[Veno-occlusive disease after bone marrow transplantation: preventive effect of heparin]

Liver veno-occlusive disease is a severe toxic effect observed after bone marrow transplantation. Clinical manifestations are jaundice, painful liver enlargement, and fluid-sodium retention. Histologically there is non-thrombotic obliteration of the centro-lobular veins associated with centro-lobular necrosis. This severe complication of bone marrow transplantation occurs early and is caused by a toxic processing effect. Incidence is variable, 2 to 50% in reported series, depending on patients, type of marrow provessing and on diagnostic criteria (which hinders comparison between studies). According to most studies, veno-occlusive disease regresses spontaneously. Mortality, depending on the severity of the symptoms, varies from 20 to 50%. Pathogenesis remains under debate: the initial event would occur in the sinusoid endothelium creating a state of local hypercoagulability by release of tissue factors favoring deposit of coagulation factors, especially factor VIII, in the subendothelial region of the veinules. There is also a direct toxic effect on centro-lobular hepatocytes which is further aggravated by ischemia and venous stasis. use of heparin to prevent veno-occlusive disease was proposed by the Besançon group in 1985 after they observed a low incidence (1 case in 65) in patients who were given low doses of heparin to maintain patent central catheters. The same team confirmed in 1992 the low incidence in a large retrospective series of 444 patients given either an autograft (3 cases in 253 patients, i.e. 1.2%), or an allograft (5 cases in 191 patients, i.e. 2.6%). Two single-center studies, one in Seattle and the other at the Saint-Antoine hospital in Paris, published in 1990 and 1991, did not show any difference in patients given heparin or not. Inversely, a randomized study published by Attal in 1992 including 161 patients showed a significant difference in the incidence of veno-occlusive disease between patients given heparin (2.5%) and those who were not given heparin (13.7%; p < 0.01). All these studies show that with low doses (100-150 U/kg) the risk is very very low. The mechanism of action of heparin would appear to be related to its protective effect on the endothelium rather than its hemostasis effect. The vascular protective effect of prostaglandin E1 suggests it might also be useful in preventing veno-occlusive disease.