Ann Thorac Surg 2000;69:1867-1872
© 2000 The Society of Thoracic Surgeons
Original articles: Cardiovascular
Immunochemical analysis of polyspecific antibodies in patients exposed to bovine fibrin sealant
Dana R. Fastenau, MSa,
John A. McIntyre, PhDa
a Center for Reproduction and Transplantation Immunology, Methodist Hospital, Indianapolis, Indiana, USA
Address reprint requests to Dr McIntyre, HLA-Vascular Biology Laboratory, St. Francis Hospital, 1600 Albany St, Beech Grove, IN 46107
e-mail: jmcintyre{at}iquest.net
 |
Abstract
|
|---|
Background. Patients exposed to bovine thrombin preparations in fibrin sealant often develop antibodies to bovine coagulation proteins, which cause significant bleeding by cross-reacting with human homologues. Recipients of our left ventricular assist system (LVAS) routinely are exposed to fibrin sealant; therefore, we determined whether they developed antibodies.
Methods. We compared sera from 6 LVAS recipients exposed to fibrin sealant (THROMBOGEN, Johnson & Johnson, Arlington, TX ) during LVAS placement to that of 5 nonexposed LVAS recipients. Pre-LVAS and weekly post-LVAS sera were tested for immunoglobulin (Ig)G, IgA, and IgM reactivity to THROMBOGEN by enzyme-linked immunosorbent assay. Peak IgG and IgA reactive sera were characterized by immunoblotting.
Results. All patients exposed to THROMBOGEN developed antibodies: 5 developed IgG, 4 IgA, and 3 IgM. In contrast, nonexposed patients did not develop antibodies. Only some antibody reactivity was contributed by antithrombin or antifactor V antibodies. Silver stain sodium dodecyl sulfate-polyacrylamide gel electrophoresis analyses of THROMBOGEN showed more than 18 bands, many of which were recognized in Western blot by positive patient sera.
Conclusions. We found both IgG and IgA polyspecific antibody responses in patients exposed to bovine thrombin preparations.
|
Introduction
|
|---|
Although fibrin sealant is used successfully in a variety of operations for wound closure and bleeding control [1], the topical bovine thrombin component of fibrin sealant has been shown to be immunogenic [2–4]. Patients exposed to fibrin sealant often develop antibodies to plasma proteins in the bovine thrombin preparation, many of which are clotting factors or glycoproteins involved in coagulation [5, 6]. Antibodies to these bovine proteins reportedly cause significant bleeding by cross-reacting with human homologues [7, 8]. Recipients of a left ventricular assist system (LVAS) often experience bleeding and thrombosis [9]. Our LVAS recipients routinely are exposed to fibrin sealant during LVAS placement operations; therefore, we determined whether LVAS recipients developed antibodies to proteins in the bovine thrombin preparations. We used both the enzyme-linked immunosorbent assay (ELISA) to identify seroconverted patients and immunoblotting to characterize their antibody specificities. We found both IgG and IgA polyspecific antibody responses in patients exposed to bovine thrombin preparations.
|
Material and methods
|
|---|
Patients
Serum samples were collected from 11 patients who were hospitalized for cardiac failure and subsequently received a Novacor LVAS while awaiting cardiac transplantation. Samples were collected before placement and at least weekly thereafter until the time of discharge after transplantation. The first 6 patients were exposed to THROMBOGEN topical bovine thrombin (Johnson & Johnson, Arlington, TX) during LVAS placement, and the last 5 patients intentionally were not exposed to THROMBOGEN.
All patients were men, and the median age was 49 years (range, 32 to 54 years). None of the patients had preexisting renal or liver disease, pulmonary parenchymal disease, or fixed hypertension. None of the patients had preexisting coagulopathy, platelet disorders, or cerebrovascular or peripheral vascular disease. None had previously had LVAS placement. Two of the 6 exposed patients (patients 1 and 6) had coronary artery bypass grafting, during which they might have been exposed to fibrin sealant. Records were not available to exclude the possibility. None of the remaining patients had any other operations during which they might have been exposed to fibrin sealant.
All patients received intravenous heparin in the immediate postoperative period. Five of the 6 exposed patients (patients 1–5) were later switched to a regimen of Coumadin for the duration of LVAS support. The remaining exposed patient and all 5 nonexposed patients continued to receive heparin. The activated partial thromboplastin time and prothrombin time were monitored three times daily for all patients to ensure proper administration of anticoagulants. Because of the anticoagulant therapy all patients remained hospitalized during the LVAS support until after cardiac transplantation.
Enzyme-linked immunosorbent assay
Sera were tested by ELISA for immunoglobulin (Ig)G, IgA, and IgM antibodies to the bovine thrombin preparation. Flat-bottom microtiter plates were coated overnight at 4°C with 2.5 µg per well of THROMBOGEN topical bovine thrombin (Johnson & Johnson, Arlington, TX) in Tris-buffered saline (TBS). Nonspecific binding control plates also were prepared with TBS only. Sera from LVAS recipients (1:100) in 1% bovine serum albumin (Sigma, St. Louis, MO) were added in triplicate to plates that had been blocked with 10% bovine serum albumin. Rabbit antihuman alpha thrombin (1:800; American Diagnostica, Greenwich, CT) was used as a positive control, and serum from a healthy individual (1:100) was used as a negative control. Bound human antibodies were detected with alkaline phosphatase-conjugated, affinity-purified goat antibodies to human IgG, IgA, or IgM (1:1000; Sigma) in 1% bovine serum albumin, and rabbit immunoglobulin was detected with an alkaline phosphatase-conjugated goat antirabbit antibody (1:1000; Sigma). All incubations were performed for 1 hour at room temperature followed by three washes with TBS. Substrate consisting of para-nitrophenyl phosphate tablets in diethanolamine buffer was incubated on plates at 37°C in the dark. IgG plates were developed for 20 minutes, IgM plates for 40 minutes, and IgA plates for 60 minutes. Reactions were stopped with 3 mol/L sodium hydroxide. The optical density (OD) was read at 410 nm, and nonspecific binding control plate OD410 values were subtracted from THROMBOGEN plate OD410 values when nonspecific binding control values were at least 0.100. Sera from 56 normal individuals also were tested, and the ELISA values obtained from the normal subjects were used to establish cutoff values. Patient results were considered positive if they were greater than the OD410 values that included 99% of the normal controls.
The ELISA also was performed with peak IgG-, IgA-, and IgM-positive patient sera using microtiter plates coated with 2.5 µg per well of purified bovine thrombin (Enzyme Research Laboratories, South Bend, IN) or purified bovine factor V (Enzyme Research Laboratories). Rabbit antihuman alpha thrombin (1:800; American Diagnostica) and rabbit antihuman factor V (1:100; Dako, Carpinteria, CA) were used as positive control antibodies and were detected with the alkaline phosphatase-conjugated goat antirabbit antibody (1:1000; Sigma). The ELISA was performed similarly with the purified human proteins (Enzyme Research Laboratories).
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis
The THROMBOGEN was electrophoretically separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under nonreducing conditions [10]. One microgram was electrophoresed using an 8% gel with 120 V for 1.5 hours at 4°C to prevent degradation of THROMBOGEN proteins. Proteins were visualized using a modified silver staining method [11]. Fixation was done overnight (50% methanol, 12% acetic acid, 1.85% formaldehyde) followed by three washes (50% ethanol) for 20 minutes per wash. After pretreatment for 1 minute (0.02% sodium thiosulphate) impregnation was performed for 20 minutes (0.2% silver nitrate, 2.78% formaldehyde). After two washes (dH2O) for 20 seconds per wash, the gels were developed (6% sodium carbonate, 0.8% sodium thiosulphate, 1.85% formaldehyde) until bands appeared. The reaction was stopped (50% methanol, 12% acetic acid) for 10 minutes after which the final wash was done (50% methanol) for 20 minutes. All steps were done at room temperature.
Western blot
THROMBOGEN was electrophoresed as indicated, after which the proteins were electrophoretically transferred to nitrocellulose in 3-[cyclohexylamino]-1-propanesulfonic acid buffer with 90 V for 30 minutes at 4°C. Peak IgG- and IgA-positive patient sera (1:100) in TBS-Tween were incubated with the membrane that had been blocked overnight at 4°C with 1% bovine serum albumin in TBS-Tween. Bound human antibodies were detected with alkaline phosphatase-conjugated goat antihuman IgG or IgA (1:1000; Sigma) in TBS-Tween. Incubations were done for 2 hours at room temperature and were followed by three 20-minute washes with TBS-Tween. The membrane was incubated with substrate consisting of 5-bromo-4-chloro-3-indolyl phosphate and nitroblue tetrazolium in alkaline phosphatase buffer until bands appeared and then rinsed in dH2O. The western blot also was done with sera from normal individuals and with the following antibodies or antisera: goat antihuman factor XI antisera (Nordic Immunology, Capistrano Beach, CA), rabbit antihuman factor X antisera (Nordic Immunology), rabbit antihuman factor IX (Dako), mouse antihuman factor VIII (Behring Diagnostics, Montreal, Quebec, Canada), rabbit antihuman factor VII (CalBiochem, La Jolla, CA), rabbit antihuman factor V (Dako), goat antihuman alpha thrombin (Sera Lab, Crawley Down, Sussex, England), rabbit antihuman fibrinogen (Dako), goat antihuman high molecular weight kininogen antisera (Nordic Immunology), and rabbit antibovine IgG (Sigma). The antibodies were detected with alkaline phosphatase-conjugated goat antirabbit, rabbit antigoat, or goat antimouse immunoglobulin (1:1,000; Sigma).
|
Results
|
|---|
After LVAS placement all 6 patients exposed to THROMBOGEN developed antibodies to it. Five patients developed IgG, 4 developed IgA, and 3 developed IgM antibodies. Development of IgG occurred on average in 19.8 ± 13.2 days (range, 5 to 34 days), IgA in 29 ± 30.4 days (range, 10 to 64 days), and IgM in 29.5 ± 27.6 days (range, 10 to 49 days). After patients were discharged, weekly samples were no longer collected; however, testing was done whenever blood was collected for other purposes. Positive IgG levels and IgA levels were maintained until well after LVAS explant and cardiac transplant. Many of the exposed patients are presently positive. Figure 1 is a representative example of ELISA results using patient serum, which shows IgG, IgA, and IgM anti-THROMBOGEN antibody reactivity. In contrast, of the five patients not exposed to THROMBOGEN, none developed antibodies.
View larger version (20K):
[in this window]
[in a new window]
|
Fig 1. Immunoglobulin (Ig)G, IgA, and IgM anti-THROMBOGEN antibodies detected in a patient serum by enzyme-linked immunosorbent assay. The dotted line indicates the positive-negative cutoff. Placement of left ventricular assist system (LVAS) and cardiac transplantation (Heart) are noted. (OD 410 = optical density at 410 nm.)
|
|
Table 1 shows that some of the antibodies in exposed patients were reactive with purified bovine factor V, and some were reactive with human thrombin or factor V. Comparison of the antibody reactivity to THROMBOGEN and to the four purified proteins with a representative IgA-positive patient serum is shown in Figure 2. Figures 3 and 4 show the analyses of THROMBOGEN using SDS-PAGE and silver stain. Western blot analysis of THROMBOGEN with a representative IgG-positive patient serum and a representative IgA-positive patient serum compared with normal human serum also are shown in Figures 3 and 4, respectively. The same banding patterns were observed for all five IgG-positive patient sera and three IgA-positive patient sera; serum from the remaining IgA-positive patient was not analyzed. IgM-positive patient sera were not analyzed. No bands were observed for the 1 nonexposed, antibody-negative patient and the 3 normal individuals analyzed. Figure 5 shows the reactivity of antibodies to human coagulation proteins in a western blot analysis of THROMBOGEN. The antibody to bovine IgG also was reactive with THROMBOGEN (data not shown).
View larger version (11K):
[in this window]
[in a new window]
|
Fig 2. Immunoglobulin (Ig)A anti-THROMBOGEN antibodies compared with antithrombin and antifactor V antibodies detected in a patient serum by enzyme-linked immunosorbent assay. (BFII = purified bovine thrombin; BFV = purified bovine factor V; HFII = purified human thrombin, HFV = purified human factor V, OD = optical density, THROMB = THROMBOGEN.)
|
|
View larger version (33K):
[in this window]
[in a new window]
|
Fig 3. Immunoglobulin (Ig)G anti-THROMBOGEN antibodies detected by Western blot. The lines drawn between the Western blot and the gel indicate corresponding proteins that are not aligned due to shrinkage of the gel during the silver staining procedure. Lane 1, Silver stain sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of THROMBOGEN. Lane 2, Western blot analysis of THROMBOGEN with patient serum. Lane 3, Western blot analysis of THROMBOGEN with normal serum.
|
|
View larger version (41K):
[in this window]
[in a new window]
|
Fig 4. Immunoglobulin (Ig)A anti-THROMBOGEN antibodies detected by Western blot. The lines drawn between the Western blot and the gel indicate corresponding proteins that are not aligned due to shrinkage of the gel during the silver staining procedure. Lane 1, Silver stain sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of THROMBOGEN. Lane 2, Western blot analysis of THROMBOGEN with patient serum. Lane 3, Western blot analysis of THROMBOGEN with normal serum.
|
|
View larger version (72K):
[in this window]
[in a new window]
|
Fig 5. Western blot analysis of THROMBOGEN with antihuman coagulation protein antibodies. Arrows indicate the correct size of each protein. Lane 1, antifactor XI; Lane 2, antifactor X; Lane 3, antifactor IX; Lane 4, antifactor VIII; Lane 5, antifactor VII; Lane 6, antifactor V; Lane 7, antifactor IIA (thrombin); Lane 8, antifactor I; Lane 9, anti high molecular weight kinnogen.
|
|
After LVAS placement, all 6 exposed patients and 3 of 5 nonexposed patients had significant bleeding complications. The number of red blood cell and platelet units transfused as well as the procedures performed because of bleeding are outlined in Table 2. There were no significant differences in number of red blood cell units transfused or procedures between exposed and nonexposed patients (p = 0.137 and p = 0.382, respectively). The difference in number of platelet units transfused was statistically significant (p = 0.0362). In addition to bleeding complications, 3 exposed patients had neurologic complications, either cerebrovascular accidents (n = 2) or transient ischemic attacks (n = 1). Although none of the nonexposed patients had neurologic complications, the difference between groups was not significant (p = 0.182). Other complications experienced by patients in either group included cardiac arrhythmia (n = 2), respiratory dysfunction (n = 2), and hepatic dysfunction (n = 3). All complications are indicated in Table 2. The average time of LVAS support (time to cardiac transplantation) was 78.5 ± 30 days (range, 28 to 138 days), with no significant difference between exposed patients (76 ± 18 days; range, 56 to 101 days) and nonexposed patients (81 ± 42 days; range, 28 to 138 days; p = 0.804). The average time to recovery (time to discharge after cardiac transplantation) was 22 ± 41 days (range, 7 to 144 days) with no significant difference between exposed patients (31 ± 56 days; range, 4 to 144 days) and nonexposed patients (12 ± 4 days; range, 8 to 12 days; p = 0.463).
View this table:
[in this window]
[in a new window]
|
Table 2. Number of Red Blood Cell and Platelet Units Transfused and Procedures Performed Because of Bleeding Complications
|
|
|
Comment
|
|---|
The ELISA data show that LVAS recipients exposed to THROMBOGEN during LVAS placement developed IgG, IgA, and IgM antibodies to THROMBOGEN, whereas nonexposed patients did not develop antibodies. Historically the described antibodies to topical bovine thrombin preparations were of IgG and IgM isotypes [12]. We have been unable to locate reports of IgA antibodies in response to THROMBOGEN. Development of IgG and IgM antibodies to bovine thrombin and factor V, which cross-react with the human homologues, have been reported previously [2, 5–8, 13], but antibodies that do not cross-react also have been reported [14]. We found that 2 patients developed IgG and IgA antibodies that reacted with human thrombin. These same 2 patients also developed IgG or IgA antibodies, or both, to human factor V, along with IgM antibodies. An additional patient also developed IgA and IgM antibodies to human factor V. None of the patients developed antibodies to bovine thrombin, but all 5 patients developed antibodies of at least one isotype to bovine factor V.
Although many of the patients with antibodies to THROMBOGEN had antibodies to thrombin, factor V, or both, the observation that not all patients did suggests that THROMBOGEN stimulated the production of antibodies to proteins other than thrombin and factor V. This is further supported by the observation that the antibody reactivities to purified thrombin and factor V were much lower than to THROMBOGEN, indicating that only some antibody reactivity was contributed by antithrombin or antifactor V antibodies.
Silver stain SDS-PAGE analysis of THROMBOGEN showed the presence of many proteins, with 18 bands ranging from 20 kDa to more than 205 kDa. In contrast to previously reported Western blot analyses of sera from similarly exposed patients, which used purified bovine or human coagulation proteins [2, 6, 8], we used the commercially prepared product for western blot analyses. Western blot analysis of the THROMBOGEN preparation with heterologous antibodies to several human coagulation proteins demonstrated the presence of these proteins in THROMBOGEN, only some of which have been reported previously [2, 5, 6, 8]. Western blot analyses of THROMBOGEN with IgG- and IgA-antibody-positive patient sera demonstrated that exposed patients developed antibodies to many of the proteins present in the preparation. Thus, THROMBOGEN stimulated the production of several antibodies in addition to antithrombin and antifactor V. If these antibodies are cross-reactive with human homologues, they might be just as pathogenic as antibovine thrombin and antibovine factor V antibodies.
Bovine IgG also was found in immunoblots with antibodies specific to bovine IgG. On the basis of molecular size, many patients appeared to develop antibodies to bovine IgG. We speculate that such antibovine IgG responses might ameliorate the benefits of intravenous Ig in these patients on the basis of cross-reactivity with human IgG. In support of that possibility, 5 of 6 patients who were exposed to THROMBOGEN also received intravenous Ig. None of those patients treated with intravenous Ig had arrested antibody development or lower antibody titers, despite its successful use reported elsewhere [13].
Recently, antiphospholipid antibody (aPA) has been found in patients who had LVAS placement [15]. We were particularly interested to learn whether the proteins in bovine thrombin preparations could be implicated in aPA production, because certain coagulation proteins bind to phospholipids [16]. Indeed, aPAs associated with thrombosis often recognize complexes of phospholipid and their respective phospholipid-binding plasma proteins [16]. We have shown that proteins in the topical bovine thrombin preparations bind to phospholipid and support aPA ELISA reactivity; thus the proteins could stimulate development of aPA that are indistinguishable from aPA found in other patient populations [17]. In support of our proposal, comparisons of LVAS recipients exposed to THROMBOGEN and LVAS recipients not exposed have shown the frequency of aPA to be significantly greater in LVAS recipients exposed to THROMBOGEN [18]. These data suggest that, in addition to topical bovine thrombin preparations stimulating development of antibodies that inhibit coagulation factors thus causing significant bleeding [6–8], immunization by topical bovine thrombin preparations appears to stimulate the development of antibodies that recognize certain coagulation proteins bound by phospholipid that might be involved in thrombosis.
Exposed and nonexposed patients had fairly comparable complications after LVAS, except for bleeding or thromboembolic complications. Although the number of red blood cell units transfused and the number of procedures performed in response to bleeding were not different between the two groups, the number of platelet units transfused was statistically significantly greater for exposed patients than nonexposed patients. A clinically meaningful number of exposed patients had thromboembolic complications compared with nonexposed patients, although statistical significance was not reached. Although exposed patients developed antibodies that reacted with human coagulation proteins, correlating these antibodies with changes in hemostasis can be difficult. All patients were treated with coumadin or heparin after LVAS placement until the time of cardiac transplant. The activated partial thromoplastin time and prothrombin time were monitored daily for determining proper anticoagulant dosing. Thus, changes in these values cannot be used to determine the effect of antibodies apart from changes in anticoagulant dosing.
There were no preoperative placement differences between exposed and nonexposed patients that would have contributed to the bleeding and thromboembolic complications. Other than use of THROMBOGEN the only other postoperative difference between the two groups was the anticoagulant therapy. The majority (5 of 6) of exposed patients were switched from heparin to coumadin after the immediate postsurgical period, whereas the remaining exposed patient and all 5 nonexposed patients remained on intravenous heparin therapy. It is more likely that the antibodies stimulated by THROMBOGEN exposure would be responsible for the increased bleeding and thromboembolic complications than would the use of coumadin. Despite these antibodies and the possibly correlated complications, the exposed patients did not have prolonged times to cardiac transplantation or recovery after transplantation.
We have shown that LVAS recipients exposed to bovine thrombin preparations produced polyspecific antibodies to IgG, IgA, and IgM isotypes, whereas nonexposed LVAS recipients did not produce antibodies. The demonstration of both IgG and IgA polyspecific antibody responses by western blot indicates that exposed patients can develop antibodies other than antithrombin and antifactor V that may be detrimental. As a result of these studies, the cardiothoracic surgeons in our program have discontinued or markedly decreased use of bovine thrombin preparations during LVAS placement. To date, none of the nonexposed patients has seroconverted.
|
Footnotes
|
|---|
This study was supported in part by grants from the Showalter Foundation and Baxter Novacor, Inc.
|
References
|
|---|
-
Jackson M.R., MacPhee M.J., Drohan W.N., Alving B.M. Fibrin sealant. Blood Coagulation Fibrinolysis 1996;7:737-746.[Medline]
-
Stricker R.B., Lane P.K., Leffert J.D., Rodgers G.M., Shuman M.A., Corash L. Development of antithrombin antibodies following surgery in patients with prosthetic cardiac valves. Blood 1988;72:1375-1380.[Abstract/Free Full Text]
-
Flaherty M.J., Wener M.H. Antibodies to thrombin in postsurgical patients. Blood 1989;73:1386.[Free Full Text]
-
Stricker R.B., Corash L. Response to Flaherty and Wener. Antibodies to thrombin in postsurgical patients. Blood 1989;73:1386.
-
Berruyer M., Amiral J., Ffrench P., et al. Immunization by bovine thrombin used with fibrin glue during cardiovascular operations. J Thorac Cardiovasc Surg 1993;105:892-897.[Abstract]
-
Chouhan V.D., DeLaCadena R.A., Nagaswami C., Weisel J., Kajani M., Rao A.K. Simultaneous occurrence of human antibodies directed against fibrinogen, thrombin, and factor V following exposure to bovine thrombin. Thromb Haemost 1997;77:343-349.[Medline]
-
Rapaport S.I., Zivelin A., Minow R.A., Hunter C.S., Donnelly K. Clinical significance of antibodies to bovine and human thrombin and factor V after surgical use of bovine thrombin. Am J Clin Pathol 1992;97:84-91.[Medline]
-
Zehnder J.L., Lueng L.L. Development of antibodies to thrombin and factor V with recurrent bleeding in a patient exposed to topical bovine thrombin. Blood 1990;76:2011-2016.[Abstract/Free Full Text]
-
Griffith B.P., Kormos R.L., Nastala C.J., Winowich S., Pristas J.M. Results of extended bridge to transplantation. Ann Thorac Surg 1996;61:396-398.[Abstract/Free Full Text]
-
Laemmli U.K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680-685.[Medline]
-
Oakley B.R., Kirsch D.R., Morris N.R. A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal Biochem 1980;105:361-363.[Medline]
-
Flaherty M.J., Henderson R., Wener M. Iatrogenic immunization with bovine thrombin. Ann Intern Med 1989;111:631-634.
-
Tarantino M.D., Ross M.P., Daniels T.M., Nichols W.L. Modulation of an acquired coagulation factor V inhibitor with intravenous immune globulin. J Pediatr Hematol Oncol 1997;19:226-231.[Medline]
-
Carroll J.F., Moskowitz K.A., Edwards N.M., Hickey T.J., Rose E.A., Budzynski A.Z. Immunologic assessment of patients treated with bovine fibrin as a hemostatic agent. Thromb Haemost 1996;76:925-931.[Medline]
-
Fastenau D.R., Wagenknecht D.R., McIntyre J.A. Increased incidence of antiphospholipid antibodies in left ventricular assist system recipients. Ann Thorac Surg 1999;68:137-142.[Abstract/Free Full Text]
-
McIntyre J.A., Wagenknecht D.R., Sugi T. Phospholipid binding plasma proteins required for antiphospholipid antibody detection - an overview. Am J Reprod Immunol 1997;37:101-110.
-
Fastenau D.R., Hormuth D.A., McIntyre J.A. Antiphospholipid antibodies in left-ventricular assist system recipients after exposure to topical bovine thrombin. Transplant Proc 1999;31:141-142.[Medline]
-
Fastenau D.R., Hormuth D.A., McIntyre J.A. Exposure to THROMBOGEN is implicated in the development of antiphospholipid antibodies in left ventricular assist system recipients. Lupus 1998;7(Suppl):S197.
Accepted for publication December 23, 1999.
Top
Abstract
Introduction
