Treatment of allograft nonunions with recombinant human bone morphogenetic proteins (rhBMP)

Fractures and nonunions are the main complications associated with bone allografts. Although the osteogenic role of recombinant human bone morphogenetic proteins (rhBMPs) has been demonstrated in experimental models and human tibial nonunions, the results are unknown for allograft nonunions. In this study, the efficacy of rhBMPs was evaluated in nonunions of femoral allografts. The results of six allograft nonunions in five patients who underwent resection of malignant bone tumours and allograft bone transplantation were analysed one to five years following application of rhBMPs at the nonunion site. There were two osteoarticular allografts and three intercalary allografts. Of three intercalary allografts, one demonstrated nonunion at both ends. Four patients received adjuvant chemotherapy and three had additional radiation therapy. There were two allograft fracture nonunions and four nonunions at the allograft-host junction. Two allograft fracture nonunions and one nonunion at the allograft-host junction were treated with 12 mg of rhBMP-2. The remaining three nonunions were treated with 7 mg of rhBMP-7 (Osigraft). The outcome and radiological evidence of healing were evaluated at a minimal follow-up of twelve months. There was neither healing of allograft fractures nor union of allograft-host junction. There was elongation or enlargement of the callus from the host. One patient continued to develop resorption of the allograft, which led to allograft fracture. Two patients who were treated with rhBMP-7 and corticocancellous allografts developed sterile drainage. There was no tumour recurrence with the use of rhBMPs after a mean follow-up of 39+/-25 months. rhBMP's alone were not sufficient to achieve healing in allograft nonunions and fractures following wide resection including periosteum and soft tissues.     

Vascularised free fibular flap in bone resection and reconstruction.

This paper compares allograft alone and in combination with vascularised free fibular flaps (FFF) to reconstruct long bone defects after tumour excision. We present 33 cases, 21 of these patients had reconstruction with an allograft alone as the initial procedure. Nine patients underwent reconstruction with FFF plus allograft plus iliac crest bone graft (ICG), two patients underwent reconstruction with a FFF and ICG and one patient underwent reconstruction with an allograft, a pedicled fibular flap and a FFF. The allograft was obtained from the Queensland Bone Bank and had been irradiated to 25 000Gy. In our experience (N=21) the complication rates with allograft alone were: delayed union 3, nonunion 7, fractured allograft 6, infection requiring resection of the allograft 3, other infections 2. The revision rate was 48% (10 cases of which five required a free fibular flap) and an average of 1.8 revision procedures were required. In the lower limb cases, the mean time to full weightbearing was 20 months and 40% were full weightbearing at 18 months. We felt that the high complication rate compared with other series may have been related to the irradiation of the graft. FFFs were used in 18 cases, 12 cases were primary reconstructions and six were revision reconstructions. The mean fibular length was 19.4 cm (range 10-29 cm). There were no flap losses and the FFF united at both ends of 11 of 12 primary reconstruction cases. One case had nonunion at one end, giving a union rate of 96% (23 of 24 junctions). When a FFF was used in combination with an allograft as a primary reconstruction, the allograft nonunion rate was 50% (five of 10 cases). The mean time to full weightbearing in the lower limb cases was 7.5 months and 100% were full weightbearing at 18 months. The FFF hastens time to full weightbearing but does not appear to affect the complication rates of allograft. The number of revision procedures required is reduced in the presence of a FFF and is the latter is a useful technique for the salvage of refractory cases.     

Revision total hip arthroplasty: The use of solid allograft

Intract solid allografts was used to augment severe bone loss in 8 acetabula and 15 femora in 22 patients who underwent revision total hip arthroplasty with noncemented implants. The average follow-up period was 4 years. The average time to radiographic union of whole acetabular allografts was 11 months. Superior migration of the allografts occurred in four patients. Varus tilt of the acetabular component within the allografts was noted in three patients. Eleven patients underwent entire proximal femoral allograft reconstruction, and four patients had femoral head allograft reconstruction. Nine patients with entire proximal femoral allografts achieved radiographic union at an average of 13 months and two failed by nonunion. Only one of the four patients with femoral head, allograft reconstruction achieved union. Femoral component subsidence was noted in seven whole proximal femoral allografts (64%). Entire proximal femoral allograft reconstruction was complicated by at least one episode of postoperative dislocation in 6 of 11 procedures. The authors recommend that femoral head allografts should be used with caution to reconstruct proximal femoral deficiencies in which structural, support is required for stability of the implant. Successful use of acetabular allografts requires that the majority of the allograft be contiguous to host bone and not to soft tissue. With entire proximal femoral reconstruction, the tip of the femoral, component should not reside at the host-graft junction. All components should be cemented into allograft bone; and revision surgery should be performed before osteolytic destruction of bone advances to the point where allograft reconstruction is required.     

The use of periosteally vascularized autografts to augment the fixation of large segmental allografts.

Large segmental bone grafts are a standard of reconstructing long bone defects. Nonunion or delayed union at the host-graft junction is a major complication of these procedures. In six patients, a nonunion was treated by locally available bone, vascularized by its periosteum as an onlay autograft to improve fixation and speed incorporation of the allograft into the host bone. At three months, all were pain free and using their allografts without limitation. By four months, all had roentgenographic evidence of allograft and autograft incorporation. During the follow-up period from two to four years, all have incorporated the allografts and are pain free.     

Reconstruction after intercalary resection of malignant bone tumours: comparison between segmental allograft and extracorporeally-irradiated autograft

We reviewed 29 patients who had undergone intercalary resection for malignant tumours. Of these, 14 had received segmental allograft reconstruction and 15 extracorporeally-irradiated autograft. At a mean follow-up of 71 months (24 to 132), 20 were free from disease, five had died and four were alive with pulmonary metastases. Two patients, one with an allograft and one with an irradiated autograft, had a local recurrence. Reconstruction with extracorporeally-irradiated autograft has a significantly lower rate of nonunion (7% vs 43%, p = 0.031) but an insignificantly higher rate of fracture (20% vs 14%, p = 0.535) than that with segmental allograft. Using the Enneking functional evaluation system, the mean postoperative score for the patients without local recurrence was 87% (80% to 96%) and was similar in both groups. Extracorporeally-irradiated autograft could be an acceptable alternative for reconstruction after intercalary resection, especially in countries where it is difficult to obtain allografts.     

The use of structural proximal femoral allografts in complex revision hip arthroplasty

Between April 1992 and November 1998 we used 34 massive proximal femoral allografts for femoral reconstruction at revision hip arthroplasty. Seven patients have died and two have been lost to follow-up. There were thus 25 grafts in 24 patients for review. The mean follow-up was 53 months (16 to 101). By the time of the review two patients had undergone a further revision for failure of the allograft. Another had required secondary plating and grafting at the graft-host junction for symptomatic nonunion. One had recurrence of deep sepsis and was being managed conservatively. Trochanteric union was considered to have occurred radiologically in 16 of the 25 grafts and union at the host-graft junction in 20. Resorption of the allograft was significant in only two hips. We recommend this technique in cases in which femoral bone loss has been catastrophic.     

Freeze-dried segmental fibular allografts in azathioprine-treated dogs

The successful use of a bone allograft may be negated by the host's immune response. This investigation assessed the efficacy of combining freeze-dried cortical allografts in three to six weeks azathioprine-immunosuppressed dogs. Forty-eight of 94 adult mongrel dogs were initiated for this study, and 46 of 94 were previously published and recompiled. The dogs were divided into five groups and followed for six months: Group I consisted of bilateral fresh autografts as an external control; Group II assessed the effect of freeze-drying on autogenous bone; Group III compared fresh autografts with fresh allografts; Group IV assessed the effect of freeze-drying on allografts; and Group V assessed the combined effect of placing freeze-dried allografts in immunosuppressed hosts. Biweekly roentgenograms were made to evaluate the time to union and the incidence of graft fatigue failure. Mechanical graft strength was assessed by rapid torsional loading to failure at the time of sacrifice. Biologic repair was assessed with the use of tetracycline and microradiographic techniques. The incorporation and repair of a fresh cortical autograft is better than that of a freeze-dried autograft because of fractures, nonunion, or delayed union of graft-host junctions; freeze-dried autografts have increased peripheral and internal resorption, yet an increased peripheral bony callus maintains normal graft strength; freeze-dried and fresh allografts are similar in roentgenographic characteristics, mechanical strength, and in the mechanism of graft incorporation; the use of three or six weeks azathioprine therapy did not improve the fate of freeze-dried allografts.     

大段异体骨移植后骨端不愈合的外科治疗

目的 探讨应用自体骨植骨再固定的方法治疗大段同种异体骨移植后骨端不愈合的手术方法和疗效.方法 1994年1月至2006年12月北京积水潭医院所行大段同种异体骨移植患者176例中41例不愈合,22例进行了骨接触端的重新植骨治疗,其中15例临床资料完整.再植骨治疗时年龄15～34岁,平均为24岁.初始疾病为:骨肉瘤5例、骨巨细胞瘤4例、皮质旁骨肉瘤2例、血管内皮瘤2例、骨原始神经外胚瘤(PNET)2例.发病部位:股骨下端7例、肱骨中段3例、股骨中段2例、胫骨上端2例、肱骨上端1例.15例患者中,8例单纯不愈合行植骨术,另7例不愈合并发原内固定断裂行植骨和再固定手术.结果 随访时间18～148个月,平均47个月.15例中13例骨愈合,占86.7%.愈合时间5～20个月,平均13个月.其中8例无原内固定失败者均重新愈合,愈合时间平均为14个月.另7例原内固定失败者5例重新愈合,愈合时间平均为12个月,与上述8例比较愈合时间相似.无感染等并发症发生.2例仍不愈合,最终行人工假体置换术.13例骨愈合的患者MSTS评分平均25.1分,8例无原内固定失败者25.4分,5例原内固定失败者24.6分,评分基本相同.结论 应用自体骨植骨再固定的方法治疗大段异体骨移植后骨端不愈合手术简单,并发症少,愈合率高,再手术后功能影响小,较人工假体置换相比有明显优势,对于大段异体骨移植后骨端不愈合患者,应为首选治疗方法.     

Support of the anterior column with allografts in tuberculosis of the spine

Fresh-frozen allografts from the humerus were used to help to stabilise the spine after anterior decompression for tuberculosis in 47 children with a mean age of 4.2 years (2 to 9). The average angle of the gibbus, before operation, was 53 degrees; at follow-up, two years later, it was 15 degrees. Rejection of the graft or deep sepsis was not seen. Cross trabeculation between the allograft and the vertebral body was observed at six months, with remodelling occurring at approximately 30 months.     

Allograft reconstruction for bone sarcoma of the tibia in the growing child

The outcome of tibial allograft reconstruction after resection of a tumour is inconsistent and has a high rate of failure. There are few reports on the use of tibial allografts in children with open growth plates. We performed 21 allograft reconstructions (16 osteoarticular, five intercalary) in 19 consecutive patients between seven and 17 years of age. Two had Ewing's sarcoma, one an adamantinoma and 16 osteosarcoma, one with multifocal disease.Five patients have died; the other 14 were free from disease at the time of follow-up. Six surviving patients (eight allograft reconstructions) continue to have good or excellent function at a mean of 59 months (14 to 132). One patient has poor function at 31 months. The other seven patients have a good or excellent function after additional procedures including exchange of the allograft and resurfacing or revision to an endoprosthesis at a mean of 101 months (43 to 198). The additional operations were performed at a mean of 47 months (20 to 84) after the first reconstruction.With the use of allograft reconstruction in growing children, joints and growth plates may be preserved, at least partially. Although our results remain inconsistent, tibial allograft reconstruction in selected patients may restore complete and durable function of the limb.     

Surgical treatment of nonunion and avascular necrosis of the proximal part of the scaphoid in adolescents

BACKGROUND: The purpose of this retrospective study was to examine the clinical and radiographic results of treatment of proximal scaphoid nonunion and avascular necrosis with vascularized bone graft from the distal part of the radius in adolescent patients. METHODS: Between 1993 and 1996, three adolescents with a proximal scaphoid nonunion and avascular necrosis underwent vascularized bone-grafting and internal fixation. The mean age at the time of the fracture was 14.8 years (14.4, 14.6, and 15.3 years), and the mean time interval between the fracture and the surgery was 19.3 months (six, seventeen, and thirty-five months). We retrospectively reviewed all available clinical and radiographic data from the time of fracture to the time of the last follow-up. All patients were examined clinically and radiographically at the time of the review, at a mean of 5.5 years (five, five, and 6.5 years) after surgery. RESULTS: All fractures healed at a mean of 3.4 months (2.75, 3.0, and 4.5 months) postoperatively. Final follow-up radiographs showed union and revascularization of the proximal part of the scaphoid with no evidence of degeneration of the radiocarpal joint. None of the patients had limiting pain or scapholunate instability demonstrated on physical or radiographic examination. Dorsiflexion and radial deviation of the affected wrist were decreased by a mean of 22 degrees (10 degrees, 22 degrees, and 35 degrees ) and 15 degrees (5 degrees, 20 degrees, and 20 degrees ), respectively, compared with those of the normal wrist. CONCLUSION: Grafting with vascularized radial bone is an effective treatment, leading to union and good function, for nonunion and avascular necrosis of the proximal part of the scaphoid in adolescents.     

Circumferential allograft replacement of the proximal femur. A critical analysis

The use of proximal femoral structural allografts in revision hip arthroplasty remains controversial. The current study constitutes the mean 8.8 years followup (range, 3-12.5 years) of a consecutive series of 55 proximal femoral allografts in 51 patients. In 46 patients the implant was cemented into the allograft and the distal femur, and the host proximal femur was resected at the time of reconstruction in all but seven patients. Five patients underwent revision surgery for acetabular failure, and six additional patients underwent revision surgery for failure of the proximal femoral allograft. Three patients underwent successful revision surgery and had additional proximal femoral allografts. Failure was caused by graft fracture in one patient, by deep infection in two patients, and by junctional nonunion in three patients. Junctional nonunion was seen in five patients (9%), two of whom were treated successfully with bone grafting and bone grafting and plating, respectively. Instability was observed in six patients (11%). Trochanteric nonunion was seen in 22 patients (43%) and trochanteric escape was seen in 14 patients (27%). The mean Harris hip score improved from 39 to 79 points. Resorption involving the full thickness of the allograft in at least one zone was seen in seven patients. This progressed rapidly and silently within the first 3 years but has yet to lead to the failure of any of the reconstructions. Infection was ruled out in every case. Allograft resorption was seen in seven patients and may be related to a combination of factors. It is most likely that this is an immunologic problem of slow rejection, but it is possible that the distal cement fixation led to stress shielding and resorption attributable to mechanical disuse. The possible protective role of retaining the bivalved host bone as a vascularized onlay autograft remains to be clarified. Although these results justify the continued use of structural allografts for selected patients, continued followup is warranted.     

Humeral shaft aseptic nonunion: treatment with opposite cortical allograft struts

Plate fixation with cortical allograft struts has been used at our Institute for decades to treat aseptic shaft nonunion. The aim of this study was to assess the results of this technique in humeral nonunion. We retrospectively reviewed 57 consecutive patients with humeral diaphyseal nonunion treated by internal fixation combined with cortical allograft struts in the last 7 years in our Department. The patients were followed-up for a mean of 48 months. We had union in 53 cases out of 57. There were 3 cases of infection out of 15 patients previously treated with an external fixator. In our experience the cortical allograft strut is a well standardised and reproducible technique that enables the treatment of severe atrophic non-union with a relatively low complication rate and quick functional recovery.     

Reinsertion of the distal head of the biceps using mini-anchors and the anterior access: a retrospective study

Plate fixation with cortical allograft struts has been used at our Institute for decades to treat aseptic shaft nonunion. The aim of this study was to assess the results of this technique in humeral nonunion. We retrospectively reviewed 57 consecutive patients with humeral diaphyseal nonunion treated by internal fixation combined with cortical allograft struts in the last 7 years in our Department. The patients were followed-up for a mean of 48 months. We had union in 53 cases out of 57. There were 3 cases of infection out of 15 patients previously treated with an external fixator. In our experience the cortical allograft strut is a well standardised and reproducible technique that enables the treatment of severe atrophic non-union with a relatively low complication rate and quick functional recovery.     

Transfracture abduction osteotomy: A solution for nonunion of femoral neck fractures

Background:

Nonunion and avascular necrosis (AVN) of the femoral head remains one of the major complications following femoral neck fractures. Despite various surgical techniques and internal fixation devices, the incidence of nonunion and AVN has remained unsolved. Neglected nonunion of femoral neck fracture is common in the developing world. Treatment options include rigid internal fixation with or without bone grafting, muscle pedicle bone graft, valgus osteotomy of the proximal femur with or without bone graft, valgus osteotomy or hip arthroplasty. We conducted a retrospective analysis of cases of nonunion of femoral neck fracture treated by transfracture abduction osteotomy (TFAO).

Materials and Methods:

Over a period of 35 years (1974-2008), 30 patients with nonunion of femoral neck fractures were treated with TFAO over a period of 35 years (1974-2008), All patients were less than 50 years of age. Absence of clinical and radiological signs of union after four months was considered as nonunion. Patients more than 50 years of age were excluded from the study. Union was assessed at 6 months radiologically. Limb length was measured at six months. The mean duration of femoral neck fracture was 19 months (range 4 months 10 years). Results were analyzed in terms of radiological union at six months. Average followup was five years and six months.

Results:

Consistent union was noted at the followup after six months in 29 cases. One case was lost to followup after five and one-half months postoperatively. However, the fracture had united in this case at the last followup. Average shortening of the limb at six months was 1.9 cm. Average neck shaft angle was 127° (range 120-145°). Five cases went into AVN but were asymptomatic. Two cases required reoperation due to back out of Moore''s pins. These were reopened and cancellous screws were inserted in the same tracks.

Conclusions:

Consistent union of nonunion femoral neck fracture was noted at the followup after six months in 29 cases. The major drawback of the procedure is immobilization of the patient in the hip spica for eight weeks.     

The treatment of nonunions following intramedullary nailing of femoral shaft fractures

OBJECTIVE: To assess the effectiveness of different surgical options in the treatment of nonunion of a femoral shaft fracture after initial intramedullary nailing. DESIGN: Retrospective. SETTING: University hospital. PATIENTS AND METHODS: During a seven-year period a total of 278 skeletally mature patients with 280 fresh femoral shaft fractures were treated by intramedullary nailing. Of these patients, a subgroup of consecutive patients with nonunion of the fracture were subjected to a detailed analysis and were followed until the fracture was united (mean thirty-three months). Injury mechanism, fracture pattern using various established classifications, any possible concomitant injuries, complications, and subsequent surgical interventions were recorded. RESULTS: Of the total of 280 fractures, nonunion was observed in thirty-four patients with thirty-five fractures (12.5 percent). To achieve solid union, one reoperation was sufficient in twenty-five fractures, six fractures had to be operated on twice, and four needed three operations. There were five patients with autogenous bone grafting alone, and all five required a further reoperation for the nonunion. After a dynamization procedure, four of seventeen patients required a further reoperation. After eight exchange nailing procedures, further surgery for nonunion was necessary in only one case. Solid union was achieved within six months after the final successful reoperation. A marked shortening of the femur developed as a local complication in six cases, four of which had undergone dynamization as final treatment before solid union. CONCLUSIONS: Exchange nailing without extracortical bone grafting seems to be the most effective method to treat a disturbed union of a femoral shaft fracture after intramedullary nailing. Autogenous extracortical bone grafting alone proved to be insufficient. Dynamization predisposed to shortening of the bone.     

锁定钢板内固定治疗四肢骨不连的临床研究

目的 评价锁定钢板内固定治疗四肢长骨骨不连的临床疗效.方法 对2003年2月至2006年10月应用锁定钢板内固定治疗61例骨不连患者的临床资料进行回顾性研究.其中男性44例,女性17例;年龄7～70岁,平均38岁.其中5例肱骨骨不连,33例股骨骨不连,23例胫骨骨不连.骨不连的原因包括内固定失效47例,外固定失败5例,感染9例.骨不连病程为10～156个月,平均19个月.42例使用LCP钢板内固定,19例使用LISS钢板内固定.55例采用自体髂骨植骨,3例在自体植骨的同时结合同种异体松质骨移植,3例采用同种异体松质骨结合人工骨移植.手术前、后根据膝关节协会评分系统(KSS)评分对47例膝关节周围骨不连患者的关节功能进行评估,85～100分为优,70～84分为良,60～69分为一般,＜60分为差.术前优29例,良8例,一般4例,差6例.结果 所有患者均获得随访,随访时间6～24个月,平均12个月;骨折均在4～6个月内牢固连接,平均愈合时间4.8个月,无内植物松动、断裂等并发症.术后膝关节KSS评分,优35例,良7例,一般1例,差4例.结论 锁定钢板内固定结合植骨术是治疗四肢长骨骨不连的有效方法.     

Structural proximal femoral allografts for failed total hip replacements: a minimum review of five years

There are few medium- and long-term data on the outcome of the use of proximal femoral structural allografts in revision hip arthroplasty. This is a study of a consecutive series of 40 proximal femoral allografts performed for failed total hip replacements using the same technique with a minimum follow-up of five years (mean 8.8 years; range 5 to 11.5 years). In all cases the stem was cemented into both the allograft and the host femur. The proximal femur of the host was resected in 37 cases. There were four early revisions (10%), two for infection, one for nonunion of the allograft-host junction, and one for allograft resorption noted at the time of revision of a failed acetabular reconstruction. Junctional nonunion was seen in three patients (8%), two of whom were managed successfully by bone grafting, and bone grafting and plating respectively. Instability was observed in four (10%). Trochanteric nonunion was seen in 18 patients (46%) and trochanteric escape in ten of these (27%). The mean Harris hip score improved from 39 to 79. Severe resorption involving the full thickness of the allograft was seen in seven patients (17.5%). This progressed rapidly and silently, but has yet to cause failure of any of the reconstructions. Profound resorption of the allograft may be related to a combination of factors, including a slow form of immune rejection, stress shielding and resorption due to mechanical disuse with solid cemented distal fixation, and the absence of any masking or protective effect which may be provided by the retention of the bivalved host bone as a vascularised onlay autograft. Although continued surveillance is warranted, the very good medium-term clinical results justify the continued use of structural allografts for failed total hip replacements with severe loss of proximal femoral bone.     

Unicondylar osteoarticular allografts of the knee

BACKGROUND: In the management of a resected distal femoral or proximal tibial condyle as the result of tumor or trauma, a unicondylar osteoarticular allograft is currently the only reconstructive option that avoids the sacrifice of the unaffected condyle. The purposes of this study were to perform a survival analysis of unicondylar osteoarticular allografts of the knee and to evaluate the complications. METHODS: We retrospectively reviewed the results of forty large unicondylar osteoarticular allograft procedures in thirty-eight patients who were followed for a mean of eleven years. Twenty-nine allografts were femoral transplants and included eleven medial and eighteen lateral femoral condyles. Eleven allografts were tibial transplants, including four medial and seven lateral tibial condyles. The procedure was performed after a tumor resection in thirty-six patients and to replace condylar loss after a severe open fracture in the remaining two patients. Complications were analyzed, and allograft survival from the date of implantation to the date of revision or the time of the latest follow-up was determined. Functional and radiographic results were documented according to the Musculoskeletal Tumor Society scoring system at the time of the latest follow-up. RESULTS: One patient died of tumor-related causes without allograft failure before the two-year follow-up evaluation. The global rate of allograft survival at both five and ten years was 85%, with a mean follow-up of 148 months. In six patients, the allografts were removed at an average of twenty-six months (range, six to forty-eight months) and these were considered failures. All six patients underwent a second allograft procedure including two new unicondylar and four bicondylar reconstructions. The mean radiographic score for the thirty-three surviving allografts evaluated was 89%, with an average functional score of 27 of a possible 30 points. CONCLUSIONS: Unicondylar osteoarticular allografts of the knee appear to be a reliable alternative for patients in whom reconstruction of massive osteoarticular bone loss is limited to one condyle of the femur or the tibia.     

Nonunion of fractures of the subtrochanteric region of the femur

There are no large clinical series to guide the clinician treating a subtrochanteric nonunion. Deformity, bone loss from previous hardware, and the high stresses in the subtrochanteric region all pose challenges to achieving successful bony union with reoperation. The purpose of this study was to retrospectively review a consecutive series of patients treated with reoperation using contemporary techniques for subtrochanteric nonunion. Between 1992 and 2002, 23 patients with a mean age of 55 years (range, 16-88 years) with 23 subtrochanteric nonunions were treated with additional attempts to achieve union. Two patients were lost to followup. The remaining 21 patients were followed up for a mean of 12 months (range, 6-39 months). Implants used for revision internal fixation were as follows: eight patients were treated with a cephalomedullary nail, seven patients were treated with a standard antegrade femoral nail, five patients were treated with a 95 degree angled blade plate, one patient was treated with a sliding hip screw, one patient was treated with a 95 degree dynamic condylar screw, and one patient was treated with dual large fragment plates. Eighteen of 23 patients had bone grafting: eight had autograft, six had allograft, and both were used in three patients. One patient had free vascularized fibular transfer. Twenty of 21 nonunions healed (95%). At last followup, all patients with healed fractures had no or minimal pain. All were ambulatory. There were no intraoperative complications. There was one postoperative complication (4%), an adynamic ileus that was treated medically. Revision internal fixation and selected bone grafting for subtrochanteric nonunion led to a high rate of fracture union and functional improvement. Intramedullary devices with fixation into the femoral head and neck and fixed angled devices were effective in achieving stable fixation of the proximal bony fragment.