Introduction. Fracture non-union is a devastating cause of patient morbidity. The cost of NU treatment ranges from £7,000 to £79,000. With an estimated 11,700 cases in the UK pa the financial implications are huge, potentially costing several hundreds of million of pounds annually. Successful outcome in the management of non-union is based upon correctly identifying the underlying cause(s) and addressing them appropriately. Aim. The aim of this study was to assess the causative factors in non-union in order to optimize the management of non-union. The causes of NU were categorized into 4 groups (infection, dead bone/gap, host factors, mechanical). Method. 100 consecutive patients who had surgery for long bone non-union were analysed. Information was obtained from the patient clinic visits, notes, radiographs and laboratory results. The cause(s) of the non-union were identified, recorded and divided into 4 groups; host, mechanical, dead bone/gap at NU site and infection. Results. The mean age at time of injury was 41.4(±16.7)years; male/female ratio was 3:1, 80% were lower limb (52% of all cases were tibial). 69% were high energy, 38% were open. 26% of patients had a single attributable cause, 59% had two causes, 14% had three causes and one had all four. Mechanical causation was found in 56% of cases, dead bone/gap in 50%, host factors in 44%, infection in 40% of patients. 5.7% of the infections were unexpected new/occult positive findings. 73% of patients with previously treated infection but without ongoing infection had multiple positive cultures. Conclusion/discussion. Surgical procedures for non-union often address a single aspect yet 74% had more than one attributable cause. With a 6% occult infection rate multiple tissue samples for microbiology and pathology investigation should be carried out routinely in every patient with non-union. The multi-factorial nature of non-union makes meticulous patient assessment vital to maximise the chance of treatment success

Introduction. The Ilizarov method for non-union comprises a range of treatment protocols designed to generate tissue, correct deformity, eradicate infection and secure union. The choice of specific reconstruction method is difficult, but should depend on the biological and mechanical needs of the non-union. We present a prospective series of patients with non-union of the tibia managed using a treatment algorithm based on the Ilizarov method and the viability of the non-union. Patients and methods. Forty-four patients (34 men and 10 women) were treated with 26 viable and 18 non-viable non-unions. Mean duration of non-union was 19 months (range 2-168). 25 patients had associated limb deformity and 37 cases were infected. 42 patients had undergone at least one previous operation. Bone resection was dictated by the presence of non-viable and infected tissue. Four Ilizarov protocols were used (monofocal distraction in 18 cases, monofocal compression in 11 cases, bifocal compression-distraction in 10 cases and 5 bone transports) depending on the stiffness of the non-union or the presence of segmental defect. Results. Union was achieved in 40 cases (91%) with Ilizarov method alone. Infection was eradicated in 35 cases (95%). Monofocal distraction produced union in all 18 stiff non-unions with little morbidity. Monofocal compression was successful in only 82% of mobile non-unions but failed to eradicate infection in 2 of 10 cases. Bifocal techniques allowed infection-free union in all 15 difficult segmental infected non-unions but required prolonged treatment times and bone grafting to docking sites. Complications included pin infection, hindfoot stiffness, refracture, one below-knee amputation and residual limb length discrepancy. Conclusion. We do not recommend monofocal compression in the treatment of infected, mobile non-unions. Distraction (monofocal or bifocal) was effective and is associated with high rates of clearance of infection

Conservative management remains the gold standard for many fractures of the humeral diaphysis with union rates of over 90% often quoted. Success with closed management however is not universal. Phase 1. A retrospective review of all conservatively managed fractures between 2001 and 2005 was undertaken to investigate a suspected high non-union rate and identify possible causes. The overall non-union rate was 39.2% (11 of 28 cases). There was no difference in axial distraction at presentation, however following application of cast there was significantly more distraction in the non-union group (1.2 v 5.09mm, p<0.01). Changes to practise. All humeral fractures were admitted, lightweight U-slabs were applied by a technician, distraction was avoided, patients abstained from NSAIDS, consultant reviewed radiographs before discharge and patients were converted early to functional brace. Phase 2. Prospective collection of data over the following two year period showed a decrease in the amount of distraction when first placed in cast (2.73 v 0.74, p<0.05), a reduction in NSAID use (89% v 38%, p<0.01) and earlier conversion to brace (37 v 20 days p<0.01). These changes to practise led to a dramatic reduction in non-union rate from 39.2% to 4.9% (p<0.01). Conclusion. Over-distraction at first application of cast is a causal factor in the development of non-union. Lightweight cast, avoidance of distraction, abstinence from NSAIDS and early conversion to functional brace is recommended. The initial surgical management of the patient who displays evidence of distraction will prevent evolution of non-union with conservative management. This will avoid lengthy delays in the treatment of the non-union and also help to prevent secondary stiffness in adjacent joints and disuse osteopenia

Definitions and perceptions of good and poor outcome vary between patients and surgeons, and perceived inadequate outcome can lead to litigation. We investigated outcomes of litigation claims relating to non-union and deformity following lower limb long bone fractures from 1995 to 2010. The database of all 10456 claims related to Trauma and Orthopaedic Surgery was obtained from the NHS Litigation Authority. Data was searched for “deformity, non-union and mal-union”, excluding spine, arthroplasty, foot and upper limb surgery. The type of complaint, whether defended or not, and costs was analysed. 241 claims met our criteria, 204 of which were closed, and 37 unsettled. Deformity/mal-union constituted 97, and non-union 143. Coronal/sagittal deformity cost £4.2 million, mean £45,487 (60% received compensation). Rotational mal-unions cost £1.6 million, mean £114,263 (87% received compensation). Non-union cost £5.3 million, mean £75,866 (60% received compensation). Mean legal fees for coronal/sagittal deformity was £18,772, rotational deformity £37,384, and non-union £24,680. The total cost of litigation was £12.2 million, with a mean of £59,597 per settled claim. The mean pay-out for all confirmed negligence/liability was £56,046 (£1,300–£500,000, median £21,500) per case. Non-union is an accepted complication following fracture surgery. However, this does not mitigate against non-union being seen as representing a poor standard of care. While it is unclear whether the payouts reflect a defensive culture or were due to avoidable errors, and notwithstanding the limitations of the database, we argue that failure of the index surgery should prompt a referral to a specialist centre. The cosmetic appearances of rotational malalignment results in higher compensation, reinforcing outward perception of outcome as being more important than harmful effects. We also note that the database sometimes contained conflicting and incomplete data, and make a case for standardisation of this component of the outcome process to allow for learning and reflection

The purpose of this study is to describe the use of intramedullary distraction coupled with an additional osteotomy to achieve union with simultaneous deformity correction and lengthening in femoral non-union. Femoral non-union is a difficult problem often associated with shortening, angulation, and mal-rotation. We report the use of an intramedullary distraction device, with additional osteotomy, to achieve union, restore femoral length and alignment. Simple distraction in femoral non-union is often ineffective, possibly because the non-union site is relatively avascular. Osteotomy is known to increase blood flow and, with lengthening, promote union through distraction histiogenesis. 7 patients with posttraumatic diaphyseal femoral non-union with shortening were studied. Pre-operative planning included long leg standing views, with CT to measure mal-rotation. 6 patients were treated with the Intramedullary Skeletal Kinetic Distractor (ISKD) with an osteotomy distant from non-union site within the parameters required for the device. One patient underwent distraction without osteotomy. Patients were followed to union. Complications and adjuvant interventions were recorded. All 7 patients with femoral non-union treated with ISKD were included and followed up. Patients treated with osteotomy united at average of 9 months with satisfactory deformity correction and lengthening. However patient who underwent pure distraction failed to unite. Complications included failure to lengthen, requiring manipulation, and delayed consolidation of regenerate requiring bone graft. The procedure was well tolerated. The initial results of the management of femoral non-union with deformity by intramedullary distraction coupled with osteotomy are encouraging. Complications were minor and readily manageable. We believe there is an important role for this method in the management of femoral non-unions associated with deformity and length discrepancy

Tibial non-union causes significant morbidity and functional impairment. Circular frames are valuable tools in the treatment of non-union, however prolonged treatment often causes patients increased morbidity due to pin site problems together with personal and emotional strains. The purpose of this study was to assess patient centred outcomes following treatment of tibial non-union in circular frame. We identified 21 patients who had undergone treatment of tibial non-union using a circular frame. Patients were sent questionnaires utilising the Enneking scoring system and EQ-5D general health questionnaire. Fourteen patients responded. There were 3 females, mean age was 48.2 years. The average number of previous operations was 1.2. All patients went on to achieve union with a mean 10.1 months (6–20) in frame. The mean Enneking score was 58.0% (34.3–77.1). Two patients were enthusiastic about their treatment, two liked it, three were satisfied, four accepted and would do it again while three accepted it but would not do it again. The Euroqol questionnaire results showed that 8 patients had some difficulty with mobility, 10 had some difficulty with usual activities and 12 moderate pain. There was no significant difference in the EQ VAS score of overall health state for treated patients compared with predicted scores for an age and sex matched UK population (77.7 vs 83.1, p=0.07). Our study shows that many patients undergoing limb salvage with circular frames for tibial non-union continue to have clinically significant symptoms, however the majority would undergo similar treatment again and we found no difference in overall health state compared with age and sex matched predicted scores. Circular frames are undoubtedly a valuable tool in the management of non-union however patients should be given realistic counselling regarding the treatment and expected outcome

Introduction. Humeral non-union may present a challenging problem. The instability from the un-united fracture leads to pain, disability and significant morbidity. The incidence of humeral shaft non-union as a complication of both operative and non-operative treatment is approximately 8% to 12%. This retrospective study reviews the results of surgical treatment of humeral fracture non-union performed by a single surgeon with a consistent surgical protocol. Material and Method. We present a retrospective analysis of a series of 51 consecutive cases of humeral fracture non-union treated in our limb reconstruction centre. Data were collected on mechanism of injury, associated co-morbidities, smoking, use of NSAIDs and treatment before referral. Patients were followed up to clinical and radiological union. Results. From 1994-2008, 48 patients with established humeral non-union were referred to our unit following initial management locally. Three patients were managed in our unit from the outset. There were 20 male and 28 female patients with a mean age of 53 years (range 15-86 years). There were 34 (68%) diaphyseal, nine (17%) proximal and eight (15%) distal humeral non-unions. The treatment in our unit involved plate fixation in 44 (86%) cases, intramedullary nailing in three (6%) and external fixation in three (6%) patients. Iliac crest bone graft, bone morphogenetic protein or a combination of these were utilised in 44 (86%) cases. Thirty-six patients required at least one surgical intervention to achieve union. Twelve patients had more than one operation. The average follow-up was 19.6 months. Union was achieved in all but one case at an average time of 9.8 months (range 3-24 months); one case developed a functionally inconsequential pseudoarthrosis. Conclusion. Union of ununited humeral fractures can be achieved consistently by providing appropriate mechanical stability and biological environment at the fracture site

Introduction. We describe a minimally invasive technique that permits intra-focal bone graft of non-union sites with minimal disturbance of soft tissues and vascularity, and present the results of this technique. Materials and Methods. 10 patients with established tibia fracture non-union were judged suitable for the technique, and were treated in our limb reconstruction unit between January 1995 to June 2007. Eight patients were male, 2 were females with a mean age of 37.4 years (27–64). Five fractures were in the distal tibia and five were diaphyseal fractures. Five fractures were as a result of high velocity and 6 fractures were open. Average number of previous operations were 3 (range 1–7). Time lapse between injury to trephine grafting procedure was mean of 34 (6–168 months). 5 patients had a sedentary job, 2 were labourers and 2 were not working. There were 5 smokers and 2 obese patients. Six cases were of infected non-unions. Operative technique. Under general anaesthesia, the graft(s) was first obtained from the iliac crest using a guide wire centred trephine. A stab incision was made at the non-union site under image control. A core was taken across the non-union, taking care to breach both bone fragments. A trephine core was rolled in Bone Morphogenetic Protein (BMP-7). This was then introduced through the trephine in to the non-union site and pushed with the plunger into the defect created by the cylindrical plug taken from the site. After this, the trocar with the stylet in place was gradually withdrawn with small oscillating motions to fill in the path of the trocar. This was repeated about 3–4 times at the non-union site. The stab wound were closed with interrupted Nylon. Results. 10 patients underwent treatment for complex non-union after initial operation (average number 3) at the referring centre. Treatment in a circular fixator ranged from 10–33 months (20.7 months) with reapplication in 2 patients. Average number of procedures were 4 (range 1–7). Trephine grafting was performed in all cases; BMP was used in 4 cases along with iliac crest bone graft. All patients attained clinical and radiological union with CT confirmation required in 2 patients. 4 patients with sedentary job returned to work, manual workers returned to low demand occupation. Conclusion. The technique described is a safe and effective treatment method for this complex problem with limited donor site morbidity and avoids prolonged hospital stay

Ultrasound treatment can be used as an alternative to surgical methods for treating non-union or to enhance healing in a delayed union. This study presents our short-term results of using low intensity pulse ultrasound stimulation in long bone non-union. 18 patients with surgically treated long bone non-union were treated using the Exogen® ultrasound stimulator (Smith & Nephew Inc., UK). The average age of patients was 48 years (20–73 yrs). There were 8 femur fractures, 9 tibial fractures and 1 knee arthrodesis. 5 of these patients had infected non-union. The average follow up after initiation of treatment was 4 months (range 2–8 months). They received 20 minutes stimulation daily at the fracture site with regular follow up in clinic. We reviewed their serial radiographs and clinical progress. The mean interval to initiation of the ultrasound treatment was 7.9 months (2–27 months). Complete bony union was obtained in 6 of the 18 cases (33%) within 4 months of initiation of treatment. In 7 of 18 cases (38%) there was good evidence of progress towards bony union, while in 4/18(22%) patients there was no progress towards union at average of 4 months (range 2–8 months). There were no complications noted with this treatment. Most non-unions are treated by surgical revision, with consolidation rates ranging from 85 to 100% according to previous studies. Our study suggested 72% patients showing progress towards union with 33% achieving bony consolidation at 4 months. Patients with infected non-union also benefitted from this treatment. In summary ultrasound treatment can be a viable option to treat long bone non-unions, which may avoid the morbidity and complications associated with surgery

We report the results of anterior plate fixation for symptomatic, mid-shaft clavicle non-union. The superior surface is most commonly used for plate fixation. To the best of our knowledge, there are no clinical reports where anterior plate fixation of the clavicle was used. We included 12 consecutive patients, with symptomatic mid-shaft clavicular non-union, aged between 23 and 56 years during a four-year period (1998-2002). The injury was secondary to RTA in 6 cases, sports-related in 5 and skiing in one. In three patients, the non-union was secondary to superior plating using one third tubular plate, in acute fractures. The most common complaint was anterior shoulder pain (12 cases) followed by brachialgia (4 patients). The operation was performed through an anterior approach. A 3.5mm reconstruction plate was contoured and fixed onto the anterior surface of the clavicle. Bone graft was used in all cases. The average follow up was 22 months. All 12 patients achieved union at an average union time of seventeen weeks. Compared to superior plating, anterior plating has the distinct advantage that the longer screws can be used (as the clavicle is a flat bone, and the AP diameter is larger compared to superoinferior diameter) thus improving the stability of fixation. Our results show that anterior clavicle fixation is safe and effective in achieving union, even in cases following failed superior plate fixation. We therefore recommend anterior plate fixation and bone grafting in symptomatic nonunions of mid third clavicle fractures

Background. Fracture non-union is still a major challenge to the orthopaedic surgeon and established non-union has zero probability of achieving union without intervention. Aim. The purpose of this study was to evaluate the effect of low intensity ultrasound for the treatment of established long bone non-union. Methods. The study was done retrospectively. The case notes of all the patients who were treated by low intensity ultrasound therapy for a period of four years were used for data collection. The radiographs were reviewed to determine that the healing process had not progressed for a minimum of three months before start of low intensity ultrasound treatment. There were 11 patients who met the inclusion criteria of which 9 were male and the rest female (mean age of 50.1 years, range 27-74 years). Results. There were seven closed and four open fractures. The fractured bones included seven tibias, two humerus and two femur. The initial treatment was varied and included surgical treatment (ORIF – 1, IM nailing – 4, External fixation – 3) in 8 patients and conservative treatment (MUA+ POP) in 3 patients. The average time to commencement of ultrasound treatment after the initial fracture was 38 weeks (range 20-72 weeks). All patients were compliant with the use of the device and no side effects were recorded. Bony union was achieved in 8 out of 11 patients (73%). The average time to healing was 27 weeks (range 12 to 48 weeks). The fractures that healed successfully were 6 tibias, one femur and one humerus, and included two infected non-unions. Conclusion. This study shows low intensity ultrasound is an effective way of treating established long bone non-unions particularly tibial non-unions

We evaluated the efficacy of using the expandable nail for treating non-union and malunion of the tibial and femoral shafts. Records of 20 patients were retrospectively reviewed: 12 had femoral non-union, 7 had tibial non-union, and one had tibial malunion. The bones underwent reaming and the largest possible nail sizes were inserted during reoperation. The mean age of the patients was 35 years (26-49) in the tibia group and 53 years (23-85) in the femur group. The fractures were defined according to AO/OTA classification and divided between open and closed. The initial treatment was 6 interlocking intramedullary nails and 2 external fixation in the tibia group, and 6 interlocking intramedullary nails, 3 plates and screws and 2 proximal femoral nails in the femoral group. The respective intervals between the original trauma and re-operation were 12 months and 15 months and the respective operation times were 59 minutes (35-70) and 68 minutes (20-120). All fractures healed satisfactorily without the need for an additional procedure. Healing time was 26 weeks (6-52) and 14 weeks (6-26) in the tibia and femur group, respectively. Limb shortenings of 10cm and 4cm were recorded in one patient each in the tibia group and of 3cm in one patient in the femur group. Using the expandable nail system permitted us to ream the bone and expand the nail to its maximal diameter, i.e. up to 16mm in the tibia and 19mm in the femur. We believe that using a bigger nail diameter contributes to better stabilisation of the fracture and promotes better and faster bone healing. Based on our experience, we recommend the use of the expandable nail system to treat tibia and femur shaft non-unions and malunions

Purpose. The aim of this study was to report the outcomes of a series of patients with clavicle fracture non-union who had undergone open reduction and internal fixation using a contoured locking plate without the use of distant bone graft. Methods. Patients were identified using the hospital database. Records were reviewed to determine basic demographics, operative findings, and radiological outcome. Patients were contacted and details about initial injury and treatment, and return to work and sport were recorded. Disabilities of Arm, Shoulder and Hand (DASH) for both operated and non-operated shoulders were completed. Results. 15 patients with at least 6 months follow-up (average 12.4) were identified. The average age was 39 years. All patients were initially treated in a broad arm sling. All fractures were fixed with a pre-contoured locking plate and all went on to achieve clinical and radiological union. The average DASH on the operated side was 14.5 and 4.2 on the contralateral. All patients had returned to work and regular sport activities. One patient required plate removal due to local irritation. Conclusion. The results of this small series suggest that use of distant bone graft is not necessary when performing open reduction and internal fixation for symptomatic non-union of the clavicle with appropriate preparation of fracture ends and adequate fixation

Despite the publication of numerous studies, controversy regarding the non- operative treatment of type II dens fractures remains. The halo-thoracic vest (HTV) and cervical collar are the most commonly used devices. We sought to compare the outcomes of patients managed with these devices in terms of non-union risk factors and associated complication rates. This study was a retrospective review of adult patients with type II dens fractures treated non-operatively at a level one trauma centre between 2001 and 2007. The patients were identified using a hospital trauma database. Each patient included in the study had a minimum follow up of six months. Patient medical records and imaging studies were reviewed. Union was defined as stable fibrous union or bony union, measured at three months. A p-value of < 0.05 was considered statistically significant. Sixty-seven patients were included. Thirty-five patients were treated using a HTV and 32 with a collar. Non-union was found to be associated with increased time in HTV or collar (p = 0.011) and with a mechanism of injury involving a low fall (p = 0.008). In addition, the proportion of patients with stable union at three months was 60% for the HVT group versus 35% for the cervical collar group (p = 0.10). There were trends to support an increased risk of non-union with a patient age of greater than or equal to 65 years at the time of presentation (p = 0.13) as well as with a fracture displacement of greater than or equal to 2 mm at time of presentation (p = 0.17). Clinically significant complications of the HTV were of greater prevalence than those experienced by collar patients. Sixty percent of HTV patients suffered one or more complications compared with 6% of collar patients. We were unable to demonstrate any clear advantage or disadvantage of either device. Further investigation of mortality would be beneficial, particularly in the patient group injured with a mechanism involving a low fall (which tends to include more elderly patients)

The purpose of this study is to describe the use of the PHILOS plate (Synthes) in reverse configuration to treat complex distal humeral non-unions. Non-union is a frequent complication of distal humeral fracture. It is a challenging problem due to the complex anatomy of the distal humerus, small distal fragment heavily loaded by the forearm acting as a long lever arm with powerful forces increasing the chances of displacement. Rigid fixation and stability with a device of high “pull-out” strength is required. The PHILOS plate has been used in reverse configuration to achieve good fixation while allowing central posterior placement of the implant. 11 patients with established non-union of distal humeral fractures were included in this study. No patient in whom this implant was used has been excluded. Initial fixation was revised using the PHILOS plate in reverse configuration and good fixation was achieved. Bone graft substitutes were used in all cases. Patients were followed to bony union, and functional recovery. All fractures united. One required revision of plate due to fatigue failure. Average time to union was 8 months with excellent restoration of elbow function. A reversed PHILOS plate provides an excellent method of fixation in distal humeral non-union, often complicated by distorted anatomy and previous surgical intervention. It has a high “pull-out” strength and may be placed in the centre of the posterior humerus, allowing proximal extension of the fixation as far as is required. It provides secure distal fixation without impinging on the olecranon fossa. It is more versatile and easier to use than available pre contoured plating systems

Introduction. It is well established that non-union of the scaphoid requires operative intervention to achieve stable union, restore scaphoid anatomy and prevent further degenerative change. Acutrak screw has been shown to have better biomechanical compression properties than the Herbert screw in the laboratory setting. The aims of the study were to assess the rate of union, the functional outcome and post- operative complications of patients with the two different screw systems. Methods. A retrospective review of the patients who had undergone surgery for non-union of scaphoid treated by a single surgeon. The first group consisted of 61 patients who were treated with Herbert screw and iliac crest bone graft between July 1996 and June 2000. The rate and time to union were assessed clinically and radiologicaly. Their post-operative functional outcome was assessed with modified Mayo wrist score. Results were compared to second group of 71 patients treated with Acutrak screw plus iliac crest bone graft between July 2000 and December 2005. Results. The mean ages of patients (25yrs in Herbert Vs 27yrs in Acutrak) were similar in both groups. The mean time interval between injury and surgery was 12.2months for Herbert group and 17months in Acutrak group. Herbert group had 77% (47) union rate compared to 93% (66) for the Acutrak group. There was persistent nonunion in 14 (22.9%) and 5 (7%) patients in Herbert and Acutrak groups respectively. Functional outcome were excellent in 41% of cases, good in 26% of cases, fair in 22%, and poor in 11 % of cases in Herbert group. For Acutrak group, the outcome score were better with excellent 60% of cases, good in 25%, fair in 10% and poor in 5% of the cases. Wrist fusion was performed in 4 and 1 cases for Herbert and Acutrak group respectively due to progressive wrist pain. Conclusion. Acutrak screw offer better union rate and functional outcome following surgery for scaphoid nonunion

Autologous bone graft has been used in the treatment of complex bone defects for more than a century. Morbidity associated with the harvest of this bone graft has led orthopaedic surgeons to seek alternative therapies in the treatment of long bone non-unions. The aim of this study was to determine whether the use of demineralised bone matrix as a bone healing adjunct improves clinical outcomes in adult patients with long bone non-union. A systematic search was carried out of the peer-reviewed English language literature to identify all relevant studies. The search strategy returned a total of 47 studies. Five of these studies were relevant to the research question. The studies were critically assessed and where appropriate combined in a meta-analysis. 4 non-comparative studies and one comparative study were reviewed. An overall estimate of the rate of union for the five studies was 86% (95%CI: 71–94%). The one comparative study demonstrated the relative risk (RR) of healing was not significantly better than in patients treated with autologous bone graft; RR=1.03 (95%CI 0.96–1.12). There are limited data to support the use of demineralised bone matrix in the treatment of long bone non-union. Demineralised bone matrix is likely to be similarly effective to other treatments in the management of non-union. This study confirms the clinical and ethical requirements to proceed with a randomised controlled trial to test the effectiveness of this intervention

Scaphoid injuries remain a challenge in both diagnosis and subsequent management. Untreated scaphoid non-union leads to inevitable osteoarthritis. Non-unions are picked up via an established management pathway in out trust and treated surgically with a standardised technique of non-vascularised bone grafting and retrograde screw fixation. Patients that underwent primary screw fixation and grafting for scaphoid non-union between 2004-2008 were reviewed. These patients were followed up clinically and radiologically. They were assessed for time to union, complications, DASH score, patient satisfaction and return to work. In addition comparison was made between the use of Herbert screw and Twinfix screw fixation and between tricortical iliac crest graft and corticocancellous distal radial bone graft. Forty-two consecutive patients underwent surgery of which 35 (83%) were available for follow up. Mean follow up was 39 months (range 13-72). 31 (89%) patients had gone on to radiological union. Of those united the mean DASH score was 11. Mean time to union was 4.4 months (range 2-15 95% CI 3.3 – 5.5). 95% of patients were satisfied with their operation and would recommend it to a friend in the same situation. No significant difference in outcomes were identified between Herbert screw and Twinfix screw fixation or between iliac crest or distal radial bone grafting techniques. This study has shown encouraging results for an established protocol for management of scaphoid non-union. Outcomes are as a good as any published comparable series in the literature and provide further support to the use of this technique

Aim

To analyse the results of the use of Recombinant Bone Morphogenic Protein (BMP-7) for treatment of fracture nonunions at our institution.

Statement of purpose

To determine the outcome of the use of Bone Morphogenetic Protein 7 (BMP7) as a replacement for bone graft in a limb reconstruction unit