Assessment of lower-extremity alignment in the transverse plane: implications for management of children with neuromotor dysfunction.

Assessment of Lower-Extremity Alignment in the Transverse Plane: Implications for Management of Children with Neuromotor Dysfunction

The authors present nine clinical assessment procedures designed to detect factors contributing to transverse-plane structural and joint alignment abnormality in children with neuromotor dysfunction. Where applicable, each assessment is accompanied by a discussion of the normal features of pediatric lower-extremity torsional and rotational alignment; limitations of the assessment procedures as regards reliability, specificity, and age-related normative findings; and clinical management suggestions. The authors urge clinical evaluators to expand the existing knowledge base. [Cusick BD, Stuberg WA. Assessment of lower-extremity alignment in the transverse plane: implications for management of children with neuromotor dysfunction. Phys Ther. 1992;72:3-15.]

Key Words: Lower extremity, general; Neuromotor disorders, general; Orthopedics, general; Pediatrics, evaluation.

Children with chronic neuromotor dysfunction commonly exhibit deviations in transverse-plane (torsional and rotational) alignment of the pelvis and lower extremities (Fig. 1). Abnormal weight bearing and muscle action and shortened soft tissues impose forces that combine to impede and impair normal maturation of torsional and rotational features of skeletal structure and joint alignment. Unlike children with normal neuromotor status, most children who lack normal neuromuscular function do not develop efficient mechanisms to compensate for these structural abnormalities.[1] Alignment problems generally worsen with increasing age.[2-19] The clinician should base management decisions on findings gained through a combination of gait analysis, posture evaluation, and a thorough biomechanical assessment.[3,17,20,21] To determine whether a clinical biomechanical finding indicates the presence of pathology, a clinician must identify and acknowledge normal maturational changes in skeletal architecture and joint configuration as they are currently understood.[1] We believe that research is needed to establish adequate norms and to evaluate the efficacy of interventions. The purpose of this article is to raise the therapist's awareness of certain biomechanical components that can influence lower-extremity transverse-plane alignment and function and of the need to expand the existing body of research pertaining to the significance of clinical findings. Following a brief review of terminology, we present nine clinical biomechanical assessment procedures. The discussion pertaining to these assessments covers normative data, normal maturational changes in structural and joint configuration, limitations of the procedures as regards reliability and validity issues, and clinical management implications of assessment results.

Terminology

Rotation--a movement of one or both joint segments in a plane that is perpendicular to the axis of the motion.[22,23] Medial rotation and lateral rotation occur when a joint segment is rotated toward and away from midline, respectively, around a vertical axis.

Torsion--a structural, osseous state of twist in a bone along its longitudinal axis.[24] "Femoral antetorsion" and "medial femoral torsion" are synonymous terms describing a femur that bears a medial twist of the distal-on-proximal ends.[3,25,26] "Femoral retrotorsion" and "lateral femoral torsion" are also synonymous terms that describe a deformity ranging from a lack of normal medial torsion to a true lateral twist of the distal-on-proximal ends of the femur.[3,25,26] Clinical tibiofibular torsion describes a state of twist, either medial or lateral, in the long axis of the tibiofibular unit.

Version--"The act or process of turning something or changing direction."[24(p1830)] For example, the femoral head and neck are described as anteverted when the head lies anterior to the frontal plane and retroverted when the head lies posterior to the frontal plane. Version and torsion are not identical, although they may occur together. Version describes a position in space relative to a plane. Torsion describes a twist in structure.

Genicular position--an abnormality in the relative ranges of medial or lateral axial tibiofibular rotation with the knee joint flexed 90 degrees.[25]

Clinical Biomechanical Assessment Procedures

At least 13 biomechanical factors can be cited as contributing to lower-extremity malalignment in the transverse plane: (1) pelvic rotation in the transverse plane;[7] (2) immature acetabular anteversion[19,27-32]; (3) soft tissue restriction of hip joint medial or lateral rotation mobility[19,25,33-35]; (4) abnormal magnitude of femoral torsion[*]; (5) abnormal activity of the medial hamstring muscles[3,4,21,36,39,46]; (6) medial or lateral genicular position[23,25,43,47]; (7) abnormal magnitude of tibiofibular torsion [dagger]; (8) abnormal transverse-plane alignment of the talar body within the ankle mortise[49,50]; (9) persistent talar neck adduction relative to the talar body[38]; (10) abnormal foot pronation, resulting in midtarsal joint abduction[36,45,51-53]; (11) abnormal foot supination, resulting in midtarsal joint adduction[21,52,53]; (12) metatarsus adductus[37,44,45]; and (13) overpull of the abductor hallucis muscle ("searching toe").[54,55] We refer the reader to the cited references for information that is beyond the scope of this article. We present a battery of nine assessment procedures that constitute a general screening. The battery features elements of Staheli's "Rotational Profile," which includes foot progression angle (FPA), medial and lateral hip rotation mobility with the hips extended, thigh-foot angle, transmalleolar axis (TMA)-thigh angle, and foot configuration (Fig. 2).[9,44,45,55-58] We have expanded Staheli's profile by adding pelvic rotation in the transverse plane, Ryder's test for femoral torsion, axial tibiofibular rotation mobility, and tibiofibular torsional status with the knee joint extended. The assessment process is begun by observing the foot progression angle (FPA) to provide a basis for investigating specific components of limb structure and joint mobility. The remaining eight tests proceed from proximal to distal body segments.

General Alignment

Test 1--foot progression angle. This test helps determine the angle formed by the foot and the line of forward progression. Abnormal FPA is commonly described as a gait deviation characterized by toeing-in or toeing-out beyond the normal mean. Procedure. The therapist dusts the soles of the child's feet with chalk and obtains a series of footprints. We measure FPA as the angle formed by the longitudinal bisection of the foot and the line of progression (Fig. 2). The longitudinal bisection of the foot normally falls between the center of the plantar heel and the second and third metatarsals.[59] We obtain at least six measurements for each foot and calculate the average FPA. Toeing-in is indicated by a negative value, and toeing-out is indicated by a positive value. Norms. Normative studies[60,61] have shown a mean FPA of 4 to 10 degrees in infants, children, and adults. The variability is greatest in children younger than 2 years of age.[60,61] Limitations. Replicability of findings is hindered by the problem of establishing a precise reference for the line of progression, as few children walk in a purely straight line. Pronation and supination foot deformities induce compensatory lateral and medial deviations, respectively, of the axis of the forefoot relative to that of the hindfoot. We use the longitudinal bisection of the hindfoot to measure FPA in children with neuromotor dysfunction. The orientation of the hindfoot bisection to the sagittal plane offers a closer manifestation of proximal rotary and torsional influences on foot alignment than does that of the bisection of the compensating forefoot. The longitudinal forefoot bisection divides the second and third metatarsals. Clinical implications. Toeing-in, which typically occurs in fewer than 30% of infants without neuromotor dysfunction, usually resolves by the age of 4 years.[40,47,48] An FPA of -1 to -10 indicates a mild abnormality, an FPA of -10 to -15 degrees indicates a moderate abnormality, and an FPA beyond -15 degrees indicates a severe abnormality. Adults rarely exhibit a negative FPA.[40,43,47,48,62] Children without neuromotor dysfunction who show persistent toe-in gait typically show no evidence of functional deficit.[63] A positive FPA exceeding 15 degrees imposes abnormal pronatory forces on the foot structures, as weight is borne abnormally on the medial aspect of the midfoot.[53] Motion analysis reveals the occurrence of lateral rotation of the femur and, to a greater degree, the tibia during the mid-stance and propulsion phases of gait in all children, including early walkers.[40,64] As the ligaments of the knee joint and foot gain integrity, these stance-phase lateral torque forces descending on the foot evidently induce spontaneous correction of toeing-in.[1,51] Children with abnormal muscle activity or myelomeningocele often demonstrate an abnormal magnitude of FPA.[4,7,36,39,40,65] Abnormal FPA in children with cerebral palsy depletes stability in the early stance phase of gait; necessitates that double-support stance begin prematurely; and reduces normal acceleration, velocity, and step length.[36] The components that contribute to abnormal FPA are discussed in the context of the remaining tests.

Proximal Alignment Features

We urge therapists to observe the postural alignment of each knee joint and patella relative to the sagital plane of the limb. If the knee axis and patella align medially in quiet stance, we look for pelvic rotation, soft tissue restriction on lateral hip rotation, and evidence of abnormally increased medial femoral torsion (see tests 2-4). If medial deviation increases in gait, we assess hamstring muscle length[4,39] using the popliteal angle test as described by Jones and Knapp[21] and Rang et al.[16] If the patella and knee axis align laterally, we look for restriction on medial hip rotation mobility and femoral retrotorsion (see tests 3 and 4).

Test 2--pelvic rotation. As the child stands and walks, we observe pelvic rotation in the transverse plane and note whether the right or left anterior superior iliac spine is anterior to a frontal-plane reference line. Norms. The pelvis ideally aligns on the frontal plane.[66(p144)] Clinical implications. If abnormal rotation is evident, the acetabulum on the anterior side may consequently be more anteverted than the acetabulum on the posterior side. Thus, the femur on the anterior side might falsely appear to be medially rotated.[19,30] Acetabular anteversion normally diminishes with skeletal maturation.[30,32,67,68] Persistence of structurally increased acetabular anteversion can cause toeing-in when femoral torsion is normal, or it can occur in conjunction with increased femoral antetorsion.[19,30] We know of no clinical test to determine acetabular orientation. Radiographic confirmation is needed.

Test 3--Ryder's test. The magnitude of femoral torsion is measured as the angle formed by the transverse-plane intersection of the femur's proximal reference axis (PRA) and distal reference axis (DRA). The PRA bisects the femoral head and greater trochanter, and the DRA lies parallel to the posterior aspects of the femoral condyles and is known as the transcondylar axis (TCA). Femoral antetorsion is evident when the TCA is placed on the frontal plane and the femoral head lies anterior to the frontal plane (Fig. 3). Femoral retrotorsion is evident when the TCA is placed on the frontal plane and the femoral head lies less than 10 to 12 degrees anterior to the frontal plane (Fig. 3).[25,26,43] The orientation of these axes to each other can only be estimated on clinical examination using Ryder's test, which features palpation of the position of the greater trochanter while rotating the femur.[4,19,25] Procedure. Position the child prone, supine, or sitting with the knee flexed to 90 degrees. The findings should be consistent in all positions, as they should represent the bony configuration. The examiner holds the leg proximal to the ankle and rotates the hip while palpating the anterior and posterior borders of the greater trochanter. When the trochanter reaches the most lateral position in the arc, we presume that the femoral neck and head are aligned on or near the frontal plane (Figs. 4, 5). We then measure the resulting position of hip rotation. We use a straight line connecting the mid-patella and the mid-ankle as the DRA.[69] If measuring with a goniometer, we use the table surface as the PRA (frontal plane) and subtract the resulting value from 90. We prefer to use a 7.62-cm-diameter (3-in-diameter), gravity-driven angle finder, [double dagger] [S] adapted by holding an extended straightedge against the shorter side of the angle finder (Fig. 5). The angle finder offers a vertical-plane reference, which is a gravitational constant. Norms. The magnitude of medial femoral torsion at birth approaches a mean of 40 degrees. Thereafter, the mean torsion angle reduces to 31 degrees at age 2 years, decreases to 25 degrees by age 8 years, and drops rapidly to the adult mean of 16 degrees between ages 14 and 16 years.[5,70] Stuberg and colleagues[71] reported mean medial rotation values of 10 to 13.5 degrees (SD = 3.3 [degrees] = 3.9 [degrees]) obtained by three testers on a group of 17 subjects, aged 3 to 22 years. Limitations. Age-related reliability coefficients and normative data for Ryder's test results have not been established. Stuberg et al[71] used a different goniometric technique from the one we suggest and showed a mean intertester difference of approximately 3 degrees. Reliability improved with the testers' clinical experience. Thickness of soft tissues overlying the greater trochanter can affect the precision of palpation. Age-related torsional normative values for computed tomography (CT) or magnetic resonance imaging (MRI) findings have not been established or validated against measurements on anatomic specimens. Clinical implications. Given the current available data comparing CT results with both Ryder's test results and measurements taken on an anatomic femur specimen, one can presume that the femur is twisted medially approximately 20 degrees more than the measured hip rotation value indicates.[7,71,72] For example, 5 degrees of lateral rotation (the distal leg deviates medially) represents a femur with approximately 15 degrees of medial torsion, whereas 5 degrees of medial rotation (the distal leg deviates laterally) suggests the presence of approximately 25 degrees of medial torsion. Children with spastic cerebral palsy exhibit increased femoral antetorsion,[//] as do many children with myelomeningocele.[19,73] These children typically show persistent hip flexion contractures, with a concurrent lack of adequate extension and lateral rotation forces crossing the proximal femoral shaft.[1,3,38] Habitual sleep and play postures, such as W-sitting, correlate with persistence of femoral antetorsion.[19,41,43,74,75] Staheli[45] has described a malalignment syndrome that causes awkward gait and chondromalacia in children with normal neuromotor function. This syndrome is characterized by increased femoral antetorsion, which persists into adolescence with compensatory, excessive lateral tibiofibular torsion. Correction would require femoral and tibial derotational osteotomy. Because the most rapid decline in femoral antetorsion normally occurs during the first 4 years postnatally,[70] we advise the pediatric clinician to intervene early to gain biomechanically efficient limb alignment before this malalignment syndrome evolves in children with neuromotor deficits. No radiologic or anatomic studies have documented the effects of twister cables, antirotation braces, exercises, orthoses, splints, or shoes on the existing degree of femoral or tibial torsion.[5,25,38,41,47,57,74,76] Twister cables are not recommended for femoral antetorsion, as they may promote excessive lateral tibiofibular torsion or rotation.[38] Until research proves that exercise, positioning, and orthoses offer no benefit, we base our management recommendations on the principle that the application of correct forces is required for optimum skeletal modeling before the skeleton ossifies[6,11,12] and on findings relating habitual limb positioning to torsional abnormality. For children with adequate innervation, from birth to 8 years of age, we recommend multiple daily repetitions of resisted hip extension and lateral rotation, the reduction of hip and knee flexion and femoral medial rotation postures in stance and gait, the improvement of lateral and posterior weight transfer skills over the feet, and the institution of a consistent positioning program. Because abnormal foot pronation deformity is believed to promote medial rotation of the proximal limb structures,[52,53] we protect the foot from pronating abnormally by serial casting to gain full ankle dorsiflexion mobility (minimum of 10 [degrees]) and by combining the consistent use of appropriate splints or orthoses with weight transfer training. The general opinion among orthopedists is that femoral derotation osteotomy is the only effective treatment for antetorsion for children aged 7 years and older.[5,7,19,39,63] Tachdjian[19] and Staheli[45] discuss criteria for osteotomy for the neurologically normal child. Hoffer et al[7] report good results and minimal complications following supracondylar femoral derotational osteotomy, undertaken for 11 children with spastic cerebral palsy, aged 7 to 15 years (X [bar] = 9.5).

Test 4--hip rotation mobility test. Another way to assess femoral torsional status and to evaluate soft tissue extensibility is to measure passive hip medial rotation and lateral rotation (Fig. 2). Procedure. The therapist positions the child prone with knees flexed 90 degrees and tibias vertical. Keeping the pelvis level, the therapist allows the feet to fall apart from each other until gravity stops the motion. Range of motion (ROM) is measured using the same DRA landmarks as described for the Ryder's test (see test 3 and Fig. 5). To measure lateral rotation mobility, the pelvis is gently kept level, and the therapist laterally rotates each hip 10 to 20 times to reduce any influence of soft tissue restriction.[19] The therapist identifies soft tissue effects on rotation by noting whether the magnitude of ROM changes with the hip and knee flexed 90 degrees (sitting).[19,25] Evidence of abnormal femoral torsion is supported when the same findings are obtained with the hip flexed and extended.[25] Norms. Lateral rotation contracture normally limits medial rotation mobility in infancy.[35,77] As the contracture reduces, medial rotation increases. At age 24 months, medial rotation exceeds lateral rotation by an average of 7 degrees.[77] A normative study of medial rotation findings for school-aged children showed a mean approximating 40 to 50 degrees (range = 15 [degrees] - 65 [degrees]), whereas the mean lateral rotation value approached 70 degrees before age 1 year and decreased to approximately 45 degrees (range = 25 [degrees] - 65 [degrees]) from age 5 years through mid-adulthood. This study examined between 11 and 28 children in age groups of 1 year each greater than 1 year.[61] Limitations. Reliability studies for measurements obtained from medial rotation/lateral rotation tests on children have not been published, nor have medial rotation/lateral rotation findings been validated using CT or MRI analysis of torsion. This test cannot be used to identify abnormal femoral torsion in children less than age 2 years because of persistence of normal intrauterine lateral rotation contracture of the hip joint.[35,40,77] If lax ligaments allow abnormally increased medial rotation and lateral rotation mobility, the findings cannot be used to discern torsional status. We believe the medial rotation/lateral rotation test cannot be used to distinguish femoral torsion from abnormal acetabular anteversion or retroversion. We use Ryder's test as an adjunct to the medial rotation/lateral rotation test, and we repeat the medial rotation/lateral rotation test in the sitting position. Clinical implications. Clinically significant medial femoral torsion is evident when medial rotation is greater than 60 degrees and lateral rotation is less than 25 degrees.[38,44,45] Limitation of lateral rotation range is the principal finding.[19] If medial rotation/lateral rotation findings corroborate those obtained with the Ryder's test, we try to manage torsional deformity as suggested for the Ryder's test (see test 3). If soft tissue limitation becomes evident, we use positioning, soft tissue mobilization techniques, and exercises to gain mobility.

Distal Limb Features

If the hindfoot FPA is negative and the patella aligns on the sagittal plane, we look for a medial genicular position, excessive activity or shortening of the medial hamstring and popliteus muscles, medial tibiofibular torsion, or a combination of these factors (see tests 5-8).[25,42,50,54,78] If the forefoot FPA is negative, we look for evidence of increased talar torsion,[38] foot supination, or forefoot adductus (see tests 5, 6, and 9). If the patella aligns either medially or normally and the hindfoot FPA is abnormally positive, we look for a lateral genicular position or excessive lateral tibiofibular torsion, or both (see tests 5-8). If a line bisecting the forefoot deviates laterally on the hindfoot, we evaluate for foot pronation, which is characterized by midtarsal joint dorsiflexion and abduction (see test 9).

Test 5--thigh-foot angle test. Staheli[44,45,57] uses this test to assess tibiofibular torsional status. We suggest instead that it is a composite test of axial tibiofibular rotation position, tibiofibular torsion, subtalar joint alignment, and foot configuration (ie, adduction, supination, pronation, and so forth). Procedure. The therapist positions the child prone with the knee flexed 90 degrees, the tibia vertical, and the ankle dorsiflexed to 0 degrees. As this test is used to evaluate structures distal to the femur, we believe the position of proximal femoral rotation or hip abduction is not relevant. We hold the foot in the subtalar joint neutral position (neither supinated nor pronated) and use the bisection of the plantar surface of the hindfoot only as the DRA to eliminate forefoot deviations as a factor in measurement. We measure the angle formed by the bisecting line of the thigh (PRA) and the longitudinal axis of the hindfoot (DRA) (Fig. 2). Medial deviation (adduction) of the foot results in a negative value; lateral deviation (abduction) produces a positive value. Norms. The range of normal findings is wide, especially during infancy.[61] Through age 2 years, the mean angle lies between 0 and -- 10 degrees (range = -25 [degrees] -20 [degrees] at birth). The angle gradually increases to a mean of 10 degrees (range = -5 [degrees] -30 [degrees]) from middle childhood on.[61] Limitations. Normative clinical data represent small age-specific study groups of 11 to 28 children.[61] Reliability data are not available for this test, nor are validity data available for comparisons of thigh-foot angles and tibiofibular torsion CT or MRI findings for children of more than 9 months' gestational age.[49] Age-related normative data for CT and MRI findings are also not available. The thigh-foot angle results might reveal factors that falsely indicate the presence of medial tibiofibular torsion, such as normal infantile hyperextensibility of the knee joint ligaments and capsule, which interferes with stabilizing and identifying the rotary position of the proximal tibia[42]; normal neonatal medial rotation bias of the knee joint ligaments and capsule[79]; medial hamstring and popliteus muscle activity or shortening[25]; release of the screwhome mechanism with knee flexion, which allows medial axial tibiofibular rotation to occur[25,66,80,81]; or persistence of medial genicular position attributable to habitual sleep or play postures. Talar torsion or medial rotation of the talar body in the ankle mortise might position the well-aligned foot in medial rotation under the tibia and fibula.[49,50] We believe this factor accounts primarily for the failure of Badelon et al[49] to find a correlation between the thigh-foot angle test and the status of anatomic tibiofibular torsion in fetal cadaver specimens. Clinical implications. Staheli[57] states that, if the thigh-foot angle is greater than -5 degrees (child's age unspecified), medial tibial torsion is present. We use the thigh-foot angle as a global indicator of change rather than as a reliable or valid indicator of tibiofibular torsion. We supplement this test with the axial tibiofibular rotation test (see test 8), the TMA-thigh angle test (see test 6), and examination of tibiofibular torsion status with the knee joint extended (see test 7) before proposing a management program.

Test 6--the transmalleolar axis-thigh angle test. The TMA-thigh angle moves the scope of assessment to the segment proximal to the subtalar joint.[9,19,45,55,61] Procedure. The therapist positions the child prone with the knee and ankle flexed 90 degrees and the tibia vertical. We locate the anterior-posterior (AP) malleolar bisections and mark a line joining them on the plantar surface of the foot, representing the TMA. We also mark a perpendicular line from the TMA posteriorly on the heel. This perpendicular line is the DRA. The PRA is the long bisection of the posterior thigh. We measure the angle formed by the PRA and DRA (Fig. 2). Norms. The normative values for TMA-thigh angle are higher than those for thigh-foot angle by approximately 10 degrees.[61] The mean TMA-thigh angle at birth is about 0 degrees (range = -30 [degrees] -20 [degrees]), gradually increasing to approximately 20 degrees (range = 0 [degrees] -45 [degrees]) during and after middle childhood.[61] In our opinion, these normative values, when compared with those for thigh-foot angle, reveal evidence of a normal medial rotation of the foot under the malleoli. Limitations. The same reliability and validity issues discussed in reference to the thigh-foot angle test pertain to the TMA-thigh angle test. "Pseudolack of malleolar torsion" describes a low orientation of the TMA to the frontal plane that might falsely suggest medial tibiofibular torsion.(#) Although a rather high incidence (30%) of toe-in gait has been reported for infants without neuromotor dysfunction,[48] Rosen and Sandick found "a surprising lack of true [internal] TF torsion"[23](p854) among 2-year-old children who exhibit a toe-in gait. Clinical implications. We advise using the TMA-thigh angle to measure change, rather than to indicate tibiofibular torsion in children with neuromotor deficits. We evaluate axial tibiofibular rotation and torsion with the knee extended before intervening. Computed tomography provides the most accurate impression of torsional status, but age-related normative data are not yet available.

Test 7--tibiofibular torsion test in knee extension. Tibiofibular torsion is measured clinically as the angle (DRA) formed by the intersection of a frontal-plane bisection of the knee joint (PRA) and a line bisecting the medial and lateral malleoli (TMA). On CT scans, the frontal-plane bisection of the proximal tibial condyles forms the PRA (Fig. 6). After aligning the PRA on the frontal plane, medial tibiofibular torsion is evident when the medial malleolus lies posterior to the frontal plane. Lateral tibiofibular torsion is evident when the lateral malleolus lies posterior to the frontal plane. The collateral and cruciate ligaments and the popliteus muscle become taut with the knee extended, locking the fully extended mature knee joint against transverse- and frontal-plane motions.[42,83,84] Given that the knee joint ligaments are intact and not hyperextensible, and that the extensibility of the knee joint capsule allows full knee extension to occur, we believe that extending the knee joint to measure tibiofibular torsional status secures the proximal tibia against rotating and reduces measurement error. Procedure. The therapist positions the child sitting or supine on a firm table surface with the knee fully extended and the ankle and foot hanging off the table edge. The table surface represents the frontal plane of the limb and serves as the PRA for this test. The therapist then places the knee joint axis on the frontal plane by flexing the knee a few times in the pure sagittal plane. An assistant supports the knee at 0 degrees of extension, stabilizing it against rotating. We rotate the tibiofibular unit medially and laterally under the stabilized distal femur. If the joint allows more than 10 degrees of rotation, we carefully align the tibial tuberosity and tibial crest on the sagittal plane and have the assistant stabilize the tibia. The assistant then dorsiflexes the ankle on the sagittal plane to the comfortable end-range and holds the ankle in that position. We mark a dot on the AP bisection of the medial and lateral malleoli ([mb.sup.1] and [mb.sup.2]). We also measure the distance (X and Y) between each malleolar bisection (mb) and the table surface. Holding a medial-lateral caliper on both the transverse and frontal planes, we measure the width (W) between the malleoli (Fig. 7). All three measurements are transferred onto graph paper, using W as the baseline and setting X and Y at each end of the baseline (Fig. 8). We draw a horizontal line ([W.sup.1]) from the top of the shorter mb line to the longer mb line, parallel to the baseline. Another line (Z), connecting the dots at [mb.sup.1] and [mb.sup.2] and representing the TMA, is drawn. A goniometer is used to measure the angle formed by Z and [W.sup.1]. We believe lateral tibiofibular torsion is evident when the medial mb line is longer than the lateral mb line and is recorded as a positive value. Medial tibiofibular torsion occurs when the lateral mb line is longer than the medial mb line and is recorded as a negative value. Norms. Existing normative values vary in magnitude, having been obtained using a variety of anatomical landmarks and knee joint positions.[23,48,85-89] We refer the reader to cited references for details. Because McCrea[25] measures tibiofibular torsional status with the knee joint extended, we report his stated normative values, as follows: -5 to 10 degrees in infancy, 10 to 15 degrees by age 2 years, and 20 to 30 degrees by age 5 to 7 years through adulthood. Jakob et al[89] found a mean tibiofibular torsion of 30 degrees in adult cadavers, measuring with the knee joint extended. Badelon et al[47] assessed fetal cadavers and found a mean anatomical angle of 20 degrees (range = 10 [degrees] - 32 [degrees]) at 9 months' gestational age, suggesting that the tibiofibular unit might achieve a mature magnitude of lateral torsion at birth. If so, the reported normative values for infants and very young children may be affected by medial genicular position. Limitations. Reliability coefficients and age-related, normative values for the tibiofibular torsion assessment procedure described are not established. Age-related normative values do not exist for CT or MRI measurements of tibiofibular torsion. If the knee joint cannot be fully extended, the screw-home mechanism cannot operate to secure the proximal tibia. If the lateral malleolus is abnormally enlarged, as often occurs with supination deformity, the width of the ankle might be misrepresented. Clinical implications. If true medial tibiofibular torsion is present in young children, it usually improves spontaneously, whereas lateral tibiofibular torsion often worsens.[45] Valmassey (RL Valmassey; personal communication; June 28, 1991) and various authors[25,40,43,90] suggest intervening early to reduce a significant, negative FPA, using serial casting, night splinting bars, and counterrotation devices. The criteria for using such interventions, however, are not specified. Radiologic studies are needed to ascertain the efficacy of these interventions. It is likely that these interventions actually impose lateral rotation on soft tissues within the knee joint, as they do not secure the proximal tibia.[43,47] The immediate and long-term biomechanical influences of these interventions on tibiofibular torsion and knee joint status have not been investigated.[45,47] Staheli[45,58] has reported suggested guidelines for surgical osteotomy for children with normal neuromotor status. For children with cerebral palsy, abnormal lateral tibiofibular torsion may be a greater biomechanical problem than medial tibiofibular torsion. Ounpuu et al[65] recommend supramalleolar derotational osteotomy for children with cerebral palsy who show evidence of increased tibiofibular lateral torsion. They claim results that show improved sagittal-plane gait features at the knee and ankle after reducing the mean FPA from 31.8 to 11.5 degrees.

Test 8--axial tibiofibular rotation test. We believe this test helps to distinguish axial rotation mobility and alignment from tibiofibular torsional status. Maximum axial rotation occurs with the knee joint flexed between 60 and 90 degrees.[91] Infants normally exhibit 20 to 30 degrees of rotary mobility with the knee joint extended.[42,92] Procedure. The therapist positions the child prone with the knee flexed 90 degrees, the tibia vertical, and the ankle flexed to 0 degrees. Refer to the illustration for thigh-foot angle in Figure 2 for landmarks of measurement. Holding the foot and malleoli as a stable unit, we rotate the leg, ankle, and foot medially to the maximum comfortable end-range. We align the arms of a small goniometer on the longitudinal heel bisection and parallel to the longitudinal axis of the thigh, with the axis posterior to the heel border. We measure the resulting angle, rotate the leg and foot laterally, and repeat the measurement. We suggest repeating the measurement with the hips flexed over the edge of the mat table and the knees flexed on a bench. McCrea[25] describes a similar measurement procedure in which the child is positioned sitting and supine; however, we believe the reference axes are more difficult to ascertain with this procedure, thus impeding potential measurement replicability. Norms. McCrea's[25] clinical experience suggests that children without neuromotor dysfunction under age 3 years, positioned sitting with hips and knees flexed 90 degrees, show a normal medial rotation-to-lateral rotation ratio of between 1:1 and 1:2, with normal findings of 45 to 65 degrees for medial rotation and 45 to 95 degrees for lateral rotation.[25] By age 3 to 4 years, the knee joint typically gains integrity, and rotation range decreases.[42,92] An unpublished study, lacking specification of reference axes, revealed that the mean total range of axial tibiofibular rotation in women without neuromotor dysfunction approached 40 degrees (SD=8 [degrees]), with a lateral rotation-to-medial rotation ratio of approximately 2:1 Axial tibial rotation with the knee extended was 0 degrees.[93] Other authors report unsubstantiated values ranging from 10 degrees of medial and lateral rotation[80,81] to 30 degrees of medial rotation and 45 degrees of lateral rotation.[79,94] Limitations. Reliability studies and studies to determine age-specific pediatric normative values have not been undertaken for this test. Clinical judgment must rest on the merits of reported clinical experience. Correlation has not been established between axial tibiofibular rotation mobility and FPA or between axial tibiofibular rotation and foot deformity. Clinical implications. A medial genicular position is evident when medial rotation exceeds lateral rotation. The medial rotation bias can be caused by soft tissues or medial tibiofibular torsion, or both.[25] If lateral rotation mobility diminishes in hip flexion, then consider shortened medial hamstring muscles to be a limiting factor. Clinical findings often reveal a medial genicular position in preschool-aged children with spastic diplegia, particularly when the medial hamstring muscles are overactive and shortened and the child habitually kneel-sits with legs rotated medially. The same problem often occurs in children with meningomyelocele who can voluntarily control only the medial (and not the lateral) hamstring muscles. We believe early positioning and night splinting to maintain full medial hamstring muscle ROM and to gently rotate the leg laterally under the flexed knee may reduce the functional problems relating to this medial rotational bias. Gait studies and radiographic assessments are needed to evaluate the structural and functional effects of these interventions. Medial hamstring muscle lengthening has been shown to produce a significant and perhaps pathological increase in FPA in children with spastic cerebral palsy over age 4 years.[3,44] The resulting stability of the tibiofemoral articulation was not evaluated. We find that children with hemiparesis who exhibit equinovarus deformity at the foot and ankle often reveal increased lateral axial tibiofibular mobility. The rigid, supinating foot imposes lateral rotation forces on the tibiofibular unit during the stance phase of gait. We suggest reducing this deforming strain by gaining dorsiflexion and pronation mobility through progressive casting, use of night splints and orthoses, or, if casting yields no gain of mobility, surgical intervention.[1,95]

Test 9--foot configuration. Consider the shape of the foot as a factor affecting FPA when proximal factors are normal. Procedure. The therapist positions the child prone to allow a view of the plantar surface of the foot. We stabilize the foot in the subtalar neutral position of maximum congruity (ie, neither supinated nor pronated) and view the plantar surface. We align the arms of a small goniometer on the longitudinal bisection of the heel and parallel with a line bisecting the second and third metatarsals. Norms. The ideal hindfoot/forefoot angle is 0 degrees.[53,59,88,96] Limitations. Reliability and validity tests are needed for this procedure. Age-related normative values have not been established. Clinical implications. Metatarsus adductus and supination or pronation deformities affect the FPA.[45] A straight foot positioned in adduction (medial rotation) under a normally aligned femur and TMA suggests the presence of increased talar torsion or medial rotation of the talar body within the ankle mortise.[38,49,50] In young children with spastic diplegia who bear excessive weight on the medial forefoot, we often see foot pronation deformity, characterized by an abducted forefoot on an adducted hindfoot. The lateral forefoot deviation masks a negative hindfoot FPA. In this case, we believe that when the foot is supported in proper alignment with an orthosis, toeing-in appears or worsens. Clinicians should alert caretakers to this possibility prior to intervening to protect the foot pronation deformity from progressing, as caretakers usually perceive toeing-in as a more serious problem than progressive foot pronation. As children with spastic diplegia grow beyond preschool age, some develop lateral tibiofibular rotation and torsion as well as foot pronation, which increases the hindfoot FPA.[27] The medial knee joint ligaments and capsule become overstretched by abnormal loading forces.[19,27,38,53,54,78] Having identified the presence of shortened triceps surae muscles, we work to reduce foot pronation and strain on the knee joint and tibiofibular unit by (1) gaining ankle dorsiflexion mobility with serial casting or stretch splinting, or surgery if these measures fail; (2) applying appropriate orthoses or splints to maintain ideal foot alignment; and (3) facilitating lateral/posterior weight transfers over the foot.[1,95,97] Because the forefoot adducts on the hindfoot in individuals with foot supination deformity, the hindfoot FPA might be more positive than that of the forefoot. We reduce the supination deformity with serial casting and the use of orthoses as needed to achieve a more normal FPA.[1,95,97]

Summary

The field of biomechanical analysis as it applies to children with neuromotor deficits is clearly in its infancy. The clinical assessment procedures presented in this article require further research to establish their validity and reliability for children with and without neuromotor dysfunction. Large-scale normative studies should follow, recording coincidental measures of the FPA. Sophisticated radiographic techniques should be used to determine the efficacy of commonly prescribed splinting apparatuses and therapeutic exercise programs. Such studies would contribute greatly to our understanding of the functional significance and management of transverse-plane biomechanical factors for children with neuromotor dysfunction.

[double dagger]Macklanburg Duncan Inc, Oklahoma City, OK 73118. [S]Model AF100, Dasco Pro Inc, 2215 Kishawaukee St, Rockford, IL 61101. [//]References 2, 5, 7, 9, 10, 13, 14, 18, 41, 70.

PHOTO : Figure 1. Malalignment, characterized by medially rotated patellae (left more than right), positive foot progression angle, and bilateral foot pronation.

PHOTO : Figure 2. Stabeli's "Rotational Profile": (A) foot progression angle; (B) hip lateral rotation; (C) hip medial rotation; (D) thigh-foot angle; (E) transmalleolar axis-thigh angle.

PHOTO : Figure 3. Proximal and distal reference axes for femur of an adolescent (unshaded) and an adult (shaded): (A) normal mean antetorsion (12 [degrees] - 20 [degrees] ); (B) abnormal or immature antetorsion (>20 [degrees] ); (C) femoral retrotorsion (<12 [degrees] antetorsion).

PHOTO : Figure 4. Skeletal representation of Ryder's test in context of increased left medial femoral torsion: (A) When the distal reference axis (DRA) of the tibia is aligned on the sagittal plane, the greater trochanter lies posterior to the frontal plane; (B) when the proximal reference axis of the greater trochanter and the femoral neck is aligned on then frontal plane, the distal DRA of the tibia deviates laterally to represent a position of hip medial rotation. (V-angle of femoral torsion.)

PHOTO : Figure 5. Ryder's test for femoral torsion. Palpate the greater trochanter and align it in its most lateral position. Use the adapted angle finder to measure hip rotation. Landmarks mid-patella and mid-ankle.

PHOTO : Figure 6. (A) Proximal and distal reference axes for tibiofibular (TF) torsion; (B) medial TF torsion; (C) lateral TF torsion.

PHOTO : Figure 7. Measure the width between the malleolar bisections by carefully aligning the measuring device (photo shows a medial-lateral caliper) on both the transverse and frontal planes. Do not take a diagonal width measurement.

PHOTO : Figure 8. Calibrating tibiofemoral torsion. (X=distance between malleolar bisection ([mb.sup.1]) and surface. Y=distance between malleolar disection ([mb.sup.2])and surface. W=width between malleolar bisections measured parallel with the transverse and frontal planes. [W.sup.1] represents the frontal plane, connects [md.sup.1] and Y, and lies parallel tol line W. Z=transmalleolar axis (TMA), connecting [mb.sup.1] and [mb.sup.2]. The angle formed by the intersection of line [W.sup.1] and Z indicates that the TMA is oriented 20 [degrees] to the frontal plane.)

Reference

[1]Cusick BD. Progressive Casting and Splinting for Lower Extremity Deformities in Children with Neuromotor Dysfunction. Tucson, Ariz: Therapy Skill Builders; 1990. [2]Beals RK. Developmental changes in the femur and acetabulum in spastic paraplegia and diplegia. Dev Med Child Neurol. 1969;11:303-313. [3]Bleck EE. Orthopedic Management of Cerebral Palsy. Philadelphia, Pa: JB Lippincott Co; 1987. [4]Chong KC, Vojnic CD, Quanburg AO, Letts RM. The assessment of internal rotation gait in cerebral palsy: an electromyographic gait analysis. Clin Orthop. 1978;132:145-149. [5]Fabry G, MacEwen GD, Shands AR. Torsion of the femur. J Bone Joint Surg [Am]. 1973;55:1726-1738. [6]Frost HM. Intermediary Organization of the Skeleton. Boca Raton, Fla: CRC Press; 1986. [7]Hoffer MM, Prietto C, Koffman M. Supracondylar derotational osteotomy of the femur for internal rotation of the thigh in the cerebral palsied child. J Bone Joint Surg [Am]. 1981;63:389-393. [8]Kasser JR, MacEWen GD. Examination of the cerebral palsy patient with foot and ankle problems. Foot Ankle. 1983;4:135-144. [9]King HA, Staheli LT. Torsional problems in cerebral palsy. Foot Ankle. 1984;4:180-184. [10]Laplaza FJ, Root L, Tassanawipas A. Burke S. Femoral Torsion and neck-shaft angles in patients with cerebral palsy. Dev Med Child Neurol. 1989;31:27. Abstract. [11]LeVeau BF. Developmental biomechanics and the principles of growth. In: Wilson JM, Davis LA, eds. Orthopedic Aspects of Developmental Disabilities. Chapel Hill, NC; University of North Carolina Press; 1980:1-27. [12]LeVeau BF, Bernhardt DB. Developmental biomechanics: effect of forces on the growth, development, and maintenance of the human body. Phys Ther. 1984;64:1874-1882. [13]Lewis FR, Samilson RR, Lucas DB. Femoral torsion and coxa valga in cerebral palsy. Dev Med Child Neurol. 1964;6:591-597. [14]Majestro TC, Frost HM. Cerebral palsy: spastic internal femoral torsion. Clin Orthop. 1971;79:44-56. [15]Mann R. Biomechanics in cerebral palsy. Foot Ankle. 1983;4:114-119. [16]Rang M, Silver R, de la Garza J. Cerebral palsy. In: Lovell WW, Winters RB, ends. Pediatric Orthopaedics. 2nd ed. Philadelphia, Pa: JB Lippincott Co; 1986;345-396. [17]Samilson RL, Dillin L. Postural impositions on the foot and ankle from trunk, pelvis, hip, and knee in cerebral palsy. Foot Ankle. 1983;4:120-127. [18]Staheli LT, Duncan WR, Schaefer E. Growth alterations in the hemiplegic child. Clin Orthop. 1968;60:205-212. [19]Tachdjian MO. Pediatric Orthopedics. 2nd ed. Philadelphia, Pa: WB Saunders Co; 1990. [20]Hoffer MM, Perry J. Pathodynamics of gait alterations in cerebral palsy and the significance of kinetic electromyography in evaluating foot and ankle problems. Foot Ankle. 1983;4:128-134. [21]Jones ET, Knapp DR. Assessment and management of the lower extremity in cerebral palsy. Orthop Clin North Am. 1987;18:725-738. [22]Cole TM, Tobis JS. Measurement of musculoskeletal function. In: Kottke FJ, Lehmann JF, eds. Krusen's Handbook of Physical Medicine and Rehabilitation. 4th ed. Philadelphia, Pa: WB Saunders Co; 1990;20-71. [23]Rosen H, Sandick H. the management of tibiofibular torsion. J. Bone Joint Surg [Am]. 1955;37:847-855. [24]Dorland's Illustrated Medical Dictionary. 27th ed. Philadelphia, Pa: WB Saunders Co; 1988:1830. [25]McCrea JD. Pediatric Orthopedics of the Lower Extremity: An Instructional Handbook. Mount Kisco, NY: Futura Press; 1985. [26]Smith C, Fuller E. Biomechanics III--Syllabus.San Francisco, Calif: California College of Podiatric Medicine. [27]Feltz WL. The prenatal development of the human femur. Am J Anat. 1954;94:1-44. [28]Lloyd-Roberts GC, Harris NH, Chrispin AR. Anteversion of the acetabulum in congenital dislocation of the hip: a preliminary report. Orthop Clin North Am. 1978;9:89-95. [29]McKibbin B. Anatomical factors in the stability of the hip joint in the newborn. J Bone Joint Surg [Br]. 1970;52:148-159. [30]McSweeney A. A study of femoral torsion in children. J Bone Joint Surg [Br]. 1971;53:90-95. [31]Stanisavljevic S. Congenital dysplasia, subluxation, and dislocation of the hip in stillborn and newborn infants. J Bone Joint Surg [Am]. 1963;45:1147-1158. [32]Watanabe RS. Embryology of the human hip. Clin Orthop. 1974;98:1-26. [33]Engel GM, Staheli LT. Natural history of torsion and other factors influencing gait in childhood. Clin Orthop. 1974;99:12-17. [34]Hlavac HF. Major considerations in the clinical evaluation of the lower extremity. In: Sgarlato T, ed. A Compendium of Podiatric Biomechanics. San Francisco, Calif: California College of Podiatric Medicine; 1971:321-344. [35]Pitkow RB. External rotation contracture of the extended hip. Clin Orthop. 1975;110:139-145. [36]Gage JR. An overview of normal and cerebral palsy gait. Neurosurgery: State of the ARt Reviews. 1989;4:379-401. [37]Alexander I. Evaluation of in-toeing in children. In: The Foot: Examination and Diagnosis. New York, NY: Churchill Livingstone Inc; 1990:149-160. [38]Bleck EE. Developmental orthopaedics, III: toddlers. Dev Med Child Neurol. 1982;24:533-555. [39]Sutherland DH, Schottstaedt ER, Larsen Lj, et al. Clinical and electromyographic study of seven spastic children with internal rotation gait. J Bone Joint Surg [Am]. 1969;51:1070-1082. [40]Sutherland DH. Gait Disorders in Childhood and Adolescence. Baltimore, Md: Williams & Wilkins; 1984. [41]Swanson AB, Greene PW, Allis HD. Rotational deformities of the lower extremity in children and their clinical significance. Clin Orthop. 1963;27;157-174. [42]Valmassy RL. Biomechanical evaluation of the child. In: Ganley JV, ed. Symposium on Podopediatrics. Philadelphia, Pa: WB Saunders Co; 1984;563-579. [43]McDonough MW. Angular and axial deformities of the legs of children. In: Ganley JV, ed. Sumposium on Podopediatrics. Philadelphia, Pa: WB Saunders Co; 1984:601-620. [44]Staheli LT. Torsional deformity. Pediatr Clin North Am. 1977;24:799-811. [45]Staheli LT. The lower limb. In: Morrissy RT, ed. Lovell and Winter's Pediatric Orthopaedics. 3rd ed. Philadelphia, Pa: JB Lippincott Co; 1990;1:741-766. [46]Sullivan RC, Krzyzanowski JA, Harris GF, Rotational changes in the lower limbs following distal medial hamstrings surgery for cerebral-palsied children. Dev Med Child Neurol. 1986;28:107. Abstract. [47]Scoles PV. Lower extremity development. In: Scoles PV. Pediatric Orthopedics in Clinical Practice. 2nd ed. Chicago, Ill: Year Book Medical Publishers; 1988:82-121. [48]Hutter CG, Scott W. Tibial torsion, J Bone Joint Surg [Am]. 1949;31:511-518. [49]Badelon O, Bensahel H, Folinais D, Lassale B. Tibiofibular torsion from the fetal period until birth. J Pediatr Orthop. 1989;9:169-173. [50]Ritter MA, DeRosa GP, Babcock JL. Tibial torsion? Clin Orthop. 1976;120:159-163. [51]Jordan RP, Cusack J, Resseque B. Foot function and its relationship to posture in the pediatric patient with cerebral palsy and other neuromotor disorders: course materials and lecture notes. Symposium, presented by Langer Biomedical Group, sponsored by the Neurodevelopmental Treatment Association; May 20-22, 1983; New York, NY. [52]Oatis CA. Biomechanics of the foot and ankle under static conditions. Phys Ther. 1988;68:1815-1821. [53]Root ML, Orien WP, Weed JH. Normal and Abnormal Biomechanics of the Foot: Clinical Biomechanics. Los Angeles, Calif: Clinical Biomechanics Corp; 1977: vol 2. [54]Kumar SJ, MacEwen GD. Torsional abnormalities in children's lower extremities. Orthop Clin North Am. 1982;13:629-639. [55]Staheli LT. Torsional deformity. Pediatr Clin North Am. 1986;33:1373-1381. [56]Staheli LT, Engel GM. Tibial torsion. Clin Orthop. 1972;86:183-186. [57]Staheli LT. Rotational problems of the lower extremities. Orthop Clin North Am. 1987;18:503-512. [58]Staheli LT. Torsion: treatment indications. Clin Orthop. 1989;247:61-66. [59]Aharonson Z, Voloshin A, Steinbach TV et al. Normal foot-ground pressure pattern in children. Clin Orthop. 1980;150:220-223. [60]Scrutton D. Footprint sequences of normal children under five years. Dev Med Child Neurol. 1969;11:44-53. [61]Staheli LT, Corbett M, Wyss C, King H. Lower extremity rotational problems in children. J Bone Joint Surg [Am]. 1985;67:39-44. [63]Sgarlato T. A Compendium of Podiatric Biomechanics. San Francisco, Calif: California Collage of Podiatric Medicine; 1971:311. [63]Staheli LT, Clawson DK, Hubbard DD. Medial femoral torsion: experience with operative treatment. Clin Orthop. 1980;1546:222-225. [64]Sutherland DH, Olshen R, Cooper A, Dale D. The development of mature gait. J Bone Joint Surg [Am]. 1980;62:336-353. [65]Ounpuu S, Andrews M, Gage JR. The effects of internal tibial derotation osteotomy on the gait of children with cerebral palsy. Dev Med Child Neurol. 1989;31(suppl 59):8. Abstract. [66]Hoppenfeld S. Physical Examination of the Spine and Extremities. East Norwalk, Conn: Appleton-Century-Crofts; 1976:144, 186-189. [67]Dega W. Richerche anatomiche e meccaniche sull'anca fetale rivolte a chiarire l'etiologia e la patogenisi della lussazione congenita. Chir Organi di Movimento. 1933;18:425-505. [68]Ogden JA. Skeletal Injury in the Child. Philadelphia, Pa: Lea & Febiger; 1982:16-40. [69]Forero N, Okamura LA, Larson MA. Normal ranges of hip motion in neonates. J Pediatr Orthop. 1989;9:391-395. [70]Shands AR, Steele MK. Torsion of the femur: a follow-up report of the use of the Dunlap method for its determination. J Bone Joint Surg [Am]. 1958;40:803-816. [71]Stuberg WA, Koehler A, Wichita M, et al. Comparison of femoral torsion assessment using goniometry and computerized tomography. Pediatric Physical Therapy. 1989;1(3):115-118. [72]Murphy SB, Simon SR, Kijewski PK, et al. Femoral anteversion. J Bone Joint Surg [Am]. 1987;69:1169-1176. [73]Carroll NC. Assessment and management of the lower extremity in myelodysplasia. Orthop Clin North Am. 1987;18:709-724. [74]Crane L. Femoral torsion and its relation to toeing-in and toeing-out. J Bone Joint Surg [Am]. 1959;41:421-428. [75]Knight RA. Developmental deformities of the lower extremities. J Bone Joint Surg [Am]. 1954;36:521-526. [76]Knittel G, Staheli LT. The effectiveness of shoe modifications for intoeing. Orthop Clin North Am. 1976;7:1019-1025. [77]Phelps E, Smith LJ, Hallum A. Normal ranges of hip motion of infants between nine and 24 months of age. Dev Med Child Neurol. 1985;27:785-792. [78]Kling TF. Toe in, toe out. Emer Med. March 1986:32-46. [79]Wilkins KE. Bowlegs. Pediatr Clin North Am. 1986;33:1429-1438. [80]Grimsby O. Fundamentals of Manual Therapy: A Course Workbook. 2nd ed. Vagsbygd, Norway: Sorlandets Fysikalske Institutt A/S; 1980. [81]Schafer RC. Clinical Biomechanics: Musculoskeletal Actions and Reactions. 2nd ed. New York, NY: Marcel Dekker Inc; 1987. [82]Tax HR. Podopediatrics, 2nd ed. Baltimore, Md: Williams & Wilkins; 1985. [83]Kisner C, Colby LA. The knee. In: Kisner C, Colby LA. Therapeutic Exercise Foundations and Techniques. 2nd ed. Philadelphia, Pa: FA Davis Co; 1990:345-356. [84]Warren R, Arnoczky SP, Wickiewicz TL. Anatomy of the knee. In: Nicholas JA, Hershman EB, eds. The Lower Extremity and Spine in Sports Medicine. St Louis, Mo: CV Mosby Co; 1986:657-694. [85]Le Damany P. La torsion des tibias, normale, pathologique, experimentale. J de l'Anat Phys. 1909;45:598-615. [86]Wynne-Davies R. Talipes equinovarus: a review of eighty-four cases after completion of treatment. J Bone Joint Surg [Br]. 1964;46:464-476. [87]Dupius PV. La torion tibiale. In: Son Interet Clinique: Radiologique et Chirurgical. Paris, France: Masson et Cie; 1951:3-68. [88]Root ML, Orien WP, Weed JH, Hughes RJ. Biomechanical Examination of the Foot. Los Angeles, Calif: Clinical biomechanics Corp; 1971: vol 1. [89]Jakob RP, Haertel M, Stussi E. Tibial torsion calculated by computerized tomography and compared to other methods of measurement. J Bone Joint Surg [Br]. 1980;62:238-242. [90]McCollough MC. Orthotic management. In: Lovell WW, Winter RB, eds. Pediatric Orthopaedics. 2nd ed. Philadelphia, Pa: JB Lippincott Co; 1986:1038-1039. [91]Larson RL. Physical examination in the diagnosis of rotary instability. Clin Orthop. 1983;172:38-44. [92]Valmassy RL. Normal and Abnormal Development of the Lower Extremity in the Infant: Treatment of Common Pediatric Musculoskeletal Disorders--Course Syllabus. San Francisco, Calif: California College of Podiatric Medicine; 1990. [93]Mossberg K, Smith LK. Axial rotation of the knee in adult females. Cited in: Lehmkuhl LD, Smith LK. Brunnstrom's Clinical Kinesiology, 4th ed. Philadelphia, Pa: FA Davis Co; 1983:292. [94]Weber D, Agro M, eds. Clinical Aspects of Lower Extremity Orthotics. Oakville, Ontario, Canada: Elgan Enterprises; 1990. [95]Cusick BD. Serial Casts: Their Use in the Management of Spasticity-Induced Foot Deformity. Rev ed. Tucson, Ariz: Therapy Skill Builders; 1990. [96]Seibel MO. Foot Functions: A Programmed Text. Baltimore, Md: Williams & Wilkins; 1988. [97]Rang M, Wright M. What have 30 years of medical progress done for cerebral palsy? Clin Orthop. 1989;247:55-60.

BD Cusick, MS-CCT, PT, is Clinical Specialist, Lucille Packard Children's Hospital at Stanford, 725 Welch Rd, Palo Alto, CA 94304. She also teaches, consults, and is involved in private pratice. WA Stuberg, PhD, PT is Director of Physical Therapy, Meyer Rehabilitation Institute, Associate Professor, Division of Physical Therapy, Education, and Assistant Professor, Department of Anatomy, University of Nebraska Medical Center, 600 S 42nd St, Omaha, NE 68198-5450.