INTRODUCTION. Computer-aided systems have been developed recently in order to improve the precision of implantation of unicompartmental knee replacement (UKR). Minimal invasive techniques may decrease the surgical trauma related to the prosthesis implantation, but there might be a concern about the potential for a loss of accuracy. Mobile bearing prostheses have been developed to decrease the risk of polyethylene wear, but are technically more demanding. Navigation might help to compensate for these difficulties. We wanted to combine the theoretical advantages of the three different techniques by developing a navigated, minimal invasive, mobile bearing unicompartmental knee prosthesis. MATERIAL AND METHODS. 160 patients have been operated on at our institution with this system. The 81 patients with more than 2 year follow-up have been re-examined. Complications have been recorded. The clinical results have been analyzed according to the Knee Society Scoring System. The subjective results have been analyzed with the Oxford Knee Questionnaire. The accuracy of implantation has been analyzed on post-operative antero-posterior and lateral long leg X-rays. The 2-year survival rate has been calculated. RESULTS. We observed 8 complications related to the implant or the operative technique: 2 cases of meniscus instability (1 revision to TKR, 1 bearing exchange); 2 cases of tibia loosening (revised to TKR), 2 cases of femoral loosening (revised to TKR), 1 case of lateral disease progression (revised to TKR), 1 case of unexplained pain syndrome (revised to TKR). The mean Knee Score was 93 points, 44% had the maximum of 100 points, and only 10% have less than 85 points. The mean pain score was 48 points/50. The mean flexion angle was 128°, and 60% had at least 130° of knee flexion. The mean Function Score was 97 points, 84% have the maximum of 100 points, and only 5% had less than 85 points. The mean Oxford Knee Questionnaire score was 19 points (best result: 12 points, worst result: 60 points). Expected limb axis correction was obtained in 77% of the cases. 62% of the cases had an optimally implanted prosthesis for all studied criteria. The 2-year survival rate was 97%. DISCUSSION. Most of the revision cases were related to technical difficulties during the development phase. Fixation of the implant has been improved, and some imprecise steps of the software have been corrected. Since these changes occurred, no severe early complication related to implant or software has been observed. The current implant is considered reliable, and the current minimal invasive navigated technique is considered reliable as well

It is widely accepted by orthopaedic surgeons that antibiotics should be withheld until aspiration has been performed to increase the odds of identifying an organism in septic arthritis. Patients often present to other specialties that may not be as familiar with these principles. Twenty-five of forty-nine patients with septic arthritis of the native or prosthetic knee had received antibiotics prior to review by the orthopaedic service. Patients were significantly less likely to demonstrate an organism on initial microscopy (entire cohort p=0.001, native knees p=0.006, prosthetic knees p=0.033) or on subsequent culture (entire cohort p=0.001, native knees p=0.017, prosthetic knees p=0.012) of their aspirate if they had received antibiotics. The sensitivity of microscopy dropped from 0.58 to 0.12 when patients had received antibiotics (native knees 0.46 to 0, prosthetic knees 0.72 to 0.27). The sensitivity of the culture dropped from 0.79 to 0.28 when the patient had received antibiotics (native knees 0.69 to 0.21, prosthetic knees 0.91 to 0.36). Patients treated with empirical antibiotics are less likely to demonstrate an organism on microscopy and culture of their initial aspirate. There is a significantly high false negative rate associated with knee aspiration, particularly with prior administration of antibiotics

Aims

Several previously identified patient-, injury-, and treatment-related factors are associated with the development of nonunion in distal femur fractures. However, the predictive value of these factors is not well defined. We aimed to assess the predictive ability of previously identified risk factors in the development of nonunion leading to secondary surgery in distal femur fractures.

It is becoming increasingly common for a patient to have ipsilateral hip and knee replacements. The inter-prosthetic (IP) distance, the distance between the tips of hip and knee prostheses, has been thought to be associated with an increased risk of IP fracture. Small gap distances are generally assumed to act as stress risers, although there is no real biomechanical evidence to support this. The purpose of this study was to evaluate the influence of IP distance, cortical thickness and bone mineral density on the likelihood of an IP femoral fracture. A total of 18 human femur specimens were randomised into three groups by bone density and cortical thickness. For each group, a defined IP distance of 35 mm, 80 mm or 160 mm was created by choosing the appropriate lengths of component. The maximum fracture strength was determined using a four-point bending test. The fracture force of all three groups was similar (p = 0.498). There was a highly significant correlation between the cortical area and the fracture strength (r = 0.804, p <  0.001), whereas bone density showed no influence. This study suggests that the IP distance has little influence on fracture strength in IP femoral fractures: the thickness of the cortex seems to be the decisive factor. Cite this article: Bone Joint J 2014;96-B:1378–84

Introduction. Unicompartmental knee arthroplasty (UKA) in patients with isolated medial osteoarthritis of the knee is nowadays a standard procedure with good results, especially with the minimally-invasive approach. However, the survival rate of the unicompartmental knee prostheses is inferior to that of total knee prostheses. Therefore, further studying of UKA is still necessary. In most mobile bearing designs the femoral component has a spherical surface and therefore its positioning is not crucial. The role of the tibial slope in UKA has not been investigated so far. The manufacturers recommend tibial slopes with values between 10° positive slope and 5° negative slope. Most surgeons try to reconstruct the anatomical slope with a high failure by measuring the slope on x-rays. The aim of this study was to investigate the influence of the tibial slope on the wear rate of a medial UKA. Materials and methods. In vitro wear simulation of medial mobile bearing unicompartmental knee prosthesis with a spherical femoral surface (Univation ®) was performed with a customized four-station servo-hydraulic knee wear simulator (EndoLab GmbH, Thansau, Germany) reproducing exactly the walking cycle as specified in ISO 14243–1:2002(E). The tibial tray was inserted with 2 different medial tibial slopes: 0°, 8° (n=3 for each group). The lateral tibial slope of the space-holder was not changed (0° for every group). We performed a total of 5 million cycles for every different slope, the gravimetric wear rate was determined gravimetrically using an analytical balance every 500 000 cycles according to the ISO 14243–2. Results. The wear rate in the 0° slope group was 3.46±0.59 mg/million cycles, and in the 8° slope group it was 0.99±0.42 mg/million cycles. The difference between the 0° tibial slope group and the 8° tibial slope group was highly significant (p<0.01, alternate t-test). Discussion. An increase of the tibial slope leads to a reduced wear rate in a mobile bearing UKA. Therefore, a higher tibial slope should be recommended for mobile bearing UKA. However, the influence on the ligaments has to be considered as a higher tibial slope leads to an increased strain on the anterior cruciate ligament. This influences needs to be investigated in further studies before a definite optimal range for the tibial slope can be recommended

Background. No study compared the clinical results of the posterior-stabilized mobile-bearing knee with those of nonposterior-stabilized mobile-bearing knee in the same patients. The purpose of this study was to examine whether the clinical and radiographic results, range of motion, patients satisfaction, and complication rates would be better in the knees with a posterior-stabilized mobile-bearing knee than in the knees with a nonposterior-stabilized mobile-bearing knee. Methods. One hundred and fourteen patients (mean age, 67.9 years) received a nonposterior-stabilized mobile-bearing knee prosthesis in one knee and a posterior-stabilized mobile-bearing knee prosthesis in the contralateral knee. Seven patients were men, and 107 were women. At the time of each follow-up (mean, 7.3 years; range, seven to 7.6 years), the patients were assessed clinically and radiographically. Results. The mean postoperative Knee Society knee score (95 points versus 96 points, p=0.176), Hospital for Special Surgery knee score (92 points versus 93 points, p=0.077), and Western Ontario and McMaster University Osteoarthritis score (21 versus 20 points, p=0.785) were similar between the two group. At the final follow-up, the average range of motion was 27.7° (range, 70° to 150°) in the knees with a nonposterior stabilized mobile-bearing prosthesis and it was 132° (range, 90 to 150 °) in the knees with a posterior-stabilized mobile-bearing prosthesis. Complication rates (2.6% versus 1.8%) were similar between the two groups. The estimated survival rate was 97.4% at eight years with an overall revision rate of 2.6% (three of 114 knees) in the nonposterior-stabilized mobile-bearing prosthesis group and 98.2% at eight years with an overall revision rate of 1.8% (two of 114 knees) in the posterior-stabilized mobile-bearing prosthesis group. Conclusions. After a minimum duration of follow-up of seven years, we found no significant differences between the two groups with regard to the clinical and radiographic results, or patient satisfaction, or complication rate. However, the posterior-stabilized mobile-bearing prosthesis group had a greater range of knee motion than the nonposterior-stabilized mobile-bearing prosthesis group

Introduction. Minimally invasive implanted unicompartmental knee arthroplasty (UKA) leads to excellent functional results. Due to the reduced intraoperative visibility it is difficult to remove extruded bone cement particles, as well as bone particles generated through the sawing. These loose third body particles are frequently found in minimally invasive implanted UKA. The aim of this study was to analyse the influence of bone and cement particles on the wear rate of unicompartmental knee prostheses in vitro. Material & Methods. Fixed- bearing unicompartmental knee prostheses (n = 3; Univation F®, Aesculap, Tuttlingen) were tested with a customized four-station servo-hydraulic knee wear simulator (EndoLab GmbH, Thansau, Germany) reproducing exactly the walking cycle as specified in ISO 14243-1:2002. After 5.0 million cycles crushed cortical bone chips were added to the test fluid for 1.5 million cycles to simulate bone particles, followed by 1.5 million cycles blended with PMMA- particles (concentration of the third-body particles: 5g/l; particle diameter: 0.5- 0.7 mm). Every 500 000 cycles the volumetric wear rate was measured (ISO 14243-2) and the knee kinematics were recorded. For the interpretation of the test results we considered four different phases: breaking in- (during the first 2.0 million cycles), the steady state- (from 2.0 million to 5 million cycles), bone particle- and cement particle phase. Finally, a statistical analysis was carried out to verify the normal distribution (Kolmogorov-Smirnov test), followed by direct comparisons to differentiate the volumetric wear amount between the gliding surfaces (paired Student's t-test, p<0.05). Results. The wear rate was 12.5±0.99 mm. 3. /mio. cycles in the breaking-in phase and decreased during the steady state phase to 4.4±0.91 mm. 3. /mio cycles (not significant, p = 0,3). The bone particles did not have any influence on the wear rate (3.0±1.27 mm. 3. /mio cycles; p = 0,83) compared to the steady state phase. The cement particles, however, lead to a significantly higher wear rate compared to the steady state phase (25.0±16.93 mm. 3. /mio cycles; p<0.05). Discussion. To our knowledge this is the first study demonstrating that free cement debris which can be found after minimally invasive implanted UKA increases significantly the wear- rate. Bone particles generated for instance through sawing during implantation, however, had no influence on the prostheses wear rate. Our Data suggests, that it is extremely important to remove all the extruded cement debris accurately during implantation in order to avoid a higher wear rate which could result in an early loosening of the prostheses

Aim. To analyse our results after en-block resection of aggressive GCT during 20 years period. Methods. We review 86 patients with skeletal GCT during the last 20 years, from 1990 until 2009, retrospectively. In the cases of latent and active type, extended curettage and bone graft or cement were our treatment of choice, while in aggressive ones we performed en block resection and reconstruction by fibular autograft (e.g. in distal part of radius) or fusion/hinge joint prosthesis (e.g. in GCT around the knee joint). We describe the recurrences, metastases and complications according to treatment. Results. There was no recurrence in 18 cases of en block resection and segmental bone defects were reconstructed with fibular autograft (5), joint fusion (4) and hinge joint arthroplasty (9). We had 2 cases of pulmonary metastasis that underwent resection of the metastasis. In one case, internal fixation failed and the graft broke; thus revision was performed. The rest 68 cases underwent extended curettage and bone graft (51) or cement (17). We had 7 cases of recurrence, 6 treated with repeated curettage and bone cement and one with en lock resection. No death or major complication was reported. Conclusions. In comparison of en block resection with extended curretage, the recurrence rate is greater with the latter; however it results in good control of the disease with less morbidity. In recent years, the invention of hinge knee prosthesis has increased the quality of patients' life in whom we could not preserve the involved joint

Introduction. When introducing new joint replacement designs, it is difficult to predict with any certainty the clinical performance of the new designs. Using roentgen stereophotogrammetric analysis (RSA) to evaluate the first two years of follow-up can serve as a predictor of late mechanical loosening for hip and knee prostheses. This prospectively randomized study was designed to evaluate the clinical performance of the Triathlon total knee system and compare the results between the two versions; posterior stabilized (PS) and cruciate retaining (CR). Methods. Sixty patients were consecutively randomized (two patients left the study prior to surgery) to receive either the Triathlon total knee PS (30 patients) or the Triathlon total knee CR (28 patients). All components were cemented. The study was approved by the Ethical Committee for Lund University. All patients met the inclusion criteria. There were no statistically significant differences between the demographics for PS and CR. RSA was set to be the principal evaluation parameter. Patient outcome was assessed by the KSS and KOOS questionnaires. Results. There were no statistically significant differences in rotation or translation around or along the three coordinal axes. Neither were there any differences in the maximal total point motion (MTPM) during the 2-year follow-up (p<0.05). There were no statistically significant differences between the clinical results of both groups when using the KSS and the KOOS scoring system (p<0.05). Discussion. There were no clinically significant differences in the RSA 2-year follow-up data, nor in the clinical scores of KSS and KOOS. The results of this study suggest that there is no major differences between the posterior stabilized system and the cruciate retaining system for the Triathlon total knee neither in clinical nor in predictable long term mechanical performance

Background. A number of studies have reported on the early failure of the Oxford unicompartmental knee arthroplasty. However, less evidence is available regarding the outcome following revision of failed unicompartmental knee prostheses to total knee arthroplasty. The aims of this study were to determine the time to failure for the Oxford unicompartmental knee arthroplasty and to assess the short-term outcome following revision surgery. Methods. Details of consecutive patients undergoing revision of an Oxford unicompartmental knee arthroplasty to a total knee arthroplasty at our centre between January 2000 and December 2009 were collected prospectively. Data was collected on patient demographics, indication for revision surgery, and time to revision from the index procedure. Clinical and radiological outcome following revision arthroplasty was also assessed. Results. During the study period 22 (4.5%) of 494 Oxford unicompartmental knee arthroplasties were revised to a total knee arthroplasty. Mean age at the time of revision surgery was 61.8 yr and 13 (59%) patients were male. Mean time to revision surgery from the primary procedure was 3.0 yr (range 0.6–6.2 yr). The commonest reasons for revision were aseptic loosening of the femoral (n=9) or tibial component (n=2), and undiagnosed (n=5) or patellofemoral pain (n=2). All patients were revised to a cemented total knee arthroplasty with most not requiring bone grafts. During follow-up (range 0.5–4.5 yr) no further surgery was performed in the 22 patients. In addition, there were no major postoperative complications and no evidence of radiological failure. Discussion. The present study demonstrates most failures of the Oxford unicompartmental knee arthroplasty occur within three to four years following the index procedure and are due to aseptic component loosening. These findings are consistent with other published reports regarding the early failure of this particular prosthesis. The short-term outcome following revision surgery appears to be good, however longer follow-up periods are required to determine if these good results continue

Although the use of constrained cemented arthroplasty to treat distal femoral fractures in elderly patients has some practical advantages over the use of techniques of fixation, concerns as to a high rate of loosening after implantation of these prostheses has raised doubts about their use. We evaluated the results of hinged total knee replacement in the treatment of 54 fractures in 52 patients with a mean age of 82 years (55 to 98), who were socially dependent and poorly mobile. Within the first year after implantation 22 of the 54 patients had died, six had undergone a further operation and two required a revision of the prosthesis. The subsequent rate of further surgery and revision was low. A constrained knee prosthesis offers a useful alternative treatment to internal fixation in selected elderly patients with these fractures, and has a high probability of surviving as long as the patient into whom it has been implanted

Background. In the literature are different data about the allogenic blood transfusion rate after total knee replacement. The common intention in orthopedic surgery is to reduce the requirement for allogenic blood transfusions by optimizing the blood management. The aim of this study is to determine the efficacy of the mechanical autotransfusion system OrthoPAT® to reduce the postoperative allogenic blood transfusion (ABT) rate. Method. According to the preliminary performed power analysis we did a prospective controlled study including 151 patients which were randomized in a group A (OrthoPAT® for intra- and postoperative blood salvage and retransfusion, n=76 patients) and a control group B (no retransfusion system was used, n=75 patients). All patients had a primary osteoarthritis of the knee and were operated on without use of a tourniquet. We implanted in all patients a cemented posterior stabilized total knee prosthesis design. In group A the autotransfusion system was used for 6 hours (intra- and postoperatively) and the collected blood was retransfused. The retransfused blood was anticoagulated, filtered and centrifuged to separate waste products. Red cells were washed with saline and reconcentrated to a high hematocrit. The preoperative data for cardiopathy, angiopathy, preoperative anemia or anticoagulant treatment showed no significant differences for group A and B. Because of missing data we finally were able to use the results of 140 patients: 70 group A and 70 in group B. The indications for a blood transfusion were influenced by the clinical symptoms of anemia, the hemoglobin value (hemoglobin < 8.0 g/dl) and the anamnesis of cardiovascular diseases. Evaluation was done with the medical history and the pre-/postoperative hemoglobin values and postoperative need of allogenic blood transfusion. Results. The two groups showed no significant differences relating to the demographic data or the medical history. 23 patients (33 %) of the retransfusion group who in mean received 281 ml of salvaged blood needed allogenic blood transfusion compared with 23 patients (33 %) of the control group B (p=0,999). The hemoglobin values of group A versus the control group showed after the donation of the salvaged blood a tendency to a higher hemoglobin value (p=0,062) but no longer at the third and fifth day postoperative. Conclusions. In this clinical observation the use of the autotransfusion system does not reduce the postoperative allogenic blood transfusion rate. At the third and fifth day postoperatively no significant differences of the hemoglobin values could be stated comparing group A with group B

Introduction. Malrotation following total knee replacement is directly related to poor outcome. The knowledge of proximal and distal rotational axes and angles of the femur is therefore of high importance. The aim of the study was to determine whether the most used proximal and distal femoral angles; femoral anteversion angle (FAA) and posterior condylar angle (PCA) were different within individuals, between right, left and gender. As well, we studied whether the “inferior condylar angle” is correlated to the PCA and therefore useful in determining the rotation of the distal femur. Material and Methods. From 36 cadavers the femora were obtained and after removing the soft tissue a Computed Tomography (CT) scan was made. Three angles were measured: (i) the FAA between femoral columnar line (FHNL) and posterior condylar line (PCL), (ii) the PCA between anatomical transepicondylar line (TEL) and PCL, (iii) the inferior condylar angle (ICA) between the TEL and inferior condylar line (ICL). Statistical analysis of comparative relationships between the different angles was examined by calculating correlation coefficients and a paired t-test. Results. The mean FAA, PCA and ICA for the whole group were respectively 12.0 degrees (range 0.2–31.6, SD 8.3, 95% CI 9.6–14.4), 4.8 degrees (range 0.9–9.6, SD 2.3, 95% CI 4.1–5.4) and 4.5 degrees (range 0.1–9.8, SD 2.1, 95% CI 3.9–5.1). A strong correlation of the FAA was found within the total group and left versus right (r = 0.82; p = 0.00). A weaker relationship was found for the total group of the PCA measurements (r = 0.59; p = 0.00). When FAA compared to the PCA subdivided in only sexes, there is a weak correlation for the female group (r = 0.54; p = 0.00) Despite the small mean difference of the mean ICA and PCA, there was no correlation between these two angles. Conclusion. Considering the weak correlation of the FAA and PCA within the group but also individuals, the importance of development of more individual approaches for determining the optimal rotation of the components in total knee surgery is essential. As a result, one should be aware that the widely used, current guidelines for knee rotation of 3 degrees of external rotation in placing total knee arthroplasties shows variation between individuals. A more individual approach in total knee arthroplasty seems essential for future knee prosthesis implantations

Introduction. In valgus knees, ligament balance remain difficult when implanting a total knee arthroplasty (TKA), this leads some authors to systematically propose the use of constrained devices. Others prefer reserving higher constraints to cases where it is not possible to obtain final satisfactory balance: less than 5 of residual frontal laxity in extension in each compartment, and a tibiofemoral gap difference not in excess to 3mm between flexion and extension. The goal of the study was to assess if is possible to establish preoperative criteria that can predict a constrained design prosthetic implantation at surgery. Materials and Methods. A consecutive series of 93 total knee prostheses, implanted to treat a valgus deformity of more than 5 was retrospectively analysed. Preoperatively, full weight bearing long axis AP views A-P were performed: hip knee angle (HKA) averaged 195 (186 to 226), 36 knees had more than 15 of valgus, and 19 others more than 20 of valgus. Laxity was measured by stress radiographies with a TelosTM system at 100 N. Fifty-two knees had preoperative laxity in the coronal plane of more than 10. Fourteen knees had more than 5 laxity on the convex (medial) side, 21 knees had more than 10 laxity on the concave (lateral) side. Statistical assessment, using univariate analysis, identified the factors that led, at surgery, to an elevated constraint selection level; these factors of independence were tested by multivariate analysis. Logistical regression permitted the classification of the said factors by their odds ratios (OR). Results. High-constraints prostheses (CCK type) were used in 26 out of 93 TKA, the other TKA were regular posterostabilized (PS) prostheses. Statistically, the preoperative factors that led to the choice of a constrained prosthesis were: (1) valgus severity as measured by HKA (PS = (PS = 193, CCK = 198), (2) increased posterior tibial slope (PS = 4.8, CCK = 6.5), (3) low patellar height (using Blackburne and Peel index PS = 0.89, CCK = 0.77), (4) severity of laxity in valgus (PS = 2.3, CCK = 4.3). Among all these factors, the only independent one was laxity in valgus (convex side laxity) (p = 0.0008). OR analysis showed a two-fold increased probability of implanting an elevated constraints prosthesis for each one degree increment of laxity in valgus. Discussion. This study demonstrated that it was not the valgus angle severity but rather the convex medial side laxity that increased the frequency of constrained prostheses implantation. Other factors, as a low patellar height or an elevated posterior tibial slope, when associated, potentiate this possible prosthetic switch (to higher constraints) and should make surgeons aware, in these situations, of encountering difficulties when establishing ligament balance

Introduction. We compared standard NexGen Cruciate substituting-flex prosthesis with gender-specific NexGen Cruciate substituting flex prosthesis in patients undergoing bilateral total knee arthroplasty with regard to Coverage of the bone by femoral component, Clinical outcome, Radiographic outcome, Survival and complication rates, with special emphasis on patellofemoral complications. Material & Methods. 30 female patients with osteo-arthritis of the knees with similar deformity and preoperative range of motion were randomized to have one knee replaced with a gender non-specific Nexgen Cruciate substituting (Zimmer) prosthesis and the other with Gender Cruciate substituting (Zimmer) prosthesis. Follow-up clinical evaluation was done at 6, and 12 months postoperatively and then at yearly interval using “Knee Society” Recommendations. Clinical assessment of the patellofemoral joint of the replaced knees was done with the help of “Hospital for Special Surgery” patellar scoring system. The overall patient satisfaction after surgery was evaluated with use of the British Orthopaedic Association patient satisfaction score. Results. The mean age was 63.8 years (range 49–76 years). 14 patients had standard NexGen Cruciate substituting-flex gender knee prosthesis on right side and 16 patients had on left side. On the other sides, standard NexGen Cruciate substituting-flex prosthesis implant was used. Preop Clinical findings were similar in both groups. The average Preop range of motion was identical in both the groups. The average follow up was 2.5 years (2–3.8 years). The mean post-operative Knee Society knee scores were 88,90,94, (70 to 100) and 87,92, 93 (70 to 100) points and the function scores were 83,85,88 (60 to 100) and 84.86.88 (60 to 100) points for the standard implants and the gender-specific designs, respectively at 6 months, 1 year and 2 year follow up. Mean improvement in patella score was 65, 71.1 and 73.3 points in Gender group at the end of 6 month, 1 yr and 2 yr respectively as compared to 65.6, 71.1 and 72.2 in Nexgen group. No patient had malpositioning of components or radiolucencies at the end of 2 year follow up. Mean improvement in BOA functional score was 20.6, 22.6 and 24.1 points in Gender group at the end of 6 months, 1 year and 2 year respectively as compared to 20.7, 21.8 and 23.2 in Nexgen group. In those with a standard prosthesis, the femoral component was closely matched in 30 knees, overhung in 17 and undercovered the bone in 13. In those with a gender-specific prosthesis, it was closely matched in 45 knees and undercovered the bone in 15. Conclusion. Although gender specific knee matched the femoral anatomy of Indian female knees better than the standard nexgen Knee, we found no significant differences between the two groups with regard to the clinical and radiological results, patient satisfaction or complication rate

The treatment of infected exposed implants which have been used for internal fixation usually involves debridement and removal of the implant. This can result in an unstable fracture or spinal column. Muscle flaps may be used to salvage these implants since they provide soft-tissue cover and fresh vascularity. However, there have been few reports concerning their use and these have concentrated on the eradication of the infection and successful soft-tissue cover as the endpoint. There is no information on the factors which may influence the successful salvage of the implant using muscle flaps.

We studied the results and factors affecting outcome in nine pedicled muscle flaps used in the treatment of exposed metal internal fixation with salvage of the implant as the primary endpoint. This was achieved in four cases. Factors predicting success were age < 30 years, the absence of comorbid conditions and a favourable microbiological profile. The growth of multiple organisms, a history of smoking and the presence of methicillin-resistant Staphylococcus aureus on wound cultures indicated a poor outcome. The use of antibiotic beads, vacuum-assisted closure and dressing, the surgical site, the type of flap performed and the time from primary surgery to flap cover were not predictive of outcome.