- Email: [email protected]
Influence of muscular contractions on the stress analysis of distal femoral interlocking nailing
Available online at www.sciencedirect.com
Clinical Biomechanics 23 (2008) 38–44 www.elsevier.com/locate/clinbiomech
Influence of muscular contractions on the stress analysis of distal femoral interlocking nailing Kao-Shang Shih a
a,b
, Ching-Shiow Tseng
c,d
, Chia-Ching Lee
c,d
, Shang-Chih Lin
c,d,*
Division of Orthopedic Surgery, Department of Surgery, Far Eastern Memorial Hospital, Taipei, Taiwan b Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan c Department of Mechanical Engineering, National Central University, Taoyuan, Taiwan d Institute of Biomedical Engineering, National Central University, Taoyuan, Taiwan Received 25 February 2007; accepted 14 August 2007
Abstract Background. In the literature, the commonly assumed loading conditions on the proximal femur are hip compression and/or gluteus contractions. However, no study has discussed the influence of muscle forces on failure of distal nail holes and locking screws. Methods. This finite-element study analyzed the influence of muscular contractions on stress analysis of distal nail holes and locking screws. Three loading conditions were used for comparison, comprised of either hip compression alone or with muscle contractions. The head displacement of intact and fractured femur, the nail and screw stresses vs. fixation depth, and the stress distribution at the distal nail–screw interfaces were chosen as the comparison indices. Findings. The addition of trochanteric and diaphysial muscles showed the more physiologically reasonable displacement of the femoral head. However, all loading conditions consistently showed the hole and screw stresses increase as the nail was inserted deeper. The stress distribution at the distal nail–screw interfaces was remarkably different under the condition of with or without the muscular contractions. The exertion of muscles predicted the fatigue cracking originated at the edge of the nail holes on the medial rather than lateral side. Interpretation. Only hip compression and/or gluteus contraction generated a characteristic bending stress pattern and medially deflected nail curvature. Comparatively, the trochanteric and diaphysial muscles stabilized the femoral head and resulted in the higher stress concentration at the distally medial nail–screw interfaces. However, further experimental and clinical studies, focusing on the failure sites of the distal femoral hardware, should be undertaken to validate such findings. Published by Elsevier Ltd. Keywords: Muscle force; Femur; Interlocking nail; Finite-element model
1. Introduction Interlocking nailing has been a common technique for medullary osteosynthesis of the femoral fracture. Many clinical studies have documented high success rates associated with the treatment of femoral fracture with this technique (Giannoudis et al., 2006; Nowotarski et al., 2000; * Corresponding author. Address: Institute of Biomedical Engineering, Department of Mechanical ngineering, National Central University, No. 300, Jhongda Road, Jhongli City, Taoyuan County 32001, Taiwan. E-mail address: [email protected] (S.-C. Lin).
0268-0033/$ - see front matter Published by Elsevier Ltd. doi:10.1016/j.clinbiomech.2007.08.020
Whatling and Nokes, 2006). However, high stress concentrations at the distal nail holes and locking screws make them relatively vulnerable to fatigue cracking, especially in cases of distal femoral fracture (Bhat et al., 2006; Sancineto et al., 2001; Voor et al., 1997; Wu and Lee, 2005). Hence, many experimental and numerical studies have been devoted to the investigation of the mechanical failure of distal femoral nailing (Brumback et al., 1999; Bucholz et al., 1987; George et al., 1998; Hajek et al., 1993; Ito et al., 1998; Knothe et al., 2000; Lin et al., 2002). In the literature, the commonly assumed loading conditions on the femur are hip compression and/or gluteus
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
contractions (Antekeier et al., 2005; Cheung et al., 2004; Wang et al., 2000). The reasons for such simplified loading conditions in studies are ease in experimental set-ups and the absence of comprehensive muscle data. These oversimplified loading conditions generate a characteristic bending stress pattern along the femoral axis (Simoes et al., 2000; Taylor et al., 1996). However, some studies have concluded that other trochanteric and diaphysial muscles significantly influence stress distribution within the intact femur (Cristofolini et al., 1995; Duda et al., 1996, 1998; Mittlmeier et al., 1994). For distal femoral nailing, the load-transferring mechanism from the proximal to distal bone through the distal screw–hole interfaces is the point-contact problem in nature (Lin et al., 2002). To the authors’ knowledge, no study to date has discussed the influence of muscle forces on loosening, bending, or cracking of distal locking screws and failure of the nail holes. The aim of the current study was to investigate the effect of muscular contractions on stress distribution within the distally fractured femur and nailing system. The intact and distally fractured femurs were loaded under three loading conditions, comprised of either hip compression alone or in combination with muscular contractions. The value and location of the stress at the distal nail–screw interfaces and the displacement of the femoral head were chosen as the comparison indices between three loading conditions. The results of the finite-element analyses were aimed to provide further insight into the mechanism of failure at the distal hole–screw interfaces as well as to reveal deviceand surgery-related factors associated with better outcomes of surgical treatment of distal femoral fracture. 2. Methods 2.1. Femoral and nailing models In this study, the three-dimensional geometry of the femur utilized was attained from the International Society of Biomechanics (ISB) Finite Element Repository managed by the Instituti Orthopedici Rizzoli, Bologna, Italy (Viceconti, 1999). Two finite-element models, intact and distally fractured femurs, were developed under three loading conditions (Fig. 1a). In the distal fracture model, the transverse distal fracture with 10-mm separation was created at the location where the nail made no contact with the supracondylar cortex even upon full weight bearing. The cancellous bone contact and the adjacent fracture fragments were also assumed to have no load-bearing capacity. Hence, the mechanical contribution of the cancellous bone was excluded from the current finite-element models and the proximal loads were totally transferred into the distal locking screws. This would represent the worst-case scenario for stress analysis of distal nail holes and locking screws. The intramedullary canal of the fractured femoral model had at least 2-mm reaming space greater than the simulated nail with a 13-mm diameter. The bottoms of both intact and fractured femoral models at the knee joint were
39
immersed in a rigid box with a fixed degree-of-freedom (Fig. 1a). An interlocking nail simulated by the 316 L stainless-steel tube with 13-mm diameter, 1.5-mm wall thickness, and 1500-mm anteriorly bowing curvature was inserted into the medullary canal. Both diameters of the distal nail hole and locking screw were 6.5 mm. The separation between the two distal locking screws was 20 mm (Fig. 1b). The fixation depth of the more proximal locking screw was defined as the parallel distance between the fracture site and the more proximal locking screw (Fig. 1c). The x–y–z coordinate system with the origin at the intercondylar notch of the knee joint was defined as follows: the x-axis was along the mediolateral direction, the y-axis was along the anteroposterior direction, and the z-axis was along superoinferior direction (Fig. 1a). 2.2. Materials properties and loading conditions In the finite-element analyses, Young’s modulus and Poisson’s ratio were respectively assigned to be 240 GPa and 0.3 for the 316L stainless steel and 17.5 GPa and 0.3 for the cortical bone. All materials were assumed to have linearly elastic, homogeneous, and isotropic material properties throughout their composition. The calculated screw and hole stresses were compared with the yielding strength (= 896 MPa) of the 316L cold-worked stainless steel to validate the assumption of linear elasticity. As shown in Fig. 1a, three loading conditions were studied in the current study. The first loading condition used only vertical hip compression along the z-axis to produce compression and pure bending moment in the coronal plane. This is the most commonly used loading condition in the previously published studies. The second loading condition included oblique hip compression and gluteus contraction (Fig. 1a). In addition to the loads mentioned in loading condition 2, the iliopsoas, iliotibial tract, vastus lateralis, and adductor were included in loading condition 3. The third loading condition represented a more physiological case (Fig. 1a). The insertion points and directions of the muscle groups chosen were cited from the published literature (Brand et al., 1986; Duda et al., 1996). The vectorial components of the hip and muscle forces on the femur were selected at the static position of one leg stance (Simoes et al., 2000; Taylor et al., 1996). 2.3. Meshing strategy and interfacial conditions In the current study, an automatic mesh generation algorithm was used with COSMOSWorks Ed. 2007 software (SolidWorks Corporation, Concord, MA, USA), which provides a special element density in the vicinity of the distal nail holes and locking screws six times that of the remainder of the model with the overall average element size of 5 mm. The meshing strategy was designed for curved element boundary; thus, there were no sharp discontinuities to induce an unrealistically high stress concentration. Using the aspect ratio and Jacobian checks, all elements were within acceptable distortion limits, maximiz-
40
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
Fig. 1. (a) The finite-element models of the intact and distally fractured femurs used in this study are shown. (b) The schematic models of the locking screws and nail holes are represented. (c) Three fixation depth situations are assumed in the finite-element analyses.
ing the accuracy of the results. The model was meshed by the ten-node tetrahedral solid elements. The final meshes respectively consisted of 11,000 elements and 22,000 nodes for the intact model, and 22,000 elements and 41,000 nodes for the distally fractured model. The interaction of touching surfaces between the diaphysial canal, nail body, and distal locking screws were simulated with surface-to-surface contact elements without friction. The other interfaces were modeled as perfectly bonded. The mesh refinement in the vicinity of the distal nail holes and locking screws was executed for modeling accuracy until excellent monotonic convergence behavior (h-adaptive method) with <1% difference in the total strain energy was achieved. 2.4. Analyses In the fatigue tests of intramedullary supracondylar nailing, Voor et al. (1997) reported that nail cracking was found at the edge of the nail hole, which coincided with the location of the maximum tensile stress. Hence, in this study, the maximum principal stress was chosen as the indicator of the initiation site of fatigue cracking around nail holes and in locking screws. In our analyses, the distal nail–hole and screw stresses were analyzed as a function of fixation depth of the screw in the distal fragment under
three depth situations: 1.0, 2.0, and 3.0 cm (Fig. 1c). The displacement of the femoral head was defined as the displacement of the femoral head center after the exertion of hip compression and muscular contractions. For the distally fractured model, three comparison indices between three loading conditions were as follows: (1) the displacement of the femoral head, (2) the hole and screw stresses vs. fixation depth, and (3) the locations of the maximum principal stress at the distal nail–hole interfaces. 3. Results Higher failure rates at the first nail hole and locking screw have been reported more frequently than failure incidences at second nail holes and locking screws (Antekeier et al., 2005; Hajek et al., 1993; Lin et al., 2002; Wu and Shih, 1992). In the current study, the terms ‘‘hole stress’’ and ‘‘screw stress’’ referred to the maximum principal stresses around the first nail hole and in the first locking screw except for particular instances mentioned. 3.1. Displacement of the femoral head In the coronal (z–x) plane, the medial displacements of the femoral head in the intact model were 3.4, 4.7, and
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
1.4 mm in loading conditions 1, 2, and 3, respectively (Fig. 2a). For the fractured femur, the medial displacement of the femoral head was much greater than the corresponding intact one under three loading conditions (Fig. 2a). Among the three depth situations, the mediolateral deflection of the femoral head under loading condition 2 was significantly higher than those of loading conditions 1 and 3. For loading condition 2, the deeper the nail was inserted into the distal fragment, the greater the distance the femoral head medially displaced. However, the medial displacements of the femoral head in loading conditions 1 (6 mm) and 3 (2 mm) were almost constant in all three depth situations. In the sagittal (y–z) plane, the femoral head displaced 1.7, 1.0, and 0.5 mm under loading conditions 1, 2, and 3, respectively (Fig. 2b). Similar to mediolateral displacement, the loss of the structural integrity at the distal femur made the femoral head to be prone to anteroposterior displacement. Fig. 2b shows that the trochanteric and diaphysial muscles stabilize the fractured femur, thus providing the minimum anteroposterior deflection of the femoral head as compared with the other two conditions. In contrast with the mediolateral displacement, loading conditions 1 and 2 resulted in much greater anteroposterior displacement of the femoral head than loading condition 3. Displacement of the femoral head in loading condition 3 was almost constant (2 mm) under all three depth situations. 3.2. Hole and screw stresses vs. fixation depth
M
A
Hip A-P Displacement (mm)
P
The stress distribution of the first locking screw in loading condition 3 was remarkably different from those of the other two loading conditions (Fig. 5a and b). For loading conditions 1 and 2, the maximum screw stress consistently occurred at the contact points between the nail holes and locking screws. The inferiorly medial side and the superiorly lateral side were the typical contact points of the medially deflected nail and first locking screw. However, in loading condition 3, the first screw was highly stressed in the region within the medial and lateral nail holes. Similarly, the locations of maximum nail–hole stress were quite different between loading condition 1 and the other two loading situations (Fig. 6a and b). In loading conditions 1 and 2, the maximum principal stress of the medially deflected nail occurred at the lateral, anterior edge of the first nail hole. However, in loading condition 3, the location of maximum nail–hole stress was on the medial, anterior edge of the first nail hole.
a
10 8
Load 2
6
Load 1
4 2
Load 3 0 Intact Femur Fe mu
b
3.3. Stress distribution around nail–holes and in locking screws
1-cm 2-cm Fixation 3-cm 3-cm Fixation -cm Fixation Fixation 2-cm Depth Depth Depth Depth Depth Depth
Load 1 Load 2
2
Load 3 1 Second Screw
0
Intact Femur 2-cm Fixation Fixation 3-cm 3-cmFixation Fixation Intact Fe mur1 1-cm -cm Fixation Fixation 2-cm Depth Depth Depth Depth Depth Depth
Fig. 2. The mediolateral (a) and anteroposterior (b) displacements of the femoral head in the intact and fractured models are depicted.
Load 3 300
Load 2 200
Load 1
100
Fixation 1-cm Fixation Depth Depth
b
4 3
First Screw
400
First Screw Stress (MPa)
L
ment (Fig. 3a and b). For the second locking screw, such an increase in stress was even more significant in loading condition 3 than in loading conditions 1 and 2. For the 1-cm and 2-cm depth situations, the first locking screw was more stressed than the second screw in all three loading conditions. The stress values of both the first and second nail–holes increased in all three loading conditions for the deeper fixation depth (Fig. 4a and b). For the same fixation depth, the first nail–hole was more stressed than the second one in all three loading conditions. However, Fig. 3b shows that the second screw was subjected to the higher stress than the first one in the 3-cm depth situation under loading condition 3.
Second Screw Stress (MPa)
a
Hip M-L Displacement (mm)
All three loading conditions consistently showed that the maximum principal stress of both two locking screws increases as the nail is inserted deeper into the distal frag-
41
2-cm 2-cm Fixation Fixation Depth Depth
3-cm 3-cm Fixation Fixation Depth Depth
500
Load 3 400
Load 2
300 200
Load 1
100 1-cm 1-cm Fixation Fixation Depth Depth
Fixation 2-cm Fixation Depth Depth
3-cm Fixation Fixation Depth Depth
Fig. 3. The stresses of two distal locking screws related to the three depth situations in the distal fragment are emphasized.
42
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
a
a
First Hole Stress (MPa)
900
First Hole
MPa 900
Load 2 700
Load 3
500
600
Load 1
Superior Anterior
300
1-cm Fixation Depth
Second Hole Stress (MPa)
b
Second Hole
2-cm Fixation Depth
3-cm Fixation Depth
300
Posterior
b
800
Lateral
Medial Inferior
0
Load 3
600
Load 2
-300
400
Load 1 200
Fixation 1-cm Fixation Depth Depth
2-cm 2-cm Fixation Fixation Depth Depth
3-cm Fixation Fixation Depth Depth
Fig. 4. The stresses on the first and the second nail holes are related to the three depth situations in the distal fragment.
MPa 300
Fig. 6. The distribution of the maximum principal stress at the first nail– hole is shown. (a) For loading conditions 1 and 2. (b) For loading condition 3.
contact points and the rate of hardware failures. The influence of the reduction in bending moment and change in point-contact locations by the muscular contractions on the stress analysis of distal nail holes and locking screws were investigated in the current study.
a 100 Superior Anterior Lateral
Medial Inferior
0
Posterior
-100
b -300 Fig. 5. The distribution of the maximum principal stress at the first locking screw is shown. (a) For loading conditions 1 and 2. (b) For loading condition 3.
4. Discussion In the femur with distal nailing, stability of the fractured femur is achieved by inserting the nail into the medullary canal and the threading the screws on both sides of the fracture. Muscular contractions and articular compression above the proximal fragment of the fracture site are transmitted from the proximal intact shaft via the oblique proximal screw to the nail body anchored distally by the transverse-threaded screws. In the literature, Lin et al. (2002) showed that the proximal loads translate into quite high bending couple forces on the contact points between the distal nail holes and locking screws. Furthermore, the authors concluded that the magnitude and the direction of the assumed femoral loads affect the locations of the
4.1. Displacement of the femoral head Duda et al. (1997) showed that internal loads of the intact femur decreased as a result of muscular contractions. Experiments by Simoes et al. (2000) and finite-element analyses by Taylor et al. (1996) also confirmed this conclusion. Hence, loading condition 1 generated a characteristic bending stress pattern along the longitudinal axis of the intact femur. The hip vertical compression medially displaced the proximal femur, and thus the tensile stress occurred on the lateral side. Furthermore, gluteus contractions and hip oblique compression deflected the proximal femur more medially than loading condition 1. As shown in the literature reports (Cristofolini et al., 1995; Duda et al., 1996; Duda et al., 1998; Mittlmeier et al., 1994), the exertion of the more muscle forces balance the bending moment by hip compression and gluteus contraction. Consequently, we investigated whether muscle forces influence the stress value and distribution of the distal nail–holes and locking screws. An in vivo radiological study by Taylor et al. (1996) demonstrated that excessive medial deflection of the femoral head in the numerical analysis is physiologically unreasonable. As shown in Fig. 2a, the mediolateral deflections of the femoral head under loading conditions 1 and 2 were respectively three and four times greater than that seen under loading condition 3. For the loss of structural integrity, the medial displacement of the nailed femur was much greater than that of the corresponding intact femur under all three loading conditions (Fig. 2a and b). Among the
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
intact and fractured femurs, the medial deflection of the femoral head in loading condition 2 was about 50% and 300% greater than those in loading conditions 1 and 3, respectively. Hence, this study agrees with the biomechanical conclusion of Simoes et al. (2000) that the constrained femoral head provides a more physiologically reasonable loading condition. The addition of muscular contractions tends to reduce the bending moment in the coronal plane and decreases the medial displacement of the femoral head. In this study, the distal femur was assumed totally fixed at the region near the intercondylar notch of the knee joint (Fig. 1a). Hence, the predicted deflection of the femoral head only provided a rough estimate for in vivo femoral movement. Although the predicted mediolateral deflection of femoral head ranged only from 2 to 9 mm and seemed to be physiologically reasonable, the discrepancy among loading condition 3 and the other two loading situations might be greater in physiological loading and boundary conditions. Hence, a loading condition with only hip compression and/or gluteus contraction should be carefully assumed for a detailed evaluation of hole and screw stress. 4.2. Hole and screw stresses vs. fixation depth Two factors influence the stress around nail holes and in locking screws – geometry (size factor) and the loading conditions applied on them (load factor) (Lin et al., 2002). As the nail is inserted more distally in the femur, the locking screw needs to be longer and, thus, makes it subject to the higher stress. The size factor accounts for the increase in screw stress as the nail is inserted deeper into the distal fragment (Fig. 3a and b). The canal width is more constant within the disphysial region and rapidly increases within the supracondylar region. Hence, for the 1-cm and 2-cm depth situation, the first locking screw with a nearly constant length was more stressed than the second one with a comparatively longer length (Fig. 3a and b). Similarly, Fig. 3b shows a more remarkable change in the stress of the second screw as compared with the first one due to the expanding width of the supracondylar canal. The neglect of the stress concentration of the screw thread and some oversimplified boundary conditions were assumed in the current study. Hence, in the 3-cm depth situation, the higher stress at the second screw than the first one should be validated in order to discuss some clinical reports that have found higher failure rates of the first screw than the second one (Antekeier et al., 2005; Hajek et al., 1993; Lin et al., 2002; Wu and Shih, 1992; Wu and Lee, 2005). However, such size factor plays a minor influence on nail–hole stress because of the constant diameter of the nail hole. With a constant nail–hole size, the first nail hole was more stressed than the second one under all three loading conditions and the same fixation depth (Fig. 4a and b). The stress values of both the first and second nail–holes increased in all three loading conditions for the deeper fixation depth (Fig. 4a and b). Hence, the current study gives
43
no support to the finite-element study by Bucholz et al. (1987) that the hole stress became constant in a situation of more than 5-cm fixation depth. In this study, the mechanical contribution of the cancellous bone to the flexural stress was assumed negligible because the cyclic deflections of the interlocking nail medially compress and destroy the spongy bone. In addition, the interfacial, loading, and boundary conditions might play a significant role in the trend of the hole stress vs. fixation depth. More detailed studies with clinical and experimental methods should be performed to clear up such discrepancy. 4.3. Stress distribution around nail–holes and in locking screws Only hip compression and/or gluteal contraction result in higher medial deflection of the nail (Fig. 2a). Consequently, the nail makes contact with the first locking screw at the medially superior edge of the first nail hole (Lin et al., 2002). In loading conditions 1 and 2, the maximum screw stress occurred on the medially inferior side of the first screw (Fig. 5a). This result was consistent with the results of an axially compressive test by Hajek et al. (1993) that the failure of the distal screw is in the region between the nail and medial cortex. Comparatively, in loading conditions 1 and 2, the maximum principal stress of the medially deflected nail occurs on the lateral side of the first nail–hole (Fig. 6a). This means the initiation site of fatigue cracking around nail holes occurs on the lateral side of the first nail–hole. By means of metallurgical method, Bucholz et al. (1987) had found that the fatigue cracking originated at the edge of the nail holes on the medial side in five of six retrieved nails. Hence, as compared with the Bucholzs’ study, the distally increasing bending moment, i.e. loading conditions 1 and 2, generates inconsistent prediction of the cracking site for distal femoral nailing holes. Fig. 6b showed that loading condition 3 makes the locking screw contact with the nail holes at the anteroposterior edges on the mediolateral sides. Consequently, the maximum screw stress occurs at the region within the nail tube rather than at the localized nail–screw contact points (Fig. 5b). Under loading condition 3 (Fig. 6b), the location of the maximum nail–hole stress was on the anterior edge of the medial rather than lateral side, which was consistent with the metallurgical findings by Bucholz et al. (1987). This finding might provide further insight into the failure mechanism and design improvement at the distal hole–screw interfaces for both the clinician and bioengineers. For example, the wall thickness of the medial nail hole can be increased for further strengthening. The nail–hole size on the lateral side can be enlarged for easier screw insertion. 5. Conclusion In this study, the displacement of femoral head and stress at the distal nail–screw interfaces were calculated
44
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
with respect to fixation depth and chosen as the comparison indices between three loading conditions. The exertion of the more trochanteric and diaphysial muscles predicted remarkably different cracking sites of the distal nail–screw interfaces. However, the trends of the displacement of femoral head and stresses of nail holes and locking screws vs. fixation depth were similar among three loading conditions. From the aspect of experimental set-up, this is an unfavorable finding because the application of muscular contractions is a technical challenge. However, it is noteworthy for implant designers to evaluate the finding that fatigue fractures of the intramedullary nail are more likely to occur on the medial side of more proximal nail holes. Nevertheless, the muscular contractions and joint compression used in this study represent only a rough approximation. Hence, further clinical observations and additional biomechanical evaluations, focusing on hardware failure of distal femoral nailing, should be undertaken.
References Antekeier, S.B., Burden Jr., R.L., Voor, M.J., Roberts, C.S., 2005. Mechanical study of the safe distance between distal femoral fracture site and distal locking screws in antegrade intramedullary nailing. J. Orthop. Trauma. 19, 693–697. Bhat, A.K., Rao, S.K., Bhaskaranand, K., 2006. Mechanical failure in intramedullary interlocking nails. J. Orthop. Surg. (Hong Kong) 14, 138–141. Brand, R.A., Pedersen, D.R., Friederich, J.A., 1986. The sensitivity of muscle force predictions to changes in physiologic cross-sectional area. J. Biomech. 19, 589–596. Brumback, R.J., Toal Jr., T.R., Murphy-Zane, M.S., Novak, V.P., Belkoff, S.M., 1999. Immediate weight-bearing after treatment of a comminuted fracture of the femoral shaft with a statically locked intramedullary nail. J. Bone Joint Surg. Am. 81, 1538–1544. Bucholz, R.W., Ross, S.E., Lawrence, K.L., 1987. Fatigue fracture of the interlocking nail in the treatment of fractures of the distal part of the femoral shaft. J. Bone Joint Surg. Am. 69, 1391–1399. Cheung, G., Zalzal, P., Bhandari, M., Spelt, J.K., Papini, M., 2004. Finite element analysis of a femoral retrograde intramedullary nail subject to gait loading. Med. Eng. Phy. 26, 93–108. Cristofolini, L., Viceconti, M., Toni, A., Giunti, A., 1995. Influence of thigh muscles on the axial strains in a proximal femur during early stance in gait. J. Biomech. 28, 617–624. Duda, G.N., Brand, D., Freitag, S., Lierse, W., Schneider, E., 1996. Variability of femoral muscle attachments. J. Biomech. 29, 1185–1190. Duda, G.N., Schneider, E., Chao, E.Y., 1997. Internal forces and moments in the femur during walking. J. Biomech. 30, 933–941. Duda, G.N., Heller, M., Albinger, J., Schulz, O., Schneider, E., Claes, L., 1998. Influence of muscle forces on femoral strain distribution. J. Biomech. 31, 841–846.
George, C.J., Lindsey, R.W., Noble, P.C., Alexander, J.W., Kamaric, E., 1998. Optimal location of a single distal interlocking screw in intramedullary nailing of distal third femoral shaft fractures. J. Orthop. Trauma. 12, 267–272. Giannoudis, P.V., Papakostidis, C., Roberts, C., 2006. A review of the management of open fractures of the tibia and femur. J. Bone Joint Surg. Br. 88, 281–289. Hajek, P.D., Bicknell Jr., H.R., Bronson, W.E., Albright, J.A., Saha, S., 1993. The use of one compared with two distal screws in the treatment of femoral shaft fractures with interlocking intramedullary nailing. A clinical and biomechanical analysis. J. Bone Joint Surg. Am. 75, 519– 525. Ito, K., Grass, R., Zwipp, H., 1998. Internal fixation of supracondylar femoral fractures: comparative biomechanical performance of the 95degree blade plate and two retrograde nails. J. Orthop. Trauma. 12, 259–266. Knothe, U., Knothe, Tate M.L., Klaue, K., Perren, S.M., 2000. Development and testing of a new self-locking intramedullary nail system: testing of handling aspects and mechanical properties. Injury 31, 617–626. Lin, J., Lin, S.J., Chen, P.Q., Yang, S.H., 2002. Stress analysis of the distal locking screws for femoral interlocking nailing. J. Orthop. Res. 19, 57– 63. Mittlmeier, T., Mattheck, C., Dietrich, F., 1994. Effects of mechanical loading on the profile of human femoral diaphyseal geometry. Med. Eng. Phys. 16, 75–81. Nowotarski, P.J., Turen, C.H., Brumback, R.J., Scarboro, J.M., 2000. Conversion of external fixation to intramedullary nailing for fractures of the shaft of the femur in multiply injured patients. J. Bone Joint Surg. Am. 82, 781–788. Sancineto, C.F., Rubel, I.F., Seligson, D., Ferro, G.V., 2001. Technique for removal of broken interlocking screws. J. Orthop. Trauma. 15, 132–134. Simoes, J.A., Vaz, M.A., Blatcher, S., Taylor, M., 2000. Influence of head constraint and muscle forces on the strain distribution within the intact femur. Med. Eng. Phy. 22, 453–459. Taylor, M.E., Tanner, K.E., Freeman, M.A., Yettram, A.L., 1996. Stress and strain distribution within the intact femur: compression or bending? Med. Eng. Phy. 18, 122–131. Viceconti, M., 1999. The ISB Finite Element Mesh Repository. Available from:. Voor, M.J., Verst, D.A., Mladsi, S.W., Khalily, C., Seligson, D., 1997. Fatigue properties of a twelve-hole versus a five-hole intramedullary supracondylar nail. J. Orthop. Trauma. 11, 98–102. Wang, C.J., Brown, C.J., Yettram, A.L., Procter, P., 2000. Intramedullary femoral nails: one or two lag screws? A preliminary study. Med. Eng. Phys. 22, 613–624. Whatling, G.M., Nokes, L.D., 2006. Literature review of current techniques for the insertion of distal screws into intramedullary locking nails. Injury 37, 109–119. Wu, C.C., Lee, Z.L., 2005. Treatment of femoral shaft aseptic nonunion associated with broken distal locked screws and shortening. J. Trauma. 58, 837–840. Wu, C.C., Shih, C.H., 1992. Biomechanical analysis of the mechanism of interlocking nail failure. Arch. Orthop. Trauma Surg. 111, 268–272.
Clinical Biomechanics 23 (2008) 38–44 www.elsevier.com/locate/clinbiomech
Influence of muscular contractions on the stress analysis of distal femoral interlocking nailing Kao-Shang Shih a
a,b
, Ching-Shiow Tseng
c,d
, Chia-Ching Lee
c,d
, Shang-Chih Lin
c,d,*
Division of Orthopedic Surgery, Department of Surgery, Far Eastern Memorial Hospital, Taipei, Taiwan b Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan c Department of Mechanical Engineering, National Central University, Taoyuan, Taiwan d Institute of Biomedical Engineering, National Central University, Taoyuan, Taiwan Received 25 February 2007; accepted 14 August 2007
Abstract Background. In the literature, the commonly assumed loading conditions on the proximal femur are hip compression and/or gluteus contractions. However, no study has discussed the influence of muscle forces on failure of distal nail holes and locking screws. Methods. This finite-element study analyzed the influence of muscular contractions on stress analysis of distal nail holes and locking screws. Three loading conditions were used for comparison, comprised of either hip compression alone or with muscle contractions. The head displacement of intact and fractured femur, the nail and screw stresses vs. fixation depth, and the stress distribution at the distal nail–screw interfaces were chosen as the comparison indices. Findings. The addition of trochanteric and diaphysial muscles showed the more physiologically reasonable displacement of the femoral head. However, all loading conditions consistently showed the hole and screw stresses increase as the nail was inserted deeper. The stress distribution at the distal nail–screw interfaces was remarkably different under the condition of with or without the muscular contractions. The exertion of muscles predicted the fatigue cracking originated at the edge of the nail holes on the medial rather than lateral side. Interpretation. Only hip compression and/or gluteus contraction generated a characteristic bending stress pattern and medially deflected nail curvature. Comparatively, the trochanteric and diaphysial muscles stabilized the femoral head and resulted in the higher stress concentration at the distally medial nail–screw interfaces. However, further experimental and clinical studies, focusing on the failure sites of the distal femoral hardware, should be undertaken to validate such findings. Published by Elsevier Ltd. Keywords: Muscle force; Femur; Interlocking nail; Finite-element model
1. Introduction Interlocking nailing has been a common technique for medullary osteosynthesis of the femoral fracture. Many clinical studies have documented high success rates associated with the treatment of femoral fracture with this technique (Giannoudis et al., 2006; Nowotarski et al., 2000; * Corresponding author. Address: Institute of Biomedical Engineering, Department of Mechanical ngineering, National Central University, No. 300, Jhongda Road, Jhongli City, Taoyuan County 32001, Taiwan. E-mail address: [email protected] (S.-C. Lin).
0268-0033/$ - see front matter Published by Elsevier Ltd. doi:10.1016/j.clinbiomech.2007.08.020
Whatling and Nokes, 2006). However, high stress concentrations at the distal nail holes and locking screws make them relatively vulnerable to fatigue cracking, especially in cases of distal femoral fracture (Bhat et al., 2006; Sancineto et al., 2001; Voor et al., 1997; Wu and Lee, 2005). Hence, many experimental and numerical studies have been devoted to the investigation of the mechanical failure of distal femoral nailing (Brumback et al., 1999; Bucholz et al., 1987; George et al., 1998; Hajek et al., 1993; Ito et al., 1998; Knothe et al., 2000; Lin et al., 2002). In the literature, the commonly assumed loading conditions on the femur are hip compression and/or gluteus
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
contractions (Antekeier et al., 2005; Cheung et al., 2004; Wang et al., 2000). The reasons for such simplified loading conditions in studies are ease in experimental set-ups and the absence of comprehensive muscle data. These oversimplified loading conditions generate a characteristic bending stress pattern along the femoral axis (Simoes et al., 2000; Taylor et al., 1996). However, some studies have concluded that other trochanteric and diaphysial muscles significantly influence stress distribution within the intact femur (Cristofolini et al., 1995; Duda et al., 1996, 1998; Mittlmeier et al., 1994). For distal femoral nailing, the load-transferring mechanism from the proximal to distal bone through the distal screw–hole interfaces is the point-contact problem in nature (Lin et al., 2002). To the authors’ knowledge, no study to date has discussed the influence of muscle forces on loosening, bending, or cracking of distal locking screws and failure of the nail holes. The aim of the current study was to investigate the effect of muscular contractions on stress distribution within the distally fractured femur and nailing system. The intact and distally fractured femurs were loaded under three loading conditions, comprised of either hip compression alone or in combination with muscular contractions. The value and location of the stress at the distal nail–screw interfaces and the displacement of the femoral head were chosen as the comparison indices between three loading conditions. The results of the finite-element analyses were aimed to provide further insight into the mechanism of failure at the distal hole–screw interfaces as well as to reveal deviceand surgery-related factors associated with better outcomes of surgical treatment of distal femoral fracture. 2. Methods 2.1. Femoral and nailing models In this study, the three-dimensional geometry of the femur utilized was attained from the International Society of Biomechanics (ISB) Finite Element Repository managed by the Instituti Orthopedici Rizzoli, Bologna, Italy (Viceconti, 1999). Two finite-element models, intact and distally fractured femurs, were developed under three loading conditions (Fig. 1a). In the distal fracture model, the transverse distal fracture with 10-mm separation was created at the location where the nail made no contact with the supracondylar cortex even upon full weight bearing. The cancellous bone contact and the adjacent fracture fragments were also assumed to have no load-bearing capacity. Hence, the mechanical contribution of the cancellous bone was excluded from the current finite-element models and the proximal loads were totally transferred into the distal locking screws. This would represent the worst-case scenario for stress analysis of distal nail holes and locking screws. The intramedullary canal of the fractured femoral model had at least 2-mm reaming space greater than the simulated nail with a 13-mm diameter. The bottoms of both intact and fractured femoral models at the knee joint were
39
immersed in a rigid box with a fixed degree-of-freedom (Fig. 1a). An interlocking nail simulated by the 316 L stainless-steel tube with 13-mm diameter, 1.5-mm wall thickness, and 1500-mm anteriorly bowing curvature was inserted into the medullary canal. Both diameters of the distal nail hole and locking screw were 6.5 mm. The separation between the two distal locking screws was 20 mm (Fig. 1b). The fixation depth of the more proximal locking screw was defined as the parallel distance between the fracture site and the more proximal locking screw (Fig. 1c). The x–y–z coordinate system with the origin at the intercondylar notch of the knee joint was defined as follows: the x-axis was along the mediolateral direction, the y-axis was along the anteroposterior direction, and the z-axis was along superoinferior direction (Fig. 1a). 2.2. Materials properties and loading conditions In the finite-element analyses, Young’s modulus and Poisson’s ratio were respectively assigned to be 240 GPa and 0.3 for the 316L stainless steel and 17.5 GPa and 0.3 for the cortical bone. All materials were assumed to have linearly elastic, homogeneous, and isotropic material properties throughout their composition. The calculated screw and hole stresses were compared with the yielding strength (= 896 MPa) of the 316L cold-worked stainless steel to validate the assumption of linear elasticity. As shown in Fig. 1a, three loading conditions were studied in the current study. The first loading condition used only vertical hip compression along the z-axis to produce compression and pure bending moment in the coronal plane. This is the most commonly used loading condition in the previously published studies. The second loading condition included oblique hip compression and gluteus contraction (Fig. 1a). In addition to the loads mentioned in loading condition 2, the iliopsoas, iliotibial tract, vastus lateralis, and adductor were included in loading condition 3. The third loading condition represented a more physiological case (Fig. 1a). The insertion points and directions of the muscle groups chosen were cited from the published literature (Brand et al., 1986; Duda et al., 1996). The vectorial components of the hip and muscle forces on the femur were selected at the static position of one leg stance (Simoes et al., 2000; Taylor et al., 1996). 2.3. Meshing strategy and interfacial conditions In the current study, an automatic mesh generation algorithm was used with COSMOSWorks Ed. 2007 software (SolidWorks Corporation, Concord, MA, USA), which provides a special element density in the vicinity of the distal nail holes and locking screws six times that of the remainder of the model with the overall average element size of 5 mm. The meshing strategy was designed for curved element boundary; thus, there were no sharp discontinuities to induce an unrealistically high stress concentration. Using the aspect ratio and Jacobian checks, all elements were within acceptable distortion limits, maximiz-
40
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
Fig. 1. (a) The finite-element models of the intact and distally fractured femurs used in this study are shown. (b) The schematic models of the locking screws and nail holes are represented. (c) Three fixation depth situations are assumed in the finite-element analyses.
ing the accuracy of the results. The model was meshed by the ten-node tetrahedral solid elements. The final meshes respectively consisted of 11,000 elements and 22,000 nodes for the intact model, and 22,000 elements and 41,000 nodes for the distally fractured model. The interaction of touching surfaces between the diaphysial canal, nail body, and distal locking screws were simulated with surface-to-surface contact elements without friction. The other interfaces were modeled as perfectly bonded. The mesh refinement in the vicinity of the distal nail holes and locking screws was executed for modeling accuracy until excellent monotonic convergence behavior (h-adaptive method) with <1% difference in the total strain energy was achieved. 2.4. Analyses In the fatigue tests of intramedullary supracondylar nailing, Voor et al. (1997) reported that nail cracking was found at the edge of the nail hole, which coincided with the location of the maximum tensile stress. Hence, in this study, the maximum principal stress was chosen as the indicator of the initiation site of fatigue cracking around nail holes and in locking screws. In our analyses, the distal nail–hole and screw stresses were analyzed as a function of fixation depth of the screw in the distal fragment under
three depth situations: 1.0, 2.0, and 3.0 cm (Fig. 1c). The displacement of the femoral head was defined as the displacement of the femoral head center after the exertion of hip compression and muscular contractions. For the distally fractured model, three comparison indices between three loading conditions were as follows: (1) the displacement of the femoral head, (2) the hole and screw stresses vs. fixation depth, and (3) the locations of the maximum principal stress at the distal nail–hole interfaces. 3. Results Higher failure rates at the first nail hole and locking screw have been reported more frequently than failure incidences at second nail holes and locking screws (Antekeier et al., 2005; Hajek et al., 1993; Lin et al., 2002; Wu and Shih, 1992). In the current study, the terms ‘‘hole stress’’ and ‘‘screw stress’’ referred to the maximum principal stresses around the first nail hole and in the first locking screw except for particular instances mentioned. 3.1. Displacement of the femoral head In the coronal (z–x) plane, the medial displacements of the femoral head in the intact model were 3.4, 4.7, and
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
1.4 mm in loading conditions 1, 2, and 3, respectively (Fig. 2a). For the fractured femur, the medial displacement of the femoral head was much greater than the corresponding intact one under three loading conditions (Fig. 2a). Among the three depth situations, the mediolateral deflection of the femoral head under loading condition 2 was significantly higher than those of loading conditions 1 and 3. For loading condition 2, the deeper the nail was inserted into the distal fragment, the greater the distance the femoral head medially displaced. However, the medial displacements of the femoral head in loading conditions 1 (6 mm) and 3 (2 mm) were almost constant in all three depth situations. In the sagittal (y–z) plane, the femoral head displaced 1.7, 1.0, and 0.5 mm under loading conditions 1, 2, and 3, respectively (Fig. 2b). Similar to mediolateral displacement, the loss of the structural integrity at the distal femur made the femoral head to be prone to anteroposterior displacement. Fig. 2b shows that the trochanteric and diaphysial muscles stabilize the fractured femur, thus providing the minimum anteroposterior deflection of the femoral head as compared with the other two conditions. In contrast with the mediolateral displacement, loading conditions 1 and 2 resulted in much greater anteroposterior displacement of the femoral head than loading condition 3. Displacement of the femoral head in loading condition 3 was almost constant (2 mm) under all three depth situations. 3.2. Hole and screw stresses vs. fixation depth
M
A
Hip A-P Displacement (mm)
P
The stress distribution of the first locking screw in loading condition 3 was remarkably different from those of the other two loading conditions (Fig. 5a and b). For loading conditions 1 and 2, the maximum screw stress consistently occurred at the contact points between the nail holes and locking screws. The inferiorly medial side and the superiorly lateral side were the typical contact points of the medially deflected nail and first locking screw. However, in loading condition 3, the first screw was highly stressed in the region within the medial and lateral nail holes. Similarly, the locations of maximum nail–hole stress were quite different between loading condition 1 and the other two loading situations (Fig. 6a and b). In loading conditions 1 and 2, the maximum principal stress of the medially deflected nail occurred at the lateral, anterior edge of the first nail hole. However, in loading condition 3, the location of maximum nail–hole stress was on the medial, anterior edge of the first nail hole.
a
10 8
Load 2
6
Load 1
4 2
Load 3 0 Intact Femur Fe mu
b
3.3. Stress distribution around nail–holes and in locking screws
1-cm 2-cm Fixation 3-cm 3-cm Fixation -cm Fixation Fixation 2-cm Depth Depth Depth Depth Depth Depth
Load 1 Load 2
2
Load 3 1 Second Screw
0
Intact Femur 2-cm Fixation Fixation 3-cm 3-cmFixation Fixation Intact Fe mur1 1-cm -cm Fixation Fixation 2-cm Depth Depth Depth Depth Depth Depth
Fig. 2. The mediolateral (a) and anteroposterior (b) displacements of the femoral head in the intact and fractured models are depicted.
Load 3 300
Load 2 200
Load 1
100
Fixation 1-cm Fixation Depth Depth
b
4 3
First Screw
400
First Screw Stress (MPa)
L
ment (Fig. 3a and b). For the second locking screw, such an increase in stress was even more significant in loading condition 3 than in loading conditions 1 and 2. For the 1-cm and 2-cm depth situations, the first locking screw was more stressed than the second screw in all three loading conditions. The stress values of both the first and second nail–holes increased in all three loading conditions for the deeper fixation depth (Fig. 4a and b). For the same fixation depth, the first nail–hole was more stressed than the second one in all three loading conditions. However, Fig. 3b shows that the second screw was subjected to the higher stress than the first one in the 3-cm depth situation under loading condition 3.
Second Screw Stress (MPa)
a
Hip M-L Displacement (mm)
All three loading conditions consistently showed that the maximum principal stress of both two locking screws increases as the nail is inserted deeper into the distal frag-
41
2-cm 2-cm Fixation Fixation Depth Depth
3-cm 3-cm Fixation Fixation Depth Depth
500
Load 3 400
Load 2
300 200
Load 1
100 1-cm 1-cm Fixation Fixation Depth Depth
Fixation 2-cm Fixation Depth Depth
3-cm Fixation Fixation Depth Depth
Fig. 3. The stresses of two distal locking screws related to the three depth situations in the distal fragment are emphasized.
42
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
a
a
First Hole Stress (MPa)
900
First Hole
MPa 900
Load 2 700
Load 3
500
600
Load 1
Superior Anterior
300
1-cm Fixation Depth
Second Hole Stress (MPa)
b
Second Hole
2-cm Fixation Depth
3-cm Fixation Depth
300
Posterior
b
800
Lateral
Medial Inferior
0
Load 3
600
Load 2
-300
400
Load 1 200
Fixation 1-cm Fixation Depth Depth
2-cm 2-cm Fixation Fixation Depth Depth
3-cm Fixation Fixation Depth Depth
Fig. 4. The stresses on the first and the second nail holes are related to the three depth situations in the distal fragment.
MPa 300
Fig. 6. The distribution of the maximum principal stress at the first nail– hole is shown. (a) For loading conditions 1 and 2. (b) For loading condition 3.
contact points and the rate of hardware failures. The influence of the reduction in bending moment and change in point-contact locations by the muscular contractions on the stress analysis of distal nail holes and locking screws were investigated in the current study.
a 100 Superior Anterior Lateral
Medial Inferior
0
Posterior
-100
b -300 Fig. 5. The distribution of the maximum principal stress at the first locking screw is shown. (a) For loading conditions 1 and 2. (b) For loading condition 3.
4. Discussion In the femur with distal nailing, stability of the fractured femur is achieved by inserting the nail into the medullary canal and the threading the screws on both sides of the fracture. Muscular contractions and articular compression above the proximal fragment of the fracture site are transmitted from the proximal intact shaft via the oblique proximal screw to the nail body anchored distally by the transverse-threaded screws. In the literature, Lin et al. (2002) showed that the proximal loads translate into quite high bending couple forces on the contact points between the distal nail holes and locking screws. Furthermore, the authors concluded that the magnitude and the direction of the assumed femoral loads affect the locations of the
4.1. Displacement of the femoral head Duda et al. (1997) showed that internal loads of the intact femur decreased as a result of muscular contractions. Experiments by Simoes et al. (2000) and finite-element analyses by Taylor et al. (1996) also confirmed this conclusion. Hence, loading condition 1 generated a characteristic bending stress pattern along the longitudinal axis of the intact femur. The hip vertical compression medially displaced the proximal femur, and thus the tensile stress occurred on the lateral side. Furthermore, gluteus contractions and hip oblique compression deflected the proximal femur more medially than loading condition 1. As shown in the literature reports (Cristofolini et al., 1995; Duda et al., 1996; Duda et al., 1998; Mittlmeier et al., 1994), the exertion of the more muscle forces balance the bending moment by hip compression and gluteus contraction. Consequently, we investigated whether muscle forces influence the stress value and distribution of the distal nail–holes and locking screws. An in vivo radiological study by Taylor et al. (1996) demonstrated that excessive medial deflection of the femoral head in the numerical analysis is physiologically unreasonable. As shown in Fig. 2a, the mediolateral deflections of the femoral head under loading conditions 1 and 2 were respectively three and four times greater than that seen under loading condition 3. For the loss of structural integrity, the medial displacement of the nailed femur was much greater than that of the corresponding intact femur under all three loading conditions (Fig. 2a and b). Among the
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
intact and fractured femurs, the medial deflection of the femoral head in loading condition 2 was about 50% and 300% greater than those in loading conditions 1 and 3, respectively. Hence, this study agrees with the biomechanical conclusion of Simoes et al. (2000) that the constrained femoral head provides a more physiologically reasonable loading condition. The addition of muscular contractions tends to reduce the bending moment in the coronal plane and decreases the medial displacement of the femoral head. In this study, the distal femur was assumed totally fixed at the region near the intercondylar notch of the knee joint (Fig. 1a). Hence, the predicted deflection of the femoral head only provided a rough estimate for in vivo femoral movement. Although the predicted mediolateral deflection of femoral head ranged only from 2 to 9 mm and seemed to be physiologically reasonable, the discrepancy among loading condition 3 and the other two loading situations might be greater in physiological loading and boundary conditions. Hence, a loading condition with only hip compression and/or gluteus contraction should be carefully assumed for a detailed evaluation of hole and screw stress. 4.2. Hole and screw stresses vs. fixation depth Two factors influence the stress around nail holes and in locking screws – geometry (size factor) and the loading conditions applied on them (load factor) (Lin et al., 2002). As the nail is inserted more distally in the femur, the locking screw needs to be longer and, thus, makes it subject to the higher stress. The size factor accounts for the increase in screw stress as the nail is inserted deeper into the distal fragment (Fig. 3a and b). The canal width is more constant within the disphysial region and rapidly increases within the supracondylar region. Hence, for the 1-cm and 2-cm depth situation, the first locking screw with a nearly constant length was more stressed than the second one with a comparatively longer length (Fig. 3a and b). Similarly, Fig. 3b shows a more remarkable change in the stress of the second screw as compared with the first one due to the expanding width of the supracondylar canal. The neglect of the stress concentration of the screw thread and some oversimplified boundary conditions were assumed in the current study. Hence, in the 3-cm depth situation, the higher stress at the second screw than the first one should be validated in order to discuss some clinical reports that have found higher failure rates of the first screw than the second one (Antekeier et al., 2005; Hajek et al., 1993; Lin et al., 2002; Wu and Shih, 1992; Wu and Lee, 2005). However, such size factor plays a minor influence on nail–hole stress because of the constant diameter of the nail hole. With a constant nail–hole size, the first nail hole was more stressed than the second one under all three loading conditions and the same fixation depth (Fig. 4a and b). The stress values of both the first and second nail–holes increased in all three loading conditions for the deeper fixation depth (Fig. 4a and b). Hence, the current study gives
43
no support to the finite-element study by Bucholz et al. (1987) that the hole stress became constant in a situation of more than 5-cm fixation depth. In this study, the mechanical contribution of the cancellous bone to the flexural stress was assumed negligible because the cyclic deflections of the interlocking nail medially compress and destroy the spongy bone. In addition, the interfacial, loading, and boundary conditions might play a significant role in the trend of the hole stress vs. fixation depth. More detailed studies with clinical and experimental methods should be performed to clear up such discrepancy. 4.3. Stress distribution around nail–holes and in locking screws Only hip compression and/or gluteal contraction result in higher medial deflection of the nail (Fig. 2a). Consequently, the nail makes contact with the first locking screw at the medially superior edge of the first nail hole (Lin et al., 2002). In loading conditions 1 and 2, the maximum screw stress occurred on the medially inferior side of the first screw (Fig. 5a). This result was consistent with the results of an axially compressive test by Hajek et al. (1993) that the failure of the distal screw is in the region between the nail and medial cortex. Comparatively, in loading conditions 1 and 2, the maximum principal stress of the medially deflected nail occurs on the lateral side of the first nail–hole (Fig. 6a). This means the initiation site of fatigue cracking around nail holes occurs on the lateral side of the first nail–hole. By means of metallurgical method, Bucholz et al. (1987) had found that the fatigue cracking originated at the edge of the nail holes on the medial side in five of six retrieved nails. Hence, as compared with the Bucholzs’ study, the distally increasing bending moment, i.e. loading conditions 1 and 2, generates inconsistent prediction of the cracking site for distal femoral nailing holes. Fig. 6b showed that loading condition 3 makes the locking screw contact with the nail holes at the anteroposterior edges on the mediolateral sides. Consequently, the maximum screw stress occurs at the region within the nail tube rather than at the localized nail–screw contact points (Fig. 5b). Under loading condition 3 (Fig. 6b), the location of the maximum nail–hole stress was on the anterior edge of the medial rather than lateral side, which was consistent with the metallurgical findings by Bucholz et al. (1987). This finding might provide further insight into the failure mechanism and design improvement at the distal hole–screw interfaces for both the clinician and bioengineers. For example, the wall thickness of the medial nail hole can be increased for further strengthening. The nail–hole size on the lateral side can be enlarged for easier screw insertion. 5. Conclusion In this study, the displacement of femoral head and stress at the distal nail–screw interfaces were calculated
44
K.-S. Shih et al. / Clinical Biomechanics 23 (2008) 38–44
with respect to fixation depth and chosen as the comparison indices between three loading conditions. The exertion of the more trochanteric and diaphysial muscles predicted remarkably different cracking sites of the distal nail–screw interfaces. However, the trends of the displacement of femoral head and stresses of nail holes and locking screws vs. fixation depth were similar among three loading conditions. From the aspect of experimental set-up, this is an unfavorable finding because the application of muscular contractions is a technical challenge. However, it is noteworthy for implant designers to evaluate the finding that fatigue fractures of the intramedullary nail are more likely to occur on the medial side of more proximal nail holes. Nevertheless, the muscular contractions and joint compression used in this study represent only a rough approximation. Hence, further clinical observations and additional biomechanical evaluations, focusing on hardware failure of distal femoral nailing, should be undertaken.
References Antekeier, S.B., Burden Jr., R.L., Voor, M.J., Roberts, C.S., 2005. Mechanical study of the safe distance between distal femoral fracture site and distal locking screws in antegrade intramedullary nailing. J. Orthop. Trauma. 19, 693–697. Bhat, A.K., Rao, S.K., Bhaskaranand, K., 2006. Mechanical failure in intramedullary interlocking nails. J. Orthop. Surg. (Hong Kong) 14, 138–141. Brand, R.A., Pedersen, D.R., Friederich, J.A., 1986. The sensitivity of muscle force predictions to changes in physiologic cross-sectional area. J. Biomech. 19, 589–596. Brumback, R.J., Toal Jr., T.R., Murphy-Zane, M.S., Novak, V.P., Belkoff, S.M., 1999. Immediate weight-bearing after treatment of a comminuted fracture of the femoral shaft with a statically locked intramedullary nail. J. Bone Joint Surg. Am. 81, 1538–1544. Bucholz, R.W., Ross, S.E., Lawrence, K.L., 1987. Fatigue fracture of the interlocking nail in the treatment of fractures of the distal part of the femoral shaft. J. Bone Joint Surg. Am. 69, 1391–1399. Cheung, G., Zalzal, P., Bhandari, M., Spelt, J.K., Papini, M., 2004. Finite element analysis of a femoral retrograde intramedullary nail subject to gait loading. Med. Eng. Phy. 26, 93–108. Cristofolini, L., Viceconti, M., Toni, A., Giunti, A., 1995. Influence of thigh muscles on the axial strains in a proximal femur during early stance in gait. J. Biomech. 28, 617–624. Duda, G.N., Brand, D., Freitag, S., Lierse, W., Schneider, E., 1996. Variability of femoral muscle attachments. J. Biomech. 29, 1185–1190. Duda, G.N., Schneider, E., Chao, E.Y., 1997. Internal forces and moments in the femur during walking. J. Biomech. 30, 933–941. Duda, G.N., Heller, M., Albinger, J., Schulz, O., Schneider, E., Claes, L., 1998. Influence of muscle forces on femoral strain distribution. J. Biomech. 31, 841–846.
George, C.J., Lindsey, R.W., Noble, P.C., Alexander, J.W., Kamaric, E., 1998. Optimal location of a single distal interlocking screw in intramedullary nailing of distal third femoral shaft fractures. J. Orthop. Trauma. 12, 267–272. Giannoudis, P.V., Papakostidis, C., Roberts, C., 2006. A review of the management of open fractures of the tibia and femur. J. Bone Joint Surg. Br. 88, 281–289. Hajek, P.D., Bicknell Jr., H.R., Bronson, W.E., Albright, J.A., Saha, S., 1993. The use of one compared with two distal screws in the treatment of femoral shaft fractures with interlocking intramedullary nailing. A clinical and biomechanical analysis. J. Bone Joint Surg. Am. 75, 519– 525. Ito, K., Grass, R., Zwipp, H., 1998. Internal fixation of supracondylar femoral fractures: comparative biomechanical performance of the 95degree blade plate and two retrograde nails. J. Orthop. Trauma. 12, 259–266. Knothe, U., Knothe, Tate M.L., Klaue, K., Perren, S.M., 2000. Development and testing of a new self-locking intramedullary nail system: testing of handling aspects and mechanical properties. Injury 31, 617–626. Lin, J., Lin, S.J., Chen, P.Q., Yang, S.H., 2002. Stress analysis of the distal locking screws for femoral interlocking nailing. J. Orthop. Res. 19, 57– 63. Mittlmeier, T., Mattheck, C., Dietrich, F., 1994. Effects of mechanical loading on the profile of human femoral diaphyseal geometry. Med. Eng. Phys. 16, 75–81. Nowotarski, P.J., Turen, C.H., Brumback, R.J., Scarboro, J.M., 2000. Conversion of external fixation to intramedullary nailing for fractures of the shaft of the femur in multiply injured patients. J. Bone Joint Surg. Am. 82, 781–788. Sancineto, C.F., Rubel, I.F., Seligson, D., Ferro, G.V., 2001. Technique for removal of broken interlocking screws. J. Orthop. Trauma. 15, 132–134. Simoes, J.A., Vaz, M.A., Blatcher, S., Taylor, M., 2000. Influence of head constraint and muscle forces on the strain distribution within the intact femur. Med. Eng. Phy. 22, 453–459. Taylor, M.E., Tanner, K.E., Freeman, M.A., Yettram, A.L., 1996. Stress and strain distribution within the intact femur: compression or bending? Med. Eng. Phy. 18, 122–131. Viceconti, M., 1999. The ISB Finite Element Mesh Repository. Available from: