Arch Orthop Trauma Surg (1990) 109 : 330-333
Achve°' Orthopaedic a.dTrauma Surgery © Springer-Verlag1990
Wave plate osteosynthesis as a salvage procedure G. Blatter and B. G. Weber Department of Orthopedic Surgery, Kantonsspital, CH-9007 St. Gallen, Switzerland
Summary. The load axis of the curved proximal femur lies not inside but outside the bone. Therefore, high bending forces are acting, the medial cortex absorbing pressure and the lateral cortex absorbs tension. In a transverse fracture, a laterally applied plate will absorb the tensile stresses and the medial cortex, the pressure forces. When the medial buttress due to a bony defect is missing, the laterally applied plate is subjected to cyclic bending and will undergo fatigue fracture. This dilemma is compensated by a wave plate with bone graft: the compression forces are redirected to the lateral cortex, and the plate is again subjected to tension. Furthermore, since the plate stands away from the bone, it does not disturb the blood supply at the fracture site and thus bone healing.
The treatment of fractures of the proximal third of the femoral diaphysis should take into account the particular biomechanical local load situation [9, 10]. Because the load axis of the leg passes outside the bone, the lateral cortex is subjected to high tension stress, while the medial cortex is under a strong compressive force [8, 12]. Osteosynthesis in this area disregarding the laws of biomechanics will always fail. According to the literature a complication rate of over 20% must be expected for the osteosynthesis of fractures of the proximal femoral shaft [5-7, 11, 13]. Conservative therapy for these fractures leads to an even higher percentage of disappointing results [1, 14]. Subtrochanteric femoral shaft fractures are rarely transverse fractures; oblique and spiral fractures are more frequent. Often a zone of comminution is found medially leaving osteosynthesis without a medial loadresistant buttress. Fatigue fractures of plates, screw pullouts or a proximal pull-out of intramedullary nails are characteristic unpleasant complications. Not only medial support but also a sound lateral cortex is essential to restore the tensile strength on the lateral side. The periosteal blood supply of the lateral part of the bone tube is impaired by open reduction and apOffprint requests to." Dr. G. Blatter
plication of a plate. Partial bone necrosis and osteoporosis under the plate may result in loss of tensile strength after plate removal, and refracture can occur [2-4].
Case report A 32-year-old patient sustained a comminuted fracture of the femur while skiing in February 1974. The fracture was stabilized with an 18-hole 130°-angled plate. Because of the lack of the medial buttress, the plate fractured. Reoperation with a 12-hole dynamic compression plate (DCP) was performed in October 1974 (Fig. la). A second plate fracture 1 year later was followed by an 18-hole DCP fixation (Fig. 1). Consolidation had failed laterally and medially after three platings, the surgeon performing this time intramedullary nailing (Feb-
Fig. 1. a First and b second reoperation after plate fractures because of lacking medial buttress
G. Blatter and B. G. Weber: Wave plate osteosynthesis
Fig. 2. a Intramedullary nailing after three failed platings, b The IM nail was being pushed out proximally because of shortening at the level of the fracture Fig. 3. a A second IM nailing with several bone graftings failed; the nail fractured at two levels (b). Re-nailing was performed, an 18ram rod was reinforced with a second smaller nail impacted inside it (c) lfig. 4. a The nails were able to withstand the loads but pushed out of the bone laterally, b The femur was then stabilized with an external fixator for almost 1 year. c The fracture was consolidated neither medially nor laterally Fig, 5. a A wave plate osteosynthesis with a corticocanceUous bone graft was performed, h The non-union healed
331
ruary 1976) (Fig. 2a). Shortening at the level of the fracture occurred, the IM nail being pushed out proximally (Fig. 2b). In October 1976 a second IM nailing with an 18-mm nail was performed, and rotation was stabilised at the same time with a plate. Eight days later the nail was interlocked and the plate in the fracture zone changed. Six months later the plate was removed and a cancellous bone graft applied. One year later, the IM nail was still in place, and the proximal fragment was tilted in varus position. The cancellous bone graft in the fracture zone had been resorbed (Fig. 3a). In November 1977 another cancellous bone graft was applied, and a corticocancellous bone graft was screwed in place laterally. Despite all this, 1 year later the nail fractured at two levels (Fig. 3b). At the beginning of November 1978 re-nailing was performed, and an 18-mm nail was placed into the medullary canal. This nail was reinforced with a second smaller nail impacted inside it (Fig. 3c).
332
G. Blatter and B. G. Weber: Wave plate osteosynthesis
Fig. 6. Schema of forces acting on the femur. F, compression force; LA, load axis; FA, femoral axis; a, distance between load axis and femoral axis; b, diameter of the femur; c, angle between load axis and femoral axis Fig. 7. If the medial buttress is missing, the plate breaks after a certain number of load cycles
6
7
The nails were able to withstand the loads, but the proximal fragment suffered further varus deformity so that the nails were pushed out of the bone laterally (Fig. 4a). Early in 1985 the patient had such a feeling of instability in his left thigh that a new operation by another surgeon was considered. Stabilization of the femur was attempted with a lateral tension band fixation using a external fixator (Fig. 4b), as it was thought that the blood supply of this badly damaged bone would be totally killed by the application of another IM nail or DC plate. So as not to damage the blood supply further the pseudoarthrosis was stabilised with an external fixator. Although the fixator was left in situ for almost 1 year, no real consolidation was achieved. Figure 4c shows that the fracture consolidated neither medially nor laterally, the fragments showing osteoporosis, however, a sign of improved blood perfusion. In this situation we proceeded to a lateral tension band plate in the form of a wave plate with a corticocancellous bone graft. This operation was performed in May 1986, a corticocancellous bone graft being jammed under the wave. Cancellous bone grafts were applied medially. The proximal femur was given a valgus position. For better rotational stability a small plate was applied anteriorly (Fig. 5a). Within 3 months the patient was allowed full weightbearing, and the non-union healed. Some 13 years after the accident this comminuted fracture of the femur has healed (Fig. 5b). Despite almost 20 operations, no infection had occurred. The leg is 3 cm shorter but can bear weight fully. Both hips and knees have a normal range of motion.
Discussion This example impressively shows the consequences of disregarding basic biomechanical and biological rules when performing internal fixation of proximal femoral shaft fractures. In the proximal femoral diaphysis, lateral tension banding and medial buttressing are biomechanically fundamental. U n d e r loading the f e m u r is subjected to compression, thrust and bending. C o n t r a r y to the bending forces, compressing and thrust forces are evenly distributed over every cross-section of the femur.
Fig. 8. By applying a wave plate with a corticocancellous bone graft, the load-resistant lateral cortex supports the compressive forces instead of the medial cortex, while the wave plate absorbs the tension forces
T h e bending force is at its m a x i m u m in the area of the lesser trochanter because here the distance b e t w e e n the load axis and the femoral axis is greatest. This distance descreases down to zero in the knee, where the load axis and femoral axis intersect each other (Fig. 6). T h e resulting tension distribution is given by the superimposition of compression and bending forces. T h e forces that n e e d neutralisation by internal fixation are greatest in the proximal femur. With a laterally applied plate the tension forces can only be neutralised when sufficient medial buttressing is present. If the medial buttress is missing, the plate will be subjected not to tension but mainly to bending forces, and it will break after a certain n u m b e r of load cycles (Fig. 7). By applying a wave plate with a corticocancellous b o n e graft, the load situation in the fracture area can be altered so that the plate is subjected again to pure tension and not bending forces. T h e load-resistant lateral cortex supports the compressive forces instead of the medial cortex, while the wave plate absorbes the tension forces (Fig. 8). B o n e is an extremely sensitive tissue. A fracture can only heal w h e n the blood supply is maintained. A f t e r plate r e m o v a l the lateral tension forces in the proximal f e m u r must be a b s o r b e d by the b o n e itself. These great forces can only be neutralised by a healthy, vital bone. This means that osteosynthesis must be p e r f o r m e d taking as m u c h care for the tissue as possible and that the implant must not impede b o n e healing. F o r mechanical reasons the plate should be applied laterally to the proximal f e m u r so that it can absorb the tension forces during fracture healing. T h e plate, however, inhibits the b l o o d supply to the b o n e in the critical area of the tension side. In contrast, the wave plate remains away f r o m the b o n e in the area of the fracture so that the blood supply to the b o n e is not h i n d e r e d [15]. Histological samples taken at plate r e m o v a l at the level of the wave-plate always showed vital bone.
G. Btatter and B. G. Weber: Wave plate osteosynthesis
333
References 1. Beam HP Jr, Seligson D (1980) Nine cases of bilateral femoral shaft fractures: a composite view. J Trauma 20 : 399-402 2. Breederveld RS, Patka P, van-Mourik JC (1985) Refractures of the femoral shaft. Neth J Surg 37 : 114-116 3. Convent L (1977) On secondary fractures after removal of internal fixation material. Acta Orthop Belg 43 : 89-93 4. Frankel VH, Bnrstein A H (1968) The biomechanics of refracture of bone. Clin Orthop 60 : 221-225 5. Katzner M, Petit R, Schvingt E (1975) Osteosynthese des fractures trochantero-diaphysaires et sous-trochanteriennes (~ propos de 57 osteosyntheses). J Chir (Paris) 109 : 53-62 6. Loomer RL, Meek R, De-Sommer F (1980) Plating of femoral shaft fractures: the Vancouver experience. J Trauma 20:10381042 7. Ltischer JN, Ruedi T, Allg6wer M (1978) Erfahrungen mit der Plattenosteosynthese bei 131 Femurschaft-Trtimmerfrakturen. Helv Chit Acta 45 : 39-42 8. Mtiller KH, Witzel U (1984) Die Briickenplatte zur Osteosynthese bei oss~ren Schaftdefekten des Femur als eine Konse-
9.
10. 11. 12. 13. 14. 15.
quenz nach Fehlschlfigen von Plattenosteosynthesen. Unfallheilkunde 87 : 237-246 Pauwels F (1965) Gesammelte Abhandlungen zur funktionellen Anatomie des Bewegungsapparates. Springer, Berlin Heidelberg New York Pauwels F (1973) Atlas zur Biomechanik der gesunden und kranken Hiifte. Springer, Berlin Heidelberg New York Roberts JB (1977) Management of fractures and fracture complications of femoral shaft using the ASIF compression plate. J Trauma 17 : 20-28 Teubner E, Fischer H (1980) Die operative Behandlung instabiler per- und subtrochanterer Oberschenkelfrakturen mit der 130 Grad T-Platte. Chirurg 51 : 685-692 Thompson F, O-Breine J, Gallagher J, Sheehan J, Quinlan W (1985) Fractures of the femoral shaft treated by plating. Injury 16 : 535-538 Wardlaw D (1977) The cast-brace treatment of femoral shaft fractures. J Bone Joint Surg [Br] 59:411-416 Weber BG, Brunner C (1982) Special techniques in internal fixation. Springer, Berlin Heidelberg New York
Achve°' Orthopaedic a.dTrauma Surgery © Springer-Verlag1990
Wave plate osteosynthesis as a salvage procedure G. Blatter and B. G. Weber Department of Orthopedic Surgery, Kantonsspital, CH-9007 St. Gallen, Switzerland
Summary. The load axis of the curved proximal femur lies not inside but outside the bone. Therefore, high bending forces are acting, the medial cortex absorbing pressure and the lateral cortex absorbs tension. In a transverse fracture, a laterally applied plate will absorb the tensile stresses and the medial cortex, the pressure forces. When the medial buttress due to a bony defect is missing, the laterally applied plate is subjected to cyclic bending and will undergo fatigue fracture. This dilemma is compensated by a wave plate with bone graft: the compression forces are redirected to the lateral cortex, and the plate is again subjected to tension. Furthermore, since the plate stands away from the bone, it does not disturb the blood supply at the fracture site and thus bone healing.
The treatment of fractures of the proximal third of the femoral diaphysis should take into account the particular biomechanical local load situation [9, 10]. Because the load axis of the leg passes outside the bone, the lateral cortex is subjected to high tension stress, while the medial cortex is under a strong compressive force [8, 12]. Osteosynthesis in this area disregarding the laws of biomechanics will always fail. According to the literature a complication rate of over 20% must be expected for the osteosynthesis of fractures of the proximal femoral shaft [5-7, 11, 13]. Conservative therapy for these fractures leads to an even higher percentage of disappointing results [1, 14]. Subtrochanteric femoral shaft fractures are rarely transverse fractures; oblique and spiral fractures are more frequent. Often a zone of comminution is found medially leaving osteosynthesis without a medial loadresistant buttress. Fatigue fractures of plates, screw pullouts or a proximal pull-out of intramedullary nails are characteristic unpleasant complications. Not only medial support but also a sound lateral cortex is essential to restore the tensile strength on the lateral side. The periosteal blood supply of the lateral part of the bone tube is impaired by open reduction and apOffprint requests to." Dr. G. Blatter
plication of a plate. Partial bone necrosis and osteoporosis under the plate may result in loss of tensile strength after plate removal, and refracture can occur [2-4].
Case report A 32-year-old patient sustained a comminuted fracture of the femur while skiing in February 1974. The fracture was stabilized with an 18-hole 130°-angled plate. Because of the lack of the medial buttress, the plate fractured. Reoperation with a 12-hole dynamic compression plate (DCP) was performed in October 1974 (Fig. la). A second plate fracture 1 year later was followed by an 18-hole DCP fixation (Fig. 1). Consolidation had failed laterally and medially after three platings, the surgeon performing this time intramedullary nailing (Feb-
Fig. 1. a First and b second reoperation after plate fractures because of lacking medial buttress
G. Blatter and B. G. Weber: Wave plate osteosynthesis
Fig. 2. a Intramedullary nailing after three failed platings, b The IM nail was being pushed out proximally because of shortening at the level of the fracture Fig. 3. a A second IM nailing with several bone graftings failed; the nail fractured at two levels (b). Re-nailing was performed, an 18ram rod was reinforced with a second smaller nail impacted inside it (c) lfig. 4. a The nails were able to withstand the loads but pushed out of the bone laterally, b The femur was then stabilized with an external fixator for almost 1 year. c The fracture was consolidated neither medially nor laterally Fig, 5. a A wave plate osteosynthesis with a corticocanceUous bone graft was performed, h The non-union healed
331
ruary 1976) (Fig. 2a). Shortening at the level of the fracture occurred, the IM nail being pushed out proximally (Fig. 2b). In October 1976 a second IM nailing with an 18-mm nail was performed, and rotation was stabilised at the same time with a plate. Eight days later the nail was interlocked and the plate in the fracture zone changed. Six months later the plate was removed and a cancellous bone graft applied. One year later, the IM nail was still in place, and the proximal fragment was tilted in varus position. The cancellous bone graft in the fracture zone had been resorbed (Fig. 3a). In November 1977 another cancellous bone graft was applied, and a corticocancellous bone graft was screwed in place laterally. Despite all this, 1 year later the nail fractured at two levels (Fig. 3b). At the beginning of November 1978 re-nailing was performed, and an 18-mm nail was placed into the medullary canal. This nail was reinforced with a second smaller nail impacted inside it (Fig. 3c).
332
G. Blatter and B. G. Weber: Wave plate osteosynthesis
Fig. 6. Schema of forces acting on the femur. F, compression force; LA, load axis; FA, femoral axis; a, distance between load axis and femoral axis; b, diameter of the femur; c, angle between load axis and femoral axis Fig. 7. If the medial buttress is missing, the plate breaks after a certain number of load cycles
6
7
The nails were able to withstand the loads, but the proximal fragment suffered further varus deformity so that the nails were pushed out of the bone laterally (Fig. 4a). Early in 1985 the patient had such a feeling of instability in his left thigh that a new operation by another surgeon was considered. Stabilization of the femur was attempted with a lateral tension band fixation using a external fixator (Fig. 4b), as it was thought that the blood supply of this badly damaged bone would be totally killed by the application of another IM nail or DC plate. So as not to damage the blood supply further the pseudoarthrosis was stabilised with an external fixator. Although the fixator was left in situ for almost 1 year, no real consolidation was achieved. Figure 4c shows that the fracture consolidated neither medially nor laterally, the fragments showing osteoporosis, however, a sign of improved blood perfusion. In this situation we proceeded to a lateral tension band plate in the form of a wave plate with a corticocancellous bone graft. This operation was performed in May 1986, a corticocancellous bone graft being jammed under the wave. Cancellous bone grafts were applied medially. The proximal femur was given a valgus position. For better rotational stability a small plate was applied anteriorly (Fig. 5a). Within 3 months the patient was allowed full weightbearing, and the non-union healed. Some 13 years after the accident this comminuted fracture of the femur has healed (Fig. 5b). Despite almost 20 operations, no infection had occurred. The leg is 3 cm shorter but can bear weight fully. Both hips and knees have a normal range of motion.
Discussion This example impressively shows the consequences of disregarding basic biomechanical and biological rules when performing internal fixation of proximal femoral shaft fractures. In the proximal femoral diaphysis, lateral tension banding and medial buttressing are biomechanically fundamental. U n d e r loading the f e m u r is subjected to compression, thrust and bending. C o n t r a r y to the bending forces, compressing and thrust forces are evenly distributed over every cross-section of the femur.
Fig. 8. By applying a wave plate with a corticocancellous bone graft, the load-resistant lateral cortex supports the compressive forces instead of the medial cortex, while the wave plate absorbs the tension forces
T h e bending force is at its m a x i m u m in the area of the lesser trochanter because here the distance b e t w e e n the load axis and the femoral axis is greatest. This distance descreases down to zero in the knee, where the load axis and femoral axis intersect each other (Fig. 6). T h e resulting tension distribution is given by the superimposition of compression and bending forces. T h e forces that n e e d neutralisation by internal fixation are greatest in the proximal femur. With a laterally applied plate the tension forces can only be neutralised when sufficient medial buttressing is present. If the medial buttress is missing, the plate will be subjected not to tension but mainly to bending forces, and it will break after a certain n u m b e r of load cycles (Fig. 7). By applying a wave plate with a corticocancellous b o n e graft, the load situation in the fracture area can be altered so that the plate is subjected again to pure tension and not bending forces. T h e load-resistant lateral cortex supports the compressive forces instead of the medial cortex, while the wave plate absorbes the tension forces (Fig. 8). B o n e is an extremely sensitive tissue. A fracture can only heal w h e n the blood supply is maintained. A f t e r plate r e m o v a l the lateral tension forces in the proximal f e m u r must be a b s o r b e d by the b o n e itself. These great forces can only be neutralised by a healthy, vital bone. This means that osteosynthesis must be p e r f o r m e d taking as m u c h care for the tissue as possible and that the implant must not impede b o n e healing. F o r mechanical reasons the plate should be applied laterally to the proximal f e m u r so that it can absorb the tension forces during fracture healing. T h e plate, however, inhibits the b l o o d supply to the b o n e in the critical area of the tension side. In contrast, the wave plate remains away f r o m the b o n e in the area of the fracture so that the blood supply to the b o n e is not h i n d e r e d [15]. Histological samples taken at plate r e m o v a l at the level of the wave-plate always showed vital bone.
G. Btatter and B. G. Weber: Wave plate osteosynthesis
333
References 1. Beam HP Jr, Seligson D (1980) Nine cases of bilateral femoral shaft fractures: a composite view. J Trauma 20 : 399-402 2. Breederveld RS, Patka P, van-Mourik JC (1985) Refractures of the femoral shaft. Neth J Surg 37 : 114-116 3. Convent L (1977) On secondary fractures after removal of internal fixation material. Acta Orthop Belg 43 : 89-93 4. Frankel VH, Bnrstein A H (1968) The biomechanics of refracture of bone. Clin Orthop 60 : 221-225 5. Katzner M, Petit R, Schvingt E (1975) Osteosynthese des fractures trochantero-diaphysaires et sous-trochanteriennes (~ propos de 57 osteosyntheses). J Chir (Paris) 109 : 53-62 6. Loomer RL, Meek R, De-Sommer F (1980) Plating of femoral shaft fractures: the Vancouver experience. J Trauma 20:10381042 7. Ltischer JN, Ruedi T, Allg6wer M (1978) Erfahrungen mit der Plattenosteosynthese bei 131 Femurschaft-Trtimmerfrakturen. Helv Chit Acta 45 : 39-42 8. Mtiller KH, Witzel U (1984) Die Briickenplatte zur Osteosynthese bei oss~ren Schaftdefekten des Femur als eine Konse-
9.
10. 11. 12. 13. 14. 15.
quenz nach Fehlschlfigen von Plattenosteosynthesen. Unfallheilkunde 87 : 237-246 Pauwels F (1965) Gesammelte Abhandlungen zur funktionellen Anatomie des Bewegungsapparates. Springer, Berlin Heidelberg New York Pauwels F (1973) Atlas zur Biomechanik der gesunden und kranken Hiifte. Springer, Berlin Heidelberg New York Roberts JB (1977) Management of fractures and fracture complications of femoral shaft using the ASIF compression plate. J Trauma 17 : 20-28 Teubner E, Fischer H (1980) Die operative Behandlung instabiler per- und subtrochanterer Oberschenkelfrakturen mit der 130 Grad T-Platte. Chirurg 51 : 685-692 Thompson F, O-Breine J, Gallagher J, Sheehan J, Quinlan W (1985) Fractures of the femoral shaft treated by plating. Injury 16 : 535-538 Wardlaw D (1977) The cast-brace treatment of femoral shaft fractures. J Bone Joint Surg [Br] 59:411-416 Weber BG, Brunner C (1982) Special techniques in internal fixation. Springer, Berlin Heidelberg New York
