Home >> Surgery-and-obstetrics >> Affections Of The Ear to Wounds Of Special Regions >> Operation for Roaring

Operation for Roaring

OPERATION FOR ROARING The surgical relief of roaring in horses was first attempted in 1866 by the Professors Gunther of the veterinary school in Hanover. Further reports upon the operation were made by K. Gunther in 1893 and 1896.

The Professors Gunther experimented with a series of operations upon the vocal apparatus, in an endeavour to find a successful method of surgical interference. They removed, in part or entirely, virtually every part of the vocal apparatus, including the arytenoid cartilages, the vocal cords, and the mucous membrane lining the laryngeal ventricles. Their investiga tions were continued for about thirty years. However, the material available to them was apparently not very extensive. The details of their various plans for operating are not avail able. In their publications they give merely the outlines and do not record fully the con sequences of their operations as revealed by post-mortem examinations. Among the numer ous experimental operations which they per formed was the removal of the mucous mem brane from the laryngeal ventricle without removing the arytenoid cartilage or vocal cords. Like the other experimental operations of the Professors Gunther, the removal of the mucosa from the laryngeal ventricle did not succeed, and the literature which they have left upon the subject does not explain why the operation failed.

Following the efforts of the Professors Gunther other investigators took up the work. In general they confined themselves to one type of operation—the total or complete ablation of the left arytenoid cartilage, which necessarily involved a partial or total removal of the vocal cord, with some mutilation of the mucosa of the ventricle. Prominent among these investigators were H. Moeller in 1888, Fleming in 1889, and Cadiot in 1891, each of whom published monographs upon the subject, recom mending the particular type of operation which he described. In each case, however, the technique proved a failure. A few recoveries occurred, but so few that, one after another, each plan had to be abandoned as wholly un satisfactory. There were many more horses ruined by the operation than benefited.

Numerous veterinarians in different coun tries attempted one operation after another, as recommended by various investigators, and occasionally published their results. At first a certain -technique would be followed by what appeared to be highly favourable results, but in the end sequelae would appear and destroy the value of the patient. A great variety of operations were advised, hut none of them, except the removal of the arytenoid cartilage, gained any standing with the veterinary pro fession.

As early as 1890 I had attempted, with out success, the operation recommended by Fleming, and later had attempted other plans, consisting chiefly of attempts to fix the vocal cords against the side of the larynx by means of permanent sutures, but these efforts gener ally failed. In 1905, in the clinic of the New York State Veterinary College, at Cornell University, I began experimental work in the removal of the ventricular mucosa, and in 1906 presented the results before the American Veterinary Medical Association.' The investiga tions were continued, and the following year a further and more encouraging report was made to the same Association.' The operation then began slowly to gain a definite place in veterinary practice in the United States. The progress of the operation was exceedingly slow. Only a few veterinary practitioners attempted it. My clinic afforded but scanty material, and each change in technique required much time and some repetition before any conclusion could be reached and a new step taken. So many operations for roaring had been heralded pre maturely as successful, and later proved useless or worse, that the profession as a whole looked askance at any proposed plan of surgical inter ference. The few attempts at operating were confined to the United States. Reports of the investigations appeared in England, but attracted no serious attention. In 1909 I demon strated the technique clinically, upon two roarers, to Mr. Hobday in England. The results were satisfactory, and Mr. Hobday at once took up the operation. Largely through his influence it was soon put into practice in continental Europe.

The technique as performed in 1905 was briefly as follows. The animal was cast or placed upon the operating table and anaesthetized. He was then placed in dorsal position with the head completely extended upon the neck. After ordinary precautions in shaving and disinfect 1 Annual Reports Am. Vet. Med. Assoc., 1906, 1907.

ing the operative area, a longitudinal cutaneous incision was made on the median line, extend ing from the level of the anterior border of the thyroid cartilage backward over one or two tracheal rings. The incision was carried down to and exposed the thyroid cartilage, the crico thyroidean ligament, the cricoid cartilage, and one or more tracheal rings. Then an incision was made through the crico-thyroidean liga ment and cricoid cartilage, and sometimes through the first tracheal ring. Retractors were inserted in the wound and the margins held well apart, affording abundant room. The mucosa of the ventricle was grasped at the ventricular border by means of forceps, incised, and carefully dissected away from the under lying tissues by means of the razor-shaped scalpel of Moeller. Tracheotomy was generally performed as a safeguard against strangling during the first few days after the operation. The animal was then allowed to recover from the anwsthesia. The invading and ventricular wounds were disinfected daily for a few days.

The operation met with far greater success than had followed any previous technique, but naturally had numerous defects. These were carefully studied and attempts made to eliminate them. In this work I was ably assisted by my colleague and assistant, J. N. Frost. One of the first defects to become prominent was chondritis and collapse of the tracheal rings due to the tracheotomy. It seemed impossible to sever the tracheal rings and insert the tracheo tomy tube without encountering this danger. Two methods for overcoming this danger were devised. The flat trachea tube of M'Cully, which can be inserted between two tracheal rings without necessarily wounding either of them, offered a convenient method for avoiding the danger. It is not perfect, however, and inevitably leaves a scar, which may be detected upon examination. The other method is the laryngeal tube, which is inserted through the crico-thyroidean space—that is, through the invading incision for the operation—and does not involve any additional danger to the carti lages.

A similar defect in the operation was en countered in severing the cricoid cartilage. Like the tracheal ring it also became diseased occasionally, became thickened or collapsed, and ruined the animal. In an effort to over come this, for a time, I severed the thyroid cartilage instead, either with a heavy scalpel or with a saw. The danger was not lessened. Again there was chondritis. The thyroid cartilage is so rigid that it does not collapse, but in some cases it becomes enormously thickened and extensively ossified, and in one instance caries set in. Finally it was found that the operation could be performed through the crico-thyroidean ligament without wound ing the cartilage, and that through this wound the laryngeal tube could be inserted to guard against post-operative strangling.

The method of removing the mucosa from the ventricle has been subjected to extensive modification. Some operators still adhere to the original method of 1905. Other operators, after making the penetrant incision through the mucosa, insert the index finger through the incision and detach the mucosa with the finger tip. In this way they obviate in a large measure the hemorrhage ordinarily ensuing when the scalpel is used. Even with the scalpel, how ever, the haemorrhage was not great. Later there was introduced, by Blattenberg, the ventricular burr—a spherical steel burr similar to that used by the human surgeon in the mastoid operation. This instrument was in serted into the ventricle, pushed against its bottom, and revolved until the mucosa was caught and entangled in the burr. By revolv ing the burr the mucosa was torn away from the underlying tissues. The detachment was completed by exerting traction upon the instru ment, drawing the mucosa out of the ventricle, and everting it. The mucosa was then excised at the margin of the ventricle and removed.

It was soon found that the steel burr was quite dangerous, because it tended to mutilate the mucosa, penetrate it, and wound the peri chondrium of the thyroid or arytenoid cartilage. From these wounds there arose perichondritis, with consequent thickening and the failure of the operation. Other operators advocated the use of a pair of forceps with the jaws roughened or milled on their exterior. The instrument was introduced into the ventricle, opened, and pushed against the bottom, after which it was closed, engaging a portion of the mucosa be tween the jaws. Then the instrument was revolved upon its long axis in the same way as the Blattenberg burr, and with the aid of traction the mucosa was everted. The forceps have in a large measure the same danger as the steel burr of Blattenberg: they may wound the adjacent perichondrium, thereby inviting failure of the operation.

Very early in the history of the operation, O'Connor of Dublin, and later Maillip of Chicago, attempted to destroy the mucosa by cauterization. The attempt was unsuccessful.

After a thorough trial of the Blattenberg burr, with some very unfortunate results, I finally designed the soft rubber burr (Fig. 287). The naked burr would not promptly and effect ively grasp the mucosa, but by wrapping a small piece of dry cheese-cloth or gauze about the rubber sphere and wiping the ventricle reasonably with surgeon's cotton, the mucosa could almost always be grasped firmly and securely at the first turn. Consequently the soft rubber burr was adopted as the safest and most efficient plan for denuding the ventricle of its mucosa.

The hyaline cartilages of the trachea and larynx are, one and all, exceedingly intolerant of any surgical interference which involves an open wound with subsequent infection. Con sequently it has been the aim, from the very first, to avoid any injury to any cartilage. This is regarded as the most essential fundamental precaution in the operation.

A second question in the technique of the opera tion is the extent to which the denudation of the ventricle should be carried. The aim of the operation is the complete obliteration of the laryngeal ventricle, reliably inducing firm adhesion between the lateral surface of the arytenoid cartilage, the vocal cord, and the median surface of the thyroid cartilage, so that the arytenoid cartilage and vocal cord can no longer be forced out into the glottis to interfere with the air-stream during inspiration. If the mucosa of the ventricle is only partially removed the adhesion between the cartilage and vocal cord will be only partial, so that the arytenoid cartilage and vocal cord may drop into the glottis to some degree and interfere with respira tion. At first I removed a considerable portion of the vocal cord. This, however, left the ary tenoid cartilage somewhat dangling and tended to cause it to drop into the glottis and become fixed in a false position, partially occluding the glottis. Researches thus far indicate that the most favourable degree of denudation consists of the complete removal of the ventricular mucosa to its outermost margin, as indicated by the dotted line in Fig. 288, without removing any essential part of the vocal cord. Inevit ably the operation involves the removal of the mucosa from the lateral surface of the vocal cord, but when completed the vocal cord still remains intact and the denuded ventricle remains as deep as before the operation, plus the depth of the tissues removed. The ideal in the opera tion is so to remove the mucosa that when the healing is completed the arytenoid cartilage and the vocal cord are firmly adherent in the posi tions shown in Fig. 289.

The confinement and control of the patient have also undergone far-reaching evolution. As previously stated, at first the horse was cast or placed upon the operating table, anmsthetized, and then placed in dorsal position. Inevitably casting accidents occur, as they must in any operation. Of still greater importance were the dangers from ancesthesia. There were the ever-present primary dangers of asphyxia and syncope and the secondary dangers of chloro form, pneumonia, and hepatic degeneration. Furthermore, following the struggles and the anmsthesia, there was inevitably a greater amount of haemorrhage and the wounds did not heal so promptly.

After it had been clearly demonstrated that the operation could be performed through the crico-thyroidean ligament, and also that the mucosa could be securely and safely grasped and everted by means of the soft rubber ven tricular burr, my colleague Frost and I began operating exclusively upon the standing animal. Local antesthesia was necessarily substituted for general. It has been found that the opera tion is far more easily and quickly performed upon the standing animal. The patient is spared the physical strain of casting and the dangers of anmsthesia. The hmmorrhage is decreased fully 90 per cent. The .operation has been transferred, so far as the pain, the mutila tion, and the dangers are concerned, from a major to a minor operation. At first it was assumed that only a few of the more stoical patients could be operated upon in the standing position, but a brief experience showed that almost, if not all animals can readily be operated upon standing.

In order to facilitate the operation upon the standing patient, it was essential that a con venient form of laryngeal speculum be devised, such as that shown in Fig. 286. It is quite securely self-retaining in the standing animal, dilates the penetrating wound very effectually, and affords good opportunity for illumination of the interior of the larynx.

While the operation upon the standing patient with the removal of the ventricular mucosa by means of the soft rubber burr is decidedly the preferable method, there are nevertheless many good operators who prefer some other plan.

The following technique will answer as a general outline for the operation, and may be modified by the individual operator to suit his preferences, so long as the fundamentals are retained.

Those who prefer to operate upon the recum bent patient may secure him either upon the operating table or by casting, and may operate under either general or local anmsthesia. Local anmsthesia is never perfect in the recumbent animal, because he constantly resists the con finement, however devoid the operation may be of pain. After local or general anmsthesia has been obtained, the animal should be placed and retained in the dorsal position. Generally, operators have preferred to maintain the dorsal position by packing bundles of straw against the sides of the horse and by having one or more assistants hold the horse from either side. A better and easier way, which more thoroughly avoids the possibility of the horse's overcoming the restraint applied and turning upon his side, thus suspending the operation, is by means of pulleys attached to a beam above and to the feet or legs of the patient below. By this means the dorsal position may be securely maintained for any length of time. It has the further advantage of partially suspending the horse, so that there is not the same danger of contusion of the withers or other portions of the body.

The instruments and materials required for the operation are a razor, two or three scalpels, three pairs of Pean's 9-inch curved compression forceps, scissors, laryngeal speculum, reflecting electric lamp, two soft rubber ventricular burrs with a few pieces of cheese-cloth or gauze about three inches square for covering the rubber sphere, needles and thread, surgeon's cotton, and a laryngeal tube.

If the operation is undertaken out of doors in sunshine, the animal should be secured with his head away from the sun, so that the rays of light, coming from above and behind the animal, enter the larynx obliquely downward and forward, illuminating the field of operation. If the patient is secured with his head toward the sun, the interior of the larynx is in shadow and the light blinds the operator. If the operation is performed indoors by means of daylight through a window, the patient is to be secured with his head pointing away from the window. Upon a dark day or in a dark room the larynx may be splendidly illuminated with an ordinary reflecting electric lamp upon an extension cord, or, if electricity is not at hand, with the ordinary pocket electric light. The operation may be performed in a dark room at night with the aid of the reflecting electric lamp.

When local or general anaesthesia has been satisfactorily attained, the patient secured in the dorsal position, and provision has been made for the illumination of the interior of the larynx, the operation proceeds in the same manner as in the standing operation.

Suitable stocks afford the most convenient and efficient method for securing the patient for the standing operation, but they are not necessary. The essentials are that the patient cannot pull backward or suddenly forge forward, that he can be prevented from rearing or strik ing, and that provision shall be at hand for securely elevating and extending his head. Backing the horse into a narrow single stall suffices as a general rule. Rearing may be pre vented by passing a rope or strap over the withers and securing the ends below to the stocks, stall partition, or floor. Striking may be prevented by a rope or strap crossed in front of the upper end of the radii and secured to a post on either side. The head may be securely elevated by passing a strap or a cord over a beam or through a ring or pulley above, one end of which is securely fastened to the nose band of the halter and the other held by an assistant. The head should not he elevated by

means of the halter shank with its usual attach ment, because that tends to rotate the head and distort the operative field. It is necessary, also, that the halter shall be devoid of a throat latch, or have one which can be detached and pushed aside, leaving the ventral surface of the head and neck fully exposed. Tact and patience should be exercised, in order that the patient shall not unduly feel the confinement. The head should be elevated carefully and gradually, and the entire confinement should be carried out very gently. As soon as the head is pro perly elevated and the patient has ceased to resist, the operative area should be carefully shaved and disinfected.

Local anaesthesia is to be induced by cocaine and adrenalin, or other local anaesthetic, should the operator prefer. The cocaine-adrenalin solution should be injected subcutaneously, accurately upon the median raphe, over the entire length of the future cutaneous incision. When inserting the hypodermic needle the operator should bear in mind the fact that directly upon the median raphe of the neck there are very few nerve endings, and that consequently the patient will not resist the introduction of the needle at this point nearly as much as he will if the operator carelessly attempts to introduce it the merest trifle to the right or left of the median line. Ordinarily, the most nervous patient will readily stand for the subcutaneous injection without the applica tion of the twitch or other painful or powerful restraint. Such restraint should be avoided as long as possible, as many horses, especially those which are nervous, will resent the application of the twitch far more than the introduction of the hypodermic needle. Upon most horses the entire operation may be carried out without the use of the twitch or other painful or powerful method of control.

After the application of the local anaesthetic has been completed, the patient's head should be freed, and ample time given for the thorough development of the anaesthesia. After ten minutes or more, the skin over the operative area may be cautiously pricked, in order to determine whether the anaesthesia is satis factory. If not complete, further time should be given, and if necessary an additional applica tion of the local anaesthetic should be made.

When the local anaesthesia is complete, the patient's head should again be elevated and the cutaneous incision made, extending from a point one or two inches in front of the anterior end of the thyroid cartilage downward and backward to the level of the first tracheal ring. The incision should be carried through the skin and fascia down to the cartilages of the larynx and trachea, without wounding the peri chondrium. The operator should be careful to make the incision directly upon the median raphe, where the degree of sensation in the skin is very low and the cutaneous tissues are non-vascular. Ordinarily, only a few drops of blood escape from the cutaneous wound. Usually it is necessary merely to mop the blood away with sterile cotton or gauze. In case of necessity, the forceps should be applied to any vessels of recognizable size.

The operator should next definitely recognize by palpation the outlines of the crico-thyroidean ligament. Care must be exercised in this, or the operator may fall into error. The crico thyroidean ligament is triangular, with a rounded apex; its base rests against the cricoid cartilage and its apex against the notch in the thyroid cartilage. When the head is completely ex tended for the operation, the cricoid cartilage is drawn away from the first tracheal ring for a distance, and the intervening ligament may be mistaken for the crico-thyroidean ligament, but it is not triangular like the latter, and there should be no difficulty in differentiating between them if care is taken. Should error arise and an incision be made through the ligament between the cricoid and first tracheal ring, no harm is done except the delay and embarrass ment.

Having fully identified the crico-thyroidean ligament, the operator should place a narrow bladed scalpel with its back resting against the anterior border of the cricoid cartilage, as accurately upon the median line as possible, with its point directed upward and backward so that, when it passed through the crico thyroidean ligament, the point will project behind the arytenoid cartilages. If directed perpendicular to the long axis of the larynx, the scalpel may wound the paralyzed arytenoid cartilage or pass between it and the thyroid cartilage. After the scalpel has safely pene trated the crico - thyroidean ligament the in cision should be carried forward on the median line to the thyroid cartilage. Care should be taken not to wound the thyroid cartilage. The laryngeal speculum (Fig. 286) is then to be placed in position, as indicated in Fig. 290.

In inserting the laryngeal speculum the sliding hook is detached, the jaws of the body of the instrument are closed, and the curved spurs are inserted through the incision in the ligament and pushed backward within the cricoid cartilage. The jaws are then spread apart as widely as practicable, after which the slotted hook is inserted. The distal end of the hook is engaged within the thyroid cartilage. The hook is then pushed as far forward within the thyroid cartilage as possible, and secured.

When the speculum has been properly applied and secured in position, the operator may observe the conditions within the larynx and verify his diagnosis. The larynx should be thoroughly illuminated by means of a reflecting electric lamp, so that every part of the organ may be observed. In many cases the arytenoid carti lages and vocal cords move but little during respiration through the dilated incision, and it is difficult to determine the exact condition. If the operator will insert a long pair of forceps, or other instrument, forward through the in cision and glottis against the base of the tongue, the animal will at once swallow, and in so doing will close the glottis. It can then be seen whether the two arytenoid cartilages move properly. If the two cartilages move vigor ously and alike, the difficulty is not paralysis of the recurrent nerves, but the roaring is due to some other cause. Removing the mucosa from the lateral ventricles is unwarranted in such cases. If the pre-operative diagnosis of roaring is verified by the direct inspection of the opened larynx, the operator should proceed to induce anaesthesia of the interior of the larynx and of the ventricular mucosa. This is best done by grasping with the Pean's forceps a small pledget of surgeon's cotton saturated with the cocaine-adrenalin solution, carefully painting the solution over the interior of the larynx, and then inserting the cotton within the ventricle. The process should be continued until complete anaesthesia is obtained. If this is not done, the animal is liable, because of the pain inflicted, to jerk the head suddenly at some critical point in the operation, interfering seriously with success.

Should the surgeon prefer to operate upon the recumbent patient, and has used general anaesthesia, the speculum should be applied and other details of technique carried out the same as for the standing operation. If the operator prefers not to use the ventricular burr, he should, after the speculum has been securely applied, grasp the mucosa of the ventricle at its margin with the forceps and exert traction upon them while he penetrates the mucosa on the side of the forceps opposite to the ventricular cavity. The incision is to be carried more or less com pletely around the margin of the ventricle, after which the mucosa is to be carefully dis sected away, either with the scalpel, with a closed blunt pair of scissors, or with the index finger.

The removal of the ventricular mucosa in the standing operation is best accomplished with the soft rubber burr. When the local anaesthesia of the laryngeal mucosa is complete, the operator should wipe away any mucus or saliva, either in or about the ventricle, by means of absorbent cotton applied with forceps. The soft rubber burr, about which there has been wound a small piece of dry cheese-cloth or gauze, is then carefully inserted into the ventricle and pushed firmly against its bottom. The burr is then revolved to the right (the burr is ordinarily fastened on the handle by revolv ing the handle to the right, and becomes un screwed if turned to the left) upon its long axis, pressing gently upon the handle until the burr grasps the mucosa, as indicated by the resist ance. As soon as the operator feels that the mucosa is securely engaged, the burr should be given one-half to one complete additional turn, after which traction is exerted, instead of pushing upon the burr. The rotation upon its long axis is continued to some extent during the application of the traction, which tears the mucosa loose from the subjacent tissues. Con tinued slight rotation of the burr ensures its grasp upon the mucosa. The traction is in creased, and finally the ventricular mucosa is drawn out inverted through the mouth of the ventricle. Just as the burr is emerging from the mouth of the ventricle there is a tendency for the mucosa to slip away. In order to avoid this, it is best to press the handle of the burr outwards—that is, lateralwards and upwards— so that the soft rubber burr is forced downwards and inwards against and beyond the vocal cords. At this point the everted mucous sac should be grasped with the forceps between the burr and the vocal cords. The forceps should then be closed upon the sac and the rubber burr released by revolving it in the opposite direction —to the left. A second pair of forceps should now be placed above the first pair, in order to get a more secure hold. Traction is then applied upon the sac, and the inverted mucosa is punctured with a scalpel at the ventricular border along the vocal cord. The incision is extended around the border of the ventricular mouth, either with the razor-shaped scalpel, an ordinary straight scalpel, or blunt curved scissors, as the operator may prefer. The ventricular mucosa when perfectly removed should constitute a closed sac, which may be pressed over the operator's thumb in order to determine whether the removal is complete.

Should the mucosa, when it is being everted, accidentally slip from the burr, it is not always easy to grasp again, because some hiumorrhage may occur inside the ventricle or mucus may be thrown out in large quantity, causing the burr to slip. In such case the forceps may be introduced into the ventricle, the jaws opened, pressed against the bottom of the sac, and closed, thus engaging a portion of the sac. Traction is then exerted upon the forceps and the mucosa is everted. As soon as the everted portion appears at the mouth of the ventricle, it should be grasped above the first forceps with the second pair, and a more secure hold upon a larger part of the mucosa obtained. Traction is then applied upon the second pair of forceps, and the division of the mucosa about the margins of the ventricular mouth is carried on the same as described with the burr.

When the mucosa has been completely removed the denuded tissues in the ventricle may be mopped over with a reliable antiseptic, such as tincture of iodine.

Formerly it was advised to operate only upon the left side of the larynx. More recently it has been found advisable, in the experience of most operators, to include both ventricles, as a rule of practice. Although the paralysis causing roaring is commonly more severe on the left side, it has now come to be understood that in most cases there is a minor degree of paralysis upon the right side also. The paralysis upon the left side is usually so much more marked than that upon the right that it has been erroneously concluded that the disease exists only upon the left side. In many cases of roaring, both nerves are badly involved and little difference can be seen in the tone of the muscles of the right and the left sides. In the clearly bilateral laryngeal paralysis the question of whether one or both ventricles should be operated upon is not open to discussion. As a general rule of practice, in a recent case of roaring it would appear to be best by far to operate upon both ventricles simultaneously. No material harm is to come from the bilateral operation; the danger therefrom is not materi ally greater than from the unilateral operation; and, once the horse has recovered, he is none the worse for having both ventricles destroyed. When the operation succeeds, the only change wrought in the larynx is in the vocal apparatus. Whenever roaring becomes pronounced, the horse is virtually mute because of the paralysis, even if only one side is involved. When the bilateral operation for roaring is performed, the patient becomes permanently mute. Since the loss of the voice is of no consequence ordinarily, this factor may, as a rule, be wholly ignored.

Should one prefer to operate upon the left side only and the operation fails because of paralysis on the right side, the operation may of course be readily repeated. If the roaring is of long standing and the right arytenoid cartilage and vocal cord appear to be normal, it is reasonably safe to risk the unilateral opera tion. In a recent case of roaring, even though the right side appears to be normal, the operator must constantly face the risk of advancing paralysis of the right cartilage, not evident at the time of the operation, but which may develop at a later date and raise a question of the surgeon's prudence in operating upon one side only.

When the removal of the mucosa from the ventricle or ventricles has been completed, the larynx should be wiped free from blood and mucus, the laryngeal speculum removed, and the laryngeal tube inserted. Generally it is best to secure the laryngeal tube in its position by means of sutures in the margins of the cutaneous wound, in addition to tapes tied over the poll. The patient is then to be released and reasonably close watch kept during the following forty-eight hours. The laryngeal tube should be removed and cleansed after twenty four hours. At the end of forty-eight hours the laryngeal tube should be removed and the wound dressed antiseptically. If the respira tion is normal, the tube may be omitted. The invading and corrective wounds may be dressed antiseptically once a day for a few days, after which only the exterior or invading wound should be disturbed.

There is little special care to be taken with reference to feeding, either before or after the operation, except when the patient is cast and anaesthetized. The horse may be allowed an ordinary ration suitable to an animal at rest. Ho should be kept quiet in a box-stall for ten to fifteen days, and may then be allowed to run in a paddock or may take light exercise to the halter. For some time after the opera tion the horse should not wear a halter with a throat-latch. The throat-latch becomes soiled with the discharges and serves to retain them against or in close proximity to the wound. It is best to remove the halter or other headgear entirely.

Ordinarily the wounds heal in about twenty days. The animal may then have moderate exercise, and in four to six weeks may be placed at gentle work. When first tested, the animal may breathe imperfectly, owing to excessive granulation tissue in the ventricle pushing the cartilage and vocal cords into the glottis. Many such cases improve after a time. In some cases the improvement gradually increases for six months or a year after the operation.

Ordinarily the prognosis for the operation is good. Horses desired for racing purposes, which are put to extreme test, do not recover so satisfactorily as ordinary work-horses. In racing horses, approximately 50 per cent recover completely, so that they may race without un natural breathing. In hunters, which are exerted less severely, the percentage of recoveries is decidedly higher. In horses used for ordinary work the recoveries reach approximately 80 per cent. The animals which do not recover completely are nearly all materially improved. The failures are largely dependent upon chon dritis due to injury to the perichondrium in the course of the operation or to errors in the extent of denudation of the ventricle. The technique, as here given, is designed to avoid any and every injury to the cartilaginous tissues. The more careful the operator is in avoiding these injuries, the higher will be the percentage of recoveries. Whenever the cartilage is in jured it tends to become infected and inflamed, followed by chronic thickening, and frequently by limited necrosis surrounded by granulation tumours, which serve to occlude the glottis or trachea. When these growths occur, they are not ordinarily subject to surgical remedy. They may be removed readily enough, but the necrosis continues and the tumour-like masses again develop. Therefore it is highly important that the injuries to the cartilages shall be pre vented.

mucosa, cartilage, ventricle, burr and vocal