The Discovery Of Heparin And Its First Clinical

The following account of the discovery and first clinical development of heparin was recorded by the physiologist Best (1959), a codiscoverer of insulin as well as a pioneer in studies of heparin. Incidentally, in 1916, while working at Johns Hopkins University to characterize procoagulant substances, McLean (1916) identified a natural anticoagulant substance. Further studies of this material were performed by his supervisor, Dr. Howell, who coined the term, "heparin" to indicate its first extraction from animal hepatic tissues (Gr. h pap [hepar], liver) (Howell and Holt, 1918). Despite its in vitro anticoagulant action, the inability of heparin to prevent platelet-mediated thrombosis (Shionoya, 1927) made it uncertain whether it had antithrombotic potential. However, animal (Mason, 1924) and human studies (Crafoord, 1937) showed that heparin could prevent thrombosis. By the 1950s, heparin was established as an important therapeutic agent in the treatment of venous and arterial thrombosis.


On June 1, 1957, at the Fifth Scientific Meeting of the International Society of Angiology (North American Chapter) in New York, two physicians suggested that heparin might cause arterial embolism in some patients. Rodger E. Weismann, a 43-yr-old Assistant Professor of Clinical Surgery at the Dartmouth Medical School (Fig. 1), and his Resident in Surgery, Dr. Richard W. Tobin, presented their 3-yr experience with 10 patients who developed unexpected peripheral arterial embolism during systemic heparin therapy at the Mary Hitchcock Memorial Hospital, in Hanover, New Hampshire. Their first patient with this complication was reported in detail, and to this day represents a classic description of the syndrome:

This 62-yr-old white woman was admitted to the Hitchcock Hospital Feb. 8, 1955, with left retinal detachment, complicating longstanding myopia ... Left scleral buckling was carried out on Feb. 10, and strict bed rest was required during the ensuing 3 wk. On her beginning ambulation, on March 6, signs and symptoms of left iliofemoral thrombophlebitis were noted, for which systemic heparinization was begun (... heparin sodium in divided subcutaneous doses, totaling 150-300 mg per day ...). On March 16, after 10 days of anticoagulation therapy, sudden signs of right common femoral arterial occlusion led to the diagnosis of common femoral arterial embolism. Successful femoral embolectomy was carried out. She was kept on adequate heparinization and made a satisfactory initial recovery until March 19, ... when signs of sudden occlusion of the distal aorta appeared.

FIGURE 1 Photograph of Dr. Rodger Elmer Weismann, taken circa 1958.

... [P]rompt transperitoneal distal aortic and bilateral iliac embolectomies were performed. In the ensuing 24 h, because unsatisfactory distal circulation persisted, the patient underwent left femoral exploration, with negative findings, and right popliteal exploration, revealing an embolus. She subsequently pursued a favorable course, ... never showing more serious ischemic changes than a small area of superficial gangrene of the right great toe and several small areas of skin infarction of the right leg (Weismann and Tobin, 1958).

The report included a photograph of the emboli removed from the distal aorta and both iliac arteries, with the authors noting their "unusual length and cylindrical shape, suggesting origin in [the] proximal aorta," as well as a corresponding photomicrograph of the embolus. The thromboemboli were described by the authors as "pale, soft, salmon-colored clots" that "histologically ... were comprised mostly of fibrin, platelets and leukocytes; red cells were rare." This appearance was distinguishable from the typical appearance of thrombi originating in the heart (i.e., mulberry-colored thrombi tending to contain cellular elements of the blood in approximately normal proportions), leading the authors to propose "the source for the emboli ... to be aortic mural platelet-fibrin thrombi."

A summary of the 10 reported patients noted that the onset of arterial embolism began between 7 and 15 days, inclusive, of commencing heparin treatment (mean, day 10). Multiple thromboemboli occurred in nine patients; six of the patients died as a direct result of these complications; two survived with extensive amputations, and two were discharged with their extremities intact. The temporal time frame was consistent with the later realization by others that this syndrome represented an immune-mediated reaction initiated by the heparin.

The authors noted that further embolization stopped when the heparin was discontinued, leading to their recommendation that "heparin should be promptly reduced in dosage, and, if possible, discontinued if the presence of fibrin-platelet thrombi adherent to the intima of the aorta is suspected." Aggressive surgical management of emboli was also recommended, as some limbs were salvageable in this way. The authors summarized well the clinical dilemma: "In each instance there was a feeling of futility in the management of the problem, due to anticipation of further emboli from the same or similar sources. Heparin was badly needed to retard distal thrombosis; yet the agent was probably seriously altering the integrity and attachment of the thrombotic source" (Weismann and Tobin, 1958).

B. Roberts and ColIeagues

The communication of Weismann and Tobin was met with considerable skepticism. When a show of hands was elicited to indicate those surgeons who had also observed similar events, none were raised (Weismann, personal communication, July 1998). However, a few years later, Roberts and colleagues from the University of Pennsylvania in Philadelphia described a series of patients who were remarkably similar to those reported by Weismann and Tobin (Roberts et al., 1964; Barker et al., 1966; Kaupp and Roberts, 1972). The key features were summarized as follows:

To witness a series of apparently paradoxical events is disconcerting as well as challenging. When such paradoxes involve totally unexpected results following the use of a major therapeutic agent, it is at first difficult to know whether the relationship is causal or merely coincidental. When, however, the same series of events has been seen repeatedly it is difficult to escape the conclusion that there is some causal relationship, even though the mechanism by which it is accomplished may be unknown ... . During the last 9 yr at the Hospital of the University of Pennsylvania, we have seen a group of 11 patients who suffered unexplained arterial embolization for the first time while being treated with heparin for some condition that could not of itself reasonably be expected to cause arterial emboli ... . AH patients had been receiving heparin for 10 days or more when the initial embolus occurred ... . All emboli removed were of a light color, seemingly made up primarily of fibrin and platelets, and microscopically appeared to be relatively free of red cells. All patients in this group had multiple emboli ... . Of the four deaths, three were attributed to cerebral vascular accidents presumably embolic in origin and one was thought to have resulted from a perforation of the small bowel 2 wk after the removal of a mesenteric embolus (Roberts et al., 1964).

Roberts' group also viewed the likely pathogenesis as that of embolization of platelet-fibrin-rich material originating within the aorta, rather than the heart. Furthermore, they believed that the thrombi were initially formed on aortic ulcerations that acted as a nidus for thrombus formation. This pathogenesis was suggested by the observation that such adherent thrombi could be removed from the proximal aorta in a few of the patients (Roberts et al., 1964; Kaupp and Roberts, 1972).

C. An Immune Basis for Heparin-lnduced Thrombosis?

The delay between initiation of heparin therapy and onset of embolization caused Roberts and colleagues (1964) to speculate that the etiology could represent an "antiheparin factor," resulting perhaps from "an antigen-antibody mechanism." Furthermore, the observation that the first 21 patients reported with this syndrome from both Hanover and Philadelphia had received heparin exclusively by subcutaneous or intramuscular, rather than intravenous, injection also was offered by Roberts' group as support for immune sensitization. Apparent heparin-induced thrombosis did not seem rare to these investigators: at least 13 of 110 (12%) patients with peripheral arterial emboli managed by the Philadelphia group over a decade were believed to have been caused by preceding heparin treatment (Barker et al., 1966).

0 0

Post a comment