Effect of heparin-PVA hydrogel on platelets in a chronic canine arterio-venous shunt Cynthia H. Cholakis, Walter Zingg,"...
76 downloads
431 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Effect of heparin-PVA hydrogel on platelets in a chronic canine arterio-venous shunt Cynthia H. Cholakis, Walter Zingg," and Michael V. Seftont Department of Chemical Engineering and Applied Chemistry and the Centre for Biomateriuls, University of Toronto, Toronto, Ontario M5S 1A4 Canada Polyvinyl alcohol (PVA) hydrogel, with or without heparin, was reactive towards canine platelets in a chronic arteriovenous shunt as demonstrated by an increase in platelet regeneration time, a systemic decrease in platelet count and transient decrease in platelet serotonin content. Immobilized heparin (heparin-PVA) had n o effect w h e r e a s unmodified polyethylene was found to be unreactive despite similar levels of platelet deposition as measured by SEM and a higher in vitro reactivity (I. Biomed. Mater. Res., this issue). Twenty-centimeter lengths of hydrogel coated polyethylene tubing were inserted between the arterial and venous portions of the shunt and left in place for 4-6 days, without the complicating artifacts of anticoagulation, anesthesia, or
surgical intervention. Regeneration time was measured as the return to normal platelet cyclooxygenase (co) activity after a single 240-mg dose of aspirin, with co activity measured in uitro as malondialdehyde production. Although measuring new platelet production, regeneration time is an indirect measure of platelet consumption, so that the reduced regeneration time seen here was presumed to reflect enhanced material associated consumption and thromboembolism. Like other hydrogels, PVA does not appear to be "thromboadherent" but it does appear thrombogenic. Immobilized heparin had no additional effect, presumably because the platelet response was dominated by the reactivity of the underlying substrate.
INTRODUCTION
The interaction of heparin, in solution or immobilized, with platelets is unclear. As summarized in a previous article,' both inhibitory and enhancing effects have been noted for heparin either in an aggregometer or by direct assessment of the interaction of heparinized biomaterials with platelets in bead columns or in measurements of adhesion. In the previous article,' we were unable to observe any effect of immobilized heparin, as heparinPVA, on platelets in vitra. Platelet adhesion, release from adherent platelets, and platelet retention were identical on both a heparin-PVA hydrogel and the same hydrogel without immobilized heparin. We continue that study here by reporting on the interaction with platelets, ex vivo in dogs in a chronic AV shunt. *Hospital for Sick Children Research Institute, Division of Surgical Research. 'To whom correspondence should be addressed. Journal of Biomedical Materials Research, Vol. 23, 417-441 (1989) CCC 0021-93041891040417-25$04.00 0 1989 John Wiley & Sons, Inc.
418
CHOLAKTS, ZINGG, AND SEFTON
Heparin was immobilized through its amino acid terminus to a glutaraldehyde crosslinked PVA.* In the absence of significant heparin release, this covalently bound heparin has demonstrated a low in vitro thrombogenicity3-6(partial thromboplastin or thrombin times, thrombin inactivationichromogenic substrate assay, factor Xa inactivation). In addition, in a canine AV shunt the heparin-PVA hydrogel was shown7 to be effective in delaying thrombus formation in 1-mm-ID coated polyethylene tubes at low flow rates (-5 mL/min). The prolonged patency times of heparin-PVA coated tubes with that of control PVA coated tubes without heparin was attributed to the biological effectiveness of the immobilized heparin since the heparin release rate was ?boo of that needed to create a heparin microenvironment.' Here, the same chronic arterio-venous shunt' was used to evaluate the effect of immobilized heparin (and the underlying I'VA substrate) over a 5day period. Previously a parallel flow test section7 was used to expose 1-mm-ID tubes to canine blood at low flow rates, essentially to measure the anticoagulant activity of the immobilized heparin. Here, however, 3mm-ID heparin-PVA (or control) coated tubes were connected in series with the arterial and venous portions of the shunt to expose the entire shunt flow (-180 mL/min) to the coated tube. At this flow rate the shunt remained patent indefinitely provided the shunt material was not too thrombogenic; i.e., it appeared that intrinsic coagulation and fibrin formation were not limiting.
MATERIALS AND METHODS
Heparin-PVA solution Heparin-PVA aqueous solutions were prepared as before,' consisting of 7-10% (w/w) poly-(vinyl alcohol) (99-100% hydrolyzed PVA, J. T. Baker Chem. Co., Phillipsburg, NJ), 5% (w/w) magnesium chloride (Lewis acid catalyst, BDH Chemicals, Toronto, Ont), 0.5-2.0% (w/w) glutaraldehyde (cross-linkingagent, BDH Chemicals), 3% (w/w) formaldehyde (Fisher Scientific Co., Fairlawn, NJ), 4% (w/w) glycerol (BDH Chemicals), 2.0% (w/w) sodium heparin (usually Canada Packers intestinal mucosal, not less than 140 USP U/mg, Toronto, Ont.), and distilled deionized water. Glycerol was added to prevent precipitation of heparin during dehydration. The control hydrogel was prepared in the identical manner but without heparin. The PVA used here was fully deacetylated unlike that used in earlier studies.s7 The PVA was dissolved in double distilled deionized water at -9O"C, followed by cooling of the dissolved polymer to room temperature and then the addition of the other ingredients with extensive stirring. The gel solution was set aside overnight in the refrigerator, to allow entrapped air bubbles to escape and was used within seven days of preparation. Tubes were coated with PVA hydrogel with 2% heparin or 2% "high antithrombin 111 affinity" heparin (34.9 n mole serine/mg, Schering Ag., Berlin). Heparin content of the standard heparin tubes was -15 mg/g
HEPARIN-PVA HYDROGEL IN A-V SHUNT
419
gel (-16 pg cm') with a release rate of -3 x g/cm2 - min, all based on the toludine blue assay.4 There was a similar glycosaminoglyan content for the "high affinity" heparin surface which was examined in only one experiment.
Coating of polyethylene tubing Polyethylene (PE) tubing (Intramedic, Clay Adams, Parsippany, NJ) was pretreated with 17 M chromic acid (88.5% H,SO,, 7.1% K,Cr,O,, w/w) at 70°C for 8 min and glow discharge cleaned (Harrick plasma cleaner, Ossining, NY) in an air plasma at full power for 45 min. The tubing was then retreated with chromic acid in an identical fashion. This treatment resulted in good adhesion between the hydrogel and the underlying substrate.'' The etched PE tubes (-30 cm long) were coated by filling vertically held tubings with gel solution, followed by draining and drying for 45 min. After four coats were applied, the tubes were left overnight before curing at 70°C for 2 h. Coated tubes were trimmed, reswollen, and stored in PBS.
Chronic arterio-venous canine shunt The AV shunts were made from Silastic medical grade tubing (3.18 mm ID, 6.35 mm OD, Dow Corning, Midland, MI). The shunt was inserted between the iliac artery and contralateral vein, with a short, smaller-diameter Silastic extension (10 cm long, 1.98 mm ID, 3.18 mm OD), used for arterial cannulation. The total shunt length was -100 cm. The shunt implantation is described in detail elswhere.' Briefly, each limb of the shunt, prefilled with heparinized saline (15 IU/mL heparin) was inserted 10 cm into the blood vessel and anchored with grommets to the vessel wall at the cannulation site, and then tunneled under the skin at each side of the body to exit separately from the back of the animal. The animal was held in a snug jacket to prevent damage to the exteriorized portion of the catheter and to minimize reciprocating motion at the wound that could lead to bacterial infiltration and infection. An antibiotic (Pen Strep or Longisil, PVU Inc., Victoriaville, CT, 1 mL daily intramuscularly) and anticoagulant (coumadin, 0.25-0.5 mg/kg orally) were administered postoperatively for the first week only. The shunt flow rate was measured by Doppler ultrasound (Blood velocimeter BV 380 Sonicaid, Fredericksburg, VA). The unit was calibrated with the timed collection of blood in an in nitro peristaltic pump flow circuit.
Insertion of test sections At the start of an experiment, the animal was restrained by hand so that it was sitting comfortably, to enable connection of the 20-cm length of tub-
CHOLAKIS, ZINGG, AND SEFTON
420
ing to the exteriorized portion of the chronic shunt without anesthesia. The test section (test tubing and the connectors) was sterilized in ethylene oxide gas for 3 h, then degassed overnight prior to use. The sterilized tube was filled with sterile saline and the animal's chronic shunt was clamped with tubing clamps (Atraugrip-Doyen Paediatric Intestinal Clamp, Downs Surgical Ltd., England), at both arterial and venous limbs, to stop the flow temporarily. Upon disconnecting the chronic shunt, the test section was quickly fitted onto each limb of the chronic shunt, with the aid of two Silastic external connectors (6.35 mm ID, 9.52 mm OD, 3 cm long) slipped over the ends of the tubing (Fig. 1). After connections were secured, the tubing clamps were removed to expose the test section to the flowing blood. The total occlusion time was -5 min. At the conclusion of the test (up to 6 days) the test section was removed from the main shunt by clamping the shunt once again, and then its two limbs were reconnected with a Silastic connector slipped over the joint. Animals were not used for a new experiment for at least 2 weeks after disconnection to allow all hemostatic factors to return to normal. Blood samples from the shunt, with or without the test section, were taken regularly to measure the platelet count (and other hematological parameters). Platelet counts were determined, in all experiments, by the clinical hematology laboratory of the Hospital for Sick Children in Toronto (ELT-8, Ortho Instruments).
Polyethylenetube Sllastic protector
to vein
Polypropylenecap Silastic connector (635rnmIDx952mmOD)
external portion of chronic shunt
from artery
Figure 1. Schematic illustration of test section showing tubing segment and connectors.
421
HEPARIN-PVA HYDROGEL IN A-V SHUNT
Platelet regeneration time A nonradioisotope technique was used to measure the platelet regeneration time during the connection of test segments, to determine albeit indirectly the platelet life span and platelet consumption rate (Fig. 2). The procedure" is based on the observation that a single oral dose of acetylsalicylic acid, aspirin, irreversibly inhibits platelet cyclooxygenase for the life of the platelet. Thus the in vitro thrombin induced peroxidation of platelet lipids is inhibited for those platelets affected by ASA for the life of the platelet. A product of lipid peroxidation, malondialdehyde (MDA), was measured indirectly after reaction with thiobarbituric acid as shown in Figure 2, using a spectrophotometer. The required time for a return to baseline blood production
consumption
platelet count
(CIJ
regeneration time
-
platelet life span
aspirin blocks
cyclooxygenase
Arachidonic acid
1
x7pGG2
J
PGH~-
5 to 6 days and the mean PRT half time was 3.8 2 0.8 days suggest that the shunt did not substantially affect the regeneration time. It is clear from the regeneration time results that the presence of the hydrogels increased the consumption rate of platelets to the extent that the rate of production of new platelets had to be increased to compensate (and this was insufficient given the platelet count decrease). However, it is not possible to determine a precise platelet consumption rate from the raw data since the assay measures the increase in the percentage of circulating platelets unaffected by aspirin and not the number of new platelets brought into the circulation. Unfortunately, we cannot estimate what fraction of the new platelets are removed prior to our MDA measurement. Nevertheless if one assumed that over the first 24 h, there is no removal of the new platelets, then it is possible to estimate a consumption rate. These rates are gwen in Table 11. In contrast in baboons, Hanson” found polyethylene and polydimethylsiloxane to be the least consumptive of all materials tested (2 X lo8 pIt/cm2-day)with Biomer and polyacrylamide (grafted into Silastic) to be the highest at 19.8 and 18 x 10’ plt/cm2-day. Presumably the difference in consumption values represent in large part species differences, although this remains to be tested. Unfortunately neither the effect of the shunt itself nor normal senescence can be subtracted from these calculations. Finally it must be noted that the detailed sequence of events that occurs once a platelet has contacted a surface in such a shunt and is eventually removed from the circulation is largely unknown. Unlike adhesion or aggregation, plateIet consumption is more difficult to follow since most of the effect occurs after the platelet has left the surface. Also the clinical significance of this phenomenon (in the absence of thrombocytopenia) is unclear since it is not known whether consumption represents microemboli formation or not. These questions and others remain to be answered. CONCLUSIONS
The ex Z ~ Z D O series shunt has been found to be an excellent system for evaluating the effect of heparin-PVA and PVA hydrogel on platelets. It is presumably useful for testing other potentially compatible materials. It allows for repeated testing in conscious dogs without anesthesia and without surgical artifacts. Consequently, interanimal variability, the number of animals and the cost of experimentation are reduced. The tests employed are sensitive to the surface properties of test segments. In this shunt, in canines, the PVA hydrogel was reactive towards platelets, and the presence of the immobilized heparin had no significant effect on that reactivity. Connection of heparin-PVA and PVA tubing segments (20 crn long, 3.18 mm ID) resulted in substantial decreases in systemic platelet counts (thrombocytopenia). Reductions in counts were approximately 50% over a 48-h period. There was also a greater than 50% reduction in platelet serotonin levels 24 h after connection of heparin-
HEPARIN-PVA HYDROGEL IN A-V SHUNT
439
PVA tubing segments. Finally, platelet regeneration times were reduced (50% decrease in mean half time) during the connection of heparin-PVA and PVA test segments indicating an increase in the platelet consumption rate. Despite the ability of heparin-PVA to inhibit fibrin formation it was found to be reactive towards platelets. However, it appeared that the substrate, PVA, dominated the platelet response and that the presence of heparin per se did not alter the situation. In vitro platelet adhesion and release of adherent platelet values reported previously were not a good indicator of ex vivo performance. While the conclusion about the absence of an effect of immobilized heparin is unchanged, the platelet reactivity of PVA was not predicted on the basis of the previously reported in vitro results.' Hence the significance of in vifro measurements as a predictor of ex vivo performance is questioned. The authors acknowledge the financial support of the NIH (HL 24020) and a scholarship (CHC) from the Natural Science and Engineering Research Council of Canada. We are also grateful for the technical assistance of Wan Ip in the animal experiments.
References 1. C. H. Cholakis and M. V. Sefton, "ln vitro platelet interactions with a heparin-polyvinyl alcohol hydrogel," J. Biomed. Muter. Res., this issue. 2. C . H. Cholakis and M. V. Sefton, "Chemical characterization of an immobilized heparin: heparin PVA," in Polymers as Biomaterials, S . Shalaby, A . S. Hoffman, B. D. Ratner, and T. Horbett (Eds.), Plenum Press, New York, 1984, pp. 305-315. 3. M. V. Sefton and M. F. A . Goosen, "Heparinized styrene butadiene styrene elastomers," J. Biomed. Muter. Res., 13, 347-364 (1979). 4. M.F. A. Goosen and M. V. Sefton, "Properties of heparin-polyvinyl alcohol coating," 1. Biomed. Muter. Xes., 17, 359-373 (1983). 5. M. F. A . Goosen, M. V. Sefton, and M. W. C. Hatton, "Inactivation of thrombin by antithrombin I11 on a heparinized biomaterial," Thromb. Res., 20, 523-554 (1980). 6 . G. Rollason and M. V. Sefton, "Factor X, inactivation by a heparinized hydrogel," Thromb. Res., 44, 517-525 (1986). 7. W. F. Ip, W. Zingg, and M. V. Sefton, "Parallel flow arterio-venous shunt for the ex vivo evaluation of heparinized materials," 1. Biomed. Muter. Res., 19, 161-178 (1985). a. D.Basmadjian and M. V. Sefton, "Relationship between release rate and surface concentration of heparinized materials," J. Biomed. Muter. Res., 17, 509-518 (1983). 9. W. Zingg, W. F. Ip, M. V. Sefton, and K. Mancer, "A chronic arteriovenous shunt for the testing of biomaterials and devices in dogs," Life Support Syst., 4, 221-229 (1986). 10. R. A. Evangelista and M. V. Sefton, "Coating of two polyetherpolyurethanes and polyethylene with a heparin-poly (vinyl alcohol) hydrogel," Biomuterials, 7, 206-211 (1986). 11. M. S. Stuart, S. Murphy, and F. A. Oski, "A simple nonradioisotope technic for the determination of platelet life-span," N. Eng. J. Med., 292, 1310-1313 (1975). 12. G. G. Nenci, P. Parise, E. Capodicasa, and M. Berrettini, "Platelet malondialdehyde production kinetics after cyclooxygenase block," Haemostatis, 13, 53-60 (1983).
CHOLAKIS, ZINGG, AND SEFTON
440
13. A. H. Drummond and J. L. Gordon, ”Rapid sensitive micro-assay for platelet 5-HT,” Thromb. Diath. Haemorrh., 31, 366-367 (1974). 14. L. McTntyre, et al., Guidelines for Blood-Material Interactions, Report of 15. 16. 17.
18.
19.
20. 21. 22. 23. 24. 25. 26. 27. 28. 29.
30. 31.
the National Heart, Lung and Blood Institute Working Group, No. 852185 (1985). A. Parbtani and J . S. Cameron, ”Platelet and plasma serotonin concentration in glomerulonephritis 1,” Thromb. Res., 15, 109-125 (1979). C. Eika, ”The platelet aggregating effect of eight commercial heparins,” Scand. 1. Haernat., 9, 480-482 (1972). S. Kaplan, K.F. Marcoe, L.R. Sauvage, M. Zammit, H.D. Wu, S.R. Mathisen, and M. W. Walker, “The effect of predetermined thrombotic potential of the recipient on small-caliber graft performance,” J. Vasc. Surg., 3, 311-321 (1986). R. S. Wilson and S. L. Cooper, “In-Vitro and in-vivo test methods for assessing blood compatibility,” in Polymeric Biornaterials, E. Piskin, A. S. Hoffman, (Eds.), NATO AS1 Series, Series E: Applied Sciences No. 106, Amsterdam, Nartinus Nijhoff Publishers, 1986, pp. 15-28. L. K. Lambrecht, M. D. Lelah, C. A. Jordan, M. E. Pariso, R. M. Albrecht, and S. L. Cooper, “Evaluation of thrombus deposition onto polymeric biomaterials in a new subhuman primate ex-vivo series shunt model,” Trans. ASAIO, 29, 194-199 (1983). J. H. Lewis, “Comparative hematology: mammalian platelets,” in Platelets, 0. N. Ulutin (Ed.), Excerpta Medica, Amsterdam, 1974; pp. 18-23. S. L. Raymond and W. J. Dodds, ”Platelet membrane glycoproteins in normal dogs and dogs with hemostatic defects,” 1. Lab. Clin. Med., 93, No. 4, 607-613 (1979). W. J. Dodds, ”Platelet function in animals: Species specificities,” in Platelets: A Multidisciplinay Approach, G. deGaetano and S. Garattini, (Eds.), Raven Press, New York, 1978, pp. 45-49. L. J. Wurzinger and H. Schmid-Schonbein, ”Species differences in platelet aggregation and the influence of citrate and heparin anticoagulation thereon,” ASAIO J., 4, 149-156 (1981). M. M. Laks, L. Greenless, and D. Garner, ”Comparison of ejection fraction and segmental circumferential fiber shortening velocity in the anesthetized and conscious canine,” lap. Heart J., 20, 359-365 (1979). F. N. McKenzie, E. Svensju, and K. E. Arfors, ”Effect of sodium pentobarbital anesthesia on platelet behavior in vitro and in vivo,” Microvas. Res., 4, 42-50 (1972). K. K. Wu and J. C. Hoak, “A new method for the quantitative detection of platelet aggregates in patients with arterial insufficiency,” Lancet, 924-926 (1974). G. A. Adams and I. A. Feuerstein, ”Maximum fluid concentrations of materials released from platelets at a surface,” A m . J. Physiol., 244, H109-Hl14 (1983). S. R. Hanson, L. A. Harker, B. D. Ratner, and A. S. Hoffman, ”In vivo evaluation of artificial surfaces with a non-human primate model of arterial thrombosis,” J. Lab. Clin. Med., 95, 289-304 (1980). G. L. Schmer, X. L. Teng, J. J. Cole, J. E. Vizzo, M. M. Francisco, and B. A. Scribner, ”Successful use of a totally heparin grafted hemodialysis system in sheep,” Trans. ASAIO, 22, 654-663 (1976). E. W. Salzman, E. W. Merrill, A. Binder, C. F. W. Wolf, and T. I? Ashford, “Protein-platelet interaction on heparinized surfaces,” I . Biomed. Muter. Res., 3, 69-81 (1969). R.A. Indeglia, F.D. Dorman, A . R. Castaneda, R.L. Varco, and E.F. Bernstein, “Use of GBH-coated Tygon tubing for experimental prolonged perfusions without systemic heparinization,” Trans. ASAIO, 12, 166-173 (1966).
HEPARIN-PVA HYDROGEL IN A-V SHUNT
441
32. J . M . Courtney, G . B . Park, I . A . Fairweather, and R.M. Lindsay, "Polymer structure and blood compatibility evaluation: application of an acrylonitrile copolymer," Biomafer. Med. Dev. Art. Urg., 4(3&4), 263 (1976). 33. P. Olsson, H. Lagergren, R. Larsson, and K. Radegan, "Prevention of platelet adhesion and aggregation by a glutardialdehyde-stabilized heparin surface," Thromb. Huemostas., 37, 274-282 (1977). 34. S. Hanson and L. A. Harker, "Simultaneous 'lCr and "C-serotonin platelet survival measurements," Thromb. Haemosfus., 38, 140 (1977). 35. S. Karpatkin, "Human platelet senescence," Ann. Rev. Med., 23, 101128 (1972). 36. G. Di Minno, M. J. Silver, and S. Murphy, "Monitoring the entry of new platelets into the circulation after ingestion of aspirin," Blood, 61, 1081-1085 (1983). 37. L. A. Harker and S. R. Hanson, "Experimental arterial thromboembolism in baboons: mechanism, quantitation and pharmacologic prevention," J. Clin. Invesf., 64, 559-569 (1979). 38. R. Roncucci, R. Deperon, J. Destailleur, J. Doumont, G. Lambelin, J. Lansen, M. Moriau, F. Van Stalle, and R. Verhaeghe, "Measurement of platelet regeneration time in cardiovascular patients," Thromb. Res , 14, 3-14 (1979). 39. V. L. Porter-Fink and F. E. Eckhauser, "Ill-In-8-hydroxyquinoline platelet labeling in dogs," J. Surg. Res., 31, 87-92 (1981). 40. D. 0. Rauls, J. R. Daniels, T. J. Brown, G. Parker, A. J. Moreno, and J. Collins, "Evaluation of platelet kinetics in 4 dogs using Indium-111,'' I. Surg. Res., 33, 362-366 (1982). 41. A. R. Wilkinson, R. J. Hawker, and L. M. Hawker, "111-Indium labelled canine platelets," Thromb. Res., 13, 172-182 (1978).
Received June 28, 1988 Accepted July 27, 1988