Current Orthopaedics (2008) 22, 193e201

ava i lab le at www.sc ienced i rec t . com

journa l homepage : www.e ls ev ier . com/ loca te /cuor

BASIC SCIENCE

Biomechanics of the menisci of the knee

Ian D. McDermott a,b,*, Spyridon D. Masouros c,d, Andrew A. Amis d,e

a Northwood Sports Orthopaedics, 166 Northwood Way, Northwood, Middlesex HA6 1RB, UKb The Centre for Sports Medicine and Human Performance, Brunel University School of Sport & Education, Heinz WolffBuilding, Uxbridge, Middlesex UB8 3PH, UKc Department of Bioengineering, Imperial College London, Exhibition Road, London SW7 2AZ, UKd Biomechanics Group, Department of Mechanical Engineering, Imperial College London, Exhibition Road,London SW7 2AZ, UKe Musculoskeletal Surgery, Imperial College London, Charing Cross Hospital, London W6 8RF, UK

KEYWORDSMenisci;Knee;Biomechanics

* Corresponding author. Mr Ian McDerFFSEM(UK), Consultant Orthopaedic SAssociate, Brunel University School of SSports Orthopaedics, 166 NorthwoodHA6 1RB, UK.

E-mail address: ian.mcdermoMcDermott).

0268-0890/$ - see front matter ª 200doi:10.1016/j.cuor.2008.04.005

Summary

The menisci of the knee are complex structures with various important functions within theknee. Loss of the menisci leads to a significantly increased risk of developing degenerativechanges in the long term. To fully understand the role of the menisci it is necessary to havean appropriate understanding of their anatomy, microstructure, material/mechanical proper-ties and biomechanical function. This review will give a brief outline of both the underlyingprinciples involved as well as giving some detail explaining the behaviour in-vivo of these com-plex and important structures.ª 2008 Elsevier Ltd. All rights reserved.

Introduction

The menisci are two crescent-shaped cartilages that arefound within each knee between the femoral condyles andthe tibial plateaux. For many years the menisci were

mott MB, BS, MS, FRCS(Orth),urgeon, Honorary Professorport & Education. NorthwoodWay, Northwood, Middlesex

tt@doctors.org.uk (I.D.

8 Elsevier Ltd. All rights reserved

considered to be the functionless remains of a leg muscle.1

Indeed, in his paper in 1942 McMurray stated that ‘‘Whenthe knee-joint is opened on the anterior aspect, and thesuspected cartilage appears normal, its removal can beundertaken with confidence if the diagnosis of a posteriortear has been arrived at (clinically) prior to operation. Afar too common error is shown in the incomplete removalof the injured meniscus.’’ 2

Attitudes towards the menisci have changed dramati-cally, and since King’s pivotal paper in 1936,3 numerousstudies have shown that the menisci do in fact play impor-tant roles in load bearing4e8 and shock absorption withinthe knee,9e11 and that they are also secondary stabilisersof the joint.12e15 Further roles in joint lubrication and

.

194 I.D. McDermott et al.

nutrient distribution,16 and sensory function and proprio-ception17 have been proposed. Meniscal tears are probablythe most common intra-articular injury of the knee,18 and itis now fully appreciated that meniscectomy leads to a largeincrease in the risk of subsequently developingdegenerative changes within the joint.19

The responses of the menisci to loads applied to thetibiofemoral joint result from their macro-geometry, theirfine structure and the nature of their insertional ligaments.The menisci are fibrocartilagenous structures primarilycomposed of an interlacing network of collagen fibres (pre-dominantly type I collagen) interposed with cells, with anextracellular matrix of proteoglycans and glycoproteins.

The collagen composition varies between the surfacelayer of the menisci and the deeper tissue. The collagenbundles of the surface layer are randomly orientated witha compositional similarity to articular hyaline cartilage.20

This relates to its function, which is to allow low-frictionmotion between the meniscal tissue and the femur andtibia as the femoral condyles slide against the meniscusand move it over the tibial plateau during knee motion.

Within the bulk of the meniscal tissue, beneath thesurface layers, there are two structurally distinct regionswhere the orientation of collagen fibres is different: in theinnermost third the collagen bundles predominantly lie ina radial pattern, whereas in the outer two-thirds thecollagen bundles are orientated circumferentially (Fig. 1).This suggests that the inner third may function in compres-sion and that the outer two-thirds function in tension.Further, less-frequent, radially-orientated collagen fibrescan also be found within the bulk of the meniscal tissueand these may act as tie fibres, resisting longitudinal split-ting of the circumferential collagen bundles.21

To fully understand the functioning of the menisci of theknee, it is first essential to have a firm grasp of the basicunderlying principles of biomechanics as they relate tomeniscal tissue. These principles will be outlined and thenexpanded upon in the following sections.

The basic biomechanics of tension

When a force is applied to any material or tissue that is notcompletely free to move, there will be a resultant

Figure 1 Diagrammatic representation of the collagen fibreorientations within the meniscus. Taken from Bullough et al.21

deformation. The behaviour of tissue as a stretching forceis applied is referred to as its tensile properties.

As a stretching force is applied to a sample of tissue,such as mensical tissue, the sample will show a resultantelongation. Equally, as a tissue is stretched, a force willdevelop within the tissue, opposing its elongation. Tissueextension can be plotted graphically against the measuredresultant force, to give a load-elongation curve (Fig. 2).

The first region of the load-elongation curve is referredto as the ‘toe region’. Little force is required to elongatethe tissue initially, and the elongation at this stage occursas the wavy pattern of relaxed collagen fibres within themeniscal tissue becomes straighter.22

The second region of the curve is linear, and during thisstage thecollagenfibres themselvesbecomestretchedandareparallel, having lost their wavy appearance. During this phase,there is a linear relationship between elongation and load.

Towards the end of the linear phase of the curve, smalldips representing force reductions can be seen. These arecaused by the early sequential failureof some individual fibrebundles.23 Eventually, major failure of fibre bundles occursand complete failure ensues. The point at which this beginsto occur is referred to as the yield point for the tissue, andrepresents a change from elastic (reversible) to plastic (irre-versible) deformation. The maximum load attained is re-ferred to as the ultimate tensile load of the specimen.

The stiffness of the testing sample can be defined by

stiffnessZload

elongation; kZ

DF

DL

and can be determined by calculating the gradient of thelinear portion of the graph. The stiffness of the sample isa structural property of the specific tissue sample tested.This means that the stiffness value is dependent upon thematerial properties of the tissue itself and the dimensionsof the sample. In order to make comparisons between dif-ferent tissues or materials, measures which are indepen-dent of sample dimensions are used; these are termedmaterial properties.

To adjust for specimen cross-sectional area, a measure-ment of stress is used. This is defined as

stressZload

cross-sectional area; sZ

F

AMPaZN=mm2� �

Elongation (mm)

Lo

ad

(N

)

Figure 2 Typical Load versus Elongation curve for tensiletesting of fibrous soft-tissue samples.

Exte

nsio

n

Time

A

B

Figure 3 Plot of Compression against time for a constantcompressive load. An initial near-linear portion is seen (A), fol-lowed by a curved section of diminishing rate of compression(B), representing creep of the tissue as water is extruded.

Time

Lo

ad

Figure 4 Stress-relaxation curve, showing the behaviour ofmeniscal tissue when it is compressed to a set value ofdisplacement.

Biomechanics of the menisci of the knee 195

To adjust for specimen length, a measurement of strainis used. This is defined as

strainZelongation

original length; 3Z

DL

L

The original length of the test sample is referred to as thegauge length. Strain is often multiplied by 100 to give per-cent strain, and has no units.

Using the definition of stress and strain given above,a new ‘stress-strain’ curve can be plotted. This has thesame shape as the load-elongation curve. Calculatingthe gradient of the linear portion of this curve by dividingthe change in stress by the change in strain, now givesa value referred to as the Young’s modulus of the material:

Young0s modulusZstress

strain; EZ

s

3

The modulus is a material property and can be used asa measure of comparison of the material stiffnesses ofvarious tissues of different size and shape.

The basic biomechanics of compression

In 1980, Mow et al. investigated the behaviour of articularcartilage and showed that it could be described as a biphasicmaterial; composed of a solid matrix phase and an in-terstitial fluid phase.24 Favenesi et al., in 1983, examinedthe behaviour of bovine menisci and concluded that menis-cal tissue also followed the expected behaviour of a biphasicmaterial.25 They demonstrated that the water content ofthe meniscus was freely exchangeable with the surroundingfluid by mechanical means. The meniscal water contentcould be extruded either by compression or by direct appli-cation of a pressure differential. Favenesi et al. showed,however, that the meniscal tissue was only half as stiff,and one sixth as permeable as articular cartilage.

Spilker et al. used computer finite element analysis tomodel meniscal properties, and found that the biphasicrepresentation, whereby the tissue was described as a mix-ture of solid and fluid components, was crucial, and that thefluid phase carried a significant part of the applied load.26

Thus, when a load is applied to meniscal tissue the solidphase, consisting predominantly of circumferentially orien-tated collagen bundles, shows an elastic response. How-ever, simultaneously, the fluid pressure carries thecompressive load while the fluid component of the tissueis very slowly extruded at a rate dependent on thepermeability of the tissue and the viscosity of the fluid.The tissue therefore exhibits viscoelastic behaviour.

If a constant compressive load is applied to a sample ofmeniscal tissue, there is an initial compression that is near-linear, and which predominantly represents the solid phaseof the material, i.e. the elastic behaviour of the collagenbundles and the matrix material. This immediate phase isfollowed by a curved phase where there is a continuedcompression of the tissue, but at a diminishing rate. Thissecond stage is where the fluid-phase predominates, andfluid is moving within and being expelled from the meniscaltissue. The continued but diminishing compression isreferred to as creep, and can be seen when extension isplotted against time (Fig. 3).

Similarly, when meniscal tissue is compressed and thenheld, and the load required to maintain that compression ismeasured, it can be seen that the load diminishes withtime. If load is plotted against time, then a characteristiccurve is seen. As fluid is expelled from the meniscal tissue,the tissue relaxes and the required load to maintain anygiven compression decreases with time. The resulting curveis known as a stress relaxation curve (Fig. 4).

Creep and stress-relaxation are related characteristicsof viscoelastic behaviour; this is a time dependent behav-iour. The water content and permeability of meniscal tissueare responsible for the viscoelastic response of the tissue. Itis important to note, however, that the permeability ofmeniscal tissue is very low when compared to articularcartilage, so that the menisci effectively maintain theirvolume under load, otherwise they would soon becomefunctionless within the knee.

The material properties of meniscal tissue

Tensile material properties

As far back as 1949, Mathur et al. took fresh meniscal tissuefrom amputated limbs, held the tissue in metal clamps,

Figure 6 Orientation of tensile testing dumbbells. Takenfrom Whipple et al., 1985.29

196 I.D. McDermott et al.

suspended the tissue within a three foot high frame, andthen hung weights on the samples, measuring the ‘stretch-ing’ of the tissue and the load necessary to cause the tissueto ‘fracture’.27

However, in order to measure the material properties ofmeniscal tissue, samples of clearly-defined geometry arenecessary. The standard shape of tissue sample used bymost investigators is based on the basic materials testingdumbbell (Fig. 5). This has expanded sections at the endsfor clamping the tissue, and a narrower gauge length inbetween. The cross-sectional area in this section is uni-form, thus allowing calculation of stress. However, al-though this is appropriate for the testing of non-fibrousmaterials, dumbbell shaped test samples should be usedwith care for fibre-reinforced composites because theytend to split in the region where the width changes.28

Whipple et al. studied the tensile properties of bovinemenisci using dumbbell-shaped samples, 400 micrometersin thickness.29 They compared the properties of the tissuefrom samples taken either circumferentially or radially tothe circumferentially-orientated collagen fibres (Fig. 6)and demonstrated that in the surface layer of the menisci,where the collagen fibres are randomly orientated, therewere no differences between samples taken in differentorientations. However, in the deeper, main bulk of themeniscal tissue, samples taken circumferentially hada much greater modulus (198.4 MPa) compared to radi-ally-orientated samples (2.8 MPa). Similar findings havebeen observed by other authors.21,30,31 This directionalvariation of material properties within the tissue is referredto as anisotropy, and the tissue is described as anisotropic.

Whipple et al. also observed differences in the materialproperties of circumferentially sampled meniscal tissuedepending on the depth of the samples: tissue from thesurface layer had a modulus of 48.3 MPa compared toa value of 198.4 MPa for the middle zone and 139.0 MPafor the deepest layer sampled.

Other investigators have demonstrated regional varia-tions in the material properties of meniscal tissue. Fithianet al. found that the tensile modulus was significantlyreduced in the posterior two-thirds of the medial meniscus,compared to the anterior third of the medial meniscus orthe entire lateral meniscus.32 They inspected the meniscaltissue under polarised light to determine collagen fibre ori-entation, and found that whereas in the rest of meniscaltissue the fibres were highly orientated in the circumferen-tial direction, in the posterior two-thirds of the medialmeniscus the fibre bundles were orientated obliquely withrespect to each other. Fithian et al. felt that the capsularattachments in the posterior two-thirds of the medial me-niscus may interrupt the circumferential fibre organisation,and stated that this might explain in part the frequency oftears observed in the posterior horn of the medialmeniscus.

Figure 5 Materials testing dumbbell.

In a more recent and highly comprehensive study,Lechner et al. observed that the posterior and middleregions of human medial menisci had a higher tensilemodulus than the anterior region.33 They also noted thatas the circumferential collagen fibres are continuousthrough the meniscus from the anterior to the posterior tib-ial attachments, and the anterior horn of the medial menis-cus is not as wide in the radial direction as the central orposterior regions, the same number of collagen fibresmust be packed into a smaller cross-sectional area of tissueanteriorly. This would increase the ratio of collagen fibresto matrix tissue anteriorly, and might explain the increasedstiffness of the tissue anteriorly.

Lechner et al. also showed that the modulus of meniscaltissue varies inversely with test specimen thickness.33 Theynoted that standard materials testing assumes that thetissue being tested is actually homogenous. Meniscal tissue,however, is composed of collagen fibre bundles, rangingfrom 50 to 400 micrometers in diameter, embedded withinweak matrix tissue (Fig. 7).34 Thus, thinner samples mayconsist predominantly of matrix tissue. Lechner et al. foundthat 28% of 0.5 mm thickness samples either failed prema-turely, or could not even be tested. Only 17% of thicker,3 mm samples were untestable. Thus with thinner samples,the samples that could be tested were likely to be thosethat contained a significant amount of collagen fibre tissue,as opposed to just mainly weak matrix. Hence the thinnersamples that could be tested had an apparent highermean modulus than the thicker samples.

The above issues emphasize how complicated it is to tryto define the biomechanical properties of mensical tissue.Thus, attempts to try to quantify absolute values for

A B

Figure 7 Diagram representing how sample thickness mayaffect material properties. Sample A (thin) contains 2 collagenbundles. Sample B (thin) contains only matrix and is untest-able. Therefore of the thin samples successfully tested, thereappear to be 2 bundles per slice thickness. Sample Aþ B (thick)contains 2 collagen bundles and is testable. The mean collagenbundle count per slice thickness is 1. Therefore, the thickerslice has a lower tensile modulus.

Figure 8 Conversion of axial load into meniscal hoopstresses.

Biomechanics of the menisci of the knee 197

material properties probably have little meaning unlesstaken within the context of the overall macrostructure ofthis highly complex organ.

However, for general comparison, if a value in the regionof 150 MPa is taken as an approximation of the tensile mod-ulus of meniscal tissue, Young’s modulus for the anteriorcruciate ligament is approximately 200e300 MPa, and themodulus of high-density polyethylene is approximately1000 MPa. The ultimate stress to failure of meniscal tissueis in the region of 20 MPa. The ultimate stress of ligamenttissue is typically 50 MPa, and of mild steel is approximately250 MPa.

Compressive material properties

Favenesi et al. examined the compressive material prop-erties of bovine menisci using confined compression creeptesting and direct permeability measurement.25 They foundthat the water content of the menisci was 73.9%. The meanhydraulic permeability of the tissue was 6.4� 10�16 m4/N�S, and the mean compressive modulus was 0.41 MPa.The water content, hydraulic permeability and compressive

modulus were noted to vary according to the depth of thesample location (superficial vs deep) and also the locationof the sample (anterior vs central vs deep).

Proctor et al. also investigated the compressive behav-iour of bovine menisci to determine the compressivemodulus and permeability of the tissue.31 Proctor et al.found a mean water content of 73.8%, with a mean perme-ability of 0.81� 10�15 m4/N�S, and a mean compressivemodulus of 0.41 MPa. They found that values were uniformwithin the surface layer and did not differ according tolocation. However, for the deeper tissue, Proctor et al.did observe that the measured compressive material prop-erties differed significantly according to sample location.The posterior deep specimens had a significantly highermodulus. No significant regional variations were observedin the permeability coefficients of the tissue, but the poste-rior specimens were found to have a significantly higherwater content.

Functional biomechanics

Load transmission

The menisci transfer forces between the femoral and tibialjoint surfaces by the development of hoop (circumferen-tial) stresses within the meniscal tissue.11 These are tensilestresses transferred along the circumferential collagen fi-bres of the meniscus between insertions. As the near-circular femoral condyles bear down onto the meniscaltissue, the menisci have a tendency to be extruded periph-erally.9 This extrusion is prevented by the firm attachmentsof the anterior and posterior horns to the tibia via the ante-rior and posterior insertional ligaments, because anincrease of the radius of the circle of meniscus simulta-neously increases the circumference. Thus, a circumferen-tial tension develops within the menisci, opposing furtherdisplacement. This is referred to as ‘hoop stresses’.35

Thus the compression force across the knee, as it is trans-mitted from the femur through the meniscus to the tibia,causes tension along the circumferential collagen fibreswithin the meniscal tissue (Fig. 8).

A number of different experimental techniques havebeen used to determine intra-articular contact areas andpressure distribution. By loading cadaver knees in

198 I.D. McDermott et al.

a materials testing machine and using pressure transducerswithin the joint, Seedhom showed that on the lateral sidethe meniscus carries 70% of the load in the lateralcompartment, and that the medial meniscus carries 50%of the load in the medial compartment.36

Walker et al. used methylmethacrylate casting andshowed that under no load, contact across the kneeoccurred primarily on the menisci.8 With loads of 1472N,the menisci were shown to cover between 59 to 71% ofthe joint contact surface area.

Using a casting method employing silicone rubber,Fukubayashi and Kurosawa found that under a load of1000N, the menisci occupied 70% of the total contact area.5

Removal of the menisci more than halved the contact areaand led to a doubling of the peak contact pressures.

The importance of the menisci in load bearing wasfurther shown by Baratz et al., who, using pressure-sensitivefilm in knees loaded by a materials testing machine, showedthat after total meniscectomy, joint contact areas de-creased by approximately 75%, and peak local contactstresses increased by approximately 235% (Fig. 9).37

Shock absorption

As has been described above, axially directed energy acrossthe knee during loading of the joint is converted into hoopstresses within the circumferential collagen fibres.9

Furthermore, as well as energy being absorbed into thecollagen fibres (solid phase of the meniscus), as the tissue iscompressed energy is further absorbed by the expulsion ofthe joint fluid (fluid phase of the meniscus) out of the tissue.

The shock absorbing capacity of the menisci has beenmeasured by registration of bone vibrations resultingfrom gait, and it has been shown that without the menisci,shock absorption within the knee is reduced by approxi-mately 20%.11

Meniscal motion

The menisci are dynamic structures, and to effectivelymaintain an optimum load-bearing function over a moving,incongruent joint surface, they need to be able to move as

Figure 9 Increase in peak local contact pressures aftermeniscectomy.

the femur and tibia move, to maintain maximumcongruency.

Thompson et al. were the first to describe meniscalmovements through a full flexion-extension arc in theintact knee using MRI of cadaver knees.38 They showedthat from full extension to full flexion, there was posteriorexcursion of the medial meniscus of 5.1 mm, and of the lat-eral meniscus of 11.2 mm, with the anterior horns movingmore than the posterior horns. However, these observationswere made in unloaded cadaver knees, and may thereforenot be representative of the in-vivo weight bearing situa-tion. Furthermore, Thompson et al. failed to comment onthe medio-lateral movement of the meniscal tissue.

More recent technical advances in the field of radio-graphic imaging have lead to the development of so-called‘open’ magnetic resonance scanners. These scanners allowa subject to lie, stand, sit or squat within the imaging field,and thus permit imaging of the intact in-vivo knee underload in all positions. Using such a scanner, Vedi et al.described the details of meniscal motion in the normalknee in both the weight-bearing and non-weight-bearingsituations.39 They found that the menisci moved less thanhad been reported by Thompson et al. However, in commonwith the findings of Thompson et al., Vedi et al. observedthat the menisci do move posteriorly as the knee flexes.The anterior horns were noted to be more mobile thanthe posterior horns, and the lateral meniscus to be moremobile than the medial (Fig. 10). The posterior horn ofthe medial meniscus was found to be the least mobile.Vedi et al. also showed that there was significant move-ment of the bodies of the menisci peripherally with flexion.

Vedi et al. compared the observed values in theunloaded knee with those found in the weight-bearingsituation. They showed that statistically there was signif-icantly greater movement in the anterior horn of the lateralmeniscus when the knee was weight-bearing, but nosignificant differences were seen in the other meniscalmovements observed.

Joint stabilisation

The role of different anatomical structures in the stabilityof the knee has been studied extensively, both in cadavericwork and in-vivo. Tapper and Hoover reviewed 113 patientsat between 10 to 30 years after meniscectomy (84% medial,16% lateral) and found that 24% showed evidence ofinstability of the knee.40 In this study all patients thathad evidence of collateral or cruciate ligamentous injuriesduring initial surgery were excluded. It was therefore pos-tulated that the group of patients showing instability at fol-low-up either had unrecognised injuries, were re-injuredlater, had laxity due to degenerative changes, or thatperhaps the loss of the meniscal mass might havecaused a relative laxity of the ligaments.

Levy et al. showed that anterior-posterior laxity andcoupled rotation were unaffected by medial meniscectomyin the anterior cruciate-intact knee.13 However, after sec-tion of the anterior cruciate ligament subsequent furtherresection of the medial meniscus led to an additionalincrease in anterior laxity. In a further study, Levy et al.also showed that lateral meniscectomy after resection ofthe anterior cruciate ligament resulted in a small but

Figure 10 Diagram showing the mean movement (mm) in each meniscus from extension (shaded) to flexion (hashed) in (A) theweight-bearing and (B) the unloaded knee. Taken from Vedi et al.39

Biomechanics of the menisci of the knee 199

significant further increase in anterior translation.12 Thesmaller increase in anterior laxity associated with removalof the lateral meniscus compared to that seen with themedial meniscus may well be related to the fact that thelateral meniscus is more mobile than the medial. The rela-tively immobile posterior horn of the medial meniscus,however, most probably acts as a ‘chock block’ resistingposterior translation of the medial femoral condyle.

Further work by Allen et al. confirmed that the actualforces within the medial meniscus in response to ananterior tibial load increased by up to 197% after sectionof the anterior cruciate ligament.41 It was also shown thatafter medial meniscectomy in the anterior cruciate defi-cient knee there was a significant further increase in in-duced anterior tibial translation of up to 5.8 mm.

It can therefore be considered that the medial meniscusdoes appear to be a significant secondary stabiliser toanterior drawer, of particular importance in the ACL-deficient knee, and this may well be a factor contributingto the poor prognosis that is seen in those patients withcombined ACL tears plus medial meniscal deficiency.

Associated ligaments

To further complicate the issue of accurately describing thebiomechanical properties of the menisci of the knee, there

Figure 11 The transverse ligament on a cadaveric specimen,being held with a pair of forceps.

are various associated ligaments which, although variablein their presence, can have significant influence on thebehaviour of the menisci.

The anterior intermeniscal ligament, also known as thetransverse geniculate ligament, connects the anteriorfibres of the anterior horns of the medial and lateralmenisci (Fig. 11). In a cadaveric study of ninety-two kneesperformed by Kohn and Moreno, a transverse ligament wasfound to be present in 64% of specimens.42 However, thefunctional relevance of this ligament has not yet beendetermined.

Two ligaments have also been identified joining theposterior horn of the lateral meniscus to the lateral side ofthe medial condyle of the femur in the intercondylar notch.These are known as the meniscofemoral ligaments. Oneligament runs anterior to the posterior cruciate ligament,and is known as the ligament of Humphrey. The other runsposterior to the posterior cruciate ligament, and is knownas the ligament of Wrisberg (Fig. 12). The ligament ofHumphrey has been found to be present in 74% of knees,the ligament of Wrisberg in 69% of knees, and bothligaments together in 50% of knees.43 The relevance and im-portance of these ligaments has been demonstrated byGupte et al., who showed in a cadaveric study that the

Figure 12 The meniscofemoral ligaments. LM Z lateralmeniscus. PCL Z posterior cruciate ligament. AMFL Z anteriormensicofemoral ligament (Ligament of Humphrey).PMFL Z posterior meniscofemoral ligament (Ligament ofWrisberg).

200 I.D. McDermott et al.

meniscofemoral ligaments contributed 28% to the totalforce resisting posterior drawer at 90 degrees of flexion inthe intact knee, and 70.1% in the PCL-deficient knee.44

It should, therefore, perhaps be of some concern thatsuch little, if any, consideration is currently given tothe associated ligaments of the menisci during surgery ofthe knee and when undertaking mensical repair orreplacement.

Conclusion

This brief review has described how the morphology andstructure of the menisci are adapted to their mechanicalload-transmitting function. In particular, the circumferen-tial collagen fibres allow meniscal radial extrusion underaxial joint forces to be resisted by the build-up of hoopstresses. The anisotropic material properties are adaptedto this role, but at the expense of the tissue being weak inthe radial direction, which allows the meniscus to be tornrelatively easily by splitting along the lines of the circum-ferential fibres. Meniscal surgery should aim to maintaincontinuity of the circumferential structure in order toensure the load-bearing performance; without this, thearticular joint surfaces are subjected to excessive contactstresses, resulting in premature onset of osteoarthrosis.

The true structural complexity of the menisci, with theirmaterial inhomogeneity, their anisotropy and their variousassociated ligaments, emphasizes just how specialised theyin fulfilling their various roles within the knee. It alsoexplains how complicated a task it is to try and model themaccurately. However, a detailed and comprehensive un-derstanding of the structural and biomechanical character-istics of the menisci is vital for future work such as thetissue engineering of meniscal replacements.

References

1. Sutton JB. Ligaments: their nature and morphology. London:M.K. Lewis & Co.; 1897.

2. McMurray TP. The semilunar cartilages. Br J Surg 1942;29:407e14.

3. King D. The function of the semilunar cartilages. J Bone JointSurg Br 1936;18:1069e76.

4. Ahmed AM, Burke DL. In-vitro measurement of static pressuredistribution in synovial joints e part I: tibial surface of theknee. J Biomech Eng 1983;105:216e25.

5. Fukubayashi T, Kurosawa H. The contact area and pressure dis-tribution pattern of the knee. A study of normal and osteoar-thritic knee joints. Acta Orthop Scand 1980;51:871e9.

6. Seedhom BB. Loadbearing function of the menisci. Physiother-apy 1976;62:223.

7. Shrive NG, O’Connor JJ, Goodfellow JW. Load-bearing in theknee joint. Clin Orthop Relat Res 1978;131:279e87.

8. Walker PS, Erkman MJ. The role of the menisci in force trans-mission across the knee. Clin Orthop Relat Res 1975;109:184e92.

9. Krause WR, Pope MH, Johnson RJ, Wilder DG. Mechanicalchanges in the knee after meniscectomy. J Bone Joint SurgAm 1976;58:599e603.

10. Kurosawa H, Fukubayashi T, Nakajima H. Load-bearing mode ofthe knee joint: physical behavior of the knee joint with orwithout menisci. Clin Orthop Relat Res 1980;149:283e90.

11. Voloshin AS, Wosk J. Shock absorption of meniscectomized andpainful knees: a comparative in vivo study. J Biomed Eng 1983;5:157e61.

12. Levy IM, Torzilli PA, Gould JD, Warren RF. The effect of lateralmeniscectomy on motion of the knee. J Bone Joint Surg Am1989;71:401e6.

13. Levy IM, Torzilli PA, Warren RF. The effect of medial meniscec-tomy on anterior-posterior motion of the knee. J Bone JointSurg Am 1982;64:883e8.

14. Markolf KL, Mensch JS, Amstutz HC. Stiffness and laxity of theknee e the contributions of the supporting structures. A quan-titative in vitro study. J Bone Joint Surg Am 1976;58:583e94.

15. Shoemaker SC, Markolf KL. The role of the meniscus in the an-terior-posterior stability of the loaded anterior cruciate-defi-cient knee. Effects of partial versus total excision. J BoneJoint Surg Am 1986;68:71e9.

16. Renstrom P, Johnson RJ. Anatomy and biomechanics of the me-nisci. Clin Sports Med 1990;9:523e38.

17. Messner K, Jizong DL. The menisci of the knee joint. Anatom-ical and functional characteristics, and a rationale for clinicaltreatment. J Anat 1998;193:161e78.

18. McDermott ID. Meniscal tears. Curr Orthop 2006;20:85e94.19. McDermott ID, Amis AA. The consequences of meniscectomy.

J Bone Joint Surg Br 2006;88:1549e56.20. Beaupre A, Choukroun R, Guidouin R, Garneau R, Gerardin H,

Cardou A. Knee menisci. Correlation between microstructureand biomechanics. Clin Orthop Relat Res 1986;208:72e5.

21. Bullough PG, Munuera L, Murphy J, Weinstein AM. The strengthof the menisci of the knee as it relates to their fine structure.J Bone Joint Surg Br 1970;52:564e7.

22. Viidik A. Functional properties of collagenous tissues. Int RevConnect Tissue Res 1973;6:127e215.

23. Butler DL, Grood ES, Noyes FR, Zernicke RF. Biomechanics ofligaments and tendons. Exerc Sport Sci Rev 1978;6:125e81.

24. Mow VC, Kuei SC, Lai WM, Armstrong CG. Biphasic creep andstress relaxation of articular cartilage in compression? Theoryand experiments. J Biomech Eng 1980;102:73e84.

25. Favenesi JA, Schaffer JC, Mow VC. Biphasic mechanical proper-ties of knee meniscus. Orthop Trans 1983;8:264.

26. Spilker RL, Donzelli PS, Mow VC. A transversely isotropicbiphasic finite element model of the meniscus. J Biomech1992;25:1027e45.

27. Mathur PD, McDonald JR, Ghormley RK. A study of the tensilestrength of the menisci of the knee. J Bone Joint Surg Am1949;31:650e4.

28. Sekaran SV, Hull ML, Howell SM. Nonanatomic location of theposterior horn of a medial meniscal autograft implanted in a ca-daveric knee adversely affects the pressure distribution on thetibial plateau. Am J Sports Med 2002;30:74e82.

29. Whipple RR, Wirth CJ, Mow VC. Anisotropic and zonal varia-tions in the tensile properties of the meniscus. Orthop Trans1985;10:367.

30. Tissakht M, Ahmed AM. Tensile stress-strain characteristics ofthe human meniscal material. J Biomech 1995;28:411e22.

31. Proctor CS, Schmidt MB, Whipple RR, Kelly MA, Mow VC. Mate-rial properties of the normal medial bovine meniscus. J OrthopRes 1989;7:771e82.

32. Fithian DC, Schmidt MB, Ratcliffe A, Mow VC. Human meniscustensile properties: regional variation and biochemical correla-tion. Orthop Trans 1989;14:205.

33. Lechner K, Hull ML, Howell SM. Is the circumferential tensilemodulus within a human medial meniscus affected by thetest sample location and cross-sectional area? J Orthop Res2000;18:945e51.

34. Kelly MA, Fithian DC, Chern KY, Mow VC. Structure and functionof the meniscus: basics and clinical implications. In: Biome-chanics of diarthrodial joints. New York: Springer-Verlag;1990. p. 191e211.

Biomechanics of the menisci of the knee 201

35. Jones RS, Keene GC, Learmonth DJ, Bickerstaff D, Nawana NS,Costi JJ, Pearcy MJ. Direct measurement of hoop strains in theintact and torn human meniscus. Clin Biomech 1996;11:295e300.

36. Seedhom BB, Dowson D, Wright V. Proceedings: functions ofthe menisci. A preliminary study. J Bone Joint Surg Br 1974;56:381e2.

37. Baratz ME, Fu FH, Mengato R. Meniscal tears: the effect ofmeniscectomy and of repair on intraarticular contact areasand stress in the human knee. Am J Sports Med 1986;14:270e4.

38. Thompson WO, Thaete FL, Fu FH, Dye SF. Tibial meniscal dy-namics using three-dimensional reconstruction of magneticresonance images. Am J Sports Med 1991;19:210e5.

39. Vedi V, Williams A, Tennant SJ, Spouse E, Hunt DM,Gedroyc WM. Meniscal movement. An in-vivo study using dy-namic MRI. J Bone Joint Surg Br 1999;81:37e41.

40. Tapper EM, Hoover NW. Late results after meniscectomy.J Bone Joint Surg Am 1969;51:517e26.

41. Allen CR, Wong EK, Livesay GA, Sakane M, Fu FH, Woo SL.Importance of the medial meniscus in the anterior cruciateligament-deficient knee. J Orthop Res 2000;18:109e15.

42. Kohn D, Moreno B. Meniscus insertion anatomy as a basis formeniscus replacement: a morphological cadaveric study. Ar-throscopy 1995;11:96e103.

43. Gupte CM, Smith A, McDermott ID, Bull AM, Thomas RD,Amis AA. Meniscofemoral ligaments revisited. Anatomicalstudy, age correlation and clinical implications. J Bone JointSurg Br 2002;84:846e51.

44. Gupte CM, Bull AM, Thomas RD, Amis AA. The meniscofemoralligaments: secondary restraints to the posterior drawer. Anal-ysis of anteroposterior and rotary laxity in the intact and pos-terior-cruciate-deficient knee. J Bone Joint Surg Br 2003;85:765e73.