JBJS | Longitudinal Deficiency of the Fibula. Operative Treatment*

The Journal of Bone & Joint Surgery, Volume 78, Issue 5

Articles | May 01, 1996

Longitudinal Deficiency of the Fibula. Operative Treatment*

TREY FULP, D.O.†; JON R. DAVIDS, M.D.†; LESLIE C. MEYER, M.D.†; DAWN W. BLACKHURST, M.S.‡, GREENVILLE, SOUTH CAROLINA

Investigation performed at the Shriners Hospital for Crippled Children and the Division of Medical Education and Research, Greenville Hospital System, Greenville

The Journal of Bone & Joint Surgery. 1996; 78:674-82

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

We reviewed the results of early amputation and prosthetic fitting in twenty-five children (thirty-one extremities) who had longitudinal deficiency of the fibula and were managed between 1977 and 1992. The median age at the time of the initial operation was thirteen months (range, eight months to nine years and eight months). The median duration of follow-up was eight years and ten months (range, two years and six months to sixteen years and eleven months). A Syme amputation was performed on fifteen extremities (thirteen children), and a modified Boyd amputation (which included resection of the distal tibial physis) was performed on sixteen extremities (thirteen children). (One child had a Syme amputation on one side and a Boyd amputation on the other and is thus included in both groups.) In twenty-seven extremities, simultaneous excision of the fibular anlage was performed to prevent the development of a deformity secondary to the potential tethering effect. In twelve extremities, a diaphyseal osteotomy of the tibia also was performed to correct tibial bowing and to improve the mechanical alignment of the extremity.At the time of follow-up, the patients who had had a Syme amputation had more problems related to reformation of the calcaneus, instability of the heel pad, prosthetic suspension, and excessive length of the residual extremity. The modified Boyd amputation improved the function of the heel pad and the prosthetic suspension and provided the optimum length of the residual extremity. We also found that an early diaphyseal osteotomy of the tibia to correct severe bowing improved prosthetic fitting. This study did not support the concept that early resection of the fibular anlage or a diaphyseal osteotomy of the tibia prevents the development of hypoplasia of the lateral femoral condyle and associated genu valgum deformity.

Figures in this Article

Introduction

The management of children who have longitudinal deficiency of the fibula has been influenced by advances in operative techniques and prosthetics. In the 1950's, efforts to correct limb-length inequality and deformities of the foot and ankle in these children were associated with a high rate of morbidity and often resulted in an amputation of the affected extremity^{5,11,15,26-29}. To avoid the poor functional outcome and the high psychosocial cost to the child and family, early amputation was offered as a solution^2,8,20-23,31. The foot was amputated, and the prosthesis was modified to adjust for growth. Surgeons who noted the success of the Syme amputation in the treatment of severe problems of the foot and ankle in adults extended its use to the treatment of fibular deficiency in children. The Syme amputation seemed appropriate for children because it made it possible to avoid the problems of overgrowth associated with transdiaphyseal amputations and to preserve the distal tibial physis.

By the 1970's, the senior one of us (L. C. M.) had noted several problems associated with the Syme amputation: (1) migration of the heel pad, re-formation of the calcaneus, and poor prosthetic suspension due to distal tapering of the residual limb (Fig. 1); (2) excessive length of the residual limb, with a resultant decrease in the space between the limb and the floor, making it difficult to use the increasingly sophisticated prosthetic foot-and-ankle units; (3) severe genu valgum deformity (characterized by hypoplasia of the lateral femoral condyle, of unclear etiology) that developed when the child was eight to ten years old; and (4) unsatisfactory prosthetic fitting related to anterior bowing of the involved tibia.

Over the next fifteen years, in an effort to correct these problems, a series of modifications were made to the treatment protocol of early amputation. The changes included (1) the use of a modified Boyd amputation to improve the load-bearing characteristics of the residual limb as well as to improve prosthetic suspension; (2) resection of the distal tibial physis at the time of amputation, with subsequent proximal tibial epiphyseodesis when necessary, such that the end of the residual limb was at the level of the middle fifth of the contralateral leg at skeletal maturity; (3) resection of the fibular anlage at the time of amputation to eliminate the potential tethering effect of the fibula; and (4) a diaphyseal osteotomy of the tibia at the time of the amputation to correct clinically severe bowing and thus improve subsequent mechanical alignment as well as prosthetic fitting. We performed a retrospective review of the results for children who had been managed for longitudinal deficiency of the fibula; our purpose was to establish the validity and efficacy of each component of this modified method.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Motion Analysis Laboratory, Shriners Hospital for Crippled Children, 950 West Faris Road, Greenville, South Carolina 29605. Please address requests for reprints to Dr. Davids.

‡Division of Medical Education and Research, Greenville Hospital System, Greenville, South Carolina 29605.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Motion Analysis Laboratory, Shriners Hospital for Crippled Children, 950 West Faris Road, Greenville, South Carolina 29605. Please address requests for reprints to Dr. Davids.

‡Division of Medical Education and Research, Greenville Hospital System, Greenville, South Carolina 29605.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1 Lateral roentgenogram of the residual limb after Syme amputation, showing complete re-formation of the calcaneus. The child remained symptomatic despite numerous modifications of the prosthesis, and operative revision was eventually necessary.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 2-A, 2-B, and 2-C: Sequential lateral roentgenograms of the lower extremity in a child who had longitudinal deficiency of the fibula. Fig. 2-A: Preoperative roentgenogram showing severe tibial bowing and deformities of the foot.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-B: Postoperative roentgenogram of the extremity after a modified Boyd amputation (our preferred method of treatment) and diaphyseal osteotomy of the tibia.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C: Roentgenograms made eight years later. The calcaneus was posteriorly displaced, providing excellent prosthetic suspension. The corrected tibial alignment had been maintained.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-A Clinical photographs for comparison of Syme (Fig. 3-A) and Boyd (Fig. 3-B) amputations. The extremity that had the Syme amputation was conical and tapered distally, which compromised prosthetic suspension. The limb that had a Boyd amputation had a posterior prominence for prosthetic suspension and an intact heel cord, which may enhance proprioception. The photographs were made from the lateral side, with anterior to the left and posterior to the right.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-B Clinical photographs for comparison of Syme (Fig. 3-A) and Boyd (Fig. 3-B) amputations. The extremity that had the Syme amputation was conical and tapered distally, which compromised prosthetic suspension. The limb that had a Boyd amputation had a posterior prominence for prosthetic suspension and an intact heel cord, which may enhance proprioception. The photographs were made from the lateral side, with anterior to the left and posterior to the right.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 4-A and 4-B: Anteroposterior roentgenograms, made with the child standing and wearing the prosthesis, showing the results after two methods of correction of genu valgum deformity. Fig. 4-A: The result after proximal tibial osteotomy and epiphyseodesis. Although the tibiofemoral alignment was restored, there was persistent severe obliquity of the knee joint.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-B: The result after stapling of the medial femoral physis. Both the tibiofemoral alignment and the obliquity of the knee joint were corrected.

View Larger

TABLE I DATA ON THE PATIENTS*

*NA = not available.†Growth inhibition could not be determined in patients who had bilateral fibular deficiency.‡The ratio could not be determined accurately in patients less than six years old.§The numbers represent the value before and after subsequent procedures to correct genu valgum deformity.¶Case 13 is included in both groups because a Syme amputation was done on one limb and a Boyd amputation, on the other.

Case	Gender, Age (Yrs. + Mos.)	Side	Type of Fib. Deficiency¹	Growth Inhibition† (Per cent)		At Op.			At Follow-up			Complications	Add. Op.	Duration of Follow-up (Yrs. + Mos.)
Femur	Tibia	Fibular Anlage	Tibial Bowing (Degrees)	Tibial Osteot.	Condylar Height Ratio‡	Tibial Bowing§ (Degrees)	Tibio-femoral Angle§ (Degrees)
Syme
1	F, 0 + 11	R	II	—	—	Yes	54	No	0.95	10	7	Tibial bowing	—	12 + 3
		L	II	—	—	Yes	50	No	0.95	10	7	Tibial bowing	—
2	M, 1 + 2	L	II	5	20	Yes	15	Yes	0.66	0	20/5	Genu valgum and limb-length discrep.	4	9 + 1
3	M, 1 + 1	R	IB	10	28	Yes	40	Yes	0.70	25/10	5	Tibial bowing	2	13 + 5
4	F, 1 + 1	L	II	11	15	Yes	30	No	0.64	10	14	Tibial bowing	—	13 + 3
5	M, 0 + 11	R	II	0	27	Yes	25	No	0.64	10	16/5	Genu valgum and limb-length discrep.	2	12 + 11
6	M, 1 + 4	L	II	5	38	Yes	55	No	0.64	20	20/5	Tibial bowing, genu valgum, and limb-length discrep.	3	14 + 3
7	F, 6 + 8	R	IB	0	13	No	5	No	0.86	5	8	Limb-length discrep.	1	11 + 4
8	M, 0 + 10	R	II	—	—	Yes	60	No	NA	30	7	NA	—	4 + 10
		L	II	—	—	Yes	50	No	NA	25	0	NA	—
9	M, 0 + 9	R	II	NA	NA	Yes	NA	No	0.80	NA	10	—	—	10 + 11
10	M, 1 + 7	R	II	2	34	Yes	30	No	0.66	20	16	NA	—	8 + 8
11	M, 0 + 9	L	II	5	36	Yes	30	No	0.60	20	16	—	—	16 + 11
12	F, 2 + 2	R	IB	6	34	Yes	40	No	0.71	20/0	27/0	Tibial bowing, genu valgum, and limb-length discrep.	3	12 + 6
13¶	M, 1 + 8	R	II	—	—	Yes	65	No	0.75	40/10	6	Tibial bowing	2	15 + 6
Boyd
13¶	M, 7 + 6	L	IB	—	—	No	10	No	0.72	10	20/-4	Genu valgum and limb-length discrep.	2	9 + 6
14	F, 1 + 1	R	II	4	NA	Yes	NA	No	0.66	10	21/0	Tibial bowing and genu valgum	1	11 + 3
15	M, 2 + 0	R	II	—	—	Yes	25	No	0.56	5	15	—	—	4 + 11
		L	II	—	—	Yes	35	Yes	0.77	7	12	—	—
16	M, 3 + 3	R	II	2	35	Yes	21	Yes	0.74	10	20/NA	Genu valgum and limb-length discrep.	2	6 + 2
17	F, 1 + 1	R	II	—	—	Yes	30	Yes	NA	0	14	—	—	3 + 5
		L	II	—	—	Yes	30	Yes	NA	15	15	—	—
18	F, 1 + 1	L	II	14	21	Yes	13	No	NA	10	7	—	—	3 + 5
19	M, 0 + 9	R	II	NA	NA	Yes	50	Yes	NA	20	14	—	—	4 + 11
20	F, 9 + 8	R	II	3	18	No	5	No	0.64	5	15	—	—	3 + 1
21	M, 1 + 2	L	II	3	NA	No	NA	No	0.89	10	4	Limb-length discrep.	1	6 + 11
22	F, 1 + 4	R	II	—	—	Yes	43	Yes	NA	15	16	—	—	4 + 7
		L	II	—	—	Yes	29	Yes	NA	10	18	—	—
23	M, 0 + 8	L	II	0	43	Yes	55	Yes	NA	14	7	—	—	3 + 11
24	M, 1 + 1	L	II	3	43	Yes	50	Yes	NA	8	5	Tibial bowing	1	3 + 11
25	M, 3 + 5	R	II	10	11	Yes	10	No	NA	10	10	—	—	2 + 6

Add to My JBJS

TABLE I DATA ON THE PATIENTS*

Case	Gender, Age (Yrs. + Mos.)	Side	Type of Fib. Deficiency¹	Growth Inhibition† (Per cent)		At Op.			At Follow-up			Complications	Add. Op.	Duration of Follow-up (Yrs. + Mos.)
Femur	Tibia	Fibular Anlage	Tibial Bowing (Degrees)	Tibial Osteot.	Condylar Height Ratio‡	Tibial Bowing§ (Degrees)	Tibio-femoral Angle§ (Degrees)
Syme
1	F, 0 + 11	R	II	—	—	Yes	54	No	0.95	10	7	Tibial bowing	—	12 + 3
		L	II	—	—	Yes	50	No	0.95	10	7	Tibial bowing	—
2	M, 1 + 2	L	II	5	20	Yes	15	Yes	0.66	0	20/5	Genu valgum and limb-length discrep.	4	9 + 1
3	M, 1 + 1	R	IB	10	28	Yes	40	Yes	0.70	25/10	5	Tibial bowing	2	13 + 5
4	F, 1 + 1	L	II	11	15	Yes	30	No	0.64	10	14	Tibial bowing	—	13 + 3
5	M, 0 + 11	R	II	0	27	Yes	25	No	0.64	10	16/5	Genu valgum and limb-length discrep.	2	12 + 11
6	M, 1 + 4	L	II	5	38	Yes	55	No	0.64	20	20/5	Tibial bowing, genu valgum, and limb-length discrep.	3	14 + 3
7	F, 6 + 8	R	IB	0	13	No	5	No	0.86	5	8	Limb-length discrep.	1	11 + 4
8	M, 0 + 10	R	II	—	—	Yes	60	No	NA	30	7	NA	—	4 + 10
		L	II	—	—	Yes	50	No	NA	25	0	NA	—
9	M, 0 + 9	R	II	NA	NA	Yes	NA	No	0.80	NA	10	—	—	10 + 11
10	M, 1 + 7	R	II	2	34	Yes	30	No	0.66	20	16	NA	—	8 + 8
11	M, 0 + 9	L	II	5	36	Yes	30	No	0.60	20	16	—	—	16 + 11
12	F, 2 + 2	R	IB	6	34	Yes	40	No	0.71	20/0	27/0	Tibial bowing, genu valgum, and limb-length discrep.	3	12 + 6
13¶	M, 1 + 8	R	II	—	—	Yes	65	No	0.75	40/10	6	Tibial bowing	2	15 + 6
Boyd
13¶	M, 7 + 6	L	IB	—	—	No	10	No	0.72	10	20/-4	Genu valgum and limb-length discrep.	2	9 + 6
14	F, 1 + 1	R	II	4	NA	Yes	NA	No	0.66	10	21/0	Tibial bowing and genu valgum	1	11 + 3
15	M, 2 + 0	R	II	—	—	Yes	25	No	0.56	5	15	—	—	4 + 11
		L	II	—	—	Yes	35	Yes	0.77	7	12	—	—
16	M, 3 + 3	R	II	2	35	Yes	21	Yes	0.74	10	20/NA	Genu valgum and limb-length discrep.	2	6 + 2
17	F, 1 + 1	R	II	—	—	Yes	30	Yes	NA	0	14	—	—	3 + 5
		L	II	—	—	Yes	30	Yes	NA	15	15	—	—
18	F, 1 + 1	L	II	14	21	Yes	13	No	NA	10	7	—	—	3 + 5
19	M, 0 + 9	R	II	NA	NA	Yes	50	Yes	NA	20	14	—	—	4 + 11
20	F, 9 + 8	R	II	3	18	No	5	No	0.64	5	15	—	—	3 + 1
21	M, 1 + 2	L	II	3	NA	No	NA	No	0.89	10	4	Limb-length discrep.	1	6 + 11
22	F, 1 + 4	R	II	—	—	Yes	43	Yes	NA	15	16	—	—	4 + 7
		L	II	—	—	Yes	29	Yes	NA	10	18	—	—
23	M, 0 + 8	L	II	0	43	Yes	55	Yes	NA	14	7	—	—	3 + 11
24	M, 1 + 1	L	II	3	43	Yes	50	Yes	NA	8	5	Tibial bowing	1	3 + 11
25	M, 3 + 5	R	II	10	11	Yes	10	No	NA	10	10	—	—	2 + 6

Materials and Methods

We reviewed the medical records and roentgenograms of all patients who had been managed for longitudinal deficiency of the fibula at our institution between 1977 and 1992. We excluded children who had an associated proximal femoral focal deficiency, as our treatment protocol for this included amputation of the foot and arthrodesis of the knee, creating a situation similar to an above-the-knee amputation, with distinct prosthetic needs. Children who had associated hypoplasia of the femur (shortening of the femur of less than 15 per cent with a normal hip joint¹⁰) were included in the study group.

The medical records were reviewed to determine the age at the time of the operation, the type of procedures that had been performed, the amount of time that the tourniquet had been in place intraoperatively, the estimated amount of blood loss, the date of the most recent follow-up evaluation, the technique of prosthetic management and its related issues, and subsequent operations. The roentgenograms were reviewed to determine the type of fibular deficiency (with use of the classification of Achterman and Kalamchi¹); the percentage of growth inhibition of the ipsilateral tibia and femur, as determined from preoperative orthoroentgenograms; and the degree of tibial bowing preoperatively and at the time of follow-up. The condylar height ratio, a measure of hypoplasia of the lateral femoral condyle, was proportional to the severity of genu valgum. To determine this ratio, the greatest perpendicular height from the distal femoral physis to the joint line in the lateral and medial femoral condyles was measured, and the lateral height was divided by the medial height. The normal value for this ratio is greater than 0.80; the smaller the value, the more extensive the hypoplasia of the lateral condyle and the more severe the clinical genu valgum deformity⁷. This measurement has been shown to be accurate in children who are more than six years old⁷.

Operative Technique

The procedure used at our institution was modified from the original technique for calcaneotibial arthrodesis, as described by Boyd^6,9. The incision begins one centimeter distal to the medial malleolus and extends across the dorsum of the ankle, just distal to the tibiotalar joint, to a parallel point on the lateral aspect of the hindfoot. (As the fibula is often not palpable as a landmark, a point is chosen that is symmetrical with respect to the medial starting point.) The plantar part of the incision connects the most medial and lateral points of the dorsal part of the incision and extends distally to the level of the metatarsal heads. This creates a long plantar flap that may need to be shortened before the wound is closed. The calcaneus is preserved with the large plantar flap, and the talus and all of the bones distal to it are removed. The superior articular surface of the calcaneus is denuded of articular cartilage and is flattened with a knife. The distal part of the tibia is dissected circumferentially, identifying the distal tibial physis. A transmetaphyseal amputation of the tibia then is performed, just proximal to the physis. Care is taken not to injure the posterior neurovascular structures and soft tissues. Extension of the incision from its lateral apex proximally allows for exposure and excision of the fibular anlage, which often is nothing more than a hypoplastic band of fibrocartilage. Optimum alignment places the heel pad perpendicular to the long axis of the tibia. Residual equinus deformity (due to tight posterior soft tissues) is corrected by additional shortening of the tibia, not by release of the heel cord, as initially described by Boyd^6,9. This maintains the integrity of the gastrocnemius at the knee and preserves the potential contribution of the plantar flexors of the ankle to proprioception of the residual limb. Equinus or calcaneus alignment at the site of the arthrodesis should be avoided, as it compromises the position and subsequent function of the heel pad. The calcaneus is displaced posteriorly, which results in a posterior prominence that facilitates prosthetic suspension. Two threaded Steinmann pins are inserted to maintain the alignment.

Simultaneous correction of anteromedial tibial bowing can be performed through the incision used to excise the fibular anlage or through a separate anterior incision at the apex of the tibial deformity. Intramedullary fixation of the osteotomy site is accomplished with the Steinmann pins, which are placed antegrade through the calcaneus and heel pad and then retrograde across the calcaneotibial arthrodesis and through the medullary canal of the tibia. The Steinmann pins are removed once there is roentgenographic evidence of union, which usually occurs within six weeks (Figs. 2-A, 2-B, and 2-C).

Results

Demographics

Twenty-five children (thirty-one extremities) met the criteria for inclusion in the study (Table I). There were sixteen boys and nine girls. The left extremity was involved in eight children; the right, in eleven; and both, in six. The fibular deficiency was classified as type II in twenty-seven extremities and as type IB in four.

The inhibition of tibial growth was determined in fifteen patients (fifteen extremities); it was not possible to measure it in the six children who had bilateral involvement, and the roentgenograms were inadequate for four children. The inhibition of tibial growth (mean and standard deviation) was 28 ± 10.8 per cent (range, 11 to 43 per cent). Inhibition of the growth of the ipsilateral femur could be determined in seventeen patients; it was not possible to measure it in the six who had bilateral involvement, and the roentgenograms were inadequate for two. The inhibition of femoral growth was 5 ± 4.1 per cent (range, 0 to 14 per cent). The median age at the time of the initial operation was thirteen months (range, eight months to nine years and eight months), and the median duration of follow-up for the entire cohort was eight years and ten months (range, two years and six months to sixteen years and eleven months).

Comparison of Syme and Boyd Amputations

Thirteen children (fifteen extremities) had a Syme amputation, and thirteen (sixteen extremities) had a Boyd amputation. One child was included in both groups, as a Syme amputation was done on one side and a Boyd amputation, on the other side. Both groups were comparable with respect to gender, side of the involved extremity, type of deficiency, inhibition of tibial and femoral growth, and age at the time of the amputation. The median amount of time that a tourniquet was in place intraoperatively was fifty-seven minutes (range, forty-five to ninety minutes) for the group that had a Syme amputation and sixty-four minutes (range, forty-five to ninety minutes) for the group that had a Boyd amputation. The median estimated amount of blood loss was thirty milliliters (range, fifteen to fifty milliliters) for the group that had a Syme amputation and fifty milliliters (range, twenty to 250 milliliters) for the group that had a Boyd amputation. The median duration of follow-up was twelve years and six months (range, four years and ten months to sixteen years and eleven months) after the Syme amputations and four years and seven months (range, two years and six months to eleven years and three months) after the Boyd amputations. The difference in the duration of follow-up reflects the preference for the Syme amputation early in the study period (from 1977 to 1983).

A complication unique to the group that had a Syme amputation was partial or complete re-formation of the calcaneus within eighteen months of the amputation; this occurred in six of the fifteen extremities. The calcaneal fragment was usually posterior to the distal part of the tibia and was painful in three children (three extremities). Seventeen subsequent procedures were performed in the group that had a Syme amputation; these procedures included proximal tibial epiphyseodesis, osteotomy of the proximal part of the tibia, diaphyseal osteotomy of the tibia, stapling of the medial femoral condyle, and revision of the heel pad. In the group that had a Boyd amputation, seven subsequent operations were done; these included the same procedures, with the exception of revision of the heel pad. Because of the difference in the durations of follow-up, it would be inappropriate to draw any conclusions on the basis of direct comparison of the subsequent procedures in the two groups.

Supracondylar suspension of the prosthesis (by a cuff strap, a neoprene sleeve, or socket design) was necessary for eight extremities (seven children) that had a Syme amputation. Supramalleolar suspension of the prosthesis with an elastomer window or a contoured Pelite liner (Southern Prosthetic Supply, Alpharetta, Georgia) was used for four extremities (four children). The records for the prostheses for the three remaining extremities that had a Syme amputation were incomplete. Four of the twelve extremities were noted to have recurrent problems with prosthetic suspension that interfered with walking and recreational activities. Supramalleolar or, more accurately, supracalcaneal suspension of the prosthesis was successful for fourteen of the sixteen extremities that had a Boyd amputation; supracondylar suspension was necessary for the remaining two extremities because of recurrent problems.

Length of the Residual Limb

After five of the fifteen Syme amputations, a proximal tibial epiphyseodesis was needed to obtain the optimum length of the residual extremity at skeletal maturity. After three other Syme amputations, the cosmetic result was poor because of the increased diameter or thickness at the level of the distal third of the limb and the excessive length of the residual extremity; the prostheses for these limbs were modified to accommodate a dynamic-response foot-and-ankle unit. After three of the sixteen Boyd amputations, supplementary proximal tibial epiphyseodesis was performed to obtain the optimum length of the limb at skeletal maturity. None of the patients who had had early resection of the distal tibial physis, which had been done in all of the patients who had a Boyd amputation, had a limb that was too short at the time of follow-up.

Genu Valgum

The fibular anlage was resected at the time of the initial amputation in twenty-seven of the thirty-one extremities. The condylar height ratio could be calculated in sixteen of the twenty-seven extremities; the remaining patients (eleven extremities) were too young at the time of follow-up for this ratio to be measured. The median ratio was 0.68 (range, 0.56 to 0.95). The fibular anlage was resected without a simultaneous osteotomy of the tibia in fifteen extremities; the median ratio was 0.66 (range, 0.60 to 0.95) in the patients (eleven extremities) who were old enough for accurate measurement. A diaphyseal osteotomy of the tibia and resection of the fibular anlage was performed at the time of the amputation in twelve extremities; the median ratio was 0.70 (range, 0.56 to 0.77) in the five extremities for which it could be measured.

Tibial Bowing

A diaphyseal osteotomy of the tibia was performed at the time of the amputation to correct tibial bowing in twelve extremities, it was not performed initially in sixteen extremities, and the operative or roentgenographic data were incomplete for the remaining three. Preoperatively, the median tibial angulation was 33 degrees (range, 15 to 55 degrees) in the patients who had had a diaphyseal osteotomy of the tibia and 30 degrees (range, 5 to 65 degrees) in the patients who had not. At the follow-up evaluation, the corresponding values were 10 degrees (range, 0 to 25 degrees) and 10 degrees (range, 5 to 40 degrees). There was clinically severe persistent tibial bowing in two extremities that had had an osteotomy; a second osteotomy was done in one of them because of pain and difficulty in prosthetic fitting. Seven of the sixteen extremities that had not had an osteotomy were painful and necessitated prosthetic modifications to accommodate tibial bowing.

Discussion

The strategies of operative treatment for most orthopaedic conditions are continually being modified in response to accurate assessment of outcome, changing demands of patients, and technological advances. Although limb-lengthening procedures are being used for the treatment of longitudinal deficiency of the fibula, early amputation and prosthetic fitting is still preferred for the majority of these children^10,14. Most of the earlier reports have included children who had other major deficiencies of the limb, such as proximal focal femoral deficiency, that alter the clinical and operative issues^{3,10,12,13,16,25,30}. Few investigators have objectively analyzed the techniques of prosthetic management used for their patients. Our review was restricted to children who had deficiency of the fibula (with or without associated femoral hypoplasia), all of whom ultimately had the function of a person with a below-the-knee amputation.

The goal of early amputation is to create an end-bearing residual extremity with a stable heel pad^{3,12,17-19,24}. In adults, this can be achieved reliably with a Syme amputation, which necessitates a subperiosteal dissection of the calcaneus to maintain the integrity of the heel pad. This dissection is poorly suited for children because of the enhanced osteogenic potential of the periosteum. In the present study of children, partial or complete re-formation of the calcaneus occurred in six of the fifteen extremities that had a Syme amputation; this prevalence has been as high as nine of eleven in other studies^{3,10,12,13,19}. Migration of the heel pad, a related phenomenon, has been documented after a Syme amputation in 17 per cent (four of twenty-three) to 45 per cent (nineteen of forty-two) of patients^{3,10,12,13,16,19,30}. With a Boyd amputation, the problems of re-formation of the calcaneus and migration of the heel pad are avoided and the integrity of the heel-pad complex is preserved (presumably maintaining shock absorption and proprioception), making it the technically more appropriate procedure to perform on children who are twelve to eighteen months old. The duration of time for which the tourniquet was in place and the estimated amount of blood loss were similar for the two types of amputation, thus supporting the observation that, with expertise, the two procedures are equally easy to perform. Correct alignment of the calcaneus relative to the tibia at the time of the arthrodesis is the most technically critical aspect of the Boyd amputation, as it determines the ultimate position of the heel pad.

Prosthetic suspension after a Syme amputation for deficiency of the fibula has not been specifically examined in other studies. Syme amputation results in a residual extremity that is more conical, narrow, and distally tapered due to the absence or hypoplasia of the distal part of the fibula^{3,12,16,25,30} (Fig. 3-A). As implied by other investigators and confirmed in our study, most of the patients who had a Syme amputation (eight of twelve extremities) needed some form of supracondylar suspension of the prosthesis. The modified Boyd amputation translates the calcaneus posteriorly at the time of the arthrodesis, creating a contour in the sagittal plane that is cosmetically acceptable and from which the prosthesis can be suspended (Fig. 3-B). Supramalleolar (actually supracalcaneal) suspension of the prosthesis, with use of either a Pelite liner or an elastomer window, was successful in fourteen of the sixteen extremities that had a Boyd amputation.

The perception that disarticulation of the ankle in adults preserves the ability to walk about the house without a prosthesis led to the use of operative procedures that preserve the length of the residual limb^6,9,18. We agree with other investigators that this perception was incorrect; children who have an amputation that results in an end-bearing residual limb stop walking about the house without the prosthesis by the time of early adolescence because of discomfort and instability^3,12,30. An excessively long residual limb creates functional and cosmetic problems with regard to prosthetic fitting. In order to use a dynamic-response foot-and-ankle unit, there must be at least 15.5 centimeters of clearance between the floor and the end of the residual limb. Although longer residual limbs can be accommodated, it is at the cost of alignment and cosmetic appearance. Excessive length of the residual limb also limits the ability to taper the prosthesis at the level of the ankle, further compromising the cosmetic appearance. On the basis of these observations, we propose that optimally the residual limb should reach the level of the middle fifth of the shank segment. When the lengths of the femoral segments are equal, 40 to 60 per cent inhibition of tibial growth in the residual limb is desirable. When the lengths of the femoral segments are not equal, the minimum desired inhibition of tibial growth in the involved limb (ITa) is described by the equation ITa = 1 - ([IFa][LFo/LTo] + 0.6), where IFa is the inhibition of femoral growth of the involved limb, LFo is the length of the contralateral femur at maturity⁴, and LTo is the length of the contralateral tibia at maturity⁴. The maximum desired inhibition of tibial growth is described by a similar equation, with 0.4 substituted for 0.6. With this equation, we determined that, when the inhibition of femoral growth is 5 per cent, the inhibition of tibial growth should be 35 to 55 per cent; when the inhibition of femoral growth is 15 per cent, the desirable inhibition of tibial growth is 20 to 40 per cent.

Although the prediction of limb length is an inexact science, our protocol of resection of the distal tibial physis at the time of Boyd amputation (with subsequent supplementary proximal tibial epiphyseodesis, when necessary) resulted in an optimum length of the residual limb with respect to prosthetic fitting (no limb was too short following this protocol). Careful follow-up with orthoroentgenograms made intermittently after amputation is mandatory, as some children may need supplemental epiphyseodesis of the proximal part of the ipsilateral tibia. The optimum time for this assessment is when the child is six to ten years old. At this time, the length of the unaffected extremity at maturity can be estimated from standard growth charts, and the length of the affected limb at maturity can be estimated by the addition of the percentage of growth inhibition of the ipsilateral femur and the remaining growth of the tibia (best estimated by the rule-of-thumb technique of 0.6 centimeter per year from the proximal tibial physis). With respect to the length of the residual limb after a below-the-knee amputation, more is not necessarily better; this is reflected by the problems with the prosthesis that were related to the length of the residual limb in three of the fifteen patients in whom the distal tibial physis had been preserved with a Syme amputation.

Late, progressive genu valgum deformity has been reported in five of seventeen patients and fourteen of twenty-four patients^1,7; it is believed to be secondary to fibular tethering, suppression of growth of the lateral femoral condyle due to the altered loading of the knee after prosthetic accommodation of the anteromedial bowing of the tibial diaphysis (Heuter-Volkmann law), or simply further expression of the insult to the congenital limb bud associated with the more obvious fibular deficiency^{2,7,11,14,20,21,26,30}. Although the deformity usually has been described on the basis of the tibiofemoral angle, the condylar height ratio provides a more accurate assessment of the degree of hypoplasia of the lateral femoral condyle. Boakes et al.⁷ evaluated twenty-four affected extremities (twenty patients) and determined that the mean condylar height ratio for the uninvolved side was 0.90 and the mean value for the side with deficiency of the fibula (treated with early Syme amputation only) was 0.75. The lack of improvement in this ratio in our patients who had had early resection of the fibular anlage (median ratio, 0.68) suggests that the tethering effect of the hypoplastic fibular anlage does not contribute to the etiology of the genu valgum deformity. Failure of this ratio to improve after performance of an early diaphyseal osteotomy of the tibia to correct anteromedial bowing (median ratio, 0.70) suggests that suppression of growth from abnormal loading of the lateral compartment of the knee due to mechanical malalignment also is not a factor in the development of the genu valgum deformity. Therefore, on the basis of our results, we believe that the insult to the limb bud is the primary cause of the hypoplasia of the lateral femoral condyle that is associated with deficiency of the fibula. The association of other deficiencies of the ipsilateral femur, such as femoral hypoplasia and proximal femoral focal deficiency, with deficiency of the fibula would seem to lend additional support to this theory. We agree with other investigators that the genu valgum deformity is best treated late, with stapling of the medial femoral physis (when possible) or an osteotomy of the distal end of the femur, to improve both the mechanical alignment of the entire residual limb and the obliquity of the knee joint^1,7,13,19 (Figs. 4-A and 4-B).

The value of an early tibial osteotomy to correct anteromedial bowing has been controversial^6,7,10,16,30. Our study has shown tibial osteotomy to be reliable and effective, without problems with healing or recurrence of the deformity. Evaluation of the patients who did not have an early tibial osteotomy confirmed the observations of other investigators that the tibial bowing will correct with growth over a period of years^{3,15,20,23,30}. However, we believe that our recommendation for early correction of the tibial bowing is supported by the increased prevalence of pain and problems with the prosthesis related to the tibial bowing in patients who did not have an early tibial osteotomy.

On the basis of the findings of our study, we recommend (1) a modified Boyd amputation (instead of a Syme amputation) for technical, functional, and cosmetic reasons; (2) early resection of the distal tibial physis, with subsequent proximal tibial epiphyseodesis when necessary, to optimize the length of the residual limb and facilitate (both functionally and cosmetically) prosthetic fitting; and (3) early correction of tibial bowing to facilitate prosthetic fitting over the subsequent years. We found that early resection of the fibular anlage or early diaphyseal osteotomy of the tibia did not alter the natural history of hypoplasia of the lateral femoral condyle and associated clinical genu valgum deformity seen in children who had longitudinal deficiency of the fibula.

References

Achterman, C., and |and |Kalamchi, A.: Congenital deficiency of the fibula. J. Bone and Joint Surg.,61-B(2): 133-137, 1979.61-B(2)133 1979

Amstutz, H. C.: Natural history and treatment of congenital absence of the fibula. Proceedings of The American Academy of Orthopaedic Surgeons. J. Bone and Joint Surg.,54-A: 1349, Sept. 1972.54-A1349 1972

Anderson, L.; Westin, G. W.; and |and |Oppenheim, W. L.: Syme amputation in children: indications, results, and long-term follow-up. J. Pediat. Orthop.,4: 550-554, 1984.4550 1984

Anderson, M.; Messner, M. B.; and |and |Green, W. T.: Distribution of lengths of the normal femur and tibia in children from one to eighteen years of age. J. Bone and Joint Surg.,46-A: 1197-1202, Sept. 1964.46-A1197 1964

Arnold, W. D.: Congenital absence of the fibula. Clin. Orthop.,14: 20-29, 1959.1420 1959

Blum, C. E., and |and |Kalamchi, A.: Boyd amputations in children. Clin. Orthop.,165: 138-143, 1982.165138 1982 [PubMed]

Boakes, J. L.; Stevens, P. M.; and |and |Moseley, R. F.: Treatment of genu valgus deformity in congenital absence of the fibula. J. Pediat. Orthop.,11: 721-724, 1991.11721 1991 [CrossRef]

Bohne, W. H. O., and |and |Root, L.: Hypoplasia of the fibula. Clin. Orthop.,125: 107-112, 1977.125107 1977 [PubMed]

Boyd, H. B.: Amputation of the foot, with calcaneotibial arthrodesis. J. Bone and Joint Surg.,21: 997-1000, Oct. 1939.21997 1939

Choi, I. H.; Kumar, S. J.; and |and |Bowen, J. R.: Amputation or limb-lengthening for partial or total absence of the fibula. J. Bone and Joint Surg.,72-A: 1391-1399, Oct. 1990.72-A1391 1990

Coventry, M. B., and |and |Johnson, E. W., Jr.: Congenital absence of the fibula. J. Bone and Joint Surg.,34-A: 941-955, Oct. 1952.34-A941 1952

Davidson, W. H., and |and |Bohne, W. H. O.: The Syme amputation in children. J. Bone and Joint Surg.,57-A: 905-909, Oct. 1975.57-A905 1975

Eilert, R. E., and |and |Jayakumar, S. S.: Boyd and Syme ankle amputations in children. J. Bone and Joint Surg.,58-A: 1138-1141, Dec. 1976.58-A1138 1976

Epps, C. H., Jr., and |and |Schneider, P. L.: Treatment of hemimelias of the lower extremity. Long-term results. J. Bone and Joint Surg.,71-A: 273-277, Feb. 1989.71-A273 1989

Farmer, A. W., and |and |Laurin, C. A.: Congenital absence of the fibula. J. Bone and Joint Surg.,42-A: 1-12, Jan. 1960.42-A1 1960

Fergusson, C. M.; Morrison, J. D.; and |and |Kenwright, J.: Leg-length inequality in children treated by Syme's amputation. J. Bone and Joint Surg.,69-B(3): 433-436, 1987.69-B(3)433 1987

Greene, W. B., and |and |Cary, J. M.: Partial foot amputations in children. A comparison of the several types with the Syme amputation. J. Bone and Joint Surg.,64-A: 438-443, March 1982.64-A438 1982

Harris, R. I.: Syme's amputation. The technical details essential for success. J. Bone and Joint Surg.,38-B(3): 614-632, 1956.38-B(3)614 1956

Herring, J. A.; Barnhill, B.; and Gaffney, C.: Syme amputation. An evaluation of the physical and psychological function in young patients. J. Bone and Joint Surg.,68-A: 573-578, April 1986.68-A573 1986

Hootnick, D.; Boyd, N. A.; Fixsen, J. A.; and |and |Lloyd-Roberts, G. C.: The natural history and management of congenital short tibia with dysplasia or absence of the fibula. A preliminary report. J. Bone and Joint Surg.,59-B(3): 267-271, 1977.59-B(3)267 1977

Kruger, L. M.: Fibular hemimelia. In Selected Lower-Limb Anomalies. Surgical and Prosthetics Management, pp. 49-71. Washington, D.C., National Academy of Sciences, 1971.

Kruger, L. M.: Recent advances in surgery of lower limb deficiencies. Clin. Orthop.,148: 97-105, 1980.14897 1980 [PubMed]

Kruger, L. M., and |and |Talbott, R. D.: Amputation and prosthesis as definitive treatment in congenital absence of the fibula. J. Bone and Joint Surg.,43-A: 625-642, July 1961.43-A625 1961

McCullough, N. C.; Matthews, J. G.; Traut, A.; and |and |Caldwell, J.: Early opinions concerning the importance of bony fixation of the heel pad to the tibia in the juvenile amputee. Inter-Clin. Info. Bull.,3: 1-16, 1964.31 1964

Mazet, R., Jr.: Syme's amputation. A follow-up study of fifty-one adults and thirty-two children. J. Bone and Joint Surg.,50-A: 1549-1563, Dec. 1968.50-A1549 1968

Ollerenshaw, R.: Congenital defects of the long bones of the lower limb. A contribution to the study of their causes, effects, and treatment. J. Bone and Joint Surg.,7: 528-552, July 1925.7528 1925

Serafin, J.: A new operation for congenital absence of the fibula. Preliminary report. J. Bone and Joint Surg.,49-B(1): 59-65, 1967.49-B(1)59 1967

Thomas, I. H., and |and |Williams, P. F.: The Gruca operation for congenital absence of the fibula. J. Bone and Joint Surg.,69-B(4): 587-592, 1987.69-B(4)587 1987

Thompson, T. C.; Straub, L. R.; and |and |Arnold, W. D.: Congenital absence of the fibula. J. Bone and Joint Surg.,39-A: 1229-1237, Dec. 1957.39-A1229 1957

Westin, G. W.; Sakai, D. N.; and |and |Wood, W. L.: Congenital longitudinal deficiency of the fibula. Follow-up of treatment by Syme amputation. J. Bone and Joint Surg.,58-A: 492-496, June 1976.58-A492 1976

Wood, W. L.; Zlotsky, N.; and |and |Westin, G. W.: Congenital absence of the fibula. Treatment by Syme amputation—indications and technique. J. Bone and Joint Surg.,47-A: 1159-1169, Sept. 1965.47-A1159 1965

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-B: Postoperative roentgenogram of the extremity after a modified Boyd amputation (our preferred method of treatment) and diaphyseal osteotomy of the tibia.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C: Roentgenograms made eight years later. The calcaneus was posteriorly displaced, providing excellent prosthetic suspension. The corrected tibial alignment had been maintained.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-B: The result after stapling of the medial femoral physis. Both the tibiofemoral alignment and the obliquity of the knee joint were corrected.

View Larger

TABLE I DATA ON THE PATIENTS*

Case	Gender, Age (Yrs. + Mos.)	Side	Type of Fib. Deficiency¹	Growth Inhibition† (Per cent)		At Op.			At Follow-up			Complications	Add. Op.	Duration of Follow-up (Yrs. + Mos.)
Femur	Tibia	Fibular Anlage	Tibial Bowing (Degrees)	Tibial Osteot.	Condylar Height Ratio‡	Tibial Bowing§ (Degrees)	Tibio-femoral Angle§ (Degrees)
Syme
1	F, 0 + 11	R	II	—	—	Yes	54	No	0.95	10	7	Tibial bowing	—	12 + 3
		L	II	—	—	Yes	50	No	0.95	10	7	Tibial bowing	—
2	M, 1 + 2	L	II	5	20	Yes	15	Yes	0.66	0	20/5	Genu valgum and limb-length discrep.	4	9 + 1
3	M, 1 + 1	R	IB	10	28	Yes	40	Yes	0.70	25/10	5	Tibial bowing	2	13 + 5
4	F, 1 + 1	L	II	11	15	Yes	30	No	0.64	10	14	Tibial bowing	—	13 + 3
5	M, 0 + 11	R	II	0	27	Yes	25	No	0.64	10	16/5	Genu valgum and limb-length discrep.	2	12 + 11
6	M, 1 + 4	L	II	5	38	Yes	55	No	0.64	20	20/5	Tibial bowing, genu valgum, and limb-length discrep.	3	14 + 3
7	F, 6 + 8	R	IB	0	13	No	5	No	0.86	5	8	Limb-length discrep.	1	11 + 4
8	M, 0 + 10	R	II	—	—	Yes	60	No	NA	30	7	NA	—	4 + 10
		L	II	—	—	Yes	50	No	NA	25	0	NA	—
9	M, 0 + 9	R	II	NA	NA	Yes	NA	No	0.80	NA	10	—	—	10 + 11
10	M, 1 + 7	R	II	2	34	Yes	30	No	0.66	20	16	NA	—	8 + 8
11	M, 0 + 9	L	II	5	36	Yes	30	No	0.60	20	16	—	—	16 + 11
12	F, 2 + 2	R	IB	6	34	Yes	40	No	0.71	20/0	27/0	Tibial bowing, genu valgum, and limb-length discrep.	3	12 + 6
13¶	M, 1 + 8	R	II	—	—	Yes	65	No	0.75	40/10	6	Tibial bowing	2	15 + 6
Boyd
13¶	M, 7 + 6	L	IB	—	—	No	10	No	0.72	10	20/-4	Genu valgum and limb-length discrep.	2	9 + 6
14	F, 1 + 1	R	II	4	NA	Yes	NA	No	0.66	10	21/0	Tibial bowing and genu valgum	1	11 + 3
15	M, 2 + 0	R	II	—	—	Yes	25	No	0.56	5	15	—	—	4 + 11
		L	II	—	—	Yes	35	Yes	0.77	7	12	—	—
16	M, 3 + 3	R	II	2	35	Yes	21	Yes	0.74	10	20/NA	Genu valgum and limb-length discrep.	2	6 + 2
17	F, 1 + 1	R	II	—	—	Yes	30	Yes	NA	0	14	—	—	3 + 5
		L	II	—	—	Yes	30	Yes	NA	15	15	—	—
18	F, 1 + 1	L	II	14	21	Yes	13	No	NA	10	7	—	—	3 + 5
19	M, 0 + 9	R	II	NA	NA	Yes	50	Yes	NA	20	14	—	—	4 + 11
20	F, 9 + 8	R	II	3	18	No	5	No	0.64	5	15	—	—	3 + 1
21	M, 1 + 2	L	II	3	NA	No	NA	No	0.89	10	4	Limb-length discrep.	1	6 + 11
22	F, 1 + 4	R	II	—	—	Yes	43	Yes	NA	15	16	—	—	4 + 7
		L	II	—	—	Yes	29	Yes	NA	10	18	—	—
23	M, 0 + 8	L	II	0	43	Yes	55	Yes	NA	14	7	—	—	3 + 11
24	M, 1 + 1	L	II	3	43	Yes	50	Yes	NA	8	5	Tibial bowing	1	3 + 11
25	M, 3 + 5	R	II	10	11	Yes	10	No	NA	10	10	—	—	2 + 6

Add to My JBJS

TABLE I DATA ON THE PATIENTS*

Case	Gender, Age (Yrs. + Mos.)	Side	Type of Fib. Deficiency¹	Growth Inhibition† (Per cent)		At Op.			At Follow-up			Complications	Add. Op.	Duration of Follow-up (Yrs. + Mos.)
Femur	Tibia	Fibular Anlage	Tibial Bowing (Degrees)	Tibial Osteot.	Condylar Height Ratio‡	Tibial Bowing§ (Degrees)	Tibio-femoral Angle§ (Degrees)
Syme
1	F, 0 + 11	R	II	—	—	Yes	54	No	0.95	10	7	Tibial bowing	—	12 + 3
		L	II	—	—	Yes	50	No	0.95	10	7	Tibial bowing	—
2	M, 1 + 2	L	II	5	20	Yes	15	Yes	0.66	0	20/5	Genu valgum and limb-length discrep.	4	9 + 1
3	M, 1 + 1	R	IB	10	28	Yes	40	Yes	0.70	25/10	5	Tibial bowing	2	13 + 5
4	F, 1 + 1	L	II	11	15	Yes	30	No	0.64	10	14	Tibial bowing	—	13 + 3
5	M, 0 + 11	R	II	0	27	Yes	25	No	0.64	10	16/5	Genu valgum and limb-length discrep.	2	12 + 11
6	M, 1 + 4	L	II	5	38	Yes	55	No	0.64	20	20/5	Tibial bowing, genu valgum, and limb-length discrep.	3	14 + 3
7	F, 6 + 8	R	IB	0	13	No	5	No	0.86	5	8	Limb-length discrep.	1	11 + 4
8	M, 0 + 10	R	II	—	—	Yes	60	No	NA	30	7	NA	—	4 + 10
		L	II	—	—	Yes	50	No	NA	25	0	NA	—
9	M, 0 + 9	R	II	NA	NA	Yes	NA	No	0.80	NA	10	—	—	10 + 11
10	M, 1 + 7	R	II	2	34	Yes	30	No	0.66	20	16	NA	—	8 + 8
11	M, 0 + 9	L	II	5	36	Yes	30	No	0.60	20	16	—	—	16 + 11
12	F, 2 + 2	R	IB	6	34	Yes	40	No	0.71	20/0	27/0	Tibial bowing, genu valgum, and limb-length discrep.	3	12 + 6
13¶	M, 1 + 8	R	II	—	—	Yes	65	No	0.75	40/10	6	Tibial bowing	2	15 + 6
Boyd
13¶	M, 7 + 6	L	IB	—	—	No	10	No	0.72	10	20/-4	Genu valgum and limb-length discrep.	2	9 + 6
14	F, 1 + 1	R	II	4	NA	Yes	NA	No	0.66	10	21/0	Tibial bowing and genu valgum	1	11 + 3
15	M, 2 + 0	R	II	—	—	Yes	25	No	0.56	5	15	—	—	4 + 11
		L	II	—	—	Yes	35	Yes	0.77	7	12	—	—
16	M, 3 + 3	R	II	2	35	Yes	21	Yes	0.74	10	20/NA	Genu valgum and limb-length discrep.	2	6 + 2
17	F, 1 + 1	R	II	—	—	Yes	30	Yes	NA	0	14	—	—	3 + 5
		L	II	—	—	Yes	30	Yes	NA	15	15	—	—
18	F, 1 + 1	L	II	14	21	Yes	13	No	NA	10	7	—	—	3 + 5
19	M, 0 + 9	R	II	NA	NA	Yes	50	Yes	NA	20	14	—	—	4 + 11
20	F, 9 + 8	R	II	3	18	No	5	No	0.64	5	15	—	—	3 + 1
21	M, 1 + 2	L	II	3	NA	No	NA	No	0.89	10	4	Limb-length discrep.	1	6 + 11
22	F, 1 + 4	R	II	—	—	Yes	43	Yes	NA	15	16	—	—	4 + 7
		L	II	—	—	Yes	29	Yes	NA	10	18	—	—
23	M, 0 + 8	L	II	0	43	Yes	55	Yes	NA	14	7	—	—	3 + 11
24	M, 1 + 1	L	II	3	43	Yes	50	Yes	NA	8	5	Tibial bowing	1	3 + 11
25	M, 3 + 5	R	II	10	11	Yes	10	No	NA	10	10	—	—	2 + 6