May 28, 1894, closed by catgut suture an opening in the axillary artery, one centimenter long, which was accidentally made while removing adherent carcinomatous glands. The patient made a good recovery and circulation in the extremity was not disturbed. In the same year (1894) Robert Abbe made some very interesting and instructive experiments, placing sterilized glass tubes in vessels with the intention of uniting and re-establishing the circulation. (N. Y. Med. Jour., Jan. 13, 1894.) Von Zoege-Manteuffel in April, 1895, operated on an arterio-venous aneurism in Scarpa's triangle. He wounded the common femoral and made a successful lateral suture of the wall. (Berl. klin Woch. No. 34, 1895.) J. Israel, in discussing Gluck's paper, mentioned that he had closed the common iliac which had been torn for two-thirds of its circumference in an operation for a perityphlitic abscess, by inserting five fine silk sutures through the entire wall of the vessel. (Berlin Cent. für Chir., Dec. 7, 1895.) It does not seem possible from my observations experimentally that two-thirds of the circumference of a vessel the size of the common iliac could be closed with five sutures and have the seam blood-proof. Heidenhain endeavored to accomplish the closure of arterial wounds in the carotid of dogs by suture, but failed. Sabanyeff of Odessa successfully closed a small opening in the femoral artery with suture in 1896. Billroth, Schede, Braun of Koenigsberg, Schmidt and others have successfully sutured wounds in veins, and the closure of wounds in veins by suture is now an accepted surgical procedure. I recently closed an opening three-eighths of an inch in length in the common jugular vein, with five continuous silk stitches. Small openings in veins have frequently been closed with lateral ligature. I cannot find recorded a case in which an artery has been sutured after complete division.

For the purpose of determining whether this were practicable, and whether a considerable portion of the artery could be resected, I have performed the following experiments:

Experiment No. 1, March 4, 1896.-Male dog, weight 40 pounds. An oblique incision dividing one-third of the circumference of the left common carotid was made and closed with a continuous silk suture, penetrating all the coats of the artery; only a small portion of artery was denuded. There was some hemorrhage from needle punctures after the clamps were removed, which subsided rapidly under digital pressure: the wound was closed. The dog showed no unpleasant symptoms after the operation. There was a slight swelling on the left side which lasted about a week.

Post-mortem, March 25. 21 days after operation: There was some newly formed connective tissue at the seat of operation. The sutures were not infected; the caliber of the vessel was only slightly diminished. The silk

was covered and could be seen under a thin layer of exudate from the inner side of the vessel. There was no thrombosis. The intima had been fractured by one of the clamps, at which point it was rough and somewhat thickened.

Experiment No. 2, March 4, 1896.-Dog, weight 50 pounds. The right common carotid was exposed and opened by a longitudinal incision one-third of an inch long. The opening was closed by interrupted sutures involving all of the coats. There was some hemorrhage after the clamps were removed, which ceased after a digital pressure of two minutes' duration. The external wound was closed with silkworm gut.

A post-morten was made March 25, 21 days after the operation. There was considerable exudate around the position of the sutures, but

Fig. I (A). (B.) Fig. I (A) shows method of inserting sutures produce invagination of artery.

to

Fig. I (B) shows method of inserting external su tures.

the sutures were not infected. The wall of the vessel was thickened on the side of the suture. The lumen of the vessel was diminished one-third; the suture was not exposed in the vessel, but it was covered with a deposit. There was no thrombus. There was some roughening of the intima at the position where the clamps had been placed.

Experiment No. 3, March 7, 1896.-Hunting dog, weight about 50 pounds. The left femoral artery was exposed and an incision one-half an inch long was made in its wall: it was closed with a continuous silk suture; vessel one-eighth inch in diameter. Vessel was clamped with eight-inch Billroth forceps covered with rubber, care being taken not to compress the artery sufficiently to injure the intima. There was no difficulty in producing an accurate approximation of the margins of the wound with silk suture. A very slight hemorrhage took place through the needle holes after the clamps were removed, which ceased in less than a minute. The external wound was closed with silkworm gut suture.

Examination of left femoral March 15, eight days after operation: Union, no hemorrhage. no suppuration, the vessel was completely thrombosed.

Experiment No. 4, same day, same dog.— Invagination of right femoral after complete division. Three double needled silk sutures were used to draw the intussusceptum onethird inch into the intussuscipiens as shown in Fig. I A. Several interrupted sutures bind the end of the intussuscipiens to the circumference of the intussusceptum as shown in Fig. I B.

Examination March 15, eight days after operation: There had been no hemorrhage from the vessel. A slight infection had taken place around the seat of operation. The artery was thrombosed.

Microscopic examination showed the lumen filled with a red thrombus and the wall to consist of two distinct muscular layers, both diffusely infiltrated with round cells. The lumen The lumen was very small and the wall very thick.

Experiment No. 5, March 11, 1896.--Dog, weight 60 pounds, large head and neck. Resection of three-fourths of an inch of right common carotid. The proximal end was invaginated with three silk sutures; several sutures penetrating the adventitia and media. were used to secure the end of the intussuscipiens to the periphery of the intussusceptum. The sheath was sewed over the vessel. The adventitia was separated from the media on either end. The Billroth clamp forceps were used to suppress the hemorrhage.

Post-mortem April 15, thirty-five days after operation. Some periarteritis and adhesions existed. The sutures in the skin had suppurated. The area around the vessel was not infected; there was but little contraction at the position of approximation.

Microscopic examination: There was some thickening in the wall. The thickening in the media was more pronounced than in the intima; the process was not progressive. There was nothing to indicate that the artery would not have remained patulous. wall of the artery showed no evidence of injury by the clamps.

The

Experiment No. 6, March 11, 1896.--Left common carotid in same dog as in experiment No. 5. Resection of one-half inch of artery. End-to-end suture without invagination. The suture penetrated all the coats; it was continuous and contracted the artery considerably at the line of approximation. It was made. with two needles on one thread and a continuous figure-of-eight stitch. There was a little hemorrhage from the needle holes after the sutures were placed. This subsided under a sponge pressure of two minutes' duration. The sheath was not sutured around the vessel. Clamps were the same as used in experiment No. 5.

Post-mortem April 15, thirty-five days after operation: Wound suppurating down to artery. Pocket of pus existed deer in the neck. All of the arterial sutures were exposed in the bottom of the wound. The artery was thrombosed.

Microscopic examination: The artery presented just the opposite condition from its fellow on the left side described above. The intima was very much thickened and was rapidly obliterating the tube. The histology was that of organization of clot; connective tissue cells and blood vessels were taking the place

of coagulated blood. The media was likewise somewhat inflamed. There was some inflammation along the lymphatics and adventitia and there were numerous foci of infection. The case was a typical infective endarteritis with formation of coagulum.

Experiment No. 7, March 15, 1896.-Dog, weight 40 pounds. Longitudinal incision in right iliac five-eighths of an inch in length, was closed with a continuous silk suture, including all of the coats of the artery. Five days after the operation a very marked swelling appeared at the position of suture and blood began to ooze from the line of union. The dog was again anæsthetized, cutaneous sutures were removed, margin of wound separated. A decomposed clot about the size of a hen's egg was carefully elevated. There was an absence of union of the wall of the vessel, and when the clot was removed profuse hemorrhage ensued. A ligature was placed above and below the incision in the artery,

The

and that portion of the vessel excised. cause of the non-union of the vessel in this case, I believe, was infection. There was no clot in the artery.

Experiment No. 8, March 18, 1896.-Dog, weight 30 pounds. Longitudinal incision onehalf inch long was made in the abdominal aorta, which was closed with a continuous silk suture, including all the arterial coats. A slight hemorrhage occurred through needle punctures after clamps were removed, which ceased on one and one-half minutes' digital pressure. No effort was made to cover the line of suture with adventitia or peritoneum.

The dog died March 28. Post-mortem: Retro-peritoneal abscess around artery. Complete occlusion of lumen of artery. General suppurative peritonitis was the cause of death.

Experiment No. 9. March 26, 1896.-Dog, weight 60 pounds. Invagination of left carotid by three double-needled threads attached to the intussusceptum, and interrupted silk suture of the intussuscipiens, involving only the tunica adventitia and media of the proximal side.

Post-mortem April 25, 1896: Left common carotid surrounded by considerable exudate and suppuration around one of the cutaneous stitches. A small sinus led down to one of the sutures at the position of approximation. Vessel wall considerably thickened at the line of union, as shown in Fig. III. The vessel admitted a considerable stream of water through it immediately after its removal. It was somewhat dilated on the proximal side. The collateral branches were enlarged.

Microscopic examination: Most evidence of endarteritis at 3 and 4, Fig. III. Much of the tissue between 2 and 5 seemed accidental. The invaginated walls could be outlined, but the line of demarcation between them was scarcely discernible. The connective tissue around the artery showed a large number of blood vessels and a cross-section of a nerve of considerable size. There was a small amount of inflammation of the vascular and lymph channels at that point. In

Fig. IV. Shows thickening of artery from end arteritis, from Experiment No. 10.

the adventitia there was a violent acute inflammatory process with foci which resembled those of suppuration. The showing in the media was the same as in the right carotid. In the intima the process was still more violent.

The inner elastic lamina was so broken up as to be unrecognizable.

Experiment No. 10.-Same dog as experiment No. 9, same day. Resection of one-half inch of right carotid. Adventitia peeled back. Three double-needled sutures were attached to the intussusceptum and put through the intussuscipiens from within outward, invaginating intussusceptum one-third inch; then sutured the intussuscipiens to the periphery of the intussusceptum, as shown in Fig. I B. The sheath of the artery was sewed around it: the adventitia was not sutured.

Post-mortem on right carotid, April 25, showed no suppuration around the artery; considerable inflammatory exudate; artery enlarged at position of invagination as shown in Fig. 4. Nerve and vein involved in exudate. The vessel admitted a stream of water about one-half its normal diameter. It was

enlarged on the proximal side of the union. Microscopic examination showed marked evidence of endarteritis. Line just below 3, Fig. IV, indicates a suture. At 2 much exudate but less than at 3. At 4 the lumen is about one-half as large as it is at I and 5. Lumen at 3 is one-half of what it is at 5. Most of the tissue between 2 and 5 is foreign. There is a moderate amount of inflammation in the adventitia and a large number of new blood vessels. Muscle fibers of the media alternate with bands of small round cells. Capillaries in outer walls are apparent. The region of the intima is the seat of a very marked round celled infiltration; capillaries can be seen in this coat. The thickness of the intima is markedly increased and the lumen is about one-third of the normal size. There are no endothelia lining the intima. Another section of the same specimen presents the same appearance, except what remains of the lumen is a triangular opening. There are sections of nerves in the loose tissue surrounding the vessel.

It will be noticed that notwithstanding the invagination had been made thirty days previous to the post-mortem, there was no clot in the vessel, and the diameter of the lumen was one-third of that above and below the point of union, showing that the endarteritis had progressed slowly.

Experiment No. 11, April 26, 1896.-Calf six weeks old. Excision of one-third inch of left common carotid. Invagination of all coats of proximal into distal end for threeeighths of an inch. The artery was clamped with Billroth forceps. It was found to be elastic and free from tension when replaced. Peripheral sutures were used to secure the invaginated portion as described above. There was little hemorrhage after the operation was completed. The sheath was sewed around the artery. There was a tumor around the area of approximation, with considerable swelling, for three weeks after the operation, which disappeared almost entirely before the specimen was removed for examination.

Specimen secured June 1, 34 days after operation; one and one-half inches of carotid at point of suture was removed and ends ligated. There was considerable inflammatory deposit over the entire field of operation. The union of skin was complete. The collateral vessels around the position of approximation were considerably enlarged. The vessel was not patulous, the lumen being filled by a clot for a considerable space on the distal side of approximation, it extended but one-fourth of an inch on the proximal side.

Microscopic examination showed on the outside a loose layer of connective tissue, the fibers of which did not run in any particular direction. Internal to this was a layer of felted connective tissue interlacing and a very

rich supply of blood vessels, richer than that which is normal in the adventitia. Most of these fibers were of the yellow elastic variety and interlaced in the general trend of the artery. Internal to the adventitia was the tunica media, consisting of muscular fibers, elastic tissue, and a great abundance of inflammatory cells. Internal to this was the inner elastic lamina. Within this there was a zone, which at some of the points of its circumference was composed of mature connective tissue fibers; at others it consisted of a pool of blood. Internal to this was a dense mass of muscle fibers. The fibers of the media were swollen and granular and not in a good state of nutrition. Internal to this was the inner elastic lamina of the invaginated portion of the artery, and the lumen of the tube was filled with a clot. There was some evidence of connective tissue growth in this clot. The union between the artery ends was incomplete.

Experiment No. 12.-Right common carotid of calf, same as used in experiment II operated May 3, 1896. Resection of one-half inch of artery united with three invaginating sutures. Three peripheral sutures were made as above; but very little bleeding followed the operation. Sheath was drawn over artery. Clamps used as temporary hæmostatics, as above. The skin was approximated with catgut. There was no swelling nor tumefaction after operation, as on opposite side.

June 1, 1896, one inch of artery at point of union was removed and artery was again sutured (see experiment 13). But little inflammatory deposit was found around the seat of wound. Primary union of skin and subcutaneous tissue had taken place. Some reactive inflammation had taken place around the artery. Water passed freely through the artery when removed.

Microscopic examination: The section showed a mass of granulation older at the periphery than at the center. The inflammation was somewhat nodular; there were some places where it was more intense than at others. In the center there was a small opening in the wall of which there were young embryonal cells, the artery was being obliterated by a general arteritis.

Experiment No. 13, June 1, 1896.-The second experiment on right common carotid of calf. In first experiment of May 3, one-half inch of artery was removed. One June I the specimen was removed, taking out one inch at seat of union. The artery was then sutured end-to-end, with interrupted silk suture without invagination. There was some tension on the sutures and the artery was somewhat diminished in size. The left carotid had previously been removed and ends ligated. When clamps were taken off there was no hemor

rhage at the line of suture. The calf rallied after the operation and was perfectly well.

Post-mortem, June 5, four days after operation: Primary adhesion of wound; the artery was carefully dissected out; there had been no hemorrhage around the line of union. The vessel was patulous; its lumen was slightly diminished. A small white exudate covered the line of union. The sutures were not exposed in the lumen of the artery, being covered by the exudate. The intima had not been fractured at the point where the clamps were placed.

Experiment No. 14, Aug. 26, 1896.Brown dog, weight 40 pounds. Median abdominal incision. The abdominal aorta was located just above its iliac bifurcation and exposed for two inches. Billroth clamps were placed in position. A transverse incision was made in the artery for two-fifths of its circumference. It was sutured with kangaroo tendon, involving all the coats. There was

Fig. V. Shows external and internal clot from Experiment No. 14.

some hemorrhage after removal of the clamps. An addition superficial suture was inserted. The retro-abdominal peritoneum was closed over the artery. Abdominal wound was closed with silkworm gut sutures. The dog died three days after operation.

Post-mortem: Abdominal cavity was filled with blood; no peritonitis. An opening was found in the posterior abdominal peritoneum leading to the vessel. The vessel was removed. A clot of considerable size existed in the retro-peritoneal space around the opening of the vessel; this was continuous with a small clot within the vessel shown in gelatine specimen No. 2, Fig. V. The clot on the inner side of the vessel measured three-eighths of an inch in length and one-eighth of an inch in diameter. It was not adherent to the wall of the vessel, but was continuous with the clot from the outside. There was some discoloration of the intima from the upper clamp and a roughening of the surface at that point, but no clot was formed. The lumen of the artery above and below the artery was free.

The kangaroo tendon was softened and would not support its own weight, which accounted for the hemorrhage. On the inner surface of the artery was seen the openings of the small arteries which were free. There was a ragged transverse opening which was gaping widely at the center. The point of adhesion of the clot was without the arterial wall. The remnants of the sutures were seen penetrating the intima. One and one-half centimeters above there was a point of ecchymosis. One centimeter above that a second point of ecchymosis could be seen, the result of the clamps. The ecchymotic patches were transverse in their long diameter and three-fifths centimeter wide. In the one next to the incision there had been rupture of the deep arterial coats, but the intima had been preserved except at one small point; in the other, fracture of the intima and media had taken place. There was no clot at either of these places.

Experiment No. 15.-Black dog, weight 40

r Fig. VI. Shows edge-to-edge union of abdominal aorta showing internal and external clot, from Experiment No. 17.

pounds. Operated, August 26, 1896. Median abdominal incision. Exposure of abdominal aorta two inches above its iliac bifurcation. A longitudinal incision was made in artery one centimeter in length. Sutured with kangaroo tendon taken from same bottle as for experiment 14. Arterial clamps were placed on artery as above. Slight hemorrhage from the wound after the clamps were removed. This was controlled by digital pressure for one minute. Posterior peritoneum was not sutured over artery. A great deal of difficulty was experienced in retaining artery in an elevated position sufficient to admit of suture, as the dog was fat and his abdomen deep. Dog died fortyeight hours after operation.

Post-morten: Abdomen full of blood. Loose large clot existed in retro-peritoneal space. The artery was removed for several inches above and below the clot; its lumen was free. There was no clot on the inner side. The kangaroo tendon was completely disintegrated. Cause of death, hemorrhage.

The general appearance of the artery was normal. Small arteries were seen emerging at

different points. The line of incision could be readily recognized. Its edges were roughened and projected about one millimeter into the lumen of the vessel. The ragged incision extended throughout the wall of the artery; in it were seen the remains of the tendon sutur. There was no evidence of regeneration, that is, no reaction process in the walls of the vessel. There was some fibrinous deposit on the roughened edges; no tendency to coagulation on the inner side of the artery. About one centimeter below the line of incision was a small ecchymosis where the intima was roughened and endothelia were absent; there was also a loss of substance showing a fracture of the intima. There was a small fibrous deposit on this area.

Experiment No. 16, September 6, 1896.Black male dog, weight 70 pounds. Transverse incision dividing one-half of the anterior wall of the abdominal aorta; the openng was closed by a contnuous suture of silk with round needle, involving all the coats of the artery. After the clamps were removed there was bleeding from the angle of the wound; an additional suture was applied without the aid of the clamps. After this was done the bleeding ceased and abdominal wound was closed. Dog died fifty-four hours after operation.

Post-mortem revealed the abdomen full of blood. An examination of the artery showed three-fourths of the line of suture approximated and surrounded by a clot. The last suture which was inserted in the angle of the wound was shown to have included on one side only the tunica adventitia, it did not grasp the median or the intima. This suture gave way and from this point hemorrhage took place. There was no regenerative reaction in the intima around the punctures made by the needle. There was no clot formed on the silk that was exposed in the lumen of the vessel. From the ragged edge at the position of penetration a small clot was adherent and suspended within the artery, but not adherent to the intima: it was continuous with the external clot. This clot was about a centimeter in length and a millimeter in thickness; smooth on its outer edge. The endothelium in the immediate neighborhood was milky white and somewhat roughened.

Experiment No. 17, September 6, 1896.Black and white female dog, weight 45 pounds. A complete transverse section of abdominal aorta. Continuous silk suture with three needles: edge-to-edge union of artery, Fig. VI. The stitches were closely applied. After the clamps were removed free hemorrhage took place, particularly at one point, where the beginning and end of the sutures met. The clamps were reapplied and one stitch placed at this point. It would have been better if I had made the final stitch overlap the primary stitch. as the artery was contracted when the arterial