Orthotic Implications of Lower-Limb Growth



Longitudinal growth adjustment continues to receive attention as a commonly encountered design limitation of children's orthoses. However, the amount of accommodation for length adjustment by the usual overlapping of side bars is generally much greater than actually required during the usual wear period of the average lower-limb orthosis, resulting in unduly heavy, bulky appliances. Similarly, circumferential alteration by reshaping metal or plastic cuffs is usually minimal during the appliance lifetime. Foot growth is generally slow enough to warrant the use of foot plates. Thus one can modify the plates rather than having to replace the entire orthosis or resorting to stirrups riveted to shoes.

On the basis of the data presented, a knee-ankle-foot orthosis being provided for a child aged 4-14 years needs to accommodate about 1.6 cm of tibial longitudinal growth, an average of 0.25 cm maximal calf circumferential growth, and about 0.9 cm foot growth, per year of expected brace usage.


Lower-limb growth rates have implications for the suitability of plastic orthoses for disabled children. In conventional metal orthoses, orthotists usually accommodate growth by providing redundant materials, such as overlapping uprights secured by a half-dozen or more steel screws, which the child must then continually transport as he walks. Accurate prediction of the maximum accommodation necessary during the estimated life of the orthosis allows braces to be constructed with only a minimum of materials. The weight savings would be gratefully received by the disabled child.

Plastic knee-ankle-foot orthoses (KAFO's) made for the Child Prosthetic and Orthotic Studies at New York University often weigh half the amount of the child's previous appliance. Despite these savings, skeptics complain that plastic orthoses are not easily amenable to growth adjustment. They note the lack of overlapping side bars and the intimacy of fit in the foot, calf, and thigh sections7. The use of plastic appliances thus requires greater orthotic skill in fitting the device, and subsequent vigilance by the clinic team, to insure continued good fit as the wearer grows.

The existing literature reveals no studies of lower-limb growth rates (i.e., change in dimension per year) of disabled children. The heights of the vast majority of disabled children are less than or equal to their age-predicted norms; there are no reports of groups of disabled children who are taller than their age-matched normal peers. Disabled children do not grow as fast during the normal growth spurts from birth to age 3 or 4, nor during adolescence; at all other times the disabled grow at the same rate as normal children4,5,6,8,12,13,14,20. That is, they "stabilize their percentile rank" of mean age-predicted height in relation to other children5. Thus, normal growth rates may be used to predict the upper limit of expected growth for the children most often seen in the orthotic clinic.

Anderson and his colleagues studied normal femoral and tibial growth longitudinally, relating the mean absolute length of these bones to age1. A longitudinal study serially follows the growth of the same children over a period of years19. Anderson, et al.2, and Stuart, et al.17,18, semilongitudinally studied foot length and calf circumference, respectively, and provided age-related measures of central tendency. Semilongitudinal studies combine serial measures and cross-sectional measures (children of one age are compared to different children of another age19), statistically treating both measures as being equivalent. Regression lines obtained for the years of linear (stable) growth would, of course, have been preferable, since their calculated slopes would relate growth velocity to any given length or circumference rather than being yoked to a particular age. Such information is not now available. Ironically, Karl Pearson invented regression for the purpose of studying height but never completed these studies".

The purpose of this article is to explore the normal growth rates of children, relating the information obtained to suggestions for orthotic accommodation of lower-limb growth. These considerations have important functional, cosmetic, and economic implications for the orthotic clinic team, the young patient, and his family.


Age-referenced means or medians from the existing data1,2,17,18 were serially subtracted to obtain annual increments in foot, femoral, and tibial length, and calf circumference. The growth changes per year were then averaged over the periods of linear growth, providing a mean rate and standard deviation for each growth phenomenon.


The results are summarized in Table 1 , Table 2 , Table 3 .


For the purposes of this discussion it is assumed that lower-limb orthotic devices should last at least two years. For clarity of presentation, each section discusses each relevant category of growth separately, even though growth rates of the component segments are highly related to one another.

Foot Growth

Normal children's feet grow quite predictably (Table 1 ). Rossi reports that by age 11 or 12 for girls and age 14 for boys, the foot will have reached 90 per cent of its adult size15. The mean foot-growth rate for girls aged 4 to 12 is 0.90 cm ± 0.18 per year; boys aged 4 to 14 exhibit foot growth an average of 0.88 cm ±0.12 per year. There is some evidence that in 1324 Edward II decreed that three average barley corns, placed end to end, equalled one inch. Thirty-nine corns were needed to measure the length of the King's "normal" foot; one barley corn was thus defined as one shoe size. Today, one shoe size still equals 0.85 cm (1/3 in.)15. A child's foot, therefore, grows about one American shoe size per year, from age 4 to 14.

The relative contribution of the tarsals and metatarsals to total foot length is constant throughout the growing years2. Thus, the foot plate of a (K)AFO need accommodate less than one centimeter growth per year. Although no reports appear in the literature comparing the growth of disabled children's feet to those of normal children, clinical experience indicates that paralyzed feet grow at rates equal to or less than normal; hence, even less frequent foot plate and shoe change should be necessary for the disabled child's foot. Accepted orthotic practice allows the distal end of the plate to terminate as far forward as the proximal edge of the first interphalangeal joint. Foot-plate length is not inadequate until the plate fails to contact the proximal edge of the first metatarsal head. Thus, the orthotist can easily provide sufficient foot-plate length accommodation for the life of the appliance. Larger shoes, incidentally, will be required no more than two or three times during the two-year wear period of the orthosis.

Femoral and Tibial Growth

Probable longitudinal increase of the lower limb can be estimated in two ways. Although chronological age is the simpler guide, skeletal age is a more accurate predictor of changes in leg length. A high correlation exists between skeletal age and sexual maturity3. Normal children's femora and tibiae grow at age-predictable rates, although there is a difference between the rates of growth of boys and girls (Figure 1 , Figure 2 , Figure 3 , Figure 4 ).


Lower-limb growth of disabled children may not follow the normal age-predicted growth curves. Several reports have demonstrated that disabled children are, on the whole, significantly shorter than their normal peers6,8,20. Stature is inversely related to degree of disability20.

The distal femoral and proximal tibial growth plates normally contribute about 37 and 28 per cent, respectively, of total longitudinal growth of the lower limb10. Most pathology, however, occurs elsewhere in the limb, such as Legg-Perthes disease, diaphyseal fracture, or surgical fixation of a slipped capital femoral epiphysis. Consequently, approximately 65 per cent of femoral and tibial growth occurs in places usually unaffected by neuromusculoskeletal disorders. Thus, it has been reported that no difference in rate of growth, compared to normals, exists for cerebral palsy20,12, Down's Syndrome4,5,12,13,14, or prematurely born children6,8. In fact, complete denervation of the limb fails to retard the growth rate in experimental animals16. Thus, in the absence of a roentgenographic skeletal-age estimate for the individual child, the use of normal growth velocities currently provides the most feasible approximation for orthotic purposes.

Table 2 reveals that, although growth rates decline slightly from age 4 to age 14 (girls) or 15 (boys), normal leg growth is linear during most of the childhood years. The greatest relative growth periods occur in the first years after birth, and during the adolescent growth spurt. The presumed linearity of growth during the adolescent growth spurt is the result of statistical insensitivity to the varying years of onset of the spurt being averaged over time with the ensuing sharp decrease in growth velocity19. These averaged slopes may be quite descriptive of disabled children who may not have an adolescent growth spurt12. The slope of the lines becomes less steep at age 4 and remains constant until age 15 for boys and 14 for girls, at which point growth rate declines. The mean femoral growth per year from age 4 to 15 for boys is 2.00 ± 0.20 cm, while their tibiae grow an average of 1.61 ± 0.15 cm per year. Girls grow less rapidly and for a shorter period. Their femora grow an average of 1.92 ± 0.44 cm per year, and their tibiae grow an average of 1.56 ± 0.38 cm per year from age 4 to 14. These data were obtained from a longitudinal study of 67 boys and 67 girls1. Maresh9 studied 113 healthy children semilongitudinally and did not separate the rates for boys from those of girls, but obtained results comparable to those of Anderson, Messner, and Green1.

Orthotic linear accommodation, therefore, can be relatively small. For example, to accommodate growth in a girl from age 4 to 6 years, the femoral portion of a KAFO would need to be adjusted 5 cm and the tibial portion 4 cm. A boy of age 7 would need approximately 4 cm for his femoral-growth accommodation but only 3.2 cm for his tibial growth, over a two-year period.

If a discrepancy of 0.8 cm were allowed between the heights of the mechanical and the anatomical knee joints, the mechanical knee joints could initially be made 0.8 cm too high. The tibia would, therefore, grow 1.6 cm before the mechanical knee joint would be unacceptably low. Thus, the tibial portion of a KAFO will need to have additional longitudinal adjustments only after its first year of use. Dimeglio, et al., provide a brief review of current orthotic linear-accommodation techniques for ankle-foot orthoses, but provide no information on KAFO's.

The femoral portion of the orthosis may require little or no longitudinal adjustability if there is no orthotic hip joint above this section. It should be noted that although the linear and circumferential dimensions of the orthosis can be systematically varied according to the growth requirements of the child, only a refitting of the three-point pressure system will reestablish the maximum orthotic efficacy.

Calf Circumferential Growth

Growth involves changes in circumferential shape as well as length. Although the limbs pass through predictable stages of angular and torsional morphology, these are not problems in orthotic growth accommodation, since orthoses are often prescribed to correct or accommodate the deformities of improper torsional and angular relationships. Tobis, et al.20, found that cerebral palsied children, when compared with standardized ("Z") scores, were much shorter than their age-matched normal counterparts, but only slightly less heavy. Therefore, the average cerebral palsied child is heavier than a normal child of the same height. Thus, circumferential growth may be a relatively greater problem than longitudinal growth for disabled children.

Most juvenile lower-limb orthoses prescribed today are either ankle-foot orthoses or knee-ankle-foot orthoses. Thus, calf-breadth growth is critically important to orthotic accommodation for the growing child. In the absence of edema, soft-tissue growth in the child is a combination of collagen and adipose tissue growth, and muscle hypertrophy10.

It is obvious that a great deal of individual variation exists in normal populations. That lower-limb circumferential growth in disabled populations has not been studied should not be surprising10. Stuart, Hill, and Shaw17,18 have examined maximum normal calf breadth using the soft-tissue shadow of the roentgenograph. Their findings support empirical impressions that the rate of calf growth remains fairly constant-throughout the growing years. Table 3 contains the calculated values for both groups. From their data17,18 it is clear that the average rate of maximal calf-diameter increase is 0,3 ± 0.08 cm per year in boys from 2 to 6 years old, and 0.23 ± 0.09 cm per year from age 7 to 10. Girls' calves grow more rapidly from age 2 to 6, at a rate of 0.33 ± 0.11 cm per year but thereafter grow more slowly than boys', viz., 0.14 ± 0.09 cm per year. Although Tobis, et al.20, found no difference in the standard deviations of cerbral palsied children's weight compared with normals, caution must be exercised in interpreting these data. Normal children probably have less variation in lower-limb circumferential growth than do most groups of disabled children due to the latter's widely disparate exercise rates and nutritional habits10. The data should, therefore, be considered to be a conservative estimate of soft-tissue growth in disabled populations.

Plastic shells may be one answer to the problem of circumferential accommodation for the growing child. The intimate fit required of plastic shells may be troublesome for the insensitive limb, however, and requires vigilance on the part of the parents and the clinic team to prevent skin irritation. Plastic shells allow a greater area of contact with the skin, thus permitting force distribution over larger areas. The edges of the shells can be flared away from the skin when the brace is fabricated, to obviate sharp changes in skin pressure from the forces applied by the orthosis.

In conclusion, the amount of growth accommodation of the child's orthosis should be based on age-predicted norms, thus permitting a minimum of bulk and extra weight to be incorporated into the device which the child must bear continually as he or she walks. More research is needed to provide empirical validation of these figures in orthopedically and neurologically disabled populations.


I would like to thank Sidney Fishman, Ph.D., and Joan Edelstein for their editorial assistance in the preparation of this manuscript, and Marshall Kaufman and Eric Hoffman for their technical suggestions.



    1. Anderson, M., M. B. Messner, and W. T. Green, Distribution of lengths of the normal femur and tibia in children from one to eighteen years of age. J Bone Joint Surg, 46A: 1197-1202, 1964.


    1. Anderson, M. A., M. Blais, and W. T. Green, Growth of the foot during childhood and adolescence. Am J Phys Anthropol, 14:287-308, 1956.


    1. Bayley, N.,, Tables for predicting adult height from skeletal age and present height. J Pediatr, 28:49-64, 1946.


    1. Chumlea, W. C., R. M. Malina, G. L. Rarick, and V. D. Sefeldt, Communalities for rates of diaphyseal elongation of short bones of the hand of children with Down's syndrome. Am J Phys Anthropol, 53:129-131, 1980.


    1. Cronk, C. C., Growth of children with Down's Syndrome: birth to age 3 years. Pediatrics, 61:564-568, 1978.


    1. Dann, M., S. Z. Levine, and E. New, Development of prematurely born children with birth weights or minimal postnatal weights of 1000 grams or less. Pediatrics, 22:1037-1053, 1958.


    1. Dimeglio, P. J., M. Taylor, C. Thomas, and J. Merchand, Growth extensions and adjustments in childrens polypropylene ankle-foot orthosis. Orth and Prosth, 33:4:49-50, 1979.


    1. Drillien, C. M., Growth and development in a group of children of very low birth weight. Arch Dis Child, 33:10-18, 1958.


    1. Maresh, M. M., Growth of major long bones in healthy children. Am J Dis Child, 66:227-257, 1943.


    1. Moseley, C. F., Growth. In Lovell, W. W. and R. B. Winter, Pediatric Orthopedics. Lippincott, Philadelphia, 1978.


    1. Pearson, K., and A. Lee, On the laws of inheritance in man. I. Inheritance of physical characters. Biomedtrika, 2:357-462, 1902.


    1. Pryor, H. B., and H. E. Thelander, Growth deviations in handicapped children: an anthropometric study. Clin Pediatr, 6:501-512, 1967.


    1. Rarick, G. L., I. F. Rapaport, and V. D. Sefeldt, Long bone growth in Down's syndrome. Am J Dis Child, 112:566-571, 1966.


    1. Rarick, G. L., and V. Sefeldt, Observations from longitudinal data on growth in stature and sitting height of children with Down's syndrome. J Ment Defic Res, 18:63-78, 1974.


    1. Rossi, W. A., How fast does a child's foot grow? J Am Pod Assoc, 69:278-279, 1979.


    1. Selye, H.. and E. Bajusz, Effect of denervation on experimentally induced changes in the growth of bone and muscle. Am J Physiol, 192:297-300, 1958.


    1. Stuart, H. C, P. Hill, and C. Shaw, The growth of bone, muscle and overlying tissues as revealed by studies of roentgenograms of the leg area. Monogr Soc Res Child Dev, 5:1940.


    1. Stuart, H. C, and P. Hill-Dwinell, The growth of bone, muscle and overlying tissues in children six to ten years of age as revealed by studies of roentgenograms of the leg area. Child Dev, 13:195-213, 1942.


    1. Tamer, J. M., Charts for the Diagnosis of Short Stature and Low Growth Velocity. In Bergsma, D., and R. N, Schimke (eds.). Growth Problems and Clinical Advances, 12:6. Alan R. Liss, New York, 1976.


  1. Tobis, J. S., P. Saturen, G. Larios, and A. O. Posniak, Study of growth patterns in cerebral palsy. Arch Phys Med Rehabil, 42:475-481, 1961