Alexis Carrel

Nobel Lecture

Nobel Lecture, December 11, 1912

Suture of Blood-Vessels and Transplantation of Organs

The idea of replacing diseased organs by sound ones, of putting back an amputated limb or even of grafting a new limb on to a patient who has undergone an amputation, is far from being original. Many surgeons before me have had this idea, but they were prevented from applying it, owing to the lack of a method for reestablishing immediately a normal circulation through the transplanted structures. It was of fundamental importance to first discover a suitable method of uniting the blood-vessels of the new organ to those of its host. In 1902, therefore, I began to investigate by what means a vascular anastomosis might be effected without producing either stenosis or thrombosis. Many surgeons had previously to myself performed vascular anastomosis, but the results were far from satisfactory. I began by using Payr’s and Murphy’s methods, after which I proceeded to study the principles for a new technique on human cadavers. I next performed some vascular anastomoses on living dogs at the University of Lyons in the laboratory of Professor Soulier and with the collaboration of Dr. Morel. This study was continued at the University of Chicago in Professor Stewart’s laboratory and with the collaboration of Dr. Guthrie. Later, at the Rockefeller Institute for Medical Research, the causes of all possible complications were analysed and greater perfection of methods was obtained. With this modified technique a great many experimental operations were performed and their clinical and anatomical results were observed during a period of three and four years. As a result the study of vascular anastomosis can today be considered as completed from the standpoints both of the technique and of the experimental results.

Vascular anastomoses

General rules
In operations on blood-vessels certain general rules must be followed. These rules have been adopted with the view of eliminating the complications which are especially liable to occur after vascular sutures, namely, stenosis, haemorrhage, and thrombosis. A rigid asepsis is absolutely essential. Sutures of blood-vessels must never be performed in infected wounds. It seems that the degree of asepsis under which general surgical operations can safely be made may be insticient for the success of a vascular operation. It is possible that a slight non-suppurative infection, which does not prevent the union of tissues “per primam intentionem”, may yet be sufficient to cause thrombosis. The obliteration of the vessel also follows injuries to its walls. The arteries and veins can be freely handled with the fingers without being injured, but it is often harmful to use forceps or other instruments. If a forceps be used, it must take between its jaws nothing but the external sheath. When temporary haemostasis is obtained by means of forceps or clamps, these instruments must be smooth-jawed and their pressure carefully regulated. The dessication of the endothelium may also lead to the formation of a thrombus. Therefore, during the operation the wall of the vessels must be humidified with Ringer’s solution or be covered with Vaseline. The presence of coagulated blood, of fibrin ferment, or of foreign tissues or tissue juices on the intima can determine the production of a thrombus. It is, therefore, necessary to resect or to remove the external sheath from the edges of the vessels, for if, during the suture, it gets between the edges of the vessels or around the silk threads, it very quickly leads to a deposit of fibrin, and possibly to an obliterative thrombus. By washing the vessel carefully with Ringer’s solution and coating it and the surrounding parts of the operating-field with Vaseline, we can efficiently protect the endothelium against coagulating blood and the juices of tissues. Vaseline also prevents the fingers of the operators and the threads from becoming soiled by fibrin ferment. As perforating stitches are always used, the endothelial layer is necessarily wounded by the needle. These wounds, however, are rendered as harmless as possible by the use of very fine and sharp round needles. Extremely small wounds are made. The threads are sterilized in vaseline and kept heavily coated with it during the suture, thus preventing the wounded tissue and the foreign material from coming into actual contact with the blood. Stenosis is very liable to occur at the point of anastomosis, therefore the suture must be performed while the wall is under tension. The tension can easily be obtained by traction on retaining stitches properly located. Great care is taken to obtain a smooth union and approximation of the internal coats, in order to produce as small a scar as possible. When these general rules are observed the operation can be safely performed.

The instruments employed are very simple. Temporary haemostasis in the case of small vessels is secured by an ordinary serrefine, the spring of which is carefully regulated, and in the case of large vessels by Crile clamps or smooth-jawed Gentile forceps. These instruments are all smooth-jawed and are applied to the vessel without a cover of rubber. The vessel is washed in Ringer’s solution by means of a Gentile syringe, which is composed of a rubber bulb and a glass-tube narrowed at its end. The anastomosis is performed with round, straight Kirby needles. No.16 is used for the small vessels and No.12 for the large ones. Short, curved needles can also be used, but the straight ones are always preferable. The needles are threaded and the thread is rolled on cardboard, after which the needles and threads are placed in a hermetically closed jar, which is put in the sterilizer under fifteen or eighteen pounds’ pressure. The needles and thread are removed from the jar whilst the Vaseline is still warm and deposited on a black towel. During the operation the threads are fixed by means of small Gentile haemostatic forceps. The operating-field is circumscribed by a black Japanese silk towel on which the fine threads can easily be seen.

(1) Temporary haemostasis and preparation of the vessels. The vessels are exposed by a large incision and they are freely dissected. The dissection of a very long segment of artery or vein is not dangerous. Careful haemostasis of the wound is made, because during the suture the operating-field must be free from blood. The temporary haemostasis is secured by clamps placed at a few centimeters’ distance from the location of the suture. If there are collateral branches between the forceps and the point of section, they are ligated or clamped. Then the vessels are cut and the external sheath is resected from the ends of the vessels. The end of the syringe is introduced into the lumen of the vessel and the blood is washed out of the vessel and from the operating-field. The fluid is removed with dry gauze sponges. Then the vessels and the surrounding parts are covered with vaseline and a black Japanese silk towel is placed around the ends of the vessels.

(2) Suture. We shall now proceed to describe the technique of the suture, selecting as our type the termino-terminal anastomosis. Then we will point out the modifications which take place in the termino-lateral and latero-lateral anastomoses.

The termino-terminal anastomosis is effected by bringing the extremities of the vessels into contact, no traction being necessary. The ends are united by three retaining stitches located in three equidistant points of their circumference. By traction on the threads the circumference of the artery can be transformed into a triangle, and the perimeter can be dilated at will. Then the edges of each side of the triangle are united by a continuous suture whilst they are under tension. During the suture great care is taken to approximate exactly the surfaces of section of the wall. Before the last stitch is made, the remaining Vaseline is removed by pressure from the lumen of the vessel. In venous anastomoses the ends of the veins are also united by three retaining stitches. A venous suture, however, requires more stitches than an arterial suture, on account of the thinness of the walls. The union of the extremities is made by eversion of the edges, which are united not by their surface of section, but by their endothelial surfaces. An inversion of the edges would be very dangerous and would provoke the formation of a thrombus. In arterio-venous sutures the vein is generally larger than the artery. After the ends of both vessels have been united by the retaining stitches each side of the venous triangle is longer than the corresponding side of the artery. By traction on the retaining stitches the arterial lumen can be greatly dilated and be brought nearer to the size of the vein. Each stitch of the continuous suture is made larger on the vein than on the artery, and the size of the vein is thus progressively reduced and a good union ensured. During the suture the venous wall is turned outward and its endothelial surface is applied to the surface of section of the arterial wall.

The termino-lateral anastomosis consists in implanting the end of one vessel on to the wall of another vessel. A triangular or elliptic incision is made in the wall of one vessel, whereupon the edges of the opening are united to the extremity of the second vessel by three or four retaining stitches, and the operation finally concluded by a continuous suture. If the vessel is of small caliber it is better to perform the termino-lateral anastomosis by the patching method. This consists in dissecting the small vessel to the level of its implantation in a larger trunk and in resecting the patch of the wall of this trunk which surrounds the mouth of the small artery. This patch must be elliptic in shape and its edges as regular as possible in order that a smooth anastomosis can be made. The patch is then inserted into an opening made through the wall of the larger vessel and fixed by four retaining stitches and a continuous suture. This method is safer than the direct implantation, for the formation of a slight thrombus on the line of suture would not impede the circulation of the large vessel, whereas the lumen of the smaller vessel would be obliterated if the thrombus were formed at its mouth.

The latero-lateral anastomosis is performed by placing two vessels parallel to one another and opening them longitudinally by incision or resection of an elliptic flap. The ends of the openings are united by two posterior retaining stitches and by a continuous suture performed on the interior of the vessels whilst their walls are under tension. A third retaining stitch unites the middle points of the anterior edges, the posterior line of suture being still stretched by traction on the two retaining stitches. Finally by traction on the posterior and anterior retaining stitches the opening is transformed into a triangle and the anastomosis is completed by a continuous suture running along both anterior sides.

(3) Reestablishment of the circulation. Before the circulation is reestablished the line of suture must be carefully examined, and if a gap is found between two stitches it is immediately closed. Next, gauze sponges are placed on the line of suture and the clamps are removed, while a gentle pressure is exercised on the sponges. There is almost always some leakage during the first minutes. After two or three minutes the sponges are removed, and if the haemorrhage persists one or two complementary stitches are added. It is of fundamental importance that the wound be not closed before complete haemostasis of the anastomosis has been obtained. Finally, the vessels and the operating-field are washed in Ringer’s solution and the wound closed without drainage.

A great number of vascular sutures and anastomoses were performed by this method on dogs and cats, and some on man. The operations were made on arteries and veins of both large and small caliber. The simplicity of the technique makes it readily adaptable to all vessels. It is as easy to make a circular suture of the delicate renal vein of a small-sized dog as of the thoracic artery of a large dog, or even of the femoral artery of man. In aortic sutures, where the walls are very friable, the edges of the vessel are jointed by u-shaped stitches, instead of by a continuous suture. If it is necessary not to interrupt the circulation during the suturing, a diversion of the flow can be practised by means of a paraffined tube, which is inserted into the lumen of the vessel and afterwards removed by means of an incision in the wall. But in nearly all cases it is possible to interrupt the circulation for the space of a few minutes. Even during the operation of the Eck fistula or the circular suture of the thoracic aorta, the circulation in the portal vein or in the aorta was completely interrupted. The operation is always simple, except in those cases in which the region to be operated upon is deeply situated. It is possible to perform the suture of the carotid artery of a medium-sized dog in two and a half minutes, and of the aorta of a dog or of the femoral artery of a man in five or six minutes. Generally there is nothing to be gained by hastening the operations, as they could last 10 or 20 minutes without any harm resulting.

The results were as simple as the technique. No haemorrhage was ever observed and no stenosis was ever produced at the location of the anastomosis, provided a proper tension was made on the retaining stitches during the suture. Occasionally I noticed in the Eck fistulas a reduction in the size of the vascular opening, which was due, doubtless, to deposits of fibrin. The commonest complication which was apt to take place in connection with vascular sutures, viz. thrombosis, never occurred when this method was properly employed. In sutures of vessels of the caliber of the carotid of a middle-sized dog, as well as in those of smaller vessels, it may be said that an obliteration occurred in hardly five out of one hundred cases. In the sutures of arteries or veins of larger caliber deposits of fibrin could sometimes be detected on the line of the suture, but never an obliterative thrombosis. In short, we can say that thrombosis has become an altogether unusual complication in cases of vascular operations.

The anatomical evolution of vascular sutures was studied during a period of time varying from several days to over four years. One week after the operation all the stitches could still be seen on the endothelial surface of the vessel. In certain cases the line of suture was covered by a very fine layer of fibrin. Little by little the threads became invisible and at the end of some months no trace of the suture could be perceived except a more or less indistinct transverse line. Sometimes this line was slightly depressed, but there was never any trace of dilatation or of stenosis at the location of the suture. Sometimes, again, the line of juncture became entirely invisible, and this occurred in the case of the suturing of the renal artery, for, six months after the operation, it was impossible to locate the place where the anastomosis had taken place. Careful examination disclosed the place where the wall of the vessel had become slightly thicker, but no signs of the operation could be detected on the inner surface. Similarly, after the extirpation and replantation of the spleen, performed with a circular suture of the splenic vessels, an examination of the endothelial surface of the vessels, conducted twentytwo months after the operation, likewise failed to show where the anastomosis had taken place. It was only after the vessel had been placed in formalin, that two faint whitish lines began to appear, indicating the place where the suture had been performed. In the case of veins, where the walls are thinner, the marks of the operation remain for a longer time. Nearly all the stitches of a suture of a renal vein could still be seen as long as ten months after the kidney had been replanted. Certain animals on which circular sutures of the thoracic or abdominal aortas or of the carotid artery had been performed, and which we had retained in health two, three, and even four years afterwards, showed the same results. A histological examination of the sutured vessels readily explains the perfection of the clinical results obtained. The union produced by the suture is so exact that the scar resulting from the junction of the extremities of the vessels is in consequence very slight, and in some cases the medias become directly united without the interposition of any fibrous tissue. A longitudinal section of the anastomosis of the abdominal aorta of a cat, to which a segment of the aorta from another cat had been transplanted, showed that the wall at one of the extremities was slightly everted and had united with the surface of the section of the intima at the other extremity, without the interposition of fibrous tissue. At a short distance from the line of suture the elastic tunic was seen to be interrupted, because the section had encroached upon one of the silk threads surrounded by fibrous tissue. This result was exceptionally good. In most cases the elastic tunic was interrupted by a faint scar at the place of the anastomosis. The faintness of the scar explains why there was never any stenosis or dilatation at the level of the suture lines.

By means of this technique I have been able to successfully perform sutures of incisions made on the ascending part of the aortic arch, and likewise to perform circular anastomoses of the thoracic aorta, of the abdominal aorta, of the vena cava, and of the majority of the vessels of the organism. Lateral anastomoses were employed in arterio-venous sutures and in the operation of the Eck fistula. Finally, by applying this technique to operations on the human body, I have shown it to be equally simple and efficacious as in the case of animals. It has made possible the successful transplantation of arterial and venous segments, of organs and members.

Transplantation of blood-vessels
It is possible to perform two kinds of vascular transplantations, uni-terminal transplantation and bi-terminal transplantation. The uni-terminal transplantation consists in grafting one extremity of a vessel on to another part of the vascular system, and includes many different varieties. The main result of a uni-terminal transplantation is to change the quality of the blood circulating through a certain vessel and to modify its pressure. If the peripheral end of the external jugular vein is united to the central end of the carotid artery, the vein becomes filled with red blood and its pressure is augmented, whilst the pressure of the artery diminishes. The vessels readily adapt themselves to these modifications. An augmentation of the pressure in the carotid artery, produces a thickening of its walls. After several months, hypertrophy of the muscular coat and sclerosis of the intima arise. If on the other hand, the pressure of the carotid artery is diminished by anastomosing it with a large venous system, such as the superficial veins of the neck, its wall becomes thinner, and hypertrophy of the intima is produced. When an arterial circulation is established through a vein, the wall of the latter rapidly responds to the change by becoming thicker. When the carotid vein and the external jugular vein are united, the wall of the jugular vein becomes much thicker, although its pressure is less than ordinary arterial pressure. When the wall of the vein has to bear still more pressure, as, for instance, when the central end of the carotid artery is united to the peripheral end of the inferior thyroid vein, the reaction of the venous wall is very much greater. The lumen of the vein diminishes and the wall shows an enormous sclerosis.

These uni-terminal transplantations were used in order to modify the circulation of certain vascular regions. Thus it was possible to produce a reversion of the circulation in the posterior limb by performing a cross anastomosis of the femoral artery and vein. In many instances we have modified the circulation of the veins of the neck and head as well as of the veins of certain organs. The union of the carotid artery and of the external jugular vein allowed the blood to pass from the carotid into the superficial veins of the neck and of the head, whereupon the latter pulsated like arteries. It was also possible to diminish the pressure of the upper carotid by uniting its central end to the central end of the jugular vein, and thereby the arterial blood was made to return immediately to the heart through this vein. In various other experiments I was able to reverse the direction of the circulation of an organ, such as the thyroid, in which case the thyroid vein was united to the carotid and the thyroid artery to the jugular.

It is possible that some of the numerous operations performed by means of uni-terminal transplantation of blood vessels may some day be of practical use. So far these operations have above all shown how sensitive the walls of the vessels are to changes in the pressure of the blood, and how readily, by becoming thinner or thicker, they adapt themselves to these modifications.

The bi-terminal transplantation consists in interposing a segment of a vessel between the ends of another vessel. This transplantation can be carried out in various ways, according as the segment is interposed between the cut ends of an artery or vein, or between two different regions of the circulatory system, or, again, between a serous membrane and a vein. If the transplanted segment is from the same animal, or from another animal of the same kind, or from a different kind of animal, the transplantations are autoplastic, homoplastic or heteroplastic respectively.

The first arterial transplantation was performed in 1896 by Jaboulay. He thought that this method could be applied to the treatment of aneurisms, but the anastomoses were imperfect and in each case thrombosis of the vascular segment occurred. Nine years later Hoepfner successfully performed the transplantation of segments of arteries by using the method of Payr. In a remarkable series of experiments he demonstrated the possibility of transplanting arteries without any obliteration resulting therefrom. In 1905, at the University of Chicago, with the help of Dr. Guthrie, I took up this study. In a large number of experiments conducted subsequently at the Rockefeller Institute, the results of this method were studied in greater detail. Segments of the carotid artery or aorta of a dog and a cat were transplanted on to the same animal, another animal of the same species, and an animal of a different species. The vessels used were either fresh or had been preserved in cold storage. These operations performed with fresh vessels of proper caliber invariably showed positive results. The circulation continued in a normal manner and at the end of several months the transplanted segment usually presented the same appearance as the normal vessel.

In the case of autoplastic transplantations there was no change in the appearance of the vessel, which preserved its normal diameter and its elasticity. The only trace left by the operation was two whitish transverse lines, which were visible on the endothelial surface of the wall of the vessel. The histological examination likewise showed that the wall of the vessel had undergone no change.

In homoplastic transplantations the appearance of the vessel also remained normal. When the vessel was examined on the living animal six months or more after the operation it was usually impossible to locate the transplanted segment. Even in cases in which the vessel had been preserved for several weeks in cold storage the results were the same. But a histological examination made three months or more after the operation often showed that the histological constitution of the wall had become modified. In some cases the wall had remained normal, but in the majority of cases the muscular fibres had completely disappeared and the intima had thickened. The elastic fibres of the media always remained in a normal condition, even after several years. Hitherto it has not been possible to recognize before operating the animals that may undergo an exchange of vascular segments without suffering degeneration of the muscular fibres, and those whose muscular fibres would, after such an exchange, completely degenerate.

In heteroplastic transplantations the results were different. Segments of the carotid from dogs were grafted on to the aorta of a cat or the segments of the aorta of a cat on to the carotids of dogs. Also segments of human aortas were transplanted on to the aorta or carotid of dogs. Obliteration often occurred in cases where human vessels were transplanted on to other human beings, but this obliteration may possibly be attributed to secondary causes, as the vessels employed had been preserved in cold storage. In nearly all cases in which the carotid of a dog had been transplanted on to the aorta of a cat the results were positive and the circulation remained normal. The clinical results were observed during a period of time varying from several days to more than four years. The anatomical evolution of the heterologous vessels is different from the evolution of the autologous and homologous vessels. Their caliber becomes progressively larger, but this augmentation of the diameter does not augment indefinitely and never results in the formation of aneurisms. Similarly the wall becomes thinner. The histological examination showed that, during the weeks and the months following the transplantation, not only the muscular fibres had degenerated but the elastic framework had also completely disappeared. Thus the wall only consisted of connective tissue. A dog, which in 1907 had undergone the resection of the abdominal aorta and the transplantation of a segment of a human popliteal artery, died four years after the operation. The vascular segment was slightly dilated and the anastomoses were normal. But the wall contained neither muscular nor elastic fibres. The same results were observed after the arteries of a cat or of a rabbit were transplanted on to a dog. The lesions consisted in the progressive disappearance of the elastic and muscular elements of the wall, of an augmentation of interstitial connective tissue and of a thickening of the intima. But the resistance of this fibrous wall was sufficient and the segment of the artery was able to take the place of an artery during several years. In these arterial transplantations fresh segments, or such as had been preserved in cold storage, were used. The clinical results were identical in both cases, that is to say, when the period of preservation in cold storage had not exceeded two or three months. Vessels preserved in formalin or by means of a system of thorough drying were also used. In the latter experiments the wall of the vessel, which was devitalized, had become regenerated at the expense of its host.

The transplantation of veins was attempted for the first time by Glückin 1898, but the segment of the jugular vein which he transplanted on to the carotid artery of a dog became quickly obliterated. Exner attempted to perform the same experiments, as did also Hoepner and Goyanes. The results of these operations, however, were always negative. In 1905 I succeeded with Guthrie in transplanting segments of the jugular veins of a dog on to the carotid artery. Later on I was able to graft segments of the vena cava on to the abdominal aorta or on to the vena cava of dogs or cats. Finally I succeeded in performing the same operation on the thoracic aorta of a dog. The upper part of the descending aorta of a fox-terrier having been cut transversely, a segment of the jugular vein of a large dog was interposed between the extremities of the vessel. The animal remained in an excellent condition. More than two years after the operation he died of poliomyelitis, which at that time was epidemic in dogs. The autopsy showed that the venous segment was not dilated and that the valves were in the same condition as at the time of the operation. The muscular fibres of the wall had disappeared, but the diameter of the vessel was practically the same as that of the neighbouring parts of the aorta.

I have performed many venous transplantations, both autoplastic and homoplastic. In the case of the autoplastic transplantations the vein immediately underwent important anatomic modifications. The wall became thicker while the lumen became slightly dilated; but aneurism never occurred. On the contrary, the vein adapted itself to its arterial function, and responded to the pressure of the blood by thickening its wall. This thickening was mainly produced by an augmentation of the connective tissue of the adventitia and of the intima. Sometimes, too, the number of muscular fibres increased. Four months after the transplantation of the segment of the jugular on to the carotid of a dog, I removed a small piece of the wall and sutured the wound. The histological examination showed a large increase of the connective tissue of the adventitia and of the connective tissue of the intima, but the muscular fibres were normal. Twelve months later the venous segment was examined again. It was found that the wall had become still thicker and that all the muscular fibres had degenerated. In the homoplastic transplantations the clinical evolution showed the same results, but all the muscular fibres disappeared more or less rapidly. These experiments served to demonstrate, therefore, that the segment of a vein, notwithstanding the thinness of its wall, may perform the function of an artery. From a practical point of view it is possible to replace segments of arteries by sections of veins. On human beings this operation would be easier than the transplantation of arteries, as it is always possible to find on the patient himself a segment of vein suitable for transplantation.

In still other experiments a piece of the wall of an artery, such as the aorta or carotid, was extirpated and replaced by a fragment of vein or peritoneum. In some cases the entire section of a vessel was removed and replaced by a tube of peritoneum, but in this case thrombosis of the tube occurred in a few days. When a piece of the wall was resected and replaced by a patch of perito-neumor vein, the results were always positive. The adaptation of the pieces of peritoneum or vein to their new conditions was so perfect, that in less than two years after the operation the wall of the vessel was almost normal, although it showed a slight thickening at the point where the operation had been performed. The exterior appearance of the abdominal aorta of a dog, on to which a patch of peritoneum had been grafted, was in no wise modified, although the wall was composed of a different tissue. It possessed the same thickness and appearance as the adjoining normal portion of the wall. By a kind of mimetism the connective cells of the peritoneum had taken on the aspect of muscular fibres. In spite of the absence of muscular and elastic fibres, such as those possessed by the normal aorta, the new wall supported the pressure of the blood without dilating. Aneurism was never produced unless there was bacteriological infection. The only way in which I succeeded in provoking fusiform aneurisms was by transplanting tissues which had been infected. I also succeeded in repairing the wall of an artery with the aid of foreign organic and inorganic substances. A portion of the anterior wall of the abdominal aorta of a dog was replaced by a piece of rubber. Fifteen months after the operation the function of the aorta had not become modified and its lumen was normal. A thick layer of connective tissue had formed on the inside and outside of the venous piece. In other experiments glass and metal tubes covered with paraffin were introduced into the thoracic aorta of the dog. The blood did not coagulate as long as the wall of the artery was not ulcerated. A dog whose thoracic aorta contained a piece of glass remained in perfect health for 90 days and finally died of an ulceration of the aorta. These experiments showed that the presence of foreign bodies in the lumen of a vessel does not necessarily produce a thrombus, as was formerly believed. This observation caused a simplification of the technique of the transfusion of blood to human beings.

These various experiments on blood-vessels demonstrated that arterial or venous transplantations could be performed with safety when a proper technique was used. It was of little importance whether the muscular fibres were preserved or not, as the vascular wall adapted itself to the pressure, and no dilatation occurred. Autoplastic and homoplastic transplantations of arteries and veins may, therefore, be adopted in surgical practice for those cases in which it is necessary to reestablish the continuity of an artery.

Transplantation of limbs and organs
The question of the transplantation of tissues and organs must be considered from two points of view, the surgical and the biological. I began by seeking to develop a technique which would enable the circulation to be reestablished in the organs without delay. After that I endeavoured to ascertain to what extent the organs fulfilled their functions, according as the transplantations were autoplastic or homoplastic. I found that an entire anatomic region could be transplanted from one animal to another. The following observation is an illustration of the transplantation from one dog to another of the anatomic structures supplied by the external carotid artery. From a first dog the right part of the scalp and the auricle were extirpated in one mass with the auditory canal, the connective tissue and the glands of the retro-maxillary space and the upper portions of the external jugular vein and of the common carotid artery. The dog was then chloroformed. The anatomic specimen was perfused with Locke’s solution. Then it was wrapped in a silk towel saturated with Vaseline and placed on a table at the temperature of the laboratory. On a second dog the auricle and the adjoining portion of the scalp were extirpated and the right part of the neck opened through a longitudinal incision. The anatomic specimen was then placed in the wound, and the peripheral ends of the carotid artery and of the jugular vein were united to the central end of the corresponding vessels of the host. The circulation was then reestablished. An abundant haemorrhage occurred from the little peripheral branches of the carotid artery, which were immediately ligated. Then the neck was closed by two rows of suture. The purpose of making the vascular anastomosis very far from the transplanted ear was to prevent the occurrence of thrombosis, if infection should be produced at the level of the external auditory canal. A few minutes after the establishment of the circulation the ear and the scalp assumed the normal appearance of the internal auditory canal. The new ear was fixed to the external auditory canal of the host. The auricular muscles were sutured and the operation was completed by continuous suture of the skin without drainage. During the days following the operation oedema of the auricle occurred, but the circulation remained excellent. The temperature of the transplanted ear was much higher than that of the normal ear. The cutaneous wounds united by first intention, but subsequently a voluminous abscess appeared under the scalp and had to be drained eight days after the operation. The oedema of the auricle diminished and disappeared, but the animal had some fever and a secondary abscess appeared in the left leg. The animal developed a slow form of pyaemia. Nevertheless, the local condition became excellent. The temperature of both auricles, normal and transplanted, were about the same. The transplanted ear was as thin and glossy as the normal one and its circulation was normal. The animal finally died, three weeks after the operation, of pyaemia. The transplanted parts were normal. I performed several other operations similar in character, without any infection arising. The transplanted ear remained normal for from six to eight days. After that the transplanted tissues became oedematous and in every case necrosis of the skin developed. The animals had then to be chloroformed. In no case did the lesions appear to have been caused by a defect of the circulation, which was apparently normal. It is therefore safe to assume that they resulted from the reaction of the host against the transplanted anatomic region, a reaction which failed to take place in the first experiment.

In 1907 I commenced the study of the transplantation of limbs. This study had been preceded by various experiments conducted by other surgeons who had studied the subject of the replantation of limbs. In 1891 Robert Abbe amputated almost completely the fore-limb of a dog. He cut the entire limb with the exception of the femoral vessels. Then he sutured the bone, the nerves, the muscles and the skin, and the animal recovered completely. If he had had at his disposal an efficient method of anastomosing blood vessels, he would have doubtless accomplished a real replantation of the limb. Hoepfner in 1903 amputated the thigh of a dog and replanted it immediately by suturing the bones and all the anatomic structures of the limb, and by uniting the vessels according to Payr’s prothesis. In one case the circulation remained excellent. Unfortunately the animal died eleven days after the operation. The tissues of the limb were well united, but the bone had very little tendency to callus formation. In 1905, in Chicago, I performed with Guthrie several experiments of the replantation of the thigh. The tissues and vessels healed by first intention, but we were unable to keep any animal alive longer than eleven days. I performed no more replantations and after this I proceeded to study the transplantation of limbs from one animal to another. These experiments were made at the Rockefeller Institute in 1907 and 1908, and they consisted in transplanting the front or back limbs from one dog to another. The animals used for these experiments were of no special variety, and they belonged neither to the same race nor to the same family. I merely endeavored to procure two animals of the same size. First of all a dog was killed with chloroform, then the limb was amputated circularly in its middle part, minute aseptic precautions being observed. In the case of the transplantation of the thigh a circular incision was first made at the level of the upper part of the limb. The sartorius muscle was cut and the femoral vessels dissected in such a way as to enable them to retain a considerable length. The limb was then amputated and the surface of the section, covered with a silk towel saturated with Vaseline. By means of a canula introduced into the femoral artery the limb was completely perfused with Ringer’s solution. It was considered advisable to eject all the blood from the limb, as by so doing there was no danger of embolisms arising upon the reestablishment of the circulation. The animal upon which the limb was to be grafted was anaesthetized in the usual manner and the thigh was amputated transversely after dissection of the femoral vessels. It was necessary to calculate the exact length of the vascular pedicle, for tension on an arterial or venous suture is a dangerous thing. Moreover, if after a suture has been performed a vein is left too long, it doubles up and the circulation is impaired. Thus, whether the vessels be too short or too long, in either case congestion of the limb would result. The operation was then begun by fixing the new limb to the stump of the host by means of an Elsberg splint. When the junction of the femur had been secured, the periosteum was sutured. After this the internal muscles of the thigh were sutured and the femoral vessels and nerves of the limb were united to the corresponding vessels and nerves of the host. The circulation was reestablished and the haemostasis of the surface of the section of the member was performed. The operation was concluded by suturing with catgut each muscle and aponeurosis and then the superficial aponeurosis and the skin. The whole operation lasted about one hour and a half.

The technique of the transplantation of the front limb is somewhat different. The humerus was amputated a little above the elbow, whilst the nerves and vessels were severed much higher up. In some experiments the leg alone was transplanted and the tibia was cut a little below the knee. In this operation a large strip of skin was preserved, containing the saphenic vessels, which were dissected at the level of the femoral vessels. The circulation was reestablished by uniting these segments of the femoral vessels with the femoral vessels of the host. But the general lines of the operation were similar to those above described. The limb was next confined in a plaster-of-Paris cast and the animal was laid on a small bed, consisting of a sheet stretched on a metal frame. The animal was fixed there by means of two broad linen bandages and was watched night and day by a nurse. From time to time it was liberated and made to walk round the room, thus taking some exercise each day. It is difficult to keep an animal in good condition, after it has suffered transplantation of the thigh. Little by little we were led to observe all these minute details, in order to avoid various complications which invariably arose in cases where a simpler procedure was adopted. The immediate results of these operations were excellent. The limb resumed its normal temperature and color. The arterial pulsations could be felt, not only in the femoral artery, but also in the branches of the saphenic artery and in the interdigital spaces. During the first few days there was no sign of oedema, the animal suffered no pain, ate in a normal manner, walked about and remained in excellent health. But after seven or eight days a difference was apparent, which produced very marked results. In nearly every case the oedema appeared at the level of the toes, ascended to the leg and thigh and stopped at the line of suture. The circulation, however, remained normal. Next, the oedema became harder and the skin became bluish in places. Notwithstanding this fact, however, an incision made in the skin produced a flow of red blood. After several days the skin became necrosed, although the inner circulation remained normal. At this stage the animals were chloroformed. It seems that this necrosis of the epithelium was not produced by the circulation but by a reaction of the organism against its new limb. In certain rare instances these phenomena did not occur and the limb seemed to adapt itself to its new conditions of life.

The best result was obtained in one of the first experiments, performed in April, 1908. The leg of a fox-terrier was amputated at the level of its upper third. This was then immediately transplanted on to a white, young female fox-terrier of the same size and shape as the first dog. In the evening of the day of operation the circulation was excellent and the foot was pink and warmer than the fellow foot. The animal recovered rapidly. She had a keen appetite and was gay and affectionate as before the operation. Several times she broke the plaster cast and even twisted her leg a little. During the first few days the temperature of the new foot was higher than that of its fellow; it was oedematous too, but this feature entirely disappeared after a few days. A week later the temperature of the feet was the same. The new leg united with the stump of the host by first intention, and fifteen days after the operation it had quite healed. The skin had the same appearance above the line of suture as below it and there seemed to be no vasodilatation of the new limb. Owing to the fact that the Elsberg splint had become twisted, the extremities of the bones had set somewhat imperfectly and in consequence the tibia was bent outwards a little. The other parts of the limb looked quite normal and it would have been impossible for anyone unacquainted with the fact to detect that the leg did not in reality belong to the animal. At this stage the animal began to cough. Then it refused to eat and rapidly grew thinner. On the twentieth day it was very sick. The plaster-of-Paris cast was removed and the leg examined. The new limb was in excellent condition and there was neither oedema nor any kind of trophic trouble. The temperature was the same as that of the normal leg. The bone had hardened and the animal walked round the room on its three sound legs without attempting to use the new limb. The cutaneous anaesthesia persisted. The skin was as soft and the coat as bright on the transplanted leg as on the normal one. The general health of the animal rapidly grew worse and it died on the twenty-second day after the operation. The circulation of the leg remained normal until the end. The autopsy showed diffuse broncho-pneumonia, which carried off a number of other dogs about the same time. In two other experiments the results were almost identical. It should be mentioned that the three animals which showed the best local results were animals suffering from a general infection of a serious nature. The same fact had previously been observed in the transplantation of the auricle. This may be a matter of pure coincidence, although one may be permitted to conjecture that the tolerance shown by an organism towards its new member may vary in degree according to certain general conditions.

During the last seven years I have performed a large number of experiments in the replantation and transplantation of organs, in the course of which the technique relating thereto has been thoroughly worked out. In 1902, in conjunction with M. Morel, I performed my first experiment on the extirpation and replantation of the thyroid gland of a dog. The vessels were sutured, but thrombosis speedily occurred. In 1905, with the collaboration of Mr. Guthrie, I succeeded in replanting a thyroid gland, accompanied by a reversion of the circulation. The circulation remained normal. These experiments were confirmed in 1907 by Stich and Makkas, who found that the replanted glands remained normal from a histological point of view. After that I performed other experiments in the transplantation of the thyroid gland, the replantation and transplantation of the suprarenal gland, of the ovaries, the intestine, the heart and the spleen. Some of the animals operated on lived a long time and the anatomic results were examined several years after. The spleen is anatomically well adapted for use in these experiments, and the replantation of this organ can be taken as characteristic of this kind of operation. The abdomen of a large or medium-sized dog was cut open by a longitudinal incision and the spleen was drawn to the exterior. The gastrosplenic vessels were then separated, ligated and cut. The temporary haemostasis of the splenic vessels was secured by means of two serrefines, after which the pedicle of the spleen was cut and the organ extracted. The organ was next washed as thoroughly as possible in Locke’s solution, by means of a cannula placed in the splenic artery of the organ, after which it was placed in a jar containing the same solution. The central ends of the splenic vessels were prepared for anastomosis. The spleen was taken out of the jar, replaced in the abdominal cavity and the central ends of the splenic vessels were united to the vessels of the organ. Immediately after the circulation had been reestablished the spleen became red and it greatly increased in size. The nerves were sutured, the organ was suspended in the large curve of the stomach by suturing the pedicles of the gastrosplenic vessels, and the abdominal wound was closed up. This operation was performed in two cases, and both animals completely recovered. One of them, which was small in size, was examined fifteen months after the operation. The spleen had entirely disappeared. This atrophy was doubtless caused by thrombosis of the splenic artery, which was very small. In the second case the animal was a large dog and the splenic vessels were of proper dimensions. The anatomical result was examined twenty-one months after the operation. Both the spleen and its vessels were normal. The sutures were almost invisible and their exact location could only be detected after a close examination had been made of the internal wall of the vessels.

The largest number of experiments were performed on the kidney. These can be divided into two classes – autoplastic transplantations and homoplastic transplantations. The first autoplastic transplantation of a kidney was performed by Ullmann in 1902. He took out a dog’s kidney and transplanted it into the neck, suturing the blood-vessels by means of Payr tubes. During the same year I similarly undertook two autoplastic transplantations of a dog’s kidney, but in each case septic complications arose. Similar experiments were also conducted by de Castello, Karl Beck and Floresco. In 1905, with the help of Guthrie, I succeeded for the first time in witnessing the functioning of a kidney that had been transplanted into the neck. We also succeeded in performing the first homoplastic transplantation of the kidney, which consisted in grafting both kidneys of one dog “en masse” on to another dog. The majority of the subsequent operations were conducted at the Rockefeller Institute, where I was able to make a systematic study of the results of replantation and transplantation of the kidney. The operation of replanting the kidney was performed principally on dogs. After a few experiments it was found that the only way of enabling the kidney to carry on its normal function was to replace it in its proper position in the lumbar region. This operation consisted in removing one of the kidneys of a dog and of grafting it anew in the lumbar egion. The other kidney was removed during the same operation or fifteen days later. The technique of this operation, when performed on medium-sized dogs, was comparatively simple. The abdominal cavity was opened by means of a transverse incision, then the kidney with its peritoneum was extirpated, together with its cellulo-adipose atmosphere and its blood-vessels. By means of a cannula introduced into the renal aorta the organ was then perfused by Ringer’s solution, whereupon it became whitish-yellow in color and a pale liquid escaped from the renal vein. Then the kidney was enveloped in a silk veil saturated with Vaseline and placed upon a table, while the central ends of the vessels were prepared for the anastomosis. After this the kidney was replaced in the abdominal cavity in its normal position and the arterial and venous ends were immediately sutured circularly. No suture of the nerves was made. After this, the serrefines, which served to secure the temporary haemostasis, were removed and the circulation was reestablished, after having been interrupted for 50 or 60 minutes more or less. The artery at once began to beat, the kidney became filled with blood, first blue then red, after which it resumed its normal size. After half an hour or an hour its colour had become the same as that of the other kidney. The phenomena of vasodilatation, which resulted after the transplantation of limbs usually failed to appear in he case of transplanted kidneys. The ureter was sutured circularly, after which the kidney was fixed in its normal position by means of a suture of the peritoneum. These anastomoses were performed with very fine thread, the result of which was that the traces of the operation had almost completely disappeared in a few months’ time. In some cases the other kidney was removed immediately, in other cases fifteen days afterwards; in others again, it was left intact, so that after several months a comparative study could be made of the anatomy of the replanted and of the normal kidney. A dozen of these operations were performed, and in every case the animals recovered. After the animals had undergone a double nephrectomy, their general health remained normal until some ureteral complication arose. In the first experiments the anastomosis of the ureter produced stenosis and the animals died after a few weeks or months, showing lesions of hydronephrosis. The technique of the ureteral anastomosis was then modified and all the animals experimented on since have retained their health. The urine was normal and contained no albumen. The kidneys were examined after several months and their histological condition was always found to be normal. Ten months after the replantation of a kidney both the normal and the replanted organs were examined. They looked identical, and in the histological examination it was impossible to distinguish the normal kidney from the replanted one. The caliber of the renal artery and vein had not changed at the point of the anastomoses, which were hardly discernible. These results may safely be considered permanent. A female dog, upon which a double nephrectomy and the replantation of one kidney had been performed, continued to live in perfect health for two years and a half, at the end of which time she died of a disease unconnected with the operation. The replanted kidney was normal. These experiments thus served to show that the extirpation of a kidney, its perfusion in Locke’s solution, the complete interruption of the circulation for a period of 50 minutes and the subsequent suturing of its vessels and of its ureter did not occasion any interference with its functions, even after a considerable interval of time. It was thus shown that from a purely surgical point of view the grafting of an organ is possible.

In other series of experiments I examined the influence exercised by a transplanted kidney upon its host. The kidneys of dogs and cats were removed and replaced in the abdomen, not of the same animal, but of another animal of the same species. I used the same technique as that just described. One of the kidneys of a dog was removed and transplanted on to another dog. Seven experiments were conducted. During the first week more or less the animals remained in good health and the urine contained no albumen. During the second week a little albumen was found in the urine. Towards the twentieth day it was noticed that the kidney had slightly increased in volume. Then the quantity of albumen augmented more and more and in certain cases some haematuria even was observed. The animals, notwithstanding, remained in good health, for they had retained one normal kidney alongside the transplanted kidney.. After six months all the animals operated upon showed albumen in their urine. The quantity of albumen progressively diminished and the urine became normal. The anatomical examination performed a year or two after the operation showed that the transplanted kidneys had become completely atrophied.

The interactions of the host and of its new organs was further studied by means of the transplantation “en masse” of kidneys, these experiments being conducted on cats. The transplantation in mass of the kidneys consisted of extirpating from a first animal both kidneys, their vessels and the corresponding segments of the aorta and vena cava, their nerves and nervous ganglia, their ureters and the corresponding part of the bladder; of placing this anatomic specimen in the abdominal cavity of a second animal whose normal kidneys had been previously resected and the aorta and vena cava cut transversely; and also of suturing the vascular segments between the ends of the aorta and vena cava, and of grafting the flap of the bladder on to the bladder of the host. In every case the reestablishment of the renal functions was observed. The secretion of urine began as soon as the arterial circulation was reestablished. In several cases clear urine flowed from the ureters while the flap of the bladder was being grafted on to the host. Oftener no urine was seen flowing from the ureters immediately after completion of the operation, but the secretion always began during the first twenty-four hours. In all the experiments the urinary secretion went on as long as the animal lived. Every cat urinated abundantly every day, but there arose in the animals sooner or later some complication, which modified in some measure the renal functions. The color of the urine was generally yellow, and often less dark than the normal urine of the cat. Its reaction was acid. Its quantity for the twenty-four hours varied from 120 to 160 cc, but, exceptionally, was as little as 25 and even 15 cc, though in another case, as much as 215 or 255 cc for the twenty-four hours. In this last case there was congestion of the kidneys produced by venous compression. The density was very far from constant; generally it varied between 1.018 and 1.030, going sometimes as high as 1.035 and 1.050. Among the abnormal constituents of the urine, albumen only has been looked for. In some cases there was a little albumen during the first days, ranging from 0.50 to 0.25 per 1,000 cc. In other cases the albumen disappeared about one week after the operation.

The general condition of the animal can, in some measure, be said to indicate the perfection of the urinary excretion. In one case, the animal was in apparently normal condition four days after the operation. She walked about the room, played, and ate a great deal of raw meat, and her condition remained excellent for several weeks. Twenty days after the operation she was in good health, had glossy hair, was very fat, ate all kinds of food with a good appetite, and urinated normally. There was, however, albumen in the urine, and slow and progressive enlargement of the kidneys took place, which showed that she was not in an entirely normal condition. The animal remained in excellent health until the twenty-ninth day after the operation. Then gastro-intestinal symptoms appeared and death occurred on the thirty-first day after the operation. Some of the animals were able to live in an excellent condition of health for 15 or 25 days and more after a double nephrectomy had been performed and they excreted every 24 hours from 120 to 160 cc of urine through their new kidneys. Nevertheless, all the animals died less than 40 days after the operation. The histological examination of the transplanted kidneys showed that the organs presented some lesions, very slight in some cases and more marked in others. The lesions were of diffuse nephritis. It is very probable that in the transplantation of the kidneys, as in the transplantation of the limbs, a reaction of the organism against the new organ takes place after a few days. The new organ may have a marked influence on the general condition of its host. In one experiment, the animal was a female cat which lived in the laboratory for several months. She was in excellent condition when she was operated on and recovered very quickly from the operation. Her life went on just the same as before. The kidneys were movable and small. She looked in excellent health and lived like any ordinary cat. On the eighteenth day after the transplantation albumen appeared in the urine and a direct examination of the kidneys was made to ascertain the cause. The general condition was little affected by the operation and the albumen disappeared almost completely. On the thirty-fifth day the animal was very weak and emaciated. She died on the thirty-sixth day. The autopsy showed that the whole arterial system was calcified. By exercising slight pressure the carotid artery or the aorta could be broken like a piece of glass tubing; but the renal arteries and the transplanted aortic segment were normal. The histological examination of the kidneys showed that they had undergone very slight morphological changes and that they were almost normal. This observation demonstrated that a change of organs may produce very marked modifications in the organism. The interactions of the organ and of its host have not been sufficiently studied thus far. In the course of these experiments the technique of transplantation was developed and their surgical study is now completed. Hereafter it will be necessary to investigate the reciprocal influences of an organism and of its new organ, and to ascertain under what conditions an organ taken from one individual can adapt itself to another individual.

It is for the above-described experiments that the Caroline Institute has honored me by conferring upon me the Nobel Prize. Many of these experiments were performed more than six years ago and have been applied in human surgery. The technique by means of which the blood-vessels are anastomosed or transplanted could be used with complete success and absolute safety on human beings. It has been shown to be possible, in cases of surgical and accidental wounds, to suture the humeral and femoral arteries, the common carotid, and even the abdominal aorta and other vessels. In cases of aneurisms several surgeons have been able to reestablish the continuity of the artery by the transplantation of a segment of vein. The principles found out in the course of the experiments will make possible all kinds of operations on the blood-vessels of human beings. In his admirable method for the transfusion of the blood Crile first used suture for the anastomosis of the blood-vessels. Although the suture is difficult on very small vessels, it has, nevertheless, been used with success in the transfusion of the blood in infants. The study of the circulation of the blood through metallic tubes has led to a simpler technique, which has been used by Dr. Tuffier and which will increase the efficiency of Crile’s method. It is probable that many other practical applications will be found, since the experimental studies have definitely demonstrated that operations on blood-vessels, and transplantation of the same, can be made without danger.

It has further been demonstrated that autoplastic transplantation of organs like the kidney, for example, is always successful; but this fact is of itself of little practical interest. However, a limb which has been accidentally severed from the body, as for instance happens with a circular saw, etc. might easily be replanted on its owner. But it is not yet known whether surgeons will ever be able to perform a homoplastic transplantation with permanent success. Before this is accomplished it will be necessary to discover a means of recognizing the individuals, if such exist, between whom organs can be interchanged with impunity. Thus, while the problem of the transplantation of organs has been solved from a surgical point of view, we see that this by no means suffices to render such operations of definite surgical practicability, and it will only be through a more fundamental study of the biological relationships existing between living tissues that the problems involved will come to be solved and thereby render possible the benefits to humanity which we hope to see accomplished in the future.

From Nobel Lectures, Physiology or Medicine 1901-1921, Elsevier Publishing Company, Amsterdam, 1967

The Nobel Foundation's copyright has expired.

To cite this section
MLA style: Alexis Carrel – Nobel Lecture. Nobel Prize Outreach AB 2024. Tue. 16 Jul 2024. <>

Back to top Back To Top Takes users back to the top of the page

Nobel Prizes and laureates

Eleven laureates were awarded a Nobel Prize in 2023, for achievements that have conferred the greatest benefit to humankind. Their work and discoveries range from effective mRNA vaccines and attosecond physics to fighting against the oppression of women.

See them all presented here.

Explore prizes and laureates

Look for popular awards and laureates in different fields, and discover the history of the Nobel Prize.