SPINE Volume 31, Number 4, pp 406 – 413 ©2006, Lippincott Williams & Wilkins, Inc. Inﬂuence of Screw Positioning in a New Anterior Spine Fixator on Implant Loosening in Osteoporotic Vertebrae Maximilian Reinhold, MD,* Karsten Schwieger, PhD,† Joerg Goldhahn, MD,‡ Berend Linke, PhD,† Christian Knop, MD,* and Michael Blauth, MD* Study Design. A biomechanical study was designed to assess implant cut-out of three different angular stable anterior spinal implants. Subsidence of the implant rela- tive to the vertebral body was measured during an in vitro cyclic loading test. Objectives. The objective of the study was to evaluate two prototypes (Synthes) of a new anterior spine ﬁxator with different screw angulations in comparison to the established MACSTL® Twin Screw Concept (Aesculap). The inﬂuence of factors like load-bearing cross-sectional area, screw angulation and bone mineral density upon implant stability should be investigated. Summary of Background Data. Epidemiologic data predict a growing demand for appropriate anterior spinal ﬁxation devices especially in patients with inferior struc- tural and mechanical bone properties. Although different concepts for anterior spinal instrumentation systems have been tried out, implant stability is still a problem. Methods. Three angular stable, anterior spinal im- plants were tested using 24 human lumbar osteoporotic vertebrae (L1-L5; age 84 (73–92)): MASC TL system (Aes- culap); prototype 1 (MP1) with 18° and prototype 2 (MP2) with 40° screw angulation (both Synthes). All implants consisted of two screws with different outer screw diam- eters: 7-mm polyaxial screw with 6.5-mm stabilization screw (MASC TL), two 5-mm locking-head screws each (MP1 and MP2). Bone mineral density (BMD) and verte- bral body width of the three specimen groups were evenly distributed. The specimens were loaded in cranio- caudal direction (1Hz) for 1000 cycles each at three con- secutive load steps; 10 –100 N, 10 –200 N and 10 – 400 N. During cyclic loading subsidence of the implant relative to the vertebral body was measured in the unloaded condi- tion. Cycle number at failure (deﬁned as a subsidence of 2 mm) was determined for each specimen. A survival analysis (Cox Regression) was performed to detect differ- ences between implant groups at a probability level of 95%. Results. High correlations were found between BMD and number of cycles until failure (MP1; r = 0.905, P = 0.013; MP2: r = 0.640, P = 0.121; MACS TL: r = 0.904, P = 0.013) and between load bearing cross sectional area and number of cycles until failure (MP1: r = 0.849, P = 0.032; MP2: r = 0.692, P = 0.085; MACS TL: r = 0.902, P = 0.014). Both Prototypes survived signiﬁcantly longer than the MACS TL implant (MP1: P = 0.012, MP2: P = 0.014). The survival behaviour of MP1 and MP2 was not signiﬁcantly different (P = 0.354). Conclusions. Implant stability within each implant group was inﬂuenced by BMD and load bearing cross- sectional area. The angulation of the two screws did not have a signiﬁcant inﬂuence on cut-out. As conclusion from this study, promising approaches for further implant development are: 1) increase of load-bearing cross-sec- tional area (e.g., larger outer diameter of the anchorage device), 2) screw positioning in areas of higher BMD (e.g., opposite cortex, proximity to pedicles or the endplates). Key words: biomechanics, implant performance, osteo- porosis, spinal fusion, fracture. Spine 2006;31:406 – 413 Vertebral fractures as a major clinical consequence of osteoporosis are becoming more frequent in an aging population. Epidemiologic data of patients with osteo- porotic vertebral fractures underline the importance of fracture stabilization on health-related quality of life, back pain, and disability in everyday life. 1–3 Especially in osteoporotic thoracolumbar spine frac- tures, conventional anterior spinal implants are still prone to failure by implant pull- or cut-out. 4–6 Therefore, en- hancing the strength of the implant– bone interface is essential and mandatory for the development of new an- terior implants for endoscopic approaches with similar reduction and stabilization capabilities compared with conventional dorsal internal ﬁxators. Different concepts and ideas have been tried out and evaluated to accomplish this goal: various implant de- signs, e.g., blade-type implants 7 or hollow screws to pro- mote bony ingrowth, 7–10 modiﬁcation of various geomet- rical screw parameters (cross-sectional area, diameter, thread, cannulation, and pitch 11–13 ), mono versus bicorti- cal screw purchase, 7,14 and different screw locking mech- anisms. 15–18 Huang et al analyzed various conﬁgurations of ante- rior vertebral double screw ﬁxation and found that tri- angulation of two anterior vertebral screws (22° inter- section angle) without penetration of the cortex achieved pullout strengths similar to that of two-parallel double- cortical screws. 19 Locking-head screws act mechanically as an angular stable, unitary implant with its connecting device and have signiﬁcantly increased the rigidity of unicortical screw-plate systems initially and after cyclic loading in cervical vertebrae. 17 Therefore, the subject of our investigation was the testing of the anchoring elements of three angular stable From the *Innsbruck Medical University, Department of Trauma Sur- gery and Sports Medicine, Innsbruck, Austria; †AO Research Institute, Davos, Switzerland; and ‡Schulthess Clinic, Research Department, Zurich, Switzerland. Acknowledgment date: August 2, 2004. First revision date: January 23, 2005. Acceptance date: March 1, 2005. The device(s)/drug(s) that is/are the subject of this manuscript is/are not FDA-approved for this indication and is/are not commercially avail- able in the United States. Institutional funds were received in support of this work. No beneﬁts in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Address correspondence and reprint requests to Maximilian Reinhold, MD, Innsbruck Medical University, Department of Trauma Surgery and Sports Medicine, A-6020 Innsbruck, Anichstrasse 35, Austria. E- mail: maximilian.reinhold@uklibk.ac.at 406