Allografts are used to compensate for bone defects resulting from revision surgery, tumor surgery and reconstructive bone surgery. While it is well known that the reduction of fat content of allografts increases mechanical properties, the content of liquids were not assessed with known grain size distribution so far. The aim of the study was to compare the mechanical properties of dried allografts (DA) to allografts mixed with a saline solution (ASS) to allografts mixed with blood (AB) having a similar grain size distribution. Fresh-frozen morsellized bone chips were cleaned chemically, sieved and reassembled in specific portions with known grain size distribution. A uniaxial compression was used to assess the yield limit of the three groups before and after compaction with a fall hammer apparatus. No statistically significant difference could be found between all three groups for the yield limit (p=0.339) before compaction. After compaction no statistically significant difference could be found between DA and ASS (p=0.339) and between ASS and AB (p=0.554). AB showed a statistically significant higher yield limit than DA (p=0.022). At the yield limit 3 outliers were removed in DA, 1 in ASS and 1 in AB before compaction and 2 in DA and 1 in AB after compaction. Excluding the effect of the grain size distribution on the mechanical properties it was shown that allografts have a lower yield limit, when lipids are present. The liquid content of allografts seems to play an inferior role as no statistically significant difference could be found between DA and ASS. It is suggested in accordance with other studies to chemical clean allografts before implantation to reduce the contamination risk and the fat content. An optimum liquid level still remains to be defined. The considerations here described are relevant for filling up bigger bone defects, while in smaller defects the differences between different preparation methods may be less prominent

The management of severe acetabular bone defects poses a complex challenge in revision hip arthroplasty. Although biological fixation materials are currently dominant, cage has played an important role in complex acetabular revision in the past decades, especially when a biological prosthesis is not available. The purpose of this study is to report the long-term clinical and radiographic results of Paprosky type Ⅲ acetabular bone defects revised with cage and morselized allografts. We retrospectively analyzed 45 patients who underwent revision hip arthroplasty with cage and morselized allografts between January 2007 and January 2019. Forty-three patients were followed up. There were 19 Paprosky type IIIA bone defect patients and 24 Paprosky type IIIB bone defect patients and 7 patients of the 24 were also with pelvic discontinuity. Clinical assessment included Harris Hip Score (HHS) and Short Form-12 (SF-12). Radiographic assessment included cage stability, allografts incorporation, and center of rotation. All patients were followed up with a mean follow-up of 10.6 years, HHS and SF-12 improved significantly at last follow-up in comparison to the preoperative. There were 2 re-revisions, one at 5 years after surgery, another at 13.6 years after surgery. Two patients had nonprogressive radiolucency in zone III and the junction of zone II and zone III at the bone implant interface. Allografts of 40 (93%) cases incorporated fully. The combination of cage and morselized allograft is an alternative option for acetabular revision with Paprosky type III bone defects with satisfactory long-term follow-up results

Osteogenic augmentation is required in various orthopaedic conditions. Autograft is the gold standard but has limitations of increased morbidity and limited amount. Bone graft substitutes are costly and limited and don't integrate with host bone. Deep freezed allografts are a viable option, though not widely used in India and there are sparse reports in literature. This paper studies early efficacy of deep freezed bone allografts in treatment of fractures requiring bone graft. This is a prospective descriptive study. Strict inclusion and exclusion criteria as per standard guidelines were followed. We have a in-house facility of gamma irradiated deep freezed allografts available in hospital. 20 patients with comminuted fracture, delayed / malunion / nonunion, depressed intra articular fractures were operated during one year and followed up for at least 24 weeks. Sloof's Criteria was used for assessing osteointegration of grafts. Efficacy was authenticated by observing complications like serous discharge from surgical site, infection (superficial/deep), rejection of graft, clinical and radiological integration of graft, maintenance of articular reduction etc. Allografts have not only accepted well but fractures have healed and bone integration is at various stages. Only one patients got infected (5%). The overall success rate in terms of adequate osteointegration is 95 %. 19 out of 20 patients in our study group had either attained or at various stages of osteointegration and healing. We concluded that deep freezed bone allografts is a viable option in patients with fractures requiring bone grafts, thus give satisfactory surgical outcome, with no serious side effects

Introduction and Objectives: The success rate of bone allografts in the medium term when used in cancer surgery is 63% to 90% according to the different series. Our aim was to analyze the results and the complication rate seen in bone allografts as used in our center. Materials and Methods: We collected follow-up data from 35 patients who received 37 allografts. The variables analyzed include diagnosis, age, bone and side affected, type of allograft, complications, additional surgery, time of follow-up, and allograft and patient survival. Results: Mean age at surgery: 10.6 years. 48.64% were osteosarcomas, 48.64% were Ewing sarcomas, 2.7% were other diagnoses. 52.94% were strut grafts, 29.41% were osteoarticular, 2.94% were composite, 2.94% were arthrodesis and 11.76% were other types. Of these, 88.88% suffered some type of complication and 81. 48% required additional surgery. We achieved allograft survival in 85.29% of cases with a mean follow-up of 55.45 months. Most frequent complications were non-union (25%), postoperative metastasis (25%) length discrepancy (25%), followed by degenerative arthritis (24%) graft resorption and infection (16.6%). Discussion and Conclusions: Allografts are a reconstructive option with good results in the medium term. In spite of the high complication rate, additional surgeries applying rescue procedures have made it possible to obtain allograft survival similar to that seen in published series

Introduction and Aims: Autologous bone is the preferred method of providing structural support in spinal surgery. The disadvantages are donor site morbidity and limited bone available to reconstitute the anterior column. We evaluated fresh frozen femoral allografts following anterior column reconstruction for lumbar burst fractures with neurological deficit. Method: Twenty-seven patients with neurological deficit ( Frankel grade A(3), B(7), C(16) D (1) due to burst fractures of the dorsolumbar junction were treated with fresh frozen allografts following anterior spinal decompression. The average age was 28 years, and 19 patients sustained the injury following a road traffic accident. The mean pre-operative kyphosis measured 190. A corpectomy was performed in all patients and femoral allografts were positioned by interference fit and the spine stabilised with an anterior rod screw construct. The radiographs were reviewed at three-monthly intervals and the fusion graded by an independent radiologist. Results: The follow-up in 24 patients ranged from 29 to 72 months (mean 43 months) and three patients were excluded due to inadequate follow-up. Allograft incorporation was assessed by criteria of Bridwell et al grade 1 fused with remodelling with cross trabeculae into the adjacent vertebral bodies, grade 11 graft intact, not fully remodelled and incorporated, no lucenies, grade 111 graft intact, but a definite lucency at the top or bottom of the graft, grade 1V not fused with resorption and collapse of graft. The allografts were stable and evidence of graft incorporation and remodelling were observed between eight and 24 months. Grade 1 fusion was seen in 23 patients at two years and subsequent follow-up revealed no fracture, resorption or collapse. The average neurological recovery, which was 1.4 Frankel grades (range 0–2 grades), occurred within seven weeks following surgery (range 11–74 days). Nine patients (37%) made a complete recovery and in four patients (16%) there was no improvement. The mean post-operative kyphosis at two years was 80 (range 2–180). At seven-year follow-up one patient had an asymptomatic grade 11 fusion following secondary infection due to TB which was successfully treated. Conclusion: The indications for the operative treatment of thoracolumbar burst fractures remains controversial. The increased compressive strength of allografts, the large surface of contact, and the stability with instrumentation created a stable construct, which permitted early mobilisation. Allografts were inexpensive biological alternatives to reconstruct the anterior column in burst fractures

Introduction: Unicompartmental osteoarticular defects of the knee are challenging due to demands of stability and function of this weight-bearing joint. Prostheses reconstruction often requires sacrificing the uninvolved compartment. Osteoarticular allograft reconstruction can restore the anatomy, and allows reattaching soft tissue structures such as meniscus and ligaments from the host. The purpose of this study was to perform a survival analysis of unicompartimental osteoarticular allografts of the knee and evaluate their complications. Material and Methods: Forty unicompartmental osteo-articular allograft of the knee performed in 38 patients during the period 1962–2001, were followed for a mean of 11 years. In 36 patients, the bone defect was created by the resection of a tumor (33 giant cell tumors, 1 osteogenic sarcoma, 1 chondrosarcoma and 1 malignant fibrous histiocytoma) and in the remaining two by an open fracture. Twenty nine transplants were located at the femur that includes 11 medial and 18 lateral condyles. Eleven transplants were located at the tibia, including 4 medial and 7 lateral tibial plateaus. According to the reconstructed compartment, host meniscus and ligaments were reattached to the graft. Rigid internal fixation with plates and screws were used in each patient. Allografts survival from the date of implantation to the date of revision or the time of the latest follow-up was determined with the use of the Kaplan-Meier method. Complications as local recurrence, fracture, articular collapse and infection were analyzed. Results: The global rate of allograft survival was 85% at five years. There were 8 complications in 6 patients: 2 local recurrences, 2 infections, 1 fracture, 1 massive resorption and 2 articular collapses. In 6 patients the allograft was removed and they were considered as failures. All these patients required a second allograft that included 2 unicompartmental and 4 bicompartimental reconstructions. The two patients with articular collapses required a regular total knee resurfacing prosthesis. Discussion: Although the incidence of reoperations due to allograft complications may be high, the allograft survivor rate at five years was 85%. Unicondylar allografts, appear to be an alternative in those situations in which the massive osteoarticular bone loss to be reconstructed, is limited to one knee compartment

Pupose: The purpose of this study was to analyze the outcome of proximal tibia osteoarticular allografts after tumor resections. Material amd Methods: We performed a retrospective study over 58 patients in which a proximal tibia allograft reconstruction was undertaken. All patients were followed for a minimum of 5 years and allografts survival from the date of implantation to the date of revision or the time of the latest follow-up was determined with the use of the Kaplan-Meier method. In all patients, the patellar tendon from the host was reattached by suturing to overlapped donor flaps. Patients were clinically evaluated with the MSTS score system. Results: The global rate of allograft survival was 65% +/− 12% (+/− 2 SE) at five and ten years, with no significant difference between patients who received chemotherapy and those who did not. Allografts needed to be removed in twenty patients due to 12 infection, 4 local recurrences and 4 fractures. Active knee extension was restored in all patients with an average functional score of 26.3 points. Discussion: Survival analysis showed that 65% of proximal tibia osteoarticular allograft reconstructions remain stable at five and ten years. Patellar tendon reconstruction with allogeneic tissue in proximal tibia allograft restores active knee extension with an excellent functional result. Despite the incidence of complications, proximal tibia osteoarticular allografts continue to be a very valuable reconstructive procedure for large defects after resection of bone tumors

Despite the growing success of OCA transplantation in treating large articular cartilage lesions in multiple joints, revisions and failures still occur. While preimplantation subchondral drilling is intended to directly decrease allograft bioburden and has been associated with significant improvements in outcomes after OCA transplantation, the effects of size, number, and spacing of subchondral bone drill sites have not been fully evaluated. This study aimed to investigate the effects of drill size with or without pulse-lavage of OCA subchondral bone by quantifying remnant marrow elements using histomorphometry.

With IRB and ACUC approvals, human and canine OCAs were acquired for research purposes. Portions of human tibial plateau OCAs acquired from AATB-certified tissue banks that would otherwise be discarded were recovered and sectioned into lateral and medial hemiplateaus (n=2 each) with a thickness of 7 mm. Canine femoral condyles and tibial plateaus were split into lateral and medial components with a thickness of 7 mm (n=8). Using our clinical preimplantation preparation protocol, holes were drilled into the subchondral bone of each condyle and hemiplateau OCA using either 1.6 mm OD or 3.2 mm OD drill bits from the cut surface to the cortical subchondral bone plate. One femoral condyle and one hemiplateau per drill bit size were pulse-lavaged while the corresponding OCAs were not. The mean total %-fill remaining marrow elements for each treatment group was calculated.

Little to no quantifiable bone marrow element retention was noted to remain within the subchondral bone of human or canine OCA specimens after subchondral drilling of allograft bone with either drill bit size evaluated and with or without pulse-lavage. The %-fill was consistent across zones, ranging from 1-5%.

This project was designed to provide a preliminary histologic evaluation of the effects of drill size on OCA preimplantation preparation efficacy based on amount of remaining bone marrow elements in human and canine femoral condyle and tibial plateau specimens. Based on these initial findings, choice of drill bit size for OCA subchondral drilling may need to be based on the associated biomechanical effects rather than effects on donor bone marrow element removal.

Purpose: To describe our experience with vascularised fibulas used in sarcoma limb salvage surgery using standardized patient outcome measures. Methods: All vascularised fibulas and osteochondral allografts performed in the Capital District Health authority were assessed. A complete chart review and current functional assessment of the patients using the Toronto Extremity Salvage score (TESS) and the Musculosketal Tumour Society (MSTS) score were performed. Results: Nineteen patients with 19 tumors were recorded. The tumors range from 11 osteosarcomas, 4 Ewing’s sarcoma, 3 Malignant Fibrous Histiocytoma’s and 1 Chondrosarcoma. Average age was 23. The patient demographics are 75% male, 42% smokers, 86% femoral lesions and 13 % presented with pathological fracture. There were 9 hip fusions, 3 knee fusions, 6 intercalary grafts and one osteochondral graft. There was 21 % mortality with 21% lung mets, 20% local recurrence, 15.7% rates of amputation or infection or and non union. Allograft fracture rates of 10% were noted. Two patients underwent numerous operations (18) due to non-compliance. Rate of surgical failures defined as patients requiring re-operation after 2 years is 21%. Of 19 patients 10 are working, 4 are unable and 4 are deceased and 1 lost to follow up. Average follow up is 9.8 years (range of 4–18). Our functional results include TESS averaging 57.5 with a range of 30–105 and MSTS scores of average of 16.8 with a range of 3–28 and a percent score average of 55.8. The average score on the subjective assessment question was 4 equaling a response of accept it and would do it again. The Halifax outcome and functional data corresponds well with that in the literature. Conclusions: The biological repair of a combination of large Allografts with Vascularised Fibula’s is an excellent long term solution for construct survival with unrestricted patient activity

Purpose of the study: Loss of acetabular bone stock is a very common finding at revision total hip arthroplasty (rTHA). The acetabular bone defect can be filled with an autograft or with cyropreserved or lyophilized and radiated allografts. The permanent availability lypophylized radiated allografts is a certain advantage. For more than ten years (1994), we have used Phoenix® (TBF) lyophylized radiated bone grafts. Material and methods: We conducted a retrospective study of all patients who underwent rTHA for aseptic loosening between 1994 and 1999 with replacment of the acetabular implant requiring use of a lyophyilized radiated allograft (TBF, Phoenix®) fashioned from femoral heads and cut to fit. Grafts were impacted followed by acetabulra replacement with a cemented polyethylene (PE) cup or a Kerboull retaining ring, or an ace-tabular grid as needed. This procedure was used for 18 hips (16 patients). The Postel-Merle-d’Aubigné (PMA) clinical score and radiographic assessment were noted at five years with the Paprovsky classification. In addition, the status of the allograft (homogeneous aspect) and the presence of a lucent line between the host bone and the allograft were noted. Results: One patient was lost to follow-up. The analysis thus included 17 of 18 hips. Mean age was 63 years at rTHA surgery and 55 years at primary surgery. The reason for revision was cup loosening (n=13), isolated PE wear (n=4) with acetabular bone defects. The mean preoperative PMA score was 10.4 (range 5–18). At three months, the PMA score was 15.2 (range 12–18), at one year 16.2 (range 15–18), and at five years 17.2 (range 16–18). Implant migration was not observed on the five-year x-rays. Allografts were incorporated for seven hips which presented a homogeneous graft image. Five hips presented a partial lucent line and five a complete lucent line but with no evidence of implant instability. None of the patients required surgical revision to change the implant or for a new bone graft. Discussion: Acetabular revisions are often associated with bone defects which can be filled with allografts. This study demonstrated the good incorporation of lyophilized radiated allografts. This incorporation is progressive with good implant stability at more than five years. Use of this graft material for filling acetabular defects can correct for the bone deficiency. Conclusion: Use of lyophilized grafts gives satisfactory results with reliable outcome at five years comparable with other bone replacement methods, particularly cyropreserved femoral grafts used before 1994

Osteochondral allografts (frozen uncontrolled, or cryo-protected with dimethyl sulfoxide) were transplanted into medial femoral condyles of eighteen sheep. Cores from the ipsilateral graft site served as autografts for the contralateral limb. Analysis of graft and host cancellous bone microarchitecture by μCT at three months post transplant demonstrated no significant differences among the treatment groups. Dramatic bone resorption at the graft–host interface, however, occurred in up to 1/3 of condyles from all treatment groups, including fresh autografts suggesting that factors other than donor source or tissue storage played an important role in the bone incorporation of osteochondral grafts. The purpose of this study was to study the effect of different freezing protocols on periarticular cancellous bone architecture after osteochondral allograft transplantation. There were no significant differences in graft or host cancellous bone architecture among the groups (autografts, frozen allografts, cryopreserved allografts). Dramatic resorption of graft bone in condyles from all treatment groups suggested that factors other than donor source or tissue storage played important roles during incorporation of osteochondral grafts. Graft positioning, graft orientation, and recipient bed necrosis may play significant roles during incorporation of osteochondral graft bone. Osteochondral allografts (10 mm diameter) were transplanted into medial femoral condyles of eighteen skeletally mature Suffolk ewes. Allografts were frozen (–80°C) without cryoprotectant (FROZ) or treated with dimethyl sulfoxide (cryoprotectant) and frozen (–80°C at 1°C · min. −1. ) (CRYO). Osteochondral cores removed from ipsilateral graft sites served as fresh autografts (AUTO) for the contralateral medial femoral condyles. Condyles were harvested at three months and scanned (micro computed tomography –μCT). Three dimensional μCT data of graft and host cancellous bone regions were analyzed for bone volume fraction, trabecular thickness, bone surface–volume ratio, and trabecular anisotropy. No morphological differences were found among treatment groups. Excessive bone resorption of graft and interface precluded analysis of some samples from each group (ALLO — 2/9, CRYO — 3/9, AUTO — 6/18). Dramatic bone loss did not correlate with poor graft orientation, placement, infection, or recipient–bed necrosis, but a combination of these factors may contribute to excessive cancellous bone resorption in osteochondral grafts. Funding: Medical Research Council of Canada, Canadian Institutes of Health Research, No commercial funding. Please contact author for figures and/or diagrams

Introduction and Objective

Osteochondral allograft (OCA) transplants have been used clinically for more than 40 years as a surgical option for joint restoration, particularly for young and active patients. While immediate graft rejection responses have not been documented, it is believed that the host's immunological responses may directly impact OCA viability, incorporation, integrity, and survival, and therefore, it is of the utmost importance to further optimize OCA transplantation outcomes. The influences of sub-rejection immune responses on OCA transplantation failures have not been fully elucidated therefore aimed to further characterize cellular features of OCA failures using immunohistochemistry (IHC) in our continued hopes for the successful optimization of this valuable surgical procedure.

Purpose

The aim of this study was to compare the clinical outcomes of the revision TKA in which trabecular metal cones and femoral head allografts were used for large bone defect.

Purpose of the study: Reconstruction of the acetabulum is becoming increasingly necessary due to the longer life expectancy of patients with a total hip arthroplasty (THA); it can also be needed after trauma or resection of a bone tumour.

Material and methods: For 39 cases, we used a bone bank acetabulum with variable size for simple reconstruction of a part of the acetabulum (n=6) or to replace the entire acetabulum (n=19) or even an entire hemi-pelvis (n=14) in patients with more or less extensive bone loss. For 18 cases, resection was necessary to remove a bone tumour (16 chondrosarcomas and two giant-cell tumours), in 19 cases the reconstruction was necessary after multiple operations, and in four others due to traumatic destruction.

Results: Allograft integration was successful in nearly all patients (two cases of necrosis required a secondary prosthesis, 12.5%) and in two cases we noted immunological reactions with a serous effusion. There were no local infections. For osteocartilaginous acetabuli (n=7) the integration was quite satisfactory, certain patients have been followed for more than 26 years with no evidence of osteoarthritis.

Discussion: Use of a well-established protocol for hypothermia and the absence of secondary sterilisation yields grafts with preserved mechanical properties. The use of cyropreserved osteocartilaginous allografts offers hip for good integration free from degenerative osteoarthritis.

Purpose of the study: The objective of this work was to analyze outcome in a series of patients with bone anomalies after failed knee arthroplasty who were treated in the same institution with revision prosthesis with use of fragmented and structured bone grafts.

Material and methods: Between April 1994 and June 2004, a total of 515 knee arthroplasties were performed at the Italian hospital of Buenos Aires. Among these, 27 were revisions after failure of a primary arthroplasty. Two patients (follow-up less than one year) were excluded from the analysis. Among the 25 patients analyzed, eight were men and 17 women, mean age 67 years (range 41–87). Minimum follow-up was one year, mean 3.5 years, range 3–9 years. The cause of primary arthoplasty failure was aseptic loosening for ten knees, prosthetic infection in eleven, and pain in three and periprosthetic fracture in one. Fragmented or structured allografts were used for reconstruction in 15 patients. Reconstruction concerned the femur or the tibia after removal of the primary prosthesis. For twelve patients, a fragmented allograft was impacted and for three a combined fragmented-structured allograft was used. For 21 reconstructions, the revision prosthesis had intramedul-lary femoral and tibial stems and for 15, complementary elements were added to the prosthetic components on the femoral piece for three, on the tibial piece for four and on both for eight.

Conclusion: Use of fragmented and structured allografts combined with complementary elements offers a valid alternative for reconstruction of bone stock loss after failed primary knee arthroplasty.

Introduction. The management of periprosthetic pelvic bone loss is a challenging problem in hip revision surgery. This study evaluates the minimum 10-year clinical and radiographic outcome of major column structural allografts combined with the Burch-Schneider antiprotrusio cage for acetabular reconstruction. Methods. From January 1992 to August 2005, 106 hips with periprosthetic osteolysis underwent acetabular revision using massive allografts and the Burch-Schneider antiprotrusio cage. Forty-five patients (49 hips) died for unrelated causes without further surgery. Fifty-nine hips in 59 patients underwent clinical and radiographic evaluation at an average follow-up of 15.1 years. There were 17 male and 42 female patients, with age ranging from 29 to 83 years (mean 59). Results. Ten hips required rerevision because of infection (3), aseptic loosening (6), and flange breakage (1). Moreover, 4 cages showed x-ray signs of instability with severe bone resorption. The survivorship of the Burch-Schneider cage at 21.9 years with removal for any reason or radiographic migration and aseptic or radiographic failure as the end points were 76.3 and 81.4, respectively. The average Harris hip score improved from 33.2 points preoperatively to 75.7 points at the latest follow-up (p < 0.001). Discussion. In hip revision surgery, severe deficiency of pelvic bone stock is a critical concern because of the difficulty in providing a stable and durable fixation of the prosthesis. Although antiprotrusio cages have a limited role in acetabular revision, the use in association with massive allografts in extended bone loss demonstrated highly successful long-term results, enabling bone stock restoration and cup stability.

Purpose: To assess the use of cortical allografts (bone plates?) in hip replacement surgery.

Materials and methods: This is a retrospective study of 43 bone plates in 36 hip prostheses. In 18 cases they were implanted to treat a periprosthetic fracture (an associated replacement of the femoral component was performed in 5 cases) and in 18 they were implanted to replace a loosened stem in a hip with large bone defects. Standard long uncemented stems were implanted in 7 cases and standard cemented stems associated with morselized compacted allografts were implanted in 16 cases. 14 patients were only given bone plates and in 22 these bone plates were associated to a metal plate. The mean age was 69.1 years (range: 38–82). 61.1% were female, 18% were implanted in the right side and the mean follow-up was 45.4 months.

Results: At the time of the last review, three patients had died but for reasons not related to their hip surgery. Transient sciatic nerve palsy was observed in one patient, prosthetic dislocation in three cases (two of them were successfully treated with bracing and the other had to be given a constrained cup), there was an infection (treated with a two-stage replacement) and two re-fractures (after 3 and 13 months) treated with a new osteosynthesis with a bone plate associated to a metal plate. All the fractures healed and the imaging tests showed an integration of the bone plate with the host bone with no signs of prosthetic loosening.

Conclusions: Cortical allografts can fulfill two functions: a mechanical one (they behave as if they were a plate) and a biological one (they increase bone stock on integration).

Background

Processing of allografts, which are used to fill bone defects in orthopaedic surgery, includes chemical cleaning as well as gamma irradiation to reduce the risk of infection. Viable bone cells are destroyed and denaturing proteins present in the graft the osteoconductive and osteoinductive characteristics of allografts are altered. The aim of the study was to investigate the mechanical differences of chemical cleaned allografts by adding blood, clotted blood, platelet concentrate and platelet gel using a uniaxial compression test.

Introduction

The reconstruction of the knee in growing children considers many options and the chosen solution is often patient (or surgeon) based. Megaprostheses represent a reliable solution but quite expensive in the non-invasive growing version and not free from complications. In an Italian reference center for Bone and Soft tissue sarcomas, following the experience of Rizzoli Institute in Bologna, we performed the reconstruction with a resurfaced allograft for the distal femur or the proximal tibia in selected patients. The aim of the study is to confirm the reliability of this technique and to identify its potential advantages and indications.

Introduction: knee revision in absence of Extensor Mechanism has been always a challenging problem in Orthopaedics. Many authors are in favour to abandone any endoprosthetic substitution in front of such a situation.

We think osteotendinous allografts, in this particular case whole Extensor Mechanism allografts, could play an essential role before any Knee Arthrodesis.

Material and Method: From 1999 up to 2004 11 patients (4 male, 7 female) (mean age 72, range 68 to 86) underwent to a whole Extensor Mechanism allografting procedure. Mean follow up was 2.7 years (1 to 5 years).

In the first four cases a whole Extensor Mechanism allograft was implanted, while the next seven cases the allograft was reinforced by means of a Leeds-Keio Dacron band.

Results: There was no infections in this serie. The mean obtained R.O.M. in the first three months was – 5 of active extension (range 0 to −15) and 95 active flexion (range 80 – 110). However 3 from the 4 former operated cases had a progressive loss of active extension up to −25 (range −20 to −35) at 18 months, that did not increase after this period. Ultrasonic exams showed a lengthening of the patellar tendon in these cases. None of these 3 patients wished to undergo to a patellar tendon reinforcement.

On the other hand those later cases, where patellar tendon was reinforced did not show any change over the time (at 18 months mean active extension was maintained to −5 (range 0 to 15)

Conclusions: Extensor mechanism allografts are very useful in difficult knee revisions with absence of extensor mechanism, so that knee arthrodesis is not the method of choice for these patients. However augmentation of patellar tendon is necessary to maintain with the years an active extension.