618

OSTEOARTHRITISBy G. C. LLOYD-ROBERTS, M.B., F.R.C.S.

Lately Nuffield Fellow in Orth9paedic Surgery at the Institute of Orthopaedics, Royal National Orthopaedic Hospitaland St. George's Hospital, London; Orthopaedic Surgeon, The Hospital for Sick Children, Gt. Ormond Street

Osteoarthritis is a clinical entity the diagnosisof which implies the presence of symptoms andphysical signs referable to a degenerating joint.This simple statement is not the truism that it mayseem for progressive joint degeneration is part ofour growing older. It normally begins in the kneein the second decade which thereafter deterioratesslowly but relentlessly throughout life, until in oldage we may suffer a degree of symptomless jointdegeneration similar to that which may causesevere disability in middle age. (Bennett et al.,1942). It would seem that a balance is struck be-tween joint degeneration and decreasing activityin many ageing people so that a joint though ab-normal by radiological and pathological standards,continues to meet the reduced demands imposedupon it without hindrance to its unsuspectingpossessor. We must constantly be on our guardlest we disturb this equilibrium between the ageingindividual and his ageing joint, by erroneouslydiagnosing osteoarthritis on the evidence of minordegenerative changes on the radiograph if these areunaccompanied by pain and the signs of arthritis.This error is commonly made always at the expenseof the patient's peace of mind and often of hispurse. If the diagnosis is warranted a full ex-planation must always be given lest these unfor-tunate consequences be exaggerated.Aetiology

Osteoarthritis is tho sequel to articular cartilagedamage however induced, and so may follow manyapparently unrelated conditions provided that thejoint is used, for degeneration does not progress inthe knees of bedridden paraplegics. Broadlyspeaking, however, causes fall into two groups, theknown and the unknown.Known causes. Anatomical abnormalities are

predominant and include those which directlyinjure the bearing surfaces of a joint and causeosteoarthritis for mechanical reasons. Examplesare intra-articular fractures which have been im-perfectly restored and avascular necrosis causingloss of subchondral support and collapse of theoverlying articular cartilage, which may complicate

traumatic dislocation of the hip. In some themechanical abnormality is congenital, an import-ant example being congenital subluxation of the hipwhich is responsible for about a quarter of all casesof disabling osteoarthritis of the hip. In others thecause is found at a distance from the joint. Mal-union bf the tibia resulting in a loss of weight-bearing alignment between knee and ankle jointsmay be followed by osteoarthritis in either, anddevelopmental deformity as in knock-knee maypredispose to early degeneration on the lateral.side, later spreading to involve the whole joint.Again previous disease may have damaged the

joint beyond repair (articular cartilage has neg-ligible powers of regeneration) so that degenerativearthritis may be superimposed upon rheumatoid,pyogenic or haemophiliac arthritis and presentatypical clinical and radiological features.

Lastly, joint degeneration may follow instabilitydue either to neglected ligamentous injury ormuscle imbalance following poliomyelitis.

Unknown causes. Here we are on more dan-gerous ground for of the many theories that havebeen advanced in explanation, few can be criticallyassessed, being often ' so meaningless as to be irre-futable' (Bennett et al., I942). They includeconstipation, toxic products of metabolism, hel-minthic disease, hormonal dysfunction and so on.There is no evidence that any of these conditionscommonly cause osteoarthritis. Obesity may ad-versely affect a weight-bearing joint already de-generate for mechanical reasons, but there is nodefinite causal relationship. The theory that minordegrees of vascular insufficiency are responsible isnot supported by angiographic (Harrison et al.,1953) and histological studies.

In some cases of idiopathic monarticular arthritisminor anatomical abnormalities may have been thecause, but are often unrecognizable when theradiograph is complicated by the changes of ad-vanced degenerative arthritis. This difficultyarises in the hip where minor displacement of thefemoral epiphysis may have been overlooked inchildhood or a mild congenital dysplasia producessome increase in the angle of anteversion, a valgoid

by copyright. on M

ay 10, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.31.362.618 on 1 D

ecember 1955. D

ownloaded from

December 1955 LLOYD-ROBERTS: Osteoarthritis 6I9

·* ··1.···. "IY

:.r·::· :·: ·:··""' "7.CI ·L .C;i.ZI.1E.i.

.ff.i.i.B.lg·L ..::::::·.;::::rlF.T.t I[i.·[.ii.ljiiii.&.It?BRI. i· F1Pllliii:·i·

t ;i;'%:'":·.

.IE;'::·;;I"L 'IIC·' ;.F.a%lE.i.l::·::I:::. ·:···

::::': IL·EGi''l(iBr* 'Ills.I·%YP8·I.Ullsl· 1I&FIFii :iiiiiI.i:·.· ·:::::::·;':::·.r.i'.D.$i,i:.i. ·.eBg[e.ii.B.Bbli.."'";.i 1YIEPI!::.A"' ;liiii6iE:I

·::· iJ': :. ·:·:·.

'ii'aiiii·:'iiiii:.:· :::: .::::::.:.I'··. .l;:.s ::::.:?. ·':l::·:·:n·:. ·:·.

:: ''.·:·i ::..:

··: ··li·I.a88iiaiiili...i.IF:'::i'::iliiiii .:..,. i.

·;.::··

::i iiiiii:ii

I::···"· ··::·

FIG. I.-Articular cartilage in osteoarthritis. Note transverse and horizontal fissuringand tenuous connections of the fragments produced. In superficial area the cellsare scanty and the matrix does not take on stain as darkly as the deeper layers.X o00.

deformity of the femoral neck or a shallow aceta-bulum. These changes are sometimes found atoperation, though often unrecognizable in the pre-operative radiograph.

Kellgren and Moore (1952) have emphasized theimportance of polyarthritis of a degenerative typeoccurring usually in middle-aged women and inassociation with Heberden's nodes. If this syn-drome is remembered many cases of' idiopathic'osteoarthritis will be found to be part of it. Thesediffer in their natural history, response to surgicaltreatment and associated joint involvement whencompared with joints degenerate for mechanicalreasons. It may be that herein lies the clue thatwill reveal the cause, or causes, of prematuresenility of articular cartilage, and thereby of mostcases of osteoarthritis of unknown aetiology.Morbid Anatomy

Osteoarthritis involves the cartilaginous andosseous elements of the joint, the synovial mem-brane and capsule, and the surrounding muscles.The primary abnormality is in the articular carti-lage and secondary changes follow this. ,

Arttctar-eartifage. Matthews (I953) has shownthat articular cartilage normally in contact withinthe habitual range of joint movement differs in

structure from non-articulating areas. There is arelative increase in the chondroitin sulphate con-tent and a decrease in the concentration of collagenin the contact area, a ratio which becomes reversedwhen cartilage degenerates. Ingelmark andEkholm (1948) have demonstrated that cartilagethrives on moderate activity and may degenerate ifthis is denied. This may be due to the repetitivepumping action of intermittent pressure aiding thepassage of fluid and nutrients to the avascular sur-face layers of the cartilage. It is of interest, there-fore, that Harrison et al. (1953), having examinedhips after death, found the earliest degenerativechanges in that part of the articular cartilage whichis not weight-bearing when there was no demon-strable mechanical fault. Where such a fault existsthis change is of course first manifest in the weight-bearing area.The naked eye appearances of degenerate arti-

cular cartilage are well known and blistering,opacity, fibrillation, ulceration, fissuring and frag-mentation require no further description. Micro-scopically degenerate cartilage becomes fibrillar,losing its superficial chondrocytes and its power toreact to metachromatic stains. The surface isfragmented or papilliferous; fissures extend downto the subchondral osseous plate (Fig. i). The

E1

by copyright. on M

ay 10, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.31.362.618 on 1 D

ecember 1955. D

ownloaded from

620 POSTGRADUATE MEDICAL JOURNAL December 1955

.ii.Be..*·

I.···

·..·.· .·rc.

r

"h-.g.;·;·

;i

:·;·

.

FIG. 2.-Osteoarthritric hip. Synovial membrane and capsule from inferior part ofjoint.Note the diffuse fibrosis which obscures the junction between synovial membraneand capsule. X 64.

intra-osseous capillaries extend outwards into thewidened basal calcified layer of the cartilage whichis often thereby replaced by cancellous bone,appearing on the radiograph as linear subchondralsclerosis.

Subchondral bone. Whilst the articular cartilageis undergoing these changes the underlying boneis not quiescent. Vascular granulation tissue con-taining osteoblasts and osteoclasts fills the spacesbetween the trabeculae. The cells arrange them-selves around the bone trabeculae and take part inan intense and vital process of bone remodelling.Concurrently cartilaginous metaplasia may occurwithin the granulation tissue, and enchondralossification follows, or bone may be laid downdirectly in the newly formed fibrous tissue.The stimulus which brings about these changes

is unknown, but they may be the result of the lossof articular cartilage and the increased vulnerabilityof bone to the stresses of weight-bearing or move-ment. The final result is that the vulnerable areais replaced by dense cortical bone (eburnation)better able to withstand strain than the delicatecancellous bone normally found here.This process is not always successful in prevent-

ing disorganization of the joint and displacement ofarticulating surfaces, for during its formation this

new bone may fail to withstand the pressure that itmeets and so may wear away allowing partial dis-location to occur. This is frequently seen in thehip joint in which upwards subluxation of the headof the femur follows degeneration in a previouslystable joint because of wear between the upperweight-bearing quadrant of the femoral head andthe opposed acetabular roof.

This remodelling process also accounts for thedevelopment of osteophytes and cysts. Osteo-phytic outgrowths occur when new bone form-ation is not prevented by weight-bearing or frictionand so are usually the result of exuberant extensionof subchondral bone into the non-articular areas ofarticular cartilage at the periphery of the joint, orby metaplasia at the lines at which the capsule isattached to bone. Cysts probably represent areasin which osteoclastic resorption has outstrippednew bone formation. The cysts are initially filledwith vascular granulation tissue which graduallybecomes less vascular and more fibrotic, until acavity with a necrotic centre enclosed in a densefibrous envelope and surrounded by compact boneis formed. Such subchondral cysts weaken thebearing surface and predispose towards disintegra-tion. The cysts often communicate with the jointsurface and this has led Lindells (I953) to the con-

by copyright. on M

ay 10, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.31.362.618 on 1 D

ecember 1955. D

ownloaded from

December 1955 LLOYD-ROBERTS: Osteoarthritis 621

·."

..

~.Z' ~i~.... ·....i~~zr'!'i~j~ '* ·: '· .,.:':ii!~'~,,:..

'!:.

FIG. 3.-Osteoarthritic hip. A group of cartilage fragments lying beneath the sunovialsurface. Note their fibrillar appearance and the partial dislocation which hasoccurred during preparation. X 280.

We wish to thank the Journal of Bone and Joint Surgery for their kind permission to reproduce theillustrations used in this article.

clusion that they are due to the pumping action ofsynovial fluid which is driven into the subchondralbone.

It should be emphasized that these changes inthe subchondral bone follow cartilage degenerationand are marked only when cartilage damage issevere.

Synovial fluid. The synovial fluid is often in-creased in quantity in the early stages, but tendsto decrease as the disease progresses. It is nevercompletely absent even in a creaking joint (Collins,I949). It is frequently greenish in colour, andcontains white flakes which on histological ex-amination of the centrifuged deposit, prove to bepieces of degenerate articular cartilage detachedfrom the joint surfaces. These pieces, if largerthan usual, may persist as cartilaginous loosebodies which are sometimes seen in osteoarthritis.Such bodies would, of course, be invisible on theradiograph, but occasionally osteo-cartilaginousopacities are met which may have developed withinsynovial membrane by metaplasia, or preceded thedegenerative changes being in part responsible forthem as in osteochondritis dissecans. Fragmentsof osteophytes must very rarely, if ever, enter thejoint as a result of injury, for the osteophytes donot protrude in areas where the joint surfaces are

in contact and mobile. Lastly, the fluid may havea chocolate tinge due to altered blood. This fol-lows haemorrhage from the congested synovialmembrane, and probably explains the episodes ofirritability which occur in some osteoarthriticjoints.

Synovial membrane and Capsule. The synovialmembrane in osteoarthritis is hyperplastic, vascularand thrown into papilliferous folds. The synovialvessels are frequently surrounded by a cuff oflymphocytes, an appearance which is certainly notcharacteristic of rheumatoid arthritis. The deeperlayers may contain haemosiderin if a haemarthrosishas been present. The boundary between synoviaand capsule becomes fibrotic (Fig. 2) and thisfibrosis later spreads to involve the joint capsule,being apparently a non-specific reaction in thecapsule to sustained synovial hyperplasia (Key,I929).These changes follow articular cartilage de-

generation and seem to bear some quantitative re-lationship to it, the tendency being for synovialhyperplasia to predominate in the early stages,becoming later less conspicuous as capsular fibrosisdevelops. This synovial abnormality is evident atoperation, but it may be readily overlooked in thepallid post-mortem joint,

by copyright. on M

ay 10, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.31.362.618 on 1 D

ecember 1955. D

ownloaded from

POSTGRADUATE MEDICAL JOURNAL

This synovitis and secondary capsular fibrosis isbelieved to be due very largely to the synovial re-action that accompanies the phagocytosis of jointdebris, which arising from the degenerate jointsurfaces enters the joint cavity and finally becomesattached to and digested by synovial membrane(Fig. 3). A similar sequence has been observedwhen joint debris is introduced into the joints ofanimals (Lloyd-Roberts, 1953).

The muscles. The muscles which by reflex con-traction dynamically determine the direction ofdeformity become in time structurally shortenedby fibrosis at their musculo-tendinous junctions,and those which invest the joint capsule closely,become adherent to it by fibrosis. The adductorsof the hip behave in the first manner and shortexternal rotators in the second.

The Relationship of the Symptoms to theMorbid AnatomyThe symptoms of osteoarthritis are pain, stiff-

ness and deformity. It is believed that thesesymptoms are largely attributable to fibrosis andshortening of the joint capsule and this relation-ship will now be discussed using the hip as anexample. The following points require emphasis.

i. The entire capsule of the hip is tight in ex-tension and becomes looser as the joint flexes.Other movements are restrained by the appropriatearea of the capsule checking the movement at ex-treme range. Abduction and internal rotation are,for example, checked by the inferior part of thecapsule which lies below the neck of the femur.

2. Particulate joint debris and the synovitiswhich follows its ingestion are found mainly in thesynovial recesses below the neck of the femur,possibly as a result of gravity.

3. The capsule is copiously supplied by somaticsensory nerve fibres, and is sensitive to painfulstimuli, especially stretching.

Pain. Pain is felt either in the hip or referred toareas supplied by its sensory nerves when anattempt is made to stretch this shortened capsule.At the onset this may be felt only at the extremesof movement and particularly when extension isattempted as in standing on one leg. The tractionforce is then the body weight. If capsular shorten-ing increases, other movements, especially ab-duction and internal rotation strain the ligamentsand cause pain. Subluxation will clearly aggravatethe strain on the capsule and so is associated withpainful arthritis. This explanation accounts for thepain of movement and on weight-bearing, but doesnot explain pain at rest, This is possibly due tothe vascular engorgement of the subchondral bone,pain being transmitted by the perivascular nerves(Harrison et al., 953). It has been suggested that

a similar mechanism accounts for pain felt beforechanges in the weather.

Stiffness and deformity. Limitation of movementbears little relationship to anatomical abnormalitybefore osteoarthritis develops. This paradox isoften seen in healed Perthes' disease, when grossdeformity may be compatible with a near full rangeof movement, until secondary arthritic changesappear. A consideration of the radioluscent cap-sule which is shortened mainly in the lower partof the joint, explains the early loss of extension,abduction and internal rotation. A continuationof this process will clearly produce deformity inthe opposite direction, i.e. flexion, adduction andexternal rotation. Spasm and later contracture ofthe adductor muscles is, however, the main de-forming force. These muscles supplied by theobturator nerve contract reflexly when painfulstimuli are transmitted by this nerve which issensory to the lower part of the capsule. A similarmechanism may also occur through the nerves topectineus and quadratus femoris which are alsosensory to this area of the capsule. Such reflexcontraction will reinforce the deformity caused bythe shortening capsule in the direction, onceagain, of flexion, adduction and external rotation.

The Relationship of Symptoms toRadiological Appearances

It is often said that the radiograph in osteo-arthritis bears little relationship to the clinicalstate. This is not entirely true, for the changes inthe invisible capsule can be inferred from themanifest bone abnormalities. Provided that theage of the patient is known and he is not sufferingfrom any other condition which will restrict hisactivity, such as cardio-vascular disease or contra-lateral arthritis, a fair correlation will be foundbetween the radiological appearances and thesymptomatology.The degree of diminution of joint space, sub-

chondral sclerosis and cyst formation indicate theamount of cartilage degeneration and loss. Inaddition, if there has been loss of contour of arti-culating bony surface, a considerable amount ofdebris must have entered the joint to cause cap-sular fibrosis. Secondary subluxation may follow,and instability will impose a further strain on theshortened capsule. The formation of new sub-periosteal bone on the under surface of the neckof the femur indicates capsular adhesion and isassociated with gross limitation of movement andusually adduction deformity. Calcification in thecapsule is usually a sign of severe stiffness. Osteo-phytes, though conspicuous, seem to bear no re-lationship to the symptomatology, though theymay indicate the duration of the disease,Continued on page 634.

622 D)ecemlber I 955

by copyright. on M

ay 10, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.31.362.618 on 1 D

ecember 1955. D

ownloaded from

634 POSTGRADUATE MEDICAL JOURNAL December I955

impetus the critical examination of fibrositis led itto totter and then to crumble. Muscular pain nowdemands careful examination and investigationbefore correct treatment can be prescribed. Theawakening awareness of the various syndromeswhich can be evoked by protrusions of the inter-vertebral discs, both cervical and lumbar, havedirected therapy to the cause, not to the site of thepain, and the increase in knowledge of the im-portance of the anatomical structures surroundingthe shoulder joint have led us to stop diagnosingperiarticular fibrositis of the shoulder and toattempt a more accurate assessment of the patho-logical condition which exists. Many other in-stances could be given of advances in knowledgewhich have led to more accurate diagnosis. Such arejection of the concept of fibrositis has dealt avery severe blow to the prescription of massage asa therapeutic weapon. Massage may have alimited part to play in medicine, but the masseusesare fighting a rearguard action in its defence. Ithas little part to play in the treatment of rheu-matism. Heat still seems to have some success inrelieving the pain of muscular spasm, and is stillextensively prescribed. On the other hand,clinicians are apt to wonder if one is really justifiedin relieving a protective muscular spasm by suchmeans if the underlying cause of the spasm is leftuntreated. Other treatments involving the electriccurrent, ultra violet rays and suchlike seem nowa-days to have very little part to play. But again

physical methods of treatment are still of greatimportance in the soft tissue rheumatisms. In thecase of the protruded disc we know what spinalmovement leads to the danger of this occurring,we know what positions and postures increase ordecrease this danger and we think that by correctexercise and by postural instruction we can relievethe immediate condition and educate the patient inayoiding further recurrences. Thus the physio-therapist comes into her own once more and in amuch more scientific way. Similarly in the treat-ment of the various other types of soft tissuerheumatism the basic principles are the same.One needs an accurate diagnosis of the cause of thetrouble; one aims at the relief of the pain, if thisshould be necessary, either by the exhibition ofsalicylate or possibly by heat; and one needscareful maintenance of mobility and muscle powerby properly carried out muscular contraction.

ConclusionsThere has been a recent tendency to clarify and

appraise the true value of physical methods oftreatment. Many claims to the efficacy of varioustreatments have not stood the test of scientificscrutiny. The fundamental principles which guideus now are accurate diagnosis; careful main-tenance of muscle power and mobility by means ofpainstaking supervision on the part of the physio-therapist and active effort on the part of the patient.

RUTHIN CASTLE, NORTH WALESA Clinic for the diagnosis and treatment of Internal Diseases (except Mental or Infectious Diseases). The

Clinic is provided with a staff of doctors, technicians and nurses.The surroundings are beautiful. The climate is mild. There is central heating throughout. The annual

rainfall is 30.5 inches, that is, less than the average for England.The Fees are inclusive and vary according to the room occupied.

For particulars apply to THE SECRETARY, Ruthin Castle, North Wales.

Telegram: Catle. Ruthin. Telephone: Ruthin 66

Continued from bpae 622-G. C. Lloyd-Roberts, M.B., F.R.C.S.

AcknowledgmentsIt gives me pleasure to acknowledge the help

given to me by members of the staffs of the RoyalNational Orthopaedic Hospital and St. George'sHospital whilst I was carrying out an investigationinto osteoarthritis.

REFERENCESBENNETT, G. A., WAINE, H., and BAUER, W. (I942),' Changes

in the Knee Joint at Various Ages,' New York: CommonwealthFund.

COLLINS, D. H. (1949), 'The Pathology of Articular and SpinalDiseases,' London, Edward Arnold & Co.

HARRISON, M. H. M., SCHAJOWICZ, F., and TRUETA, J.(195 3), y. Bone J. Surg., 35-B, 598.

INGELIARK, B. E., and EKHOLM, R. (1948), 'A Study onVarittions in the Thickness of Articular Cartilage in Associationwith Rest and Periodical Load: and Experimental Investi-gation on Rabbits,' Upsala Lakareforenings Fordhandlinger,53, 61.

KELLGREN, J. H., and MOORE, R. (I952), Brit. med. J., i, x8z.KEY, J. A. (I929), . Bone . Surg., 11, 705.LINDELLS, J. W. (1953), Ibid, 35-B, 643.LLOYD-ROBERTS, G. C. (950), Ibid, 35-B, 627.LLOYD-ROBERTS, G. C. (1955). Ibid, 37-B, 8.MATTHEWS, B. F. (I953), Brit. med. J., ii, 66o.

by copyright. on M

ay 10, 2021 by guest. Protected

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.31.362.618 on 1 D

ecember 1955. D

ownloaded from