Journal of Applied Biomechanics, 2012, 28, 568-578 © 2012 Human Kinetics, Inc. An Official Journal of ISB www.JAB-Journal.com ORIGINAL RESEARCH 568 Vasileios I. Sakellariou (Corresponding Author), Andreas F. Mavrogenis, George C. Babis, Panayiotis N. Soucacos, and Panayiotis J. Papagelopoulos are with the First Department of Orthopaedics, Athens University Medical School, Athens, Greece. Evangelos A. Magnissalis is with the Laboratory for the Research of Musculoskeletal System, University of Athens, Athens, Greece. Comparison of Four Reconstructive Methods for Diaphyseal Defects of the Humerus After Tumor Resection Vasileios I. Sakellariou, 1 Andreas F. Mavrogenis, 1 George C. Babis, 1 Panayiotis N. Soucacos, 1 Evangelos A. Magnissalis, 2 and Panayiotis J. Papagelopoulos, 1 1 Athens University Medical School; 2 University of Athens The objective of the current study was to compare quantitative data on the biomechanical analysis of different techniques for fxation of intercalary bone defects of the humerus, by means of consistently applied methodology on composite models. A total of 25 humeral specimens of composite models were used. An intercalary defect was created and reconstructed using plates, intramedullary nails, external fxators and segmental prosthetic implants. The specimens were loaded under axial compression, four-point bending and torsion within the linear elastic region. Modular segmental implants and intramedullary nails were able to compensate signifcantly greater amounts of compressive loads compared to locking plates and external fxators. However, in fexion and torsion, the modular segmental implants and the external fxators were signifcantly better load-bearing devices compared to the intramedullary nails and plates. Early mobilization of the upper limb in patients with diaphyseal bone defects of the humerus could probably be more safe and tolerable when reconstructed with modular segmental implants. Keywords: intercalary, diaphyseal, defects, biomechanics, endoprosthesis Reconstruction of large diaphyseal bone defects in long bones is generally addressed by several treatment options which are generally divided into bone preserv- ing and segmental bone substituting techniques (de Pablos et al., 1994). Bone-preserving techniques involve vascularized or nonvascularized autografts (Chang & Weber, 2004; de Pablos et al., 1994; Nair et al., 2009), allografts—stabilized with plates (Fuchs et al., 2008; Virkus et al., 2008; Wingerter et al., 2007), intramedullary nails (Fuchs et al., 2008; Virkus et al., 2008; Wingerter et al., 2007)—and distraction osteogenesis using unilateral rail external fxators (de Pablos et al., 1994) or Ilizarov apparatus (Fuchs et al., 2008; Hasenboehler et al., 2006; Meffert et al., 2000). Bone-substituting techniques include the application of segmental diaphyseal implants (Aldlyami et al., 2005; Fuchs et al., 2008; Henry et al., 2002) or interposition of metallic spacers (Bullens et al., 2009a, 2009b; Fujibayashi et al., 2003; Lindsey et al., 2006) or scaffolds [Kuzyk et al., 2009; Nair et al., 2009). It is generally thought that using an endoprosthesis the operated limb could be ambulated more safely and rapidly (Fuchs et al., 2008; Henry et al., 2002) comparing to reconstruction with allografts which necessitates for more cautious primary mobilization until allograft incor- poration is evident (Fuchs et al., 2008; Pollock et al., 2005). In the existing literature there are only few reports regarding the biomechanical evaluation of endoprosthe- ses or the internal fxation of intercalary defects (Heck et al., 2010; Henry et al., 2002). However, there is no study comparing simultaneously all these reconstructive options on the same composite model with specifcally sized bone defect. The hypothesis of our study was that modular seg- mental implants should have better biomechanical proper- ties comparing to internal and external fxation techniques. The purpose of our study was to verify this hypothesis by developing a biomechanical project consisting of humeral composite models with specifc mid-diaphyseal defects that were reconstructed using four different fxation options (modular intercalary implants, interposition of allograft and fxation with locking plates, intramedullary nails or external fxators). By submitting these constructs to dif- ferent modes of loading we intended to reveal potential statistically signifcant differences among the collected data of axial, bending and torsional loads, axial stiffness, fexural and torsional rigidity of each construct. Methods Fourth-generation bone composite models (Sawbones Europe AB, Pacifc Research Laboratories, Inc. Sweden) were selected instead of cadaveric bones as they simulate