SPINE Volume 33, Number 22, pp 2387–2393 ©2008, Lippincott Williams & Wilkins A Repeatable Ex Vivo Model of Spondylolysis and Spondylolisthesis Katie Beadon, MASc,* James D. Johnston, MSc(Eng),* Kevin Siggers, MASc,*† Eyal Itshayek, MD,‡ and Peter A. Cripton, PhD* Study Design. An ex vivo biomechanical study using porcine spinal segments. Objective. To produce a biomechanical model of both spondylolysis and spondylolisthesis using an accelerated cyclic loading model with intermittent impulse loads. Summary of Background Data. Only a few models of spondylolisthesis appropriate for biomechanical testing have been presented previously. Past modeling attempts have largely required nonphysiologic gross fracture of the pars before testing and have resulted in nonphysi- ologic endplate fracture. In these tests no clinically rele- vant spondylolisthesis was seen at the end of testing. A reproducible, clinically relevant model of both spondylol- ysis and spondylolisthesis would allow study of these disease processes, and facilitate the development and evaluation of advanced spinal implants optimized specif- ically for these pathologies. Methods. Five porcine lumbar functional spinal units were tested (2 L4 –L5, 3 L6 –S1) after small notches had been created in the pars and after the disc had specific collagen fibers in the anterior anulus sectioned. Speci- mens were loaded with a constant cranial-caudal com- pressive force of 300 N and the application of cyclic an- terior shear loads between 300 and 600 N with intermittent impulse loads to 1500 N until pars fracture occurred. Elevated cyclic loading then continued between 500 and 800 N. Results. All specimens displayed bilateral pars frac- ture with the fractures passing through the points of notching and no damage to endplates or facet joints. Clinically-relevant Grade II spondylolisthesis was achieved in all 5 specimens. The mean slip at the con- clusion of testing was 33%. Conclusion. Cyclic shear loading with intermittent im- pulse loads can reliably create fracture in the pars inter- articularis in ex vivo porcine spine segments. Subsequent cyclic anterior motion of the superior vertebra results in clinically-relevant spondylolysis and spondylolisthesis. Key words: spondylolysis, spondylolisthesis, cyclic, ex vivo, lumbar spine, biomechanical model. Spine 2008;33: 2387–2393 Fracture of the neural arch at the pars interarticularis, known as spondylolysis, 1 is the most common serious injury to the low back of athletes. 2 Spondylolysis (lysis) occurs in 5% of white North Americans 3,4 but is present in as many as 47% of adolescent athletes with low back pain. 5 Eighty percent of patients with spon- dylolysis also have anterior slippage of the superior vertebra relative to the subjacent vertebra, known as spondylolisthesis (listhesis). 6 While most only show minor slip, one fifth show slip of over 25% (defined as the displacement of the cranial vertebra with respect to the subjacent vertebra as a percentage of the subjacent vertebra’s AP width), which can lead to severe pain and neurologic complications. 6 Lysis and listhesis primarily occur in the lower lumbar re- gions where lumbar motion and shear loads are highest 7,8 and are hypothesized to be related to cyclic loading, 8,9 repetitive microtrauma of the pars interarticularis, 10 increased sacral in- clination 11–13 and deficiencies in the disc, 14 –17 facets 18 and pars interarticularis. 9,14 –16 Additionally, the amount of ini- tial slippage at the onset of lysis is thought to be directly related to the severity of listhesis. 19 Meyerding’s spon- dylolisthesis grading system 10 classifies the severity of listhesis based on the magnitude of forward slippage rel- ative to the anterior-posterior (AP) width of the subja- cent vertebra: Grade I listhesis is defined as slippage not exceeding 25%; Grade II is defined as slippage between 26% and 50%. Flexion, extension and anteroposterior shear forces have been suggested as causative factors in the develop- ment of lysis and listhesis 20 –22 and previous biomechani- cal studies have attempted to create a model of listhesis using combinations of these loading protocols. In order to simulate the risk factors that may predispose the spine to listhesis, some previous studies have experimentally weakened or sectioned the intervertebral disc 6,19,23–25 and thinned or fractured the pars 6,19,23,24,26 before test- ing. The majority of these studies applied a quasi-static shear loading regime, and this most often resulted in endplate fracture. 6,19,26 Recently, Patwardhan et al 24 in- duced high Grade I listhesis in human ex vivo specimens using a cyclic shear model. In this study, 24% slip was seen at shear loads of over 1200 N, but this decreased to 17% when loading was removed. A functional model of spondylolysis and spondylolis- thesis is necessary to isolate the loading configurations and predispositions that may lead to these conditions. In this way, screening procedures and interventions maybe developed to prevent them. A successful model may also From the *Injury Biomechanics Laboratory and Division of Orthope- dic Engineering Research, Departments of Mechanical Engineering and Orthopedics and International Collaboration on Repair Discoveries (ICORD), †Department of Materials Engineering, and ‡Department of Orthopaedics, Division of Orthopaedic Spine Surgery, University of British Columbia, Vancouver, Canada. Acknowledgment date: August 29, 2007. First revision date: April 12, 2008. Acceptance date: May 26, 2008. The manuscript submitted does not contain information about medical device(s)/drug(s). No 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 Peter A. Cripton, PhD, Department of Mechanical Engineering, University of British Colum- bia, 6250 Applied Science Lane, Vancouver, BC, Canada, V6T 1Z4; E-mail: cripton@mech.ubc.ca 2387