G20 nonunions with defects

Editor's Notes

• #3: Nonunion with significant bone loss can occur in a variety of ways. Perhaps most commonly, it is the result of high energy open fracture, in which bone fragments are stripped out of the soft tissues and left at the accident scene. This results in an “instant nonunion” when the size of the defect exceeds the healing capacity of the individual. Not rarely, badly stripped and contaminated bone fragments must be debrided to prevent infection. Occasionally, bone is lost by high velocity or close range gunshot injury or other blast. Infection can lead to loss of bone. Badly infected and necrotic bone must be debrided. Resection of tumors may lead to loss of large bone segments, often including joints. Osteo necrosis from radiation, vasculitis or other rare etiologies can lead to fracture, infection, nonunion and the requirement for extensive bone debridement. A cadaveric study showed that metaphyseal defects around the knee entail the loss of 12 cc of bone volume per cm. lost, while diaphyseal defects average 7 cc/cm in the femur and 5 cc/cm in the tibia.
• #4: This classification scheme was proposed by Salai et al. In the Archives of Orthopedic and Trauma Surgery in 1999. It consists of 3 types - Type I defects are those which are minor loss of cortical bone or articular surface. Type II defects involve major loss of cortical bone, and type III involve major loss of articular surface. Subtypes A are open fractures, Subtypes B are closed. This system is not widely used or validated, and seems to have some gaps. It doesn’t make outcome or treatment predictions that are particularly useful. A better way to describe the defect is simply to specify location, volume or length, associated tissue loss, presence or absence of infection.
• #5: These problems are always complex and many factors have to be evaluated before treatment decisions can be made. Some of these are: 1. The status and integrity of the soft tissue envelope - open wounds,scars, previous incisions, draining sinuses. A healthy soft tissue envelope is important to almost all forms of bone reconstruction to some degree. 2. Presence or history of infection. This may require ESR, C-reactive protein, nuclear medicine studies, or biopsy if not obvious. 3. Contracture of adjacent joints greatly impacts on the ultimate function, and may get worse with certain forms of treatment. 4. Status of the nerves. An insensate, paralysed and/or painful extremity may be no value to the patient. 5. Status of the blood supply is particularly important if free tissue transfer is anticipated. 6. The location and size of the defect will be obtained from plain films usually. Deformity and bone quality for fixation should be considered. 7. Presence of hardware needing removal must be identified and planned. 8. Medical co-morbidities or other injuries may compromise the result. Chronic medical conditions including addictions to tobacco, alcohol or narcotics must be treated. 9. The social, economic and psychological resources of the patient must be assessed to determine what they can tolerate - reconstruction is a long and harrowing process.
• #6: The first issue to consider when confronted by any severe open fracture is whether the limb is salvageable or whether the patient would be better served with an amputation. This is a difficult decision sometimes and should be made with as much input from the patient and/or the family as possible; whenever you think you have to amputate a limb emergently, it is helpful to get a confirmatory consult from a partner, or another surgeon, perhaps the general surgery trauma guy or a Plastics consultant. A variety of considerations go into this decision. The first is the vascular status of the limb and the duration of ischemia. If the warm ischemia time has been greater than 6-8 hours, the chance of successful revascularization is poor. Likewise, complete loss of sensation in the tibial nerve distribution has been considered an indication for amputation. The extent and severity of other injuries may dictate amputation as the quickest way to stop ongoing blood loss and save the life of an unstable patient. Host health factors may influence the decision process. An elderly patient with vascular disease, diabetes or otherwise immunocompomised can be predicted to have a prolonged and comlicated course of healing after attempted limb salvage. One must balance the magnitude of the reconstructive effort with an estimation of the patient’s emotional and intellectial resources. The ultimate functional outcome of complex reconstructive procedures is often less than early amputation and this is a difficult judgement. We shouldn’t do things just because we can. Despite the proliferation of postulated predictive scoring systems like the MESS, no system has achieved universal acceptance as a way of helping to make this decision.
• #7: It is important to prioritize a sequence of goals and communicate them to the patient. The process goes through stages and will often take years to reach a final state. This list of goals represents both a priority schedule and a temporal sequence. Failure at any step along the way will probably preclude advancement to the next goal.
• #9: Temporary bridging external fixation can maintain length and alignment and add stability to protect soft tissues from further damage.
• #11: Closure or coverage with healthy soft tissue helps promote healing of the bone and prevention of infection. Rotation or free flaps are often required for the more severe tissue defects. Soft tissue envelope integrity should be re-established within a week if possible.
• #13: Multiple case reports exist of re-implantation of extruded diaphyseal bone segments that accompany the patient to the hospital after severe open fractures. The risk of infection is very high with this approach, however, if successful it can avoid the substantial difficulties and risks associated with reconstruction of these defects. Some criteria for when this approach might be acceptable: a young and healthy host, a well vascularized soft tissue bed like the thigh, a single cleanable segmental fragment rather than pieces. If this is attempted, there must be aggressive meticulous wound care and stabilization of the limb, usually with external fixation initially. The fragment must be carefully and thoroughly debrided of all contamination and dead soft tissue, cleaned and then sterilized. The techniques used have included autoclaving, boiling and soaking in antiseptics. The patient must be treated with systemic and possibly local (bead) antibiotics.
• #14: When the bone loss includes major portions of articular surface, the treatment options for reconstruction include osteochondral allograft, custom or standard joint replacement, or arthrodesis. These techniques are beyond the scope of this lecture and will not be discussed further.
• #15: The reconstructive options for loss of large amounts of bone from the diaphysis or metaphysis of long bones are listed on this slide. Autogenous bone grafting involves harvesting bone from the patient’s own body and moving it to another location. The bone can be cancellous, which usually comes from the iliac crest, although other sites include trochanter, distal femur, or proximal tibia. Cortical bone can be harvested from pelvis, tibia or fibula. When bone segments are transferred with attached blood vessels using microvascular techniques, these are called vascularized bone grafts. Autogenous bone graft heals quickest and best, but is limited in quantity and has donor site morbidity. Allograft bone graft comes from a cadeveric human sources. It also can be cancellous or cortical, or processed components of bone such as demineralized bone matrix. Distraction osteogenesis is a technique for generating new bone by applying tension to healing mesnchymal tissues using external fixation techniques. Pioneered by Ilizarov in Russia, this is a very useful and versatile method of handling bone loss. In certain situations, other approaches may be appropriate, such as primary shortening, or creation of a 1 bone forearm.
• #16: Bone grafts have 3 potential functions. The first is osteogenesis, the formation of new living bone. This can occur in two ways. Surface osteoblasts on the graft can survive the transplantation by receiving nutrition through diffusion, and then proliferate to form more living bone tissue. There is more surface area in cancellous graft, and thus it has more potential for surviving cells than cortical graft. The other method of osteogenesis is by osteoinduction, which is the process of recruitment and stimulation of osteoprogenitor mesnchymal cells from the host tissue. This process is stimulated by graft derived growth factors such as bone morphogenetic proteins, transforming growth factor beta, insulin-like growth factors 1 and 2, platelet derived growth factor and others. The second function of a bone graft is osteoconduction, or serving as ascaffold or conduit for the growth of host bone. This 3 dimensional process involves vascular proliferation and ingrowth of capillaries along the open spaces in the graft, followed by the differentiation of bone cells and production of bone tissue, which is subsequently remodeled. The third function of the graft is to provide structural support. In the diaphysis, this requires cortical graft. In filling metaphyseal voids to support articular surfaces ( e.g. tibial plateau) cancellous graft can be used.
• #17: Graft incorporation proceeds in 5 stages of host response, with the relative speed of each phase depending on the type of graft. Hemorrhage and inflammation are the initial 2 phases and many active cytokines are produced which help inititiate later stages. The third stage is vascular proliferation and ingrowth. The Invading capillaries bring perivascular tissue with mesenchymal cells that can differentiate into osteoprogenitor cell lines. The fourth stage consists of osteoclastic resorption of the avascular graft bone lamellae and simultaneous production of new bone matrix by osteoblasts. In the final stage, the newly formed bone is remodelled and reoriented based on the mechanical environment of the host site. This process is fastest in autogenous cancellous bone, and slowest in avascular cortical bone. In a canine model, the same phases occurred with allograft as with autograft, but the rate was about half as fast. In humans, the process seems to be slower than in dogs, and avascular cortical strut allografts may take years to incorporate by “creeping substitution” through the haversian system of the graft - in many cases, complete substitution will never occur.
• #18: Fresh cancellous autograft provides the quickest and most reliable type of bone graft. It’s open structure provides for rapid revasuclarization - a 5 mm graft may be totally revascularized in 20-25 days. The high surface area allows for potential survival of the largest number of graft cells. Handling of the graft is important to optimize this - the graft should be kept in chilled saline or blood and not allowed to dry out after harvesting. It does not provide structural support, except in the case of a contained metaphyseal defect such as the void left after elevation of a depressed tibial plateau fracture, in which it can support an articular surface if packed firmly. These grafts depend on ingrowth of host vessels and do best in well vascularized beads. There is the risk of donor site morbidity. Iliac donor site complications include pain, neurolovascular injury ( lateral femoral cutaneous, iliohypogastric, ilioinguinal, cluneal, superior gluteal), fracture including avulsion of the ASIS, infection, hematoma, herniation of abdominal contents, gait disturbance, violation of SI joint, and ureteral injury. Due to limited quantity available, it has been thought that this technique is restricted to short defects - some authors say under 6cm. However, one study of 8 large tibial defects averaging 10 cm were all reconstructed successfully using autogenous cancellous bone. ( Christian et al JBJS 71-A, 1989)
• #19: A technique you may hear about is call Papineau grafting. In this method open wounds with underlying bony defects are treated with wound care until a granulation begins, the the open wound is directly packed with cancellous bone graft and left open. Granulation grows through the graft, it is incorporated and skin grafting can be done on top. Although said to be very successful, this is rarely used today due to the availability of a wide variety of flap techniques to cover open wounds.
• #27: This case shows long term outcome (10 years) after cancellous grafting of a post- motorcycle accident distal tibial defect of about 8 cm. The patient ws full weight bearing and working as a truck driver.
• #28: Nonvascularized cortical strut grafting can be done, usually with autogenous fibula, although anteromedial tibia as the graft source has been described. This type of graft provides structural support of diaphyseal defects and is weekly osteogenic. It revascularizes very slowly, because vascular penetration of the graft proceeds through the existing volkman’s and Haversian canals, widened by osteoclastic resorption. The slow progress of this process means that the graft gets weaker initially as it increases in porosity, and this weakness may persist for months. In the classic article on this technique, Enneking et al described 40 patients who had diaphyseal defects after tumor resection and were reconstructed using autogenous fibula. Twelve required secondary cancellous grafting to achieve union. Eighteen had stress fractures in the graft after union, of which 15 healed and 3 became persistent nonunions despite treatment. The risk of stress fracture was related to length of the graft. Overall, there were 30 good or excellent results, 7 fair and 3 poor, including 1 infection. (Enneking, JBJS 62-A, 1980)
• #29: Frozen or freeze-dried allograft can be used to reconstruct large bone defects. These grafts incorporate by the same mechanisms and stages as autograft, but much slower - in canine models it takes about twice as long. In humans, large cortical allografts may never be completely replaced with living host bone. No bone cells survive the treatment process, and so the graft cannot form bone directly. It may be weakly osteoinductive in some forms. Cortico-cancellous chips may be used to support and fill contained metaphyseal defects. Cortical segments or strips can be used as structural elements. They will frequently require supplemental autogenous cancellous graft to heal to the host, and essentially always require support by internal fixation. The advantages of allograft are unlimited quantity, no size restriction, and they can include joint surfaces. The disadvantages are the risk of infection (about 5-12%), incomplete incorporation, healing problems and the risk of disease transmission. The risk of viral transmission from allograft is approximately 1 in 600,000. One long term study of intercalary allografts after tumor resection showed a success rate of 84%. Of the 15 that failed, half were ultimately salvaged with another graft or procedure. 31 failed to unite at one or the other end, requiring 81 additional procedures to achieve union. Once the first 3-4 years were passed, the grafts seemed to have good durability. Fixation with one long plate seemed superior to two plates with unsupported bone in between. Allograft bone struts can be combined with autograft to provide structural support along with osteoinduction.
• #30: This 35 year old male was involved in a high speed motor vehicle accident in which his family was killed. He suffered an open segmental femur fracture with significant diaphyseal and metaphyseal bone loss, but relative sparing of the joint. The limb was neurologically and vascularly intact. He was treated with irrigation and debridement of the grossly contaminated fracture, placement of an external fixator, and referral to the trauma center. On subsequent trips to the operating room for wound care, antibiotic beads were placed.
• #31: After the wound seemed clean and healthy, the external fixator was removed and the fracture was fixed with a long blade plate. The defect was grafted with a fibular allograft strut, cancellous autogenous bone, CaSO4 pellets and a bone stimulator was placed.
• #32: After 8 months, he had a solid union clinically, had formed a medial column of bone uniting proximal and distal segments, and was able to bear full weight and return to work as a delivery man.
• #33: Vascularized grafts consist of cortico-cancellous bone portions that are transferred with their vascular attachments. If the blood vessels are left intact to the native circulation and only the bone is moved, these are called “pedicled grafts” This is most commonly done with the fibula to reconstruct an ipsilateral tibial defect, and was first described by Huntington in 1905. If the blood vessels are divided and then reattached to a different source at a different location, it becomes a free vascularized graft. Most commonly, this is done with the fibula as well, transferred with the peroneal vessels, but a portion of iliac crest can be transferred with attached deep circumflex iliac artery, and the rib can be moved with posterior intercostal vessels. In some cases, skin and/or muscle can be transferred at the same time as part of a composite flap. Rib is rarely used due to high donor site morbidity. Iliac crest gives at most 5-6 cm of straight segment, and incorporates slower than fibula - up to half require cancellous graft. Iliac crest vascularized graft harvest has a relatively high morbidity rate, including hernia of abdominal contents. The free fibula is the real workhorse for reconstructing diaphyseal defects. You can get up to 20 cm of bone, and in some cases, a growing epiphysis can be transplanted as part of the free fibula. Donor site morbidity is minimal. If handled correctly, up to 90% of cells in the graft will survive. Vascularized bone grafts offer rapid incorporation, independence from host bed vascularity, and structural support. The will hypertrophy with time to handle increased loads.
• #34: The vascularized fibula graft was first described by Taylor in 1975 in a case where it was used to restore integrity in a tibial defect. Since then it has become the most commonly used free bone flap for diaphyseal defects due to it’s length and minimal donor site morbidity. In cases of large defects (>6 cm), many authors consider it the treament of choice, particularly in sites with poor host bed vascularity. Nutrition for the graft comes from the branches of the peroneal vessels that run along with it, and from periosteal vessels. Because the graft has it’s own blood supply, it is relatively independent of the vascularity of the host bed. It can be transferred with some soleus muscle and/or some skin (osteoseptocutaneous graft) to reconstruct both bone loss and soft tissue defects with a single graft. The skin paddle allows monitoring of the vascular supply to the graft. Jupiter et al reported on the use of this graft for reconstruction of segmental radius defects in 9 patients with successful healing in 8 of the 9. Heitmann reported on the use of this graft in 8 segmental humerus defects. Seven of the 8 had early fixation failure or fracture of the graft, but were fixed with ORIF and cancellous grafting. One infected case required a second free fibula. Donor site problems, exaggerated by some authors promoting other techniques, are generally mild and include moderate gait disturbance for the first 18 months (particularly difficulty with stairs), minor gait problems thereafter, slight decrease in calf strength and ankle eversion, FHL contracture and paresthesias of the peroneal nerve.
• #35: A healthy 29 year old right had dominant woman was injured in an accidental shooting which damaged her left proximal humerus shaft. The pulses and perfusion to the hand were intact, and motor function of all 3 nerves was retained, although there was some weakness of the ulnar innervated muscles.
• #36: She was initially treated with irrigation and debridement and application of an external fixator. Amputation was discussed, but she was adamantly opposed. The wound was treated with dressing changes for a prolonged period.
• #37: At 5 months post-injury, a free fibula graft was performed using a long lateral T plate for fixation.
• #38: The distal junction healed, but the proximal portion gradually developed into a nonunion and lost fixation into the humeral head. Vigorous rehabilitation of the arm was continued and pain was minimal, but the radiographic picture deteriorated.
• #39: AT 2 years post-injury, revision fixation was performed proximally, along with cancellous grafting, using a blade plate for better fixation into the head.
• #40: At 3 years post-injury, consolidation between the graft and the humeral head had occurred, and the arm was pain-free. Due to a malunion into varus, her abduction was limited to about 50-60 degrees, but she was able to get her hand to her face and the back of her head, and it was very helpful in activities of daily living.
• #41: This case shows a 10 year follow-up of a free fibula graft done for segmental bone loss in the femur following a motorcycle accident. The woman is fully weightbearing without any aid to ambulation and works as an over the road truck driver.
• #42: Distraction osteogenesis was discovered in the late 1940’s and early 1950’s in Russia by Ilizarov. This was an entirely new aspect of the biological behavior of bone. He discovered that new bone formation could be induced by very slowly pulling apart two well vascularized bone segments using external fixation. This process involves local neovascularization and increased biosynthetic activity. Osteoprogenitor cells are recruited and activated from local mesenchymal sources. It results in intramembranous bone formation in the gap between the segments.
• #43: The Ilizarov technique begins with application of an external fixator. The classic fixator used tensioned wires through the bone segments connected to rings around the limb, which are then connected to each other with threaded rods. Half pin monolateral fixators can be used in similar fashion for many applications of the technique. The bone is cut with a percutaneous corticotomy, usually in the metaphysis, preserving as much as possible of the intramedullary and periosteal blood supply. After corticotomy, there is a latency period of 5 -14 days before distraction is begun. Once the distraction process is started, the cut bone surfaces are slowly pulled apart at a rate of about .25 mm every 6 hours. Increments of distraction too large (e.g. 1 mm once a day) inhibit osteogenesis. If the rate of distraction is too slow the gap may close by normal fracture healing. Any instability producing shear stress inhibits osteogenesis. The tissue between the cut bone surfaces develops into a bipolar fibrovascular zone with collagen fibers oriented parallel to the direction of pull. Bone begins to form by intra membranous ossification arising from the full width of the cut bone surface. This bone forms in a highly uniform, ordered fashion of columns or cones about 200 microns in diameter, surrounded by microvasuclar channels. Mineralization proceeds in proximity to the vessels, which grow parallel to the distraction force.
• #44: There are 2 strategies for us of distraction osteogenesis in the face of bone deficits. The first involves acute shortening and compression at the fracture site after contouring the bone ends for stability, followed by corticotomy and lengthening at a separate metaphyseal location. This allows for reduction of the size of the soft tissue defect, and will often allow delayed primary closure or skin grafting of a defect that otherwise would require a soft tissue flap. A frame can be constructed to simultaneously compress at the fracture site and distract at a separate location. The other strategy involves putting on the frame with the limb at the correct length and alignment,and then using an internal lengthening of one or both segments to fill the gap. This is called bone transport, and the advantage is that the limb can be functional, even weightbearing, during the process.
• #45: Bone transport has a high rate of ultimate success with many series reporting upwards of 90% eventual healing with arrest of infection. There is no donor site morbidity associated with transport, as all the new bone comes from the injured leg. In addition, the leg can be functional and weightbearing during treatment. However, the treatment does require prolonged time in the external fixator, in some series up to 2 months per cm of gap filled. A large part of the frame time is due to delayed healing of the docking site, and this problem frequently requires bone grafting. Docking site healing problems occur in up to half the cases in some series. The prolonged time in the frame contributes to a high rate of complications, such as pin site infections, cellulitis, contracture, edema. Technique modifications to shorten the time in the frame include double segment transport, pre-grafting the docking site, or transport over an unreamed IM nail. Called the “monorail technique”, transport of a segment over a nail helps minimize malalignment, and once the docking occurs and is compressed, the nail can be interlocked and the frame removed. A similar approach uses transport under a MIPO bridge plate, which provides stability after thesegment is transported and allows earlier frame removal.
• #47: Initial treatment consisted of irrigation, debridement and placement of a simple half-pin unilateral fixator.
• #49: This image shows the regenerate consolidating after retrograde unifocal transport across a 14 cm gap
• #52: There are 3 studies which compare Ilizarov techniques for handling bone gaps with conventional techniques. All are published in the same volume of Clinical Orthopaedics and Related Research. They each use separate outcome measures and define things differently in terms of treatment success and complications, so it is very difficult to compare the studies. For example, conventional techniques for Green consist of Papineau grafting, while for Cierny they include cancellous grafting and free vasuclarized bone tranfer. Two of the studies are retrospective, while one reports on a “prospective protocol” compared with historical controls. None compare concurrently treated patients randomized to different groups. It is also important to realize that the authors are all Ilizarov advocates to some degree.
• #53: Nonetheless, one can combine the series and compare the treatments. In all 3 studies, there are only about a hundred patients total, pretty equally divided between conventional and Ilizarov treatment. The average defect was about the same and the ultimate success rate was about the same. The conventionally treated patients seemed to need more secondary procedures. One study (Cierny)showed that conventional treatment resulted in many more transfusions, longer hospitalizations and more total OR hours. Another study (Marsh) found an distinct advantage for Ilizarov treatment in terms of ultimate leg length discrepancy, but it included patients in the conventional group who had been treated with intentional shortening. Many of the Ilizarov patients needed bone grafting for problems at the docking site, but the authors noted that those bone grafts were much less extensive, required less volume, than the grafts in the conventional group, and thus presumably had a lower risk of morbidity.
• #54: A variety of other methods can be combined to treat difficult nonunions. Bone graft expanders can be added to autogenous bone graft. Most function in osteoconduction, with variable and rather unpredictable degrees of osteo-induction. Examples include ceramics such as calcium phosphate, hydroxyappetite, tricalcium phosphates or calcium sulfate. Bovine collagen composites with calcium phosphate (e.g. collagraft) and demineralized bone matrix products function the same way. Alone, they are not able to stimulate sufficient bone to fill major gaps, but they may have a role when mixed with autograft. The exact indications and efficacy are not clear. Similarly, bioactive substances that stimulate bone healing, such as bone morphogenic protein (BMP-7 or OP-1) and Platelet Rich Plasma preparations may have a role in helping to heal these difficult problems. Electrical or ultrasonic bone stimulation is unlikely to heal significant gaps, but may help promote bone healing in association with bone grafting procedures. A single case report in the Journal of Orthopaedic Trauma discusses the use of a cylindrical titanium mesh cage around an intramedullary nail to contain cancellous bone graft in a case of segmental diaphyseal loss.