Post-Transplant Complications

Retrospective multivariate analysis of hepatic veno-occlusive disease after blood or marrow transplantation: possible beneficial use of low molecular weight heparin

Article metrics


This retrospective cohort study of 462 consecutive adult allogeneic and autologous blood or marrow transplantation (BMT) patients compared the incidence of hepatic veno-occlusive disease (VOD) after BMT with three prophylactic regimens. Patients receiving heparin (Hep), heparin + prostaglandin E1 (Hep + PGE1) or low molecular weight heparin (LMWH) as a prophylactic VOD regimen were compared to a historical cohort receiving no VOD prophylaxis. Of 462 BMT patients, VOD was diagnosed in 22% (31 of 142) of the no prophylaxis group, 11% (11 of 104) of the Hep, 12% (13 of 110) in the Hep + PGE1 and 4% (four of 106) of the LMWH group (P = 0.0002). VOD was the primary cause of death in 20% (12 of 59). By multivariate logistic regression, independent risk factors for developing VOD were: no VOD prophylactic regimen, unrelated allogeneic BMT, Karnofsky performance score (KPS) <80 and aspartate aminotransferase (AST) 50 U/l. There was no increase in the rate of death due to hemorrhagic events or VOD in any prophylaxis group compared to the control group. Prospective randomized trials of Hep vs LMWH vs placebo are warranted to assess the efficacy of heparin compounds in the prevention of VOD. Bone Marrow Transplantation (2001) 27, 627–633.


Veno-occlusive disease (VOD) of the liver is a clinical syndrome, characterized by hyperbilirubinemia, painful hepatomegaly and fluid retention.1 It occurs in up to 54% of patients and is a leading cause of blood or marrow transplantation (BMT)-related death with a mortality rate up to 47%.23 Diagnostic criteria for VOD have been previously described by McDonald et al124 and Jones et al.3 The etiology of VOD is not clearly defined, however, drug-induced hepatocellular damage occurs in the centrilobular zones of the liver. The coagulation cascade may have a role in VOD pathophysiology due to clotting factor activation leading to fibrin deposition in the central veins. It has been postulated that anticoagulant therapy could prevent fibrin deposition and subsequent hepatic damage.56 Several clinical trials of prophylactic heparin (Hep) during BMT have shown contradictory results.789101112 Nevertheless, these studies demonstrated that low doses of Hep could be safely administered to BMT patients. Prostaglandin E1 (PGE1) was used as a VOD prophylactic regimen by Gluckman et al,13 who found a decreased incidence of VOD in patients treated with PGE1, whereas Bearman et al,14 concluded that PGE1 was too toxic. Previous work has demonstrated the efficacy and safety of low molecular weight heparin (LMWH) for deep venous thrombosis prophylaxis and treatment.1516 A small prospective study comparing LMWH to placebo revealed a shorter duration of VOD-related symptoms after BMT in the LMWH treated patients.17 In our retrospective study, we analyzed three VOD prophylactic regimens in 462 patients undergoing allogeneic or autologous BMT. The VOD preventive effects of Hep, Hep + PGE1 and LMWH were compared to each other and to a historical control group. This is the first retrospective multivariate analysis of three prophylactic regimens used in one patient population.

Patients and methods

Study design

This retrospective cohort study included all adult patients (18 years old) who underwent autologous or allogeneic BMT between 1991 and 1997. Patients received VOD prophylaxis consisting of (a) None (1991–1997), (b) Hep (1994–1997), (c) Hep + PGE1 (1994–1997) or (d) LMWH (1996–1997). The Hep and Hep + PGE1 treated patients were part of a randomized phase II study,18 the remaining data were based on retrospective analysis. The no VOD prophylaxis group was primarily a historical cohort of patients transplanted before initiation of the VOD prophylactic regimens. During 1996–1997, patients were either enrolled on the randomized phase II study or given LMWH based on the discretion of the treating physician and patient refusal to participate in the randomized study. Our study does not include patients treated after 1997 because of major changes in BMT conditioning regimens, graft-versus-host disease (GVHD) prophylaxis and supportive care that may have affected VOD incidence.

Study population

Four hundred and sixty-two consecutive adult patients underwent allogeneic (n = 216) or autologous (n = 246) BMT at Roswell Park Cancer Institute from 1991 to 1997 for the treatment of leukemia (n = 169/462 = 37%), lymphoma (n = 159/462 = 34%), breast cancer (n = 83/462 = 18%), multiple myeloma (n = 36/462 = 8%) and other hematologic disorders or selected solid tumors (n = 15/462 = 3%).

Diagnosis of VOD

VOD was defined according to previously published criteria as the onset of two of the following occurring before day 30 post-BMT: (a) jaundice (total serum bilirubin >2 mg/dl), (b) hepatomegaly and right upper abdominal pain, (c) ascites and/or unexplained weight gain.1 No other identifiable cause for liver disease including infection or GVHD could be present.

Prophylactic regimen administration

Patients in the randomized study received Hep 5 U/kg/h i.v. continuous infusion 12–24 h prior to the start of their myeloablative regimen, keeping the daily PTT 50. If the PTT was >50, the Hep was adjusted daily until a PTT of 50 was obtained. PGE1 was started a minimum of 12 h prior to conditioning therapy and was given via continuous infusion at a rate of 0.1–0.4 μg/kg/h depending on patient tolerance. Therapy with Hep ± PGE1 was continued until day 30 or day of discharge with a minimum treatment of 15 days post BMT. In the LMWH group, 30 mg enoxaparin was administered subcutaneously twice a day within 24 h of initiation of the conditioning regimen and continued until day of discharge or day 28 post BMT.

BMT conditioning regimens

The conditioning regimens included: (1) thiotepa and TBI;19 (2) thiotepa and carboplatin;19 (3) busulfan and TBI;19 (4) etoposide, cyclophosphamide and TBI;20 (5) etoposide, cyclophosphamide and carmustine;21 (6) busulfan and cyclophosphamide;22 (7) cyclophosphamide;23 (8) melphalan and TBI;24 (9) thiotepa and cyclophosphamide.19All regimens, except 7 and 9, were used for autologous and allogeneic BMT (7 for allogeneic and 9 for autologous BMT only). No pharmacokinetic measurements were performed during any of the conditioning regimens. All time-points were calculated from the day of the infusion of blood or bone marrow (day 0).

Donor types and GVHD prophylaxis

Allogeneic BMT patients received bone marrow, peripheral blood or cord blood from related or unrelated donors. Allogeneic BMT patients received GVHD prophylaxis with (1) cyclosporine and methotrexate;25 (2) cyclosporine and methylprednisolone;26 (3) cyclosporine, OKT3 and methylprednisolone;27 (4) cyclosporine, methotrexate and methylprednisolone;28 or (5) FK506 (tacrolimus) and methotrexate.29 FK506 was substituted if the patient exhibited cyclosporine toxicity. Two syngeneic BMT patients analyzed with the allogeneic group received no GHVD prophylaxis. Acute GVHD was treated with high dose methylprednisolone from 2–10 mg/kg/day and cyclosporine 2–4 mg/kg/day. Grade I GVHD was treated with topical steroids only.

Other BMT supportive care

Autologous and allogeneic BMT patients received antimicrobial prophylaxis with fluconazole (or ketoconazole prior to 1995) and acyclovir. All patients received acyclovir prophylaxis from day 0 to day of discharge for a minimum of 21 days. In addition, allogeneic BMT patients received ganciclovir prophylaxis from day 30 to day 100 following BMT if the donor or recipient were positive for cytomegalovirus. Autologous BMT patients with TBI- or carmustine-containing regimens and all allogeneic BMT patients received intravenous immunoglobulin (IVIG) infection prophylaxis beginning on day −1 at a dose of 500 mg/kg then weekly × 8 doses, then biweekly × 4 doses. Acute GVHD was classified according to published criteria; after day 100, chronic GVHD was graded according to published criteria.3031

Statistical analysis

VOD was evaluated as a binary endpoint. Univariate analysis used the Fisher's exact test or analysis of variance (ANOVA) where appropriate (SPSS v7.5, Chicago, IL, USA). To determine VOD risk factors, forward stepwise logistic regression models were used conditional on the Wald score with each variable having a P value <0.1 to enter and >0.1 to remove. For each significant variable in the model, 95% confidence intervals of the relative risk estimates were calculated. To adjust for multiple comparisons, statistical significance was considered at P < 0.01. P values between 0.01 and 0.10 were considered as trends toward an association and are presented in the Tables.


Study patients

Of the 462 BMT patients, 246 underwent autologous and 216 allogeneic BMT. One hundred and forty-two patients did not receive any VOD prophylaxis either because they underwent BMT between 1991 and 1993 (n = 127) when no prophylaxis was routinely used, or between 1994 and 1997 (n = 15) if they were (a) at low risk for VOD; (b) at high risk for hemorrhagic complications; or (c) refused prophylaxis. One hundred and four patients received Hep and 110 patients received Hep + PGE1 as part of a phase II study.18 One hundred and six patients received LMWH during 1996 and 1997.

Table 1 summarizes the patient characteristics for all patients by prophylactic regimen. Although this was not a randomized study, most patient characteristics were similar between VOD prophylactic groups. There was no significant difference with regard to age, gender, diagnosis, TBI, pre-transplant aspartate aminotransferase (AST) level, or BMT type (autologous vs allogeneic) with a trend toward statistical significance for Karnofsky performance score (KPS) (P = 0.05) and donor type (autologous vs related allogeneic vs unrelated allogeneic, P = 0.04). The no prophylaxis group had the lowest proportion of unrelated allogeneic BMT; the Hep group had the highest proportion of unrelated allogeneic BMT patients; the Hep + PGE1 and LMWH groups had similar proportions of unrelated allogeneic BMT patients. There was a significant difference between VOD prophylactic regimens with respect to BMT conditioning regimen, stem cell source and GVHD prophylaxis. These variables were considered in the univariate and multivariate analyses.

Table 1  Summary of patient characteristics by prophylactic regimen

VOD incidence by prophylactic regimen

Table 2 shows the incidence of VOD by prophylactic regimen in all BMT patients. Of 462 BMT patients, VOD was diagnosed in: (a) 31 cases (22%) in the no prophylaxis group; (b) 11 cases (11%) in the Hep group; (c) 13 cases (12%) in the Hep + PEG1 group; and (d) four cases (4%) in the LMWH group (P = 0.0002). Since heparin compounds can increase the risk of hemorrhage, we examined the rate of death due to hemorrhagic events (diffuse alveolar hemorrhage, gastrointestinal bleeding, or central nervous system hemorrhage) for each VOD prophylactic group and found no significant difference in the rate of primary cause of death due to a hemorrhagic event (P > 0.1).

Table 2  Incidence of VOD and cause of death due to VOD and hemorrhage by prophylactic regimen in all BMT patients

Univariate analysis of VOD incidence

Table 3 summarizes the patient characteristics that were tested for an association with developing VOD. Age, gender, time from diagnosis to BMT, busulfan and thiotepa containing conditioning regimens, methotrexate- or OKT3-containing GVHD prophylaxis and time to neutrophil and platelet engraftment were not significantly different between patients who did and did not develop VOD. A significantly higher proportion of patients who developed VOD had an allogeneic BMT (P = 0.0002), a related or unrelated donor (P = 0.0007), stem cells derived from bone marrow (P = 0.0019), no VOD prophylaxis (P = 0.0002), and a KPS <80 (P = 0.003). There was a trend toward a significant difference with a diagnosis of leukemia (P = 0.05), conditioning with etoposide, cytoxan and TBI (VCT, P = 0.04), TBI-containing regimens (P = 0.06), and median pretransplant AST (P = 0.02).

Table 3  Univariate analysis of patient characteristics by development of veno-occlusive disease (VOD)

Multivariate analysis of risk factors for VOD

The following factors were considered in the multivariate logistic regression model: diagnosis (leukemia vs lymphoma vs myeloma vs breast cancer vs other), time from diagnosis to BMT (continuous), KPS ( 80 vs <80), conditioning regimen (busulfan-containing vs thiotepa-containing vs VCT vs other), donor type (autologous vs related allogeneic vs unrelated allogeneic), stem cell type (BM vs PB vs BM + PB vs CB), methotrexate-containing GVHD prophylaxis, OKT3-containing GVHD prophylaxis, VOD prophylaxis (LMWH vs Hep ± PGE1 vs None) and pretransplant AST (elevated vs normal). Table 4 summarizes the variables that were significant predictors of VOD. The data shown evaluate AST as a dichotomous variable (elevated vs normal AST). Substituting AST as a continuous variable in the same model results in the same independent predictors and a finding that for every increase in AST by 10 U/l, there is a corresponding increase in risk of developing VOD of 16.5% (95% confidence interval 1.0–32.4%, P = 0.04).

Table 4  Multivariate analysis of risk factors for developing VODa

Since none of the autologous BMT patients received cyclosporine as part of the GVHD prophylaxis regimen, we were unable to differentiate the effect of cyclosporine vs donor type as predictive of developing VOD. The increasing use of PB and decreasing use of BM as the stem cell source over time yielded a difference in the proportion of each by VOD prophylactic regimen in the univariate analysis, however, stem cell type was not shown to be an independent risk factor for developing VOD in the multivariate analysis.

Treatment of VOD

Of the 59 patients who developed VOD, nine patients received tissue plasminogen activator (tPA)32 and 39 received tPA + PGE1 as therapy for VOD. Eleven had close observation and no drug therapy after developing VOD. In addition, 26 of the 403 patients who did not develop VOD were treated with PGE1 for suspected VOD (ie transient bilirubin elevation). Of the 48 patients who were treated with tPA ± PGE1 for VOD, three (6%) died of hemorrhagic events (one from each VOD prophylaxis group: LMWH, Hep, Hep + PGE1). The use of tPA ± PGE1 for VOD therapy was not associated with an increased risk of hemorrhagic death in any prophylaxis group.

Rate of death due to VOD in the VOD cases by prophylactic regimen

Overall, 20% (12 of 59) of VOD cases died of VOD as the primary cause of death. There was no statistically significant difference in the rate of death due to VOD between the prophylactic and control groups (no prophylaxis: 8/31 (26%), Hep: 1/11 (9%), Hep + PGE1 1/13 (8%), LMWH 2/4 (50%), P > 0.1). There was also no statistically significant difference in the rate of death due to a hemorrhagic event between the groups (no prophylaxis: 4/31 (13%), Hep: 1/11 (9%), Hep + PGE1: 2/13 (15%), LMWH: 1/4 (25%), P > 0.10).


Given the severity and incidence of VOD and the lack of effective treatment, pursuit of prophylaxis is important. This and previous studies used very low doses of heparin for VOD prophylaxis because of the risk of bleeding. LMWH can be safely given at higher doses with a lower risk of hemorrhagic complications. Thus, the LMWH doses used in this study potentially had a higher prophylactic effect than unfractionated heparin with or without PGE1. Despite the heterogeneity of conditioning regimens, patient characteristics, and GVHD prophylaxis, the VOD incidence was significantly decreased with all of the prophylactic regimens. The greatest decrease was seen in the LMWH group.

Although our study was not randomized, we found a similar incidence of VOD in the no prophylaxis group as several other studies.13781013 We confirmed previous findings that unrelated donor, decreased performance status and elevated AST are significant risk factors for developing VOD. We concluded that a significantly lower incidence of VOD developed in patients treated with Hep ± PGE1 or LMWH than in the control patients. We also found that LMWH was more effective than unfractionated heparin; PGE1 had no additional benefit.

The encouraging results of our study justify the role of Hep and LMWH in the prophylaxis of VOD, however prospective randomized trials should verify the superior efficacy and low risk of hemorrhagic complications of LMWH in preventing VOD after BMT.


  1. 1

    McDonald GB, Sharma P, Matthews DE et al. Venoocclusive disease of the liver after marrow transplantation: diagnosis, incidence, and predisposing factors Hepatology 1984 4: 116–122

  2. 2

    McDonald GB, Hinds MS, Fisher LD et al. Veno-occlusive disease of the liver and multiorgan failure after bone marrow transplantation: a cohort study of 355 patients Ann Intern Med 1993 118: 255–267

  3. 3

    Jones RJ, Lee KSK, Beschorner WE et al. Veno-occlusive disease of the liver following bone marrow transplantation Transplantation 1987 44: 778–783

  4. 4

    McDonald GB, Sharma P, Matthews DE et al. The clinical course of 53 patients with veno-occlusive disease of the liver after marrow transplantation Transplantation 1985 39: 603–608

  5. 5

    Shulman HM, McDonald GB, Matthews D et al. An analysis of hepatic venocclusive disease and centrilobular hepatic degeneration following bone marrow transplantation Gastroenterology 1980 79: 1178–1191

  6. 6

    Shulman HM, Fisher LB, Schoch HG et al. Veno-occlusive disease of the liver after marrow transplantation: histological correlates of clinical signs and symptoms Hepatology 1994 19: 1171–1181

  7. 7

    Attal M, Huguet F, Rubie H et al. Prevention of hepatic veno-occlusive disease after bone marrow transplantation by continuous infusion of low-dose heparin: a prospective, randomized trial Blood 1992 79: 2834–2840

  8. 8

    Rosenthal J, Sender L, Secola R et al. Phase II trial of heparin prophylaxis for veno-occlusive disease of the liver in children undergoing bone marrow transplantation Bone Marrow Transplant 1996 18: 185–191

  9. 9

    Marsa-Vila L, Gorin NC, Laporte JP et al. Prophylactic heparin does not prevent liver veno-occlusive disease following autologous bone marrow transplantation Eur J Haematol 1991 47: 346–354

  10. 10

    Carreras E, Bertz H, Arcese W et al. Incidence and outcome of hepatic veno-occlusive disease after blood or marrow transplantation: a prospective cohort study of the European Group for Blood and Marrow Transplantation Blood 1998 92: 2599–3604

  11. 11

    Cahn JY, Flesch M, Brion A et al. Prevention of veno-occlusive disease of the liver after bone marrow transplantation: heparin or no heparin? Blood 1992 80: 2149–2150

  12. 12

    Bearman SI, Hinds MS, Wolford JL . A pilot study of continuous infusion heparin for the prevention of hepatic veno-occlusive disease after bone marrow transplantation Bone Marrow Transplant 1990 5: 407–411

  13. 13

    Gluckman E, Jolivet I, Scrobohaci ML et al. Use of prostaglandin E1 for prevention of liver veno-occlusive disease in leukaemic patients treated by allogeneic bone marrow transplantation Br J Haematol 1990 74: 277–281

  14. 14

    Bearman SI, Shen DD, Hinds MS et al. A phase I/II study of prostaglandin E1 for the prevention of hepatic venocclusive disease after bone marrow transplantation Br J Haematol 1993 84: 724–730

  15. 15

    Hull RD, Raskob GE, Pineo GF et al. Subcutaneous low-molecular-weight heparin compared with continuous intravenous heparin in the treatment of proximal-vein thrombosis New Engl J Med 1992 326: 975–982

  16. 16

    Spiro TE, Johnson GJ, Christie MJ et al. Efficacy and safety of enoxaparin to prevent deep venous thrombosis after hip replacement surgery Ann Intern Med 1994 121: 81–89

  17. 17

    Or R, Nagler A, Shpilberg O et al. Low molecular weight heparin for the prevention of veno-occlusive disease of the liver in bone marrow transplantation patients Transplantation 1996 61: 1067–1071

  18. 18

    Schriber J, Milk BJ, Baer MR et al. A randomized phase II trial comparing heparin (Hep) +/− Prostaglandin E1 (PG) to prevent hepatotoxicity (HT) following bone marrow transplantation (BMT): preliminary results Blood 1996 88: (Suppl. 1) 413a (Abstr. 1642)

  19. 19

    Alidina A, Lawrence D, Ford LA et al. Thiotepa-associated cardiomyopathy during blood or marrow transplantation: association with the female sex and cardiac risk factors Biol BMT 1999 5: 322–327

  20. 20

    Brown RA, Wolff SN, Fay JW et al. High-dose etoposide, cyclophosphamide and total body irradiation with allogeneic bone marrow transplantation for patients with acute myeloid leukemia in untreated first relapse: a study by the North American Marrow Transplant Group Blood 1995 85: 1391–1395

  21. 21

    Jagannath S, Armitage JO, Dicke KA et al. Prognostic factors for response and survival after high-dose cyclophosphamide, carmustine, etoposide with autologous bone marrow transplantation for relapsed Hodgkin's disease J Clin Oncol 1989 7: 179–185

  22. 22

    Santos GW, Tutschka PJ, Brookmeyer R . Marrow transplantation for acute nonlymphocytic leukemia after treatment with busulfan and cyclophosphamide New Engl J Med 1983 309: 1347–1353

  23. 23

    Thomas ED, Storb R, Fefer A et al. Aplastic anaemia treated by marrow transplantation Lancet 1972 1: 284–289

  24. 24

    Jagannath S, Vesole DH, Glenn L et al. Low-risk intensive therapy for multiple myeloma with combined autologous bone marrow transplantation and blood stem cell support Blood 1992 80: 1666–1672

  25. 25

    Nash RA, Pepe MS, Storb R et al. Acute graft-versus-host disease: analysis of risk factors after allogeneic marrow transplantation and prophylaxis with cyclosporine and methotrexate Blood 1992 80: 1838–1845

  26. 26

    Shepherd JD, Shore TB, Reece DE et al. Cyclosporine and methylprednisolone for prophylaxis of acute graft-versus-host disease Bone Marrow Transplant 1988 3: 553–558

  27. 27

    Alam A, Baer MR, Bernstein S et al. Phase II clinical trial of OKT3 for acute graft-versus-host disease (GVHD) prophylaxis in allogeneic bone marrow transplantation (alloBMT) Blood 1997 90: (Suppl. 1) 102a (Abstr. 449)

  28. 28

    Chao NJ, Schmidt GM, Niland JC et al. Cyclosporine, methotrexate and prednisone compared with cyclosporine and prednisone for prophylaxis for acute graft-versus-host disease New Engl J Med 1993 329: 1225–1230

  29. 29

    Ratanatarathorn V, Nash RA, Przepiorka D et al. Phase III study comparing methotrexate and tacrolimus (prograf, FK506) with methotrexate and cyclosporine for graft-versus-host disease prophylaxis after HLA-identical sibling bone marrow transplantation Blood 1998 92: 2303–2314

  30. 30

    Glucksberg H, Storb R, Fefer A et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HLA-matched sibling donors Transplantation 1974 18: 295–304

  31. 31

    Shulman HM, Sullivan KM, Weiden PL et al. Chronic graft-versus-host syndrome in man: a long-term clinico-pathologic study of 20 Seattle patients Am J Med 1980 69: 204–217

  32. 32

    Schriber J, Milk B, Shaw D et al. Tissue plasminogen activator (tPA) as therapy for hepatotoxicity following bone marrow transplantation Bone Marrow Transplant 1999 24: 1311–1314

Download references


The authors acknowledge Sandra Dascomb for her assistance with data management.

Author information

Correspondence to PL McCarthy Jr.

Rights and permissions

Reprints and Permissions

About this article


  • veno-occlusive disease
  • prophylaxis
  • heparin
  • blood and marrow transplantation

Further reading