Special Features of Estimating Leg Length Discrepancies in Limb Deficient Children



The problems of estimating leg length differences in children is well covered in standard texts of pediatric orthopedics, however, the materials are about children who have both femurs and both tibias intact. For these children, most orthopedic surgeons use estimates made from Moseley's "Straight Line Graph". In those with limb deficiencies, however, one is faced with making an appropriate decision for children with problems which are not covered in the usual texts. Some examples are: "How can I get the right length of stump after doing a Syme's amputation in children with PFFD?" "How do I get the appropriate length after a van Nes procedure in a child with a PFFD or after a "Turnabout" procedure following a resection of a tumor?" "When is the optimum time to plan an epiphyseodesis for a child with a Through-Knee amputation so that the affected limb will be sufficiently short to accommodate a standard knee-hinge below the stump socket at maturity?"

Even if all of the lower extremity bones are present, it is important, within the first year or two, to be able to make an estimate of the leg length difference at maturity in children who have marked deformities, so that appropriate longrange planning can be made. For example, in a child with a Congenital Short Femur the question needs to be resolved whether lengthening later on is an appropriate option or whether the parents should be counselled to proceed ahead to have their child undergo an appropriate conversion amputation for prosthetic fitting.


When making estimates for leg equalization, it is important to establish appropriate goals. The scanogram is not used as a more accurate means of measuring the length of the extremities, but rather only the length of the femur and tibia. It excludes the pelvis and foot. A boy who has had his talus shot away will have a clinically evident leg length discrepancy, but the scanograms will show no difference between the two sides. In the same way in dealing with limb deficient children, the appropriate goal must be set. When anticipating an appropriate epiphyseodesis for a child with a Through-Knee amputation, one must keep in mind that it is not the length of the bone that is important, but the length of the stump. If the knees are to be at an appropriate level, then one must allow the one to three centimeters of soft tissues distal to the bone and must be accommodated within the stump socket. In addition, the appropriate estimation of leg length discrepancy must not be for the present but for when the child is skeletally mature.

While we all prefer to have our estimates of leg length come out as close to the ideal as possible, the clinical situation in limb deficient children, fortunately, is forgiving in that possible errors in estimation can be made up for by adjustments in the prosthesis.

Integration With Other Anticipated Surgery

Children with limb deficiencies frequently have other anomalies. One may have to plan for a proximal femoral osteotomy changing the femoral neck from varus into valgus. This may be counted on to give a few centimeters in length. A major potential problem in leg length is the requirement for fusing the knee in children with PFFD. The knee is usually in flexion. The knee, if it is to be fused, should be fused in extension. When the limb is brought into extension at the hip region, there is considerable gain in length. Conversely, there may be loss of bone at subsequent surgery. At the time of rotation osteotomies (e.g. van Nes procedures for PFFD), a wafer of bone is often excised to facilitate the rotation. The amount of bone excised can be greater if one needs to reduce the length. If knee fusion is part of the plan, a segment of bone, together with either the epiphysis of the distal femur or the proximal tibia or both, can be resected at the time of surgery. All of these aspects need to be integrated at the time the initial plan is conceived.

Principle of Proportional Growth

It was initially demonstrated by Amstutz1 and now has become an accepted principle that the growth of a congenitally short limb will progress at the same proportional rate as the difference between the limbs at the time of birth, (e.g. if the Short Femur is 50% of the length of the Normal Side at birth, it can be expected to be 50% of the length of the normal side at skeletal maturity). This is an especially useful estimate when seeing an infant (e.g. "Is the child a candidate for leg lengthening at a later age or should one proceed ahead with a conversion amputation now?").

It is essential when making this original estimate that the appropriate view is taken so that the full length of the bone is measured. In children with congenital anomalies, there is often a knee flexion contracture so that the true AP length of the femur or tibia is foreshortened. Lateral x-rays of both legs are the useful views from which to take the measurements.

In making estimates of potential leg length discrepancy for limb deficient children the Green-Anderson "Growth Remaining" Charts2 are usually the most useful guide. When looking at these charts, one sees five lines for the femur and also for the tibia. The central line is the Mean and there is a plus and minus one, and plus and minus two, Standard Deviation line above and below the Mean. These Standard Deviation lines are developed from the Anderson/ Messner/Green Charts3 that most orthopedic surgeons seldom use. These charts give the total length of the femur or tibia at any given bone age for each sex. The Normal Side bone is measured by scanogram and then compared (at the appropriate Bone Age) to the Anderson-Messner-Green Charts to find out whether the child's Normal Side is growing at the Mean or plus or minus one or two Standard Deviations from the Mean. Knowing this, the appropriate line can be selected on the Green-Anderson "Growth Remaining" Charts for the Normal Side. For the Short Side, the amount of "Growth Remaining" in the Normal Side is multiplied by the proportional difference between the sides. (E.g. If the growth remaining for the Normal Side is ten centimeters, and the Short side is growing at 50% of the Normal side, the "Growth Remaining" for the Short side is, obviously, five centimeters.)

The Green-Anderson growth data were developed from numbers measured at specific landmarks, which are not those which most orthopedic surgeons use. The femur is measured from the top of the head of the femur to the bottom of the lateral femoral condyle ossification (not the medial condyle). The tibia is measured from the "mid-point of a line drawn cross the proximal condyles" (not the tibial spines) to the mid point of the distal articulating surface ossification.3

Length Measurements from X-rays

Scanograms or Orthoroentgenogram are more difficult to obtain in a child with congenital anomalies of the lower extremities because the level of the knees and ankles may be at markedly different levels than the other side. This may require taking separate films for each extremity. Furthermore, because flexion deformities at the knee are so common in these children, it may be necessary to take the views laterally. If this is done, then a separate x-ray needs to be taken for the femur and for the tibia so that each one is taken in line with the ruler or else trigonometric calculations need to be used to work out the actual lengths.

Bone Age Estimates

Classically, all the growth data are measured in relation to Bone (or Skeletal) Age rather than to Chronologic (or Calendar) Age. This is characteristically done using the standards in the Atlas developed by Greulich and Pyle of selected x-rays of the left hand and wrist. Children with limb deficiencies often have no left hand or if they do, it may be considerably deformed. What then? If the right hand is available, it can be used with a very small margin of error. x-rays of the knees have been compared to standard x-rays of normal knees at different Bone Ages but these are not related to the data generated from the wrist films, so are best not used. Furthermore, children with limb deficiencies often have anomalies in the knee so this option is doubly not useful.

Commonly, the estimate of Bone Age is just a guess and Chronologic Age is used. One should make the estimate toward the correctable side, i.e. which will be less of a problem for the child, overcorrection or undercorrection?

Bone Age estimates have a considerable measurement error. Those which were used to create the Green-Anderson Charts are more consistent than that found in most clinical evaluations because all of the Bone Ages were read by one person4 who had been doing the estimates for years, thus providing a systematic error.

Children with PFFD

A. Those needing Syme's amputations: If there is anything which characterizes the treatment of children who undergo Syme's amputations for PFFD, is that the stumps are too long. Generally, the length of the tibia of the affected side is close to normal. In adults, the tibia is approximately 10 cm shorter than the femur in males and 9 cm. shorter in females. This means that the tibia on the affected side (if at normal length) is at the appropriate length for a stump (if one allows a minimum of 7 centimeters difference in the knee heights plus two to three centimeters of flesh on the end of the bone for the placement of an internal mechanical knee joint in the prosthesis). Therefore, if any segment of the femur remains on the affected side, the stump will be too long. This extra length is often not recognized before a knee fusion is done. Consequently, the overall plan must include the excision of parts of the physes around the knee at the time of the knee fusion. The "thigh" segment, after a Syme's amputation with a knee fusion, can readily be a little short rather than long and still provide ample leverage to propel a prosthesis adequately.

B. "Turnaround" Procedures for Tumor Resection (as different from van Nes procedures, see below). Ordinarily, the distal tibia grows at approximately one-half the rate of the distal femur on the normal side. In planning for a resection, one should get a Scanogram or Orthoroentgenogram pre-operatively and by the Bone Age and the "Growth Remaining" Chart, estimate how much further growth there will be in the distal femur on the Normal Side at the time of surgery. The Resection Side should be left long, that is, the tip of the lateral malleolus (which is a reasonable estimate of the location of the old ankle joint, which is now the "knee" i.e. the "KAnkle"5) of the Resected Side beyond the level of the knee joint on the Normal Side by half the distance that the femur has yet to grow (i.e. if the Normal distal femur has 8 cms. to grow, the Resection Side should be left 4 cms. longer).

C. PFFD with van Nes Procedure: These patients are hard to predict because the short side tibia may not be growing at a normal rate. Serial scanograms are needed to establish the growth rate or the relative lengths of the tibias used as an estimate. Furthermore, one also needs to plan for the effects of excising one or more physes proximally and the need for one or more rotation procedures. At the time of the final rotation, the distal tibia (the "KAnkle") should be left long and will usually need to be left longer than for a "Turnabout" procedure for tumor patients who are otherwise growing normally. How much longer should it be? It needs to be longer by the difference in the growth remaining between the Normal Distal Femur and the abnormal Short Side Distal Tibia. This will be estimated based on the principle of the Normal Distal Tibia growing at one-half of a Normal Distal Femur, but this value must be multiplied by the percentage that the Short Side Tibia length is of the Normal Side Tibia. For example, let us make estimates for a child who has a Normal Side femur which is estimated to have 9 cm. of growth remaining, the Short Side Tibia measures 20 Cm. and the Normal Side Tibia measures 30 Cm. If both of the tibias were Normal then the van Nes Side should be fixed longer by one half of the Femoral growth remaining, (i.e. 4.5 Cm.) However, in our example where the Short Side tibia is not growing normally but is growing at 2/3 the rate of the Normal Tibia, the growth remaining on the Short Side Distal Tibia is V/2 of 9.0 Cm. multiplied by 213 i.e. 3 Cm. Therefore, at the time of the van Nes, the Short Side should be left longer by 6 cm. (i.e.: 9 Cm. minus [9 Cm x 1/2 x 2/3]).


  1. Amstutz HC: "The morphology, natural history, and treatment of proximal femoral focal deficiency" in: A Symposium on Proximal Femoral Focal Deficiency: a Congenital Anomaly. Washington, DC: National Academy of Sciences, 1969: p 50-76.
  2. Anderson M, Green WT, Messner MB: Growth and Predictions of Growth in the Lower Extremities. JBJS, 45A:114, 1963.
  3. Anderson M, Messner MB, Green WT: Distribution of Lengths of the Normal Femur and Tibia in Children from One to Eighteen Years of Age. JBJS 46A:1197-1202, 1964.
  4. Personal communication from M. Anderson.
  5. There is often confusion between the orthopedist and the patient or when one orthopedic surgeon talks with another when using the terms "knee" or "ankle" after doing a van Nes or "Turnabout" procedure. This is especially so if the anatomic knee has not yet been fused and the child has incorporated the idea of "Knee" into his body image as representing the old "Ankle". The term "Kankle" was introduced by Colin Moseley when he worked at the Sick Children's Hospital in Toronto and has worked well for me. I commend its use to others.
  6. Shriners Hospital, 3160 Geneva Street, Los Angeles, CA. 90020