A cadaveric experimental methodology for performance evaluation of intramedullary nails for femur and tibia.
Intramedullary nails are implants placed within the intramedullay canal of a bone to stabilize long bone fractures. These nails are most often made of either stainless steel or titanium and are generally secured to the bone via sets of screws inserted through the implant at either end . The interlocking nail widens the range of indications for medullary osteosynthesis of femoral and tibial shaft fractures . A low rate of complications and non-unions has been reported as well as excellent and prompt return of function . Other implanted devices which end in the subtrochanteric region such as femoral neck pins and short ante grade intramedually nails, have well established history of increased fracture risk . Unlike plate fixation, intramedullary nailing does not provide absolutely rigid fixation, so that some motion at the fracture site exists . While the effectiveness of intramedullary nails has been demonstrated clinically, a number of complications including bone refracture and implant failure persist. Some explanations for these failures, both in the bone and in the implant include high stress concentration in certain regions, torsional deformity and stress shielding. Unfortunately, the detailed load sharing between bone and nails and the locations of high stress concentration in response to realistic loading and configuration of the screws remains largely unknown. Also some suggested explanations are difficult to evaluate objectively . The development of long term successful orthopaedic implants requires knowledge of the physiological environment to which these implants are exposed . Fracture fixation by means of metal implants provides an opportunity to measure the implant-born forces and to estimate the long bone forces before healing occurs. The load changes during fracture healing may provide additional information. 
Forces acting on long bones
Musculoskeletal loading influences the stresses and strains within the human femur. It is essential for implant design and simulations of bone modeling processes to identify locally high or low strain values, which may lead to bone resorption and thereby affect the clinical outcome . Significant forces are present in the long bones, but their magnitudes have so far only been estimated from mathematical models.
Using one of the Engineering Tools, Finite Element Analysis (FEA) three dimensional study of the strain in the proximal portion of the standardized femur implanted with a retrograde intramedullary nail in response to simplified axial, bending and compression loading, The authors found that varying the length of the nail alters the strain distribution in the proximal femur geometry and that increasing the nail length introduces more stress shielding into the proximal region . Moreover many of these FE work hasn't been experimentally supported and hugely depend on the software capability and then there are many assumptions added to it .
Strain Measurement Studies
In vivo studies on animals were done to measure the successive strength changes of the callus after a fracture. Specifically designed Kuntscher nails were made to meet the anterior convexity and the size of the medullary canal of the femora of goats. Three self-temperature-compensating semi-conductor strain gauges were glued on the inner surface of the nail and coated with
epoxy-resin and silicone rubber. There was a tendency that the less loosening is found between the nail and the bone, the larger is the strain and the smaller becomes the dispersion of the strains. On the three point bending test, the post-fracture strain changes were not directly proportional to the lapse of time. In the early phase, the strains remained large for about four weeks and tended to rapidly decrease in the middle and late phases in accordance with the development of callus formation on x-ray. A similar strain change was observed on the walking test, although the animal presented with some painful limp. The results suggest that early excessive exercise to be avoided when an intramedullary fixation is not sufficiently rigid . X-rays and scan datas have been used for studying the failure modes and healing process . These studies do depend on the experience of interpretation. 
Literature studies showed that less work has been done on the strain changes happening in the cortical bone around a crack site. There is no study to measure the load supported by an intramedullary nail under various sized cracks. Few authors have carried out in vivo strain measurements by customizing the implant to fix the strain gages. But this has a limitation of testing with the regularly used implants. As these studies have been carried out in vivo, there is always a risk of experimenting more loads during the early postoperative phase. Thus these studies have been limited to around 250N only during this phase. So any further in vivo telemetrized work needs to have an idea on the behaviour of these bones to more loading data. In effect a cadaveric experimental study will give an insight on the behaviour of the implant during high loads, which could then be interpreted for evaluation of the implant. The strain changes around the crack site are bound to show the behaviour of the bone during the various loads. These data will also help orthopedic surgeon to plan their postoperative protocols for fracture healing and implant protection .
The scope of this project is to measure the strain changes on the cortical bone of middle aged cadaveric femur and tibial bone around various crack sizes using rosette type strain gages under various compressive loads using the fit implant manufactured indigenously in India. This study would help us
1. To evolve a methodology to validate the effectiveness of the implant
2. To study the weight bearing capability of a long bone fracture of femur and tibia
Methods and Materials
Selection of cadaveric bone
Cadaveric specimen to be selected has to be representative and of a healthy one. Bone mineral density for the cadaveric bones were undertaken to establish its calcium deterioration and was found to be very minimal when compared to middle aged normal healthy person (Table 1). The selected size of the femur was 430mm in length and the diameter at midshaft section was 32mm and that of tibia was 370mm and diameter at midshaft section was 28mm. Interlocking nail and its fixation in femur and tibia
Interlocking fixation has become increasingly popular for femoral and tibial shaft fracture treatment. These implants offer improved load sharing biomechanics. STAAN interlocking stainless steel nail (STAAN Surgicals, Coimbatore, India) was used for our experiments.
The nail was implanted by orthopaedic surgeons using STAAN surgical instrumentation designed for use with this particular implant. The surgical technique guide provided by the manufacturer was followed during the procedure for reaming through both femur and tibia.
The strains on a cadaveric femur bone were recorded by M.Papini et al and the study shows that the proximal region, midshaft region and the distal region can be taken as the representative on the strain distribution on long bones. So oblique cracks were generated in the midshaft region of femur as well as in tibia using hacksaw. The sizes of the crack were 0%(no crack), 30%, 60%, 90% with respect to the diameter of the bone midshaft with 90% crack being representative of a comminuted midshaft crack . The experiments were conducted on these bones from 0% to 90% crack size .
Instron Tensile testing machine was used for various compressive loads. The bones were loaded in the 11[degrees]abduction position as shown in fig (1).
The experiments was designed with the femur and tibial bones compressive loaded with progressive loads of 500N, 1000N, 1500N, 2000N, 2500N and reversed with same loads and their strain measurement studied : Case 1 without the fixation of implant, Case 2 with implant no crack, Case 3 with implant 30% crack, Case 4 with implant 60% crack, Case 5 with implant 90% crack.
[FIGURE 1 OMITTED]
Strain gaging on the bones
Strain gages are used for finding the strain rate in any object, which is affected by some load. The change in the value of resistance by straining the gauge may be partly explained by the normal dimensional behavior of elastic material. The strain measured by the gages is principal strains. One of the types of the strain gage is the rosette type strain gage where three strain gages are found in the lateral, longitudinal and shear directions separated by 45 degrees each, which can measure strains in all the three directions. Rosette type strain gages were used for our study. The strain gages were fixed to the bones using the methacrylate bonding material. The interface wires were soldered to the gages and interfaced with the strain meter to readout the strain values from all the three directions in the rosette.
In case 1, of the three strain gages one was fixed proximally and second one in the anterior aspect in the midshaft region and the third on Lateral aspect in the midshaft region for both femur and tibial bones. For case2 to case 5 for femur bone two strain gages were fixed one inch above the fracture site in the anterior aspect and one inch below the fracture site in the anterior aspect.
In the case 2 to case 5 for tibial bone too two strain gages were fixed one inch above the fracture site in the antero medial aspect of the midshaft region and another gage one inch below fracture site in the antereomedial aspect in the midshaft region. The strain readings were observed for lateral, shear and longitudinal strains in all the gages for all the cracks during the compression loading and the retraction cycle too, to analyse the behaviour of the bone during the stress relaxation.
Compressive loading of femur and tibia without implant
The study of the strain pattern of a bone without any fracture is necessary to validate the experimental setup and for comparison with the results of fractured bone for performance evaluation. It was found that at 2.5kN, strain measuring upto 1500microstrain was observed in the femur as shown in Fig2 (a). At 500N, femur has 500 micro strain, which broadly agrees with the experimental validation done by M. Papini et al. and with the work of Brian Geer et al. Also the strain pattern is broadly in agreement with the work done by Duda et all for their simplied loading condition for the midshaft section in intact condition. This validates our experimental setup as well as to the extent that the bone selected for this work is representative. The surface strain on the tibial bone under similar loading conditions was found to be only 50% of the strain value of femur with a maximum strain of 300ustrain at 1.5kN as shown in Fig 3(a)
The longitudinal strains taken on femur and on tibia in the proximal region is less than the other regions and the shear strain is nearly 50% of the lateral strain.
[FIGURE 2 OMITTED]
Compressive loading with implant without a fracture
This loading condition without any crack shows strain reaching upto 500microstrain in the midshaft region that agrees with the experimental validation done by G.Cheung et all as shown in Fig2 (b). Femur when loaded at 1500N, 350microstrain above and below the crack and for 500N it is 100microstrain above and below the crack was observed (Fig. 2b, Fig.2c). In tibia for 1500N it is 250microstrain above the crack and 230 microstrain below the crack and at 500N it is 50microstrain below and above the crack (Fig.3a, 3b, 3c). The shear and longitudinal strains in the femur and tibia also follows the same behaviour.
Compressive loading with implant in fractured condition
When a crack was introduced, predictably the strains got reduced. This is because as the crack size increases the stress on the surface of the bone is relieved. This is further emphasized by the lower strains found in 60% crack model. Since the 90% crack model is a comminuted a discontinuous, the strain values vary. A sharp decrease in the strain values is noted which signals that as the crack grows the effect becomes more prominent. Thus the implant starts sharing the load and which may show the effectiveness of the implant.
In femur and tibia it is observed that the longitudinal strain is 50% of lateral strain and the shear strain remains in the same level of lateral strain. In this condition the strains below the crack is found to around 50% less than those above the crack. Also another interesting observation was, as the load increases the strain movement was around a narrow band of 50microstrain only in all the three directions. It is also found that the strain band lies between 25microstrain to 50microstrain for all sizes of cracks during the loading and unloading cycles.
[FIGURE 3a OMITTED]
[FIGURE 3b OMITTED]
[FIGURE 3c OMITTED]
The cadaveric femur bone was implanted with interlocking nail in vitro conditions. The strains observed in the surface of the femur and tibia bones in the implanted condition where nearly 60-70% of the bones loaded without an implant. This shows that a percentage of load is shared by the implant. Different crack sizes were initiated as 30%, 60%, and 90% with respect to the diameter of the bone in the midshaft zone of the bones to study the strain pattern under these conditions.
The behaviour of femur and tibia under various fractured condition has brought unique strain patterns for various physical loads. The strains observed in femur without implantation ranged from 1000microstrain to 1500microstrain in the loading zone of 1000-1500N.
In the comminuted fracture condition of the bone an unstable strain pattern is observed in both femur and tibia. Whereas during the fracture consolidation its observed that strain becomes stable and a definite pattern is observed. Moreover during this process the surface strain increases which shows that healing progress as well as the effectiveness of the implant. Strains observed above the fracture are 50% more than those observed below the fracture, but as the fracture consolidates this difference decreases.
From Table 1 one find that 50% more lateral strain in femur than in tibia in minor cracks but as crack size increases tibial has more strain in longitudinal and shear strain. In case of minor cracks the loading and unloading is stable but as crack size increases this band increases. Tibia becomes more unstable than femur when the crack opening is larger .Its found that strains in both above and below the cracks the strain is nearly 50% less in below than above and is observed to be the same in longitudinal and shear strain. Performance Evaluation of the Implant The strain values are found to be around a band of 200 to 400 microstrain for femur and 50 to 200microstrain for tibia at 1000N, above and below the fracture area respectively. As the implant has the ability to share the load, progressive weights can be given. The work of M.Papini et al. do not consider fracture, are therefore relevant for the post fracture consolidation period only .
The performance graph (fig 4) brings out that the bone surface strain observed above and below the fracture reduces as the fracture size increases. The strain below the fracture is found to be nearly 50% of the strain found above the fracture .The same trend is seen in tibia as well but as the crack size increases there is instability of the bone under loading which showed erratic strain values for the 90% fractured condition. The differences between the strains found at the maximum crack size and the one without a crack was nearly 50%. This band difference and the parallelism between the curves of the strain values for above and below the fracture will in essence define the effectiveness of the implant.
[FIGURE 4 OMITTED]
The strain results shows that the intramedullary nail used by us, manufactured locally in India implanted were found to be good load sharing implants when the results of this work are compared with comparative literatures. This study demonstrates strain pattern that is found around the crack site during the fracture consolidation phase for various physical loads. These results are limited by the fact that the experiments were carried out on one representative cadaveric femur and tibia bone and the muscle loading has been neglected. If the muscle reaction forces had been accounted too there would be a proportional reduction in the strain measurements as the surface strains are overestimated under simplified loading of the tibia . These results also bring insight into loading conditions during the fracture consolidation.
The performance graph in fig 4 shows the simple methodology for performance evaluation, the primary objective of this work. The conclusion from this graph is that, if the difference of the measured surface strains between that found without fracture and with maximum fracture is greater than 50% and if the curve for the strain values above and below is more parallel the implant can be safely concluded to be highly effective.
Similar studies can be undertaken for other implants as well. These simplified evaluation procedures can be used for validating the implants and orthopaedic Surgeons and patients will have confidence on the implants they use for their surgical procedures.
The results of these studies will also be a lot helpful for further improvising on this experimental work and can be used for validating the implant design at design stage itself using FEM softwares, which shall then be used for optimizing the current implant design.
Acknowledgement: Authors would like to gratefully acknowledge GKNM Hospital, Coimbatore, India, STAAN biomed (P) ltd., Coimbatore, India & Welding Research Institute, Trichy for their support to use their facilities. We are greatly indebted to Karunya Deemed University, Coimbatore, India for funding and their motivation to undertake this work.
Received 6 July 2007; published online 1 July 2008
(1.) M.Papini, JK Spelt et all " Finite element analysis of a femoral retrograde intrameduallary nail subject to gait loading", Med. Engineering & Physics 26 (2004) 93-108
(2.) R.Neiman, S.J.Hazelwood et all "strain analysis of the proximal femur after retrograde femoral nailing: A 3 dimensional finite element study" 25th ASB conference, 2009.Vlad Morcovesu "the stress values numerical computing in the proximal femur under physiological load conditions ","Zilele Academice Timisene" Symposium, "Biomechanics Section"
[4.] Georg N Duda et al "Influence of muscle forces on femoral strain distribution", Journal of biomechanics 31 (1998) 841 846
[5.] Kessler SB, Hallfeldt KK, Perren SM, Schweiberer L." The effects of reaming and intramedullary nailing on fracture healing, Clin Orthop Relat Res. 1986 Nov; (212): 18-25.
[6.] Sitthiseripratip K, Van Oosterwyck H, Vander Sloten J, Mahaisavariya B, Bohez EL, Suwanprateeb J, Van Audekercke R, Oris P." Finite element study of trochanteric gamma nail for trochanteric fracture, Med Eng Phys. 2003 Mar; 25(2): 99-106.
[7.] Schneider E, Michel MC, Genge M, Zuber K, Ganz R, Perren SM, "Loads acting in an intramedullary nail during fracture healing in the human femur", Journal of biomechanics 2001,Jul; 34(7): 849-57.
[8.] Obara T., A biomechanical study on the fracture treatment--intravital measurement of the strain on an intramedullary nail in the healing process of the femoral fracture in goats (author's transl), Nippon Seikeigeka Gakkai Zasshi. 1979 Feb; 53(2): 199-212.
[9.] Harper, M.C., 1985 Fractures of the femur treated by open and closed intramedullary nailing using fluted rod. Journal of Bone and joint surgery 67-A, 669-708
[10.] Klemm KW, Borner M. Interlocking nailing of complex fractures of the femur and tibia. Clin Orthop Relat Res. 1986 Nov; (212):89-100.
[11.] Duda GN, Mandruzzato F, Heller M, Goldhahn J, Moser R, Hehli M, Claes L, Haas NP. Mechanical boundary conditions of fracture healing: borderline indications in the treatment of unreamed tibial nailing. J Biomech. 2001 May;34(5):639-50.
[12.] S.K.Senapathi, S.Pal, Finite element analysis: an effective tool for prostheses design, Trends in biomaterials artificial organs Vol17 (2) pp141-148, 2004
[13.] Brian Greer, Eric Wang and Yanyae Jiang, on the appropriation of using standardized femur for FEA in the proximal region
[14.] Georg N.Duda, Francesco Mandruzzato, Markus Heller, Jorg Goldhahn et all, Mechanical boundary conditions of fracture healing: borderline indications in the treatment of undreamed tibial nailing, Journal of biomechanics 34(2001) 639-650
D. Davidson Jebaseelan (a, #), N. Mathiarasu (b), Sudhakar (c), Clement Joseph (d), N. Raju (e), D.G. Harris Samuel (f)
(a) School of Mechanical Sciences, Karunya University, Karunya Nagar, Coimbatore 641 114
(b) Department of Bioengineering, University of Illinois at Chicago, USA
(c) 02-04 ME (cad/cam) c/o KSMS, KITS, Karunya Nagar, Coimbatore 641 114
(d) GKNM Hospital, PN Palayam, Coimbatore 641 037
(e) Stress Analysis Group, Welding Research Institute, Trichy 620 014
(f) Kings Engineering College, Chennai 602 105
(#) corresponding author e-mail: email@example.com
Table 1: Comparison of BMD of a live and the cadaveric bone used in the experiment studies BONE MINERAL DENSITY Region along CADAVERIC BONE the bone LIVE BONE Femur Tibia Head Top 0.53 0.548 0.464 Neck 0.762 0.727 0.637 Below Neck 0.817 0.997 0.796 Center 1.127 1.154 1.031 Below Center 1.26 1.282 1.086 Bottom Most 1.272 1.255 1.006 Table 1 list out the strain data found in femur and tibia Femur--strain in ustrain Above 0[degrees] 45[degrees] 90[degrees] No crack 380 200 400 30% crack 350 150 200 60% crack 320 125 200 90% crack 220 100 Unstable Femur--strain in ustrain Below 0[degrees] 45[degrees] 90[degrees] No crack 270 100 200 30% crack 250 100 100 60% crack 240 150 100 90% crack 210 100 -- Tibia--strain in ustrain Above 0[degrees] 45[degrees] 90[degrees] No crack 180 400 -- 30% crack 200 250 75 60% crack 100 100 70 90% crack 50 200 200 Tibia--strain in ustrain Below 0[degrees] 45[degrees] 90[degrees] No crack 200 100 -- 30% crack 150 150 100 60% crack 140 100 60 90% crack 80 400 150
|Printer friendly Cite/link Email Feedback|
|Author:||Jebaseelan, D. Davidson; Mathiarasu, N.; Sudhakar; Joseph, Clement; Raju, N.; Samuel, D.G. Harris|
|Publication:||Trends in Biomaterials and Artificial Organs|
|Date:||Jul 1, 2008|
|Previous Article:||pH sensitive succinyl chitosan microparticles: a preliminary investigation towards oral insulin delivery.|
|Next Article:||Experimental determination of material properties of cortical cadeveric femur bone.|