chapter XXXIII

.[32]) It may be enforced by fixed dressings of plaster, etc.

[32] The fundamental idea expressed in all of the methods for enforcing rest by traction is of American origin, and constitutes one of the advances in surgery for which the world is indebted to America. For a long time it was referred to in Germany as the American method, and yet now the Germans claim so much for it that one of their surgeons has written a book of 600 pages devoted to the employment of traction for various surgical purposes, in which but very little credit is given to the men who originated it.

[Illustration: PLATE XXXIV

Normal Knee-joint. (Child, seven years old.)

Tuberculosis of Knee, with Partial Dislocation.

(Child, seven years old.)]

With a better appreciation of the pathology of the condition numerous methods were devised by which the germs should be attacked _in loco_. Thus various antiseptics have been injected in varying strengths, either into joint cavities or around them. Lannelongue devised a “sclerotic method,” by which zinc chloride solutions were injected into the peri-articular tissues, to so condense and harden them as to imprison and destroy their contained germs. The method, however, is an extremely painful one and has not found general favor. For a long time iodoform was employed for the same purpose, in emulsions of 10 per cent. and 20 per cent. strength, in sterilized glycerin or olive oil. It affords a curious paradox that the iodoform itself must be sterilized before being thus used. This emulsion has been injected into the peri-articular tissues or into joint cavities, which, when containing appreciable amounts of fluid, should be first emptied and washed out; all of which can be done through the same small trocar used for introduction of the iodoform. The verdict of surgeons today is rather against the employment of iodoform, since they have learned to not rely upon it because of disappointment so often following its use.

Bier, in 1891, advised the so-called _congestion treatment_ of tuberculous joints, basing it upon the fact that tuberculosis does not develop in lungs which are the seat of venous stasis from valvular heart disease. He proposed to produce an artificial stasis, in the joint structures and about them, by which living germs should be destroyed and their disease products encapsulated, claiming that as the result of the hyperemia thus produced the alexins are thus brought into more complete contact with the bacilli. The method is applicable to the limbs below the shoulder and hip. It consists in the application of an Esmarch bandage above the affected joint, applied with sufficient firmness to obstruct the returning blood, but not to interfere with the arterial supply. If there be room the limb is also bandaged below the joint with an ordinary cotton roller. This congestion is kept up at daily intervals for increasing periods, beginning with perhaps half an hour and continuing until it is in operation at least half of the time. Meantime other methods of treatment are not interdicted. In the earlier stages of tuberculous joint disease this method has given very encouraging and pleasing results. (See Fig. 208.)

_Tuberculous hydrops_ may be treated by aspiration and elastic compression. Should fluid distend the joint it should be opened and thoroughly cleaned, then closed and perhaps drained.

The treatment of _pyarthrosis_ and of _peri-articular cold abscess_ has long been a mooted subject. The orthopedic surgeons still adhere to mildly or absolutely non-operative measures, whereas the general surgeon prefers to adopt more radical methods. Each case should be judged on its own merits, and these should include a careful estimation of the general condition of the patient. Should evidences of septic intoxication be present or the ordinary general signs of the presence of pus, then these collections should be opened and cleaned out. If hectic can be excluded, then other considerations will indicate what is best. At all events there will be seen many cases where a delay in operation will be advisable, in order to permit of improvement of the general condition by measures above described. To merely open up a tuberculous focus and leave at least two fresh raw surfaces exposed to contamination is rather to invite the spread of the disease than to correctly meet the indication. Every old focus will be lined or surrounded with a more or less dense membrane formerly called _pyogenic_, but now more correct knowledge shows it to be _pyophylactic_. (See p. 113.) To leave this _in situ_ is to leave germ-laden walls, while to dissect it thoroughly is to make a larger, fresh raw surface and to open up innumerable absorbent vessels. Thus, whether it be removed in whole or in part, or allowed to remain, some sufficiently strong caustic material should be promptly employed, by which both destruction of living residual germs and closure of the mouths of the absorbents shall be effected. This has been set forth more fully when dealing with cold abscesses in general, but is of so much importance that it may be reiterated here. Whether the actual cautery, pure carbolic acid, strong zinc chloride solution, or some other agent be used should depend upon circumstances, but every portion of the surface which it is proposed to leave more or less exposed to the possibility of infection should be thus protected. In proportion to the intensity of the caustic action there will be separation of more or less cauterized and sloughing material, for whose escape provision should be made; but it will be separated by the granulation process, aided by an active phagocytosis, and when removed will leave a granulating surface which is but slightly absorbent. These facts pertain to small incisions for drainage as well as to extensive arthrectomies.

_The operative treatment_, then, of tuberculous arthritis varies from tapping, with or without drainage, to complete arthrectomy or amputation. When the joints of the foot or ankle are extensively diseased, and the patient, as usually happens, is in poor condition, it may appear that amputation will afford the most complete relief, and that a stump with an artificial member will be of much more use to the individual than a mutilated, tender, and disabled foot.

[Illustration: FIG. 208

Calcified mass in old “cold abscess” about hip-joint. (Buffalo Clinic. Skiagram by Dr. Plummer.)]

To incision with or without drainage is given the name _arthrotomy_. When the joint is widely opened and portions removed with the sharp spoon or otherwise, it is known as _arthrectomy_. When bone is removed irregularly the measure is called _atypical resection_. When entire bone ends are removed the operation becomes an _exsection_ or _resection_. The ordinary arthrectomy is not sufficient when foci are present in the epiphyses. Here at least atypical resection is called for. Arthrectomy may properly include a wide exposure of articular surfaces and the removal of the thickened and diseased synovia, with its fringes, or with the cartilages, by which cancellous structure is more or less widely exposed. When arthrectomy is undertaken it should be thoroughly made and by a large incision, since the more completely the joint cavity can be inspected and attacked the better are the interests of the patient subserved. All fresh or cold abscess cavities which connect with the joint or lie in contact with it should also be attacked at the same time, and those which do not communicate with it should be separately drained. While drainage by tube or other means will usually suffice, there are cases where the disease is so extensive that it will pay to pack the cavity with balsam gauze for a few days, placing secondary sutures by which the incision can be closed after its removal. In the shoulder and hip, for instance, such a method will give satisfactory results.

The advantage of avoidance of resection is the non-interference with the epiphyses and their junctions, thus permitting the growth of the bone to continue. _Therefore complete and typical excisions should be practised as seldom as possible, especially in growing children._ They may be practised to advantage even in advanced age, and the writer has seen satisfactory results after complete excision of tuberculous joints in senile cases. When operating upon a tuberculous tarsal joint the surgeon is likely to find one or more of the tarsal bones so much involved in the tuberculous disease that he is compelled to scrape it out and thus leave a cavity almost the size of the bone itself. Should he have to do this to a series of the bones it would be better to make a formal resection of the tarsus or possibly an amputation. The cavity should be left open with a sufficiently large incision so that it may be easily packed. A cavity of this kind left unpacked will fill up with clot, which will disintegrate and the result will be much less satisfactory. In the former case there is an open cavity which fills with granulations, but this can be kept accessible under observation and with more effect and comfort. This is equally true of those cavities where both arthrectomy and bone curettage have been practised.

MOVABLE BODIES IN THE JOINTS.

Several different terms have been applied to loose and movable bodies, even in the various joints, depending on their size, arrangement, and appearance. Thus we have the _rice-grain_ or _melon-seed bodies (corpora oryzoidea)_, which have already been described and are now supposed to indicate a form of tuberculous synovitis which has undergone a partial if not complete subsidence. Again we have larger masses occurring singly or in very small number, especially in the knee, to which the Germans have given the significant name of _joint mice_. Also in the knee, owing to its peculiar construction, another form of movable body is met with, _i. e._, a displaced and more or less _motile semilunar cartilage_. This condition was first described by Hey, and especially studied by Allingham, who made it a prominent feature of what he described as “internal derangement of the knee.” Lastly, in those joints in which _synovial fringes_ occur, the knee especially, it is held that portions may become detached by having been infiltrated and cast off or broken loose, and thus form a fourth variety of floating body. The joints most often affected are the knee and the elbow. In many instances there is a history of injury, especially when the mass is of considerable size. The theory of an “osteochondritis dissecans” has also been invoked to account for the resemblance between some of these bodies and the articular cartilages. Some pathologists have held that they may result from the organization of clots, which are subsequently rounded off and shaped by attrition (Fig. 209). These bodies then may consist of condensed fibrinous material, of cartilage, of true bone, or of hyperplastic and fatty synovial tassels. To these may be added rare instances of mucoid connective tissue.

[Illustration: FIG. 209

Floating bodies--“joint mice”--from knee-joint. (Lexer.)]

=Symptoms.=--_Rice-grain bodies_ may be suspected in cases of chronic tuberculosis and often in arthritis deformans, while in many instances they may be felt gliding beneath or between the joint structures. A perfectly loose _floating body_ will produce symptoms which are quite distinctive. They consist of sudden and intense pain, with such muscle spasm as to fix the joint and prevent its use, thus “locking it.” Occurring at the knee the individual is instantly disabled, but usually learns by some peculiar manipulation, with or without assistance, to “unlock” the joint, and after a few moments to resume its use. Such a complaint as this should always suggest the condition. Patients who have had it for a long time learn how to avoid it as well as how to relieve it, and will often discover and be able to indicate to the surgeon the existence of a movable body, and even to describe its usual resting place.

## Partial or complete _dislocation of a semilunar cartilage_ in the knee

is usually the result of traumatism, a distinct history of which can generally be obtained. It may not have been discovered at the time, owing to swelling or tenderness, but will produce its peculiar symptoms later, _i. e._, after use of the joint is resumed. Here, again, so long as it remain in proper position, it interferes but little; with a misstep or sudden movement, however, the patient is seized with sudden and painful disability. Here the movable cartilage may be felt projecting near its proper location. In such cases as these it is movable only to a certain extent and makes no free excursion about the joint. When not detected it may be suspected from the description which the patient gives of his seizures.

=Diagnosis.=--So far as diagnosis is concerned, when a movable body can be felt all doubt is set at rest. When it cannot be discovered its existence may be inferred with an accuracy proportionate to the patient’s description of his difficulties.

=Treatment.=--The treatment of rice-grain bodies is essentially that of the chronic hydrarthrosis and probably tuberculous condition which have led to their formation. It will consist usually in arthrotomy, with thorough irrigation; often in some form of arthrectomy. With the larger floating bodies, the “joint mice,” the most radical measures are the best. In most of these instances there will be some degree at least of hydrarthrosis. The joint cavity being distended and relaxed, the indication for arthrotomy is the more urgent, since it will permit also of irrigation or of dry sponging, with the same benefit with which analogous intraperitoneal conditions are treated by the same measures. The joint may be opened by a sufficiently ample incision, through which the foreign body or bodies may be removed. The operator should not be satisfied with mere removal of one, but should make a thorough search for others which may have escaped previous detection.

Perhaps no operative measure in surgery better illustrates the advantages of asepsis. This operation, which now can be done with impunity, was in the pre-antiseptic era one which had a discouraging fatality, death resulting from septic infection in about 40 per cent. of cases.

FOREIGN BODIES IN THE KNEE-JOINT.

“Joint mice” are of sufficient frequency and significance to justify brief separate consideration. According to Connell these may be grouped as follows:

Those composed of foreign material, fatty tissue, fibrous tissue, etc.;

Those composed of bone, cartilage, or of a mixture of the two.

Among the many explanations offered are the following:

Dry arthritis, with overgrowth of the margins of the cartilages;

Bony growths, separation from their attachments;

Infarct of the articular cartilage, with final separation;

Plate of bone formed outside of the joint and then invaginated;

Calcification or chondrification of enlarged synovial fringes;

Irritation and growth of embryonal cartilage or bone cells in the synovial fringes;

Concretions whose nuclei are clots, torn fringes, or some foreign body;

Some portion of the articular cartilages broken off by injury, or damage and subsequent separation.

Injury figures largely in the opinion of most of the authorities, it being well established that an injured portion of articular surface may become subsequently detached by a fatty necrosis, spoken of by König as osteochondritis dissecans, or by Paget as “quiet necrosis.” Others imagine that these floating bodies are rarely of traumatic origin.

Symptoms are usually marked and significant. There is sudden sharp and shooting pain, sometimes so severe as to cause faintness. Along with this there is “locking” _i. e._, fixation of the joint, usually in the flexed position, probably due to the entanglement of the floating body between the articular surfaces or between the bone and the capsule.

[Illustration: FIG. 210

Ankylosis of hip with contracture of knee, following post-scarlatinal arthritis.]

It is the smaller rather than the larger bodies which give the most acute symptoms. This “locking” may last for only a few moments or for a number of hours and may or may not be followed by acute effusion. When with the above symptoms the presence in the joint of a movable mass can be made out diagnosis is complete. Some patients discover the movable body in their own joints before they go to the surgeon.

When the diagnosis is established the removal of the offending material is imperative. In the pre-antiseptic era this was an extremely hazardous operation. It is now one involving only theoretical risks. These bodies are sometimes extremely movable and slip about within the joint in a manner to almost defy removal even after the joint cavity is open. If such a body can be felt and fixed by digital pressure, or by the method of “stockading” suggested by Andrews some years ago, _i. e._, fixation by forcing sterilized pins into the tissues around it so that it cannot escape, it is then an easy matter to cut down upon it and remove it. Otherwise incision may require to be sufficiently ample to permit insertion of a finger and the general exploration of the joint before it is encountered. These bodies sometimes exist in small numbers, and it may be possible to remove several through a single opening. If the joint be opened and explored it should be done thoroughly in order that nothing may escape. After removal the capsule is closed with buried sutures, the balance of the wound closed as usual, and the limb then dressed upon a splint with absolute fixation for several days, in order to ensure physiological rest (Fig. 209).

ANKYLOSIS.

The term _ankylosis_ implies _angular deformity_, but is used to designate partial or complete fixation of joints, such fixation being usually accompanied by more or less deformity or displacement. It is a name for a condition and not for a disease, but is always produced by the latter or by injury. The term itself implies nothing as to the nature, extent, or appearance of the exciting cause. The actual cause may have been disease of the joint, of the tissues around it, or may have been the result of injury rather than of infectious or other

## active disease.

For convenience we speak of _fibrous_, _false_, or _pseudo-ankylosis_, and of that which is _bony_ or _actual_. A more accurate use of terms would lead us to refer to the former as contracture rather than true ankylosis.

_Contractures_ are the result of acute, usually septic intra-articular and peri-articular processes, where muscle spasm is a pronounced factor and where the intensity of the process has more or less weakened the joint structures. The profession is hardly in the mood to accept acute rheumatism as an infectious process. If true or not the acute rheumatic affections are frequently followed by fibrous ankylosis with contractures. Disfigurements of this kind are often produced as the result of the surface lesions of severe burns or ulcerations, followed by cicatricial contraction and the formation of dense bands and scar tissue. This is a condition which can always be foreseen and which should be guarded against with very great care. (See Treatment of Burns.) Contractures also occur as the result of certain diseases of the spinal cord, either as the result of active contraction of one set of muscles, or of paralysis, by which the opposing muscles are deprived of resistance and thus draw the limb out of shape.

_True ankylosis_ is sometimes fibrous, sometimes osseous, and occasionally both combined. The older the case the more probable is actual osseous union of joint surfaces. Bony ankylosis implies a sharply destructive type of arthritis, which may have been originally of pyogenic, gonorrheal, or tuberculous character, or else indicates a series of very slow ossific and calcific changes, such as are connected with the osteo-arthritis already described. Many of these cases are to be referred to lesions of the cord, and many of them are of polyarticular character. Fig. 195, illustrating one of the cases of so-called “ossified men” under the writer’s observation, will portray a series of lesions of this kind, most of the vertebral as well as the other joints being involved in an absolute osseous union.

[Illustration: FIG. 211

Bony ankylosis of hip. (Ransohoff.)]

[Illustration: FIG. 212

Bony ankylosis of knee. (Ransohoff.)]

[Illustration: FIG. 213

Bony ankylosis of hip with deformity. (Ransohoff.)]

When a joint is stiff bony ankylosis may be inferred. So long as there is any motion possible it is essentially of the fibrous type. The condition is one easy of recognition, and is seen in all degrees of completeness. In many instances joint fixation is accompanied by adhesions of tendons and tendon sheaths, while as time passes all the structures around a joint thus fixed become less movable and more stiffened. Even the patella may become firmly attached to the bony surface upon which it normally rests, and thus interfere with motion of the knee almost as much as though the femur and the tibia were alone involved. Occasionally one of the acute exanthems is followed by contractures of a joint, with or without actual joint lesions, by which when neglected distressing deformities are produced; such, for instance, as partial flexion and fixation of the knees, or such stiffening of the hips as to prevent the thighs from being separated. While in such cases stiffening cannot always be prevented, deformity at least can be if suitable measures instituted sufficiently early.

Figs. 211 and 212, from Ransohoff, illustrate osseous union in the hip and the knee, while Fig. 213 illustrates the deformity which may be produced by contractures and ankylosis at the hip.

The following tabular presentation of the types of ankylosis will perhaps convey the greatest amount of information in small space:

{ {Capsular {Peri-articular { { Tendinous Ankylosis, { {Extracapsular { Tendovaginal true and false { { { Muscular { {Synovial {Articular {Cartilaginous { {Osseous

Murphy has prepared the following table of the types of arthritis which lead to some of these varieties, and which may be classed as follows:

{(_a_) Primary hematogenous fibrous arthritis {(_b_) Dry fibrous arthritis. Non-traumatic { {With fracture {(_c_) Traumatic fibrous arthritis {into joint { {Without fracture { {(contusion) { { {Cryptogenetic { { { {Typhoid { {Hematogenous {Metastatic {Scarlatina Arthritis { { { {Pyemia { { { {Gonorrhea {(_d_) Suppurative { {Traumatic { { { {Tuberculous { { {Osteitis {Osteomyelitic { {Extension { {(infective) { { {Peri-arthritis (phlegmon) { { {Panarthritis {(_e_) Ossifying arthritis (primary) {(_f_) Static adhesive

=Treatment.=--The best method of treatment should be determined by the original character of the exciting cause, the duration of the condition, the amount of deformity present, and the degree of joint fixation. That which will be possible if done early will be useless if not resorted to until the case is old and chronic. In every acute or subacute condition which may threaten ankylosis every possible precaution should be taken to prevent it. If ankylosis be inevitable it should occur with the limb in the most suitable position. At the elbow, for example, this will be the right-angle position; at the knee, one with the leg almost completely extended. In the lower extremity traction with weight and pulley will serve a useful purpose in many instances, either to overcome a threatening condition or to improve one actually existant. Mechanical measures (_i. e._, use of various splints or forms of orthopedic apparatus) will sometimes be of great use. These may be arranged for the purpose of providing absolute rest, with fixation in a desirable position rather than in one which is undesirable, or they may be made with such devices as shall permit of frequent change of position.

The mildest operative measure which can be practised in these cases is manipulation, either gentle and frequent, combined with massage, or more violent and painful, such as requires anesthesia for its performance. The question of when to resort to these manipulations is one calling for the soundest judgment, as on one side the surgeon faces the possibility of setting up a renewed and more or less acute disturbance, and on the other of seeing a joint gradually stiffen, perhaps in a bad position. There is also a third difficulty, _i. e._, the necessity for continuing motion in order to prevent the re-formation of adhesions, and this in spite of the fact that it may be intensely painful to the patient. Fortunately, however, the use of nitrous oxide anesthesia usually permits this to be done as often as may be necessary with a minimum of discomfort.

Firm, fibrous ankylosis will be attacked with great hesitation by the experienced surgeon. Even though he may succeed in restoring the limb to a better position, he may feel quite positive that the patient cannot undergo the pain of the subsequent frequent handling. With bony ankyloses he may feel that nothing short of radical measures will suffice. Here it is rarely a question of restoring motility but rather of overcoming deformity. At the knee a wedge-shaped portion of the joint may be removed, its angle corresponding to the angle of deformity, and thus a crooked leg may be restored to the straight position; in fact, with a raised heel under such a limb it may be made almost as useful as ever. At the hip one may do a subcutaneous osteotomy, dividing the femoral neck either with chisel or with a small and protected saw, and then bringing the limb down into the normal position of extension, allowing the bone to repair itself, and effecting improvement only in position, or, by constantly moving it, securing a false joint; or a more formal exsection may be made and by removing the head of the femur and clearing out the acetabulum a degree of motion may be established at this point. At the wrist, elbow, and shoulder-joint resections will usually give good results if the operation be performed before the muscles have almost disappeared by atrophic processes.

Danger attaches to the performance of the so-called bloodless operations, in that there is a possibility of laceration of nerve trunks or of large vessels which may have become fixed in the condensed tissues and be torn with them. There is more danger of this perhaps at the knee than in other joints, and ruptures of the popliteal vessels and nerves have been repeatedly reported. The first attempt in breaking up such a joint should be to increase the degree of flexion. If by efforts in this direction the tissues can be first released, then there is less danger of their yielding when extension is made. Another danger which threatens in all resistant cases, and especially in elderly people, is fracture of bones. The writer has seen the upper end of the tibia as well as the neck of the humerus yield under these circumstances. In the latter event one should endeavor to prevent bony union, and thus to gain a false joint in place of the original.

In regard to the nature of the operative attacks upon the above types, the following is copied from Murphy:[33]

A. Extracapsular {1. Tendon elongation (tendoplasty). disease {2. Tendovaginitis (exsection of sheath). {3. Cicatrices (removal).

B. Intracapsular {1. Adhesive synovitis (exsection of capsule). {2. Replacement by aponeurosis or muscle.

{1. Disconnect bones. {2. Remove neighboring bony processes or { prominences. C. Osseous {3. Liberate soft parts. {4. Prevent subsequent bony contact. {5. Interpose tissue to form hygroma or fibrous { surface.

{1. Mandibular. D. Joints suitable {2. Hip. for operation. {3. Shoulder. {4. Elbow. {5. Knee.

{1. Flap formation (skin flap with fascia, or { muscular). {2. Exposure of ankylosed area. {3. Osseous separation. E. Technique {4. Transplantation and fixation of interposition { flap. {5. Replacement of bone. {6. Fixation of parts. {7. Drainage.

{1. Passive motion F. Subsequent {2. Active motion. treatment {3. Forced traction.

[33] Journal American Medical Association, May 20, 1905, p. 1573.

To the various expedients which may be adopted for making stiffened joints more useful may be given, in a general way, the term _arthroplasty_. A variety of mechanical contrivances have been resorted to in the past, operators hoping to be able to secure, for instance, a movable knee instead of one which is stiff. Artificial joints, made of celluloid, ivory, etc., have been used for experimental purposes, but while occasionally they have given good results in animals, they have rarely been satisfactory in man. For the prevention of re-adhesion, plates of celluloid, thin metal, gutta-percha, rubber, etc., have been used. These are either wrapped around a bone end or are used for lining a bone cavity, and rapidly accumulating experience is showing that this may be done with great benefit.

Thoroughness of operative work is one of the important contributing agents to the securement of wide range of motion, especially in complete removal of synovial membrane, capsule, and ligaments. Soft parts should be liberated thoroughly. Of the materials which can be interposed between bone ends in order to prevent reunion, muscular aponeurosis, with a certain amount of fatty tissue, makes the best material for interposition. When aponeurosis cannot be secured, then muscle should be tried, with some fat, as the former flattens out and undergoes structural changes, with conversion into fibrous tissue.

It should be represented to the patient as a legitimate scientific experiment, and in such a way that no matter what may happen no blame can be attached to the operator. In general it may always be stated that the older the lesion the less satisfactory will be any measure of treatment except possibly resection and arthroplasty.

ARTHRODESIS.

This term applies to the intentional production of ankylosis in a joint previously healthy or nearly so, with the intention of stiffening a useless limb and thus enhancing its usefulness. The measure applies mainly to those cases of infantile paralysis, with loss of control of the knee or ankle, or both, when by stiffening the limb it can be made to serve the purpose of a crutch. It is the last resort in this direction when there is no possibility for tendon grafting. Long confinement of a limb in a fixed dressing will lead to considerable stiffening of the joint, yet a joint so immobilized lacks that firmness of support called for in cases above mentioned. Therefore when it is desired to perform arthrodesis the joint is usually opened and more or less of its articular surface removed, the intent being to produce the effect in the shortest time and in the best way. It can be better attained by a removal of articular surfaces with the saw and the apposition of fresh bone surfaces to each other, their retention being ensured either by sutures (tendon or wire) or accurate fixation in plaster of Paris. Under these circumstances drainage should not be necessary, and limbs can be completely enclosed in a fixed dressing.

MAJOR OPERATIONS ON JOINTS.

Aside from arthrotomy and partial or complete arthrectomy, as above mentioned, the latter, including removal of synovia or cartilage, and perhaps curetting of bone foci, the formal _resections_ or _excisions of joints_ remain to be considered. The latter is the preferable term, as it is meant to include removal of the component parts that enter into the construction of joints, while the term resection implies rather the removal merely of portions of bone.

_Joint excisions_ are practised especially for the following purposes: (_a_) To atone for the result of old unreduced dislocations; (_b_) in certain compound dislocations, with or without fracture; (_c_) in certain comminuted fractures where there is no prospect of recovery with useful joints; (_d_) in the destructive forms of acute arthritis where the entire joint is disorganized and the bone ends carious; (_e_) in tuberculous arthritis or panarthritis, with or without suppurative complications; (_f_) in occasional instances of disabling osteo-arthritis; (_g_) for relief of ankylosis, either for improvement of position (knee) or restoration of motion; (_h_) occasionally after gunshot injuries. _Excisions required by the exigencies of traumatisms should be promptly done._ If the case be complicated with septic infection the prognosis is much less favorable. For convenience of description excisions may be classified as _primary_, _intermediary_, and _secondary_. According to the joint involved, as at the knee, the purpose underlying the operation is to effect an absolutely rigid bony ankylosis.

The development and perfection of the general method of joint excisions is a matter of but little more than a century. Previous to that time amputation was almost the only resort when destruction had occurred. The most prominent surgeons in the early development of the measure were Park, of Liverpool, and Moreau, of France. During the latter part of the past century Ollier, of Lyons, greatly improved the technique by demonstrating the importance of the periosteum and by introducing the so-called _subperiosteal methods_. This is of great value in _uninfected_ cases. It is a mistake, however, to endeavor to save periosteum which has become involved in the tuberculous process; in fact, in the presence of tuberculous disease we cannot be too radical in the removal of all affected tissue.

In the so-called subperiosteal method the operator endeavors, so far as possible, to preserve the periosteum of the parts exposed to attack, and, if possible, the capsular ligament as well. Thus at the elbow the capsule, _if not diseased_ or obliterated, should be preserved, the osseous tissue being shelled out from within, so far as possible. The less, then, the connections between the capsule and the periosteum are disturbed the better. The French apply to this method the term “subcapsular periosteal.” When the bone covering can be preserved new bone is easily formed to replace that which has been lost, especially during adolescence, while the preservation of the capsule, with its ligamentous connections, affords a better joint cavity than will the substitute which results from natural processes. Furthermore the surrounding tendons are less disturbed and the condition remains more like the original. Nevertheless one does not exsect healthy joints, and the method is not always easy nor even possible of performance. It will suffice to say that it should be adhered to only as far as circumstances may justify or permit.

Surgeons, however, have not been satisfied with the older methods, and have endeavored to still further enhance motility in operated joints. (See above--Arthroplasty.) To this end the interposition of muscle, fascia, or of foreign membrane has been suggested. Thus, after removal of the head of the femur a strip of fascia lata may be interposed between the raw-bone surface and the cavity of the acetabulum, being fastened there by catgut sutures. In the shoulder a similar procedure has been carried out, utilizing a strip of deltoid muscle. At the elbow a piece of the pronator radii teres may be detached and fixed by sutures to the brachialis anticus. In every case the method should be adapted to the demands made, the intent being to cover divided bone ends with tissue which will prevent osseous union, as it is known to do in many cases of fracture where such interposition produces non-union. In so far as one attempts here to imitate conditions which are considered undesirable in certain other traumatisms, Murphy has done more than any other American surgeon, both in the experimental and clinical study of this subject. (See above.)

For the joints below the hip and shoulder the _bloodless method_ will facilitate operative work. In case of a septic joint, however, it would not be advisable to apply the elastic bandage below and then over and around the joint, as by the pressure thus made some septic material may be forced into the absorbents. In clean cases the rubber bandage is a great advantage to the operator. It has this objection, however, in that hemorrhage which does not occur during the operation has to be checked after its conclusion, and I have often thought it advisable to avoid the use of the bandage and to secure vessels as they are divided, in order that when bleeding has once ceased there be no fear of its recurrence later.

The question of _drainage_ is one of importance. In a general way one may feel that in an absolutely clean case drainage is not required, save possibly a small opening for escape of blood. If practised at all it should be thoroughly done. Drainage tubes are often too small and do not permit the escape of either clotted blood or debris of injured tissue.

The _after-treatment of excisions_ demands, first of all, _physiological rest_ of the part involved, especially if, as at the knee, sutures or other expedients for maintaining apposition have been inserted. When motion is sought there will soon come a time when passive motion can be begun. This will vary with the size of the joint and the magnitude of the procedure. Actual rest should be maintained until firm wound healing has been secured. Passive motion is then begun, to be practised daily, the sensation of the patient being the guide as to the range of the movement and extent of manipulation. Thus, after exsection of an elbow with prompt union of the wound passive motion should be begun in about two weeks, but it should not be begun for a month if the joint has been thoroughly disorganized and the cavity is still discharging. Motion should be begun as early as is considered feasible in order to guard against a false joint.

_The remote consequences of joint excisions_ are usually very satisfactory. The best results are obtained in the young, _i. e._, those whose tissues are still undergoing natural changes and whose bones are growing. In the course of time, by condensation of surrounding tissues, a new joint capsule is formed, its interior smoothed off, apparently covered with endothelium and filled with a sufficient amount of fluid, similar to that of normal joints, to serve the purpose; in this way a new joint becomes gradually substituted for the old, which serves the original purpose, in a surprising and gratifying way. Even in those of advanced years a satisfactory result is often obtained. It is often necessary to afford some support, by which too great a range of motion may be avoided; thus at the elbow the result at first is what may be called a “flail-joint,” which permits much undesirable lateral movement. This can be avoided by having light leather corsets fitted to the forearm and arm, connected by two lateral hinged braces. This being constantly worn, and no motion permitted which is not an imitation of the normal, the parts in time adapt themselves to the purpose, so that all apparatus can after a while be removed.

Excisions, like amputations, may be practised and the general methods learned on the cadaver, but their actual performance in the presence of extensive disease will be found to be a different procedure from that learned upon the dead body. For reasonably representative cases typical operations can be devised, with explicit directions. It is not advisable to try to do such work through too short incisions. A long incision heals as kindly as one shorter and affords more room for operative work. The incision should be so planned and executed as to afford the maximum of exposure with the minimum of damage to important structures. The region of the great vessels is avoided in all the classical operations, while nerve trunks, if exposed, are retracted and kept out of harm’s way. After the knife has once laid open the joint it is used but little except for the division of resisting structures, _e. g._, ligaments. The greater part of the work is then done with elevators, or periostomes with reasonably sharp edges and sufficiently broad surface, so that the periosteum can be divided with the latter and separated with the former to the necessary extent. Obviously epiphyseal junctions should be spared whenever possible, especially in the young. To remove an entire epiphysis is to materially impair the later growth of the limb. In some of the most serious cases it will be found already loosened and lying as a sequestrum in the joint cavity. In this case it may be easily lifted out of place. Tendons should never be divided unless absolutely necessary. Incisions in their neighborhood should be so planned as to be parallel with their direction and permit their displacement without division. The sharp spoon should be employed for curetting the interior of a joint capsule or cleaning out a bone focus (erasion). A capsule involved in tuberculous disease should be completely extirpated. Diseased bone ends should be sufficiently exposed to permit of the use of an ordinary saw or a chain or wire saw.[34] Considerable force will often be necessary in making bone ends accessible for this purpose. The chisel is rarely used except in cases of bony ankylosis, where it is not possible to force bone ends through the opening in order to attack them with the saw. As remarked above, clean cases may be closed without drainage. Visible vessels should be secured, and, while a certain amount of oozing may be expected, if the part be enclosed in suitable compressive dressings and elevated, it need not cause alarm. The gentle application of an elastic bandage for three or four hours may afford additional security. It should not, however, be allowed long to remain. The terminal portion of the limb will always afford an indication as to the condition of the circulation. Should it become cyanotic or cold the dressing should be renewed and the wound examined promptly.

[34] Wyeth’s “exsector” is an admirable substitute, especially at the shoulder and hip.

=Special Incisions.= =The Shoulder.=--A longitudinal incision suffices for most cases (Fig. 214). This may be made posteriorly between the fibers of the deltoid or anteriorly and externally over the bicipital groove. It is better to separate the deltoid fibers than to divide them, although they may be divided. Should the straight incision afford insufficient room another incision at right angles will afford ample access. The capsule, having been exposed, is opened, the wound widely separated with retractors, the arm rotated through a wide arc, while with a stout knife the capsular ligament and the various muscular attachments around the neck of the bone are divided. The greater and lesser tuberosities, with their muscles undivided, should be retained, when circumstances permit. The head of the bone, being freed, is dislocated and forced out through the wound, where it may be seized with large forceps and removed with a saw. The higher the bone is divided the better. Every other consideration, however, should be sacrificed to removal of all foci of disease. The capsule may then be extirpated and the glenoid cavity thoroughly cleaned out with a sharp spoon. Should the case be one of serious infection it is advisable to make a posterior opening, even through the deltoid, for purposes of thorough drainage. The greater part of the first incision is to be closed with sutures, the arm dressed in a comfortable position, with the elbow at a right angle, and the patient allowed to be up and around as soon as he feels in the mood for it.

[Illustration: FIG. 214

Excision of the shoulder: _A_, regular incision; _B_, supplementary. (Ollier.)]

=The Elbow.=--Here a variety of methods have been advised, and the extent of the operation must depend, to some degree at least, on the nature and extent of the condition which necessitates it. Partial excisions have been recommended, though in the writer’s experience incomplete operations often give less satisfaction than those which are complete. However, when it is a question of removing callus or displaced bone fragments, which, after fracture into the joint, impair its function, then partial resections may be serviceable.

[Illustration: FIG. 215

Excision of the elbow-joint: _A_, von Langenbeck; _B_, Ollier.]

[Illustration: FIG. 216

Excision of the elbow-joint: _A_, Nélaton; _B_, _C_, Hueter.]

[Illustration: FIG. 217

Osteoplastic method: _A_, by external incision; _B_, von Mosetig-Moorhof.]

The essential incision is a long posterior one, which may be somewhat modified (Figs. 215, 216 and 217). It is essential here to avoid the ulnar nerve, which passes between the internal epicondyle and the olecranon, and the vessels and nerves in front of the joint. If it be made an inviolable rule to always _keep close to the bone_ both of these dangers may be avoided. Ligamentous and muscular structures, among the latter the anconeus, should be spared as much as possible. After separating the joint surfaces thoroughly, by forced flexion, it is usually easier to force out the lower end of the humerus and first remove it, after which the upper ends of the radius and ulna are exposed and removed. When there is bony ankylosis it is preferable to divide the bones of the forearm first. The tendon of the triceps is not only detached from the olecranon, but divided by the first long incision. After concluding the incision, the capsule, if it remains, is to be closed with chromic catgut sutures and the end of the triceps tendon or some of its periosteal attachment united to the periosteum of the upper end of the ulna.

The arm is now fixed in the right-angle position and held comfortably to the body by a suitable sling.

=The Wrist.=--It is rare that in disease of the wrist-joint this is found to be limited to a single bone of the carpus. Should an _x_-ray examination indicate such limitation then the focus can be exposed and cleaned by an incision upon the dorsum of the wrist, where it may seem best adapted for the purpose. Suppurative and tuberculous affections of the wrist usually necessitate removal of the carpal bones, including, possibly, the lower extremities of the ulna and radius. When the wrist-joint is involved it may be sufficient to remove the latter with the first row of the carpus.

Fig. 218 illustrates the incisions to be recommended for wrist resection, of which the Langenbeck line is to be preferred. Occasionally two lateral incisions, with through drainage, will better serve the purpose. It may be necessary to divide the short radial extensor, but this may be united again with suture. In most instances it is possible to retract the tendons to either side and thus clear the carpal region. By hyperextension the extensor tendons are relaxed and more room is thus made. The incision marked “_A_” combined with that marked “_B_” in Fig. 218, affords the best exposure when disease is extensive. The incision along the inner border of the wrist is made 5 Cm. above the styloid process of the ulna, and between the latter and the ulnar flexor down to the middle of the last metacarpal bone. Here the tendon of the latter muscle should be divided at its insertion and lifted out of its groove in the ulna. The collection of extensor tendons is then separated from the back of the wrist and lifted up, it being usually necessary to divide the unciform process of the unciform bone with forceps. The knife should be kept from the palmar surfaces of the metacarpal bones in order to avoid injury to the deep arch. After dividing the anterior radiocarpal ligament the carpus is extirpated through the ulnar incision. The ends of the ulna and radius are now easily accessible for removal with forceps or a metacarpal saw. The same is also true of the proximal ends of the metacarpals. After spreading the hand and forearm upon a flat splint drainage can be made to the desired extent and the wound closed.

[Illustration: FIG. 218

Excision of the wrist: _A_, Lister’s radial incision; _B_, Lister’s ulnar incision; _C_, Ollier; _D_, von Langenbeck.]

[Illustration: FIG. 219

Excision of the hip: _A_, Sayre; _B_, Ollier.]

So far as the _hand and fingers_ are concerned little resecting need be done, the surgeon usually confining himself to the removal of sequestra or curetting of carious bone. In cases of compound comminuted fracture bone fragments may be removed; only in cases of lost or destroyed phalanges will amputation be necessary.

=The Hip.=--In its structure the hip-joint is one of the simplest in the body. Although it lies deeply it is easily made accessible. Fig. 219 illustrates the incisions by which the joint is attacked for the purpose of exsection. If necessary either extremity of the incision can be extended or enlarged by a cross-cut. When the joint is disintegrated by disease, especially when partially dislocated, the parts will lend themselves to an easy and simple operation. When, however, the operation is done for ankylosis or for disease, by which great thickening and fixation have been produced, the measure may become difficult. For ordinary purposes the simplest method is to drive a sharp-pointed, strong-bladed knife directly down upon the neck of the bone from a point midway between the great trochanter and the crest of the ilium; then keeping the knife-blade in contact with the bone the incision is carried downward over the trochanter and along the shaft to a length making it sufficient for easy exposure of the bone and of the joint. Nothing is gained in these cases by trying to work through a short incision. A long one heals as readily and makes the operation more simple. It is as easy to make the entire incision in one cut as to divide the muscles layer by layer. The capsule of the neck of the femur being exposed by a wide retraction of wound margins, it is necessary next to divide muscular attachments to the great trochanter by raising the periosteum to which they are attached and saving both. To expose these insertions the femur should be rotated inward and outward, while the capsule is at the same time divided. The ligamentum teres, which offers a theoretical obstacle, usually disappears in the presence of any active disease and is scarcely ever encountered; it can be divided with curved scissors. Now by more or less powerful effort, including flexion and adduction to the extreme limit, with more or less rotation, the head of the bone is forced out from its socket and through the wound. Whether the bone should be decapitated with chain saw, metacarpal saw, or by the exsector of Wyeth will depend partly upon the freedom with which it can be exposed and on the equipment of the operator. It may be advisable to divide the neck with a chisel. The trochanter major should be preserved whenever its removal is not made imperative by the progress of the disease. The head and neck of the bone having been removed, the acetabulum is now more or less easily exposed, especially with retractors, and it should be cleaned with a sharp spoon. The capsule also should be removed, at least when the operation is done for tuberculous or other infectious condition. It is advisable to irrigate, then to wipe dry all the original joint surfaces and raw bone, and finally to cauterize either with pure carbolic or with zinc chloride, which should be washed away with the irrigating stream, the intent being to close the mouths of all the absorbents and prevent absorption from fresh exposure. Sinuses if present should be thoroughly excised, scraped, and treated in the same way. A drainage tube is usually preferable to the use of gauze.

The above is the method usually relied upon for hip exsection. Other methods have been devised, especially by _anterior incision_; of these the best probably is that of Barker. The cut is made along the outer border of the anterior surface of the sartorius and rectus, and through it the femoral neck is reached. By wide retraction the anterior surface of the joint can be completely exposed and opened, and through this opening the neck of the femur can be divided with a chain saw or chisel, before removal of the head from the acetabulum. The disadvantage of anterior incision is that pertaining to drainage. Nevertheless this can be obviated with capillary drains. Its advantages are that splinting and protection can be more perfectly effected, with less necessity for frequent interference. In other words it makes the subsequent care of the patient easier. Many English surgeons are in favor of it. Ollier devised a so-called osteoplastic excision, made through a curved incision with a downward convexity, the top of the great trochanter being exposed and divided with a chisel sufficiently to permit of its being turned up with the flap, and then being reunited to the main part of the bone after the removal of the neck and head. This method has its advantages in a limited number of cases, but it has not become popular in this country. It would seem to be an advantage to preserve the trochanter, although some surgeons remove it. So long, however, as disease is confined to the head and neck of the bone it is unnecessary to remove this projection.

The _after-care_ of a hip excision is not an easy matter. Most surgeons prefer to maintain the limb in position by the aid of traction, with sufficient weight to overcome all muscle spasm. If the case be such that dressings need only be made at long intervals, then it matters little, but in a septic case in which there is considerable discharge the problem is sometimes a serious one. Various beds or suspension splints have been devised, consisting essentially of frames with cross-strips of stout material, upon which the patient lies. After raising the frame one or two of these strips are released and the parts exposed. This arrangement also permits of the easy management of a bed-pan. In young children a wire splint with a fenestrum, or a plaster-of-Paris spica or breeches with large opening cut opposite the wound, will often be serviceable. The tendency is rather toward adduction, and this should be overcome. Something will depend upon whether the surgeon is working for ankylosis or for a movable joint. In the former case a rigid dressing should be employed as soon as the condition of the wound permits. In the latter passive movement should be begun as soon as the wound is healed.

While the operation is usually performed quickly, and is not regarded as serious, it nevertheless has a considerable mortality, especially in the young and the aged, because of the conditions which necessitate it. After a complete exsection, even by the most ideal method and in the most ideal case, the limb remains somewhat shortened. This may be compensated by raising the heel of the shoe worn on the affected side. In severe cases it may be necessary to supply even two or three inches of artificial support for this purpose. Unless this is done compensatory spinal curvature will ensue.

[Illustration: FIG. 220

Excision of the knee-joint: _A_, semilunar incision; _B_, Ollier’s incision.]

=The Knee.=--The knee is generally more accessible for operation than the elbow, as the important structures which should not be disturbed lie grouped upon its posterior aspect. Protection for one of these is protection for all, and the freedom with which the joint may be opened makes it especially easy to do either complete or partial operation. Here the surgeon should endeavor to preserve the epiphyses, especially in children, as they have much to do with the growth and length of the limb. So long as incision is confined to the anterior aspect of the joint it can be made in almost any manner. The usual method is that represented by line _A_ in Fig. 220, by which a horseshoe flap is raised and the joint interior exposed. Occasionally the direction of the flap is reversed, and it is turned downward rather than upward. In the former case the ligamentum patellæ is divided; in the latter, the tendo patellæ. Whichever way the flap is turned it is made to include the patella, although this bone can be removed at any time. The lateral ligaments being divided, as well as the crucial, and the limb completely flexed, exposure of the joint surfaces is made. It is now possible to do an arthrectomy, a partial exsection or a complete one, according as the disease is more or less extensive. In the complete operation the articular surfaces of the femur and of the tibia are usually removed with an amputating saw. If this be introduced from the front and made to work its way backward the popliteal vessels should be amply protected against possible injury. Here it should be borne in mind that the leg is not constructed in a straight line, but that there is a lateral angle at the knee, as the femurs diverge as they pass upward, and this angle should be imitated in directing the saw and removing the bone end. Again, a slight bend anteriorly will make the limb more useful than one which is absolutely straight. The intent thus should be to give the knee at a slight angle anteriorly and interiorly, and the saw should be manipulated with great care. In a complete operation the patella is also removed. In tuberculous and other septic disease the capsule should be completely extirpated. This offers no difficulty, save at the posterior surface, where it may approach closely to the region of the great vessels.

Various modifications have been practised in these operations. Some open the joint by straight cross-incision with division of the patella, the latter being reunited with tendon or wire sutures. Others have practised a more complicated H-shaped incision, the transverse portion being carried either through the patella or just below it. The line marked _B_ in Fig. 220 was suggested by Ollier. It is questionable whether any of these methods offer any advantages over the one first described.

After exsection it is desirable to maintain the bone ends in an accurate position if speedy reunion be desired, and for this purpose various methods are in vogue. The bones may be drilled and fastened together with tendon or wire sutures, or ivory nails may be driven in, one on each side, directing them obliquely, so that displacement cannot easily occur, or metal nails may be used for the same purpose. Another plan is to insert two long metal drills, one on either side, which perforate the skin two or three inches above the wound, and are passed downward and toward the other side so as to fix the surfaces, as it were, by a cross-forked arrangement. After two or three weeks these drills may be withdrawn. Fixation of this kind is advantageous, for when complete excision has been practised the surrounding tissues are lax and the parts are not easily held in position by external dressings alone. In a clean case, with careful hemostasis, very little drainage will be required. What is needed can be provided by an absorbable drain passed through the lower portion of the wound on either side. In a septic case it would be well to provide for ample drainage on each side.

The limb may be dressed upon a fenestrated wire or gauze splint, which is easier when frequent change of dressing can be foreseen, or it may be immobilized in a plaster-of-Paris splint.

=The Ankle.=--The ankle is usually reached by an incision on either side, three or four inches in length, extending from above each malleolus downward and forward on to the tarsus. The knife-blade should be forced to the bone, so as to divide the periosteum, which is subsequently separated and lifted by an elevator, in order that the operation may be made subperiosteally. The fibula is usually first divided, with a chain saw or a chisel, an inch above its tip. The divided fragment is wrenched from its place with forceps, and severed from the ligaments by knife or scissors, being careful not to injure the external lateral ligament. The inner incision is made in practically the same way, the periosteum separated, the internal lateral ligament divided, and the end of the tibia forced through the incision by everting the foot. Its joint end may be removed with a saw, dividing on the same level and plane with the lower end of the fibula. Through the gap thus made the astragalus may be either removed or its upper surface divided with a metacarpal saw. The fresh bone surfaces left in this way will unite and ankylosis will result, unless fibrous or muscular tissue be interposed to favor the formation of a false joint.

As in other operations methods may be varied to meet the exigencies of certain cases. Longitudinal incisions may be placed farther forward than indicated above, as is shown in Fig. 221, which illustrated König’s method. Here the bone surfaces are divided with broad chisels. A transverse incision of the front and upper part of the ankle may be made, through which the tendons are exposed, lifted in a group out of harm’s way, and curetting and bone sawing performed. Kocher makes a semilunar incision from the outer border of the tendo Achillis to the outer border of the extensor tendons, its line passing beneath the external malleolus. By this method the joint is opened and the peroneal tendons divided, their ends being reunited after the completion of the balance of the work. This method is usually applicable in children.

Ample drainage is required in these cases, for the operation is seldom performed in the absence of septic complications. The foot should be kept in proper and right-angled position by metallic splints, or by plaster of Paris, the latter preferable, fenestra being cut in order to make access to the wound.

=Excisions of the Tarsus and Osteoplastic Excision of the Heel.=--Removal of the tarsal bones is confined usually to cases of tuberculous disease, and may be performed by a variety of methods. Thus the tissues of the sole of the foot may be divided transversely by an incision carried from the tubercle of the scaphoid beneath the sole and across to a point one inch behind the base of the metatarsal. Through this, access can be made to the inferior surface of the tarsus. Conversely the upper portion may be exposed by a similar transverse incision across the dorsum of the foot, by lateral incisions, or by a combination of both. It is seldom necessary to divide the tendons, it being nearly always possible to gather them into a group and lift them out, while the bones are attacked with a sharp spoon or a chisel.

Occasionally the calcis becomes involved in cancerous or tuberculous disease and it would appear that removal of the heel proper would be all that is required. To meet these indications Wladimirov, in 1871, and Mikulicz, in 1880, independently devised a method by which the ankle-joint may be opened and as much of the heel and adjoining tarsus as necessary removed, the foot being later fixed in the extreme equinus position. This is referred to as _osteoplastic excision or amputation of the heel_. Fig. 222 illustrates the line of incision, which extends from the tubercle of the scaphoid beneath the heel to a point on the opposite side, then obliquely upward and backward to the base of each malleolus, and then transversely and posteriorly, thus including within its line the region of the heel. These incisions extend to the bone, the ankle-joint is opened posteriorly, the lateral ligaments divided, the lower extremities of the tibia and fibula removed with a saw, the astragalus and calcis separated from their attachments, and the posterior articular surfaces of the scaphoid and cuboid also removed. The lines of division of bone are indicated by dotted lines in Fig. 222. Thus the lower ends of the leg bones are brought into contact with the upper end of the divided tarsus by straightening the foot in the extreme equinus position and maintaining this position with wire sutures or bone or metal pins.

[Illustration: FIG. 221

König’s incision for excision of the ankle.]

[Illustration: FIG. 222

Osteoplastic excision of the foot. (Mikulicz.)]

The cases in which this method is of use are rare, but when indicated it has usually given satisfactory results. It is a substitute for amputation of the leg, and it is often an open question as to which will give the most satisfactory result. It has probably not been practised a hundred times.

##