Surface passivation by human albumin of plasmapheresis circuits reduces platelet accumulation and thrombus formation. Experimental and clinical studies Juliette N. Mulvihill INSERM U.311, Biologie et Pharmacologie des Interactions du Sang auec les Vaisseaux et les Biomatlriaux, Centre Rlgional de Transfusion Sanguine, 10 rue Spielmann, 67085 Strasbourg Ckdex. France Albert Faradji and Francis Oberling Service d'Onco-Hlmatologie, Hhpital de Hautepierre, 67098, Strasbourg Cldex, France Jean-Pierre Cazenave INSERM U.311, Biologie et Pharmacologie des Interactions did Sang avec les Vaisseaux et les Biomatiriaux, Centre Rigional de Transfusion Sanguine, 10 rue Spielmann, 67085 Strasbourg Cldex, France The contact of flowing blood with an artificial surface leads to adsorption of plasma proteins, followed by platelet adhesion and aggregation and thrombus formation. This phenomenon is enhanced by turbulent flow at joints, bifurcations, and constrictions. In therapeutic plasmapheresis using an IBM blood cell separator, blockage of the extracorporeal circulation system by platelet-fibrin thrombi imposed a halt in treatment for manual clearance of the circuit for 66 in 149 cases (44%). Thus it was decided to passivate the surface of the extracorporeal circuit by filling the tubing with 4% human serum albumin 15-
20 min before the treatment session and then displacing the albumin solution with the patient's blood without creating an air-liquid interface. After introduction of this technique, a blockage was observed for only 11 in 239 cases (5%). In nitro measurements of platelet accumulation on the internal surface of the circulation system were carried out using washed human platelets labeled with ll'In-oxine in the presence of a 40% hematocrit. Preadsorption of the surface with albumin reduced platelet deposition to 4-5% that observed for an equivalent pretreatment with physiological saline.
INTRODUCTION
Thrombus formation and platelet-fibrin embolization following contact of flowing blood with an artificial surface are major problems in the development of extracorporeal circulation systems such as those used in plasmaPart of this work was presented at the 8emeCongres International sur la C.E.C., Paris, 23-25 June 1983. Journal of Biomedical Materials Research, Vol. 24, 155-163 (1990) 0 1990 John Wiley & Sons, Inc. CCC 0021-9304/90/020155-09$04.00
MULVIHILL ET AL.
156
pheresis. When blood enters into contact with a foreign surface, a series of events is initiated': (i) rapid adsorption at the surface of a layer of blood proteins, (ii) platelet and leukocyte adhesion to the protein layer with formation of platelet thrombi, (iii) activation of the coagulation system leading to thrombin generation and fibrin formation. Hemodynamic conditions influence the size, location, and structure of the resulting thrombus and its fragmentation. Under the high flow conditions of plasmapheresis systems, platelet accumulation normally predominates over the formation of a fibrin clot. Turbulent flow at joints, filters, constrictions and bifurcations may lead to the development of gross thrombotic deposits with eventual blockage of the circulation system. In therapeutic plasmapheresis using an IBM blood cell separator, we had observed, in spite of anticoagulation with acid-citrate-dextrose (ACD) and heparin and sometimes administration of the antiplatelet drug dipyridamole, the formation of platelet aggregates at joints, filters and the entry needle of the extracorporeal circulation system. Blockages often necessitated a halt to replace or clear part of the circuit. In consequence, the session was frequently shortened and the plasma exchange volume reduced. In order to improve the thromboresistance of the extracorporeal system, experimental and clinical studies of the effects of protein adsorption were undertaken. Since it has been shown that preadsorption of albumin to an artificial surface reduces platelet acc~mulation,~-~ we decided to passivate the surfaces of the IBM extracorporeal circuit by precoating it with human albumin prior to passage of the patient's blood.
PATIENTS AND METHODS
Clinical studies
Putien ts During a one-year period, 52 consecutive patients underwent 388 therapeutic plasmaphereses for various disorders (Table I). The effect of the preadsorption of human albumin was evaluated in the following way. From January to June 1982, the presence of documented platelet-fibrin thrombi in the system (fall in pressure in the IBM blood cell separator causing arrest of the session and macroscopic examination of the circuit) was retrospectively recorded from the clinical protocols of 149 consecutive apheresis sessions. In July 1982, it was decided to examine prospectively the effect of the preadsorption of human albumin on the formation of thrombi in the extracorporeal system. The study was conducted for 239 consecutive sessions during an equal 6-month period until December 1982.
157
SURFACE PASSIVATION BY ALBUMIN IN PLASMAPHERESIS
TABLE I Clinical Diagnosis in 52 Patients Submitted to Therapeutic Plasmapheresis During a One-Year Period Number of Sessions of Plasmapheresis -
Precoating of the Surface Clinical Diagnosis Myasthenia Rheumatoid arthritis Systemic lupus erythematosus Periarteritis nodosa Polymyositis Sclerodermia Chronic polyradiculoneuritis Multiple sclerosis Behcet’s disease Immune complex uveitis Immune complex glomerulonephritis Myeloma Waldenstrom’s macroglobulinemia Cryoglobulinemia Porphyria Total
Number of Patients 3 6
3 4 2 1 1 4 2 2 9 4
8 2 1 52
No Albumin (January to June 1982)
Albumin (July to December 1982)
1 25 14 9 12 14 3 -
9 29 4 10 13 48 22 23 38 18 19 12 6 239
-
42 12 15 149
Plasmapheresis At the beginning of each session, with or without albumin preadsorption, the extracorporeal circuit was filled with physiological saline. The preadsorption of albumin was carried out by displacing the saline solution with 4% pasteurized human serum albumin (Centre R6gional de Transfusion Sanguine, Strasbourg, France, purity greater than 98%). The albumin remained in the circuit, in static at ambient temperature, for an adsorption time of 15-20 min, after which it was displaced by the patient’s blood. Care was taken at all times to avoid the creation of an air-liquid interface. Plasma exchange was done by continuous flow centrifugation using an IBM 2997 blood cell separator and apheresis kits fabricated from polyvinylchloride (PVC) tubing. The length of the session varied from 1 to 2 h, the time to replace one plasma volume (40 mL/kg) by an equal volume of 4% human serum albumin. The total blood flow was in the range 1520 mL/min and the centrifugation speed 1800 rpm (550g). Before returning to the patient, the mixture of blood cells and replacement solution passed through a filter and a blood warmer at 37°C. Anticoagulant ACD (trisodium citrate dihydrate 17.8 g, citric acid monohydrate 0.3 g, anhydrous glucose 30.0 g/liter) in the proportions of 1 :9 to 1:10 and heparin (Roche, France) at
158
MULVIHILL ET AL.
20-30 U/min were infused into the extracorporeal circuit 10 cm after the entry needle.
Experimental studies To determine the effect of adsorbed albumin on platelet-surface interactions, in vitro experiments were carried out to quantitate the accumulation of platelets in the IBM circulation system.
Preparation and labeling of washed human platelets Washed human platelets were prepared according to the technique of Mustard with minor modifications6and labeled with lllIn-oxine (Amersham, France) as described by Eber et al.7 The labeled platelets were resuspended in Tyrode’s buffer (containing 2 mM Cazt, 1mM Mg2+,0.35% human serum albumin, 5 mM Hepes and 10 pl/ml apyrase, reference6)in the presence of washed red blood cells’ at 40% hematocrit.
Coating of the plusmapheresis circuit with human albumin Segments of PVC tubing (internal diameter 3 mm) of about 20-cm length from an apheresis kit were connected in parallel to a four-circuit peristaltic pump (Sage 375A, Orion Research Inc., Canada). The tubing was filled either with physiological saline for controls or with 4% human serum albumin, left at ambient temperature for 30 min and rinsed with physiological saline.
Measurement of platelet accumulation The platelet suspension was introduced into the segments of pretreated tubing by displacement of the saline rinse solution and circulated for 30 min at a flow rate of 13.3 mL/min from a reservoir thermostated at 37°C. All manipulation was conducted so as to avoid the occurrence of an air-liquid interface. After rinsing with physiological saline, the tubing was emptied, dried in an air stream, and cut into 2-cm segments for radioactive counting in a gamma-counter (1282 Compugamma, LKB, Turku, Finland). Platelet deposition on the internal surface of the segments was calculated according to the radioactivity per unit surface area and the specific activity of the platelets as described previously.’
159
SURFACE PASSIVATION BY ALBUMIN IN PLASMAPHERESIS RESULTS
Clinical studies The distribution of the 52 patients submitted to repeated therapeutic plasmapheresis during a one-year period is shown in Table I. The rationale for the treatment was to remove from the plasma of the patient either autoantibodies and immune complexes or excessive amounts of monoclonal antibodies. Most of the patients treated fall into these two categories. During the first half of 1982, 149 plasmapheresis sessions were conducted without precoating of the surface of the plasmapheresis circuit. During the second half of 1982, the same or new patients were plasmapheresed (239 sessions) after precoating of the circuit with human albumin. The clinical results of 388 consecutive plasmapheresis sessions conducted from January to December 1982 are shown in Table 11. During the first 6 months, when albumin pretreatment of the surface of the extracorporeal circuit was not employed, 66 blockages due to the presence of platelet-fibrin thrombi were recorded for a total of 149 sessions (44.3%).During the following 6 months, the extracorporeal circuit was precoated with 4% human albumin. A total of 239 sessions were carried out, in which 11 cases of blockage were observed (4.6%).
Experimental studies Since the thrombi obstructing the extracorporeal circuit were essentially composed of platelets, the effects on platelet-surface interactions of precoating the circuit with human albumin were investigated in vitro. Deposition of "'In-labeled platelets was quantitated in three separate experiments (Table 111). Albumin pretreatment reduced the concentration of platelets adhering to the surface to about 5% the concentration observed for the untreated surface.
TABLE I1 Therapeutic plasmapheresis. Effect of Albumin Pretreatment of the Extracorporeal Circulation System Precoating of the Surface No Albumin
Total number of sessions Blockage due to platelet-fibrin aggregates
149 (100%) 66 (44.3%)
Albumin 239 (100%) 11 (4.6%)*
Pretreatment of the circuits with 4% human albumin for 15-20 min in static at ambient temperature. * p < 10-~ (x2test).
MULVIHILL ET AL.
160
TABLE 111 In Vitru Effect of Human Albumin Pretreatment of the PVC Tubing of the Plasmapheresis Circuit on the Accumulation of “In-Labeled Platelets Number of Platelets Deposited per mm2 (mean t S.D. for 10 Tube Segments) Precoating of the Surface Experiment
No Albumin
1 2 3
9,100 f 500 5,000 400 16,900 t 1100
*
Albumin 400 300 800
f
-t
f
loo* 100“ 100“
Pretreatment of the circuit with 4% human serum albumin for 30 min in static at ambient temperature. Platelet suspension 180,000/mm3 containing 40% red blood cells recirculated with a peristaltic pump for 30 min at 37°C. “ p < 0.001 (paired Student’s t-test).
DISCUSSION
One of the major problems in plasmapheresis for therapeutic reasons remains the occurrence of thrombotic complications due to blood activation by the surface of the plasmapheresis circuits, in spite of the administration of anticoagulants and antiplatelet drugs to the patient.” In this respect, our experience is in agreement with that of others. We used an IBM blood cell separator to perform therapeutic plasmapheresis in patients with immunological disorders, to remove autoantibodies, immune complexes or monoclonal antibodies. Plasmapheresis was accompanied, during the first 6 months of the study, by blockage of the circuit with platelet-fibrin thrombi in 4.4% of the sessions, despite the patient anticoagulation. This is not surprising, since it is well known that immunoglobulin or immune complex adsorption on surfaces is highly t h r o m b o g e n i ~ , ~ and ~ ~our ” ~ patients had elevated levels of autoantibodies, immune complexes or monoclonal immunoglobulins in their plasma. In addition to immunoglobulin or immune complex deposition on the surface of the plasmapheresis circuit, other mechanisms may have played a role in the formation of platelet-fibrin thrombi, e.g., fibrinogen adsorption, activation of the coagulation and complement systems, adhesion of leukocytes and release of procoagulant factors, flow abnormalities or the presence of air As the adsorption of albumin on artificial surfaces is known to prevent extensive platelet a c c ~ m u l a t i o n , we ~ ~ ~decided ~ - ~ ~ to passivate the plasmapheresis circuit by prior adsorption of human albumin. Preliminary in vifro studies with a perfused PVC tubing circuit demonstrated that pretreatment of the surface by albumin preadsorption reduced considerably the number of adherent platelets in comparison to the nontreated surface (Table 111). We then used this same method to passivate the IBM plasmapheresis circuit be-
SURFACE PASSIVATION BY ALBUMIN IN PLASMAPHERESIS
161
fore the beginning of the therapeutic session. According to the results summarized in Table 11, albumin preadsorption of the extracorporeal circulation system reduced the occurrence of blockage due to the formation of plateletfibrin aggregates from 44% to 4.6% of the total number of sessions. This technique thus minimized the cases in which the plasmapheresis session was shortened and the plasma exchange volume reduced. Since the halts in treatment to clear a blocked circuit were largely eliminated, it also enabled a considerable time gain, the average number of patients treated per day passing from two to three. It is important to note that in all our studies, clinical or experimental, the patient's blood or the platelet suspension or protein solution was introduced into the extracorporeal circuit by displacement of another solution, either albumin or physiological saline, thus avoiding the formation of an air-liquid interface. This is a crucial point, since air-liquid interfaces cause activation of platelets and the coagulation and complement systems. l6 Several approaches can be used to prevent the undesirable effects of platelet-surface interactions in plasmapheresis circuits: modification of the The adsorption of surface or pharmacological modification of platelets. albumin on surfaces markedly reduces platelet deposition, in contrast to adsorption of fibrinogen or garnma-gl~bulins.~~'~~~~-~~ In addition, the adsorbed albumin competes with immune complexes and other proteins such as fibrinogen for adsorption on the surface. Thus, we decided to passivate the plasmapheresis circuit with human albumin. We are aware that the study, for historical reasons, contains a methodological flaw in that the two clinical series were sequential in time and did not overlap, but in view of the large difference between the two groups it is unlikely to be due to a mere statistical aberration. Furthermore, the same patients were treated with and without albumin passivation and the clinical data are fully supported by the results for in vitro platelet accumulation. Diluted (4%) sterile human albumin solution is cheap and well tolerated by the patient and can be used as a replacement solution for the exchanged volume of plasma. Owing to its simplicity, the passivation of plasmapheresis circuits with albumin has now been introduced as a routine procedure in our plasmapheresis unit. For short-term use, adsorption of albumin renders the artificial surface sufficiently thromboresistant and is less expensive than bonding albumin to polymer surfaces by chemical or radiation processes.20-22 The authors are grateful for the secretarial assistance of Mrs. C. Helbourg and the technical assistance of Mrs. J. Groener and Mrs. A. Sutter-Bay.
References 1. E. W. Salzman and E. W. Merrill, "Interaction of blood with artificial surfaces," in Hemostasis and Thrombosis. Basic Principles and Clinical Practice, 2nd ed. R. W. Colman, J. Hirsh, V. J. Marder, and E. W. Salzman (Eds.), J. 8. Lippincott Co., Philadelphia, 1987, pp. 1335-1347.
MULVIHILL ET AL.
162 2. 3.
4. 5.
6.
7.
8.
9.
10.
11. 12.
13. 14. 15.
16.
17.
M.A. Packham, G. Evans, M.F. Glynn, and J.F. Mustard, "Effect of plasma proteins on interaction of platelets with glass surfaces," 1. Lab. Clin. Med., 73, 686-697 (1969). J. L. Brash, "Hydrophobic polymer surfaces and their interactions with blood," Ann. NY Acad. Sci., 283, 356-371 (1977). D. E Mosher, "Influence of proteins on platelet-surface interactions," in Interaction of the Blood with Natural and Artificial Surfaces, E. W. Salzman (Ed.), Marcel Dekker, New York, 1981, pp. 85-101. K . Park, D. F. Mosher, and S. L. Cooper, "Acute surface-induced thrombosis in the canine ex vivo model: importance of protein composition of the initial monolayer and platelet activation," J. Biomed. Muter. Xes., 20, 589-612 (1986). J.-P. Cazenave, S. Hemmendinger, A. Beretz, A. Sutter-Bay, and J. Launay, "L'agrhgation plaquettaire: outil d'investigation clinique et dktude pharmacologique. Methodologie," Ann. Biol. Clin., 41, 167-179 (1983). M. Eber, J.-P. Cazenave, J. C. Grob, J . Abecassis, and G. Mcthlin, ""lIndium labeling of human washed platelets; kinetics and in vivo sequestration sites," in Blood Cells in Nuclear Medicine, Part I. Cell Kinetics and Bio-distribution, M. R. Hardeman and Y. Najean (Eds.), Martinus Nijhoff Publishers, Boston, 1984, pp. 29-43. J.-P. Cazenave, D. Blondowska, M. Richardson, R. L. KinloughRathbone, M. A. Packham, and J. F. Mustard, "Quantitative radioisotopic measurement and scanning electron microscopic study of platelet adherence to a collagen-coated surface and to subendothelium with a rotating probe device," 1. Lab. Clin. Med., 93, 60-70 (1979). J. N. Mulvihill, H. G. Huisman, J.-P. Cazenave, J. A. van Mourik, and W. G. van Aken, "The use of monoclonal antibodies to human platelet membrane glycoprotein IIb-IIIa to quantitate platelet deposition on artificial surfaces," Thromb. Haemostas., 58, 724-731 (1987). G. P. Clagett, "Artificial devices in clinical practice," in Hemostasis and Thrombosis. Basic Principles and Clinical Practice, 2nd ed. R. W. Colman, J. Hirsh, V. J. Marder, and E. W. Salzman (Eds.), J. B. Lippincott Co., Philadelphia, 1987, pp. 1348-1365. P. M. Henson and H. L. Spiegelberg, "Release of serotonin from human platelets induced by aggregated immunoglobulins of different classes and subclasses," J. Clin. Invest., 52, 1282-1288 (1973). C . R. H. Wildevuur and M. D. Kazatchkine, "Vascular biomaterials in use," in Blood-Surface Interactions: Biological Principles Underlying Haemocompatibility with Artificial Materials, J.-P. Cazenave, J. A. Davies, M. D. Kazatchkine, and W. G. van Aken (Eds.), Elsevier, Amsterdam, 1986, pp. 197-207. T. Chandy and C. P. Sharma, "The antithrombotic effect of prostaglandin E l immobilized on albuminated polymer matrix," J. Biomed. Muter. Res., 18, 1115-1124 (1984). Y. Ishikawa, S. Sasakawa, M. Takase, and Y. Osada, "Effect of albumin immobilization by plasma polymerization on platelet reactivity," Thromb. Res., 35, 193-202 (1984). K . Kottke-Marchant, J. M. Anderson, Y. Umemura, a n d R . E . Marchant, "Effect of albumin coating on the in vitro blood compatibility of Dacron arterial prostheses," Biomaterials, 10, 147-155 (1989). C . A. Ward, "Blood protein-material interactions that lead to cellular adhesion," in Proteins at Intevfaces: Physicochemical and Biochemical Studies, J. L. Brash and T. A. Horbett (Eds.), American Chemical Society, Washington DC, 1987, pp. 551-565. C. Pusineri and J.-P. Cazenave, "Techniques for modifying artificial surfaces," in Blood-Surface Interactions: Biological Principles Underlying
SURFACE PASSWATION BY ALBUMIN IN PLASMAPHERESIS
18.
19.
20.
21.
22.
163
Haemocompatibility with Artificial Materials, J.-P. Cazenave, J. A. Davies, M. D. Kazatchkine, and W. G. van Aken (Eds.), Elsevier, Amsterdam, 1986, pp. 209-235. J. A. Davies, ”Therapy affecting the blood,” in Blood-Surface Inteructions: Biological Principles Underlying Haemocompatibility with Artificial Materials, J.-P. Cazenave, J. A. Davies, M. D. Kazatchkine, and w. G. van Aken (Eds.), Elsevier, Amsterdam, 1986, pp. 237-248. S. Ben Slimane, R. Guidoin, Y. Merhi, M. W. King, and M.F. SigotLuizard, “In vim evaluation of polyester arterial grafts coated with albumin: the role and importance of cross-linking agents,” Eur. Surg. Res., 20, 66-74 (1988). G. Joseph and C. P. Sharma, “Platelet adhesion to surfaces treated with glow discharge and albumin,“ 1. Biomed. Mater. Res., 20, 677-682 (1986). R. C. Eberhart, M. S. Munro, J.R. Frautschi, and V. I. Sevastianov, “Organization of albumin on polymer surfaces,” in Proteins a t Interfaces: Physicochemical and Biochemical Studies, J. L. Brash and T. A. Horbett (Eds.), American Chemical Society, Washington DC, 1987, pp. 378-400. W. R. Gombotz and A. S. Hoffman, “Gas discharge techniques for biomaterial modification,” CRC Critic. Rev. Biocornp., 4, 1-42 (1987).
Received March 6, 1989 Accepted July 31, 1989
20 min before the treatment session and then displacing the albumin solution with the patient's blood without creating an air-liquid interface. After introduction of this technique, a blockage was observed for only 11 in 239 cases (5%). In nitro measurements of platelet accumulation on the internal surface of the circulation system were carried out using washed human platelets labeled with ll'In-oxine in the presence of a 40% hematocrit. Preadsorption of the surface with albumin reduced platelet deposition to 4-5% that observed for an equivalent pretreatment with physiological saline.
INTRODUCTION
Thrombus formation and platelet-fibrin embolization following contact of flowing blood with an artificial surface are major problems in the development of extracorporeal circulation systems such as those used in plasmaPart of this work was presented at the 8emeCongres International sur la C.E.C., Paris, 23-25 June 1983. Journal of Biomedical Materials Research, Vol. 24, 155-163 (1990) 0 1990 John Wiley & Sons, Inc. CCC 0021-9304/90/020155-09$04.00
MULVIHILL ET AL.
156
pheresis. When blood enters into contact with a foreign surface, a series of events is initiated': (i) rapid adsorption at the surface of a layer of blood proteins, (ii) platelet and leukocyte adhesion to the protein layer with formation of platelet thrombi, (iii) activation of the coagulation system leading to thrombin generation and fibrin formation. Hemodynamic conditions influence the size, location, and structure of the resulting thrombus and its fragmentation. Under the high flow conditions of plasmapheresis systems, platelet accumulation normally predominates over the formation of a fibrin clot. Turbulent flow at joints, filters, constrictions and bifurcations may lead to the development of gross thrombotic deposits with eventual blockage of the circulation system. In therapeutic plasmapheresis using an IBM blood cell separator, we had observed, in spite of anticoagulation with acid-citrate-dextrose (ACD) and heparin and sometimes administration of the antiplatelet drug dipyridamole, the formation of platelet aggregates at joints, filters and the entry needle of the extracorporeal circulation system. Blockages often necessitated a halt to replace or clear part of the circuit. In consequence, the session was frequently shortened and the plasma exchange volume reduced. In order to improve the thromboresistance of the extracorporeal system, experimental and clinical studies of the effects of protein adsorption were undertaken. Since it has been shown that preadsorption of albumin to an artificial surface reduces platelet acc~mulation,~-~ we decided to passivate the surfaces of the IBM extracorporeal circuit by precoating it with human albumin prior to passage of the patient's blood.
PATIENTS AND METHODS
Clinical studies
Putien ts During a one-year period, 52 consecutive patients underwent 388 therapeutic plasmaphereses for various disorders (Table I). The effect of the preadsorption of human albumin was evaluated in the following way. From January to June 1982, the presence of documented platelet-fibrin thrombi in the system (fall in pressure in the IBM blood cell separator causing arrest of the session and macroscopic examination of the circuit) was retrospectively recorded from the clinical protocols of 149 consecutive apheresis sessions. In July 1982, it was decided to examine prospectively the effect of the preadsorption of human albumin on the formation of thrombi in the extracorporeal system. The study was conducted for 239 consecutive sessions during an equal 6-month period until December 1982.
157
SURFACE PASSIVATION BY ALBUMIN IN PLASMAPHERESIS
TABLE I Clinical Diagnosis in 52 Patients Submitted to Therapeutic Plasmapheresis During a One-Year Period Number of Sessions of Plasmapheresis -
Precoating of the Surface Clinical Diagnosis Myasthenia Rheumatoid arthritis Systemic lupus erythematosus Periarteritis nodosa Polymyositis Sclerodermia Chronic polyradiculoneuritis Multiple sclerosis Behcet’s disease Immune complex uveitis Immune complex glomerulonephritis Myeloma Waldenstrom’s macroglobulinemia Cryoglobulinemia Porphyria Total
Number of Patients 3 6
3 4 2 1 1 4 2 2 9 4
8 2 1 52
No Albumin (January to June 1982)
Albumin (July to December 1982)
1 25 14 9 12 14 3 -
9 29 4 10 13 48 22 23 38 18 19 12 6 239
-
42 12 15 149
Plasmapheresis At the beginning of each session, with or without albumin preadsorption, the extracorporeal circuit was filled with physiological saline. The preadsorption of albumin was carried out by displacing the saline solution with 4% pasteurized human serum albumin (Centre R6gional de Transfusion Sanguine, Strasbourg, France, purity greater than 98%). The albumin remained in the circuit, in static at ambient temperature, for an adsorption time of 15-20 min, after which it was displaced by the patient’s blood. Care was taken at all times to avoid the creation of an air-liquid interface. Plasma exchange was done by continuous flow centrifugation using an IBM 2997 blood cell separator and apheresis kits fabricated from polyvinylchloride (PVC) tubing. The length of the session varied from 1 to 2 h, the time to replace one plasma volume (40 mL/kg) by an equal volume of 4% human serum albumin. The total blood flow was in the range 1520 mL/min and the centrifugation speed 1800 rpm (550g). Before returning to the patient, the mixture of blood cells and replacement solution passed through a filter and a blood warmer at 37°C. Anticoagulant ACD (trisodium citrate dihydrate 17.8 g, citric acid monohydrate 0.3 g, anhydrous glucose 30.0 g/liter) in the proportions of 1 :9 to 1:10 and heparin (Roche, France) at
158
MULVIHILL ET AL.
20-30 U/min were infused into the extracorporeal circuit 10 cm after the entry needle.
Experimental studies To determine the effect of adsorbed albumin on platelet-surface interactions, in vitro experiments were carried out to quantitate the accumulation of platelets in the IBM circulation system.
Preparation and labeling of washed human platelets Washed human platelets were prepared according to the technique of Mustard with minor modifications6and labeled with lllIn-oxine (Amersham, France) as described by Eber et al.7 The labeled platelets were resuspended in Tyrode’s buffer (containing 2 mM Cazt, 1mM Mg2+,0.35% human serum albumin, 5 mM Hepes and 10 pl/ml apyrase, reference6)in the presence of washed red blood cells’ at 40% hematocrit.
Coating of the plusmapheresis circuit with human albumin Segments of PVC tubing (internal diameter 3 mm) of about 20-cm length from an apheresis kit were connected in parallel to a four-circuit peristaltic pump (Sage 375A, Orion Research Inc., Canada). The tubing was filled either with physiological saline for controls or with 4% human serum albumin, left at ambient temperature for 30 min and rinsed with physiological saline.
Measurement of platelet accumulation The platelet suspension was introduced into the segments of pretreated tubing by displacement of the saline rinse solution and circulated for 30 min at a flow rate of 13.3 mL/min from a reservoir thermostated at 37°C. All manipulation was conducted so as to avoid the occurrence of an air-liquid interface. After rinsing with physiological saline, the tubing was emptied, dried in an air stream, and cut into 2-cm segments for radioactive counting in a gamma-counter (1282 Compugamma, LKB, Turku, Finland). Platelet deposition on the internal surface of the segments was calculated according to the radioactivity per unit surface area and the specific activity of the platelets as described previously.’
159
SURFACE PASSIVATION BY ALBUMIN IN PLASMAPHERESIS RESULTS
Clinical studies The distribution of the 52 patients submitted to repeated therapeutic plasmapheresis during a one-year period is shown in Table I. The rationale for the treatment was to remove from the plasma of the patient either autoantibodies and immune complexes or excessive amounts of monoclonal antibodies. Most of the patients treated fall into these two categories. During the first half of 1982, 149 plasmapheresis sessions were conducted without precoating of the surface of the plasmapheresis circuit. During the second half of 1982, the same or new patients were plasmapheresed (239 sessions) after precoating of the circuit with human albumin. The clinical results of 388 consecutive plasmapheresis sessions conducted from January to December 1982 are shown in Table 11. During the first 6 months, when albumin pretreatment of the surface of the extracorporeal circuit was not employed, 66 blockages due to the presence of platelet-fibrin thrombi were recorded for a total of 149 sessions (44.3%).During the following 6 months, the extracorporeal circuit was precoated with 4% human albumin. A total of 239 sessions were carried out, in which 11 cases of blockage were observed (4.6%).
Experimental studies Since the thrombi obstructing the extracorporeal circuit were essentially composed of platelets, the effects on platelet-surface interactions of precoating the circuit with human albumin were investigated in vitro. Deposition of "'In-labeled platelets was quantitated in three separate experiments (Table 111). Albumin pretreatment reduced the concentration of platelets adhering to the surface to about 5% the concentration observed for the untreated surface.
TABLE I1 Therapeutic plasmapheresis. Effect of Albumin Pretreatment of the Extracorporeal Circulation System Precoating of the Surface No Albumin
Total number of sessions Blockage due to platelet-fibrin aggregates
149 (100%) 66 (44.3%)
Albumin 239 (100%) 11 (4.6%)*
Pretreatment of the circuits with 4% human albumin for 15-20 min in static at ambient temperature. * p < 10-~ (x2test).
MULVIHILL ET AL.
160
TABLE 111 In Vitru Effect of Human Albumin Pretreatment of the PVC Tubing of the Plasmapheresis Circuit on the Accumulation of “In-Labeled Platelets Number of Platelets Deposited per mm2 (mean t S.D. for 10 Tube Segments) Precoating of the Surface Experiment
No Albumin
1 2 3
9,100 f 500 5,000 400 16,900 t 1100
*
Albumin 400 300 800
f
-t
f
loo* 100“ 100“
Pretreatment of the circuit with 4% human serum albumin for 30 min in static at ambient temperature. Platelet suspension 180,000/mm3 containing 40% red blood cells recirculated with a peristaltic pump for 30 min at 37°C. “ p < 0.001 (paired Student’s t-test).
DISCUSSION
One of the major problems in plasmapheresis for therapeutic reasons remains the occurrence of thrombotic complications due to blood activation by the surface of the plasmapheresis circuits, in spite of the administration of anticoagulants and antiplatelet drugs to the patient.” In this respect, our experience is in agreement with that of others. We used an IBM blood cell separator to perform therapeutic plasmapheresis in patients with immunological disorders, to remove autoantibodies, immune complexes or monoclonal antibodies. Plasmapheresis was accompanied, during the first 6 months of the study, by blockage of the circuit with platelet-fibrin thrombi in 4.4% of the sessions, despite the patient anticoagulation. This is not surprising, since it is well known that immunoglobulin or immune complex adsorption on surfaces is highly t h r o m b o g e n i ~ , ~ and ~ ~our ” ~ patients had elevated levels of autoantibodies, immune complexes or monoclonal immunoglobulins in their plasma. In addition to immunoglobulin or immune complex deposition on the surface of the plasmapheresis circuit, other mechanisms may have played a role in the formation of platelet-fibrin thrombi, e.g., fibrinogen adsorption, activation of the coagulation and complement systems, adhesion of leukocytes and release of procoagulant factors, flow abnormalities or the presence of air As the adsorption of albumin on artificial surfaces is known to prevent extensive platelet a c c ~ m u l a t i o n , we ~ ~ ~decided ~ - ~ ~ to passivate the plasmapheresis circuit by prior adsorption of human albumin. Preliminary in vifro studies with a perfused PVC tubing circuit demonstrated that pretreatment of the surface by albumin preadsorption reduced considerably the number of adherent platelets in comparison to the nontreated surface (Table 111). We then used this same method to passivate the IBM plasmapheresis circuit be-
SURFACE PASSIVATION BY ALBUMIN IN PLASMAPHERESIS
161
fore the beginning of the therapeutic session. According to the results summarized in Table 11, albumin preadsorption of the extracorporeal circulation system reduced the occurrence of blockage due to the formation of plateletfibrin aggregates from 44% to 4.6% of the total number of sessions. This technique thus minimized the cases in which the plasmapheresis session was shortened and the plasma exchange volume reduced. Since the halts in treatment to clear a blocked circuit were largely eliminated, it also enabled a considerable time gain, the average number of patients treated per day passing from two to three. It is important to note that in all our studies, clinical or experimental, the patient's blood or the platelet suspension or protein solution was introduced into the extracorporeal circuit by displacement of another solution, either albumin or physiological saline, thus avoiding the formation of an air-liquid interface. This is a crucial point, since air-liquid interfaces cause activation of platelets and the coagulation and complement systems. l6 Several approaches can be used to prevent the undesirable effects of platelet-surface interactions in plasmapheresis circuits: modification of the The adsorption of surface or pharmacological modification of platelets. albumin on surfaces markedly reduces platelet deposition, in contrast to adsorption of fibrinogen or garnma-gl~bulins.~~'~~~~-~~ In addition, the adsorbed albumin competes with immune complexes and other proteins such as fibrinogen for adsorption on the surface. Thus, we decided to passivate the plasmapheresis circuit with human albumin. We are aware that the study, for historical reasons, contains a methodological flaw in that the two clinical series were sequential in time and did not overlap, but in view of the large difference between the two groups it is unlikely to be due to a mere statistical aberration. Furthermore, the same patients were treated with and without albumin passivation and the clinical data are fully supported by the results for in vitro platelet accumulation. Diluted (4%) sterile human albumin solution is cheap and well tolerated by the patient and can be used as a replacement solution for the exchanged volume of plasma. Owing to its simplicity, the passivation of plasmapheresis circuits with albumin has now been introduced as a routine procedure in our plasmapheresis unit. For short-term use, adsorption of albumin renders the artificial surface sufficiently thromboresistant and is less expensive than bonding albumin to polymer surfaces by chemical or radiation processes.20-22 The authors are grateful for the secretarial assistance of Mrs. C. Helbourg and the technical assistance of Mrs. J. Groener and Mrs. A. Sutter-Bay.
References 1. E. W. Salzman and E. W. Merrill, "Interaction of blood with artificial surfaces," in Hemostasis and Thrombosis. Basic Principles and Clinical Practice, 2nd ed. R. W. Colman, J. Hirsh, V. J. Marder, and E. W. Salzman (Eds.), J. 8. Lippincott Co., Philadelphia, 1987, pp. 1335-1347.
MULVIHILL ET AL.
162 2. 3.
4. 5.
6.
7.
8.
9.
10.
11. 12.
13. 14. 15.
16.
17.
M.A. Packham, G. Evans, M.F. Glynn, and J.F. Mustard, "Effect of plasma proteins on interaction of platelets with glass surfaces," 1. Lab. Clin. Med., 73, 686-697 (1969). J. L. Brash, "Hydrophobic polymer surfaces and their interactions with blood," Ann. NY Acad. Sci., 283, 356-371 (1977). D. E Mosher, "Influence of proteins on platelet-surface interactions," in Interaction of the Blood with Natural and Artificial Surfaces, E. W. Salzman (Ed.), Marcel Dekker, New York, 1981, pp. 85-101. K . Park, D. F. Mosher, and S. L. Cooper, "Acute surface-induced thrombosis in the canine ex vivo model: importance of protein composition of the initial monolayer and platelet activation," J. Biomed. Muter. Xes., 20, 589-612 (1986). J.-P. Cazenave, S. Hemmendinger, A. Beretz, A. Sutter-Bay, and J. Launay, "L'agrhgation plaquettaire: outil d'investigation clinique et dktude pharmacologique. Methodologie," Ann. Biol. Clin., 41, 167-179 (1983). M. Eber, J.-P. Cazenave, J. C. Grob, J . Abecassis, and G. Mcthlin, ""lIndium labeling of human washed platelets; kinetics and in vivo sequestration sites," in Blood Cells in Nuclear Medicine, Part I. Cell Kinetics and Bio-distribution, M. R. Hardeman and Y. Najean (Eds.), Martinus Nijhoff Publishers, Boston, 1984, pp. 29-43. J.-P. Cazenave, D. Blondowska, M. Richardson, R. L. KinloughRathbone, M. A. Packham, and J. F. Mustard, "Quantitative radioisotopic measurement and scanning electron microscopic study of platelet adherence to a collagen-coated surface and to subendothelium with a rotating probe device," 1. Lab. Clin. Med., 93, 60-70 (1979). J. N. Mulvihill, H. G. Huisman, J.-P. Cazenave, J. A. van Mourik, and W. G. van Aken, "The use of monoclonal antibodies to human platelet membrane glycoprotein IIb-IIIa to quantitate platelet deposition on artificial surfaces," Thromb. Haemostas., 58, 724-731 (1987). G. P. Clagett, "Artificial devices in clinical practice," in Hemostasis and Thrombosis. Basic Principles and Clinical Practice, 2nd ed. R. W. Colman, J. Hirsh, V. J. Marder, and E. W. Salzman (Eds.), J. B. Lippincott Co., Philadelphia, 1987, pp. 1348-1365. P. M. Henson and H. L. Spiegelberg, "Release of serotonin from human platelets induced by aggregated immunoglobulins of different classes and subclasses," J. Clin. Invest., 52, 1282-1288 (1973). C . R. H. Wildevuur and M. D. Kazatchkine, "Vascular biomaterials in use," in Blood-Surface Interactions: Biological Principles Underlying Haemocompatibility with Artificial Materials, J.-P. Cazenave, J. A. Davies, M. D. Kazatchkine, and W. G. van Aken (Eds.), Elsevier, Amsterdam, 1986, pp. 197-207. T. Chandy and C. P. Sharma, "The antithrombotic effect of prostaglandin E l immobilized on albuminated polymer matrix," J. Biomed. Muter. Res., 18, 1115-1124 (1984). Y. Ishikawa, S. Sasakawa, M. Takase, and Y. Osada, "Effect of albumin immobilization by plasma polymerization on platelet reactivity," Thromb. Res., 35, 193-202 (1984). K . Kottke-Marchant, J. M. Anderson, Y. Umemura, a n d R . E . Marchant, "Effect of albumin coating on the in vitro blood compatibility of Dacron arterial prostheses," Biomaterials, 10, 147-155 (1989). C . A. Ward, "Blood protein-material interactions that lead to cellular adhesion," in Proteins at Intevfaces: Physicochemical and Biochemical Studies, J. L. Brash and T. A. Horbett (Eds.), American Chemical Society, Washington DC, 1987, pp. 551-565. C. Pusineri and J.-P. Cazenave, "Techniques for modifying artificial surfaces," in Blood-Surface Interactions: Biological Principles Underlying
SURFACE PASSWATION BY ALBUMIN IN PLASMAPHERESIS
18.
19.
20.
21.
22.
163
Haemocompatibility with Artificial Materials, J.-P. Cazenave, J. A. Davies, M. D. Kazatchkine, and W. G. van Aken (Eds.), Elsevier, Amsterdam, 1986, pp. 209-235. J. A. Davies, ”Therapy affecting the blood,” in Blood-Surface Inteructions: Biological Principles Underlying Haemocompatibility with Artificial Materials, J.-P. Cazenave, J. A. Davies, M. D. Kazatchkine, and w. G. van Aken (Eds.), Elsevier, Amsterdam, 1986, pp. 237-248. S. Ben Slimane, R. Guidoin, Y. Merhi, M. W. King, and M.F. SigotLuizard, “In vim evaluation of polyester arterial grafts coated with albumin: the role and importance of cross-linking agents,” Eur. Surg. Res., 20, 66-74 (1988). G. Joseph and C. P. Sharma, “Platelet adhesion to surfaces treated with glow discharge and albumin,“ 1. Biomed. Mater. Res., 20, 677-682 (1986). R. C. Eberhart, M. S. Munro, J.R. Frautschi, and V. I. Sevastianov, “Organization of albumin on polymer surfaces,” in Proteins a t Interfaces: Physicochemical and Biochemical Studies, J. L. Brash and T. A. Horbett (Eds.), American Chemical Society, Washington DC, 1987, pp. 378-400. W. R. Gombotz and A. S. Hoffman, “Gas discharge techniques for biomaterial modification,” CRC Critic. Rev. Biocornp., 4, 1-42 (1987).
Received March 6, 1989 Accepted July 31, 1989
