Bone Marrow Transplantation (2000) 26, 881–886  2000 Macmillan Publishers Ltd All rights reserved 0268–3369/00 $15.00 www.nature.com/bmt Predictive markers for hepatic veno-occlusive after hematopoietic stem cell transplantation in adults: a prospective single center study

S Tanikawa1, S Mori1, K Ohhashi1, H Akiyama1, T Sasaki1, H Kaku1, K Hiruma1, T Matsunaga2, T Morita2 and H Sakamaki1

1Bone Marrow Transplantation Team, Tokyo Metropolitan Komagome Hospital, Tokyo; and 2Pharmaceutical Development Department I, Teijin Limited, Tokyo, Japan

Summary: Hepatic veno-occlusive disease (VOD) is one of the serious complications of hematopoietic stem cell transplantation Hepatic veno-occlusive disease (VOD) is a major com- (HSCT) characterized by painful hepatomegaly, ascites and plication after hematopoietic stem cell transplantation jaundice.1 Its frequency varies greatly from 1 to 54%,2,3 (HSCT). Aetiological determinants, diagnosis and and mortality rate is high. VOD may possibly arise from treatment remain unclear. Changes in - injury or disruption of the lining the sinusoids fibrinolysis parameters and N-terminal propeptide for and pores connected to the hepatic venular lumen. Endo- type III procollagen (P-III-P) have been studied in thelial injury causes activation of the coagulation system patients with or without VOD after HSCT. We prospec- and post-sinusoidal obstruction, but the pathogenesis is not tively measured C activity, tissue plasminogen clear and the diagnosis of VOD is based only on clinical activator (t-PA), antithrombin III (AT-III), - criteria advocated by the Seattle or Baltimore groups.1,4,5 ogen activity (PLG), -antithrombin III (TAT), Thus, changes in hemostatic parameters have been exam- ␣2-plasmin inhibitor (␣2-PI),fibrinogen (Fbg) and ined after HSCT. Decreases in natural such P-III-P in 44 consecutive adult patients undergoing allo- as protein C6 and antithrombin III (ATIII)7 have been geneic HSCT. Each parameter was determined before observed after HSCT. Increases in the serum concentration conditioning, on day 0 of HSCT and weekly for 5 weeks. of the N-terminal propeptide for type III procollagen (P- Five of the 44 patients developed VOD at a median post III-P), which has been found to be a sensitive marker for HSCT of day 3 (range, day 3 to 12). On repeated analy- pulmonary and fibrosis, have been observed after sis of variance (ANOVA), there were significant differ- HSCT.8,9 However, there are few papers in which both ences between patients with and without VOD in P-III-P coagulation-fibrinolysis markers and P-III-P were prospec- (P Ͻ 0.0001), (P Ͻ 0.0001), t-PA (P Ͻ 0.0001), tively measured after HSCT. PLG (P Ͻ 0.0001), AT-III(P Ͻ 0.0001), Fbg (P Ͻ 0.0001), In this study, we prospectively measured the level of ␣2-PI (P = 0.0002). Levels of P-III-P were significantly coagulation-fibrinolysis markers and P-III-P in 44 consecu- higher in patients with VOD than without VOD, before tive patients undergoing allogeneic HSCT; furthermore, we preparative chemotherapy (P Ͻ 0.005) and on days 0 determined whether we could find an early predictive and 7 (P Ͻ 0.001). On day 0, levels of t-PA were signifi- marker for the development of VOD among these measure- cantly higher in patients with VOD than without VOD ments. (P Ͻ 0.05). On day 7, levels of protein C were signifi- cantly lower in patients with VOD than without VOD (P Ͻ 0.01). On day 0, there were trends of differences Patients and methods (P = 0.0515) between patients with and without VOD in the levels of protein C. These results suggest P-III-P, t- Patients PA and protein C are predictive markers for VOD after HSCT in adults. Moreover, the serum P-III-P level All patients more than 15 years of age who underwent allo- before start of conditioning might indicate patients at geneic hematopoietic stem cell transplantation in Tokyo risk for developing VOD. Bone Marrow Transplantation Metropolitan Komagome Hospital from January 1998 to (2000) 26, 881–886. March 1999 were enrolled into the study. Written informed Keywords: hematopoietic stem cell transplantation; consent was obtained from all patients. Forty-two of 44 VOD; P-III-P enrolled patients underwent first bone marrow trans- plantation. One patient with severe aplastic anemia suffer- ing from late graft failure after BMT underwent allogeneic peripheral blood stem cell transplantation. Only one patient Correspondence: Dr S Tanikawa, Division of Blood Transfusion, Tokyo Metropolitan Toshima Hospital, 33-1, Sakae-chou, Itabashi-ku, Tokyo, with acute lymphoblastic leukemia (ALL) in second com- 173-0015, Japan plete remission underwent a second BMT. This patient was Received 2 March 2000; accepted 5 July 2000 transplanted from an HLA-A one-locus-mismatched sister. Predictive markers for hepatic VOD S Tanikawa et al 882 The following preparation regimens were used. Patients (SRL); SRL, Tokyo, Japan). The AT-III activity, PLG and with myeloid malignancies who underwent related BMT ␣2-PI were determined by the Hitachi auto-analyzer 7070 received busulfan (4 mg/kg) for 4 days and cyclophos- with commercial kits (AT-III: Testzyme S AT-III, PLG: phamide (CY) (60 mg/kg) for 2 days (BU + CY). Patients Testzyme S PLG and ␣2-PI: Testzyme S APL; Daiichika- with myeloid malignancies who underwent unrelated BMT gaku, Tokyo, Japan). Fbg were assessed with a coagulo- received BU + CY followed by total lymphoid irradiation meter and commercial kits (Fibrinoquik; Organonteknika, (TLI), 7 Gy in two fractions. Patients with lymphoid malig- Durham, NC, USA). nancies who underwent first BMT received four doses of 2 2 g/m cytosine arabinoside over 3 days, 60 mg/kg of CY Measurement of serum P-III-P for 2 days and 12 Gy of total body irradiation (TBI) in six fractions over a period of 3 days. One patient who Serum samples for measurement of P-III-P were collected underwent second BMT received BU + CY. One patient at the same time as plasma samples for assays of coagu- with severe aplastic anemia who underwent allogeneic lation-fibrinolysis parameters and stored at −80°C until second hematopoietic stem cell transplantation received analysis. P-III-P was measured by radioimmunoassay rabbit antithymocyte (ATG) (2 mg/kg/day × 4 (Riagnost P III Pc.t.; CIS Bio International, Saclay, days) followed by TBI (2 Gy × 2/day × 2 days). All patients France). received short-term methotrexate and cyclosporine for GVHD prophylaxis. No patients received any prophylaxis Statistical analysis for VOD, which could have affected the coagulation pro- file. The repeated measures analysis of variance (ANOVA) was used to analyze differences in each parameter between Diagnosis of hepatic VOD patients with and without VOD. Statistical significance between the values of each parameter at different times in All enrolled patients were monitored prospectively for patients with and without VOD was determined by the Stu- VOD. A diagnosis of VOD was made according to the dent’s t-test. P Ͻ 0.05 was considered significant. Multi- clinical criteria3 of two of the following being present variate analysis of the established risk factors for VOD and before day 20 post transplant: (1) hyperbilirubinemia P-III-P before preparative chemotherapy was carried out (bilirubin у 2.0 mg/dl or 34.2 ␮mol/l), (2) painful hepato- with the multiple logistic regression analysis. megaly, and (3) unexplained weight gain (Ͼ2% from baseline), no other explanation for these signs and symp- toms being present at the time of diagnosis. Severity of Results VOD was classified into mild, moderate, or severe.3 Patients who met the criteria for VOD, but who were not Patient characteristics treated and whose illness was self-limited, were considered to have mild VOD. Those whose VOD resolved but who Of the 44 patients, five developed VOD (ALL 3, CML 1, RAEB in T:1) (Table 1). All five patients with VOD received treatment such as diuretics for fluid retention or narcotic analgesics for painful hepatomegaly, were con- sidered to have moderate VOD. Patients were classified as Table 1 Patient characteristics having severe VOD if they experienced adverse effects from liver disease, and signs, symptoms, and laboratory VOD (−) VOD (+) values did not resolve before day 100 or the patient died, whichever occurred first. No. (Male/Female) 38 (19/19) 5 (5/0) Age (year, median and range) 32 (15–48) 28 (18–35) Diagnosis Assays of coagulation-fibrinolysis parameters ALL 10 3 AML 11 Protein C activity, tissue (t-PA), CML 11 1 antithrombin III (AT-III), plasminogen activity (PLG), SAA 1 thrombin-antithrombin III (TAT), ␣2-plasmin inhibitor MDS ␣ RA 2 ( 2-PI),fibrinogen (Fbg) were monitored in all enrolled RAEB 1 patients. Each coagulation-fibrinolysis parameter was RAEB in T 1 determined before preparative chemotherapy, on the day of CMMoL 1 marrow or peripheral blood stem cell infusion (day 0), and Chronic myelofibrosis 1 Regimen on days 7, 14, 21, 28 and 35. Venous blood was collected BU + CY 16 in 1/10th volume of 3.8% sodium citrate. Plasma samples BU + CY + TLI 12 2 were separated by centrifugation at 2000 g for 10 min, 4°C Ara − C + CY + TBI 9 3 and stored at −80° until assayed. Protein C activity was ATG + TBI 1 determined by a Behring fibrin timer A coagulometer with Allogeneic transplants HLA-matched sibling 17 2 a commercial kit (protein C clot BW; Behringwerke, Mar- HLA-matched unrelated donor 20 3 burg, Germany). The t-PA and TAT were measured by HLA-mismatched family 1 enzyme immunoassay (EIA) with commercial kits (t-PA: Imulyse tPA; Bio-pool, Umeå, Sweden and TAT: TAT VOD(−) = patients without VOD; VOD(+) = patients with VOD.

Bone Marrow Transplantation Predictive markers for hepatic VOD S Tanikawa et al 883 received serologically HLA-matched allogeneic marrow Univariate analysis of established risk factors for VOD grafts; three were unrelated and the other two were related. and P-III-P Three of the five patients underwent conditioning with Ara- Univariate analysis of established risk factors for VOD C + CY + TBI and the rest with BU + CY + TLI. One of (serum ALT, AST3 and albumin2) and P-III-P was carried the five patients died of multiorgan failure (Table 2). Two patients who received a second hematopoietic stem cell out before preparative chemotherapy (Table 3). P-III-P before preparative chemotherapy was the only significant transplant did not develop VOD. The median onset of VOD risk factor. was day 3 (range, 3–12 days). One patient (UPN 282) received recombinant tissue plasminogen activator (rt-PA) (6 × 106 U/day × 2days) on days 10 and 11. Four other Multivariate analysis of established risk factors for VOD patients did not receive any treatments which might affect and P-III-P later coagulation profile data. Multivariate analysis of established risk factors for VOD In one patient (a 37-year-old male, CML in first chronic phase, first BMT from an unrelated donor), the serum bili- and P-III-P was done before preparative chemotherapy rubin exceeded 2.0 mg/dl and the body weight became (Table 4). P-III-P before preparative chemotherapy was more influential on occurrence of VOD than was any other more than 102% of baseline on day 12. This patient could not be diagnosed as having VOD, because he was suffering risk factor examined. from acute GVHD at the same time. Excluding this patient, statistical analysis was carried out on 43 patients. Discussion

Serum P-III-P in adults with or without VOD Coagulation-fibrinolysis parameters and P-III-P in patients undergoing allogeneic HSCT have been studied. The num- In the two-way ANOVA with repeated measurements, there bers of patients in almost all previous reports have been were significant differences between patients with and with- small and conclusions inconsistent. Although Faioni et al6 Ͻ out VOD in P-III-P (P 0.0001). Before preparative suggested that protein C levels before the start of condition- Ͻ chemotherapy (P 0.005) and on days 0 and 7 ing would indicate patients at risk for developing VOD, Ͻ (P 0.001), the levels of P-III-P were significantly higher many others have failed to demonstrate that pre-condition- in patients with VOD than without VOD (Figure 1). ing levels of various coagulation-fibrinolysis parameters can predict for the development of VOD in patients Coagulation-fibrinolysis parameters undergoing HSCT. However, decrease of natural anticoagu- lants and change of fibrinolytic activity following HSCT In the two-way ANOVA with repeated measurements, there have been shown to be relevant to the occurrence of VOD were significant differences between patients with and with- in several reports.10–15 Specifically, protein C and AT-III out VOD in protein C (P Ͻ 0.0001), t-PA (P Ͻ 0.0001), decreased usually 0–14 days after marrow infusion in PLG (P Ͻ 0.0001), AT-III (P Ͻ 0.0001), Fbg (P Ͻ 0.0001) patients with VOD.10,12,15 PAI-1 has been reported to ident- and ␣2-PI (P = 0.0002) (Figures 1 and 2). On day 0, the ify VOD in patients with post-transplant hyperbilirubin- levels of t-PA were significantly higher in patients with emia.14 Significant increase in P-III-P has been reported in VOD than without VOD (P Ͻ 0.05). On day 7, the levels patients with VOD after HSCT.8,9,16,17 of protein C were significantly lower in patients with VOD In our study, we prospectively measured the level of than without VOD (P Ͻ 0.01). On day 0, there were trends coagulation-fibrinolysis markers and P-III-P in 44 consecu- of differences (P = 0.0515) between patients with VOD and tive patients undergoing allogeneic HSCT. Levels of P-III- without VOD in the levels of protein C (Figure 1). P were significantly higher in patients with VOD than

Table 2 Clinical features of patients with VOD

UPN Age Sex Diagnosis BMT VOD GVHD Complications Outcome acute/chronic (months) Onset Maximum Maximum Tender Severity post BMT weight gain bilirubin hepatomegaly (days) (%) (mg/dl)

254 35 M CML/CP 1 Unrelated 3 3.5 2.9 + Moderate 0/− – DFS (16) 265 29 M ALL/CR 1 Related 12 2.8 2.2 + Moderate 0/− – Relapse 274 18 M ALL/CR 1 Related 3 3.5 2.1 − Moderate 0/− – Relapse 282 26 M RAEB in T Unrelated 10 5.4 2.1 + Moderate 0/− – DFS (11) 283 28 M ALL/Rel 1 Unrelated 3 6.2 3.0 + Severe grade II/− Pulmonary Dead hemorrhage

CML = chronic myelogenous leukemia; CP 1 = first chronic phase; ALL = acute lymphoblastic leukemia; CR 1 = first complete remission; RAEB in T = refractory anemia with excess of blasts in transformation; Rel 1 = first relapse; DFS = disease-free survival.

Bone Marrow Transplantation Predictive markers for hepatic VOD S Tanikawa et al 884 P-III-P tPA 2.6 20 2.4 VOD (+) * 18 2.2 ** VOD (-) 2.0 VOD (+) 16 VOD (-) 1.8 ** 1.6 * 14 1.4 12 U/ml

1.2 ng/ml 1.0 10 0.8 8 0.6 P <0.0001 P <0.0001 0.4 6 Pre 0 7 14 21 28 35 Pre 0 7 14 21 28 35 Day Day * : P <0.005 ** : P <0.001 * : P <0.05

Protein C PLG 130 120 120 VOD (+) VOD (+) 110 VOD (-) 110 VOD (-) # # 100 * 100 90 90 % 80 % 70 80 60 50 70 P <0.0001 P <0.0001 40 60 Pre 0 7 14 21 28 35 Pre 0 7 14 21 28 35 Day Day * : P <0.01 # : P =0.0515 # : P =0.098

Figure 1 Changes in N-terminal propeptide for type III procollagen (P-III-P), protein C activity, tissue plasminogen activator (t-PA) and plasminogen activity (PLG) in patients with and without VOD after hematopietic stem cell transplantation. Each P-value in the repeated measures analysis of variance (ANOVA), was represented on the right side of a figure. By means of the two-way ANOVA with repeated measurements, there were significant differences between patients with and without VOD in P-III-P (P Ͻ 0.0001), protein C (P Ͻ 0.0001), t-PA (P Ͻ 0.0001), PLG (P Ͻ 0.0001). Before preparative chemotherapy (P Ͻ 0.005) and on days 0 and 7 (P Ͻ 0.001) , the levels of P-III-P were significantly higher in patients with VOD than without VOD. On day 0, the levels of t-PA were significantly higher in patients with VOD than without VOD (P Ͻ 0.05). On day 7, the levels of protein C were significantly lower in patients with VOD than without VOD (P Ͻ 0.01). On day 0, there were trends of differences (P = 0.0515) between patients with VOD and without VOD in the levels of protein C.

ATIII a2-PI 130 120 125 VOD (+) VOD (+) VOD (-) VOD (-) 120 115 115 110 110 105 % 105 100 % 100 95 95 90 90 85 P <0.0001 P = 0.0002 80 85 Pre 0 7 14 21 28 35 Pre 0 7 14 21 28 35 Day Day TAT Fbg 11 500 10 VOD (+) VOD (+) VOD (-) VOD (-) 9 450 8 400 7 6 350 5 mg/dl ng/ml 4 300 3 250 2 P = 0.5477 P <0.0001 1 200 Pre 0 7 14 21 28 35 Pre 0 7 14 21 28 35 Day Day

Figure 2 Changes in antithrombin III (AT-III), ␣2-plasmin inhibitor (␣2-PI), thrombin-antithrombin III (TAT) and fibrinogen (Fbg) in patients with and without VOD after hematopietic stem cell transplantation. Each P-value in the repeated measures analysis of variance (ANOVA), was represented on the right side of a figure. By means of the two-way ANOVA with repeated measurements, there were significant differences between patients with and without VOD in AT-III (P Ͻ 0.0001), Fbg (P Ͻ 0.0001), ␣2-PI (P = 0.0002).

Bone Marrow Transplantation Predictive markers for hepatic VOD S Tanikawa et al 885 Table 3 Univariate analysis of established risk factors for VOD and of recombinant human TNF-␣. Salat et al14 measured t-PA P-III-P and PAI-1 in 32 patients undergoing BMT. The levels of PAI-1 were significantly higher in patients with VOD than Risk factors Patients with VOD Patients without P value without VOD on day 14. Although they did not mention (mean ± s.e.) VOD (mean ± s.e.) the fact, levels of t-PA tended to be higher in patients with VOD than those without VOD on day 0. Although PAI-1 AST (pre) (IU/1) 26.6 ± 3.9 24.5 ± 1.6 0.633 has been reported to be very useful in the differential diag- ALT (pre) (IU/1) 35.4 ± 9.4 36.3 ± 3.5 0.917 nosis of hyperbilirubinemia after BMT,14 t-PA may be an Alb (pre) (g/dl) 4.14 ± 0.07 4.07 ± 0.06 0.658 early predictive marker for the development of VOD. P-III-P (pre) 1.18 ± 0.15 0.66 ± 0.05 0.002 On day 7, levels of protein C were significantly lower (U/ml) in patients with VOD than in those without VOD in this study. This finding is consistent with previous reports,12,15 s.e. = standard error; pre = before preparative chemotherapy; AST = serum aspartate aminotransferase; ALT = serum alanine aminotransferase; Alb and may simply have resulted from decreased hepatic = serum . synthesis associated with VOD, although changes in coagulation parameters may indicate that the natural antico- agulant has some pathogenetic role in the development of Table 4 Multivariate analysis of established risk factors for VOD and VOD. Regardless of whether decreased levels of protein C P-III-P are a cause or result of VOD, they are likely to aggravate its clinical effects. Risk factors Parameter estimate P value Univariate and multivariate analysis of established risk AST (pre) (IU/l) −0.059 0.481 factors for VOD and P-III-P before preparative chemo- ALT (pre) (IU/l) 0.013 0.734 therapy were undertaken. These results suggest P-III-P is Alb (pre) (g/dl) −1.318 0.507 an independent risk factor for VOD . P-III-P (pre) (U/ml) −2.911 0.054 Our data suggest that P-III-P, t-PA and protein C are predictive markers for hepatic VOD after allogeneic hema- pre = before preparative chemotherapy; AST = serum aspartate topoietic stem cell transplantation in adults. Moreover, aminotransferase; ALT = serum alanine aminotransferase; Alb = . serum P-III-P levels before the start of conditioning might indicate patients at risk for developing VOD. Prophylactic administration of protein C in patients whose P-III-P levels without VOD, even before preparative chemotherapy, as are high before the start of conditioning or on day 0 of well as on days 0 and 7. These findings suggest that serum HSCT, should be investigated. P-III-P levels before the start of conditioning may indicate patients at risk for developing VOD, and P-III-P levels on day 0 are an early predictive marker for the development References of VOD. Although there were trends towards differences between patients with and without VOD in P-III-P levels 1 McDonald GB, Sharma P, Matthews DE et al. Venocclusive before the start of conditioning in the paper of Heikinheimo 9 disease of the liver after bone marrow transplantation: diag- et al, it has not been reported that P-III-P levels were sig- nosis, incidence, and predisposing factors. Hepatology 1984; nificantly higher in patients with VOD than without VOD 4: 116–122. before preparative chemotherapy. P-III-P has been found to 2 Locasciulli A, Bacigalupo A, Alberti A et al. Predictability be a sensitive marker for pulmonary and liver fibrosis.18,19 before transplant of hepatic complications following allo- Therefore, it can be an early predictive marker for the geneic bone marrow transplantation. Transplantation 1989; development of VOD, because fibrous obliteration of ter- 48: 68–72. minal hepatic venules and small lobular veins is the patho- 3 McDonald GB, Hinds MS, Fisher LD et al. Veno-occlusive logic hallmark of VOD. Patients who develop VOD may disease of the liver and multiorgan failure after bone marrow have fibrous change (due to prior chemotherapy etc) of the transplantation: a cohort study of 355 patients. Ann Intern Med 1993; 118: 255–267. liver before conditioning. It has been reported that poor 4 Jones RJ, Lee KSK, Beschorner WF et al. Veno-occlusive hepatic function reserve may be a risk factor for hepatic disease of the liver following bone marrow transplantation. 20,21 VOD. Our results might be relevant to this hypothesis, Transplantation 1987; 44: 778–783. because fibrous change can result in poor functional reserve 5 Schulman HM, Hinterberger W. Hepatic veno-occlusive dis- of the liver. ease-liver toxicity syndrome after bone marrow trans- On day 0, levels of t-PA were significantly higher in plantation. Bone Marrow Transplant 1992; 10: 197–214. patients with VOD than without VOD in this study. Catani 6 Faioni EM, Krachmalnicoff A, Bearman SI et al. Naturally et al22 reported t-PA and PAI-1 levels to be elevated at the occurring anticoagulants and bone marrow transplantation: onset of VOD in two VOD patients. The occurrence of plasma protein C predicts development of venocclusive dis- VOD may be closely related to hypercoagulability. Seeber ease of the liver. Blood 1993; 81: 3458–3462. 23 7 Gordon B, Haire W, Kessinger A et al. High frequency of et al reported endothelial coagulation factors such as t- antithrombin 3 and protein C deficiency following autologous PA and PAI-1 to be released by activation and/or damage bone marrow transplantation for lymphoma. Bone Marrow to endothelial cells following an increase in serum tissue Transplant 1991; 8: 497–502. necrosis factor ␣ (TNF-␣). Vanhinsbergh et al24 found a 8 Rio B, Bauduer F, Arrago JP et al. N-terminal peptide of type significant increase in t-PA and PAI-1 following infusion III procollagen: a marker for the development of hepatic veno-

Bone Marrow Transplantation Predictive markers for hepatic VOD S Tanikawa et al 886 occlusive disease after BMT and a basis for determining the 17 Schuler U, Subat S, Schmidt H et al. Evaluation of procol- timing of prophylactic . Bone Marrow Transplant lagen-III peptide as a marker for veno-occlusive disease after 1993; 11: 471–472. bone marrow transplantation. Ann Hematol 1996; 73: 25–28. 9 Heikinheimo M, Halila R, Fasth A. Serum procollagen type 18 Kirk JME, Bateman ED, Haslam PL et al. Serum type III III is an early and sensitive marker for veno-occlusive disease procollagen peptide concentration in cryptogenic fibrosing of the liver in children undergoing bone marrow trans- alveolitis and its clinical relevance. Thorax 1984; 39: 726– plantation. Blood 1994; 83: 3036–3040. 732. 10 Park YD, Yasui M, Yoshimoto T et al. Changes in hemostatic 19 Rohde H, Vargas L, Hahn E et al. Radioimmunoassay for type parameters in hepatic veno-occlusive disease following bone III procollagen peptide and its application to human liver dis- marrow transplantation. Bone Marrow Transplant 1997; 19: ease. Eur J Clin Invest 1979; 9: 451–459. 915–920. 20 Kami M, Mori S, Tanikawa S et al. Risk factors for hepatic 11 Salat C, Holler E, Wolf C et al. Laboratory markers of veno- veno-occlusive disease after bone marrow transplantation: occlusive disease in the course of bone marrow and sub- retrospective analysis of 137 cases at a single institution. Bone sequent liver transplantation. Bone Marrow Transplant 1997; Marrow Transplant 1997; 20: 397–402. 19: 487–490. 21 Hasegawa S, Horibe K, Kawabe T et al. Veno-occlusive dis- 12 Tabbara IA, Ghazal CD, Ghazal HH. Early drop in protein ease of the liver after allogeneic bone marrow transplantation C and antithrombin III is a predictor for the development of in children with hematologic malignancies: incidence, onset venoocclusive disease in patients undergoing hematopoietic time and risk factors. Bone Marrow Transplant 1998; 22: stem cell transplantation. J Hematother 1996; 5: 79–84. 1191–1197. 13 Scrobohaci ML, Drouet L, Monem-Mansi A et al. Liver veno- 22 Catani L, Gugliotta L, Mattioli Belmonte M et al. Hypercoag- occlusive disease after bone marrow transplantation changes ulability in patients undergoing autologous or allogeneic BMT in coagulation parameters and endothelial markers. Thromb for hematological malignancies. Bone Marrow Transplant Res 1991; 63: 509–519. 1993; 12: 253–259. 14 Salat C, Holler E, Kolb H-J et al. Plasminogen activator 23 Seeber C, Hiller E, Holler E et al. Increased levels of tissue inhibitor-1 confirms the diagnosis of hepatic veno-occlusive plasminogen activator (t-PA) and tissue plasminogen activator disease in patients with hyperbilirubinemia after bone marrow inhibitor (PAI) correlate with tissue necrosis factor alpha transplantation. Blood 1997; 89: 2184–2188. (TNF ␣) release in patients suffering from microangiopathy 15 Lee JH, Lee KH, Kim S et al. Relevance of protein C and S, following allogeneic bone marrow transplantation (BMT). antithrombin III, , and factor VIII for Thromb Res 1992; 66: 373–383. the development of hepatic veno-occlusive disease in patients 24 Vanhinsbergh VWM, Bauer K, Kooistra K et al. Progress of undergoing allogeneic bone marrow transplantation: a pro- fibrinolysis during tumor necrosis factor infusions in humans. spective study. Bone Marrow Transplant 1998; 22: 883–888. Concomitant increase in tissue-type plasminogen activator, 16 Eltumi M, Trivedi P, Hobbs JR et al. Monitoring of veno- plasminogen activator inhibitor type-1 and fibrin(ogen) occlusive disease after bone marrow transplantation by serum degeneration products. Blood 1990; 76: 2284–2289. aminopropeptide of type III procollagen. Lancet 1993; 342: 518–521.

Bone Marrow Transplantation