Anterior thoracolumbar instrumentation: stiffness and load sharing characteristics of plate and rod systems

STUDY DESIGN

An in vitro biomechanical study using a thoracolumbar corpectomy model to compare load sharing capabilities and stiffnesses of six different anterior instrumentation systems (three rod styles and three plate styles) for stabilizing the thoracic and lumbar spine.

OBJECTIVES

To evaluate the axial load sharing capabilities of the instrumentation in a thoracolumbar corpectomy model and to measure the bending stiffness of the anterior instrumentation systems for the axes of flexion-extension, lateral bending, and axial rotation with and without an anterior column graft in place.

SUMMARY OF BACKGROUND DATA

Prior publications have analyzed biomechanical characteristics of many spinal instrumentation systems. These reports have compared anterior instrumentation systems with posterior instrumentation systems, in situ fusion techniques, intervertebral spacers, structural allograft and instrumentation, and combined anterior and posterior instrumentation. Other reports have published data on the biomechanical characteristics of typical anterior and posterior spinal instrumentation systems. However, there are no published reports that specifically compare the characteristics of anterior plate-style with anterior rod-style systems, or examining load sharing capabilities.

METHODS

Six constructs of each of six instrumentation systems were mounted on simulated vertebral bodies. A custom four-axis spine simulator was used to apply independent flexion-extension, lateral bending, and axial rotation moments as well as axial compressive loads. Axial load sharing was measured through a range of applied axial loads from 50 N to 500 N with rotational moments maintained at 0 Nm. The bending stiffness of each construct was calculated in response to +/-5.0 Nm moments about each axis of rotation with a 50 N compressive axial load with a full-length corpectomy graft in place, simulating reconstruction of the anterior column, and with no graft in place, simulating catastrophic graft failure. Statistical significance was determined using an analysis of variance and Fisher PLSD post hoc test with an alpha <or= 0.05.

RESULTS

Load sharing results ranged from 63% to 89%. There was an inverse relationship between load sharing and stiffness. No correlation was found between load sharing and implant style (rod vs. plate). With the graft in place, stiffness result varied by instrumentation system rather than by plate/rod style. Without the graft, the stiffness of the constructs decreased approximately one-third in flexion-extension, two-thirds in lateral bending, and one-fifth in axial rotation, underlying the importance of the graft in overall construct stiffness.

CONCLUSIONS

For both load sharing and stiffness, there is more influence from the design of the instrumentation system, than whether it is a plate or rod style system. The graft contributed to overall construct stiffness, particularly in lateral bending.

Fixation of periprosthetic femur fractures: a biomechanical analysis comparing cortical strut allograft plates and conventional metal plates

This study compared the stability of periprosthetic femur fractures fixed using cortical allograft struts with a metal plate. Cadaveric specimens were loaded in single-leg stance and stair climbiing to 2250 N. Optimum stability in single-leg stance was achieved with two long struts medially and laterally. No clear advantage was noted in using a second strut in stair climbing. Cables rather than wires were useful in single-leg stance, but not in stair climbing. Allograft cortical struts are a biomechanically sound alternative to metal plates fixed with screws and cables for femur fracture fixation below a well-fixed femoral component.

Are uniform regional safety factors an objective of adaptive modeling/remodeling in cortical bone?

It has been hypothesized that a major objective of morphological adaptation in limb-bone diaphyses is the achievement of uniform regional safety factors between discrete cortical locations (e.g. between cranial and caudal cortices at mid-diaphysis). This hypothesis has been tested, and appears to be supported in the diaphyses of ovine and equine radii. The present study more rigorously examined this question using the equine third metacarpal (MC3), which has had functionally generated intracortical strains estimated by a sophisticated finite element model. Mechanical properties of multiple mid-diaphyseal specimens were evaluated in both tension and compression, allowing for testing of habitually tensed or compressed regions in their respective habitual loading mode ("strain-mode-specific" loading). Elastic modulus, and yield and ultimate strength and strain, were correlated with in vivo strain data from a previously published finite element model. Mechanical tests revealed minor variations in elastic modulus, and yield and ultimate strength in both tension and compression loading, while physiological strains varied significantly between the cortices. Contrary to the hypothesis of uniform safety factors, the MC3 has a broad range of tension (caudo-medial, 4.0; cranio-lateral, 37.7) and compression (caudo-medial, 5.7; cranio-lateral, 68.9) safety factors.

Rotational stability of a modified step-cut for use in intercalary allografts

BACKGROUND

Intercalary allografts are used for the reconstruction of major skeletal defects. Step-cuts help to provide rotational stability when intramedullary fixation is used. A modified step-cut is proposed to reduce rotation at the interface. This study compares the rotational stability of conventional and modified step-cuts.

METHODS

In Phase I, seven pairs of human cadaveric femora were divided into a conventional step-cut group (left femora) and a modified step-cut group (right femora). All femora were cut transversely at the mid-diaphysis. In the conventional group, a 1-cm step-cut was created in the exact midsagittal plane in both the proximal and distal segments. In the modified group, a 1-cm step-cut was created in the parasagittal plane, leaving 2 mm of additional bone on both the proximal and the distal fragment. Phase II was identical except that in the modified step-cut group only 1 mm of additional bone was left. Smooth femoral nails were then placed after standard reaming. Specimens were tested by fixing the proximal segment and applying +/-2 N-m (17.7 in-lb) of torque to the distal segments with ten oscillation cycles. Maximum rotation was measured. The data were analyzed with the paired Student t test.

RESULTS

The average rotation in Phase I was 23.3 degrees for the conventional step-cut group and 3.0 degrees for the 2-mm modified step-cut group; the difference was significant (p < 0.001). Four femora sustained an incomplete fracture during nail insertion. The average rotation in Phase II was 20.6 degrees for the conventional step-cut group and 0.5 degrees for the 1-mm modified step-cut group without any fractures; the difference was significant (p < 0.001).

CONCLUSIONS

Step-cut modification that leaves more bone in the sagittal plane provides rigid fixation and significantly more stability than the conventional step-cut technique.

Cyclic loading of olecranon fracture fixation constructs

BACKGROUND

Despite the good results that are usually reported after fixation at the sites of olecranon fractures and osteotomies, problems such as loss of fixation, nonunion, and the need for revision surgery are still encountered. Various types of fixation have been recommended, but few have been evaluated with use of clinically relevant cyclic load testing at appropriate levels of stress. The purpose of the present study was to test multiple olecranon fixation techniques under physiologic cyclic loads.

METHODS

We studied ten cadaveric elbows with use of cyclic loading that simulated (1) active range of motion and (2) pushing up from a chair. Each specimen underwent fixation of a simulated 50% transverse olecranon fracture with use of intramedullary and cortically fixed tension band constructs (in randomized order) followed by fixation with a 7.3-mm-diameter cancellous screw with and without a tension band. Displacement transducers were placed posteriorly on the tension side and anteriorly near the articular surface.

RESULTS

Both configurations involving the 7.3-mm-diameter cancellous screw provided the most stable fixation-nearly five times better than that provided by the Kirschner-wire techniques. Use of the tension band in conjunction with the intramedullary screw improved the stability of fixation. In none of the constructs did the AO tension band result in compression across the osteotomy gap.

CONCLUSIONS AND CLINICAL RELEVANCE

The use of a 7.3-mm screw in conjunction with a tension band provided better fixation of simulated displaced transverse fractures than did the use of Kirschner wires in conjunction with a tension band or the use of a screw only. The AO principle of converting posterior tensile forces to articular compressive forces was not demonstrated in this study. We therefore question the validity of the tension band concept in olecranon fracture fixation and recommend passive rather than active range of motion in the immediate postoperative period to limit fracture distraction.

Tibial component fixation with cement: full- versus surface-cementation techniques

Despite excellent outcomes with cemented tibial components in total knee arthroplasty, it still is debated whether the tibial stem should be cemented and what the optimal tibial stem design should be. Proponents of full cementation of the tibial stem and component state that better short-term and long-term component fixation is achieved when full cementation is used. Advocates for surface cementation contend that sufficient implant stability is achieved without the increased bone loss that occurs at revision and the stress shielding thought to be linked with cemented stems. This biomechanical cadaver study compared initial fixation and cement penetration depth in fully cemented versus surface cemented tibial trays with two different stem geometries (cruciate and I-beam) and compared two stem designs (cruciate and I-beam) fixed with surface cementation. Under an eccentric load, simulating three times body weight for 6000 cycles, there seems to be no difference in the micromotion of either tibial component implanted with surface or full cementation. Additionally, no difference in the average depth of cement penetration was detected between fixation techniques or stem types. The initial fixation stability of the surface cement technique seems correlated to the depth of cement penetration into proximal tibial surface. The current data support other studies which indicate that stability of surface-cemented tibial components may be related to the depth of cement penetration.

Glenohumeral articular contact areas and pressures following labral and osseous injury to the anteroinferior quadrant of the glenoid

The objective of this study was to determine the effect of progressive labral and bone loss on the articular contact area and pressures across the glenohumeral joint under compressive loads of 220 and 440 N. Eight fresh-frozen cadaver shoulders were used, and contact pressures in 4 quadrants of the glenoid were determined with a Tekscan flexible tactile force sensor. Testing conditions included intact glenoids, glenoids with the anteroinferior labrum removed, and glenoids with 3 sizes of bone defects in the anteroinferior quadrant. By means of Tekscan sensing equipment, the measured contact area over the glenolabral complex was between 49.0% and 61.5% of the calculated surface area for the intact specimens. Loss of the anteroinferior labrum decreased contact area by 7% to 15% compared with the intact specimens, and the mean contact pressure increased by 8% to 20%. With bone loss corresponding to a defect measuring 30% of the diameter in the anteroinferior quadrant, contact area across the entire glenoid decreased a mean of 41% compared with the intact specimens, whereas the mean contact pressure increased nearly 100%. When the anteroinferior quadrant of the glenoid was analyzed separately, loss of the anteroinferior labrum alone resulted in an increase in the mean contact pressure in this quadrant compared with the intact specimens (mean, 53%). Bone loss of 30% of the diameter resulted in mean contact pressures in this quadrant increasing by 300% to 400% compared with the intact specimens, with 2 of 8 specimens becoming grossly unstable. In addition, with 30% diameter bone loss, the mean contact pressure decreased by 26% in the posterosuperior quadrant, indicating a shift in loading of the cadaveric glenoid. Peak pressures followed similar trends, with labral loss alone increasing peak pressures in the anteroinferior quadrant by a mean of 28% of that seen for the intact specimens.

Durability and strength of Steinmann pin augmentation in cemented tibial defects

It has been argued that Steinmann pin augmentation does not improve the biomechanics of polymethylmethacrylate reconstruction for massive defects of bone. The current authors investigated whether pin augmentation of polymethylmethacrylate in the reconstruction of noncontained defects of bone improved the biomechanical properties of the reconstruction as compared with polymethylmethacrylate alone when minimal or large bone porosity is present. Large noncontained defects were created in 10 pairs of human tibias. In Group 1, five left tibias had reconstruction with polymethylmethacrylate augmented with three, 5-mm diameter by 10-mm deep holes into lateral condyle cancellous bone. Right tibias had identical reconstruction with three, 3/16-inch threaded pins placed into the medullary canal. In Group 2, three, 10-mm diameter by 10-mm deep holes were created in both pairs. The left tibia had polymethylmethacrylate reconstruction and the right tibia had polymethylmethacrylate and pin augmentation. Specimens were subjected to 2000 compressive cycles then loaded to failure. In Group 1, cycles and load to failure were significantly lower in reconstructions without pins compared with reconstructions with pins. No significant difference was observed between reconstruction techniques in Group 2. In reconstructions without pins, large diameter holes had significantly better cyclical durability. Pins improved survival compared with no pins.

The effects of medial and lateral displacement calcaneal osteotomies on ankle and subtalar joint pressure distribution

We compared the pressure distribution in the ankle and posterior facet of the subtalar joint following 1 cm medial and lateral displacement calcaneal osteotomies to the pressure distribution in the intact foot. Six cadaver specimens were loaded in neutral alignment while pressure measurements were recorded. A 1-cm medial displacement osteotomy shifted the average center of force in the ankle 1.0 mm medially (p = 0.36) while a lateral displacement osteotomy shifted the center of force 1.1 mm laterally (p = 0.42). There was also a slight shift in the percentage of pressure toward the side of the talus to which the calcaneus was shifted. For the lateral displacement osteotomy, the pressure increased 4.0% in the lateral-most quadrant (p = 0.05), while the medial osteotomy increased the pressure 1.3% in the medial quadrant (p = 0.30). In the subtalar joint, a medial displacement osteotomy shifted the pressure distribution slightly medially (5.9%, p = 0.06) and more anteriorly (9.6%, p = 0.02) while the distribution was shifted laterally (5.9%, p = 0.17) and anteriorly (5.6%, p = 0.03) with a lateral displacement osteotomy. These shifts of percentage of pressure between quadrants of the joints were slight-less than 5% in the ankle and less than 10% in the subtalar joint. Significant translation of the calcaneal tuberosity appears to have only a small effect on pressure distribution in the ankle and posterior facet of the subtalar joint in a weighted cadaver model.

Primary total knee arthroplasty in the valgus knee: creating a balanced soft tissue envelope

Lateral tissue releases in valgus total knee arthroplasty frequently produce asymmetric flexion-extension gaps and ligamentous instability. This study compared 2 lateral-release sequences and quantified the effects of sequential lateral capsular ligamentous structure release. One knee from 7 paired specimens was released according to a 4-step sequence: posterior cruciate ligament (PCL), ibiotibial tract (IT band), popliteus tendon/lateral collateral ligament (PT/LCL), and biceps femoris tendon. The contralateral knees were released according to a 5-step sequence: PCL, posterolateral capsule, IT band, PT, and LCL. After each release step, flexion and extension gaps were measured and recorded for the medial and lateral aspects. The 5-step sequence produced more symmetric flexion-extension gaps, whereas the absolute magnitudes of correction were lower than with the 4-step sequence. LCL sacrifice in both sequences produced marked lateral flexion-extension gap asymmetry.

Segmental pedicle screw fixation or cross-links in multilevel lumbar constructs. a biomechanical analysis

BACKGROUND CONTEXT

The placement of segmental pedicle screws and cross-links in short segment posterior pedicle screw constructs has been shown to increase the construct stiffness in some planes. To date, no studies have looked at the contributions of segmental pedicle screw and cross-link placement in longer constructs.

PURPOSE

To evaluate the influence of segmental pedicle screw and/or cross-link placement on flexion/extension, lateral bending and axial torsion stiffness in two- and three-level posterior pedicle screw fixation constructs.

STUDY DESIGN/SETTING

An in vitro biomechanical analysis of two- and three-level posterior pedicle screw constructs with and without segmental fixation and/or cross-links was performed using calf lumbar spines. Stiffness of the constructs was compared.

METHODS

Six calf lumbar specimens were used to test stiffness in one-, two- and three-level posterior pedicle screw fixation constructs in 12 configurations. A custom-made, four-axis spine simulator applied pure cyclical (+/-5 Nm) flexion/extension, lateral bending and axial torsion moments at 0.1 Hz under a constant 50-N axial compressive load. The stiffness of each construct was calculated about each axis of rotation. Data were analyzed using nonparametric techniques with statistical significance determined at alpha less than .05.

RESULTS

The stiffness of the instrumented spines were significantly greater than the noninstrumented intact spines in all loading conditions for one-, two- and three-level constructs. There were no significant changes in flexion/extension stiffness with the addition of either the cross-links or the segmental pedicle screws. In lateral bending, the addition of segmental pedicle screws significantly increased the stiffness in the two- and three-level constructs. The addition of two cross-links increased lateral bending stiffness in the longer three-level constructs, with little change in the two-level constructs. In axial torsion, the progressive addition of cross-links showed a tendency toward increased stiffness in both the two- and three-level constructs. Segmental pedicle screws further increased torsional stiffness of the longer, three-level constructs.

CONCLUSIONS

As the use of segmental spinal instrumentation progresses from one to two and three levels, the contribution of cross-links and segmental pedicle screws to the overall construct stiffness increases.

Spaceflight alters bone mechanics and modeling drifts in growing rats

BACKGROUND

Alterations in bone metabolism may be a particularly serious consequence of spaceflight and a major obstacle to long-term space exploration. The effects of spaceflight on bone mechanics are unclear. This study examined the effects of spaceflight on bone mechanics in a growing rat model during a 17-d mission aboard the space shuttle (STS-78).

METHODS

There were 18 rats that were divided into 3 experimental groups: flight rats (n = 6), ground-based control rats housed in an animal enclosure module (AEM, n = 6), and ground-based control rats housed in standard vivarium caging (n = 6). At the conclusion of the mission, rat femurs were tested in three-point bending followed by static and dynamic bone histomorphometry.

RESULTS

Maximum stress was unaffected by spaceflight, but flexural rigidity was significantly decreased in flight animals. Much of the decrease appeared to be the result of decreases in tissue properties (elastic modulus) rather than structural changes within the bone. No significant differences in cortical bone mass or geometry were observed. In contrast, endocortical resorption was significantly decreased in flight rats accompanied by a nonsignificant decrease in periosteal bone formation, suggesting alterations in bone modeling drifts during spaceflight. For nearly all measured indices, ground-based AEM rats displayed values intermediate to flight and ground-based vivarium rats.

CONCLUSIONS

Spaceflight can impair tissue properties in femoral cortical bone during growth without significant decreases in bone mass or geometry.

The effect of femoral prosthesis design on cement strain in cemented total hip arthroplasty

Finite element studies show that the highest cement stresses are located at the most proximal and distal ends of the prosthesis. In vitro biomechanical and histologic analyses of autopsy-retrieved cemented femoral components show these areas to be associated with cement-prosthesis debonding. In this study, cement strains were measured in 2 geometrically different femoral stems in paired cadaver femora: A straight, collared, moderately tapered stem (Centralign) was compared with an anatomically curved, collarless, dramatically tapered stem (Scientific Hip Prosthesis [SHP]). Results showed that the maximum strain and the overall strain profile differed between the 2 stems. The Centralign had peak strains located at the most proximal gauge positions, whereas the peak strains of the SHP were located around the middle of the femoral stem. Minimization of cement strain, especially at the crucial proximal and distal areas of the stem, by altering component design may be able to reduce cement-prosthesis debonding and improve clinical results.

The effects of drilling force on cortical temperatures and their duration: an in vitro study

Bone loss due to thermonecrosis may weaken the purchase of surgically placed screws and pins, causing them to loosen post-operatively. The goal of this study was to determine how differences in applied drilling forces affect the temperature of cortical tissue near the drilling site. Results from thermocouples placed into fresh cortical bone indicate that increasing the applied drilling force resulted in a significant decrease (P=0.001) of maximum cortical temperatures. Furthermore, increasing the drilling force resulted in a significant decrease (P=0.001) in the average duration of temperature elevations above 50 degrees C. The results of the current study demonstrate that by the application of a larger force to the drill, both maximum cortical temperatures and their duration above 50 degrees C may be effectively reduced, decreasing the potential for thermal necrosis in the neighboring cortical bone.

External fixation of the distal radius: to predrill or not to predrill

Using both clinical and laboratory studies we investigated whether predrilling before insertion of external fixation pins is necessary for use in treating distal radius fractures. Our clinical study included 50 consecutive external fixators (4.0- and 2.5-mm pins) using 100 predrilled and 100 direct-drilled pins placed in a randomized manner. There was no increased incidence of pin track infection or other pin problem with the direct-drilled technique. There were, however, significantly elevated temperatures with the direct-drilled technique. We therefore recommend predrilling even though the temperature differences in this bone with this fixator were not clinically evident.

Calcaneocuboid joint pressure after lateral column lengthening in a cadaveric planovalgus deformity model

The purpose of this study is twofold: first, to measure the joint contact pressure across the calcaneocuboid joint in a planovalgus deformity and compare the results to pressures measured in a normal foot; and second, to determine the change in pressure across the calcaneocuboid joint after an Evan's-type calcaneal lengthening osteotomy. The effect of this procedure on the calcaneocuboid joint was evaluated using seven cadaver feet to measure peak pressure across the calcaneocuboid joint under a constant load. Each foot was sectioned medially to reproduce a deformity consistent with an adult, acquired flatfoot. Each flatfoot deformity was then corrected using a ten-millimeter lateral column lengthening osteotomy. Joint pressures were measured in the normal foot, the created flatfoot and then in the corrected flatfoot. Peak pressures across the joint increased significantly from baseline in the flatfoot (p <0.05). However, the change in pressure from the flatfoot to the corrected foot was not significant, and in some cases peak pressures in the corrected foot were actually lower than in the flatfoot. These findings indicate that calcaneal lengthening through an Evan's osteotomy does not increase pressure across the calcaneocuboid joint beyond physiologic loads in the flatfoot.

Thermal effects of the electron beam and implications of surface damage in the analysis of bone tissue

Electron beam interactions with specimens in the scanning electron microscope (SEM) can lead to increased surface temperatures and damage. These changes may have significant consequences in the analysis of bone tissue. An investigation was performed to measure the surface temperature changes associated with the electron beam on a thermocouple with systematic variations in operating conditions. Probe currents, magnifications, and accelerating voltages were incrementally adjusted to measure the temperature changes and to make assessments for determining optimal operating conditions for the SEM in future analyses of bone tissue. Results from this study suggest that thermal effects were minimal at lower accelerating voltages (< 20 kV), lower probe currents (< 10 nA), and lower magnifications, but surface damage may still occur during the analysis of bone tissue.

Technique for identification of submicron metal particulate from implants in histological specimens

Metal implants are being used with increasing frequency for the treatment of many diseases in the field of orthopedics, cardiology, cardiovascular surgery, and otolaryngology. Unfortunately, metals can be a source of submicron particles, which may have adverse effects on tissues. This article describes a technique that uses backscattered electron imaging and energy dispersive X-ray microanalysis, which have the capacity to perform both quantitative and qualitative analysis. The particles can be characterized by size, shape, amount, and composition. Although this technique can be used near the implant interface, it is particularly helpful in tissues a great distance from the implant site with a low concentration of metal debris. In addition, the sensitivity and specificity of this technique can be adjusted to the investigator's needs.

The effect of reconstruction of the medial patellofemoral ligament on patellar tracking

We evaluated patellar tracking in six cadaveric knees with the medial restraints intact and then sectioned to determine their contribution to lateral translation of the patella with and without a lateral force on the patella. The medial patellofemoral ligament was then reconstructed with a gracilis tendon graft and patellar tracking was again evaluated. The knees were extended using a materials testing machine, and patellar tracking was measured with a position sensing system. With no lateral force applied to the patella, patellar tracking was unaffected by the presence or absence of the medial restraints or by reconstruction of the medial patellofemoral ligament. With a lateral force applied to the patella, patellar tracking was changed significantly by the loss of the medial restraints. Normal patellar tracking was substantially restored by reconstruction of the medial patellofemoral ligament.

Loading conditions and cortical bone construction of an artiodactyl calcaneus

Customary nonuniform distributions of physiological bone strains are thought to evoke heterogeneous material adaptation in diaphyseal cortices of some limb bones. Recent studies of artiodactyl calcanei have suggested that the regional prevalence of specific mechanical strain features such as mode and magnitude correlate with specific variations in cortical bone ultrastructure, microstructure and mineralization. These data are also consistent with predictions of current algorithms of mechanically induced bone adaptation. However, detailed characterization of the customary functional strain environment of these bones is needed to understand better the mechanisms of these adaptations. An in vitro loading method and rosette strain gauges were used to record principal strains, maximum shear strains and principal strain angles at multiple locations on ten calcanei of adult male mule deer (Odocoileus hemionus hemionus). Each hind limb was fixed in an apparatus to mimic the mid-support phase of the gait and loaded via the Achilles tendon over a broad range of functional loads (0 to 2943 N). Strains were recorded on the craniolateral, craniomedial, caudal, medial and lateral cortices at mid-diaphysis. Loading variations included the progressive elimination of the ligament and tendon along the caudal calcaneus. The results showed that the cranial cortex experiences longitudinal compressive strains that are nearly equal to the principal minimum strains and that the caudal cortex receives longitudinal tensile strains that are nearly equal to the principal maximum strains. With a 981 N load, the mean principal compressive strain on the cranial cortex was −636+/−344 micro(ε) (mean +/− s.d., N=9) and the mean principal tensile strain on the caudal cortex was 1112+/−68 micro;(ε)x (N=9). In contrast to the cranial and caudal cortices, principal strains in the medial and lateral cortices displayed relatively large deviations from the longitudinal axis (medial, 24 degrees cranial; lateral, 27 degrees caudal). Although shear strains predominated at all gauge sites, variations in maximum shear strains showed no apparent regional pattern or consistent regional predominance. The plantar ligament and tendon of the superficial digital flexor muscle were shown to have important load-sharing functions. These results demonstrate that the functionally loaded artiodactyl calcaneus generally behaves like a cantilevered beam with longitudinal compression and tension strains predominating in opposing cranial and caudal cortices, respectively. Differences in osteon remodeling rates, osteon morphology and mineral content reported previously between the cranial and caudal cortices correlate, in part, with the magnitudes of the principal compressive and tensile strains, respectively. However, material differences that distinguish the medial and lateral cortices from the cranial and caudal cortices could not be primarily attributed to locally increased shear strains as previously suggested. Variations in osteon and/or collagen fiber orientation may correlate more strongly with principal strain direction.

A biomechanical analysis of internal fixation of complex tibial plateau fractures

OBJECTIVE

To compare the mechanical stability of fixation of an unstable bicondylar tibial plateau fracture with several different fixation techniques in a cadaveric model.

DESIGN

Randomized laboratory investigation using a simulated bicondylar tibial plateau fracture with metaphyseal-diaphyseal dissociation.

SETTING

Complex tibial plateau fractures were instrumented and tested under ramp and cyclic loading conditions on a servohydraulic materials testing machine.

INTERVENTION

Each tibia was instrumented sequentially with a lateral buttress plate, a lateral and a medial buttress plate, and a lateral buttress and an anteromedial antiglide plate for ramp load testing. For cyclic testing, one of the three constructs was used on each specimen.

MAIN OUTCOME MEASUREMENTS

Vertical subsidence of the medial tibial plateau was measured in both ramp and cyclic loading in order to evaluate the three internal fixation techniques.

RESULTS

No significant difference was measurable between the dual buttress construct and the lateral buttress/anteromedial antiglide construct. However, the lateral buttress plate alone provided significantly less stability.

CONCLUSIONS

A lateral buttress plate with an anteromedial antiglide plate may provide equally effective fixation as compared with the dual buttress plating technique in complex tibial plateau fractures. This less invasive technique may also be associated with fewer complications due to the lack of soft tissue stripping that is required for its application.

Comparative micromotion of fully and proximally cemented femoral stems

This investigation studied the differences of in vitro micromotion between two stem designs. The two stem types investigated were a proximally cemented stem with distal press fit and a fully cemented stem. After initial micromotion testing to 2250 N in simulated single leg stance and stair climb, six of each stem type were loaded dynamically for 1 million cycles at 950 N at 1 Hz. Micromotion studies were repeated. The two stem types had similar micromotion. For the single leg stance, fully cemented implant motion averaged (+/- 95% confidence) 18 +/- 8 microns toggle, 41 +/- 5 microns axial, and 59 +/- 22 microns rotation. Proximally cemented implant motion averaged 20 +/- 6 microns toggle, 42 +/- 6 microns axial, and 31 +/- 15 microns rotation. For the simulated stair climb, fully cemented implant motion averaged 24 +/- 10 microns toggle, 45 +/- 8 microns axial, and 92 +/- 32 microns rotation. Proximally cemented implant motion averaged 19 +/- 10 microns toggle, 42 +/- 9 microns axial, and 87 +/- 53 microns rotation. For both loading conditions, there were no significant differences measurable between the two systems. After dynamic testing of the fully cemented implants, there were no significant changes in the micromotion of either the toggle or the rotation, but an average of 18 microns increase of axial motion was measured in the fully cemented stem. For the proximally cemented implants, there were no significant changes after dynamic testing. This differences was not considered clinically significant because roentgen stereophotogrammetric analysis studies have shown that more than 4 mm of migration is required before clinical symptoms manifest. The protocol developed in this study may help provide a screening process to determine the stability of femoral stem designs before these devices are used clinically.

Biomechanical evaluation of lateral patellar dislocations

This investigation was undertaken to identify the structures torn within the medial retinaculum and localize the injury site anatomically following acute lateral dislocation of the patella in a cadaver model. The patellae of 10 fresh-frozen cadavers were translated laterally 135% of the patella width on a universal testing instrument. Magnetic resonance imaging (MRI) was performed on all specimens prior to testing and immediately following testing. Anatomical dissection also was performed on the medial retinaculum following testing. Dissection revealed avulsion fractures from the inferomedial border of the patella in 8 of the 10 knees. The medial patellofemoral ligament was injured in 8 of the 10 knees; the location of the injury varied. Tears of the medial patellofemoral ligament from the femur in 6, a midsubstance tear in 1, and stretch in 1 knee were noted. In a knee with a femoral-sided tear, an avulsion fracture of the medial patellofemoral ligament was identified. None of the cadaver knees demonstrated tears of the lateral retinaculum or medial patellotibial ligaments on dissection. Review of the MRIs revealed a medial retinaculum tear in 6 of the 10 knees. Two tears from the femur, 3 from the patella, and 1 tear from both the patella and femur were noted. An avulsion fracture was noted from the inferomedial patellar border in 3 of the 10 knees. No pathology was noted on 4 of the MRIs. When anatomically correlated, the 3 patellar retinacular tears and 3 avulsion fractures noted on MRI represented a tear of the medial patellomeniscal ligament from the patella. The femoral-sided tear represented a tear of the medial patellofemoral ligament from the femur. An appreciation of the spectrum of injury to the medial retinaculum may aid in the diagnosis of an acute dislocation of the patella and help establish the anatomical structures damaged. The pathology demonstrated in this study may explain the diversity of injury seen clinically. Whereas an avulsion fracture from the patella may represent the medial patellomeniscal ligament, a femoral-sided retinacular tear may represent the medial patellofemoral ligament. This may lead to future refinements of surgical options and anatomic restoration of the damaged structure.

Dissolution of particulate hydroxyapatite in a macrophage organelle model

It is controversial as to whether debris from hydroxyapatite (HA)-coated implants jeopardizes the long-term success of total joint replacements. It has been hypothesized that liberated HA particles are engulfed by macrophages and through normal cellular digestion prevent osteolysis and third-body wear. HA particulates, however, have been observed at the interface and on polyethylene articulating surfaces. There is limited data demonstrating the ability of HA to dissolve at the acidity levels associated with macrophage organelle digestion. The objective of this study was to determine if particulate HA could dissolve at the pH levels found in macrophage organelles. Characterized HA particles were placed into buffered solutions corresponding to phagosomal organelle pH levels: cytoplasmic (pH 7), phagosomal (pH 6), and lysosomal (pH 5). Flasks were under continuous agitation in a shaker chamber at 37 degrees C. Calcium and phosphate ions were measured beyond the maximum life span of an activated macrophage. The data showed that calcium ions rose within the first 24 h and then remained constant throughout the experiment for all pH groups. Phosphate ion concentration showed a similar pattern at the lysosomal pH but remained undetected at the other organelle pH levels. The saturation point was highest at the lysosomal pH level and lowest at the cytoplasmic pH level. The results of this experiment leave the potential for HA particles to dissolve following macrophage digestion. However, caution must be exercised when interpreting the macrophage organelle digestion hypothesis; the size of the HA particle, the length of time required to completely dissolve the particle, and potential cellular toxicity all are factors that have yet to be determined before this hypothesis can be validated.

Porous-coated metal-backed patellar components in total knee replacement. A postmortem retrieval analysis

The use of porous-coated metal-backed patellar components to achieve consistent fixation by bone ingrowth and to provide relief of pain warrants serious scrutiny. We conducted a quantitative postmortem investigation of eleven consecutively retrieved components with use of high-resolution contact radiographs, electron microscopy, and histological analysis. The implants had been in situ for a mean (and standard deviation) of 45+/-36 months (range, one to eighty-four months). Analysis of the high-resolution contact radiographs revealed that a mean of 86+/-12 per cent (range, 61 to 100 per cent) of the porous coating was in contact with the host bone. Backscattered electron imaging showed that the mean volume fraction of bone ingrowth was 13+/-9 per cent (range, 0 to 30 per cent). No significant difference was detected, with the numbers available, between the volume fraction of the bone ingrowth measured in the porous coating and that of the host cancellous bone in the patellae.