SPINE Volume 33, Number 18, pp 1958–1965 ©2008, Lippincott ...1.step.nl/...a_Lumbar_Intervertebral_assym_tillen.pdf · intervertebral disc specimens. Lateral bending and ﬂex-ion

SPINE Volume 33, Number 18, pp 1958–1965©2008, Lippincott Williams & Wilkins

Biomechanical Response of a Lumbar IntervertebralDisc to Manual Lifting ActivitiesA Poroelastic Finite Element Model Study

Raghu N. Natarajan, PhD,*† Steve A. Lavender, PhD,‡ Howard A. An, MD,*and Gunnar B. J. Andersson, MD, PhD*

Study Design. Determination of damage to a lumbardisc caused by lifting using a poroelastic finite elementmodel study.

Objective. Compare the biomechanical response of alumbar disc under 8 different lifting conditions and iden-tify the loading conditions that produce the greatest de-formations and highest stresses in various tissue compo-nents of the disc.

Summary of Background Data. Lifting has been associatedepidemiologically with back injuries and back pain complaints.Forces high enough to cause mechanical damage to the dischave been predicted using analytical methods.

Methods. Using kinematic data from series of lifts ob-tained in our laboratories and an electromyography(EMG)-driven muscle optimization model, representativeforce patterns for each of 8 lifts were obtained and usedas input into a poroelastic finite element model. Disctissue displacements and stresses were determined for allthe 8 lift conditions.

Results. Lifting the box from a maximally lateral flexedposture up to waist level (Task 5) produced the largesttranslational and rotational motions of all studied. This liftactivity also produced maximum von Mises stresses in alldisc components: annulus, nucleus, and endplates. Thelargest facet joint forces were also observed during liftTask 5.

Conclusion. Asymmetric lifting involving lateral bend-ing of the trunk produced large motions that might causelocalized disc tissue injury. Stresses larger than the failurestrength of the corresponding disc tissues were experi-enced during asymmetric lifting. Lifting that involves lat-eral bending of the trunk was the most hazardous type ofloading with regard to damage to the disc.

Key words: poroelastic finite element model, biome-chanics, lumbar spine, manual material handling. Spine

2008;33:1958–1965

Low back disorders (LBDs) affect up to 47% of workersyounger than 45 years of age and who perform physi-cally demanding jobs.1–3 Retrospective studies4,5 of in-

dustrial injuries have identified manual material han-dling (MMH) as the most common cause of LBD. It isestimated that lifting and MMH account for 50% to75% of all back injuries.4–6 Unfortunately our knowl-edge on the relationship between loads acting on thespine during lifting and the occurrence of disc injuries isnot well defined. Understanding the mechanical responseof the spine to various complex loading conditions thatarise during manual lifting is necessary to understand therelationship between MMH and the associated risk offailure in various parts of the disc tissue.

Van Dieen et al7 using 10 healthy male subjects sug-gested that the variability of spinal compression duringlifting needs to be taken into account while designinglifting tasks. Davis and Marras8 studied the contributingrole of biomechanical, psychosocial, and individual riskfactors among 60 college-aged individuals and con-cluded that the weight of the load is the most importantfactor when controlling compression forces during lift-ing. An epidemiologic case-control study done by Kelseyet al9 showed that people in jobs involving lifting ofobjects weighing more than 25 lbs with the body twistedand knees not bent are at high risk for prolapsed lumbardiscs. Marras et al10 performed an in vivo study andconcluded that by properly varying the lifting frequency,load moment, trunk lateral and twisting velocities, andtrunk sagittal angle during occupational lifting, individ-uals can reduce the risk of lumbar disc injury. Videmanand Nurminen11 studied annular tears in relation to life-time frequency of back pain and concluded that in earlyadulthood, the presence of annular tears is high in peoplewith back pain. At retirement age, tears were practicallyunavoidable. Stokes and Iatridis12 reviewed the litera-ture on macromechanical factors that accelerate disc de-generation and concluded that any abnormal loadingconditions can produce tissue trauma and/or adaptivechanges that may result in disc degeneration.

Adams and Hutton13 showed that fatigue loading incompression and bending can cause gradual prolapse ofthe disc with the formation of radial fissures. Gordon etal14 investigated disc prolapse under physiologically rea-sonable repetitive loads and found disc failure invariablyoccurred either through annular protrusion or nuclearextrusion through annular tears. Adams et al15 appliedcontrolled compressive load with moderate repetitiveloading appropriate for simulation of light manual laborand found minor damage to the vertebral endplate thatlead to progressive structural changes in the adjacent

From the *Rush University Medical Center, Chicago, Illinois; †Univer-sity of Illinois at Chicago, Chicago, Illinois; and ‡Ohio State Univer-sity, Columbus, Ohio.Acknowledgment date: June 21, 2007. Revision date: December 21,2007. Acceptance date: March 12, 2008.Supported by grant NIH AR48152–02.The manuscript submitted does not contain information about medicaldevice(s)/drug(s).Federal funds were received in support of this work. No benefits in anyform have been or will be received from a commercial party relateddirectly or indirectly to the subject of this manuscript.Address correspondence and reprint request to Raghu N. Natarajan,PhD, Department of Orthopedic Surgery, 1653 West Congress Park-way, Suite 764-, Armour Academic Facility, Chicago, IL 60612-3833;E-mail: [email protected]

1958

discs. Videman et al16 examined the functional spinalunits of 86 cadavers whose work and LBD history wereknown. They found increased degeneration in the spinesof those specimens who had performed physically heavywork. The physiologic rotations and displacements thatmay place the disc at greatest risk for large tissue strainsand injury were identified by Costi et al17 using humanintervertebral disc specimens. Lateral bending and flex-ion produced the highest physiologic maximum shearstrain. Schmidt et al18 using a finite element model of alumbar motion segment concluded that the risks of discfailure and prolapse was particularly high in a load com-bination of lateral bending and axial rotation.

The literature fails to provide insight on the distribu-tion of stresses in various components of the disc, whichare essential to understand the relationship between loadand disc injury. The aim of this investigation was to usea finite element model of a human lumbar disc that mim-ics as accurately as possible the biomechanics of an invivo disc19 and determine the biomechanical character-istics when common lifting activities are performed. Theobjective was to analyze the biomechanics of a lumbardisc under 8 different lifting conditions and identify themost hazardous type of loading with regard to damage tothe disc. The hypothesis is that more damage to the discwill occur due to asymmetric lifting tasks than in sym-metric lifting.

Methods

The geometric shape of a lumbar motion segment was gener-ated from a serial computed axial tomographic scan (CT) of anL4–L5 disc body unit. Using this CT scan a 3-dimensionalfinite element model was generated for the motion segment(Figure 1) consisting of vertebra-disc-vertebra unit using aCAD station. The cortical bone, cancellous bone, posterior el-ements, endplates, facet cartilage, and nucleus pulposus weremodeled as 8-node, 3-dimensional, isoparametric elements.The left and right superior and inferior articulating surfaces ofthe facet cartilage were approximated by flat trapezoidal mov-ing frictionless contact surfaces. In the intervertebral disc, the

annulus matrix was assumed as a composite material consist-ing of fibers embedded in a homogenous matrix material. Thematrix was discritized by 8-node, 3-dimensional elements. Theannular fibers were assembled in a criss-cross fashion at anangle approximately 30° to the transverse plane and were mod-eled as 2-node, nonlinear truss elements that reacted to tensiononly. The normal nucleus, which is a gelatinous material, wasrepresented by 3-dimensional fluid elements. The 7 major lig-aments were modeled by 2-node nonlinear cable elements andtheir attachment points were taken directly from the literature.This existing L4–L5 lumbar motion segment model was mod-ified19 to include physiologic parameters such as swelling pres-sure and the effect of change in permeability of the disc tissuesdue to strain in addition to conventional poroelastic parame-ters such as pore pressures. The analyses were conducted usinga commercially available software ADINA.20 Detailed descrip-tion of the finite element model in terms of types of elementsused and description of properties of materials associated withvarious components in the disc-body-disc unit have alreadybeen published21,22 and hence only a brief summary will bepresented here. The effect of change in proteoglycan concentra-tion within the disc tissue that occurs due to fluid flow in andout of the disc created by externally applied load was charac-terized in the model by the inclusion of a pressure Pswell. Changein the size of the pores in the disc tissue due to externallyapplied load, which affects the fluid flow in and out of the disctissue, was included in the model by an additional pressureterm Ppermeability. The model was further refined22 by includingvariations in proteoglycan concentrations from the outer an-nulus to the inner annulus and in the nucleus, as well as thevariation of the porosity, permeability, and aggregate moduluswithin the disc tissue.

To show the level of validation achieved by this model, 3different analyses were conducted using this model: long-termcreep compression (circadian variation), short-term creep com-pression, and short-term cyclic loading. Diurnal changes intotal stature obtained from the poroelastic finite element modelwere compared with changes in total stature measured on 8young adults and available in the literature,23 and the resultswere similar. Percent loss of total stature predicted by the cur-rent FEM was also compared for short-term creep and cyclicloading with in vivo results23 and again the results were simi-lar.22 Total stature changes at the end of short-term loadingand unloading and short-term cyclic loading were within 4% ofthe corresponding in vivo measurements. Similar close compar-ison of results was observed during calculation of circadianvariation of lumbar disc height. Details on the validation of thecurrent model are available in our recent publication.22

The load data used in the current study were obtained froma series of 48 lifts performed by each subject: eight subjectsperformed all 8 lift activities shown in Table 1 6 times. For eachtask, a 30 lb box (13.6 kg) was lifted. Thirty pounds (13.6 kg)was selected as it is close to the mean item weight in a grocerydistribution center. The first 2 lifts served as practice to ensurethat the desired motions were produced. After a short rest pe-riod, the participants were asked to lift the object 4 times at anormal work pace. A muscle optimization model using EMGand kinematic data collected in our lifting laboratory wasused24 to calculate spinal loads as a function of time. Whileperforming a specific lift activity, considerable variability in themuscle recruitment pattern can be expected among the partic-ipants. Based on our previous EMG testing, for a specific liftactivity, it was possible to identify the most representative mus-

Figure 1. Finite element model of the lumbar spine (L4/5) used forthe analyses.

1959Biomechanical Response of a Lumbar Intervertebral Disc • Natarajan et al

cle force pattern for use in the model study. A single individu-al’s force data were selected for the specific lift activity by ran-domly sampling from the pool of individuals that used the mostfrequently observed muscle force pattern. The force data wereused in the finite element analyses. The time history data ofspinal loads were input into the poroelastic finite elementmodel to determine the biomechanics of the disc during the 8specific lifting activities considered in this study.

The task begins with the subject in an upright neutral pos-ture. The compression force at the beginning of the trial repre-sents the body weight and any compressive forces induced bythe nominal trunk activity used to maintain the standing pos-ture. Maximum shear force of 150 N was observed actingalong the lateral direction (Figure 2A) except during Task 5.During Task 5, which involves lateral bending of the trunk, amaximum lateral shear of 750 N was observed (Figure 2A). Inalmost all lifting conditions, maximum shear force along theanterior-posterior direction (Figure 2B) varied from 250 to 500N. During lifting Tasks 2 and 5, maximum compressive forceof 4500 N on the motion segment was observed (Figure 2C). Inall the other lift conditions, maximum compressive force variedfrom 2000 to 3000 N.

Results for disc displacements and corresponding rotationsin all the 3 principal directions were determined during theentire duration of lift activity. Effective stresses in the annularportion of the disc, stresses in the nucleus, annular fiberstresses, endplate stresses, and facet forces were obtained fromthe finite element model study for all 8 lift conditions.

Results

In all lifting activities, motions were largest at the pointwhere the load on the spine was the greatest. A compar-ison of this maximum biomechanical response between

different lifting activities will be presented here. The lift-ing activity that required lateral bending of the trunk(Task 5) produced the largest disc translational as well asrotational motions (Figures 3 and 4). The smallest trans-lational and rotational motions of the disc were pre-dicted during a lifting activity that involved lifting withan upright trunk posture (Task 1). Task 5 produced 6.4mm disc compression followed by 5.6 mm disc compres-sion during the lift activity that required extension fromflexed posture (Task 2). All the other tasks producedmuch smaller compression (varied from 2 to 3 mm).Shear motion of the disc directed along the right lateraldirection was 4.0 mm for Task 5, whereas a muchsmaller shear motion was seen during all the other liftactivities (0.5 mm for Tasks 1, 3, 6, 7, and 8 and 1.8 mmfor Tasks 2 and 4). Anteriorly directed shear motion was2.2 mm for Tasks 2 and 5 and 1.6 mm for Tasks 4, 6, and7. A flexion motion of 10.0° was seen during Task 5,whereas Task 2 produced a much smaller flexion of 5.8°(Figure 4). Tasks 4, 6, and 7 produced a flexion motionof 4.0°. All the other lifting tasks produced less than 1° offlexion/extension motion. Maximum torsional motion(4.4°) of the disc once again occurred during Task 5. Allthe other lift activities produced either 1.0° or muchsmaller torsional motion in the disc. Lateral bending ro-tation of the disc was highest in Task 5 (4.0°). All theother lift activities produced less than 1.0° of lateralbending.

In all 8 lift activities considered here, maximum vonMises stress occurred at the interface between the annu-

Table 1. Eight Different Lifting Tasks Considered

Task No. Trunk Posture/Motion Loading Conditions Lifting Task

Task 1 Upright Compression, flexion moment Box is lifted using arms only, upright backposture is maintained

Task 2 Extension from flexed toupright posture

Compression, flexion moment Box is lifted from mid-shank level (10” offthe floor) to elbow height

Task 3 Extension from upright posture Compression, extension moment Box is lifted from elbow height to upperreach limit while standing

Task 4 Twisting Compression,torsional moment

Box is lifted from elbow height on theright side and placed on the left side atelbow height. Total rotation of the box �180°. Subject instructed to maximallytwist spine during this task

Task 5 Lateral bending Compression, lateral bending moment Box is fitted with a handle so that it canbe picked up with one hand. Box islifted from a maximally lateral flexedposture up to waist level

Task 6 Extension from flexed postureand twisting

Compression, flexion moment, torsional moment Subject lifts box from knee level whileflexed, swings box from one side to theother without moving feet, and placesthe box at knee level

Task 7 Extension from flexed posture,twsitin, and lateral bending

Compression, flexion, lateral bending,and torsionalmoments

Like task 6, the subject lifts box from kneelevel while flexed, swings box from oneside to the other without moving feet,and places the box at knee level. Inthis test, the initial and final handpositions are adjusted so that lateralbending is required

Task 8 Extension from upright postureand twisting

Compression, extension and torsional moments Box is lifted from elbow height andplaced to the side at the subject’supper reach limit

1960 Spine • Volume 33 • Number 18 • 2008

lus and the inferior endplate of the motion segment. Lift-ing the box from a maximally laterally flexed posture upto waist level (Task 5) produced the highest von Mises

stresses in all disc tissues, annulus, nucleus, and end-plates. A maximum von Mises stress of 6.5 MPa wasobserved (Figure 5) all along the outer circumference ofthe posterior portion of the annulus in Task 5. Lifts in-volving extension of the torso from flexed to upright

Figure 2. Loading in the 3 principal directions corresponding to the 8 lifting conditions studied here.

Figure 3. The 3 principal translational motions of L4 with respect to L5at peak load corresponding to the 8 lift conditions studied. Maximumresultant displacement of the lumbar disc was seen under Task 5.

Figure 4. The 3 principal rotational motions of L4 with respect toL5 at peak load corresponding to the 8 lift conditions studied.Maximum resultant rotation of the lumbar disc was seen underTask 5.


posture (Task 2) produced the second highest von Misesstress of 4.5 MPa, which occurred in the anterior as wellas posterior quadrant of the annulus. In all the lift con-ditions considered here, maximum von Mises stress inthe nucleus was below 0.5 MPa. Lifting that involvestwisting and lateral bending of the torso (Tasks 4, 5, and7) produced a maximum von Mises stress of 6 MPa in thesuperior endplates along its outer edges. A slightly lowermaximum von Mises stress of 5 MPa in the endplateswas produced by tasks that involved extension of thespine (Tasks 2 and 8).

Stresses at interface between the annulus and endplatewere studied to find out whether any of these 8 liftingmodes can cause rim lesions. In all the lift cases, maxi-mum shear stress occurred in the annulus at the interfacebetween inferior endplate and annulus. Lifting that in-volved lateral bending of the trunk (Task 5) produced ashear stress of 3.0 MPa all along the posterior edge of theannulus (Figure 6). All the other lifting tasks produced amaximum shear stress in the annulus material varyingbetween 1.0 and 2.0 MPa. Stresses in the annular fiberswere also highest in the task that involved lateral bendingof the trunk (Task 5) (Figure 6). Maximum stress (4MPa) occurred in fibers located in the anterior and pos-terior quadrants of the disc during Task 5. Annular fiberstresses between 1.0 and 2.0 MPa were calculated duringall other lifting tasks and they occurred mostly in theouter annular fibers.

The largest facet forces were calculated at the timewhen the load reached its peak value for all 8 lift casesconsidered (Figure 7). Both facets (right and left) carriedabout the similar amount of force. The largest facetforces (350 N) were predicted (Figure 7) during lift Tasks2 and 5, which involved either extension from a flexedtrunk posture or lateral bending of the trunk. Lifts thatinvolved twisting of the trunk (Tasks 4, 6, and 7) pro-

duced a maximum facet force of 240 N. All other tasksproduced facet forces less than 100 N.

At the peak load, fluid exchanges between the nucleusand the surrounding tissues were greater with the adja-cent endplates (in the central region) than with the sur-rounding annular tissues. Fluid flow under Tasks 2 and 5was much higher than during all other tasks (Figure 8), 3times higher in lift Task 2 and 4.5 times higher in liftTask 5.

Conclusion

An attempt has been made in the current study to com-pare the relative motions of a lumbar disc and the asso-ciated stresses within the disc during various types oflifting activities. A quantitative comparison of stresses

Figure 6. Maximum fiber stress and maximum shear stress in theannulus in the L4/5 disc tissues at peak load corresponding to the8 lift conditions studied. Highest shear stress in the annulus andstress in the fiber was seen under Task 5.

Figure 5. Maximum von Mises stress in the L4/5 disc tissues atpeak load corresponding to the 8 lift conditions studied. Higheststress in the annulus was seen under Task 5. Maximum endplatestresses were seen under Tasks 4, 5, and 7.

Figure 7. Facet contact forces distribution corresponding to the 8lift conditions studied. Highest facet forces at peak load wereobserved under Tasks 2 and 5.


induced in the disc components is presented here only toarrive at a qualitative conclusion about which type oflifting is potentially most hazardous to the lumbar spine.

FEM analyses showed that of all 8 lift conditions con-sidered, the task that involved lateral bending of thetrunk (Task 5) produced the highest resultant disc defor-mations (translation of 8.1 mm and rotation of 11.6°).The lifting task that involved extending the trunk from aflexed posture (Task 2) produced the second highest re-sultant disc deformations (6.4 mm translation and a ro-tation of 5.8°). Given the failure strength of the disctissue, a task that involves asymmetric bending of thedisc might cause macromechanical tears12 (because ofhigh disc deformation) if the task was repeated over along period of time. This could initiate or accelerate theprocess of disc degeneration. Task 5, therefore may in-crease the risk of disc injury. This result supports theconclusions reached by Marras et al10 who found thatthe risk of occupationally related LBDs was associatedwith work that required frequent and large sagittal flex-ion angles.

The annulus can be considered as a composite mate-rial. This type of laminated structure requires matrixcracking, delamination, and fiber failure before thewhole tissue fails.12 In this study, lifting Task 5 producedthe highest von Mises stress (6.5 MPa) in the annularmatrix followed by a stress of 4.5 MPa during lifting thatinvolved sagittal bending (Task 2). Annular failurestrength has been reported to vary widely from 3.6 to10.3 MPa.25 Task 5 produces stresses sufficiently high tostart local yield failure (matrix cracking) in the matrixmaterial. If this lift is performed several times in a repet-itive lifting activity over a long period, yielding of thematrix material can propagate.

Animal studies have shown that when rim lesionswere introduced in the annulus, progressive failure of theinner annulus was observed.26 Further from a macro-

scopic study on lumbar motion segments originatingfrom regular autopsies it was found27 that rim lesionsexist due to age-related changes. To determine whetherany of the lift conditions performed repetitively over along period of time can induce such rim lesions, maxi-mum shear stresses in the annular matrix produced bythe 8 lift conditions were studied. The maximum shearstress (0.8 to 3.0 MPa) was higher than the shear failurestresses (which range from 0.4 to 1 MPa).28 The highestshear stress value of 3.0 MPa was seen all along theposterior edge of endplate-annulus interface duringasymmetric lifting (Task 5). Thus, Task 5 might inducerim lesions.

The maximum stresses in the annular fibers variedfrom 0.5 to 4 MPa over the 8 lift conditions. These stresslevels are quite small compared with the failure stressthat varies from 62 to 110 MPa,25 and thus none of thelifting modes is likely to cause annular fiber fracture.

Perey29 and Lotz et al30 have experimentally shownthat fracture of endplates can occur at stress levels of 3 to4 MPa. In the current study, maximum stresses in theendplate varied from 3 to 6 MPa indicating high proba-bility of fracture of endplates in almost all 8 lift condi-tions considered here when such lifts are repeated over along period of time.

The maximum predicted facet forces varied from 200to 350 N under lift conditions 2, 4, 5, 6, and 7. Wilson etal31 have measured facet forces in a lumbar cadaver spec-imen under a moment load of 7.5 Nm and found that theloads varied from 55 to 110 N in axial rotation and10–50 N in extension. These values are consistent withfindings from finite element analyses of Rohlmann etal.32 The facet forces calculated under lift conditionsstudied here are 2 to 3 times higher than facet forcesobserved during normal daily activities. Large facet loadsmean higher pressures on the cartilaginous layers of thefacet joint that might initiate facet joint osteoarthritis orcause its progression. Current analyses showed that inparticular tasks that involve sagittal bending (Task 2) orlateral bending (Task 5) of the trunk during lifting mightinduce high pressures in the cartilage layers of the facetsurfaces and might lead to facet joint osteoarthritis overa period of time if such lifting activities are repeated.

Costi et al17 found that lateral bending and flexionsegmental motions produce physiologic maximum shearstrain comparable with the known failure strain of disctissues. This study also showed that physiologic lateralrotation of the motion segment may place the disc atgreatest risk for large tissue strains and injury. Anotherrecent analytical study by Schmidt et al18 showed thatdisc failure and prolapse mostly occur under a load com-bination of lateral bending and axial rotation, whichonce again agree with the conclusions of this study.

In all 8 lift cases studied, flow of fluid between thenucleus and surrounding tissues predominantly occurredthrough the endplates, which is in agreement with datareported in the literature.33 Asymmetric bending (Task5) produced the largest amount of fluid exchange be-

Figure 8. Distribution of fluid flux in the intervertebral disc at peakload corresponding to the 8 lift conditions studied. Maximum fluidflow at peak load occurred between nucleus and top endplateunder Task 5.


tween nucleus and endplates. If the fluid exchange chan-nels between the nucleus and the endplates are restricted,as is the case with aging, higher intradiscal pressuresmight occur during the Task 5 lifting scenario causingthe annulus to bulge further.

It is important to understand that the poroelastic finiteelement model used here is generic in nature and thevarious parameters used in defining the characteristics ofthe model have been taken from the literature. In general,these values might vary widely. The constitutive proper-ties used in the current model study are assumed to berate independent whereas during dynamic lifting suchproperties do change over time. Unfortunately, currentlysuch rate-dependent material properties for various disctissues are not available in the literature. Also, in thecurrent study, the effect of complex loading arising fromvarious lifting modes was studied using one isolated mo-tion segment, and thus the effects of adjacent segmentshave been ignored. Other factors such as ligament pre-strain and facet joint interaction between levels have alsobeen ignored. However, the main aim of the currentstudy was to compare the biomechanical response of alumbar spine motion segment under 8 different com-monly occurring lifting activities keeping all other modelparameters such as geometry, material parameters, andboundary conditions the same in all these studies. Thus,the results from this model study will help to compare,on relative terms, the lumbar disc biomechanics of thevarious lift loading tasks. The current model includesregional variations of material properties that will alsohelp to better predict the biomechanics of the lumbardisc under complex loading that involves flexion com-bined with lateral bending and twisting.

The model produced an overall prediction that com-pared well with data available in literature based on mea-surements taken using human volunteers in a laboratorysetting for long-term creep (circadian variation), short-term creep, and short-term cyclic loading. Currently, themodel has not been validated for complex loading modesbecause no literature is available on the in vivo biome-chanics of a lumbar disc under loading conditions suchas those experienced during lifting.

The results suggest that the loading modes consideredcan cause local yield failure in the annulus and endplates.This conclusion was reached by comparing the stressesproduced in these components under various loadingmodes with corresponding component failure stresses re-ported in the literature. It is important to note that thefailure considered here is a localized phenomenon and itdoes not refer to total component failure. Further, theselocal tissue failures do get repaired with time, which hasnot been included in the current model analyses.

In conclusion, the current study suggests that asym-metric lifting involving lateral bending of the trunk mayincrease the risk of disc injury because it produced largetranslational as well as rotational motions in the lumbarmotion segment. Higher stresses exceeding the failurestrength of the corresponding disc tissues were predicted

during asymmetric lifting. These stresses (1) might paveway for matrix cracking and interlaminar shear failure inthe annulus and (2) cause localized fracture of endplates.Thus, asymmetric lifting might not only induce failure invarious components of the disc tissue but also enhance itsprogression if this lift is performed over a large numberof cycles.

Key Points

● Biomechanical response of the spine to loads thatarises during manual lifting.● Poroelastic finite element model study of a hu-man lumbar disc subjected to 8 different lift condi-tions.● Current study showed that asymmetric lifting in-volving lateral bending of the spine produced thehighest stresses in the annulus and endplates andmay cause macromechanical tears.● Maximum shear stress in the annulus producedby asymmetric lifting and higher than the shearfailure stress might induce delamination.● High facet forces under asymmetric lifting mightlead to facet joint osteoarthritis.● Assymmetric lifting may increase the risk of backinjury and pain.

References

1. Fathallah FA, Marras WS, Parnianpour M. The role of complex, simulta-neous trunk motions in the risk of occupation-related low back disorders.Spine 1998;23:1035–42.

2. Praemer A, Furner S, Rice DP. Musculoskeletal conditions in the UnitedStates. American Academy of Orthopaedic Surgeons. Park Ridge, IL; 1999.

3. Andersson GB, Herberts P, Ortengren R. Myoelectric back muscle activity instandardized lifting postures. In: Komi P, ed. Biomechanics. University ParkPress: Baltimore, MD; 1976.

4. Mc Gill SM. The Biomechanics of low back injury: implications on currentpractice in industry and the clinic. J Biomech 1997;30:465–75.

5. Daltroy LH, et al. Teaching and social support: effect of knowledge, atti-tudes, and behavior to prevent low back injuries in industries. Health EducBehav 1993;20:43–62.

6. Frank J, Sinclair S, Hog Johnson S, et al. Preventing disability from work-related low-back pain. New evidence gives hope – if we can just get all theplayers onside. Can Med Assoc J 1998;158:1625–31.

7. van Dieen JH, Dekkers JJ, Groen V, et al. With-in subject variability in lowback load in a repetitively performed, mildly constrained lifting task. Spine2001;26:1799–804.

8. Davis KG, Marras WS. Partitioning the contributing role of biomechanical,psychosocial, and individual risk factors in the development of spine loads.Spine J 2003;3:331–8.

9. Kelsey JL, Githens PB, White AA, et al. An epidemiological study of liftingand twisting on the job and risk for acute prolapsed lumbar intervertebraldisc. J Orthop Res 1984;2:61–6.

10. Marras WS, Lavender SA, Leurgans SE, et al. The role of dynamic three-dimensional trunk motion in occupationally-related low back disorders: theeffects of workplace factors, trunk position, and trunk motion characteristicson risk of injury. Spine 1993;18:617–28.

11. Videman T, Nurminen M. The occurrence of annular tears and their relationto lifetime back pain history: a cadaveric study using barium sulfate discog-raphy. Spine 2004;29:2668–76.

12. Stokes IAF, Iatridis JC. Mechanical conditions that accelerate interevertebraldisc degeneration: overload versus immobilization. Spine 2004;29:2724–32.

13. Adams MA, Hutton WC. Gradual disc prolapse. Spine 1985;10:524–31.14. Gordon SJ, Yang KH, Mayer PJ, et al. Mechanism of disc rupture. A prelim-

inary report. Spine 1991;16:450–6.


15. Adams MA, Freeman BJ, Morrison HP, et al. Mechanical initiation of inter-vertebral disc degeneration. Spine 2000;25:1625–36.

16. Videman T, Nurminen M, Troup JD. 1990 Volvo Award in clinical sciences.Lumbar spinal pathology in cadaveric material in relation to history of backpain, occupation, and physical loading. Spine 1990;15:728–40.

17. Costi JJ, Stokes IA, Gardner-Morse M, et al. Direct measurement of inter-vertebral disc maximum shear strain in six degrees of freedom: motions thatplace disc tissue at risk of injury. J Biomech 2007.

18. Schmidt H, Kettler A, Rohlmann A, et al. Under which load combination anddegree of disc degeneration exists the highest risk of annular failure and discprolapses? 2007, 53rd Annual meeting of the Orthopaedic Research Society,Poster No. 1032.

19. Natarajan RN, Williams JR, Andersson GBJ. Recent advances in analyticalmodeling of lumbar disc degeneration. Spine 2004;29:2733–41.

20. ADINA R&D Inc, Watertown, MA 02172, USA.21. Natarajan RN, Williams JR, Andersson GBJ. Modeling changes in interver-

tebral disc mechanics with degeneration. J Bone Joint Surg 2006;88-A(suppl2):36–40.

22. Williams JR, Natarajan RN, Andersson GBJ. Inclusion of regional poroelas-tic material properties better predicts biomechanical behavior of lumbardiscs subjected to dynamic loading. J Biomech 2007;40:1981–7.

23. Tyrrell A, Reilly T. Circadian variation in stature and the effects of spinalloading. Spine 1985;10:161–4.

24. Granata KP, Marras WS. An EMG-assisted model of trunk loading duringfree-dynamic lifting. J Biomech 1995;28:1309–17.

25. Schechtman H, Robertson PA, Broom ND. Failure strength of the bovinecaudal disc under internal hydrostatic pressure. J Biomechanics 2006;39:1401–9.

26. Osti OL, Vernon-Roberts B, Fraser RD. Annulus tears and intervertebraldisc degeneration: an experimental study using an animal model. Spine 1990;15:762–7.

27. Haefeli M, Kalberer F, Saegesser D, et al. The course of macroscopic degen-eration in the human lumbar intervertebral disc. Spine 2006;31:1522–31.

28. Iatridis JC, Gwynn IA. Mechanisms for mechanical damage in the interver-tebral disc annulus fibrosus. J Biomech 2004;37:1165–75.

29. Perey O. Fracture of the intervertebral endplate in the lumbar spine. Anexperimental biomechanical investigation. Acta Orthop Scand Suppl 1957;25:1–101.

30. Lotz JC, Glazer PA, Gryler EC. Tensile properties of the human vertebralend-plate. International society for study of lumbar Spine 1995;21.

31. Wilson DC, Niosi CA, Zhu QA, et al. Accuracy and repeatability of a newmethod for measuring facet loads in the lumbar spine. J Biomech 2006;39:348–53.

32. Rohlmann A, Zander T, Schmidt H, et al. Analysis of the influence of discdegeneration on the mechanical behavior of a lumbar motion segment usingthe finite element method. J Biomech 2006;39:2484–90.

33. Nachemson A, Lewin T, Maroudas A. In vitro diffusion of dye through theendplates and the annulus fibrosus of human lumbar intervertebral discs.Acta Orthop Scand 1970;41:589–607.


Documents

SPINE Volume 33, Number 18, pp 1958–1965 ©2008, Lippincott ...1.step.nl/...a_Lumbar_Intervertebral_assym_tillen.pdf · intervertebral disc specimens. Lateral bending and ﬂex-ion