(Note from PFFDvsg Webmaster: Due to limited disk space and scanning problems, the Figures referred to in this document are not available online. My apologies for the inconvenience - hope to get them online soon.)

This is Section 5 of 5 from PFFD: A Congenital Anomaly, National Academy of Sciences, 1969.

G. WILBUR WESTIN, M.D.
FINN 0. GUNDERSON, M.D.


PROXIMAL FEMORAL FOCAL DEFICIENCY-

A Review of Treatment Experiences


To evaluate the treatment of proximal femoral focal deficiency (PFFD), 165 cases of this relatively uncommon anomaly have been reviewed. It was believed that by gathering cases from all of the Shriners Hospitals, reviewing the history and treatment of these patients, and reporting on composite experience, a rational approach to this difficult problem would evolve. In the past, treatment had involved the accommodation of the deformity in some type of prosthesis or brace. Collective experience now favors early elimination of the defect by surgical conversion to an amputation-type stump which provides improved function and cosmesis.

MATERIAL

All seventeen units of the Shriners Hospitals for Crippled Children contributed detailed information on 165 cases of PFFD. These cases were located with the help of the central office in Chicago, where computer records of all Shriners Hospital patients, past and present, are located. A total of 400 patient records were reviewed by the doctors in each unit, and those falling into the category of PFFD were reported in detail. Further information, including copies of operative reports, x-rays, and progress notes, was obtained. Treatment has ranged from prosthetic fitting alone to multiple surgical procedures. Of the 165 patients in the PFFD category, 70 were treated without surgery and 95 with surgery. The surgical approach to these 95 patients was evaluated in detail.

THE DISEASE STATE

The etiology of PFFD is unknown. It is known that the development of the limb buds takes place early in fetal life, beginning at about four weeks' gestation. Various factors act upon the developing limb, resulting in rotation, segmentation, longitudinal growth, and differentiation of elements. The most proximal elements of the limb develop first (4, 6), and the hand and foot follow, being fully formed by the seventh week. Chemical toxicity (6), radiation (6), enzyme alterations, viral infections (6), and mechanical trauma (10) have produced limb anomalies in humans and experimental animals. Ring (14, 15) has stated that the primary problem is in the enchondral ossification of defective cartilage. Gardner (6) pointed out that failure of skeletal elements to form may be due to factors operating during the period of differentiation. This critical period-at four to eight weeks of fetal life-was defined by studies of thalidomide babies. It is apparent from these and other studies that as the severity of the defect increases, so does the incidence of associated anomalies. A multifocal process must be supposed, and it must act over a period of time, the severity and variety of involvement increasing with the duration of the process. The frequency (50 percent) of congenital absence of the fibula in cases of PFFD (Table 1) supports this theory, since the more distal portions of the limb develop last.

TABLE 1        Associated Anomalies (in percentage)
Total incidence of other defects 65
Congenital absence of the fibula, same leg                                          50
Contralateral leg anomalies 26
Arm anomalies 26
Other congenital defects 09

The theory advanced by Morgan and Somerville (10), that mechanical trauma to the advancing growth plate interferes with the development of normal infantile valgus, may be appropriate for simple coxa vara, but it does not explain the wide dissociation of fragments seen in the typical case of PFFD. Among the unexplained features of this anomaly is the frequent presence in the acetabulum of a vestigial femoral head, often completely dissociated from a small shaft element. Perhaps the head has a separate impetus to develop associated with its position in the acetabulum. Review of the prenatal and familial history in these patients reveals no consistent factor. Breech presentation was the only frequently recorded variation from normal (18 percent). Two cases of nonidentical twins were found, and in each set only one of the twins had phocomelia. There is one reference to a set of twins (14) with phocomelia present in both children. This variability raises the question of individual susceptibility, whatever the cause. We believe that the evidence points to an etiology extrinsic to the fetus.

THE CLINICAL PICTURE

The deformity is commonly noted at birth. The thigh is shortened and bulky. There are usually flexion contractures of the hip and knee. After the patient learns to walk, there is a short-leg gait, a gluteus medius limp, and the foot is in equinus. Relative shortening of the extremity is progressive, carrying the knee higher and higher. In mild cases, early x-rays may show absence of the capital femoral epiphysis and a normal femoral shaft. This picture often leads to an erroneous diagnosis of congenital dislocation of the hip (Figure 1-A).

The deformity makes bracing or prosthetic fitting increasingly difficult. The hip- and knee-flexion contractures and associated anomalies contribute to the orthotic problem. Symptoms are related to the shortening of the limb and hip instability. Hip pain is seldom a complaint (9). In spite of the complete separation of the shaft from proximal elements, little or no telescoping occurs with weight-bearing. Apparently the short rotators and iliopsoas contribute to this stability. The abductor lurch is quite prominent.

TREATMENT

Since 1950 the majority of these patients have been treated surgically. The purpose of the surgery has been to: (1) improve prosthetic fitting and use, (2) improve hip stability and prevent increasing deformity, and (3) equalize leg lengths.

Amputation, knee arthrodesis, and occasionally rotation osteotomy have been used to help prosthetic fit and use. Leg lengthening has been attempted with little real success. Hip surgery has included a number of procedures, the most common being valgus osteotomy. Arthroplasty has not been a success in reconstituting a satisfactory femoral-acetabular joint. Soft tissue release at the knee or hip done solely for flexion contracture has failed repeatedly!

AMPUTATION

The most common procedure has been amputation, which was performed on 59 patients (Table 2). The Syme's type of ankle disarticulation has been the most satisfactory operation performed. Four of the below-knee and one of the foot amputations (a Chopart's) required revision. The goal of amputation is a long, above-knee, end-bearing stump that can be easily fitted in a standard prosthesis.

TABLE 2         Amputation Levels
Syme's amputation 32                          Revisions          0
Below-knee amputation 16 Revisions 4
Foot amputation distal to ankle joint         09 Revisions 1
Bilateral Syme's amputation 02 Revisions 0
Total amputations 59


KNEE ARTHRODESIS

Knee arthrodesis was performed on 31 patients (Table 3). Amputation accompanied the arthrodesis in 25 of these cases. It was found that after knee arthrodesis in full extension, the hip-flexion contracture was spontaneously corrected. One or both of the epiphyses at the knee can be excised, the goal being a long above-knee stump at maturity. Two knee fusions were considered delayed unions after the patients fell during the healing process, requiring further immobilization to achieve union.

TABLE 3             Knee Arthrodesis
With Syme's amputation                                     16
Without amputation 05
With below-knee amputation 07
Bilateral with bilateral Syme's 02
Bilateral without amputation 01
Total arthrodesis 31
Delayed union 02


LEG LENGTHENING

Leg lengthening has been attempted in nine cases with little success and frequent complications (Table 4). In no instance was leg length equalized. One of the criteria (18) for consideration of femoral lengthening is the presence of a stable hip and knee rarely found in PFFD.

TABLE 4             Leg-Lengthening Procedures
Total patients 09
Total operations 14
Femoral lengthening operations 13
Tibial lengthening operations 01
Operations gained 1 in. or more                                 06
Patients with complications 08
Leg length equalized 00


ROTATION OSTEOTOMY

Eight patients have had the Borggreve rotational osteotomy (as described by Van Nes, 17 ) or a modification thereof; (Table 5). This operation externally rotates the foot 180 deg by tibial or femoral osteotomy so that the ankle joint functions as a knee joint. Power from the plantar flexors is used to extend the prosthetic leg. Attaining 180 deg of rotation usually requires two or more operations and frequently many months in the hospital. In our group of cases, only two of the eight patients so treated can be classified as successful. Two cases were complete failures, necessitating return to the original position. The other four patients failed to gain full rotation. The cosmetic deformity of a foot  pointing the wrong way could be distressing to some patients, especially girls. This, together with the prolonged treatment time, may outweigh the functional advantage.

TABLE 5           Rotational Osteotomy
Total patients 08
Total operations 16
180' rotation achieved                      02 patients
Failure 02 patients
Less than full rotation 04 patients


EXCISION OF THE FEMORAL FRAGMENT

In nine patients the proximal end of the tibia was placed into the acetabulum after excision of the short femoral fragment (Table 6). Deepening of the acetabulum accompanied four of these transfers. The growing tibial epiphysis tends to accommodate to the shape of the acetabulum. The procedure is usually accompanied by Syme's amputation. The abductors, when present, are transferred distally on the shaft of the tibia to give added abductor strength.

The patients had a decreased range of motion and especially had decreased flexion power in the reconstructed hip In the more severe types of PFFD there may be no acetabulum. Without an acetabulum, we would caution against this type of conversion.

TABLE 6                   Excision of the Femoral Fragment (a)
Unit
No		 Patient
No.		 Age Date Procedure Result
1 5224 4 yr 1954 Simple excision of femur. Instability persisted. "Rating excellent, with good gait."
1 6204 3 yr 1965 Simple excision of femur. Excessive motion at hip. Rated satisfactory.
1 6759 5 yr 1959  Simple excision of femur. "Gait good, rating good."
12 4404 6 yr 1958 Femur excised, Colonna arthroplasty over tibia. Hip stable with good extension power.
12 4659 5 yr 1957 Simple excision of femur. Good stability with strong extension power.
12 8798 4 yr 1960 Simple excision of femur. Operative tachycardia. Died night of surgery.
12 9223 2 yr 1960 Simple excision of femur. Increased stability. Proximal tibia well seated on x-ray.
12 9485 5 yr 1961 Excision of femur. Proximal tibia shaped and placed into acetabulum. Both hips stable. Flexion weak on left. Transfer of rectus abdominis in 1966 to increase flexion power of left hip partially successful
6 yr 1962 Bilateral procedures.
12 9936 1 yr 1963 Excision of femur and tibial-acetabular arthroplaasty Hip stable with negative Trendelenberg. Flexion limited to 15 deg.

(a) When excision of the femoral fragment is accompanied by arthroplasty, stability is increased. Good extension power is retained, but flexion is weak. Excision of the femoral fragment eliminates the hip- and knee-flexion contractures.

PROXIMAL FEMORAL OSTEOTOMY

The most common surgical procedure at the hip joint was subtrochanteric valgus osteotomy (Table 7). These procedures have been successful in correcting varus in only one third of the operations performed. The low success rate is attributed to failure to recognize the abnormal anatomy of the femur at the time of osteotomy. Some hips have no femoral head ossification or significant development of the acetabulum. Others have an intact cartilaginous continuity from the short femoral shaft to the ossifying femoral head. In all our cases where such a cartilaginous bridge was present, osteotomy was unsuccessful (Figures 1-A through 1-G). Only when ossification of the femoral head, neck, and shaft have progressed to bony continuity at the site of osteotomy (Figures 2-A, 2-B, and 2-C), or when the cartilaginous portion can be excised to produce bony continuity, will this procedure be successful.

The success rate of the osteotomies done to correct varus was directly related to the amount of unossified cartilage between the head and the shaft, and to the age of the patients. A fair success rate was achieved in coxa vara with shortening where the growth plate was narrow. In those cases where a wide, abnormal cartilaginous bridge was present, the osteotomy always failed -the pin or plate pulled out of the cartilage and the fixation was lost. Further ossification occurs with age, and the defect can then be excised with fusion of the neck or head of the femur to the shaft in valgus. This procedure is usually possible by the age of eight to ten years. The abnormal proximal growth plate does not represent a significant consideration.

TABLE 7                 Proximal Femoral Osteotomy
Unit No. Patient No. Age Date Procedure Result
1 1175 9 yr 1931 "Osteotomy for varus." Unknown.
1 6205 4 yr 1957 Subtrochanteric closing wedge vaigus osteotomy. Failed.
7 yr 1960 Repeat same operation. Successful with six-year follow-up.
1 6991 7 yr 1962 Valgus osteotomy of proximal femoral shaft. Unknown.
1 7634 13yr 1958 Anterior angulation osteotomy. Done to realign the hip fused in flexion.
4 2706 9 yr 1958 Subtrochanteric valgus osteotomy. Failed.
12yr 1961 Valgus subtrochanteric osteotomy with blade plate. Plate broke, failed varus of 75 deg.
6 12 8 yr 1954 Intertrochanteric closing wedge osteotomy. Corrected 80 to 130 deg. Hip later degenerated.
4 yr 1957 Subtrochanteric valgus osteotomy. Failed.
6 yr 1959 Repeat with 165 deg blade plate. Corrected to 150 deg with 7-year follow-up.
6 988 3 yr 1960 Inverted-V valgus osteotomy. Failed.
4 yr 1961 Valgus subtrochanteric osteotomy fixed with Steinmann pins. Failed.
8 yr 1964 Excision of neck defect with shaft fixed to head. Position of 135 deg maintained 2 yr.
6 1432 3 yr 1963 Subtrochanteric valgus osteotomy. Failed.
5 yr 1965 Repeat. 105 to 125 deg maintained 1 yr to date.
6 1971 8 yr 1965 Subtrochanteric valgus osteotomy. 85 to 165 deg maintained for 1 yr.
9 7410 14yr 1957 Subtrochanteric valgus osteotomy. Failed.
9 8550 3 yr 1962 Lateral closing wedge valgus subtrochanteric osteotomy. 85 to 115 deg later failed.
7 yr 1966 Valgus osteotomy with Blount blade plate fix. Successful to date.
12 3605 8 yr 1949 Subtrochanteric valgus osteotomy. Nonunion, grafted. failed.
13 2189 5 yr 1963 Varus osteotomy to seat the shaft better. Increased stability.
15 8170 5 yr 1964 Lateral closing wedge valgus subtrochanteric osteotomy. 80 to 110 deg maintained for 2 yr.
15 5370 7 yr 1947 Subtrochanteric osteotomy. Done to realign the leg after fusion, success.
15 9178 6 yr 1964 Subtrochanteric valgus osteotomy. Failed.
6 yr 1964 Repeat. Minimal or no correction.
16 6361 7 yr 1962 Subtrochanteric lateral closing wedge osteotomy. 60 to 90 deg improvement.
Summary of Results
Total patients            18
Total osteotomies      26
Success                      9
Failure                      11
Result unknown          7

ARTHROPLASTY OF THE HIP

In nine patients, attempts were made to obtain increased stability of the hip by arthroplasty. Six of the thirteen operations performed were reported as increasing stability (Table 8). Four of these were in the previously mentioned cases where excision of the femoral fragment was followed by acetabular deepening. One patient died during surgery. Bilateral procedures were performed in two cases. In one of these, mold-shaft arthroplasty apparently gave a fair result. Two fascia lata arthroplasties were failures. In the remaining cases the acetabulum was cleared of bone remnants and fibrous tissue and the proximal femoral shaft or trochanter reduced into it. Only one of these was reported as having a good result.

Evaluation of these results is difficult; hence, status has been indicated as accurately as possible, or questionnaire has been quoted. It appears that the majority of these procedures have not been successful in increasing function. In only one was the Trendelenberg test negative.

TABLE 8      Arthroplasty of the Hip
Unit No. Patient No. Age Date Procedure Result
6 179 10 yr 1953 Fascia Lata shaft arthroplasty. Operated hip stiffer than other in this bilateral case. Poor result.
6 1115 9 yr 1963 Fascia lata shaft arthroplasty with Pemberton osteotomy. Hip dislocated. Result poor.
10 yr 1963 Adult-type Colonna arthroplasty. Redislocated, open reduction and fixation. Hip stiff.
9 3344 12 yr 1949 Greater trochanter into acetabulum, left hip (bilateral case). Result unknown.
13 yr 1950 Mold arthroplasty right hip shaft. Flexion 30/40, abduction 15/20, IR 0/0. ExR 60/50.
13 1950 Mold shaft arthroplasty left hip. Marked waddle bilateral at age 16.
12 3370 12 yr 1957 Femur absent. Proximal tibia placed in acetabutum. Result satisfactory. "Walks well for short distances."
12 4404 8 yr 1960 Colonna arthroplasty. Good initial stability with good extension power.
12 18 mo 1958 Greater trochanter placed in acetabuium. Failed.
4 yr 1961 Colonna arthroplasty Foot externally rotated. Hip stiff and painful.
12 9485 5 yr 1961 Tibial-acetabular arthroplasty, right (bilateral case). Hip stable with good power. Good result.
6 yr 1962 Tibial-acetabuiar arthroplasty, left. Hip stable. Weak flexion. Rectus abdominis transfer 1966 to strengthen flexion.
12 9936 1 yr 1963 Tibial-acetabular arthroplasty. Good stability. Negative Trendelenberg test. Flexion 0 to 15 deg.
14 105 3 yr 1927 Trochanter into acetabulum. Operative death, no autopsy.


HIP ARTHRODESIS

Hip arthrodesis has not usually been a beneficial procedure (Table 9). Such a procedure is contraindicated if a prosthesis is to be fitted on the same side.

TABLE 9               Arthrodesis of the Hip
Unit No. Patient No. Age Date Procedure Result
3 3608 9.5yr 1949 Arthrodesis, right hip. Fusion failed. Motion was painless with moderate pistoning at age 20.
12 2448 11 yr 1949 "Bosworth" arthrodesis, left hip. Failed. Peroneal nerve palsy a complication.
4 1748 4 yr 1954 Bilateral arthrodesis with hips externally rotated. 180-deg rotation was the goal to use the knee as a hip joint. This failed and hips were rotated back. Final result was a patient with buttocks 1 ft off ground.
15 5370 9 yr 1947 Arthrodesis, left hip. Fused in too much flexion. Osteotomy done to correct this but left at 50 deg. Patient had below-knee amputation. He walks but has back strain.
8 3020 10 yr 1961 Arthrodesis, left hip. Three subsequent rotation osteotomies. Patient can only walk with crutches, with marked lurching. Poor result.


CONCLUSIONS AND RECOMMENDATIONS

Most of the Shriners Hospitals' surgeons have struggled through similar patterns of treatment. The following recommendations are made as a result of this review:

Hip Joint: No surgery except subtrochanteric osteotomy in the presence of (or being able to obtain) bone continuity of the femoral neck and shaft.

Excision of the femoral fragment has produced a satisfactory result in a few cases, especially when accompanied by acetabuloplasty. It should be considered only in the more severe cases where there is no proximal continuity, as it precludes the possibility of performing other procedures for hip stability should future bone development unite the femoral head to the shaft.

Knee Joint: Arthrodesis in full extension. Excise (neither, one, or both) adjacent epiphyses dependent upon projected future growth.

Knee fusion provides a stable extremity in good alignment under the pelvis and well suited for fitting an above-knee prosthesis. Enough bone should be removed to obtain full extension at the operating table without soft tissue release. Postoperatively it may be necessary to cast the involved extremity in considerable flexion and/or abduction. However, after removal of the cast, the extremity readily assumes full extension, thereby correcting flexion contractures of the hip as well as the knee joint.

Ankle: Disarticulation using the heel pad for end-bearing. The Syme's-type ankle disarticulation provides excellent endbearing and simplifies the fitting of a prosthesis.

Prosthesis: The prosthesis should be designed for end-bearing (total-contact socket).

Conversion of the patient to an amputee is recommended as soon as the growth potential of the extremity can be determined with reasonable accuracy. This should be done before the child attends school and hopefully may make it possible for him to proceed in an unrestricted fashion throughout growth and development.

ACKNOWLEDGMENT

The authors wish to express appreciation to each unit of the Shriners Hospitals for contributions of information and participation in the analysis of the data contained in this paper.

BIBLIOGRAPHY

1. Aitken, G. T. Amputation as a treatment for certain lowerextremity congenital abnormalities. J. Bone Joint Surg. 41A(,7): 1267-1285, 1959.

2. Amstutz, H. C., and P. D. Wilson, Jr. Dys2enesis of the proximal femur (coxa vara) and its surgical management. J. Bone Joint Surg. 44A(l): 1-24, 1962.

3. Babb, F. S., R. K. Ghormley, and C. C. Chatterton. Congenital coxa vara. J. Bone Joint Surg. 3 IA: 1 15-131, 1949.

4. Borggreve, J., Kniegelenksersatz dutch das in der Beinlangsachse um 180' gedrehte Fussgelenk. Arch. Orthopad. Chir. 28:175-178. 1930.

5. Frantz, C. H., and R. O'Rahilly. Congenital skeletal limb deficiencies. J. Bone Joint Surg. 43A(8):1202-1224, 1961.

6. Gardner, E. D. The development and growth of bones and joints. A.A.O.S. Instructional Course Lecture. J. Bone Joint Sure. 45A(4):856-862, 1963.

7. Golding, F. C. Congenital coxa vara. J. Bone Joint Surg. 30B(l): 161-163, 1948.

8. Lloyd-Roberts, G. C., and K. H. Stone. Congenital hypoplasia of the upper femur. J. Bone Joint Surg. 45B(3):557-560, 1963.

9. Mital, M. A., K. S. Masalawalla, and M. G. Desai. Bilateral con-enitai aplasia of the femur. J. Bone Joint Surg. 45B(3):561-565, 1963.

10. Morgan, J. D., and E. W. Somerville. Normal and abnormal growth at the upper end of the femur. J. Bone Joint Surg. 42B:264-272, 1960.

11. Noble, T. P., and E. D. Hauser. Coxa vara. Arch. Sur-. 12:501-538,1926.

12. Ollerenshaw, R. Congenital defects of the long bones of the lower limb. J. Bone Joint Surg. 7:528-552, 1925.

13. O'Rahilly, R. The development and the developmental disturbances of the Limbs. Irish J. Med. Sci., No. 397:30-33, 1959.

14. Ring, P. A. Congenital short femur. J. Bone Joint Surg. 41B: 73-79, 1959.

15. Ring, P. A. Congenital abnormalities of the femur. Arch. Dis. Child. 36:410-417, 1961.

16. Trueta. J. The normal vascular anatomy of the human femoral head during growth. J. Bone Joint Surg. 39B:358-394, 1957.

17. Van Nes, C. P. Rotation-plasty for congenital defects of femur. J. Bone Joint Surg. 32B: 12-16, 1950.

18. Westin, G. W. Femoral lengthening using a periosteal sleeve. J. Bone Joint Surg. 49A(5):836-854, 1967.

19. Wood, W. L., N. Zlotsky, and G. W. Westin. Congenital absence of the fibula; treatment by Syme amputation-indications and technique. J. Bone Joint Surg. 47A(6):1159-1169, 1965.

This is Section 5 of 5 from PFFD: A Congenital Anomaly, National Academy of Sciences, 1969.