Abstract
Background: A variety of methods are available for the fixation of femoral shaft fractures after total hip arthroplasty. However, few studies in the literature have quantified the performance of such repair constructs. The aim of this study was to evaluate biomechanically four different constructs for the fixation of periprosthetic femoral shaft fractures following total hip arthroplasty.
Methods: Twenty synthetic femora were tested in axial compression, lateral bending, and torsion to determine initial stiffness, as well as stiffness following fixation of a simulated femoral midshaft fracture with and without a bone gap. Four fracture fixation constructs (five specimens per group) were assessed: construct A was a Synthes locked plate (a twelve-hole broad dynamic compression plate) with locked screws; construct B, a Synthes locked plate (a twelve-hole broad dynamic compression plate) with cables and locked screws; construct C, a Zimmer nonlocking (eight-hole) cable plate with cables and nonlocked screws; and construct D, a Zimmer nonlocking (eight-hole) cable plate with allograft strut, cables, and nonlocked screws. Axial stiffness, lateral bending stiffness, and torsional stiffness were assessed with respect to baseline intact specimen values. Axial load to failure was also measured for the specimens.
Results: Construct D demonstrated either equivalent or superior stiffness in all testing modes compared with the other constructs in femora with both a midshaft fracture and a bone gap. A comparison of constructs A, B, and C demonstrated equivalent stiffness in all test modes (with one exception) in femora with a midshaft fracture and a bone gap.
Conclusions: A combination of a nonlocking plate with an allograft strut (construct D) resulted in the highest stiffness of the constructs examined for treating a periprosthetic fracture around a stable femoral component of a total hip replacement.
Clinical Relevance: A locked plate (constructs A and B) should be used with caution as a stand-alone treatment for the fixation of a periprosthetic femoral shaft fracture following total hip arthroplasty, particularly with good bone stock.
Although they are not common, femoral fractures occur in approximately 0.1% to 6% of all patients who have a total hip arthroplasty1-4. Periprosthetic femoral fractures are found mostly in elderly women with osteopenia or in patients who have had loosening of the femoral component4-8.
Femoral fractures at the tip of a stem of a total hip replacement (Vancouver type-B1 fractures9) occur in 75% of all patients with periprosthetic fractures, are the most complex to manage, are associated with the most complications (such as nonunion in 25% to 42% of all such fractures treated nonoperatively), and continue to cause controversy as to which surgical intervention is best4,5,10-13. Periprosthetic fractures around stable femoral components provide a difficult challenge for the treating surgeon, with most advocating open reduction and internal fixation of the fracture14-20. The goals of treatment for this injury pattern include the achievement of fracture union while retaining proper function of the prosthesis. The most frequently utilized extramedullary approach involves some variation of the Ogden construct, which is a metal plate placed laterally on the femur with proximal fixation with cables and distal fixation with bicortical screws21. Other similar methods, incorporating combinations of allograft struts, plates, screws, and cables, have been tested biomechanically4,20,22-25 and used in clinical practice10,18,26-31. Although these approaches are clinically available, a limited number of studies have quantitatively determined the biomechanical strength and stability of these constructs3,4,22,23,25,32,33.
The aim of this investigation was to characterize and compare the biomechanical behavior of four fixation systems used to fix periprosthetic femoral fractures near the tip of a total hip replacement.
Intact Specimen Preparation
Twenty large left adult synthetic femora (Third Generation Composite Femur, model 3306; Sawbones Worldwide, a division of Pacific Research Laboratories, Vashon, Washington) were used in this study. The femoral condyles were partly removed with use of a band saw on the lateral and medial sides for proper insertion into square-channel steel tubing. The long axes of the femora were aligned vertically in the frontal and sagittal planes, and the steel tube sections were filled with anchoring cement. Synthetic femoral analogues were used because they offered attractive advantages over human cadaveric bone, including the fact that they are easier to store, are less expensive and easier to obtain, have no biohazard, and greatly reduce interspecimen variability in physical properties34.
Construct Preparation
Femora were assigned to four groups of five femora each to receive total hip arthroplasty components and fracture fixation devices. The total hip arthroplasty components consisted of an Exeter femoral stem and head (28-mm diameter) with a standard neck in combination with an Omnifit cup insert (0°, Series II; Stryker Orthopaedics, Mahwah, New Jersey). Exeter hip stems were inserted into the medullary canal and cemented with use of polymethylmethacrylate bone cement (Simplex P; Stryker Orthopaedics). The specimens were then osteotomized with use of an industrial band saw outside the cement mantle at 2.5 cm distal to the stem tip to simulate a 45° oblique midshaft fracture (superolateral to inferomedial to stop the buttress effect of the plate during axial compression). The fracture fixation constructs applied to the intact femora were as follows:
Construct A was a Synthes locked (twelve-hole) broad dynamic compression plate (Paoli, Pennsylvania) and locked screws. Fixation proximal to the osteotomy site was done with use of four 5-mm (14-mm-long) locked unicortical screws. Fixation distal to the osteotomy site was achieved with use of four 5-mm (40-mm-long) locked bicortical screws.
Construct B was a Synthes locked (twelve-hole) broad dynamic compression plate with cables and locked screws. Proximal fixation was accomplished with two 5-mm (14-mm-long) locked unicortical screws and two 1.7-mm cables with crimps on cable mounts that locked into the plate. Distal fixation was achieved with four 5-mm (40-mm-long) locked bicortical screws.
Construct C was a Zimmer nonlocking (eight-hole) cable plate (Warsaw, Indiana) with cables and nonlocked screws. Proximal fixation was achieved with two 4.5-mm (14-mm-long) nonlocked unicortical screws and two 1.8-mm cables for the cable plate. Distal fixation was achieved with use of four 4.5-mm (40-mm-long) nonlocked bicortical screws.
Construct D was a Zimmer nonlocked (eight-hole) cable plate with a cadaveric femoral allograft strut placed anteriorly, with cables and nonlocked screws. Proximal fixation was achieved with use of two 4.5-mm (14-mm-long) nonlocked unicortical screws and two 1.8-mm cables for the cable plate. The proximal screws provided cortical fixation only, but no fixation into the cement mantle, as with constructs A, B, and C. Distal fixation was achieved with four 4.5-mm (40-mm-long) nonlocked bicortical screws and two 1.8-mm cables for the cable plate. In this construct, all cables encircled the strut allograft. The struts were embalmed, cut to length with use of a band saw, and randomly assigned for application to a host specimen.
The length of the plates (224 mm and 246 mm for locked and nonlocking plates, respectively) and of the allografts (220 mm) ensured a similar maximum working length for each construct, since this is a factor that influences construct stiffness and load to failure.
General Test Procedure
Twenty intact femora were first tested to obtain baseline values for axial compression, lateral bending, and torsional (internal rotation of the femoral head) stiffness. Following total hip arthroplasty, osteotomy, and application of the fracture fixation constructs, all specimens were retested for axial, lateral bending, and torsional stiffness. A 5-mm bone gap was then created at the osteotomy site, and all specimens were again retested for axial, lateral bending, and torsional stiffness. All specimens were finally loaded to failure in axial compression. Throughout this report, the phrase midshaft fracture refers to the initial osteotomy performed with anatomical reduction, whereas the phrase midshaft bone gap refers to the 5-mm introduced gap, which mimics a comminuted fracture or a nonreduced fracture. Each intact and construct stiffness value was obtained from an average of three trials. A materials-testing machine (model 8874; Instron, Canton, Massachusetts) was used for all experiments.
Axial Stiffness Tests
Each intact femoral specimen was oriented in 25° of adduction in the frontal plane and aligned vertically in the sagittal plane to simulate anatomical single-leg stance. Distally, the femur was mounted in an industrial swivel vise. Proximally, the femoral head was inserted into a conforming cup that was cut out of a stainless-steel cylindrical block, which was oriented with no anteversion with respect to the femoral neck. The femoral head was not fixed but was free to rotate inside the cup. Vertical load was applied at the femoral head to a maximum of 1000 N (displacement control at a load rate of 8 mm/min and a preload of 100 N) to provide axial compression. The slope in the linear region of the load-deflection curve was used to calculate axial baseline stiffness. All specimens remained within the linear elastic region to avoid permanent specimen damage, as demonstrated by the linearity coefficient during the ramp-up and ramp-down loading procedure (average R2 > 0.95). Following total hip arthroplasty and fracture fixation, the same alignment procedure was followed except that the femoral head was fitted proximally into a mating acetabular polyethylene cup, which was potted into the jig (Fig. 1). Similar loading regimes were followed except that maximum loads applied for femora with midshaft fractures and gaps were 500 N and 250 N, respectively, with a preload of 50 N. The load levels, although below what is expected physiologically for many activities of daily living, were chosen in order to prevent permanent damage to the specimens during stiffness testing.
Lateral Stiffness Tests
Intact femora were mounted horizontally onto a jig. The femoral head was free to slide under a load application plate. A support block was placed 160 mm from the top of the steel chamber used for potting the distal condyles in order to minimize long-axis bending. A vertical load of 200 N (preload, 50 N) at 8 mm/min was then applied to the femoral head to create lateral bending. The slope of the linear elastic portion of the load-deflection graph defined lateral bending stiffness. All specimens remained within the linear elastic region to avoid permanent specimen damage, with an average linearity coefficient of R2 > 0.95. The same procedure was used for constructs having midshaft fractures and bone gaps, except that the maximum load applied was 50 N (preload, 10 N). Loads were chosen to avoid permanent damage to the specimens.
Torsional Stiffness Tests
Intact femora were mounted horizontally onto a jig. The femoral head was free to slide under a load application plate. A support block was placed 305 mm from the top of the steel potting chamber, just distal to the trochanteric region, to minimize long-axis bending. A maximum vertical load of 200 N (displacement control at a load rate of 8 mm/min and a preload of 50 N) was then applied to the anterior side of the femoral head to create internal rotation. The slope of the linear elastic portion of the load-deflection graph defined torsional stiffness. All specimens were within the linear elastic region to prevent permanent specimen damage, with an average linearity coefficient of R2 > 0.95. The same procedure was used for constructs having midshaft fractures and bone gaps, except that the maximum load applied for specimens with a bone gap was 100 N. Load levels were chosen to prevent permanent specimen damage.
Axial Load-to-Failure Tests
In the final phase of testing, the load to failure for each femoral construct was determined by applying a vertical force (displacement control at a load rate of 8 mm/min and a preload of 100 N) to generate axial compression of the construct. The load was applied until catastrophic failure of the implant and/or the femur occurred. Catastrophic fracture patterns of the bone and implant were examined and noted. A construct was considered to have undergone "clinical failure" when either 10 mm of vertical deflection was reached (which is a clinically practical limit) or when the first abrupt decrease in applied force was experienced after reaching a peak load (which was deemed to indicate substantial initial structural collapse).
Statistical Analysis
Stiffness data were expressed as a percentage of baseline intact stiffness to account for interspecimen differences in properties and to detect the relative effect of the construct and fracture configuration on stiffness. For analysis of differences between all constructs, one-way analyses of variance were performed on the data with a significance level of p < 0.05. If warranted, post hoc pairwise multiple comparisons were made with use of the Tukey honestly significant difference test when homogeneity of variances between comparison groups was maintained and with use of the Tamhane method in cases when homogeneity was not maintained. For analysis of differences between midshaft fracture and midshaft bone gap stiffnesses for a given construct, Student t tests were performed with an adjusted Bonferroni significance level of p < 0.0125. This adjusted value was calculated by dividing the p value for a 95% confidence interval by the number of constructs compared, i.e., p value (Bonferroni) = p value for 95% confidence interval/number of constructs = 0.05/4 = 0.0125.
Intact Specimens
The baseline stiffness data obtained for the twenty intact femora were an average (and standard deviation) of 1416.2 ± 55.5 N/mm (range, 1322.3 to 1522.7 N/mm) for axial compression, 80.2 ± 6.0 N/mm (range, 71.9 to 95.9 N/mm) for lateral bending, and 181.6 ± 10.8 N/mm (range, 160.5 to 207.6 N/mm) for torsion. The four groups comprising five femora each were statistically equivalent to one another in terms of the baseline stiffness values in axial compression (p = 0.99), lateral bending (p = 0.85), and torsion (p = 0.20).
Axial Stiffness
Normalized axial stiffness for all constructs ranged from 1.11 to 1.36 for the midshaft fracture group and 0.67 to 1.08 for the midshaft bone gap group (Fig. 2). There were no significant differences among constructs after a midshaft fracture was created (p > 0.052). Construct D showed significantly superior axial stiffness compared with all other constructs when a midshaft bone gap was present (p < 0.05), and no difference was detected among constructs A, B, and C (p > 0.119). For intergroup comparisons, axial stiffness was higher for each construct during testing with a midshaft fracture compared with testing with a midshaft bone gap (p < 0.0125).
Lateral Bending Stiffness
Normalized lateral bending stiffness for the constructs assessed in this study ranged from 0.14 to 0.96 for the midshaft fracture group and 0.06 to 0.65 for the midshaft bone gap group (Fig. 3). Construct D showed a significant improvement compared with all other constructs after a midshaft fracture was created (p < 0.05). No difference was detected among constructs A, B, and C with regard to midshaft fracture tests (p > 0.102). Similarly, for the midshaft bone gap group, construct D showed significantly superior lateral bending stiffness compared with all other constructs (p < 0.008), while construct C was stiffer than both constructs A and B (p < 0.001). For intergroup comparison, lateral stiffness was not different when corresponding midshaft fracture and bone gap constructs were compared (p > 0.0125).
Torsional Stiffness
Normalized torsional stiffness for all constructs ranged from 0.60 to 0.91 for the midshaft fracture group and 0.66 to 0.92 for the midshaft bone gap group (Fig. 4). After fracture creation, construct D demonstrated greater stiffness than construct B (p = 0.014), although no other differences were detected. After bone gap creation, construct D showed superior stiffness compared with all other constructs (p < 0.015), whereas no difference was detected among constructs A, B, and C (p > 0.919). However, for intergroup comparison, torsional stiffness was not different when the effects of a midshaft fracture and a bone gap for a given construct type were compared (p > 0.0125).
Axial Load to Failure
Axial load-to-failure levels, based on the criterion defined above, ranged from 5561.5 to 6700.2 N (average and standard deviation, 6074.6 ± 296.1 N) (Fig. 5). No significant differences were detected among any of the constructs tested (p = 0.6).
Axial catastrophic failure modes demonstrated several patterns common to all specimens and all construct groups (Fig. 6). Gradual application of axial load caused bone gap closure and progressive plate bending at the gap. Catastrophic failure would then occur. There was no evidence of permanent plate damage, screw loosening or pullout, cable loosening or damage, or hip stem loosening within the cement mantle.
Failure patterns specific to each construct were also evident (Fig. 6). Construct A demonstrated catastrophic failure at the proximal part of the femur at the tip of the total hip replacement because of an oblique femoral crack originating at the most proximal screw (four of five specimens). Catastrophic failure of the distal aspect of the femoral shaft, characterized either by a transverse femoral break near the support base (three of five specimens) or an oblique break at the most distal screw (one of five specimens), occurred in most specimens.
Construct B demonstrated minimal proximal femoral damage, with either no visible evidence of any damage (three of five specimens) or a superficial stable transverse crack originating at the second most proximal screw (one of five specimens). Only one specimen had an unstable proximal fracture, due to a femoral crack extending from the second most proximal screw to the distal cable. All specimens experienced catastrophic failure of the distal end of the femur, either by an oblique break at the most distal screw (three of five specimens) or by a transverse break near the support base (two of five specimens).
Construct C demonstrated a spectrum of degrees of proximal femoral damage: no visible evidence of any damage (two of five specimens), a superficial stable transverse crack originating at the most proximal screw (one of five specimens), and an unstable proximal fracture due to a crack extending from the second most proximal screw to its neighboring cable (two of five specimens). However, all specimens underwent catastrophic failure of the distal end of the femur, either by an oblique break at the most distal screw (three of five specimens) or by a transverse break near the support base (two of five specimens).
Construct D demonstrated minimal proximal damage, with no visible evidence of any damage (two of five specimens), a superficial stable transverse crack of the femur originating at the most proximal screw accompanied by a small transverse or longitudinal crack in the allograft (two of five specimens), and a superficial oblique femoral crack extending from the second most proximal screw to its neighboring cable (one of five specimens). Every specimen underwent catastrophic failure of the distal end of the femur, either by a transverse break near the support base (three of five specimens) or by an oblique break at the most distal screw (two of five specimens).
A comparison was made between construct stiffnesses obtained from the slope of the linear portion of the force-deflection graph from the final axial load-to-failure tests (i.e., "destructive test stiffness") and the initial axial stiffness tests performed for specimens with a midshaft bone gap (i.e., "nondestructive test stiffness") (Fig. 7). The results revealed that the ratio of nondestructive to destructive test stiffness for construct A was greater than that for construct C (p = 0.02) and for construct D (p = 0.05). All other comparisons showed no differences.
With the advent of locking plate technology, attention has turned to its use in the periprosthetic fracture setting, although its use in total hip arthroplasty is limited23,29. Reported advantages include a toggle-free construct for osteoporotic unicortical fixation, decreased soft-tissue and periosteal stripping with preservation of fracture site vascularity, a decreased need for plate contouring, and the potential for minimally invasive insertion techniques23,35,36.
The current study suggests that locking plate technology may not provide superior fracture fixation stiffness immediately postoperatively, which is what the study models. However, it must be noted that reoperation for failure occurs at an average of twenty-two months after initial fixation of a periprosthetic femoral fracture, making the immediate postoperative stability of constructs of moderate importance clinically37. For most relevant statistical comparisons made in the current study, allograft struts with nonlocking plates (construct D) afforded superior stiffness compared with both locked plates with use of screws alone (construct A) and locked plates with use of a combination of screws and cables proximally (construct B). There was also no significant difference in stiffness during any of the evaluated testing modes among constructs A, B, and C. The one exception was that construct C (an Ogden-type plate) was superior to constructs A and B during lateral bending for the midshaft bone-gap group. This latter result differs from that described by Fulkerson et al.23, who found locked plates to be superior to an Ogden-type construct in axial and torsional testing but not in lateral bending. There may be two methodological reasons for the discrepancy. First, Fulkerson et al. used a previously shown inferior Ogden construct for comparison, which may not be ideally suited for such an assessment. Second, they used cadaveric osteoporotic bone, in which the difference between locked and conventional nonlocking technology would be more pronounced.
The present investigation suggests that locked plates provide no immediate advantage in good bone stock, as modeled by the synthetic femora. Locked plates (construct A) and locked plates with proximal cables (construct B), particularly for the midshaft gap model, demonstrated stiffnesses that were lower than those of the allograft-plate system (construct D) and lower or equivalent to those of a nonlocking plate system. In addition, no difference in axial load to failure was detected among any of the constructs. In osteoporotic bone, however, locked system constructs provide increased stability compared with their nonlocking counterparts23.
Our results showed that the use of an allograft strut combined with a nonlocking plate may provide superior biomechanical stability compared with other methods in some situations. For most relevant statistical comparisons made in this investigation, construct D demonstrated greater stiffness than constructs A, B, and C in axial compression, lateral bending, and torsional testing. However, construct D was not significantly different in axial load to failure compared with constructs A, B, or C (Fig. 5). Moreover, it was unable to retain its axial stiffness better than constructs B and C and was, in fact, outperformed by construct A (Fig. 7). The use of an allograft strut effectively makes construct D a two-plate system compared with the single-plate systems of constructs A, B, and C. Although it may seem initially intuitive that more plates would mean better performance, these results show that this is not always the case and that they will, rather, depend on the specific loading conditions. This agrees with the results of a study of the repair of periprosthetic fractures near the tip of the femoral stem in total hip replacements by Wilson et al.33, who compared six fixation techniques, namely, a single allograft strut applied anteriorly and a lateral plate (with and without proximal screws), a single lateral plate only (an Ogden-type plate constructed with and without proximal screws), and two allograft struts placed anteriorly and laterally (short and long struts). Those investigators found that the combined allograft-plate constructs offered superior resistance to the relative segment rotation and translation at a simulated osteotomy site compared with the other methods.
Our results provide some relevant information regarding surgical procedures. First, locked plates alone (construct A) and cable-plate systems (constructs B and C) yielded similar stiffnesses in several test modalities. Thus, the advantage of the use of locked plates may be that they can offer sufficient stability while avoiding the excessive soft-tissue stripping and strangulation associated with cables. Second, the only fixation system that demonstrated proximal femoral failure was the locked plate (construct A). Thus, the disadvantage of the use of locked plates may be that their potential failure will be difficult to deal with surgically because of the proximity of the plate to the tip of the femoral stem.
Artificial femora have recently been used in several studies dealing with periprosthetic femoral fractures around knee or hip arthroplasty components3,4,38. We and others have shown that the Third Generation Composite Femur used in this study demonstrates good agreement with regard to axial and torsional stiffness compared with normal human cadaveric femora39,40. The material and geometric properties stated by the manufacturer for these synthetic bones may be considered "normal," in that they model good healthy bone stock rather than osteoporotic bone. Therefore, the use of synthetic bones may be more appropriate in modeling the mechanical stiffness of the long bones of the younger and more active patients receiving a total hip arthroplasty who are increasing in number. Moreover, although these synthetic femora demonstrate similar cortical screw shear stress during pullout relative to that of human cadaveric femora41, no study that we know of has conclusively shown that the failure mechanisms between these artificial femora and human femora are equivalent. Consequently, our results for the load-to-failure tests should be considered with circumspection.
There were several limitations to this study. First, the lack of soft tissues allowed for optimal plate, screw, and cable positioning. This also meant that the short and long-term role of the soft tissues in the stability and union of the fractures could not be assessed. Second, the synthetic femora used do not simulate the osteoporotic quality of the bone often encountered in patients with periprosthetic fractures, although they may model good bone stock adequately. The behavior of locked screws, nonlocked screws, and plates may be different in these synthetic femora compared with osteoporotic bone. Third, the effect of cyclic loading was not examined in this study. Fourth, the combination of a locked plate and an allograft strut was not considered. Such a fixation system should be the subject of a future study, given the interest of many surgeons in locked plates and their intuitive appeal in producing biomechanically superior constructs.
In conclusion, this study demonstrated that a combination of a nonlocking plate with an allograft strut had the highest stiffness of the constructs examined for the treatment of periprosthetic fractures around stable femoral components of total hip arthroplasties. This is consistent with the findings of other researchers33. Because the advantage of locked plates may be in their use with osteoporotic bone or when anatomic plate contouring is difficult, their use as a stand-alone treatment may not be as optimal as that of a nonlocking plate combined with an allograft strut in more healthy bone stock. 
Note: Sawbones Worldwide (Vashon, Washington), Stryker (Mahwah, New Jersey), Synthes (Paoli, Pennsylvania), and Zimmer (Warsaw, Indiana) donated the implants, equipment, and supplies used in this study.
Lowrey CE, Nord KD, Wilson MR, Trick LW. Periprosthetic femur fractures: a complication of rigid cylindrical reaming? Orthop Trans.1991;15:342-52.15342
1991
Scott RD, Turner RH, Leitzes SM, Aufranc OE. Femoral fractures in conjunction with total hip replacement. J Bone Joint Surg Am.1975;57:494.57494
1975
[PubMed]
Stevens SS, Irish AJ, Vachtsevanos JG, Csongradi J, Beaupré GS. A biomechanical study of three wiring techniques for cerclage-plating. J Orthop Trauma.1995;9:381-7.9381
1995
[CrossRef]
Dennis MG, Simon JA, Kummer FJ, Koval KJ, Di Cesare PE. Fixation of periprosthetic femoral shaft fractures occurring at the tip of the stem: a biomechanical study of 5 techniques. J Arthroplasty.2000;15:523-8.15523
2000
[CrossRef]
Bethea JS 3rd, DeAndrade JR, Fleming LL, Lindenbaum SD, Welch RB. Proximal femoral fractures following total hip arthroplasty. Clin Orthop Relat Res.1982;170:95-106.17095
1982
Chapman MW. Diaphyseal fractures of the femur. In: Chapman MW, editor. Chapman's Orthopaedic Surgery. 3rd ed. Philadelphia: Lippincott Williams and Wilkins; 2001. p 671-708.
2001
Partridge A. Nylon straps for internal fixation of bone. Lancet.1976;2:1252.21252
1976
Tower SS, Beals RK. Fractures of the femur after hip replacement: the Oregon experience. Orthop Clin North Am.1999;30:235-47.30235
1999
[CrossRef]
Brady OH, Kerry R, Masri BA, Garbuz DS, Duncan CP. The Vancouver classification of periprosthetic fractures of the hip: a rational approach to treatment. Techn Orthop.1999;14:107-14.14107
1999
[CrossRef]
Beals RK, Tower SS. Periprosthetic fractures of the femur. Analysis of 93 fractures. Clin Orthop Relat Res.1996;327:238-46.327238
1996
[CrossRef]
Cooke PH, Newman JH. Fractures of the femur in relation to cemented hip prostheses. J Bone Joint Surg Br.1988;70:386-9.70386
1988
Johansson JE, McBroom R, Barrington TW, Hunter GA. Fracture of the ipsilateral femur in patients with total hip replacement. J Bone Joint Surg Am.1981;63:1435-42.631435
1981
Zenni EJ Jr, Pomeroy DL, Caudle RJ. Ogden plate and other fixations for fractures complicating femoral endoprostheses. Clin Orthop Relat Res.1988;231:83-90.23183
1988
Berman AT, Zamarin R. The use of Dall-Miles cables in total hip arthroplasty. Orthopedics.1993;16:833-5.16833
1993
Brady OH, Garbuz DS, Masri BA, Duncan CP. Classification of the hip. Orthop Clin North Am.1999;30:215-20.30215
1999
[CrossRef]
Brady OH, Garbuz DS, Masri BA, Duncan CP. The reliability and validity of the Vancouver classification of femoral fractures after hip replacement. J Arthroplasty.2000;15:59-62.1559
2000
[CrossRef]
Duncan CP, Masri BA. Fractures of the femur after hip replacement. Instr Course Lect.1995;44:293-304.44293
1995
Serocki JH, Chandler RW, Dorr LD. Treatment of fractures about hip prostheses with compression plating. J Arthroplasty.1992;7:129-35.7129
1992
[CrossRef]
Tsiridis E, Haddad FS, Gie GA. The management of periprosthetic femoral fractures around hip replacements. Injury.2003;34:95-105.3495
2003
[CrossRef]
Wilson D, Masri BA, Duncan CP. Periprosthetic fractures: an operative algorithm. Orthopedics.2001;24:869-70.24869
2001
Ogden WS, Rendall J. Fractures beneath hip prostheses: a special indication for Parham bands and plating. Orthop Trans.1978;2:70.270
1978
Dennis MG, Simon JA, Kummer FJ, Koval KJ, Di Cesare PE. Fixation of periprosthetic femoral shaft fractures: a biomechanical comparison of two techniques. J Orthop Trauma.2001;15:177-80.15177
2001
[CrossRef]
Fulkerson E, Koval K, Preston CF, Iesaka K, Kummer FJ, Egol KA. Fixation of periprosthetic femoral shaft fractures associated ith cemented femoral stems: a biomechanical comparison of locked plating and conventional cable plates. J Orthop Trauma.2006;20:89-93.2089
2006
[CrossRef]
Kelley SS. Periprosthetic femoral fractures. J Am Acad Orthop Surg.1994;2:164-72.2164
1994
Schmotzer H, Tchejeyan GH, Dall DM. Surgical management of intra- and postoperative fractures of the femur about the tip of the stem in total hip arthroplasty. J Arthroplasty.1996;11:709-17.11709
1996
[CrossRef]
Chandler HP, King D, Limbird R, Hedley A, McCarthy J, Penenberg B, Danylchuk K. The use of cortical allograft struts for fixation of fractures associated with well-fixed total joint prostheses. Semin Arthroplasty.1993;4:99-107.499
1993
Emerson RH Jr, Malinin TI, Cuellar AD, Head WC, Peters PC. Cortical strut allografts in the reconstruction of the femur in revision total hip arthroplasty. A basic science and clinical study. Clin Orthop Relat Res.1992;285:35-44.28535
1992
Haddad FS, Duncan CP, Berry DJ, Lewallen DG, Gross AE, Chandler HP. Periprosthetic femoral fractures around well-fixed implants: use of cortical onlay allografts with or without a plate. J Bone Joint Surg Am.2002;84:945-50.84945
2002
Jimenez ML, Sferra TA, Twigg SL. Morbidity and mortality associated with periprosthetic fractures of the femur. Read at the Annual Meeting of the Orthopaedic Trauma Association; 2004 Oct 8-10; Hollywood, FL. Paper no 63.
2004
Tadross TS, Nanu AM, Buchanan MJ, Checketts RG. Dall-Miles plating for periprosthetic B1 fractures of the femur. J Arthroplasty.2000;15:47-51.1547
2000
[CrossRef]
Tsiridis E, Haddad FS, Gie GA. Dall-Miles plates for periprosthetic femoral fractures. A critical review of 16 cases. Injury.2003;34:107-10.34107
2003
[CrossRef]
Peters CL, Bachus KN, Davitt JS. Fixation of periprosthetic femur fractures: a biomechanical analysis comparing cortical strut allograft plates and conventional metal plates. Orthopedics.2003;26:695-9.26695
2003
Wilson D, Frei H, Masri BA, Oxland TR, Duncan CP. A biomechanical study comparing cortical onlay allograft struts and plates in the treatment of periprosthetic femoral fractures. Clin Biomech (Bristol, Avon).2005;20:70-6.2070
2005
[CrossRef]
Cristofolini L, Viceconti M, Cappello A, Toni A. Mechanical validation of whole bone composite femur models. J Biomech.1996;29:525-35.29525
1996
[CrossRef]
Frigg R. Locking Compression Plate (LCP). An osteosynthesis plate based on the Dynamic Compression Plate and the Point Contact Fixator (PC-Fix). Injury.2001;32 Suppl 2:63-6.3263
2001
[CrossRef]
Haidukewych GJ. Innovations in locking plate technology. J Am Acad Orthop Surg.2004;12:205-12.12205
2004
Lindahl H, Malchau H, Odén A, Garellick G. Risk factors for failure after treatment of a periprosthetic fracture of the femur. J Bone Joint Surg Br.2006;88:26-30.8826
2006
[CrossRef]
Cusick RP, Lucas GL, McQueen DA, Graber CD. Construct stiffness of different fixation methods for supracondylar femoral fractures above total knee prostheses. Am J Orthop.2000;29:695-9.29695
2000
Heiner AD, Brown TD. Structural properties of an improved redesign of composite replicate femurs and tibias. In: Transactions of the 29th Annual Meeting of the Society for Biomaterials; 2003 Apr 30-May 3. Mt. Laurel, NJ: Society for Biomaterials; 2003. Vol. 26. p 702.
2003
Papini M, Zdero R, Schemitsch EH, Zalzal P. The biomechanics of human femurs in axial and torsional loading: comparison of finite element analysis, human cadaveric femurs, and synthetic femurs. J Biomech Eng.2007;129:12-9.12912
2007
[CrossRef]
Zdero R, Rose S, Schemitsch EH, Papini M. Cortical screw pullout strength and effective shear stress in synthetic third generation composite femurs. J Biomech Eng.2007;129:289-93.129289
2007
[CrossRef]