Journal of Orthopaedic Research 9414421 Raven Press, Ltd., New York 0 1991 Orthopaedic Research Society

Bypassing Femoral Cortical Defects With Cemented Intramedullary Stems "J. E. Larson, *?E. Y. S. Chao, and SR. H. Fitzgerald *Department of Orthopedics and fBiomechanics Laboratory, Mayo Clinic, Rochester, Minnesota, and $Department of Orthopedic Surgery, Wayne State University, Detroit, Michigan, U.S.A.

Summary: The objective of this investigation was to examine the effect of cemented intramedullary stem bypass on bone torsional property in the presence of femoral cortical defect. We intended to test two hypotheses; first, intramedullary fixation without bypass will accentuate the stress concentration effect and second, there will be an optimal length of stem bypass beyond the defect. Cemented intramedullary stems were used to bypass 50% diaphyseal diameter unicortical defects in paired, fresh-frozen canine femora. One member of each pair served as an unaltered control and all specimens were tested to failure in torsion. Both single-tailed paired t-test and analysis of variance were used for data analysis. Bones subjected to the cortical defect and no bypass were substantially weakened, exhibiting only 44 2 8% of control side maximum torque (p < 0.001). Positioning of the intramedullary stem with its tip at the center of the defect provided a small degree of strength improvement, achieving 60 ? 7% of control side maximum torque. One and two diaphyseal diameter bypass, however, significantly (p < 0.01) improved bone torsional strength, resulting in 80 -+ 6% and 84 13% of control side maximum torque, respectively. Three diameter bypass achieved only 68 8% of control side maximum torque. This significant (p < 0.05) decline in strength when compared to two diameter bypass appears to indicate that the length of stem bypass beyond the cortical defect does have an optimum, probably due to the geometric characteristics of the canine femora. Key Words: Cortical defectFemoral prosthesis-Stem bypass.

*

*

biomechanical effects of various defect sizes, shapes, and locations (8,9,16). The clinical consequence of weakening imposed by cortical defects is apparent in numerous reports detailing fractures through bone biopsy sites, screw holes, cortical windows and metastatic lesions (6,12,14,15,28). In order to improve stress transfer and reduce the risk of fracture through cortical defects, clinicians frequently utilize intramedullary devices such as the intramedullary nail or the longstem femoral prosthesis, with or without bone cement augmentation (11,24,29). As suggested in the revision hip arthroplasty series presented by Callaghan et al. (7), use of the long-stem femoral com-

The topic of bone weakening imposed by cortical defects such as screw holes, cortical windows, cortical perforations and regional endosteal thinning has inspired considerable interest from both clinicians and basic science investigators over the past 20 years (4,8,9,16,17,25,27). The engineering concepts of the stress riser and the open section effect have been applied to bony defects (13) and multiple studies have been undertaken to characterize the Received April 25, 1990; accepted October 25, 1990. Address correspondence and reprint requests to Dr. E. Y. S. Chao, Biomechanics Laboratory, Mayo ClinidMayo Foundation, Rochester, MN 55905, U.S.A.

414

BYPASSING FEMORAL CORTICAL DEFECTS

.

ponent, to bypass recognized cortical defects, decreases the risk of postoperative femoral fracture. Unfortunately, recommended bypass distances vary widely, and few basic science data has been generated to differentiate the relative value of different stem lengths in improving the structural strength of bone with a cortical defect. Panjabi et al. (18) specifically addressed the bypass issue in their study. Testing unpaired, embalmed cadaveric femora with below failure bending loads, Panjabi concluded that 1.5 diameter bypass was sufficient to eliminate the stress riser effect created by a cortical drill hole or endosteal cortical thinning. However, the effects of intramedullary bypass of defects under torsional loads, capable of producing bone fracture, were not addressed by Panjabi's study or by other reports in the literature. The purpose of this investigation was to study the effect of cemented intramedullary stem bypass distance on bone torsional properties in the presence of a femoral cortical defect. Attention was specifically directed at determination of bone-implant composite strength change due to intramedullary bypass and study of the effect of stem tip placement in relation to the level of a unicortical defect. MATERIALS AND METHODS A model for the bypassing of femoral cortical defects was established using paired, fresh-frozen canine femora (Fig. 1). Testing of paired specimens allowed one member of each pair to serve as a control for the opposite side and thereby helped to obviate differences in bone structure, nutritional background, and age between animals. Paired femora from mixed-breed dogs ranging in weight from 19 to 30 kg were harvested and dissected free of soft tissues. After careful recording of bone length and curvature, visual and radiographic inspection was used to detect any bony abnormalities. The accepted pairs were wrapped in salinesoaked towels. These bones were then labeled and double wrapped in tightly sealed plastic bags for storage at - 20°C until used. Freezing and thawing a single time has been shown to have little effect on the biomechanical properties of bone (10). The investkation was divided into three phases. First, five intact pairs were tested to failure in torsion to observe the mode and site of fracture, and to validate side-to-sidevariation. Next, five pairs were used to determine the weakening imposed by a 50%

Defect and IM Rod

415

Intact Control

FIG. 1. A model for testing the effects of bypass distance upon bone weakened by a cortical defect was developed with fresh-frozen canine femoral pairs. One side was subjected to the defect and bypass whereas the opposite side served as an unaltered control.

circular unicortical defect (the diameter of the defect was half of that of the mediolateral femoral cortical diameter at the same level) located anterolaterally below the lesser trochanter. Finally, the effects of stem bypass distances of zero, one, two, and three diaphyseal diameters beyond the defect were tested in 24 pairs. Defect location in the anterolateral cortex in this study was chosen because of its clinical correlate of a window for cement or prosthesis removal in revision hip arthroplasty. An effort was made by determination of cortical thickness and polar area moment of inertia, to objectively identify any region of canine femoral diaphysis that might be inherently weak. Anteroposterior and lateral contact radiographs were taken of all bones to rule out the presence of bone defects in terms of pathologic involvement, previous fracture malunion, presence of shotgun pellets, etc., and to allow for measurement of bone dimensions. Measurements taken from radiographs were corrected for magnification. The femoral length was determined by measuring parallel to the femoral shaft on the anteroposterior projection from the top of the head to the most distal point on the femoral condyles. The location of the cortical defect along the longitudinal axis of the femur was then established at a distance of 20% of the femoral

J Orthop Res, Vol. 9, NO. 3, 1991

416

J . E . LARSON ET AL.

length below the tip of the lesser trochanter. Placement of the defect at this level was selected to model the relative location of the tip of a primary hip arthroplasty component (18). The mediolateral outer cortical diameter was measured at the level of the defect on the anteroposterior radiograph. The diameter of the defect was then calculated as one half of this mediolateral outer diameter. The mediolateral inner cortical diameter was measured at the level of the isthmus on the anteroposterior radiograph. The isthmus represented the narrowest portion of the medullary canal and therefore was the limiting factor in sizing the intramedullary stem. To approximate the relative thicknesses of cortical bone, cement, and intramedullary stem seen in human hip arthroplasty, the intramedullary stems were sized to allow a 1-mm cement mantle at the isthmus. Bones were thawed at room temperature and were wrapped in saline-soaked cloth strips and plastic film to prevent dehydration. To facilitate uniform alignment and secure fixation, a distal condylar osteotomy was performed and two screws inserted into the distal metaphysis. Using two precision mounting jigs with fine parallel vertical reference lines engraved on plastic endplates, the middiaphysis of the bone model was aligned in both

anteroposterior and lateral planes (Fig. 2). Rotational alignment was verified based on the position of the distal fixation screws. The bone ends were potted in Wood’s metal (Cerrobend Alloy, Minneapolis, MN, U.S.A.) within the plastic mold boxes of the mounting jig so that the central axis of the middiaphysis was coaxial with the axis of torsion. The cortical defect was carefully created in the anterolateral cortex using a milling machine. Precise localization of the defect was attempted. Defect creation was achieved by sequential milling using gradually increasing cutter diameters under high speed and low feeding velocity while allowing constant saline irrigation in order to minimize microcrack formation around the cortical defect. Access to the medullary canal was achieved by drilling through the trochanteric fossa at a point coaxial with the longitudinal axis of the middiaphysis. The access site was enlarged using progressively larger drill bits until the stem plus a 1-mm cement mantle could be accommodated. The contents of the medullary canal were removed by flexible reaming up to the size of the inner cortical diameter at the isthmus. In this way, no diaphyseal cortical bone was removed by reaming. Prior to stem placement, the medullary canal was thoroughly cleansed by pulsed lavage with normal saline.

FIG. 2. In preparation for torsion testing, bones were carefully aligned in anteroposterior and lateral planes using a precision mounting jig with fine parallel vertical reference lines engraved into plastic side plates. Bone ends were potted in Cerrobend Alloy within plastic mold boxes to create uniform fixation blocks.

J Orthop Res, Vol. 9, No. 3, 1991

417

BYPASSING FEMORAL CORTICAL DEFECTS

Individually sized intramedullary stems were prepared from stainless steel rods of circular cross section. A transfixing rod of 6 mm in diameter and 5 cm in length was rigidly fixed to the proximal end of each stem. Both the proximal stem and the transfixing rod extended above the proximal femur to achieve secure fixation within the Wood's metal fixture block. Fixation in this manner ensured that the stem and proximal femur rotated together during torque application through the bone-implant composite. Such a loading mode, though not the same as the in vivo situation, was able to produce a uniform torque at the section with circular defect in order to test the effect of stem bypass on the strength of bone-implant composite. Standardized cement technique included the use of a rubber cement plug, a vent tube, and pressurized injection of polymethyl methacrylate (PMMA). After the bone model was accurately positioned on the specimen mounting platform of the Bridgeport Milling Machine, the stem was carefully guided by the drill press mechanism down the central portion of the medullary canal to the required distance. Two aspects of the cementing process merit additional comment. First, cement was cleared from the defect site after insertion of the intramedullary stem so that it would not be filled by excess cement. Second, the use of canal plugs was intended to allow a uniform 2-mm cement mantle beyond the stem tip. Because of plug migration due to injection pressure, cement was noted in some specimens to extend up to 20 mm beyond the stem tip. In order to investigate the effects of excessive distal cement extension on the torsional strength of the composite, five additional femoral pairs were studied. In this group, the stem tip was placed at the defect, and cement was intentionally allowed to extend two diaphyseal diameters beyond. The results could then be compared with the group of five pairs where both the stem tip and distal cement mantel ended at the level of the defect. The proximal and distal Wood's metal fixation blocks were seated securely into specially adapted fixtures on an MTS universal testing machine. Testing to failure in torsion was accomplished at 30" of external rotation per second with the load cell programmed to maintain zero axial load. A plot of rotation versus torque was generated for each specimen. Maximum torque and angular deformation to failure were determined directly from the plot. Energy absorption to failure as represented by the area under the rotation-torque curve

was determined by digitization of the curve and computer calculation software. Torsional stiffness was determined by the slope of the initial portion of the rotation-torque curve using a linear regression analysis program. Single-tailed paired t-tests were applied to compare right and left members of intact pairs. The influence of different bypass distances was assessed using analysis of variance and Tukey's studentized range test for post-hoc testing (30).

RESULTS Maximum torque values for the five intact canine paired femora had a 10% standard deviation (right versus left ratio: 1.047 ? 0.104), whereas angular deformation and energy absorption to failure were less reliable, with standard deviations of 17% (right versus left ratio: 0.940 k 0.161) and 20% (right versus left ratio: 1.034 0.215), respectively. The degree of variability between right and left sides in this model compares favorably to other reports of torsional tests on paired animal bones (9,20,26). In view of the insignificant differences observed between right and left sides, the usefulness of this paired model in providing meaningful information was established. Bones subjected to the 50% cortical defect alone consistently experienced spiral fractures through the defect with significant reductions (p < 0.001) in maximum torque, angular deformation, and energy absorption to failure (Table 1). Bone strength, as reflected by maximum torque, was found to be only 44 2 8% of control in the presence of the defect. The defect did not alter the torsional stiffness. The results of zero, one, two, and three diameter bypass are presented in Table 2. Positioning of the intramedullary rod with its tip at the center of the cortical defect provided a small improvement in torsional strength of the bone-implant composite, achieving 60 7% of control side maximum torque. One diameter bypass, however, provided significant (p < 0.01) recovery of bone strength to 80 f 6% of control side maximum torque, in five pairs. Two diameter bypass showed a modest additional improvement, achieving 84 ? 13% of control side maximum torque. Three diameter bypass, performed in nine femoral pairs, improved bone strength only to 68 ? 8% of control side maximum torque. Although this demonstrated a significant improvement (p < 0.01) over the femur with a defect and no bypass, it also represented a significant

*

*

J Orthop Res, Vol. 9, No. 3, 1991

J . E. LARSON ET AL.

418

TABLE 1. Change of mechanical properties of canine femora under torsion due to a diaphyseal defect Of 50% cortical diameter Mechanical properties Maximum torque at fracture (Nm) Maximum angular rotation at fracture (deg) Energy absorption at fracture (Joules) Torsional stiffness (Ndradian)

50% Defect (n = 5)

Intact (n = 5)

23.9 2 6.1

47.5 2 9.6

9.1 2 3.6 1.9 t 1.2 159.5 f 33.2

24.8 f 3.9 12.4 2 3.7 162.7 f 42.7

Defecvintact ratio (paired sample) 0.44

p value"

f 0.08

Bypassing femoral cortical defects with cemented intramedullary stems.

The objective of this investigation was to examine the effect of cemented intramedullary stem bypass on bone torsional property in the presence of fem...
778KB Sizes 0 Downloads 0 Views