetal musculoskeletal system depend on fetal movements, which start by the second half of the first trimester. In the absence of normal fetal motion, the bones and muscles will be underdeveloped, the chest will be narrow, and joint contractures and postural deformities can occur.

Extremity Measurements

It is a standard practice to assess femur length (FL) as part of the evaluation of fetal size and morphology. Although measurement of all the long bones is not required in a routine obstetric ultrasound, an overall evaluation of the fetal skeleton should be performed to ensure the presence and bilateral symmetry of the tubular bones. Available charts provide guidance for correlating the length of the extremities with the gestational age ( Table 40.3 ).

The longest femur measurement, excluding both proximal and distal epiphyses, is usually chosen. The inclusion of the distal femur point, or the specular reflection of the lateral aspect of the distal femoral epiphysis cartilage, is the most common reason for overestimating FL ( Fig. 40.2A ). An oblique FL measurement will result in underestimate of length. The lateral border of the femur in the near field of the transducer appears straight, whereas the medial border of the femur in the far field has a curved appearance ( Fig. 40.2B ).

In the lower extremity the lateral bone is the fibula and the medial bone is the tibia. The tibia and fibula end at the same level distally. In the upper extremity, pronation may cause the radius and ulna to cross, so it can be difficult to distinguish the ulna from the radius using lateral and medial positions. The ulna is distinguished from the radius by its longer proximal extent and its relationship to the fifth digit distally. The radius and ulna end at the same level distally. Demonstration of this relationship will effectively exclude the majority of radial ray defects.

The clavicles grow in a linear fashion, approximately 1 mm per week, and the gestational age in weeks is approximately the length of the clavicle in millimeters from 14 weeks to term. By 40 weeks' gestation, the clavicles measure approximately 40 mm.

Foot length is measured from the skin edge overlying the calcaneus to the distal end of the longest toe (the first or second toe) on either the plantar or the sagittal view ( Fig. 40.3 ). The ossified FL is almost equivalent to the foot length, resulting in a normal femur-to-foot length ratio of approximately 1.0. This ratio remains relatively constant from the 14th week of gestation onward. If the fetus is constitutionally small or there is symmetrical IUGR, the ratio is generally 0.9 or greater. In most skeletal dysplasias characterized by short limbs, the ratio is generally less than 0.9 because of the relative sparing of the hands and feet. The greater the deviation is from the lower limits of the norm, the greater is the disproportion and usually also the severity.

Positive Family History

A positive family history of a sibling or parents affected by a skeletal dysplasia or consanguineous parents should prompt an intensive ultrasound investigation with a focus on targeted abnormalities and serial measurements. A history of consanguinity is important because many of the skeletal dysplasias have an autosomal recessive mode of inheritance. Heterozygous achondroplasia, the most common nonlethal skeletal dysplasia, has an autosomal dominant pattern of inheritance. Family history may not be helpful because 80% of cases are caused by a new mutation.

Abnormal Bone Length or Appearance

The fetal femur is often the only long bone routinely measured at the second-trimester ultrasound evaluation. An abnormal FL is traditionally defined as below 2 standard deviations (SD) for gestational age. Using this cutoff, 2.5% of all fetuses would be classified as having short limbs. This exceeds the expected frequency of skeletal dysplasias, and thus additional investigations are needed to separate fetuses with a skeletal dysplasia from those with no underlying pathology and those with other diagnoses associated with growth restriction.

When one or all of the long bones measure less than 2 SD for gestational age, a follow-up ultrasound should be done in 3 or 4 weeks to evaluate the interval growth. If the interval femur growth is normal, there is a high likelihood that the fetus does not have skeletal dysplasia. However, further deviation from the mean by at least 1 SD suggests the presence of a skeletal dysplasia or severe IUGR. When FL measures below 4 SD for gestational age, there is a high likelihood of a skeletal dysplasia. Kurtz et al. have shown that the number of millimeters below the −2 SD line is a simple screening tool to evaluate femoral shortening with the following guidelines:

• If FL is 1 to 4 mm below the −2 SD point, further serial measurements are required to determine if a skeletal dysplasia is present.

• If FL is more than 5 mm below the −2 SD point, there is a high likelihood of a skeletal dysplasia.

The most common cause of a so-called short femur is either inaccurate dating or a normal variant in a constitutionally small fetus, which may be associated with a parental or family history of less-than-average stature. In about 13% of cases, a remeasurement will bring the FL into a normal range, likely representing a false-positive diagnosis rather than a growth spurt. Isolated, symmetrical short femurs identified at the second midtrimester ultrasound evaluation are helpful in identifying a group of fetuses at increased risk for low birth weight, small for gestational age, or severe IUGR. Typically, these fetuses will also have small abdominal circumference measurements.

Occasionally, fetuses with severe IUGR will have greatly shortened long bones. Associated findings of normal or decreased skin fold measurements, oligohydramnios, abnormal placental morphology, and abnormal Doppler waveforms suggest the diagnosis of IUGR, whereas redundant, thickened skin folds and polyhydramnios typically accompany short-limb dysplasias.

Nonlethal skeletal dysplasias such as heterozygous achondroplasia are often not evident before 20 weeks' gestation. The findings of short long bones before 20 weeks indicate a more serious and usually fatal skeletal dysplasia. As a rule, the earlier the detection of limb shortening, the worse is the prognosis. Virtually all cases diagnosed in first trimester are considered severe skeletal dysplasias, with the large majority representing lethal conditions. First-trimester skeletal biometry tables are available. Although mild, isolated shortening of the femur indicates increased risk for trisomy 21 by 1.5-fold, other factors are more important for assessing this risk.

The pattern of limb shortening should be assessed to determine which long-bone segments are most severely affected ( Fig. 40.4 ). Rather than millimeters, we find it useful to standardize measurements to “weeks of size” to determine disproportion. There are four main patterns of shortening of the long bones: rhizomelia, shortening of the proximal segment; mesomelia, shortening of the middle segment; acromelia, shortening of the distal segment; and micromelia, shortening of the entire limb (mild, mild/bowed, or severe).

The shape, contour, and density of the bones should be assessed for the presence of bowing, angulation, fracture, or thickening. Bowing is a nonspecific finding, typically caused by underlying osseous fragility. Although more than 40 distinct disorders are associated with bowed, bent, or angulated femurs, most cases can be accounted for by three relatively frequent disorders: campomelic dysplasia, thanatophoric dysplasia, and osteogenesis imperfecta (OI). Patients with OI types I and IV can present with apparently isolated in utero bowing of the long bones (especially the femur) without frank fractures (see Fig. 40.2D ). Anterior bowing of the tibia, femur, and humerus may suggest the diagnosis of campomelic dysplasia; other more specific findings such as hypoplastic scapulae and cervical kyphosis are typically present and help distinguish campomelic dysplasia from OI and thanatophoric dysplasia, which are both more common. Bone fractures may appear as angulations or interruptions in the bone contour or as thick, wrinkled contours corresponding to repetitive cycles of fracture and callus formation (see Fig. 40.2F ). Decreased or absent acoustic shadowing is a marker for decreased mineralization of the long bones. When evident, this is helpful in narrowing the differential diagnosis, but in its absence the bone mineralization may still be abnormal.

The spine is assessed for segmentation anomalies, kyphoscoliosis, platyspondyly (flattened vertebral bodies), demineralization, myelodysplasia, and caudal regression syndromes. Although platyspondyly is the most common spine abnormality, it is a challenging prenatal diagnosis. Demineralization of the spine can result in the appearance of ghost vertebrae or nonvisualization of one or all of the three ossification centers ( Fig. 40.5 ). A progressively narrowed lumbar interpedicular distance is associated with homozygous achondroplasia; a widened interpedicular distance is associated with myelodysplasia. Three-dimensional imaging may provide a large field of view and additional details that may be helpful in achieving a diagnosis ( Fig. 40.6 ).

The most reliable prognostic indicator of the lethality of a given skeletal dysplasia is pulmonary hypoplasia. Ultrasound is 85% to 95% accurate in the diagnosis of a lethal skeletal dysplasia on the basis of pulmonary hypoplasia. The thoracic circumference is measured at the level of the four-chamber heart and compared to nomograms (see Chapter 36 , Table 36.2 ). A thoracic-to-abdominal circumference ratio of less than 0.8 is considered abnormal. The thoracic length (from the neck to the diaphragm) is also measured, and the ribs are assessed to determine if they are short. At the level of the four-chamber cardiac view, the ribs should normally encircle at least 70% of the thoracic circumference. The ribs remain in a relatively horizontal plane, as does the cardiac axis, facilitating this evaluation. In a sagittal view, a markedly narrowed anteroposterior diameter of the thorax is associated with pulmonary hypoplasia. In the coronal view, a concave or bell-shaped contour is associated with pulmonary hypoplasia ( Fig. 40.7 ).

The hands and feet are examined for deformities such as clubfoot or clubhand. A hitchhiker thumb, or abducted thumb, is associated with diastrophic dwarfism. Fixed postural deformities may suggest the diagnosis of arthrogryposis multiplex congenita. Polydactyly is associated with short-rib polydactyly syndromes, Ellis–van Creveld syndrome, asphyxiating thoracic dystrophy, and some chromosomal abnormalities.

The fetal cranium is assessed for the presence of macrocranium, frontal bossing, cloverleaf skull deformity, underlying brain abnormalities, and facial abnormalities, such as saddle nose, hypertelorism, and cleft lip and palate. An abnormal cranial contour may indicate craniosynostosis or premature fusion of the sutures. The most reliable sonographic sign of demineralization is increased compressibility of the calvarium. This finding is typically present in fetuses with osteogenesis imperfecta type II and hypophosphatasia. The falx may appear abnormally bright or echogenic compared to the demineralized calvarium.

The ribs are assessed to ensure an adequate length, thus minimizing the risk of pulmonary hypoplasia, and are examined for abnormal number or appearance. The finding of an abnormal number of fetal ribs is an isolated finding of no clinical importance in the majority of cases, associated with minor anomalies in a smaller group (29%) and only occasionally associated with severe malformations. Associated syndromes include Poland syndrome, VACTERL association (vertebral abnormalities, anal atresia, cardiac abnormalities, tracheoesophageal fistula, renal agenesis and dysplasia, limb defects), campomelic dysplasia, and chromosome abnormalities. Three-dimensional ultrasound volume images aid in accurately counting the number of ribs ( Fig. 40.8 ).

Ultimately, a detailed examination of each bone may be required to determine the fetal condition. Specific dysmorphic features of bones (e.g., clavicular or scapular hypoplasia; aplasia of fibula, tibia, or radius; platyspondyly) can be helpful to further define a specific skeletal dysplasia. A detailed evaluation of the cardiovascular, genitourinary, gastrointestinal, and central nervous system should be done concurrently with the musculoskeletal evaluation.

Dizygotic twin pregnancies are at similar risk for skeletal abnormalities as singleton pregnancies, and the frequency is increased two to three times in monozygotic twins. Both monozygotic and dizygotic twins can be discordant for genetic and nongenetic skeletal abnormalities. Twin pregnancies are generally discordant and overall, about 15% of twins are concordant for the same anomaly.

Additional Diagnostic Techniques

Three-Dimensional Ultrasound

Three-dimensional ultrasound is becoming an increasingly useful complement to two-dimensional ultrasound in diagnosis of skeletal dysplasia and pulmonary hypoplasia. High-contrast structures such as the fetal skeleton are especially amenable to three-dimensional ultrasound-rendering software and postprocessing techniques. Surface-rendering capabilities are particularly useful for visualizing subtle facial dysmorphism (such as low-set or deformed ears, micrognathia, or flattening of the facial profile associated with midface hypoplasia), evaluating cranial distortion due to craniosynostosis, or assessing hand and foot abnormalities. However, there are insufficient published data to gauge the diagnostic performance of this modality ( Fig. 40.9 ).