TRENDS IN THE TREATMENT OF OSTEOARTHRITIS OF THE …

397

PRESENT TRENDS IN THE TREATMENT OFOSTEOARTHRITIS OF THE HIP

With an addendum for Fellowship candidates

By M. H. M. HARRISON, Ch.M., F.R.C.S.Senior Registrar in Orthopaedics, Westminster Hospital, London

The practice of orthopaedic surgery is under-going extensive changes and the treatment ofosteoarthritis of the hip reflects this in a challeng-ing manner. Systematic surgery for this diseasebegan to be practised on an increasing scale in theyears following the first world war. An arthro-desis, once successfully achieved, was appreciatedto be of great value; doubtless it was in part thedifficulties and morbidity of this operation whichled surgeons to work out other methods to relievethe pain of the osteoarthritic hip. So-called re-constructions of the Whitman type made a crudearticulation from the distorted joint; McMurraydescribed the use of intertrochanteric osteotomyand Girdlestone evolved the pseudarthrosis opera-tion. These were the main procedures practisedby British orthopaedic surgeons during most of theyears between the wars; it was essentially salvagesurgery.

Hopelessly deranged hip joints were eitherstiffened by arthrodesis or were refashioned oradjusted by the other relatively crude methods.Both surgeon and patient recognized that the endresult fell far short of a normal hip-a painless,mobile and stable joint. Each method aimed atproducing a painless joint but differed from theothers in the mobility and stability that followedits performance. Considered as salvaging operationsthey were, and as we shall see below still are,very successful.

In the last decade another era of therapeutics forthe osteoarthritic hip has been established. Twofactors combined to increase both the scope ofsurgery and the number of patients to whomoperation could be offered. Firstly, advances inanaesthesia and in transfusion techniques havemade operations much safer. Secondly, muchhas become known about the tolerance of inertmaterials by the tissues of the body, and operationsof prosthetic replacement have become common-

place in the treatment of osteoarthritis of the hip.Such operations aim at reconstructing a normal ornear normal hip from the joint wreckage, to pro-vide hips not only painless but stable and mobile as

well-clearly different from the salvage operations.There are two main operative techniques: In

the interposition arthroplasty of Smith-Petersena metal cup is placed between the remodelledfemoral head and acetabulum, the cup is fixed toneither bone and induces a fibro-cartilaginouscovering on both bone ends. In the replacementarthroplasties a portion of the upper end of thefemur is resected and replaced by a prosthesisfitted into the bone. The first of such techniquesto find wide application was the acrylic femoralhead introduced by the Judets; seldom can anoperative technique have been so widely employedin so short a time.

In deciding the best treatment for any particularpatient the following points must be considered:

i. No patient is operated upon until a thoroughtrial of physiotherapy has been given, and ifnecessary repeated, without obtaining satisfactoryrelief from pain.

2. The great majority of operations are per-formed to relieve severe and disabling pain, a muchsmaller number to correct deformity. Accordinglyit is essential to estimate how much pain the patientsuffers and what inroads it makes into his life.

3. The surgeon explains to the patient the levelof functional activity likely to follow operation;the patient can then see how this will fit in withhis needs at work and at home.

4. The lumbar spine, the other hip and theknees are examined; the findings may indicatewhether the affected hip can be stiffened or mustbe left mobile.

5. A general physical examination determineswhether the patient can stand a major hip opera-tion. An assessment of the patient's emotionalmake up is important when considering arthro-plasty as this operation requires considerable co-operation throughout a long post-operative courseof physiotherapy.The final decision regarding operation is made

with reference to these five points and to thepreference of surgeon and patient. The tech-niques now in common use will be briefly discussed.

copyright. on June 2, 2022 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.31.358.397 on 1 August 1955. D

ownloaded from

http://pmj.bmj.com/

398 POSTGRADUATE MEDICAL JOURNAL August 1955

ArthrodesisAlthough arthrodesis destroys the joint, success

depends upon obtaining bony union betweenfemur and acetabulum; this requires considerableskill. Osteogenesis is hindered by the long leverof the lower limb which tends to disturb thejunctional tissues at the slightest movement.Intra-articular arthrodesis is the method of choiceand most surgeons reinforce this by metallic in-ternal fixation, by a bone graft or by both. Thedisadvantages of arthrodesis are the difficulty ofsecuring bony ankylosis, the duration of plasterimmobilization which is relatively ill tolerated byolder patients and the risk of stiffening the knee bysuch immobilization. Even a partially stiff andpainful knee is serious in the presence of anarthrodesed hip, and new techniques are stillbeing devised to overcome these difficulties. Thefunctional result of an arthrodesis is excellent andthis operation finds its greatest application in theunilateral osteoarthritic hip of the youngish patient.The hip is completely stable and painless, ex-cellent for walking and standing, less well adaptedto sitting. The optimum position of ankylosis ofthe adult hip is neutral rotation, flexion 25° and

:i:

··:. *:

.:·;:··:,::·

iiiit::iii·ii

'i

':?

;C';"

·?:

FIG. I.-Central dislocation arthrodesis (Mr. J.Charnley's case). The hip having been widelydislocated the femoral head and neck are cut to theform of a cylinder and pushed through a hole madein the acetabular floor. The medial end of thereshaped head now lies within the true pelvis. Ametal clamp, not shown here, which compresses thegreat trochanter to the ilium is buried in the woundin those cases where bony ankylosis is sought; theclamp is removed after six weeks.

abduction only to compensate for any realshortening.

Charnley (I953) has approached the problem ofhip arthrodesis in a new fashion. His central dis-location technique either produces an arthrodesisof the hip (Fig. i) or results in a fibrous ankylosisallowing 10° to 30° of flexion, but without move-ment in other directions. This desirable state ofaffairs prevents a Trendelenberg gait and is due tothe central dislocation creating a bone block tolateral and rotary motion. The procedure hasmany impressive features; in Charnley's hands itappears technically straightforward, is not shock-ing, fixes the knee post-operatively for only onemonth, all plaster is discarded at six weeks andthe patient leaves hospital in about a furtherfortnight. The author was recently privileged tosee a follow-up review of 30 patients who had beenoperated during the last six years; the degree offreedom from pain and the high level of functionalactivity of the group as a whole was mostimpressive.Resection of the Femoral Head and Neck,the Pseudarthrosis Operation

This is neither a difficult nor a shocking pro-cedure and is well tolerated by the reasonably fitpatient of 70 years (Fig. 2). It was practised andtaught by G. R. Girdlestone, whose technique andresults are described by Taylor (I950). The opera-tion results in a mobile hip, a short leg (I- in.) anda limp due to a positive Trendelenberg; the

FIG. 2



j.bmj.com

/P


j.31.358.397 on 1 August 1955. D

ownloaded from

http://pmj.bmj.com/

August 1955 HARRISON: Present Trends in the Treatment of Osteoarthritis of the Hip 399

majority but not all the patients thus have per-inanent need of a stick. It is effective in relievingpain and younger patients have been enabled towalk long distances, drive a tractor, etc., after itsperformance. Patients sit very well after thisoperation; it can be performed bilaterally. Theprocedure is very useful for the older patient andthose cases where it is obligatory not only to relievepain but to leave a mobile hip. Batchelor (I940)has added an abduction osteotomy to the originalprocedure in an attempt to increase the stability.Intertrochanteric Femoral Osteotomy

This was first performed for osteoarthritis of thehip by McMurray (i939), who aimed to displaceShe distal fragment medially beneath the acetabu-um (Fig. 3). It is a straightforward and simpleoperation which aims to relieve pain and correctdeformity; the range of motion in the hip is notincreased. As originally performed the operationhad the same disadvantages of immobilization asarthrodesis, but now some surgeons are usinginternal fixation to obviate this. As the popularityof acrylic arthroplasty wanes osteotomy is beingincreasingly performed in many centres; a mostinteresting fact which has become apparent is thatpain relief does not seem to depend upon themaintenance of displacement at the osteotomy site

I

FIG. 3.-A tracing of a radiograph showing the dis-placement that has been obtained after intertro-chanteric osteotomy; the femoral fragments unitein this position.

-equally satisfactory results have been obtainedafter osteotomies in which the bone fragmentsappear to have returned to their original position.ArthroplastyOne problem is common to both types of arthro-

plasty, the method of approach. Both the anterior(Smith-Petersen) and the posterior (Gibson)approaches and their modifications are used ex-tensively, neither enjoys an overwhelming superi-ority over the other. In reconstruction of a hip itis necessary to preserve normal gluteus mediusaction to prevent a Trendelenberg gait, and tomaintain the tension in the short rotator musclesand thus maintain stability. New operativeapproaches are still being devised in an attempt toensure these desirable happenings (McFarland andOsborne, I954).Cup arthroplasty entails a thorough exposure of

the hip joint, remodelling of both femoral headand acetabulum and, finally, the insertion of avitallium cup which covers the femoral head andfits into the acetabulum. The technique of Smith-Petersen (I939) has remained largely unmodified.Adams (I953) has recently introduced the con-centric cup which aims to provide a bettermechanical arrangement within the new joint; thelong-term results must be awaited with interest.The original replacement arthroplasty utilized

the Judet acrylic prosthesis (Judet, J. and R., etal., 1954), and this is the one which has been themost popular in this country. The hip is dis-located, the femoral head is resected and theacetabulum is remodelled. The prosthesis isthen inserted by driving its stem into a channelprepared in the length of the femoral neck.

If the acrylic femoral head had an enthusiasticreception, it has equally had extensive modifica-tion. Between 1950 and I953 it largely supplantedthe cup but the popularity of replacement arthro-plasty is now very seriously threatened. Innumer-able alterations have been made since the prototypewas introduced; they take two chief forms. Theconstitution of the prosthesis is varied wherebydifferent materials are used either singly or incombination. More significant, however, is thechanging shape of the prosthesis-the overallchange is towards the use of bigger prostheses en-tailing resection of more of the original articulation(Fig. 4). Thus some of those who are exploringthe possibilities of this type of surgery are replac-ing the whole of the femoral head and neck by aprosthesis whose fixation has extended from theoriginal 24 in. stem of the Judet prosthesis to amedullary nail which goes almost halfway downthe femoral shaft. The prosthetic joint has alreadybeen used in the knee and the elbow and is beingcontemplated for the hip!



j.bmj.com

/P


j.31.358.397 on 1 August 1955. D

ownloaded from

http://pmj.bmj.com/

400 POSTGRADUATE MEDICAL JOURNAL August i955

A. C

FIG 4.-Diagrams of some prostheses used for osteo-atthritis of the hip:

A, early, and B, the later type of Judetprosthesis.

C, Thompson type.D, Merle d'Aubign6 type.

These changes reflect both the disappointmentsand the hopes of those engaged in this phase oforthopaedic surgery. The results of replacementarthroplasty have disappointed a majority of sur-geons who had hoped so much from it, and thisdisappointment is being met by some who aretrying to find the way whereby the occasionalexcellent result may be achieved in large numbers.

Replacement arthroplasty may fail due to:I. The use of materials which fracture, deform

or produce tissue reaction.2. Post-operative subluxation or dislocation of

the new joint. It is not difficult to insert anacrylic head into a femoral neck, but it is difficultto place it in a predetermined position after havingremoved just the right amount of bone from thejoint to leave a fit which is neither too loose nortoo tight.

3. Dislocation is prevented by skilled surgeryand nursing; late loosening of the prosthesis isprobably only in part dependent upon the tech-

nique of operation. The evolution of prostheseswhich replace more and more of the upper end ofthe femur and whose fixation extends progressivelydown the shaft is one way whereby this problem oflate loosening of the prosthesis is being attacked.These rather formidable looking appliances are-probably receiving a slow and cautious applicationin the traditionally conservative schools of Britishorthopaedics.The commonest cause of disappointment after

arthroplasty is the onset of pain and stiffness in thereconstructed joint. This is in part connectedwith the problems just discussed, but will be con-sidered again below.

In summary, the arthroplasties require thegreatest surgical skill of all operations on theosteoarthritic hip if they are to be successful.They require prolonged post-operative physio-therapy; the early results are frequently very en-couraging, the late results are not infrequentlydisappointing. While the two-year follow up re-



j.bmj.com

/P


j.31.358.397 on 1 August 1955. D

ownloaded from

http://pmj.bmj.com/

August 1955 HARRISON: Present Trends in the Treatment of Osteoarthritis of the Hip 401

B AFIG. 5.-Arteriograms, obtained postmortem, of the femoral heads of a subject aged 76. The right, A, is

the site of advanced osteoarthritis. The left, B, is essentially normal. The arteries within the osteo-arthritic head are seen to be dilated and increased in number. (Reproduction by kind permission ofthe Editor of the Journal of Bone and Joint Surgery.)

view is of interest, it is the five to ten year resultwhich is of significance.CommentThe difficulties and disappointments of the

orthopaedic surgeon and his patient assume acorrect perspective when viewed against certainfacts. Osteoarthritis is a painful disease, surgeryaims to relieve the pain and yet we are completelyignorant of the origin of this symptom. Perhapsthe capsule, synovium and the bone all contribute,but to say more is speculation. This is less sur-prising when one remembers that very little isknown of the nerve supply of bone.We can only guess at the factors controlling the

resorption and production of bone, how thenshould we prevent the loosening of prostheses ornew bone formation around reconstructed joints?

It is well recognized that certain methods em-ployed in physiotherapy relieve joint pain. Heat inone form or another is very often effective, yetwhat is known of the mode of action of the justlypopular short wave diathermy? Recent work hassuggested that as usually applied its heating effectis largely confined to the superficial tissues (Scott,I955), and that whereas it produces an increasedcirculation in the skin the blood flow in deeperorgans may be decreased (in the kidney, Kottkeet al., 1949; in muscle, Harris, I955). It isinteresting to compare this last piece of evidencewith the findings that the vascular tree within theosteoarthritic femoral head is grossly abnormal,being in a condition of widespread arterial hyper-aemia and venous dilatation (Fig. 5) (Harrison,

Schajowicz and Trueta, I953). It seems possiblethat at least some of the pain of this diseasemight have a vascular basis and Trueta is atpresent exploring the possibilities of relieving thepain of osteoarthritis by local intraosseous X-irradiation. The finding of this deranged vas-cular bed perhaps throws some light on thesuccesses which have been known for a long timeoccasionally to follow external X-irradiation inthis disease.

In the light of these deficiencies in knowledgeour attempt to build new joints by arthroplasty canbe compared to the attempted construction of abridge designed to carry heavy traffic withoutintimate knowledge of the principles of engineer-ing and mechanics. Small wonder that the result isunpredictable.

It is fascinating to speculate on the difficultiesand the future of arthroplasty. The problem of thereconstruction of a durable joint after the resectionof an osteoarthritic one is a biological problem,mechanical solutions must be biologically accept-able. There are at least two major difficulties tobe considered.

Firstly, evidence has been offered elsewhere forthe thesis that osteoarthritis represents the re-sponse of the tissues of a joint to unfavourablemechanical circumstances (Harrison, Schajowiczand Trueta, 1953). If this is correct, then evengranted a perfect prosthesis, what is to prevent there-appearance of osteoarthritic lesions in the livingtissues adjacent to the prosthesis, if it merelysubstitutes a portion of the diseasedjoint and leavesthe original abnormal anatomy and mechanics un-



j.bmj.com

/P


j.31.358.397 on 1 August 1955. D

ownloaded from

http://pmj.bmj.com/

402 POSTGRADUATE MEDICAL JOURNAL August 1955

changed? Such lesions are found after arthro-plasty both in the acetabulum and the femoralneck stump (sclerosis and eburnation of bone,cyst formation and osteophytes) and may wellaccount for a return of symptoms. Operationswhich would correct the underlying anatomy inaddition to joint reconstruction might be con-sidered by many to be altogether too formidable.

Secondly, there arises the problem of a methodof durable and perfect fixation of an inert materialto living bone. It is precisely here that ortho-paedic surgery differs from carpentry and engineer-ing. Great forces are transmitted by the hip butthe tissues of the normal joint successfully standup to them. However securely a prosthesis may bebolted or otherwise fixed to the femur, it seemslikely that the trabeculae at the site of fixationcannot survive unless the forces to which they areexposed can be kept down to something like thosenormally prevailing. It may very well be that suchcan never be obtained in this, the most majorweight bearing joint in the body. Scales (i954) isattempting to overcome these difficulties by usingmetal plate fixation only as a temporary measurecontriving finally to secure the shaft of the pros-thesis by new bone formation.Cup arthroplasty should theoretically be success-

ful in osteoarthritis secondary to a local deformityof femoral head or acetabulum which can be cor-rected at the time of operation (viz., old Perthe'sdisease, slipped epiphysis, malunion of intra-articular fractures, congenital subluxation of thehip). For the reasons just advanced replacementarthroplasty may prove unsuccessful in these con-ditions, as may both forms of arthroplasty inosteoarthritis due to anatomical defects occurringat an extra-articular level-old congenital disloca-tion, coxa vara and valga, anteversion of thefemoral neck. In such patients, arthrodesis,osteotomy or pseudarthrosis may still provide thegreatest percentage of satisfying results.The follow-up studies of De Vas (I954) and

Shepherd (1954) lend some support to thesebeliefs in that they show the results of cup arthro-plasty to be more durable than those of replace-ment. What would be most interesting, however,would be to compare the results of cup arthro-plasty in the different aetiological groups justmentioned, i.e. intra-articular causes and 'path-ology at a distance.'The most worthwhile approach to the problem

of osteoarthritis of the hip is that of prophylaxis.Increasing skill in the treatment of dislocation andother congenital deformities of the hip, of Perthe'sdisease and slipped epiphysis in the first twodecades, will reduce the incidence of clinicalosteoarthritis in the adult.

AddendumIf a candidate in a higher surgical examination

has been given 15 to 20 minutes to examine apatient with an osteoarthritic hip he should facehis examiners armed with the following informa-tion:

I. How is the affected hip or hips troubling thepatient? What inroads is it making into his life?Does it disturb sleep, necessitate analgesics,affect employment, cause him to use a stick? Isthere any history of bone disease or injury in earlieryears which might be of aetiological significance inthe present arthritis?

2. Clinical examination should yield:(a) The degree of any fixed deformity of the

affected hip-most osteoarthritic hips lie in aposition of flexion, adduction and external rota-tion. The examinee will be familiar withThomas' test for flexion deformities of the hip.

(b) The amount of any real or apparentshortening. The examinee should be able todiscuss the reasons for any differences in thesemeasurements. If there is any real shorteningit will be necessary to determine in which seg-ment of the lower limb this is occurring. Thefirst step in such elucidation is to constructBryant's triangles which will reveal any shorten-ing occurring above the trochanteric region ofthe femur. Should Bryant's triangles be ofequal dimensions on both sides, the cause ofany real shortening must be sought for moredistally than the hip joints.

(c) The range of motion in the affected hip,considering flexion, abduction, adduction, medialand lateral rotation and extension in turn. Thecardinal physical signs of an arthritis is thelimitation of motion in the affected joint to anapproximately equal extent in all three planes.

(d) The presence of wasting of the thigh.(e) Is Trendelenberg's sign positive or nega-

tive?(f) An estimate of the mobility of the lumbar

spine and the joints in both lower limbs.(g) It is a useful precaution to conclude the

examination by eliciting the ankle and pupillaryreflexes; a Charcot joint may be painful in itsearly stages of development but will allow anabnormal range of movement.

AcknowledgmentsI would like to express my thanks both to Miss

J. Hassell for her drawings and to the PhotographyDepartment of the Westminster Hospital.Bibliography continued on page 413



j.bmj.com

/P


j.31.358.397 on 1 August 1955. D

ownloaded from

http://pmj.bmj.com/

August I955 CLINTON-THOMAS: Regional Colitis 4.3

PrognosisWhile the immediate results of resection in

acute regional colitis appear to be good, few caseshave been reported and none has had a longfollow up. In the case reported here and the onedescribed by Thompson (personal communication)there has been no evidence of recurrence afterfour and five years respectively. In chronicregional colitis, recurrence following excision orshort circuiting is unfortunately by no meansuncommon. Neuman et al. (1954), record arecurrence rate of 37 per cent. after surgicalprocedures but they point out that their figuresare based on a small number of cases and that thefollow up has been short in most. Relapse aftera long interval of apparent cure is not unknownand Butler (1953) records a case in which thisoccurred after eight years.SummaryA case of regional colitis, which first presented

as an acute abdominal emergency, is reported.Following a caecostomy the affected segment ofbowel rapidly stenosed and was subsequentlyresected. A four year follow-up shows noevidence of recurrence.The literature is reviewed and the pathology

and treatment are discussed.I wish to thank Mr. J. A. McLauchlan, under

whose care the patient was admitted, for permis-

sion to publish this case; Dr. T. Culoty for theradiological reports; and Dr. L. Steingold for thepathological report.

I am indebted to Dr. Lynne Reid and thePhotographic Department of the BromptonHospital for the photomicrography.

BIBLIOGRAPHY

ARMITAGE, G., and WILSON, M. (1950), Brit. J. Surg., 38, 182.BARBOSA, J. DE C., BARGEN, J. A., and DIXON, C. F. (x945),

Surg. Clin. N. Amer., 25, 939.BARGEN, J, A. (1943), 'The Moder Management of Colitis,'

C. C. Thomas, Springfield, Ill.BARGEN, J. A., and WEBER, H. M. (1930), Surg. Gyn. Obst.,

50, 964.BROWNLEE, T. J. (x95 ), Brit. J. Surg., 38, 507.BUTLER, E. C. B. (I953), Proc. Roy. Soc. Med., 46, 69.CORBETT, R. S. (1945), Ibid., 38, 277.CROHN, B. B., and BERG, A. A. (1938), J. Amer. med. Ass.,

O10, 32.CROHN, B. B., GARLOCK, J. H., and YARNIS, H. (1947),

Ibid., I34, 334.CROHN, B. B., GINSBERG, L., and OPPENHEIMER, G. D.

(1932), Ibid., 99, 1323.CROHN, B. B., and ROSENAK, B. B. (1936), Ibid., 136, I.DALZIEL, T. K. (1913), Brit. med. J., i, 425.LUMB, G. (1951), Brit. J. Surg., 39, 233.MOSCHOWITZ, E., and WILENSKY, A. G. (I923), Amer. J.

med. Sci., i66, 48.MOYNIHAN, B. G. A. (I907), Edin. med. J., 21, 228.NEUMAN, H. W., BARGEN, J. A., and JUDD, E. S. (1954),

Surg. Gyn. Obst., 99, 563.NEUMAN, H. W., and DOCKERTY, M. B. (1954), Ibid., 99,

572.ROBSON, A. W. MAYO (1908), Brit. med. J., i, 425.THOMPSON, H. R. (1950), Proc. Roy. Soc. Med., 43, 685.TIETZE, A. (1920), Erg. Chir. Orthop., 12, 211.WELLS, C. (1952), Ann. Roy. Coll. Surg. Eng., Ii, o05.

NOTICE OF SPECIAL INTEREST TO SUBSCRIBERS: WHY N"WHY NOT HAVE YOUR COPIES OF THISJOURNAL BOUND INTO YEARLY VOLUMES?"

Arrangement have now been made to have the twelve monthly issues fully bound HAVE Uin dark green pin head coth, lettered in gilt on spine with name of Journal,Volume Number and year, complete with index at front, 17s. 6d. per Volume, postfree. A limited number of out of print journals are available to bind into volumesand make your library complete. Price on application giving details of issues J URrequired to complete back volumes.

THE FELLOWSHIP OF POSTGRADUATE MEDICINE BOUN60 PORTLAND PLACE, LONDON, W.I

Bibliography continued from page 402-M. H. M. Harrison, Ch.M., F.R.C.S.

BIBLIOGRAPHYADAMS, J. C. (1953), J. Bone Jt. Surg., 35B, 199.BATCHELOR, J. S. (I948), Postgrad. med. J., 24, 24I.CHARNLEY, J. (1953), 'Compression Arthrodesis.' E. and S.

Livingstone, London.DEVAS, M. S. (1954), J. Bone Jt. Surg., 36B, 561.HARRIS, R. (1955), Personal communication.HARRISON, M. H. M., SCHAJOWICZ, F., and TRUETA, J.

(x953), Y. Bone t. Surg., 35B, 598.JUDET, J., JUDET, R., LAGRANGE, J., and DUNOYER, J.

(1954), ' Resection-reconstruction of the hip; arthroplasty withan acrylic prosthesis.'

KOTTKE, F. J., KOZA, D. W., KUBICEK, W. G., and OLSON,M. (I949), Arch. phys. med., 30, 431.

McFARLAND, B., and OSBORNE, G. (1954), J. Bone Jt. Surg.,36B, 364.

McMURRAY, T. P. (I939), J. Bone Jt. Surg., 21, I.SCALES, J. (I954), Personal communication.SCOTT, B. O. (x955), Personal Communication.SHEPHERD, M. M. (s954), J. Bone Yt. Surg., 36B, 567.SMITH-PETERSEN, M. N. (1939), J. Bone Jt. Surg., 21, 269.TAYLOR, R. G. (I950), J. Bone Yt. Surg., 32B, x6I.



j.bmj.com

/P


j.31.358.397 on 1 August 1955. D

ownloaded from

http://pmj.bmj.com/

Documents

TRENDS IN THE TREATMENT OF OSTEOARTHRITIS OF THE …