that "it is plain that the risk of sloughing would be diminished if the flap were made shorter, and not extending the horns of the incision so high up the limb upon either side." He proposes a modification of Carden's long flap, making it shorter, with a short posterior flap sufficient to cover the stump well. The advocates of this joint amputation, which became a comparatively modern operation through the influence of Velpeau, admit the great danger to the long flap by sloughing, hence have endeavored to so modify it as to prevent it. (See Pollock, Medico Chi. Trans., 1870, on Amputation at the Knee Joint.)

AS TO PROPRIETY OF SECTION OF THE CONDYLES OF THE FEMUR.

Another question to be mentioned in the method by Carden is the risk to be apprehended in section of the femur. If the section is made so as to open cellular structure in the expanded end of the femur, ordinarily no great risk would occur (some surgeons suppose less than the solid structure higher up); nevertheless, in case we had inflammatory action there is danger of it extending to the cellular spaces, and thus making the bone in disease not easily controlled. No doubt this was comprehended by Guerin in advising that the structures of the joint, viz, articular cartilages synovial membrane, be allowed to remain, while others permit them to be removed by sawing around the condyles of the femur, thus removing the synovial covering by the aid of Butcher's saw; or again, by cutting through the condyles, after Carden. Statistics as given by Dr. Brinton show that it is better to allow the condyles to remain. Therefore, an amputation through the joint, omitting section of the femur, would be better if the patella is removed, thus securing such modifications in the execution of the operation in all its steps as is desirable to make it the most successful operation which can be made at this point. There is one thing suggested by Erichsen to prevent retraction of the patella, viz, division of the insertions of the quadriceps femoris and retain the patella. So, too, *Pollock advises the retention of the patella, but says nothing about the separation of its muscular attachments. If Carden's plan be adopted in any case, viz, the supra condyloid, then the patella should be removed, and in any case if diseased, so

See Pollock's Medico Chi. Trans., Op. Cit., 1870.

that by it we should get rid of difficulties which are not likely to occur or spring in that event. This, however, is a question sub judice and must be determined by the judgment of the operator. I would remove or retain it according to the circumstances in the case, which might indicate one or the other, as well as being governed by the kind of operation about to be made.

STATISTICS.

As far back as 1856, Dr. Markoe, of the New York Journal of Medicine and Surgery, gave most valuable statistics, from which we find, by comparison of operations at the knee and in the thigh, that those of the knee have a mortality 7 per cent. less than those in the thigh, being 37 per cent. for knee-joint amputations and 44 per cent. in those in the continuity of the thigh.

As late as 1868, Dr. Brinton, in the American Journal of Medical Science, in an elaborate paper gives fuller statistics of this operation. In the latter the mortality per cent. is 32 31. Of 211 cases found in the reports of the Surgeon-General's office, U. S. A., 106 died, or 50.2 per cent. These, of course, were for traumatic causes. By a comparison of operations made in the leg, at the knee, and in the thigh, as taken from various sources, viz: Reports by Malgaigne; Archives of Med. St. George's Hospital Reports, vol. 1; Medico Chi. Trans., vols. 42 and 47; Trans. American Med. Ass'n, Vol. 4. These are taken from tables of the published statistics of French, British and American hospitals, from those cases in the army reports-U. S. A.-and reports of the war in the Crimea, by Legouset; in surgery of the Crimean war; also Guy's Hospital Reports, vol. 15, p. 630. We group together the facts, and deduce the mortality rate per cent. :

For chronic disease, by Brinton and Legouest, Op. Cit., P. 735:

MORTALITY

PER CT.

Knee Joint, disease..

Traumatic Causes, injury.

SUMMARY.

22.58 40.62

We now come to the question which has been frequently

asked: "What method of operation is best?" Widely diverse opinions are held by many surgeons upon this subject. The methods used and commended are the long anterior, short posterior flaps, the circular method of Berand and Heulte, the lateral flap, the long posterior flap with section of the condyles (Syme), and the long anterior flap with section of the condyles (Carden). Markoe says (New York Medical Journal, March, 1868, P. 515,): "The mode of operative procedure is so frequently determined by the condition of the parts after injury, that, I doubt not, most surgeons will arrive at the same conclusions that have gradually forced themselves upon us, viz: That it is really of comparatively little consequence what particular method of cutting the flaps be adopted provided only that enough integument be left amply and easily to cover the expanded extremity of the femur." Brinton says (American Med. Journal, p. 336,): "The main indication in the disarticulation of the knee is to obtain a sufficient amount of integumental covering to admit to the lips of the wound being brought together without strain. Whether this be done by the circular oval method, or by lateral or antero posterior flaps, matters but little."

Brinton himself inclines to the method of antero-posterior flaps. Smith, surgeon to Belleview Hospital, favors lateral flaps, and says: Something more is required in all amputations than mere easy adjustment of the flaps," and speaks of drainage during healing. The question resolves itself into this: "What method of operating will most effectively meet two important conditions, viz, drainage and such a line of union as relieves the stump from pressure upon the cicatrix ?"

Smith: "The flap methods, viz, antero-posterior, may most effectively prevent drainage if there should be extensive suppuration, as I found in my case unavoidable. If, however, the posterior flap is nearly circular, you may have good and sufficient drainage, and the cicatrix is brought posterior to the axis of the limb. This introduced the idea of a long anterior skin flap, with the added danger of sloughing, which is bringing this method into deserved disrepute. The circular and oval methods are less adapted to meet the conditions required, hence are not much in vogue at the present time. The methods of Syme and Carden, involving

excision of the condyles, are, I believe, decidedly objectionable, even unnecessary, except it is found that too little flap has been secured in making the operation, a mistake which may be made.

Hence we find that but two methods are left which can secure all the advantages required, viz, the latteral flaps of Smith, or the long flap with a modification in favor of a slightly longer circular posterior flap. The lateral flaps properly made will combine the advantages of both methods with the least dangers of any, and will secure best drainage, a line of union not in the way of pressure, and safety from sloughing in the long anterior skin flap. These are very great advantages in selecting an operation, and which is an operation confessedly beyond those other methods. mentioned as to dangers. Easy adjustment of flaps, quick union, completing a well-rounded or covered stump, is a sine qua non in this as in other amputations, and is more readily obtained by the lateral flap method than any which are presented to us.

CASE.

Mr. C., Clay City, Ind., came to the city to consult me by the advice of his physician, Dr. Erskine, for an injury received by a railroad accident five years ago, in which his left limb was crushed, producing a compound-comminuted fracture of the bones of the lower leg. He was treated, and slowly recovered, with an impaired condition of nutrition of all the soft parts and bone. The skin and muscles were destroyed in the middle of the lower leg, the cicatricial tissue adherent to the bone, and the skin in a state of ulceration, being unable to heal it up permanently, although skingrafting had been repeatedly tried. The heel was drawn up by the contraction and adhesion of the gastrocnemius, and requiring the use of crutches for locomotion.

In this condition he sought advice, and was advised to have an amputation, as the difficulties could not be cured. February 23, before the class of the Central Medical College, I amputated at the knee joint, making Syme's amputation, viz, a long anterior flap, disarticulation, and a short posterior flap. It was neatly adjusted by sutures and supporting adhesive strips. On the 26th, union was effected nearly throughout the extent of the wound, except upon the inner angle, where I observed discoloration and a

disposition to slough at the inner margin of the long flap, and this continued till the whole anterior flap at its base had become dead and was separated, leaving the condyles of the femur exposed. Thus I was face to face with re-amputation or cure by a process of granulation to secure such a stump as was desirable. I had a large surface to granulate, condyles exposed and a very unfavorable condition of things. Rather, however, than submit the patient to the risks of an amputation in the continuity of the femur, with all the increasing dangers, I determined to endeavor to get a cure by the granulation process. I did so, and succeeded well, save an attack of erysipelas, affecting the stump and extending along the muscular interspaces half way to the hip. Free incisions into the tissues, promptly made, saved the damage which threatened, viz, sloughing and extensive destruction of soft tissue along the outer side of the limb. Position, drainage, antiseptic, dressings, tonics, good food, completed the work of cure, and at this writing, May 6, 1881, is recovered, and returned to his home, and "nothing to complain of," using his own language of this date. The point of the management is the granulation of an open stump, while fighting an erysipelas above the stump, and not touching the granulating surface.