Proximal Tibio-Fibular Bifurcation Synostosis for the Management of Longitudinal Deficiency of the Tibia

Jon R. Davids, M.D. Leslie C Meyer, M.D.


Three adults with severe longitudinal deficiency of the tibia (LDT), in which an unossified proximal tibial anlage was present, who had been managed with proximal tibio-fibular bifurcation synostosis (PTFBS) in early childhood, were evaluated between 19 years and 31 years following the index procedure. All 3 were found to be functioning well as below-the-knee (BK) amputees. Medio-lateral stability and anteroposterior instability of the knee were present in all cases. Instrumented motion analysis revealed diminished loading characteristics of the prosthetic limb, similar to that described for BK amputees in general. The most significant gait deviations at the knee unique to this study group were a quadriceps avoidance gait pattern and an increased dynamic varus alignment. Instrumented muscle testing suggested that these deviations were a consequence of ligamentous instability. This study supports the concept that the presence of a proximal tibial anlage in severe LDT is indication for a surgical strategy that preserves the biological knee joint. The PTFBS maintains the integrity of the knee extensor mechanism, the fibular collateral ligament, the tibiofemoral joint capsule, and the medial collateral ligament, enhancing the long term stability and function of the knee joint. Key Words: Longitudinal Deficiency of the Tibia-Proximal Tibio-fibular Bifurcation Synostosis- Gait Analysis.


Comprehensive surgical management of longitudinal deficiency of the tibia (LDT) must consider the issues of [1] knee joint deformity and stability, [2] ankle/foot deformity and stability, and [3] limb length inequality. When possible, a strategy that preserves or reconstructs the knee, and stabilizes the ankle/foot, which will allow the individual to function as a below-the-knee (BK) amputee, is favored. (6,16,17,18,21,23,27,31,33). When this is not possible, a through-the-knee (TK) amputation will provide the best level of function (6,10,17,18,20,27,31,33). The optimal plan is determined based upon the magnitude of the tibial deficiency. When the tibia is completely absent, TK amputation is the current treatment of choice. Surgical centralization of the fibula beneath the femur, while attractive in theory, has been associated with significant complications and no longer performed at most major centers, (10,21,31). When the proximal tibia is present and the distal tibia is either absent or dysplastic, stabilization of the ankle/foot by disarticulation (Syme amputation) or fibulo-calcaneal fusion (modified Boyd amputation) is performed in conjunction with side-to-side proximal tibio-fibular synostosis (17,18,23,27,33). Individuals managed by this protocol have been found to function well as BK amputees ;

In some cases, the tibia appears to be completely absent on plain radiographs, yet a cartilaginous anlage of the proximal tibia is appreciated by other imaging modalities (e.g. ultrasound, magnetic resonance imaging, anthrog-raphy) or at the time of surgery (8,13,17, 23,27,33). Multiple management strategies have been proposed for this circumstance, including TK amputation, "modified" fibular centralization (without description of the actual procedure) and observation (6,17,18,23,27,31,32,33,35). At our institution, a procedure for this circumstance, the proximal tibiofibular bifurcation synostosis (PTFBS), has been developed based upon an appreciation of the developmental pathoanatomy of the knee joint associated with LDT and the functional demands during BK amputee gait. A description of this procedure and long-term follow up, including objective functional assessment of outcome, is reported for three individuals managed with PTFBS in early childhood.

Materials and Methods

A retrospective review of the medical records of all children with the diagnosis of LDT treated at our institution over the last 35 years was performed to identify those who had been managed with a BK amputation strategy. Of the 52 children with this diagnosis, 18 were functioning as BK amputees at the most recent follow up. Within this group were three individuals who had been managed with PTFBS. All three were contacted and brought back for a detailed clinical history, physical examination, radiographic examination, instrumented motion analysis, and instrumented muscle testing. Instrumented motion analysis was performed using a six camera retro reflective system (Vicon 370, Oxford Metrics Limited, Oxford, UK), a two camera editing quality picture video system. (Camscope S-VHS AG-455M, Panasonic, Se-caucus, NJ, USA), and two force platforms (Advanced Medical Technologies, Newton, MA,USA). The data was processed and joint kinematics and kinetics were calculated using Vicon Clinical Manager software (Vicon 27, Oxford Metrics Limited, Oxford, UK.). Instrumented testing of the quadriceps and hamstring muscles was performed using the Kin-Corn II (Chattecx Corp., Chattanooga, TN, USA). Concentric and eccentric muscle function were tested at a constant velocity through a 69 degree arc of motion for both lower extremities, and the involved side was then compared to the normal, unin-volved side.

Surgical Technique:
The knee joint is approached through a slightly curved, transverse incision just distal to the femoral condyles. (23) The extensor mechanism is identified, preserved, and followed distally to identify the location of the tibial anlage. The distal most portion of the tibial anlage is dissected free and a tunnel is created centrally at the inferior margin. Care is taken not to disrupt the tibiofemoral articulation and associated soft tissue structures, particularly at the medial margin, where the medial collateral ligament is located. The proximal fibula is identified and the periosteum is opened longitudinally at the medial margin, beginning just distal to the proximal fibular physis. A transverse metaphyseal osteotomy of the proximal fibula is performed, and the shaft of the fibula is translated medially, leaving the fibular sleeve intact (Fig. 1. ). The proximal portion of the shaft of the fibula is placed within the tunnel in the tibial an-lage, and the latter is impaled upon the former. This aligment is maintained with deeply buried crossed Kirschner wires. The extremity is immobilized for 6 weeks, after which gentle active and passive knee range of motion is begun. Weight bearing is begun 12 weeks following the surgery, once the periosteal sleeve has reconstituted a new segment of the proximal fibula. Until the tibial anlage has ossified, union at the synostosis site is difficult to determine. The Kirschner wires may be removed once bony union is appreciated.


The study group consisted of 2 females and 1 male. Age at the time of follow up ranged from 23 years and 6 months to 36 years and 9 months. Age at the time PTFBS ranged from 4 years to 6 years and 7 months. Follow up after this procedure ranged from 19 years and 6 months to 31 years and 2 months. All patients had a Syme amputation prior to the PTFBS (age range 1 year and 1 month to 2 years and 9 month; interval between procedures 2 years and 3 months to 2 years and 11 months). Two of the 3 patients had a proximal fibula epiphy-seodesis for a prominent fibular head following the PTFBS (age range 6 years and 11 months to 7 years and 1 month; interval between procedures 2 years to 2 years and 11 months).

Clinical History:
Two of the 3 subjects were gainfully employed outside the home at the time of follow up. The third was currently a homemaker with 3 children and had been employed prior to having children. Employment included bartending, auto mechanic, retail sales, restaurant food preparation, and grocery bagger/stocker. All 3 enjoyed recreational activities such as bowling, swimming, golf, volleyball, racquetball, tennis, and badminton. None of the subjects felt that they were limited by their limb deficiency, though one did complain of occasional mild knee pain and swelling after standing for periods greater than 5 hours. Two of the 3 preferred prostheses with thigh lacers, 1 since childhood, the other since her teenage years. The third favored a patellar tendon bearing, supracondylar prosthetic design and used a neoprene sleeve when playing sports. Prosthetic maintenance and replacement had been funded by Medicaid, state vocational rehabilitation services, and work related private insurance.

Physical Examination:
The pelvis, both hips, and the contralateral knee and ankle were all within normal with respect to range of motion, stability, and strength. The arc of active motion of the affected knee ranged from 105 degrees to 145 degrees. Knee extension ranged from -15 degrees to 0 degrees. Knee flexion ranged from 115 degrees to 150 degrees. Patellofemoral examination was normal in 2 of the subjects. The third subject (who complained of mild knee pain and swelling) had obligatory subluxation of the patella. The patella was nontender, irreducible, and had a negative apprehension test. The medial and lateral collateral ligaments were stable to stress testing with the knee flexed to 5 degrees for all of the subjects. The clinical impression was that the lateral knee stabilizers were somewhat "tighter" than the medial stabilizers. Increased anteroposterior laxity was appreciated for all subjects on the Lachman's, anterior drawer, and tibial sag tests. Despite excessive excursion of the tibia both anteriorly and posteriorly, firm endpoints in both directions were appreciated in each case. Manual muscle testing of the quadriceps, performed with knee fully extended, was grade 4 (possible to "break" the muscle) in 1 case. Manual muscle testing of the hamstrings, performed with the knee flexed to 90 degrees, was similar to that of the quadriceps. The clinical impression was that the hamstrings were generally somewhat stronger than the quadriceps. Observational gait analysis revealed a relatively normal, symmetric, and comfortable gait pattern.

Radiographic Examination:
Radiographs taken within the first year of life showed no evidence of a proximal tibia (Fig. 2. ). The proximal fibula was proximally migrated in 1 of the 3 subjects, and the dimensions of the distal femur on the affected side were comparable to the opposite side in only one of the cases. At the time of PTFBS a faint ossification center of the proximal anlage was present in 2 of the 3 cases. In the third case, the tibial anlage was completely unossi-fied. Union between the proximal tibia and the translated fibular shaft could not be confirmed radiographically until between 24 and 36 months following PTFBS, by which time the tibial anlage had ossified completely. At the most recent follow up, the tibio-femoral articulation was congruous in 2 cases and incongruous in 1 case. Squaring of the medial femoral condyle and narrowing of the medial compartment joint space was present in the oldest subject.

Instrumented Motion Analysis:
Time-distance parameters for the >subjects are summarized in Table 1. . Self-selected walking velocity ranged from 80% to 96% of age matched normals. Diminished velocity was a consequence of decreased stride length and cadence, the latter being diminished to a greater degree. Diminished duration of single limb stance for the affected side, suggesting some degree of pain, weakness, or instability of the prosthetic limb, was present for all cases, and led to decreased contralateral step length. Both periods of double limb support were increased, particularly the initial period for the affected side, suggesting altered ability to load the prosthetic limb. Force plate analysis revealed diminished magnitude of both the early stance phase vertical force peak and deceleration shear force peak, further reflecting the limited loading characteristics of the affected extremity. None of these deviations were major and all were difficult to appreciate on real time observational gait analysis.

Dynamic knee range of motion for the affected extremity was normal for all of the cases (approximately 0 to 60 degrees flexion). Kinematic analysis of the knee in stance phase revealed a quadriceps avoidance pattern in all cases, with absence or diminution of the sagital plane flexion wave during loading response (Fig. 3. ). Kinetic analysis confirmed this deviation in 2 of the subjects, exhibiting an increased internal flexion pattern (i.e. increased external extension moment pattern ) during stance phase. A stance phase varus thrust was appreciated on kinetic analysis for all of the subjects. (Fig. 4. ) Kinetic analysis confirmed an increased internal valgus moment pattern relative to the contralateral knee (i.e. increased external varus moment pattern) in 2 of the subjects. The absolute stance phase external varus moment was diminished for all of the cases. Swing phase kinematics were within one standard deviation of normal for the timing and magnitude of peak knee flexion and the magnitude of knee extension at terminal swing. Instrumented Muscle Testing: Quadriceps strength of the affected extremity in the concentric testing mode ranged from 17 to 29% of the contralateral, otherwise normal extremity. When tested eccentrically, quadriceps strength was determined to be 17 to 22% of the opposite extremity. Concentric strength of the hamstring muscles on the affected side ranged from 52 to 86% relative to the opposite side. Eccentric testing of this muscle group showed the strength of the affected side to range between 49 and 81% of the unaffected side.


Early, definitive surgical management of the deformities associated with LDT, designed to facilitate prosthetic fitting and age appropriate ambulation, is the current strategy favored at most centers (10,17,18,27,31,33). Surgical decision making is based upon determination of the degree of tibial deficiency: when the proximal tibia is obviously present, a strategy that preserves the knee joint is appropriate; when the tibia is completely absent, a TK amputation is performed. The optimal management strategy when a non-ossified, cartilaginous tibial an-lage is present has not been well established. The procedure favored for this situation at our institution, the PTFBS, was developed based upon the embry-ologic pathoanatomy associated with LDT. Embryologic development is a continuous sequence of structural alterations that proceeds in an orderly fashion. During the embryologic period, the extremities develop in a cranio-caudal sequence, with each segment of the extremity developing in a proximodistal direction. (15,6) About the knee chon-drification of the distal femur, proximal tibia, proximal fibula, and the patellar tendon occurs sequentially around day 28. (11, 12,14,15) The femoral condyle, the patella, and the lateral collateral ligament develop over the next few days. (11,12,14,15) By day 38, the medial collateral ligament, the knee joint, the anterior and posterior cruciate ligaments, and the menisci have sequentially become distinct structures. (11,12, 14, 15) At this point, all of the anatomical components of the knee joint are present. The subsequent fetal period is associated primarily with growth of the existing structures. Although the exact etiology of LDT is not known, anatomical studies of specimens obtained at the time of amputation suggest arrested development during the embryologic period, with the persistence of primitive neurovascular and musculoskeletal structures. (22,34,36 ) From this information, it can be deduced that the presence of a distinct proximal tibia anlage implies the presence of an intact knee extensor mechanism, joint capsule, and medial collateral ligament. This observation is supported by the majority of clinical reports describing the findings at the time of surgery when a non-ossified tibial anlage has been appreciated, and by ultrasound imaging studies performed to determine the presence and characteristics of the tibial anlage. ( 8, 13,18,21,23,27,31,33) The clinical experience at our institution supports the corollary observation that the presence of an intact extensor mechanism implies the presence of a proximal tibial anlage.

The function of the joint and the extremity following PTFBS, as assessed by instrumented motion analysis, was generally comparable to that described for BK amputees. (1,7,9,19) Diminished loading characteristics of the prosthetic limb, including decreased self-selected velocity, diminished stance time, and decreased rate and magnitude of vertical loading, were appreciated for all of the cases. Decreased power generation by the prosthetic ankle/foot was compensated for by increased power generation by the ipsilateral hip extensors during stance phase. The most significant gait deviations unique to the study group were a quadriceps avoidance gait pattern and an increased stance phase dynamic varus alignment at the knee. The quadriceps avoidance pattern, characterized by a flattened sagittal plane stance phase knee flexion wave and increased external knee extension moment during stance phase, was similar to that seen in adults with acquired anterior cruciate ligament deficiency. (3) Quadriceps weakness can also lead to to the incorporation of a quadriceps avoidance pattern. (30) Although static manual muscle testing suggested strong quadriceps and hamstrings in all of our subjects, dynamic instrumented muscle testing revealed significant strength deficiencies relative to the contralateral, normal side. The inadequacy of manual testing of large, strong muscle groups in the lower extremity has been objectively documented. ( 2,28, 29) Additionally, the dynamic nature of the instrumented muscle testing highlighted the significance of the anteroposterior knee ligament instability, which clearly compromised the ability of the quadriceps and hamstring muscles to perform optimally. These observations imply that the quadriceps avoidance gait pattern seen in the current study was more a consequence of knee joint instability than knee extensor mechanism weakness. The increased stance phase varus alignment seen at the knee may be a consequence of the contour of the residual limb, prosthetic malalignment, or a compensatory gait deviation. The residual limb alignment and the prosthetic fitting were judged to be adequate on the physical examination. Clinical and radiographic analysis of the knee suggested that the lateral compartment and associated soft tissue stabilizing structures were generally better developed than on the medial side. As a compensatory dynamic gait deviation, an increased varus alignment at the knee during stance phase would preferentially load the lateral stabilizers, which might provide enhanced stability.

The current study supports the algorithm that the presence of a proximal tibial anlage is indication for a surgical strategy that preserves the biological knee joint. In such a circumstance, we favor a Syme or modified Boyd amputation at 10 to 12 months of age, which generally facilitates ambulation with a BK style prosthesis. Once the tibial anlage is large enough (usually between 3 and 7 years of age), PTFBS is performed to further stabilize the knee joint. Subsequent proximal fibular epiphyseodesis for a prominent fibular head may be necessary. The PTFBS preserves the integrity of the knee extensor mechanism, the fibular collateral ligament, the tibiofemoral joint capsule, and the medial collateral ligament, enhancing the long term stability of the knee joint. The sagittal plane instability of the knee appreciated in all of the cases suggests the absence of the anterior and posterior cruciate ligaments, which are among the latest structures to develop during the embryologic period. Functional analysis of gait following PTFBS suggests that an optimal outcome at the knee is a reflection at joint stability, muscle strength, and range of motion. The persistent confusion in the literature regarding an alternative procedure utilized in the management of severe LDT, fibular centralization as popularized by Brown, can be resolved by applying the principles appreciated in the analysis of the PTFBS proce-dure.(4,5,24,25) Proponents of fibular centralization suggest that an essential prerequisite for the procedure is the "good" quadriceps function. (5,16,32,35) While the presence of anti-gravity knee extensor function can often be appreciated early on, we have generally found it difficult to accurately assess quadriceps function in the first months of life. When such function is present, embryologic data, noninvasive imaging studies, and the bulk of the reported clinical cases suggest that a tibial an-lage is also present. (11,13,17,18,27) In fibular centralization, the lateral collateral ligament and all of the soft tissue structures between the distal femur and the fibular head are excised to allow for the creation of a femora-fibular articulation. (4,5,3235) Femoral shortening is recommended to further facilitate femora-fibular alignment. (5,32) Although impressive modelling of the proximal fibula following centralization has been reported, the inevitably associated joint instability and muscle weakness significantly compromise the ultimate functional outcome. (6,10,16,18,21,27,31,33) Based upon the excellent and predictable outcome following TK amputation in children ( an end-bearing residual limb, possible prosthetic suspension off the distal femoral condylar flare, and ability to establish the optimal length of the residual limb with distal femoral epiphyseodesis), and the good long term functional outcome reported here for PTFBS, we agree with other investigators who suggest that the fibular centralization procedure as popularized by Brown has no place in the current algorithm for the surgical management of severe LDT. (10,18,20,21,27,31) Finally, we favor the classification of LDT proposed by Kalamchi and Dawe over the more widely accepted classification proposed by Jones et al because the former is a more consistent guide to prognosis and surgical management. (Table 2. ) (17,18) In our current protocol, a Jones la deficiency is managed by early TK amputation, while a type of lb will be imaged or explored surgically to determine the absence or presence of a tibial anlage. If present, knee salvage with PTFBS will be performed. This algorithm is more consistent with a slightly modified Kalamchi and Dawe's classification, in which all type 1 deficiencies (tibia is completely absent) are managed by TK amputation, and all type II deficiencies (2b proximal tibial is present but not ossified, 2b proximal tibia is present and ossified) are managed by a variety of knee sparing, BK amputation strategies.

Shriners Hospital for Children,Greenville, South Carolina

Bagley AM, Skinner HB. Progress in gait analysis in amputees: a special review. Cri Rev Phys and Rehabil Med 1991;3:101-120
Beasley WC. Quantitative muscle testing: principles and applications to research and clinical services. Arch Phys Med rehabil 1961: 42: 398-425
Berchuk M., Andriacchi TP, Bach BR, Reider B. Gait adaptations by patients who have a deficient anterior cruciate ligament. J Bone Joint Surg {Am} 1990;871-877
Brown FW. Construction of a knee joint in congenital total absence of the tibia (paraxial hemimelia tibia). J Bone Joint Surg {Am} 1965;47:695-704.
Brown FW, Pohnert WH. Construction of a knee joint in meromelia tibia (congenital absence of the tibia). J Bone Surg {Am} 1972; 54: 1333.
Christini D, Levy EJ, Facanha FAM, Kumar SJ. Fibular transfer for congenital absence of the tibia. J Pediatr Or-thop 1993; 13:378-381.
Czeriecki JM, Gitter AJ. Gait ananly-sis in the amputee: has it helped the amputee or contributed to the development of improved prosthetic components? Gait & Posture 1996;28:258-268.
Dennison WM. Delayed ossification of the tibia in apparent congenital absence. Br J Surg 1940; 28:101-105.
Engsberg JR, Lee AG, Tedford KG, Harder JA. Normative ground reaction force data for able-bodied and below-knee -amputee children during walking. J Pediatr Orthrop 1993: 13: 169-173 l0.Epps CH, Tooms RE, Edholm CD, Kruger LM, Bryant DD. Failure of centralization of the fibula for congenital longitudinal deficiency of the tibia. J Bone Joint Surg {Am} 1991; 73: 858-867
Finnegan MA, Uthoff HK. The development of the knee. In:Uthoff HK, ed. The embryology of the human locomotor system. Berlin: Springer-Verlag,
1990: 120-140.
Garnder E, O'Rahilly R. The early development kof the knee joint in staged human embryos. J Anat 1968; 102: 289-299.
Grisson LE, Harcke HT, Kumar SJ. Sonography in the management of tibial hemimelia. Clin Orthop 1990; 251:266-270.
Haines RW. The early development of the femorotibial and tibio-fibular joints. J Anat 1953; 87:192-206. 15. Hosea Tm, Tria AJ Jr, Bechler JR. Embryology of the knee. In: Scott WN, ed. The knee. St Louis: Mosby-Year Book, Inc. 1994: 3-13.
Jayakumar SS, Eilert RE. Fibular transfer for congenital absence of the tibia. Clin Orthop 1979; 139: 97-101. 17.
Jones D, Barnes J, Lloyd-Roberts GC. Congenital aplasia and dysplasia of the tibia with intact fibula. J Bone Joint Surg {Br} 1978;60:31-39. 18.Kalamchi A, Dawe RV. Congenital deficiency of the tibia. J Bone Joint Surg {Br} 1985;67:581-584.
Lewallen R, Dyck G, Quanbury A, Ross K, Letts M. Gait kinematics in below-knee child amputees: a force plate analysis. J Pediatr Orthrop 1986; 6: 291-298
Loder RT, Herring JA. Disarticulation of the knee in children. J Bone Joint Surg {Am} 1987; 69: 1155-1160.
Loder RT, Herring JH. Fibular transfer for congenital absence of the tibia: a reassessment. J Pediatr Orthop 1987;7:8-13.
Miller LS, Armstrong PF. The morbid anatomy of congenital deficiency of the tibia and its relevance to treatment. Foot & Ankle 1992; 13: 396-399.
Meyer LC, Sehayik RI, Davis H. The telescoping bifurcation synostosis in the treatment of incomplete longitudinal tibial deficiency. Inter-Clin Info Bull 1980; 17: 1-8. " 24.Myers TH. Congenital absence of the tibia: transplantation of the head of the fibula: arthrodesis at the ankle joint. Am J Orthop Surg 1905; 3: 72-85.
Myers TH. Further report on a case of congenital absence of the tibia. Am J Orthop Surg 1910; 8: 398-400.
O'Rahilly R, Gardner E. The timing and sequence of events in the development of the limbs in the human embryo. Anat Embryol 1975; 148: 1-23.
Pattinson RC, FIxsen JA. Management and outcome in tibial dysplasia. J Bone Joint Surg {Br} 1992; 75: 893-896.
Perry J, Giovan P, Harris LJ, Montgomery J, Azaria M. The determinants of muscle action in the hemiparetic lower extremity and their effects on the examination procedure. Clin Orthop 1978;131:71-89.
Perry J, Ireland ML, Gronley J, Hoffer MM. Predictive value of manual muscle testing and gait analysis in normal amkles by dynamic electromyography. Foot & Ankle 1986; 6: 254-259.
Perry J. Gait Analysis Normal and Pathologic Function. Thorofare: Slack Inc, 1992: 171-182.
Schoenecker PL, Capelli AM, Millar EA, Sheen MR, Haher T, Aiona MD, Meyer LC. Congenital longitudinal deficiency of the tibia. J Bone Joint Surg {Am} 1989; 71:278-287.
Simmons ED Jr, Ginsburg GM, Hall JE. Brown's procedure for congenital absence of the tibia revisited. J Pediatr Orthop 1996; 16: 85-89.
Sulamaa M, Ryoppy S. Congenital absence of the tibia. Acta Orthop Scan-dinav 1964; 34: 337-348.
Turker R, Mendelson S, Ackman J, Lubicky JP. Anatomic considerations of the foot and leg in tibial hemimelia. J Pediatr Orthop 1996; 16: 445-449.
Wehbe MA, Weinstein SL, Ponseti IV. Tibial agensis. J Pediatr Orthop 1981;1:395-399.
Williams L, W'eintroub S, Getty CJM, Pincott JR, Gordon I, Fixsen JA. Tibial dysplasia a study of the anatomy. J Bone Joint Surg {Br} 1983; 65:157-159.