Activation of the coagulation cascade after infusion of a factor... concentrate in congenitally deficient patients
Transcription
Activation of the coagulation cascade after infusion of a factor... concentrate in congenitally deficient patients
From www.bloodjournal.org by guest on October 28, 2014. For personal use only. 1994 84: 1314-1319 Activation of the coagulation cascade after infusion of a factor XI concentrate in congenitally deficient patients PM Mannucci, KA Bauer, E Santagostino, E Faioni, S Barzegar, R Coppola and RD Rosenberg Updated information and services can be found at: http://www.bloodjournal.org/content/84/4/1314.full.html Articles on similar topics can be found in the following Blood collections Information about reproducing this article in parts or in its entirety may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests Information about ordering reprints may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#reprints Information about subscriptions and ASH membership may be found online at: http://www.bloodjournal.org/site/subscriptions/index.xhtml Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036. Copyright 2011 by The American Society of Hematology; all rights reserved. From www.bloodjournal.org by guest on October 28, 2014. For personal use only. Activation of the Coagulation Cascade After Infusion of a Factor Concentrate in Congenitally Deficient Patients XI By Pier Mannuccio Mannucci, Kenneth A. Bauer, Elena Santagostino, Elena Faioni, Samad Barzegar, Raffaella Coppola, and Robert D. Rosenberg Virally inactivated, high-purity factor XI concentrates are available for treatment of patients with factor XI deficiency. However, preliminary experience indicates that some preparations may be thrombogenic. We evaluated whether a highly purified concentrate produced signs of activation of the coagulation cascade in two patients with severe factor XI deficiency infused before and after surgery.Signs of heightened enzymatic activity of the common pathway of coagulation (elevated plasma levels of prothrombin fragment 1+2 and fibrinopeptide A) developed in theearly postinfusion period, accompanied by more delayed signs offibrin formation with secondary hyperfibrinolysis (elevated D-dimer and plasmin-antiplasmin complex). These changes occurred in both patients, but were more severe in the older patient with breast cancer when she underwent surgery, being accompanied by fibrinogen and platelet consumption. There were no concomitant signs of heightened activity of the factor VII-tissue factor mechanism on the factor Xase complex (plasma levels of activated factor VI1 and of factor IX and X activation peptides did not increase).The observed changes in biochemical markers of coagulation activation indicate that concentrate infusions increased thrombin generation and activity and that such changes were magnified by malignancy and surgery. Because somefactor XI concentrates may bethrombogenic, they should be used with caution, especially in patients with other risk factors for thrombosis. 0 1994 by The American Society of Hematology. P heightened factor Xa and thrombin enzymatic activity accompanied by secondary fibrinolysis. In one patient with breast cancer, the laboratory abnormalities were more severe, particularly during and after surgery when signs of fibrinogen and platelet consumption developed. ATIENTS WITH congenital factor XIdeficiency are usually treated with fresh-frozen plasma, which contains the protein lacking in these patients and necessary for hemostasis. Large amounts of plasma, suchas those that may be repeatedly needed after major surgical procedures, may lead to volume overload. Most importantly, plasma carries a risk of transmitting bloodborne viral infections. To circumvent these problems, virally inactivated plasma concentrates, purified with chromatographic techniques to contain almost exclusively factor XI, have been developed and used clinically, particularly during and after surgical procedures.’.* However, a preliminary report has shown that in two patients concentrate infusions were accompanied by the occurrence of laboratory signs of intravascular coagulation with secondary fibrinolysis, expressed as a postinfusion increase in fibrinogen-fibrin degradation products and D-dimer.3 In one patient, who died after coronary artery bypass surgery, all the grafts were found during emergency thoracotomy to be occluded by thrombi.3 Prompted by this report indicating that some factor XI concentrates may be thrombogenic, we have monitored the hemostatic systems of two patients with congenital factor XI deficiency during treatment with large doses of a factor XI concentrate.’ Both patients developed laboratory signs of pathologic activation of coagulation after the infusion, with From the Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Institute of Internal Medicine, IRCCS Maggiore Hospital and University of Milan, Italy; the Charles A. Dana Research Institute and the Harward-Thorndike Laboratory, Department of Medicine, Beth Israel Hospital and Harvard Medical School, Boston, MA; and the Department of Biology and Whitaker College, Massachussetts Institute of Technology, Cambridge, MA. Submitted October 27, 1993; accepted April 22,1994. Address reprint requests to Pier M. Mannucci, MD, Via Pace 9, 20122 Milano, Italy. The publication costsof this article were defrayedin part by page chargepayment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 1994 by The American Society of Hematology. 0006-497//94/8404-0037$3.00/0 1314 PATIENTS AND METHODS Patients. Patient no. I was a 19-year-old womanwith severe factor XI deficiency ( < l U/dL), expressed clinically as prolonged bleeding after dental extractions and superficial cuts. She was referred for the surgical removal of a large pilonidal cyst of the anus. Patient no. 2 was a 69-year-old woman with severe factor XI deficiency (< 1 U/dL), expressed as prolonged bleeding after appendectomy and tooth extractions, which on one occasion required transfusions of whole blood. She was referred with a diagnosis of breast adenocarcinoma for total mastectomyand axillary lymphadenectomy. The patient was in good clinical condition and there was no evidence of metastatic disease by computed tomography scans or bone scintigraphy. Neither of thetwo patients was anti-human immunodeficiency virus seropositive, nor was there clinical or serologic evidence of previous infection with the hepatitis B and C viruses. Measurements and methods. Factor XI functional activity was measured in plasma using the microplate assay described by Scott and C ~ l m a n In . ~ brief, plasma factor XI was activated by Kaolin, in the presence of soybean trypsin inhibitor to inhibit formation of Kallikrein and in the presence of corn trypsin inhibitor to inhibit activated factor XII. Cleavage of the chromogenic substrate S-2366 (Chromogenix, Molndal, Sweden) was then evaluated by measuring absorbance at 405 nm. The between-assay coefficient of variation of this method was 7%, which is lower than that of factor XI assays based on the activated partial thromboplastin time (15% to 20%). Activated factor XIwas measured in the concentrate withan amidolytic method, using as a reference standard a purified preparation of activated human factor XI (Enzyme Research Laboratories. South Bend, IN). Aliquots of different dilutions of activated factor XI were incubated at 37°C with S-2366. Absorbances were read at 405 nm over 10 minutes and plotted against concentrations of activated factor XI on a logarithmic scale. The factor XI concentrate was treated in the same way, and concentrations of activated factor XI were read on the standard curve and expressed as a percentage of factor XI by weight. Measurements were performed in triplicate and the results were averaged. Samples of reconstituted concentrate were electrophoresed on SOdium dodecyl sulfate ( l % ) polyacrylamide gel (10%) (SDS-PAGE) under reducing and nonreducing conditions and protein bands were Blood, Vol 84, NO 4 (August 15), 1994: pp 1314-1319 From www.bloodjournal.org by guest on October 28, 2014. For personal use only. ACTIVATION COAGULATION FACTOR AFTER XI CONCENTRATE shown by silver staining. Purified preparations of antithrombin 111, factor XI, and activated factor XI (Enzyme Research Laboratories) were run on the same gel. The factor IX activation peptide (FMP; the product of the enzymatic action on factor M of factor VIWIIa-tissue factor or of factor XIa) and the factor X activation peptide (FXP; the product of the enzymatic action on factor X of factor VIWIIa-tissue factor or of the factor Ma-VIIIa-activated surface complex) were measured by double-antibody radioimmunoassay (RIA) in plasma stored at -80°C until assay and extracted to eliminate nonspecific effects from other plasma proteins.5,6The plasma half-lives of these moieties are 15 and 30 minutes, respectively?.6 Prothrombin fragment 1+2 (F1+2), a measure of the enzymatic action of factor Xa on prothrombin with a plasma half-life of 90 minutes, was measured on stored plasma by double-antibody RIA.' Fibrinopeptide A (FPA), a measure of the enzymatic action of thrombin on fibrinogen with a plasma half-life of 3 to 5 minutes, was measured by enzyme immunoassay (Stago, Asnieres, France) in plasma extracted twice with bentonite to remove fibrinogen. The plasma concentrations of FXP and FIXP are given in picomoles and those of FPA and F1 + 2 are given in nanomoles. For these methods, venous blood was collected into an anticoagulant mixture containing aprotinin, a synthetic thrombin inhibitor, and EDTA (purchased from Byk-Sangtec, Dietsenbach, Germany). Factor VIIa, the two-chain enzymatic form of factor VI1 with a plasma half-life of 180 minutes, was measured in citrated plasma by a onestage, prothrombin time-based assay using a truncated and soluble form of recombinant tissue factor that, on relipidation, reacts with factor VIIa but not with one-chain factor VII.* The normal laboratory ranges for these measurements (see Table 1) are based on the calculation of log means -C SD. We also measured fibrinogen, antithrombin 111, and platelets by previously published standard methods'; D-dimer and plasmin-antiplasmin complex were measured by enzyme immunoassays (Dimertest; Ortho Diagnostic System, Milan, Italy; Behring Institute, Scoppito, Italy). Concentrate. The factor XI concentrate, not licensed in Italy, was provided by the Centre Regional de Transfusion Sanguine (Lille, France; courtesy of Prof M. Goudemand) as a single lot (30010226) in vials containing approximately 1 0 0 W/mL factor XI after reconstitution in 10 mL of distilled water. The concentrate was produced from cryoprecipitate-poor pooled plasma obtained from whole blood donations or by plasmapheresis of unpaid donors. A detailed description of the manufacturing process has been published by BurnoufRadosevich and Burnouf.' In brief, the process involves filtration on a negatively charged cellulose filter followed by chromatography on a cation exchange resin. A solvent-detergent mixture [0.3% tri(-nbutyl) phosphate and 1% Tween 801 was added before chromatography to inactivate lipid-enveloped viruses." The concentrate has a high specific activity (120 to 160 IU of factor XI per milligram of protein) and before lyophilization pasteurized antithrombin 111 and heparin are added, both at final concentrations of 2 U/mL of reconstituted concentrate.' According to the manufacturer, the concentrate lot used in this study contained no measurable levels of activated factors XI, M, X, thrombin, and plasmin.' Activated factor XI was also measured by us and was found to be present in small amounts (0.003% wt/wt of factor XI). On SDS-PAGE performed under reducing and nonreducing conditions, the only bands shown by silver staining corresponded in electrophoretic mobility to purified factor XI and antithrombin I11 run in the same gel, as previously demonstrated by manufacturers.*We detected a small amount of prothrombin F1 +2 in the concentrate (0.87 pmol/L). No thrombogenic activity was found in a stasis model after infusing rabbits with concentrate doses of 900 to 1,100 U of factor XI per killogram of body weight. The only lot used in this study was not thrombogenic after infusion in rats at a dose of 300 Ukg, in contrast with a prothrombin complex 1315 d M m ? c m m m o w O W ( D W J N P o x N N N N N U p ', a 2 2 P- .-E; m- .g P $ N N w m d 5 m m m m m I "!"!$~ I 2 r d, m m 5 3 L E 2 2, cq 7 x 85 U .-C E 3 2s .-P O, 0 "C P) c .E 8 h e c S 2 b -- I a ti 5 s U N x ;i &3 N N ~ F m 2m I 0 - l . ( ~ m m w .E g, d m m o c o 22 U. .-S m o m o m r r o d m - t m i < g N ~ N - e42 + o cLrINl.0.- EE 0 0 0 N - O - m m ~ 2 J 19'4797 2 c qmq9cqk $2 u E- a! - 9 ? 2 6 r m" F Sc .-c P) L L s -0 : ge c Q -I C E €1 w m m m m zzzzz 0 9 52 I N N c c m 2 0 E 1 m o m w V I 0 2 2 0 " z2Mz:A 0 C $x m c? S C m w- B< .E F n ._Y z mmm ~m -mt c l . R- c v m m m m m m W xz r 2 P) &$ $ g ~ ~ - t - w m m 0m z m - t m c - -"c %2 0 (D U C .- 30 c .E + a r C o'Ef;=$ ' 5 0 m u) -P 3 m 5 n m .- a v) F $7 z $ $ W E c 2 .g z n - 2-0 c ,P .-e U m m - L Y - t w m z g E Z From www.bloodjournal.org by guest on October 28, 2014. For personal use only. 1316 MANNUCCI ET AL concentrate, which at the same dosage generated more than 600 ng/ mL of FPA (unpublished data from the manufacturers). CASE REPORTS Patientno. 1. The first concentrate infusion (50 U k g ) was administered on an out-patient basis to evaluaterecovery and half-life. The total amount of heparin infused with this dose (50 IU) is not large enough to affect coagulation per se. Table 1 shows that factor XI levels increased to greater than 100 U/dL in the first postinfusion day, with increments of factor XI of 1.9 U/dL per U k g administered. Factor XI declined slowly in the next 10 days (Table l), with a halflife of 61 hours, but8 U/dL couldstill be measuredin plasma 14 days after the infusion (Table 1). A second concentrate infusion was then administered (50 U k g ) and surgical removal of the cystwas performed under localanesthesia. At 6 hours postinfusion, factor XI was 138 U/dL, with an increment of 2.4 U/dL per U k g administered. Although the patient did not bleed abnormally during or after surgery, a supplementary dose of factor XI was administered on postoperativeday 4(day 19fromthe first infusion)tokeep factor XI levels above 50 U/dL (preinfusion levels, 62 U/ dL; postinfusion levels, 120 U/dL). The wound healed uneventfully and thepatient was discharged on the postoperative day 7. Table 1 also shows that, after the first infusion of factor XI, there was no important change in routine measurements of hemostasis (platelets,fibrinogen,andantithrombin 111). However,whenmoresensitivemarkers were measured, there were highpostinfusion values of D-dimer from normal preinfusion levels (41 to 319 ng/mL), plasmin-antiplasmin complex (258 to 385 ng/mL), and F1+ 2 (0.61 to 2.07 nmol/ L). There were also high postinfusion values of FPA (up to 3.6 nmoVL), althoughno preinfusionvalue had been obtained. The increase of FPA and F1+2 occurredalready 1 hour after the infusion, whereas those of the remaining measurements were more delayed (at 2 to 6 hours). There were no changes in FXP, FIXP, and factor VIIa, which remained normal. After the second and third infusion, we monitored only fibrinogen, antithrombin 111, and platelets, all of which remained normal (data not shown). Patient no. 2. This patient received the first concentrate infusion 12 days before surgery. As in patient no. 1, factor XI increased to more than 1 0 0 U/dL in the earlypostinfusion period (with an increase of 2.1 U/dL per U k g infused) and declined slowly over thenext 12 days, with a half-life of 62 hours. There were slight and delayed decreases in platelet 154 X 109/L) and plasmafibrinogen count(from209to (from 355 to 289 mg/dL)4 to at 6 hours, but not inantithrombin 111 (Table 2). On the postinfusion day 13, when factor XI levels were still 10 U/dL, the patient received a second infusion (50 U k g ) and underwent mastectomy and lymphadenectomy. Histologic examination confirmed the presence of breast adenocarcinoma, with no metastatic involvement of axillary lymphnodes. With a peak factor XI level of 159 U/dL attained in the early postinfusion period, the patient did not bleed abnormally at the time of surgery. However, there were clear signs of consumption coagulopathy during the day of operation, seen as a progressive decline of fibrinogen from 371 to 143 mg/dL and of platelets from 190 to 130 X 10y/L over the first X postoperative hours, with little change in antithrombin 111. Although there was some oozing from the drainage during the postoperative period,it was only on postoperative day 6 (day 19 after the first infusion) that the hematocrit decreasedto 22% and alarge wound hematoma developed. Because plasma factor XI was 16 U/ dL, with a normal platelet count and fibrinogen levels ( I 88 X IO’/L and 402 mg/dL),bleeding was thought to bemainly caused by low factor XI levels rather than by consumption of coagulopathy, so thatpatientreceivedathirdinfusion factor XI concentrate and 2 U of packed red blood cells. In the first postinfusion hours there were marked decreases in fibrinogen (to a nadir of 187 mg/dL at 8 hours) and platelets ( 1 12 X 109/Lat 12 hours), with little change in antithrombin 111. In the next few days, there was no further decrease in hematocrit, the wound hematoma absorbed slowly, and the patient was discharged 20 days after surgery. At all times, renal and liver functions (as assessed by measuring plasma creatinine and liver function tests) remained normal. The results for the more sensitive molecular markers of coagulation activation are shown in Table 2. After the first test infusion, there were clear signsof heightened factor Xa and thrombin activity (increasing levels of FPA and F1 +2) and of secondary hyperfibrinolysis (increasing levels of Ddimer and plasmin-antiplasmin complex). FPA and F1 +2 peaked earlier (at 1 hour) than the remaining measurements (at 4 to 6 hours). The same pattern of changes was seenafter the second infusion,administered atthetime of surgery, and after the third infusion, administered at the time of the hematocrit decrease. At the time of the latter infusion, however,some values were alreadymarkedlyelevatedbefore infusion (FPA, D-dimer, plasmin-antiplasmin complex, and, to alesser extent, F1+2). At no time were there any important changesintheplasmalevels of FIXP and factor VIIa, which remained normal. There were slightly elevated values of FXP in some samples, with no clearpattern of change in relation to infusions and surgery (Table 2). DISCUSSION The infusion into two patients with congenital factor X1 deficiency of large doses of factor XI concentrate was followed by the development of signs of heightened activity of the enzymes of the common pathway of blood coagulation, accompanied by signs of fibrin formationand secondary fibrinolysis. In patient no. 1, heightened enzymatic activity was discrete, developedprogressively throughout theearly postinfusion period, and could only be detected by sensitive markers such as F1 +2, FPA, and D-dimer, with no signs of consumption coagulopathy (normal fibrinogen, antithrombin 111, and platelets). The patient was young andhealthy, with no underlying condition predisposing to hypercoagulability, and developed the abnormalities after a test concentrate infusion administered on anout-patientbasis,when shewas fully mobile and considerably before the planned surgical procedure. Exogenous F1 +2 in the concentrate cannot explainthe postinfusion increase of this peptide in plasma, because a From www.bloodjournal.org by guest on October 28, 2014. For personal use only. COAGULATIONACTIVATION AFTERFACTOR 1317 XICONCENTRATE d z2:zz:: m m m m m ~ N N N N " U r - m - r - m m m - o * o ? - N ~ O m m l n O * W N r - m ~ l m - m m c n w r - m m m ;DP - xL N N N . - " N N N N ' m a 2 F L > C O .g m - : N : ? Im222m:::m - r m m ~ g g 2 z $ $ 4 g ~ z z c,.-=O)=O) n .? v)*r-P-r.r(O " " " " " F & m -m 72 Wo????eq m E 2 Or m m m w "qrn-?"?? ~ ( . ~ N ~ N ?o?9"9"999""$ N e ~ N e e m ~ ~ m m m o? m m m N U C ._ € 2 2& Q " c $ 3 ,iE 'Z Q U c U c 1 v) v) v) H LL c W. c & 2 O z€ E4 U 5 e 3 - -. W ~ N N Lap wx mm?? ~ v)*(DrPlr- N N . 'i/ - ZLgs - a-mm zm 9 ???9? . - - - ( D - k € - m * m r W W ~ ~ N - . ~ ) ~ ) ~ o N ~ W 7 7 ? o ? 7 9 " - - r - m * r - r N W * * W dui N N N 0 - N N N r N N 7 O ) o N N d " " " " ? 9 y ? X I ?FoDl..** v)N x ; 2 C E z: .52 m m m 0 N N - m N O ) * m m o O ) w zzO)zzz2z o ) ~ o ) o ) o ) ~ o ) ;;%Es;ggEE%y m o ~ c ~m m b o m m m e m m m ~ m v ) m m ~ F 2 N P c .-Q U 0 E 2 o %S 25 - ) ~ o o szlq2:zzz 7 N * m ENsEEc2 O ) O " r " " N W N r ~ ~ W l NNg F n L 9 (I) 3 3 gc c * A C v) g2 ,gG SJ E vm )mc - or -- ~w ) m g* *de c v * o ~ ~ ~ l m w - * r - + m m m m ~ m m m - r - m o ~ l m r - o P l r W - Y ) e : N m m N m - N -" (N"" n7 oO oO vO )) oW mU oY g $m 2w ~ om o m e mo * ~ ~ mo v m ) ~w O) Sfi%%NZZR%ZRm 2 U - x- 0 b? g 2 "" ~m o rawn -m r n z 0 W Y hi Q 2 ._ S .-C 20 .-C .-C E z z c s c r E z z x c r c .E ._ 1+ c O E - N e r n eN 0 m 7 ~ m w E c z c r x r r r o ~ - m ~ m 2N dd e 0 m N 6 L m F a e CI) e 3 - D 3 ._ g $ 2 .E 2 c U).- U) .g -I3 m f -$ 3 n o Ep 20) e .-U L Ei3 m $ 2 !. .-0) 1 Q v1 C, W .c % S ..: m C, .-m ua) a + 2 0 .- z .E e -,&-, m c m r m 'f: b .-E : U)P Ea Xo .? Q2 .r l- N o ) o -N.N.N.-.LO~~ """ r -CV- 3 ~ m m m c u ~ ~ e o m w ? ~ ~ $ x x ?z ' 0~0 ? " " ? 9 " 7 2 q? " N qNr o r c Wu l ? 4 % o "" C" + o L f I m a m q q - - q a m m m gj Q c e W- *m".- h Ng .q 03 N . 0- -(om S 2 b '9 z2'5;gggg s zm sm 4e. g a em. r r eo No W m m rv N) F n - 0 - W O ) z N N N N b z From www.bloodjournal.org by guest on October 28, 2014. For personal use only. 1318 small amount of F1+2 wasmeasuredin theconcentrate (0.87 pmoVL) and the F1+ 2 peak occurred relatively late in the postinfusion period (at 2 hours), suggesting endogenous generation triggered by the concentrate itself. Patient no. 2 was much older than patient no. I and had localized breast cancer, two conditions predisposing to hypercoagulability. The elevations of FPA, F1+2, D-dimer, and plasmin-antiplasmin complex that occurred afterfirst the concentrate infusion, administered on an out-patient basis, were much more marked than those elicited in patient no. 1 by thesameconcentratedoseand were accompanied by signs of platelet and fibrinogen consumption comparable to those seen in patientswith disseminated intravascular coagulation. It is unlikely thatincreases of hypercoagulability markers were caused by impaired clearance from plasma, because renal and hepatic functions remained normal. It appears, therefore, that in the patient pre-existing hypercoagulability caused by malignancyinteracted with the concentrate infusion to increase theseverity of the hemostatic alterations. This is further illustrated by the occurrence of even more marked and persistent alterations after the two subsequent infusions, administeredin connection with major surgery, ie, another thrombogenic stimulus. The alterations of the coagulation cascade seen were characterized by the delayed appearance,at 1 to 2 hours postinfusion, of signs of heightened thrombin activity (expressed as elevated FPA) and heightened factor Xa activity (elevated F1 +2), with resolution beginning by 4 to 6 hours. There was not heightened factor X activation, as demonstrated by small and inconsistent changes of factor X activation peptide. A similar pattern of ITA and F1+2 increases has previously been observed in patients with hemophilia B infused with potentially thrombogenic prothrombin complexconcentrates,butin themtherewasalso a delayedincrease of the factor X activation peptide." There are several possible mechanisms to explain the hypercoagulable state observed in ourpatientswithfactorXI deficiency. Giventhatthe product infused was a factor XI concentrate,activated factor XI might be expected to bea trigger for hypercoagulability. We have directly measured factor XIa in the concentrate and found this enzyme to be present in small amounts (0.003% of the zymogen). These may not be sufficient to activate massively factor IX, because the factor IX activation peptide (ie,theproduct of the enzymatic action of factor XIa on factor IX) was normal in both patients postinfusion. Massive contamination of the concentrate with activated factor IX is also unlikely, because this enzymatic activity was unmeasurable according to the concentrate manufacturer and the postinfusion plasma levels of factor Xactivation peptide (the product of the enzymatic action of factor IXa on factor X) measured by us werenormal or only modestly and inconsistently elevated. Because factor VIIa was also normal, the mechanisms by which the concentrate caused the activation of the last phaseof coagulation remainunclear, even though the assaysused may be too insensitive to exclude that activation proceeded through the generation of trace quantities of activated factors IX or X. Signs of secondary fibrinolysis and thrombotic complications have been observed by Gitel et al' after the infusion of MANNUCCI ET AL a factor XI concentrate produced by a different manufacturer. Because the method of preparation of that concentrate has not been reported and its production has been discontinued, we cannot fully compare the characteristics of the two concentrates. However, weknow that thatconcentrate was much less pure than the one we used (specific activity, 7.5 v 120 to 160 U of factor XI per milligram of protein) and that it exhibited proteolytic activity against S-2238, a chromogenic substrate sensitive to the enzymatic activity of thrombin.' No such activity could be detected in our concentrate.' The biochemical characteristics of another factor XIconcentrate, produced by a British manufacturer, are knownin detail,' but there is no study of the postinfusion behavior of coagulation activationmarkers.' The main biochemicaldifference between the British product and our product is that the former contains a more than 50-fold molar excess of antithrombin 111over factor XI, meant to inactivate any coagulation serine protease that may be generated during concentrate manufacture.' Even though the clinical record of the British product indicates safety in humans,' some batcheselicited significant thrombogenic responses, similar in magnitude to those seen after prothrombin complex concentrates, in rabbits infused with doses 2 to 4 times the highest dose recommended for humans (50 Ukg).'' Following these observations, the manufacturer has recently adopted the precautionary measure of adding approximately 10 IU/mL of heparin to the concentrate to accelerate inhibition of coagulation proteases by antithrombin" (theconcentratewe used contains a smaller amount of heparin, 2 IU/mL). On thebasis of theseresults, we recommend that manufacturersperform furtherbiochemicaland animal studiesto identify thrombogenic component(s) that may be present in some factor XI concentrates, and try to to eliminateit. Meanwhile, concentrates should be used only when repeated infusions of fresh-frozenplasma might cause significant fluid overload. Plasma virally inactivated with solvent/detergent'' or paste~rization'~ is now availableand should be considered as an alternative treatment. When concentrates are needed, it is advisable to monitorpatients for clinical and laboratory signs of coagulation activation, particularly patients undergoing surgery or carrying other risk factors for thrombosis. ACKNOWLEDGMENT We thank Dr Myriana Burnouf-Radosevich (Lille, France) for providing information on the concentrate and for stimulating discussions. REFERENCES 1 . Bolton-Maggs PHB, Wensley RT. Kernoff PBA, Kasper CK, Winkelman L, Lane RS, Smith J K Production and therapeutic use of a factor XI concentrate from plasma. Thromb Haemost 67:314, 1992 2. Burnouf-Radosevich M, Burnouf T: A therapeutic, highly purified factor XI concentrate from human plasma. Transfusion 32:861, 1992 3 . Gitel SN, Varon D, Schulman S , Martinowitz U: Clinical experiences of a FXI-concentrate:Possible side effects.Thromb Haemost 65:1157, 1991 (abstr) 4. Scott CF, Colman RW: A simple and accurate microplate assay From www.bloodjournal.org by guest on October 28, 2014. For personal use only. COAGULATION ACTIVATION AFTERFACTOR XI CONCENTRATE for the determination of factor XI in plasma. J Lab Clin Med 1 1 1:708, 1989 5. Bauer KA, Kass BL, ten Cate H, Bednareck MA, Hawiger JJ, Rosenberg RD: Detection of factor X activation in humans. Blood 74:2007, 1989 6. Bauer KA, Kass BL, ten Cate H, Hawiger JJ, Rosenberg RD: Factor M is activated in vivo by the tissue factor mechanism. Blood 76:731, 1990 7. Teitel JM, Bauer KA, Lau HK, Rosenberg RD: Studies of the prothrombin activation pathway utilizing radioimmunoassays for the F2/Fl+2 fragment and thrombin-antithrombin complex. Blood 59:1086, 1982 8. Wildgoose P, Nemerson Y, Hansen LL, Nielsen FE, Glazer S, Hedner U: Measurement of basal levels of factor VIIa in hemophilia A and B patients. Blood 80:25, 1992 9. Horowitz B, Wiebe ME, Lippin A, Stryker MH: Inactivation of viruses in labile blood derivatives. I. Disruption of lipid eveloped 1319 viruses by tri(n-butyl) phosphate detergent combinations. Transfusion 25:516, 1985 10. Mannucci PM, Bauer KA, Gringeri A, Santagostino E, Bottasso B, Barzegar S , Rosenberg RD: Thrombin generation is not increased in the blood of hemophilia B patients after infusion of a purified factor IX concentrate. Blood 76:2540, 1990 11. Mannucci PM, Bauer KA, Gringeri A, Barzegar S, Santagostino E, Tradati FC, Rosenberg RD: No activation of the common pathway of the coagulation cascade after a highly purified factor IX concentrate. Br J Haematol 79:606, 1991 12. Winkelman L, McLaughlin LF, Gray E, Thomas S: Heattreated factor XI concentrate: Evaluation of in vivo thrombogenicity in two animal models. Thromb Haemost 69:1286, 1993 (abstr) 13. Horowitz B, Bonomo R, Prince AM, Chin SN, Brotman B, Shulman RW: SolvenUdetergent-treated plasma: A virus inactivated substitute for fresh-frozen plasma. Blood 79:826, 1992 14. Bumouf-Radosevich M, Bumouf T, Huart JJ: A pasteurized therapeutic plasma. Infusionstherapie 19:91, 1992