Growth in the child with a tumor in the limb

The scenario of the growing child with a tumor in the limb is one of the most difficult challenges in musculoskeletal oncology. When tumors are close to a center of a bone responsible for growth, removing them can result in stunted growth in the limb as the other limbs grow normally. In specific areas furthermore, these challenges can result in dysfunction around adjacent joints (eg. hip dislocations). A number of solutions have been proposed for this situation. In some centers amputations are resorted to fearing for the crippling effect misplaced solutions are apt to cause. In the very young child amputations and their variations do remain a very viable solution and should be seriously considered.


A number of physiological norms need to be born in mind when planning surgery in such children. Ordinarily the limbs of children are known to stop growing at a certain age. Girls are generally accepted to conclude gain in height due to limb growth at 14 and boys at 16 years of age. The rate of growth of the respective centers (ie. growth plates or physes) also needs to be considered. Accordingly, per year up to the age of maturity the proximal femur grows 4 mm, the distal femur 10 mm, proximal tibia 7 mm and ankle 5 mm. Therefore a 10 year old girl who loses a proximal tibial growth plate would be expected to have a limb length discrepancy of 28 mm (ie. 4 years of growth up to 14 at the rate of 7 mm per year). There is furthermore a tendency for a child to stop growing while he is undergoing chemotherapy but there is a rapid period of catch-up growth to his normal potential afterwards. Finally one must consider that in the upper limb length discrepencies are easily compensated for and there is no real need to correct for length discrepencies especially if the shoulder has not been fused. The very fact that such a shoulder would dangle would be able to make up for any length discrepancy (Figure 1). The more critical issue is length discrepancy in the lower limb to which the rest of this article is focused.

Figure 1. Upper limb tumors often do not pose the same issues for growth in limb salvage surgery. While the final result is often shorter than the normal side, the child adapts quickly. It is therefore critical to provide a painless functional arm and the child will do the rest.

How much of a discrepancy should be corrected?

It may be surprising to note that many seemingly normal people walk around with marginal limb inequalities. Therefore while it seems intuitive that limb length should be corrected until both limbs are perfectly equal the reality is this is not actually necessary. Considering that these factors significantly impact on management decisions, it has been determined that a limb length discrepancy of up to 2 cm is quite readily compensable. More than 2 cm up to 4 cm discrepancy can be corrected by shoe raises or with contralateral limb growth retardation (eg. epiphysiodesis) and anything more than 4 cm requires active surgical correction (ie. lengthening).

Therefore working backwords from the growth norms given above for example, a distal femur tumor would really only require correction in a boy younger than 12 years old because one would anticipate 2 cm to be compensable and another 2 cm to be correctable. In practice however, if a device or intiative has been incorporated to correct for length then one would aim for complete correction. For example a 10 year old boy with a growing prosthesis would be aimed for correction within 2 cm of normal although a correction to within 4 cm would suffice.

Anatomical implications

As mentioned above, one is usually not too concerned about corrected length in tumors in the upper limb.

In the hip, the anticipated length discrepancy is actually quite minimal in a given case considering the rates of longitudinal growth. What is far more important in this stage is to ensure that the hip stays wthinin its socket during the period of growth. Often attempts at lengthening a hip joint results in stresses on the hip increasing the chance of dislocation. In practical terms, if one is aiming for a 4 cm discrepancy (ie. 10 years of growth), a boy with a hip tumor would need a lengthening procedure if he presented at 6 and a girl at 4. Thankfully, this presentation is very rare with such tumors more typically occurring nearly a decade later. If a hip is to remain mobile (as opposed to a fused hip) it is often more desirable to keep the hip short by up to 2 cm. This has the effect of stabilizing the hip.

The knee contributes the most to linear growth. Unfortunately when resorting to prosthetic solutions here one needs to destroy both the tibial and femoral growth plate amounting to 1.7 cm of growth a year. By preserving one or the other growth plate however this loss while inevitable could be halved.

The ankle contributes minimally to linear growth and tumors around this region in the actively growing child is extremely rare. Considering how functional below knee amputees are, serious thought should be given to amputations (Figure 2).

Figure 2. It is rarely if ever necessary to perform an amputation in this era of modern medicine for bone tumors unless the tumor has fungated through the skin. This patient who had a neglected bone tumor and had sought a second opinion from a neighbouring centre was offered an amputation above the knee (a). We agreed that an amputation was necessary but offered the option of replanting the amputated segment using microsurgical techniques (b). We were able to re-anastamose all his vessels and nerves which were lost in the amputated segment (a so-called amputation-replantation procedure). The huge advantage of this procedure is that the ankle now functions as a knee and can take a below knee prosthesis(c). The variants of this procedure and how it compares to an above knee amputation(d) illustrate how this procedure may be superior over an above knee amputation.

Methods available for correcting limb length

Physis sparing resections

The growth plate preserving resection is when feasible the best of these options. Apart from the readily addressable intercalary reconstructive issues this is the only modality that allows the limb to grow as per normal with a fully functional joint (Figure 3).

Figure 3. This 4 year old child had Ewing’s sarcoma of the tibia (a). It was about 15 mm from the growth plate (b). In view of her age, she was offered an amputation by her referring physician. Nevertheless we were able to offer her a limb sparing procedure after resecting the tumor (c). The only thing that would fit the dimensions of this toddler was the humerus of an adult (d). By doing an intercalary replacement, the author was able to save the growth plate (e). Surgery was uneventful and the epiphyseal screws were removed to allow growth (f). Six months later (g) the growth in the epiphysis was appreciable (arrowed).

Slowing down the opposite limb

A similarly useful procedure is the contralateral epiphysiodeses. In this modality, the growth plate of the contralateral knee is surgically disrupted. This needs to be timed accurately and tends to be really only practical in the older child where limited further growth is possible.

Bone transport

Bone transport is another option. This yields good results but the child must remain in the apparatus for long periods of time. At an elongation rate of 1 mm per day the child with a 9 cm discrepancy may remain in the apparatus for 3 months for the elongation and a further 3 months for consolidation of the regenerate. There would be a significant risk of infection if this method is adopted while the child is on chemotherapy. A variant of this method with fully implantable devices has been developed in our clinic and is described here.

Growing prostheses

The extensible prosthesis is a marvel of modern science that is presently undergoing teething issues. The manual designs require repeated surgical procedures to periodically lengthen the limb to keep pace with normal growth. Self-extending designs work through electromagnetic couplers or heating coils that allow motors with heat release springs to extend the implant. All these designs have poor longevity. Ironically, especially around the knee the use of prostheses results in destruction of an additional growth plate (for example, for a distal femoral tumor to be reconstructed with a knee endoprosthesis one would have to destroy both the distal femur and proximal tibial physis). In other words it adds to the problem it attempts to correct!

As such there are many solutions to the problem but no perfect one.


From the foregone discussion it is apparent that the surgeon dealing with this condition must be able to perform or at least facilitate any one of these procedures. Any one of these procedures is applicable to the individual case and they remain complementary to each other. Managing (especially parental) expectations is an important part of treating this group of patients. While length is one aspect of growth, the limb even if it is restored to normal length cannot be totally normal because the tumor when removed takes with it a fair bit of normal tissue as well. It is common for salvaged limbs therefore to be smaller than their normal counterpart (eg. the foot may be smaller). Nevertheless, length restoration does provide significant advantages to the overall function of the child.