Bone Marrow Transplantation (2002) 29, 403–408 2002 Nature Publishing Group All rights reserved 0268–3369/02 $25.00 www.nature.com/bmt Post-transplant complications Acquired factor VII deficiency in hematopoietic stem cell transplant recipients AA Toor1, A Slungaard1, U Hedner2, DJ Weisdorf1 and NS Key1 Department of Medicine, University of Minnesota Medical School, Minneapolis, Minnesota, USA; and 2Novo Nordisk Pharmaceuticals Inc., Gentofte, Denmark Summary: Isolated acquired FVII deficiency in the absence of vit- amin K deficiency, coumadin therapy, synthetic liver dys- Acquired factor VII (FVII) deficiency in the absence of function, or overt DIC is rare, having been described in vitamin K deficiency, oral anticoagulant therapy, syn- only a handful of cases. Associated disease states have thetic liver dysfunction, or DIC is rare, with only a included malignancy,2 aplastic anemia,3,4 and antiphospho- handful of cases thus far reported. In the period from lipid antibodies.5,6 In stem cell transplant (SCT) patients, a 1990 to 1996 we identified eight patients with acquired modest reduction in FVII levels (in the 25 to 50% range) FVII deficiency, all of whom presented with pro- has been described in the first 2 weeks following SCT.7–9 longation of the prothrombin time (PT) in the first 2 In these cases it has often been associated with deficiencies weeks following stem cell transplantation (SCT). The of other factors such as FXII or protein C.10 It has been mean plasma FVII clotting activity (FVII:c) was 22% suggested that the fall in FVII and protein C levels in this (range 8–35%) with an approximately equivalent setting may be a harbinger of veno-occlusive disease of the reduction in FVII antigen (FVII:Ag) level. Mean plasma liver (VOD).11 levels of fibrinogen and factors II, V, IX, and X were Bleeding after SCT has been identified as a significant normal. Protein C activity was significantly depressed in risk factor affecting outcome.12–14 Commonly encountered only one of the three patients in whom it was measured. sites of bleeding include hemorrhagic cystitis, diffuse Several patients experienced bleeding complications, alveolar hemorrhage, and oro-mucosal and gastro-intestinal and hemorrhage directly accounted for death in two sources. Although thrombocytopenia is clearly a major con- cases. Veno-occlusive disease of the liver developed in tributing factor, other reasons, such as toxicity related to three patients. We conclude that FVII deficiency should the preparative regimen, infection, and graft-versus-host be considered in the differential diagnosis of prolonged disease, have all been cited. Acquired FVII deficiency, at PT in patients who have recently undergone SCT. The least of the degree previously reported, has not been impli- mechanism of this acquired deficiency state remains to cated as a risk factor for hemorrhage following SCT. be defined. In this communication, we describe the clinical and lab- Bone Marrow Transplantation (2002) 29, 403–408. DOI: oratory features of a series of patients with acquired FVII 10.1038/sj/bmt/1703381 deficiency following SCT, who have been managed at our Keywords: factor VII; stem cell transplantation; institution. The frequency and severity of bleeding compli- hemorrhage; acquired deficiency; acute myelogenous leu- cations in these patients suggest that this coagulopathy may kemia have contributed to adverse outcomes. Methods Factor VII is a vitamin K-dependent zymogen which when activated binds to tissue factor (TF), a transmembrane pro- tein strongly expressed by cells in the adventitia of blood Patients vessels. The TF–VII(a) complex activates both factors IX We conducted a retrospective chart review of eight patients and X in the presence of phospholipid and calcium, and known to have developed acquired FVII deficiency during catalyzes the auto-activation of more zymogen FVII. The SCT at the University of Minnesota between 1990 and TF–VII(a) pathway is now known to be the dominant 1996. The transplants were carried out using a myeloabl- 1 hemostatic mechanism in vivo. ative regimen, followed by allogeneic or autologous hema- topoietic SCT (using bone marrow or peripheral blood stem cells). Post-transplant care consisted of GVHD prophylaxis, Correspondence: Dr NS Key, Department of Medicine, Division of Hema- tology, Oncology, and Transplantation, University of Minnesota Medical anti-microbial prophylaxis and therapy as needed, School, MMC 480, 420 Delaware St SE, Minneapolis, MN 55455, USA nutritional support, and transfusion support with red cell Received 24 October 2000; accepted 12 October 2001 and platelet products. Routine parenteral vitamin K supple- Factor VII deficiency following SCT AA Toor et al 404 mentation was administered to all patients. Daily blood reagent referred to above; (2) APTT-FSL reagent (Sigma counts were obtained and coagulation tests were performed Diagnostics, St Louis, MO, USA); (3) Dade Actin FS if clinically indicated, eg in patients with bleeding, septice- (Dade-Behring Inc); and (4) Organon Teknika (Durham, mia, or suspected DIC. In non-bleeding patients, platelets NC, USA) automated APTT reagent. When a prolonged were transfused to keep the platelet counts greater than 10– APTT was detected with one or more reagents, a 1:1 mix 20 ϫ 109/l. Hemoglobin was routinely maintained at Ͼ8.0 of patient and normal pooled plasma was performed, and G/dl. These parameters were revised upwards for patients the APTT repeated using that same reagent(s). Failure of with clinically overt bleeding. Coagulopathy was assessed the mixing study to correct led to the performance of the initially by a battery of tests that included prothrombin time platelet neutralization procedure16 (with one or more (PT), activated partial thromboplastin time (APTT), throm- reagents) to confirm the diagnosis of a lupus inhibitor. After bin time (TT), Clauss fibrinogen, and fibrin(ogen) degra- converting our procedure to include the dilute Russell viper dation product titer. In the event of a clotting time being venom time (dRVVT), with mix and confirm steps, sub- prolonged, a 1:1 mix with normal plasma was performed sequent to 1996, we were able to determine that the mul- to differentiate the presence of a coagulation inhibitor from tiple APTT screening approach used in this study was a a factor deficiency. This was followed by specific factor very specific, if somewhat less sensitive technique for assays if indicated. At the discretion of the attending phys- screening for lupus anticoagulants. ician, coagulopathy was treated with fresh frozen plasma (FFP), and/or cryoprecipitate infusions with a therapeutic FVII antibody screening methods goal of clinical hemostasis (in the case of clinical bleeding) and/or normalization of laboratory parameters of coagu- Two separate methods were used to screen for the presence lation. In some cases, additional doses of parenteral vitamin of an auto-antibody to FVII. Firstly, we used a previously K were also administered. None of the patients were receiv- described ELISA method in which patient plasma was ing any form of anti-thrombotic therapy nor anti-platelet screened for the presence of anti-FVII(a) IgG.17 Reference agents, such as aspirin or non-steroidal anti-inflammatory ranges and cut-off limits were established by screening agents. multiple normal donors. Secondly, in order to screen for a neutralizing antibody to FVII, we used a plasma mixing Factor assays assay. Specifically, FVII:c was measured in patient and standard (pooled normal) plasma, which were then mixed Measurement of specific clotting factor activity was in all in equal proportions. FVII:c was assayed both immediately cases performed by a one-stage clotting assay (PT-based after mixing and following a 1 h incubation at 37°C. A assay for factors II, VII and X, and APTT-based assay for positive result (ie presence of an inhibitory antibody) was FIX), using a mixture of the appropriate factor-deficient suspected if the FVII:c in the mixed sample (either at time substrate plasma and test plasma. All factor assays were zero or after 1 h incubation) was approximately 15% or performed at 1/10, 1/20, 1/40, and 1/80 dilutions. The more below the expected value. This arbitrary cut-off value results are expressed as a percentage, with 100% equivalent was chosen taking into account the fact that FVII:c in the to the appropriate factor level in pooled plasma samples mix sample might be slightly less than expected – in the from 40 normal volunteers. Reference ranges for factor lev- absence of an inhibitor – simply due to the reproducibility els were established as the mean Ϯ 2 s.d. of 40 normal of the factor VII assay. In order to minimize artifactual loss individuals. Prior to January 1993, rabbit brain thrombo- of activity of labile clotting factors (such as FVIII) in the plastin (Thromboplastin C, Dade Laboratories, Deerfield, mix, a 1 h incubation is routinely used in our laboratory to IL, USA) was used in our laboratory for PT and PT-based screen for possible coagulation factor inhibitory antibodies. factor assays. The stated International Sensitivity Index The exception to this is in the case of suspected FVIII (ISI) of this reagent was 2.71. Subsequently, all PT assays inhibitors, where a 2 h incubation, as described in the ori- were performed using recombinant re-lipidated tissue factor ginal18 (and subsequently modified19) Bethesda assays, is (Innovin, Dade) with an ISI of 1.0. APTT was performed used. using an in-house reagent in which 0.075% aluminum sili- cate was used as an activator, and the chloroform extract Statistical analysis of acetone-dried rabbit brain thromboplastin as the phos- pholipid source.15 The immunoreactive level of factor VII Significance testing of the differences observed between antigen was measured using a commercial ELISA kit clotting factor levels seen in our patients was performed (Diagnostica Stago, Asnieres, France). using a two-tailed, paired Student’s t-test, using Microsoft Plasma fibrinogen was assayed by the Clauss method.