 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Magnetic Resonance Imaging of Skewfoot* , PHILADELPHIA, PENNSYLVANIAInvestigation performed at the Departments of Radiology and Orthopedics, Children's Hospital of Philadelphia, and the University of Pennsylvania, Philadelphia
Magnetic resonance imaging was used to visualize the ossified and unossified portions of the bones and soft tissues of the feet in order to evaluate the tarsometatarsal anatomy in sixteen children, three months to six years old (mean, fifteen months old), who were seen in the orthopaedic clinic with a suspected diagnosis of skewfoot. Twenty-seven feet were clinically abnormal and five were normal. Of the abnormal feet, twenty-six had a radiographic diagnosis of skewfoot and one, of simple metatarsus adductus. Of the skewfeet, seven had a talocalcaneal angle of 45 degrees or more as measured on the lateral radiograph and six had a talocalcaneal angle of 45 degrees or more as measured on the anterior radiograph. Valgus deformity of the hindfoot was not apparent on clinical examination in any of the children. The talocalcaneal angles measured on the magnetic resonance images corresponded poorly with those measured on the radiographs, possibly because it is not possible to simulate weight-bearing during magnetic resonance imaging or because of the effect of partial volume averaging on thin sections. However, magnetic resonance imaging demonstrated the shapes of the bones and the positions of the unossified portions of the bones. Magnetic resonance imaging showed lateral subluxation of the navicular in twenty-four skewfeet, plantar subluxation in ten, and medial subluxation of the first metatarsal on the medial cuneiform in twenty-five. The alignment of the lateral margin of the calcaneus and cuboid on the magnetic resonance images was normal in all patients. Magnetic resonance imaging has the unique ability to show the cartilaginous and ossified portions of the developing bones of the foot.
Adduction of the forefoot is the most common deformity of the foot in children3. The deformity may be isolated or associated with abnormalities of the midfoot or hindfoot. Discussion of adduction of the forefoot is complicated by the confusing variety of terms associated with it, such as metatarsus adductus, metatarsus varus, metatarsus adductovarus, skewfoot, z foot, and serpentine foot. Metatarsus adductus is simple medial deviation of the metatarsals on the cuneiform in the plane of the foot. Metatarsus varus and metatarsus adductovarus also include adduction of the forefoot. Supination of the sole of the foot has occasionally been described in association with these abnormalities. The term skewfoot, also known as hooked foot, z foot, or serpentine foot, was coined by McCormick and Blount to describe adduction of the forefoot with many associated deformities of the midfoot and hindfoot, including varus deformity of the forefoot and cavus deformity. Other authors have redefined the term skewfoot by adding valgus deformity of the hindfoot to the definition1,9. At birth, there is only partial ossification of the talus, calcaneus, metatarsals, and cuboid, and the remainder of the midfoot is unossified4. Until recently, operative findings and autopsy studies were the only methods available for the evaluation of the unossified bones of the foot. Now, with magnetic resonance imaging, the anatomy of the unossified portion of bones and the soft tissues can be shown. The purpose of this study was to evaluate the tarsometatarsal anatomy in children in whom skewfoot was suspected.
Children who were seen in our orthopaedic clinic between March 1990 and June 1992 and in whom skewfoot was suspected on the basis of the clinical examination were referred for magnetic resonance imaging. On physical examination, there was adduction of the forefoot as well as a palpable step-off on the medial border of the midfoot at the talonavicular joint or a stiff uncorrectable deformity of the midfoot involving either the first metatarsal-medial cuneiform joint or the navicular-cuneiform joint. Scanning was performed for sixteen children, who ranged in age from three months to six years (mean, fifteen months). The children had no other physical deformities. Only one child had received any therapy before the magnetic resonance imaging. During the same period, an equal number of children seen in our orthopaedic clinic had simple metatarsus adductus on clinical examination and no visible or palpable abnormality of the midfoot. These children were not the focus of the study and were not studied radiographically or with magnetic resonance imaging. Radiographs of thirty-one feet were available for retrospective analysis and comparison with the magnetic resonance images. Children who were able to stand had had anterior or lateral transmalleolar radiographs, or both, of the feet made while standing, and children who were unable to stand had had anterior or lateral radiographs, or both, made with the foot in forced dorsiflexion. One clinically normal foot was not examined radiographically. The talocalcaneal angle and the talus-first metatarsal angle were measured on the lateral radiograph, and the talocalcaneal angle and the angle between a line drawn through the mid-talar long axis and the first metatarsal were measured on the anterior radiograph (Fig. 1-A). On the anterior radiograph, the foot was believed to have a skew shape if the mid-talar line was medial to the first metatarsal line and there was an apparent step-off in the soft tissues on the medial border of the foot between the first metatarsal and the talus. In a normal foot and in metatarsus adductus, the mid-talar line hits the base of the first metatarsal or is lateral to the first metatarsal line, and no step-off is noted in the soft tissues of the medial border of the foot. The calcaneocuboid alignment was evaluated on the anterior radiograph by drawing a line through the mid-calcaneal long axis and observing whether it bisected the ossification center of the cuboid.
All of the children were sedated for the magnetic resonance imaging. Children who were less than eighteen months old received sixty-five to 100 milligrams of chloral hydrate per kilogram of body weight orally, and those who were more than eighteen months old received two to six milligrams of pentobarbital per kilogram of body weight intravenously. All patients were monitored with pulse oximetry. Both feet were scanned simultaneously with an extremity coil in a 1.5-tesla Magnetom unit (Siemens, Erlangen, Germany). The feet were placed on a plastic foot-holder to maintain a constant position and to try to prevent plantar flexion. The foot-holder consisted of a piece of moldable plastic with the foot-plate at 90 degrees to the back-plate and multiple Velcro straps that crossed the feet and ankles. Three-millimeter-thick T1-weighted angled coronal (long-axis) and straight sagittal images were made throughout the feet. The talocalcaneal angle was measured on the coronal images. A line was drawn through the mid-talar axis to determine if it passed through or was medial to the base of the first metatarsal. Medial displacement of the base of the first metatarsal on the medial cuneiform was graded as mild, moderate, or severe by dividing the base of the metatarsal into thirds and determining which third aligned with the medial edge of the cartilage of the medial cuneiform. The cartilage of the head of the talus was divided into fourths, and the position of the medial edge of the cartilage of the navicular was assigned to one of these quadrants (Fig. 1-B). The shape of the navicular was evaluated. A navicular of normal shape appears as a smooth arch of uniform thickness on the coronal image and as a slight wedge on the sagittal image, with the plantar aspect of the navicular being slightly smaller5. The alignment of the lateral margin of the cuboid with respect to the calcaneus was evaluated. A line was drawn through the mid-calcaneal long axis to see if it bisected the ossification center of the cuboid (Fig. 1-A).
With the coronal examination, the talonavicular relationship was seen best on angled coronal images made along the axis of the talus and the first metatarsal. The calcaneocuboid relationship was seen best on angled coronal images made through the base of the calcaneus (Fig. 1-C).
The talocalcaneal angle and the position of the navicular with respect to the talar head were measured on the sagittal images. The alignment of the lateral margin of the calcaneus and cuboid and the shape of the navicular were also evaluated. In both planes, the course and caliber of the muscles and tendons of the foot and ankle were noted. The osseous positions were evaluated only if the bones or joints were visualized on two consecutive images. This was done to help to eliminate errors caused by partial volume averaging on thin sections.
Eight of the sixteen children were less than eleven months old and were not yet walking. Seven children were thirteen to twenty-six months old and were walking. One child was six years old and had been managed for presumed metatarsus adductus with reverse-last shoes in infancy at another institution (Fig. 2-A and Fig. 2-B). Thirty-two feet were scanned; five had been considered normal and twenty-seven had been considered abnormal on the basis of clinical examination. Of the clinically abnormal feet, twenty-six had evidence of skewfoot on the anterior radiograph and one had evidence of simple metatarsus adductus (Table I). Four clinically normal feet had normal findings on the radiographs, and one foot had not been assessed radiographically.
The talocalcaneal angle, as measured on the coronal magnetic resonance images, ranged from 30 to 47 degrees (mean, 38 degrees) in the clinically diagnosed skewfeet and from 30 to 45 degrees (mean, 39 degrees) in the clinically normal feet (Table I). Twenty-four of the skewfeet had lateral subluxation of the navicular with respect to the talar head. There was no subluxation in the normal feet. The medial edge of the navicular with respect to the talar head ranged from alignment with the medial one-fourth to alignment with the medial one-half of the talar head; the mean position of the medial edge of the navicular was at the junction of the medial fourth and the second fourth of the talar head. The medial one-fourth of the talar head was uncovered. All of the patients had a normally shaped navicular. Twenty-five of the skewfeet had medial subluxation of the first metatarsal on the medial cuneiform; it was mild in nine, moderate in twelve, and severe in four. The feet that had moderate or severe subluxation had mild rounding off of the distal part of the medial edge of the medial cuneiform (Figs. 3-A and 3-B). There was no medial subluxation of the first metatarsal on the cuneiform in the normal feet.
As measured on the sagittal images, the talocalcaneal angle ranged from 15 to 42 degrees (mean, 26 degrees) in the skewfeet and from 18 to 30 degrees (mean, 22 degrees) in the normal feet. The alignment of the navicular with respect to the talar head in the sagittal axis was normal in thirteen of the skewfeet. Ten skewfeet had plantar displacement of the navicular with respect to the talar head. The plantar displacement of the navicular seen on the magnetic resonance image did not correlate well with the midfoot angle measured on the lateral radiograph. The shape of the navicular and the calcaneocuboid alignment appeared normal in all of the feet in the series. The course and caliber of the major tendons (the anterior tibial, posterior tibial, and peroneus tendons) appeared normal in all feet. The talocalcaneal angle, as measured on the anterior radiograph, ranged from 28 to 52 degrees (mean, 39 degrees) in the skewfeet and from 25 to 42 degrees (mean, 34 degrees) in the four normal feet for which radiographs were available. The talus-first metatarsal angle ranged from 0 to 50 degrees (mean, 19 degrees) in the skewfeet and from -21 to -7 degrees (mean, -15 degrees) in the four normal feet for which radiographs were available. The talocalcaneal angles measured on the radiographs did not correlate well with the angles measured on the magnetic resonance images. The differences between the angles on the anterior radiographs and those on the coronal magnetic resonance images ranged from 0 to 18 degrees. The differences between the angles on the lateral radiographs and those on the sagittal magnetic resonance images ranged from 5 to 37 degrees. On the anterior radiograph, the ossification center of the cuboid was bisected by the line drawn through the mid-calcaneal long axis in twenty-two feet, was medial to the line in three, and was lateral to the line in five. There was discordance between the position of the line through the mid-calcaneal long axis with respect to the cuboid on the magnetic resonance image compared with that on the radiograph of nine feet. However, the lateral margin of the calcaneus and cuboid was aligned normally on the coronal magnetic resonance image of all feet. Of the skewfeet with plantar displacement of the navicular evident on the sagittal magnetic resonance image, six were seen to have a normal talocalcaneal angle on the anterior and lateral radiographs and four had an angle of more than 44 degrees on the anterior or lateral radiograph, or both. Of the skewfeet with no plantar displacement of the navicular evident on the sagittal magnetic resonance image, ten were seen to have a normal talocalcaneal angle on the anterior and lateral radiographs and three had an angle of more than 44 degrees on the anterior or lateral radiograph, or both. Twenty-four of the skewfeet had lateral subluxation of the navicular on the coronal magnetic resonance image. The talocalcaneal angle was increased on the anterior and lateral radiographs of three of these feet, only on the anterior radiograph of three, and only on the lateral radiograph of four feet; it was normal on both radiographs of fourteen feet. The talus-first metatarsal angle as measured on the anterior radiograph was more than 20 degrees in three of the skewfeet with plantar subluxation of the navicular and in six of the skewfeet without plantar subluxation of the navicular. Five of the nine feet with a talus-first metatarsal angle of more than 20 degrees were skewfeet with moderate or severe lateral subluxation of the navicular.
The term skewfoot was first used by McCormick and Blount in 1949. However, the deformity was first mentioned in the American literature, in 1933, by Peabody and Muro. They described a deformity involving primarily the forefoot, with the base of the first metatarsal articulating with the medial aspect of the medial cuneiform, causing a marked varus angle with respect to the midfoot, and a prominent concavity between the first metatarsal and the navicular. Pronation with inward rotation of the talar head and lateral displacement of the navicular and medial cuneiform as well as a flat sole have also been described as part of the skewfoot deformity7. Valgus deformity of the heel was added to the definition of skewfoot by Berg as well as by Peterson. The etiology of this deformity is uncertain. Asymmetrical pull of the muscles and tendons has been suggested. Reimann and Werner postulated that metatarsus varus was a congenital deformity that started as subluxation of the tarsometatarsal joints in the dorsiflexed foot, which caused secondary changes in the shapes of bones and soft-tissue contractures. Abnormalities of the insertions of the anterior tibial8 and posterior tibial2 tendon attachments on the cuneiform and first metatarsal have been described in dissection specimens. No abnormalities in the course of the tendons were noted in the patients in our study. However, these abnormalities in the tendinous insertions in the infant foot may be too subtle to discern on magnetic resonance images. The abnormalities of the hindfoot and midfoot in skewfoot have been attributed to intensive treatment with casts or to walking on the medially deviated forefoot9. In the present study, only one child, who was six years old, had been previously managed with reverse-last shoes. Eight of the children were less than eleven months old and had not begun walking. This suggests that the deformity that we identified in this group of children was congenital and was not the result of treatment or abnormal weight-bearing. Radiography remains the most efficient means of evaluating the talocalcaneal and talus-first metatarsal relationships, as these bones are partially ossified at birth. The navicular and cuneiforms normally ossify between the ages of eighteen months and five years4. However, ossification of the tarsals is frequently delayed in the deformed foot6. The advantage of magnetic resonance imaging is its ability to show the unossified portions of the tarsals and their interrelationships. It was difficult to measure the angles on the magnetic resonance images because of the partial volumes obtained when thin sections are made. Simulated weight-bearing, which is the standard for radiographic evaluation of a deformed foot, cannot be performed with magnetic resonance imaging; thus, comparisons with the angles measured on radiographs are suspect. All radiographic determinations are affected by errors in measurement caused by variability in the positioning of the patient and in the locations where the measurements are made. However, the purpose of this study was not to duplicate the information available on radiographs but to obtain information regarding the shape and position of the unossified portions of the tarsals. It is reasonable to assume that the shapes of the tarsals would be constant with or without weight-bearing. Adduction of the forefoot was the presenting deformity in all patients. It is difficult to determine on the magnetic resonance image if displacement of the metatarsals (varus deformity of the forefoot) represents a true subluxation of the metatarsal-cuneiform joint or merely a varus deformity in the shape of the cuneiform or the base of the first metatarsal, or both, secondary to remodeling of the bone. The shape of the cuneiform was abnormal in the feet with severe adduction of the forefoot. The definition of simple metatarsus adductus is medial displacement of the metatarsals on the cuneiform in the plane of the sole. On the radiograph of an otherwise normal foot with simple metatarsus adductus, the mid-talar axis should hit the base of the first metatarsal or be lateral to it3. In the patients in the present study who had suspected skewfoot, the mid-talar axis hit or was medial to the base of the first metatarsal. On the coronal magnetic resonance image, this relationship of the forefoot to the hindfoot appeared to be related to the lateral displacement of the navicular on the head of the talus. Lateral displacement of the navicular was a common finding on the magnetic resonance images of the skewfeet. In the five normal feet, a relatively straight medial margin in the alignment of the tarsals and metatarsals was demonstrated on the coronal images. However, as magnetic resonance imaging was performed for only a small number of normal feet in this study, it may be difficult to make valid comparisons. Plantar displacement of the navicular with respect to the talus was seen on the sagittal magnetic resonance image of less than half of the skewfeet (Figs. 4-A, 4-B, 4-C through 4-D). In the five normal feet, a smooth curved arch in the midfoot was evident on the sagittal magnetic resonance image. The plantar displacement of the navicular was not obvious on the radiographs as the midfoot was only partially ossified. However, the talus-first metatarsal angle was increased on the lateral radiograph. A deformity of the midfoot with an increase in the talus-first metatarsal angle on the lateral radiograph is part of the spectrum of flatfoot deformities. We attempted to subclassify the skewfeet into two groups: the feet in which the navicular was displaced laterally and the talocalcaneal angle was increased on both the anterior and the lateral radiograph and those in which the navicular was displaced in a plantar direction and, thus, laterally, without any change in the talocalcaneal angle. However, we were not able to confirm such a pattern.
Flatfoot deformity has been described in association with skewfoot7, although it was not found on clinical examination in any of our patients. There may be subtypes of flatfoot deformity, with one type resulting when the talocalcaneal angle increases as the calcaneus everts through the subtalar joint, causing valgus deformity of the heel. Another type of apparent flatfoot may result from plantar flexion of the navicular with dorsiflexion of the metatarsals on the midfoot. In this study, the range of talocalcaneal angles on the magnetic resonance images and radiographs was large, as is also the case for normal feet, according to Vanderwilde et al. They limited the definition of valgus deformity of the hindfoot, to more than two standard deviations above the mean11. Their measurements on radiographs of normal feet revealed that the fiftieth percentiles in newborns for the lateral and anterior talocalcaneal angles were 39 and 42 degrees, respectively, with upper limits of 55 and 56 degrees, respectively. When defining valgus deformity of the hindfoot, Berg considered 45 degrees on the lateral radiograph and 35 degrees on the anterior radiograph as the upper limits of normal for the talocalcaneal angle. Therefore, more patients are considered to have a valgus deformity of the hindfoot with the use of Berg's definition than with the use of the definition of Vanderwilde et al. In our study, only seven of the skewfeet had a talocalcaneal angle of 45 degrees or more on the lateral radiograph. Valgus deformity of the heel was not apparent in these patients on clinical examination. Lateral subluxation of the cuboid with respect to the distal end of the calcaneus was described by Berg in children who had a skewfoot deformity. This abnormality in the calcaneocuboid relationship was used to predict the outcome of therapy. The alignment of the lateral margin of the calcaneus and cuboid appeared normal on the magnetic resonance images of all of the patients in our study (Fig. 5), even when the line drawn through the mid-calcaneal long axis did not bisect the ossification center of the cuboid on the radiograph. It was not possible to assess whether there was a slight change in the shape of the calcaneus that would account for the apparent malposition of the calcaneocuboid joint in the study by Berg. The apparent lateral position of the cuboid on the radiographs may also have been caused by changes in positioning of the foot, as the position of the cuboid varied on radiographs of the same patient made at different times in our study. Also, pronation of the foot may have caused an apparent lateral displacement of the cuboid with respect to the calcaneal axis, as viewed on the anterior radiograph.
It has been suggested that lateral displacement of the navicular must be accompanied by either a lateral shift in the position of the cuboid or an increase in the talocalcaneal angle. This was not documented on the magnetic resonance images in our study. However, it may be difficult to distinguish between a change in the shape of a small bone and rotation of that bone on a thin-section two-dimensional image. The appearance of lateral displacement of the navicular may actually represent plantar displacement of the navicular with rotation. We were not able to define a distinct subset of feet with plantar displacement of the navicular and an increase in the talocalcaneal angle. Of the skewfeet with lateral displacement of the navicular, only ten had an increase in the talocalcaneal angle on either the anterior or the lateral radiograph. We did document a change in the shape of the medial border of the medial cuneiform in the feet with moderate or severe displacement of the first metatarsal. There may have been other changes in the shape of the tarsals that were too subtle for us to detect on the magnetic resonance images that may account for the tarsal relationships that we found. Although, to our knowledge, changes in the shape of the bones have not been described previously in abnormal specimens, few specimens have been available for that type of anatomical study. Magnetic resonance imaging allowed us to evaluate the position of the unossified bones of the midfoot with respect to the partially ossified talus, calcaneus, cuboid, and first metatarsal. Lateral displacement of the navicular and medial subluxation of the first metatarsal on the magnetic resonance images correlated well with adduction of the forefoot and a step-off in the midfoot on the radiographs. Skewfoot deformity may be due to the varus angle of the medial cuneiform, medial subluxation of the first metatarsal on the medial cuneiform, lateral subluxation of the navicular on the talar head, or plantar subluxation of the navicular on the talar head. Although magnetic resonance images also showed lateral and plantar subluxation of the unossified bones of the midfoot in these feet, we no longer perform such studies for children who have a suspected skewfoot deformity. We believe that the positions of the unossified bones of the midfoot can be approximated on the basis of the positions of the talus, calcaneus, and first metatarsal on the radiographs. There is a spectrum of deformities between metatarsus adductus and skewfoot that cannot always be determined precisely from the clinical and radiographic examination. We believe that, if the base of the first metatarsal is lateral to the mid-talar axis, lateral subluxation of the navicular on the talus is to be expected and may help to distinguish metatarsus adductus from skewfoot.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |