Virchow's triad revisited.

To the Editor: Three factors predispose patients to thrombosis--stasis, hypercoagulability, and vessel wall injury--which have commonly been termed Virchow's triad. However, this is not what Rudolf Virchow originally described.

In the mid-19th century, the very existence of pulmonary emboli was debated. It was known that some patients died with thrombi in the pulmonary arteries, but it was widely thought by Virchow's contemporaries, such as Paget, Bochdalek, and Cruveilhier, that these thrombi formed de novo in the pulmonary circulation and did not migrate from distant sites. They thought that the inflammation of vessel walls was the primary impetus for thrombosis in the pulmonary arteries or in the venous tree. Virchow endeavored to show that inflammation of the vessel wall was instead a secondary phenomenon--that the inflammatory vessel wall changes observed at autopsy are a result of thrombosis. Cruveilhier wrote, "I believe that to form a spontaneous non-traumatic phlebitis, it is necessary to have a reason for the irritation that affects the wall or lining of the veins .... The only difference between my viewpoint and that of my opponents [ie, Virchow's] is that I believe that phlebitis precedes the coagulation of blood, while the other viewpoint sees the coagulation as before the phlebitis." (1)

Virchow responded, "I have endeavored to clear up the doctrine of phlebitis ... by substituting for the mysticism which pervaded Cruveilhier's interpretation, merely a statement of the real facts. We do not know that inflammation as such has any necessary connection with coagulation." (2)

In the setting of this debate, Virchow performed a series of experiments in which he introduced foreign bodies, such as elderberry cores, into the veins of dogs. (3) He showed that the sluggish blood flow in veins was indeed able to deliver foreign bodies to the heart and the pulmonary arteries. On page 294 of his opus, Virchow noted that once these foreign bodies lodged in the pulmonary circulation, there was a triad of adverse consequences: "[I]n all cases the blood formed more or less extensive clots around the introduced body .... [T]he list of possible consequences of the obstruction could be grouped into three categories: 1. Phenomena due to the irritation of the vessel and its surroundings; 2. Phenomena due to blood-coagulation; 3. Phenomena due to the interruption of the blood-stream." (3)

These experiments showed that pulmonary artery thrombosis and inflammation result from venous emboli. Virchow's original triad actually referred to the consequences of thrombosis rather than to its precipitants. He even endeavored to disprove the concept that vessel wall irritation could induce thrombosis. After instilling ice, "one of the most irritating agents," into veins, he found that "there was neither an inflammation nor any coagulation which proves that in situ clot formation after the irritation of the inner surface of the vessel does not occur." He also recognized that humans with pathologically identified pulmonary thrombi almost always have thrombi in the deep veins, supporting his view that pulmonary artery thrombosis generally results from venous embolism rather than from in situ thrombosis within the pulmonary circulation.

Decades later, Aschoff--an admirer of Virchow's teachings--criticized him for dismissing the role of the vessel wall in thrombosis: "We have up to this point considered two of the conditions which have to do with the building of thrombi, namely, changes in the blood stream, and qualitative and quantitative alterations in the platelets. We will now take up a third condition, which used to play the chief role in the teaching about thrombosis, but the significance of which was soon greatly limited by Virchow. I refer to alteration of the vessel wall itself." (4)

It is now recognized that endothelial damage, hypercoagulability, and stasis lead to thrombosis. (5), (6) As Virchow showed, thrombosis leads to endothelial damage, hypercoagulability, and stasis. So, is Virchow's triad the chicken or the egg? Perhaps both. The formation of thrombus and the consequences of thrombosis may be inseparable, because the same three factors resulting from thrombosis also cause thrombosis. Thrombus does beget thrombus; clots do propagate and recur. Because abnormal blood flow, intrinsic properties of the blood, and vessel wall damage lead to thrombosis, whereas thrombosis itself leads to stasis, endothelial damage, and further coagulation, it is easy to appreciate how a vicious cycle might occur. This might in part explain the high recurrence rate after initial thrombosis, especially when residual thrombus is still present. (7) This also might be extrapolated to argue for aggressive thromboprophylaxis to prevent situational venous thrombosis (8) and for aggressive treatment of small thrombi of uncertain clinical consequence, such as calf vein thrombosis. Virchow was correct in asserting that thrombosis has important consequences.

A century and a half ago, no effective treatments for venous thromboembolism were available. Now, we have anticoagulants. Although these drugs are not "treatments" for venous thromboembolism in that they do not dissolve thrombi, they do prevent thrombus propagation and thereby limit the sequelae of thrombosis. In this sense, anticoagulants help to prevent Virchow's original triad (ie, the consequences of thrombosis) from becoming the contemporary triad (ie, the precipitants of thrombosis). After 150 years, it turns out that Virchow's original triad and its modern interpretation may be one and the same.

We thank Jodith Janes and Doris Haag for their assistance in locating pertinent historical writings, and we appreciate the helpful comments of Dr. P.C. Malone.

References

(1.) Cruveilhier J. Traite D'Anatomie Pathologique Generale. Paris, Chez J.-B. Bailliere, 1852.

(2.) Virchow RLK. Cellular Pathology [F Chance, transl]. Philadelphia, J.B. Lippincott, 1963, ed 2.

(3.) Virchow RLK. Gesammelte Abhandlungen zur Wissenschaftlichen Medicin. Frankfurt, Meidinger Sohn & Co., 1856. [Reprint edition: Virchow RLK. Thrombosis and Emboli (1846-1856) (AC Matzdorff, WR Bell, transl). Canton, MA, Science History Publications, 1998.]

(4.) Aschoff L. Thrombosis, in Lectures on Pathology. New York, Paul B. Hoeber, Inc., 1924, pp 253-278.

(5.) Nielsen HK. Pathophysiology of venous thromboembolism. Semin Thromb Hemost 1991;17(Suppl 3):250-253.

(6.) Lip GY, Gibbs CR. Does heart failure confer a hypercoagulable state? Virchow's triad revisited. J Am Coll Cardiol 1999;33:1424-1426.

(7.) Prandoni P, Lensing AWA, Prins MH, et al. Residual venous thrombosis as a predictive factor of recurrent venous thromboembolism. Ann Intern Med 2002;137:955-960.

(8.) Thromboembolic Risk Factors (THRIFT) Consensus Group. Risk of and prophylaxis for venous thromboembolism in hospital patients. BMJ 1992;305:567-574.

Daniel J. Brotman, MD

Department of General Internal Medicine

Cleveland Clinic Foundation

Cleveland, OH

Steven R. Deitcher, MD

Department of Hematology and Medical Oncology

Cleveland Clinic Foundation

Cleveland, OH

Gregory Y.H. Lip, MD

Haemostasis Thrombosis and Vascular Biology Unit

University Department of Medicine City Hospital

Birmingham, UK

Axel C. Matzdorff, MD, PHD

Department of Hematology, Oncology, and Hemostasis

Caritasclinic Saarbruecken

Saarbruecken, Germany