Rotation Of Congenitally Hypoplastic Lower Limbs To Use The Ankle Joint As A Knee

John E. Hall, M.D., B.A., C.M., F.R.C.S.(C), F.A.C.S.

The concept of rotating a short lower extremity in which the level of the ankle joint approximates the level of the knee joint on the opposite (sound) limb was first advanced by Borggreve1 in 1930. His patient had a short femur secondary to infection, and the rotation was performed through the femur.

In 1950 Van Nes2 described three cases in which he had rotated congen-itally short lower extremities so that the ankle joint could be used to give active control of the knee joint of the patient's prosthesis. The ages of his patients were 1, 9, and 15 years. In the 1-year-old patient, part of the rotation was performed through a pseudarthrosis at the upper end of the femur, and the rotation was completed through a knee arthrodesis. In the 9-year-old, part of the rotation was performed through a knee arthrodesis, and subsequently the remainder was obtained through a tibial rotation osteotomy. In the third child, all the rotation was performed through a knee arthrodesis with sufficient resection of the lower end of the femur and the upper end of the tibia to effect the rotation safely. In each of these patients the final level of the ankle joint approximated the level of the opposite knee joint. The patients learned to control the knees of their prostheses well.

Faced with several children having similar problems at the Amputee Clinic of the Ontario Crippled Children's Centre, it was decided to use the principles of Borggreve and Van Nes to provide the anatomical basis for a better type of prosthesis than these children had previously worn. One of the principal difficulties with prosthetic fitting in this type of deformity is that the prosthetic knee joint, if one is provided at all, must be either very close to the level of the anatomical hip joint or below the level of the foot held in equinus. If no knee joint is used, it is possible to make a reasonably shaped leg with the foot held in full equinus and the prosthesis shaped over it; but the long, straight segment makes for difficulties in sitting and a rather awkward gait.

The patients to be discussed had a variety of deformities at the upper end of the femur. In one patient the condition appeared to be a simple coxa vara, but in several others a rather severe, largely cartilaginous deformity was present, combined with what was, in effect, a pseudarthrosis at the upper end of the femur (Fig. 1 ). In many patients this hip defect was an additional cause of instability and limp. In three of our patients, it has been treated by a separate procedure. Three others will require some type of surgical procedure on their hips before completion of their treatment.

Two of our patients have been successfully fitted with prostheses, while four others were still in the process of being fitted when this article was submitted. The following case reports are of the two patients who were fitted during the past year.

Case 1 (S.N.)

Surgery. This girl, who was 10 years of age at the time of the operation, had been wearing a very cumbersome built-up boot, which only partially compensated her 6-inch leg-length discrepancy. She had a very unsatisfactory gait and was extremely concerned about her appearance. It was decided to begin her prosthetic programme by attempting to use her ankle joint as a knee, and in August 1965 a rotation osteotomy was performed

through the left tibia (Fig. 2 ). A 3-inch segment of the tibia was resected, and the rotation was maintained by using the fibula as an intermedullary bone graft with a staple passed through drill holes in the proximal and distal fragments.

The limb was immobilized in plaster for six weeks. X-rays at the end of that time showed that the osteotomy had united solidly, and the child was admitted to the Ontario Crippled Children's Centre. Within two weeks, she had regained good active control of flexion and extension in her ankle joint, which she learned to consider as her knee joint, and the fitting of her prosthesis was begun (Fig. 3 ). Because of her hip instability, a quadrilateral upper socket with ischial bearing was used. It is planned to stabilize her hip joint later by correcting the coxa vara and doing an innominate osteotomy to improve the acetabular roof.

Prosthesis. Because of hip instability, a quadrilateral type of weight-bearing plastic thigh cuff was used. The University of California at Berkeley (UCB) brim-fitting technique was employed in the fitting. The lower socket was made from a plaster cast, with the foot in the equinus position. In the initial fitting the thigh cuff and lower socket were connected by two supporting sidebars. A Winnipeg-type wedge-disc coupling was used for foot adjustment (Fig. 4 ). The prosthesis was finished by foaming a shank and applying a surface laminate. A back check was used as in a conventional below-knee prosthesis. Approximately 90 degrees of flexion at the knee can be obtained with the prosthesis to permit normal sitting (Fig. 5 ).

Case 2 (D.N.)

Surgery. This boy was 3 1/2 years of age at the time rotation osteotomy of his left tibia was performed in the same manner as in the previous patient (Fig. 6 ). One year earlier he had had an osteotomy of the upper end of the femur to correct a coxa vara. This had been only partially successful, in that postoperatively the coxa vara had begun to recur. His leg was immobilized in a plaster spica for six weeks, and at the time of plaster removal he was admitted to the Ontario Crippled Children's Centre for physiotherapy and prosthetic fitting.

During his postoperative convalescence he walked on and broke the plaster (which was intended as a nonweight-bearing cast). Some of the rotation was lost, but this loss has been successfully handled prosthetically by using a pelvic band. Further surgery is anticipated in the region of his hip joint as he gets older, and more rotation will be obtained at that time.

The knees of both of these patients are stabilized in the thigh section of the prosthesis. Whether or not they will ever require a knee arthrodesis is a decision for the future.

Prosthesis. The construction of the prosthesis was similar to that of Case 1 except, as stated above, that a pelvic band was added to control rotation (Fig. 7 ).

Additional Operations

Four additional patients were operated upon during July 1966, and although the prosthetic portion of their programme is not yet complete, details of their surgical procedures are reported.

Case 3 (M.W.)

Surgery. This girl was nearly 15 years of age at the time of operation in July 1966. Previously she had had four operations at the upper end of the femur, which had been the site of a severe deformity. The defect consisted of a coxa vara and a rather large cartilaginous section comprising a pseudarthrosis (Fig. 8 ). In June 1965 an extensive procedure was performed on the left hip. At the time of the operation, she had a severe varus deformity, with an ununited area just below the level of the greater trochanter. The entire upper end of the femur was exposed to the level of the hip joint capsule, which was partially opened in front. The upper fragment was freed from the lower, and a Steinmann pin was passed along the neck of the femur through the greater trochanter. By the use of this pin, the neck of the femur was brought down and rotated into its proper position. The redundant portion of the distal fragment was excised, and with some difficulty the remainder was fastened to the proximal fragment by means of screws and a wire. The femur united satisfactorily after ten weeks in plaster, and during the following year the child developed a normal range of movement of her hip joint with good stability.

Rotation of the limb was performed through the knee joint in June 1966 by resecting 2 inches at the lower end of the femur and an inch at the upper end of the tibia. The two raw cancellous surfaces were then held together by staples. At the same operation an amputation was done through the metatarsophalangeal joints, removing all the toes to make a neater-looking stump (Fig. 9 ).

Case 4 (J.C.)

This boy's story was very similar to that of the previous patient's. He had had two procedures on his left hip, which had been unstable, with an ununited area below the lesser trochanter (Fig. 10 and Fig. 11 ). In June 1965 a procedure similar to that described for M.W. was performed, which resulted in a stable hip with an excellent range of movement (Fig. 12 ).

In July 1966 a rotation of his limb through the tibia was performed, with 2 1/2 inches of the bone being resected at the time of the operation. This shortening permitted rotation of the limb without any danger to the circulation or to the nerve supply. The incision was made diagonally across the leg at the level of the anticipated osteotomy and slanted so that the skin incision would open up during rotation. The rotation in this child was maintained, as in the previous patient, by intermedullary insertion of the resected segment of fibula. A staple was drilled into the upper and lower fragments. At the same time, this boy's toes were removed by disarticulation through the metatarsophalangeal joints.

Case 5 (P.A.)

This 4-year-old girl had a 4-inch leg-length discrepancy at the time of her operation in July 1966. Although she had a slight coxa vara, her hip joint was not unstable. It was decided that the first approach to her treatment should be rotation of her tibia. This was performed with a resection of 2 1/2 inches. Her toes were also removed through the metatarsophalangeal joints.

Case 6 (D.N.)

This 3-year-old girl had clinical findings almost identical with those of Case 5. She was managed in exactly the same way, by rotation through her tibia, resecting enough of the bone so that rotation could be effected without any danger of stretching the neurovascular tissues. Her toes were also removed as part of the same operative procedure.

It is anticipated that both these children may require procedures on their hips in the future and all may require knee arthrodeses, but at the present moment the presence of the knee joint does not interfere with the fitting of a prosthesis. In fact, the use of a small strap fitted over the patella inside the prosthesis aids in suspension.


It is believed that the principle of rotating a hypoplastic lower extremity which is too short for adequate mechanical lengthening is a valuable one, and permits more efficient prosthetic fitting. The procedure is worth considering before using the more standard method of performing a Syme's amputation and fitting the patient with a knee-disarticulation-type prosthesis.

John E. Hall, M.D., B.A., C.M., F.R.C.S.(C), F.A.C.S. is the Chief, Amputee and Prosthetic Clinic Ontario Crippled Children's Centre, Ontario


1. I. Borggreve, Archiv für Orthopädische Chirurgie, 28, 1930 (quoted by Van Nes).
2. CP. Van Nes, "Rotation-plasty for Congenital Defects of the Femur, Making Use of the Ankle of the Shortened Limb to Control the Knee Joint of a Prosthesis," Journal of Bone and Joint Surgery, 32B:12,