A History of Vascular Surgery

Prologue

Although some might argue that Guy de Chauliac or Ambroise Paré should properly be called the sires of surgery, John Hunter is the prototype of the modern vascular surgeon. He was an unbelievably productive and tireless worker, cut from the same Scottish mold as his brother William, who was 10 years older. John was largely unlettered, whereas William had become sophisticated through his education at Glasgow, yet they shared a frenetic capacity for work and an incurable curiosity.

To place the Hunters in a clear perspective in regard to nonmedical history, one should note that they were contemporaries of George Washington and Benjamin Franklin. William Hunter was born in Scotland in 1718, his brother John was born 10 years later; William died in 1783, and John died in 1793. John was even made a member of the American Philosophical Society, although he never attended a meeting.

William Hunter preceded John to London, where he soon established a busy medical practice and interested himself in many subjects, including aneurysms. In fact, William proposed the concept that a lancet used carelessly during bloodletting might enter both artery and vein, and after healing, the two channels might be connected. He thus imagined an arteriovenous fistula. He soon found just such a patient and described the clinical manifestations with great accuracy. William's primary activity, however, was focused on obstetrics and on the teaching of anatomy. John became his assistant in this latter project.

John Hunter is remembered for many things, but especially for his studies of the dynamics and efficiency of collateral arterial circulation, which he described in the vessels feeding the antlers of a stag after he had interrupted the major arteries in its neck. More renown came from his ligation of the femoral artery in its subsartorial course at a distance above a popliteal aneurysm—in Hunter's canal.

To be sure, others had preceded him in performing proximal ligation of arteries to treat aneurysms. In the third century, a Roman surgeon named Antyllus had described proximal and distal ligation of the artery, followed by incision of the aneurysm and removal of its contents—a formidable operation without either anesthesia or asepsis. In 1680, Purmann, faced with a large aneurysm in the antecubital space, performed ligation of the vessels and excision of the aneurysmal mass. In 1714, Anel described an operation in which he placed one ligature on the artery at the proximal extent of the aneurysm. Hunter, however, had found that the ligature would sometimes cut through the artery when it was placed too close to the popliteal aneurysm; therefore he chose a site that was more remote, but was easily reached by the surgeon and would preserve collaterals. Most of Anel's patients suffered from false aneurysms caused by bloodletting in otherwise healthy arteries. The femoropopliteal aneurysms treated by Hunter were due to degenerative processes, probably a mixture of syphilis and trauma.

Many other surgeons were ligating aneurysms in various anatomic sites at this time. Cooper, one of John Hunter's students, was soon established as one of the early vascular surgeons when he ligated the carotid artery for an aneurysm in 1805, as well as the aorta for an iliac artery aneurysm. Only these few important events occurred before the latter part of the nineteenth century.

At the time, ligation was virtually the only procedure available to surgeons for the management of arterial problems, and those problems were limited to the control of hemorrhage and the treatment of aneurysms. Hallowell in Newcastle-on-Tyne performed one arterial repair of an artery torn during bloodletting. The laceration was a short one, and at the suggestion of Lambert, he placed a short ( inch) steel pin through the edges of the wound and looped a ligature around it in a figure-of-eight pattern, approximating the edges of the wound with apparent success. Hallowell wrote to William Hunter concerning this operation in 1761, foreseeing that if this were a successful technique, “we might be able to cure wounds of some arteries that would otherwise require amputation, or be altogether incurable.” That Hallowell wrote to William instead of John is probably due to William's published work on arteriovenous fistulas secondary to inept bloodletting. Twelve years later in 1773, Asman reviewed the Newcastle repair, attempted some experiments of his own that were disastrous, and concluded that such a procedure could not work and that Lambert and Hallowell's efforts had probably failed as well. After Asman's criticism, the matter of arterial repair rested quietly for nearly another 100 years.

John Hunter's less widely known contributions are scattered throughout the immense museum he left to the Royal College of Surgeons of England, and they hint at an understanding of arterial pathology that would not be general knowledge for half a century. They include dissections of several atherosclerotic aortic bifurcations (specimens P.1177 and P.1178), showing the atheromatous lesion at the aortic bifurcation that Leriche would describe 150 years later; a carotid bifurcation with an ulcerated atheroma from a patient who died of a ruptured syphilitic thoracic aneurysm (specimen P.1171); and an extracranial internal carotid aneurysm (specimen P.282) in a patient whose neatly described symptoms are almost typical of what today are recognized as classic transient ischemic episodes. Regrettably, most of Hunter's notes did not survive to provide more than this fragmentary view of his understanding of vascular disease. To cap it all, in a postmortem specimen, Hunter had dissected the atheromatous layers (although the term atheroma had not yet come into use) from the remaining intact wall of an atherosclerotic terminal aorta (specimen P.1176), foreshadowing dos Santos by 150 years.

Both Hunter and Cooper seemed to hold with the teleologic belief of the times that when senile or spontaneous gangrene occurred in older persons, thrombosis of the major vessels supervened so that the patient would not bleed to death when the gangrenous part separated. It was Cruveilhier who first clearly stated that the phrase “gangrene due to obstruction of the arteries” by thickening and by thrombosis should replace the terms spontaneous and senile gangrene, but he attributed the concept to Dupuytren.

The recognition that arterial obstruction causes functional disability that limits the use of the affected part may have arisen in the veterinary world. Bouley described the clinical picture in a horse in 1831.

Four years later in 1835, a nearly anonymous physician on the ward of a Professor Louis provided the first clear description of human claudication. Barth's patient was a 51-year-old woman who died of heart failure resulting from mitral valvular disease. His report described her incidental history of claudication in terms that we would recognize today. In the postmortem report, he noted thrombosis of the terminal aorta and included a sketch suggesting that the lesion was a thrombosed hypoplastic terminal aorta, a contracted atherosclerotic lesion, or a combination of both. Barth also repeated Hunter's observation that the obstructing material could be separated easily from the residual intact arterial wall. Barth was never identified further, not even by an initial.

Charcot is often erroneously given credit for recognizing the syndrome of intermittent claudication caused by arterial insufficiency in humans. Charcot described, just as Bouley had done, the vanishing pulses, the cold extremity, and what is now recognized as the loss of sympathetic tone in a horse in the throes of a spasm of severe claudication; he reported a human case as well. Homans liked to joke that Charcot observed the former because he spent so much time at the horse races.

As a neurologist, Charcot was familiar with intermittent claudication in humans caused by various neurologic processes. The patient Charcot described, however, suffered claudication in one leg secondary to an old gunshot wound that resulted in occlusion of the iliac artery and an aneurysm proximal to the occlusion. The aneurysm, which was adherent to and in communication with the jejunum, gave rise to a series of small gastrointestinal hemorrhages before the final fatal episode. Charcot thus deserves credit for identifying the herald hemorrhages that often presage major bleeding from an aortoenteric fistula. (Charcot credited both Bouley and Barth with their prior observations regarding claudication.)

Successful Arterial Suture

Such information was of little utility to surgeons, however, until arterial repair became a reality. Consistent with the observations of Asman, several German masters had deemed arterial repair (as opposed to ligation) to be impossible. Langenbeck stated in 1825 that, because the primary requirement for healing is perfect rest, an arterial incision could never heal as long as the pulsatile movements of the arterial wall continued. Heinecke was certain that the patient would bleed to death through the suture holes and the apposed edges of the arterial wall.

Repair of small injuries to veins, however, was becoming an established procedure. The lateral ligature, in which a clamp is placed on the defect in the venous wall and a ligature is tied around the puckered wall, had been performed in 1816. The first lateral suture of a venous defect (an erosion of the common jugular vein from an infected neck wound) was undertaken by Czerny in 1881, but the patient died of sepsis and hemorrhage. Jassinowsky credits Schede with the first successful repair of a large venous injury (to the common femoral vein) by lateral sutures.

Going beyond the stage of venous repair, Eck reported the experimental creation of a portocaval fistula in dogs. The original description hints that he had little to confirm his success. Among a series of eight dogs, one died within 24 hours, six lived 2 to 6 days, and the one survivor “tired of life in the laboratory and ran away after 2 months.” The doctoral dissertation of Jassinowsky, written in 1889 and based purely on library research, reviewed the published information on arterial suture and concluded that it could not be successful at that time, but that there might be hope in the future.

Only 2 years later, however, Jassinowsky himself succeeded. In 1891, he reported his successful animal experiments involving arterial suture. The suture he described was passed carefully only two-thirds of the way through the media; he tried to avoid penetrating the intima, except in very thin-walled vessels. This effort should be recognized for its intrinsic difficulty using even the finest milliner's needles, because without sutures swaged onto needles, two pieces of suture have to be dragged through the arterial wall. Dörfler modified Jassinowsky's method and passed the suture through all thicknesses of the arterial wall. He also recognized that the arterial suture exposed in the lumen of the vessel did no harm if uninfected. He observed that it soon became covered with a glistening membrane. Shortly thereafter in 1896, Jaboulay and Briau described successful end-to-end carotid arterial anastomoses in animals using an everting U-shaped suture.

Jaboulay was one of the surgeons in Lyon, France, under whom Carrel studied. When Sadi Carnot, the president of the Republic of France, was wounded by an assassin and died because no one dared to try to repair his portal vein, Carrel was highly critical, because he believed that blood vessels could be sutured as well as any other tissue. He soon undertook experimental arterial anastomoses; some of the earliest of these were arteriovenous communications in which the high-flow system ensured patency. Carrel's contributions to technical arterial surgery included methods that vascular surgeons routinely use today. He devised the triangulation suture to facilitate end-to-end anastomosis, described the patch technique to anastomose a small vessel to the side of a larger one (as in replantation of an inferior mesenteric artery), and pioneered the use of vessel grafts and organ transplantation. His work, however, was not fully accepted in the United States for many years. In part, this stemmed from disputes that arose between him and Guthrie, who was his coworker for 1 year.

In contrast, European surgeons not only accepted Carrel's work but also began to follow his lead. In 1906, Goyanes of Madrid, Spain, resected a popliteal aneurysm, then restored arterial continuity with an in situ venous graft using the popliteal vein, which was probably the first successful clinical vascular replacement.

Surgeons in the United States were beginning to perform vascular surgery in their own way. In New Orleans in 1888, Matas described a landmark operation. He stumbled onto the surgical procedure for which he is commonly remembered, endoaneurysmorrhaphy, when an aneurysm for which he had ligated only the proximal brachial artery, with apparent initial success, began to pulsate again 10 days later. Reportedly, it was a medical student who called this to the professor's attention. He chose to reoperate and to ligate the brachial artery distally. Even after this distal ligation, the aneurysm continued to pulsate, and he was forced to open the aneurysm, clean out the sac (the operation performed by Antyllus), and oversew the other arteries feeding the aneurysm from inside the sac. This foreshadowed the problems with endoleaks that confound vascular surgeons who place endovascular aortic prostheses today.

Matas's operation differed from that of Antyllus, in that Matas used a suture within the aneurysmal sac to obliterate the feeding vessels instead of ligating them outside the sac. The extensive dissection that would have been required outside might have damaged the collateral circulation and other adherent anatomic structures. It was many years before Matas performed another endoaneurysmorrhaphy, because most patients were treated successfully by simple proximal ligation. Matas ultimately expanded the descriptions of his technique to include “restorative” and “reconstructive” modifications, and he reported an approach to the arteriovenous fistula through the venous component, as had been proposed by Bickham.

Murphy, of Chicago, performed a series of experiments on animals in which he successfully restored continuity by invagination of the proximal into the distal vessel. In 1897, he presented a successful human case. Edwards briefly revived this anastomotic technique of invagination when he recommended the use of the first braided nylon grafts.

Murphy's invagination techniques were reflected in other nonsuture methods of anastomosis: Nitze and Payr used small metal or ivory rings through which the vessel was drawn, everted, and tied in place; this unit was then inserted into the mouth of the distal vessel, and another ligature secured it there. This is substantially the Blakemore tube, used during World War II, albeit without signal success.

During his tenure at Johns Hopkins Hospital, W.S. Halsted had an abundance of traumatic and syphilitic aneurysms commanding his attention. In the early 1900s, Carrel visited Halsted and described his own technical experiments, including his early arteriovenous anastomoses. As a result, Halsted almost made history in 1907 when he faced the dilemma of a patient whose popliteal artery and vein had been sacrificed during an en bloc dissection of a sarcoma of the popliteal space. Halsted went to the other leg, took the saphenous vein, reversed it, and anastomosed the distal saphenous vein to the proximal femoral artery. For his distal anastomosis, however, he chose the popliteal vein. Although the graft pulsated for 40 minutes, it soon thrombosed. It is possible that Halsted was pursuing the chimera of reversal of arterial flow through the venous bed. One can only imagine what a dramatic leap forward vascular surgery would have made if Halsted, with his superb supporting cast of talented surgeons, had chosen the popliteal artery for the distal anastomosis and had achieved a truly successful arterial reconstruction in the pattern of the modern vascular surgeon.

There is considerable literature on attempts to revascularize ischemic extremities via arteriovenous anastomoses. San Martín and A. E. Halsted attempted to improve the distal circulation using arteriovenous anastomoses.

Meanwhile, German surgeons such as Höpfner, Lexer, and Jeger had become familiar with the use of short (<10 cm) vein grafts. Höpfner described the bypass procedure, which was illustrated in an encyclopedic book by Jeger. Jeger's book, republished posthumously in 1937, included a foreword that described Jeger's replantation of the completely severed arm of a German soldier, which he had performed in 1914. One year later, Jeger came to an untimely death from typhus while on the Russian front.

Lexer collected and reported on 65 vein transplants, 13 of which were his personal cases. In 8 of these 13 cases, Lexer had obtained a distal pulse. This report prompted a Polish surgeon, Weglowski, to present his own personal series of 51 vein grafts, mostly for trauma, operated on between 1914 and 1921; in 40 patients he could document good distal pulses and normal arterial tracings. Yet all this seemed to be forgotten for the next 25 years as Germany suffered the agonies of the interbellum years, and as the forceful and charismatic personality of Leriche appeared on the scene (Leriche's role is described in a later section).

Abdominal Aortic Aneurysms

Beyond the management of trauma to the arteries, the aneurysm is clearly one of the great surgical challenges. The previous section detailed early attempts to treat peripheral aneurysms, but these were sporadic and lacked a continuing series.

Vesalius is said to have been the first to describe an abdominal aneurysm. The successful management of the abdominal aneurysm is certainly one of vascular surgery's major accomplishments. The technical maneuvers described previously concerning the ligation of aneurysms in various anatomic sites usually involved aneurysms of the peripheral vessels; aneurysms of the trunk were sacrosanct, because proximal control was not feasible. Cooper had continued many of Hunter's studies, including evaluation of collateral arterial supplies. In 1805, he had ligated the common carotid artery for an aneurysm, but he opened the door for even wider surgical applications when, in 1818, he ligated the abdominal aorta to control external hemorrhage from an aneurysm of the external iliac artery that had eroded to the surface of the skin of the flank, bleeding openly at that site.

Interest in the treatment of major vessel aneurysms lagged for almost a century. Eventually Colt, at the end of the nineteenth century, used wire to pack an aneurysm and then heated the wire. Blakemore and King revived interest in this technique in 1938, and many surgeons undertook modifications of the wiring technique, largely without success. Meanwhile, more direct attempts were being made by the major actors in the next scene: Matas of New Orleans and Halsted of Baltimore. Their interest in the management of vessel trauma, and in the management of late sequelae of such trauma, provided material for the fertile imaginations of the many surgeons who were emboldened to follow in their footsteps. Reid reported the experience of the Johns Hopkins Hospital (headed by Halsted) with aneurysms in 1926. The aneurysms treated included many varieties, both anatomic and etiologic, but treatment of abdominal aneurysms was substantially a failure. These operations were only preparation for the end of ligation as a treatment for aneurysms of the abdominal aorta.

Matas finally accomplished a successful aortic ligation (just below the renal arteries) for an aneurysm at the bifurcation of the aorta. He reported it first in 1925 and then again in 1940. In the issue of Annals of Surgery that contained Matas's second report was a similar paper by Elkin, as well as a hint of the coming era of vascular reconstruction in a report by Bigger of Virginia. Bigger had ligated the neck of an abdominal aneurysm using fascia that he expected to loosen gradually and allow restoration of flow. With the protection of this temporary control, he performed a plication of the aneurysm, restoring the aorta to its proper caliber. The patient had a protracted survival without recurrence of the aneurysm and also with restoration of femoral pulses.

About this time, however, cardiac surgery began to emerge. During the first decade of the twentieth century, Jeger had proposed valved venous grafts between the left pulmonary veins and the left ventricle to bypass mitral stenosis, and a valved venous graft from the left ventricle to the innominate artery to bypass aortic stenosis. In the mid 1920s, Cutler and colleagues had attempted to treat mitral stenosis surgically, but with minimal success. A valvulotome was used through a ventricular approach.

Nonetheless, the influence of these attempts led Gross to the successful ligation and, 5 years later, division of the patent ductus arteriosus. In Baltimore, Blalock and Taussig began their series of pioneering surgical procedures for various cardiac anomalies, the first and most dramatic of which was the “blue baby” operation—the creation of a systemic shunt from the subclavian artery to the pulmonary artery in patients with congenital pulmonic stenosis.

Crafoord and Nylin reported the successful end-to-end anastomosis of the aorta after resection of an aortic coarctation at the same time that Gross and Hufnagel carried out their first case. This last operation demonstrated that lesions of the thoracic and abdominal segments of the aorta were amenable to a surgical approach.