In vitro platelet interactions with a heparin-polyvinyl alcohol hydrogel.

No difference in in vitro platelet reactivity was found between an immobilized heparin containing hydrogel (heparin-PVA) and the hydrogel without heparin (PVA), in a variety of experimental assays. There was no significant difference between the heparin-PVA and PVA coated polyethylene tubing in the number of 51Cr-labeled platelets, the extent of 14C serotonin release by the adherent platelets or in the degree of platelet count decrease after 1 h exposure to citrated canine whole blood in a Chandler loop system. Furthermore, adhesion and release values were lower than those observed with the uncoated polyethylene tubing (e.g., 9.3 +/- 4.3 plt/10(3) microns 2 on PVA; 18.3 +/- 4.6 plt/10(3) microns 2 on polyethylene). There was also no significant difference between heparin-PVA and PVA in bead column retention values with canine blood and with the previously reported washed human platelet adhesion/release values. Thus there appears to be no effect of the immobilized heparin by itself on the in vitro interactions of PVA with platelets, with the reactivity towards platelets dominated by that of the underlying substrate (i.e., PVA).

Effect of heparin-PVA hydrogel on platelets in a chronic canine arterio-venous shunt.

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 no effect whereas unmodified polyethylene was found to be unreactive despite similar levels of platelet deposition as measured by SEM and a higher in vitro reactivity (J. 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 vitro 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.

Water content and compression modulus of some heparin-PVA hydrogels.

The effects of changes in the heparin-PVA formulation on the water content, compression modulus and molecular weight between cross-links of the resulting gel was determined. Molecular weight between cross-links was calculated directly from the compression modulus and the swelling ratio. Glutaraldehyde and MgCl2 displayed the effects expected of them as cross-linking agent and catalyst respectively. On the other hand, formaldehyde appeared to be a non-essential component since its absence had little effect on gel properties and its presence did not affect the water content as originally predicted. Understanding the effect of each formulation component on gel properties enables alteration of gel water content and hence permeability for special applications.