Five randomized controlled trials (RCTs) (Bell and Royall, 1980; Green et al., 1984; Powers et al., 1984; Ansell et al., 1985; Bailey et al., 1986) and one nonrandomized study (Cipolle et al., 1983; Ramirez-Lassepas et al., 1984) compared the frequency of HIT during treatment with UFH that was derived from either bovine lung or porcine intestinal mucosa. In addition, the frequency of HIT was evaluated in normal volunteers in one RCT (Schwartz et al., 1985) and one nonrandomized prospective study (Saffle et al., 1980) involving porcine and bovine heparins. The study of Bell and Royall (1980) has been excluded from primary analysis because neither laboratory testing for HIT antibodies nor data on the timing of onset of thrombocytopenia were provided.
Taken together, the four RCTs in medical patients strongly suggest that bovine UFH is more likely to cause HIT than porcine UFH, as all nine patients with HIT had received UFH of bovine origin (p = 0.0059 by Mantel-Haenszel) (Table 2). Similarly, an increased frequency of HIT in patients receiving bovine lung heparin was suggested in the nonrandomized comparison by Cipolle and colleagues (1983) (6/100, 6% vs. 1/111, 0.9%; p = 0.055), as well as in the study of Bell and Royall (1980) (26% vs. 8%; p < 0.005), although this latter study probably included patients with nonimmune HAT.
A significantly higher frequency of HIT antibody formation with bovine UFH was observed in a study of cardiac surgical patients randomized between bovine and porcine UFH (Francis et al., 2003). Although a smaller cardiac surgery study (Konkle et al., 2001) failed to detect a difference in antibody formation, blood samples were obtained only until postoperative day 5, which may have been too early to detect the majority of HIT antibodies (Warkentin, 2003).
A higher frequency of immune HIT with bovine lung heparin is biologically plausible. Bovine heparin has a higher sulfate:disaccharide ratio than does porcine heparin (Casu et al., 1983), and it is better able to activate platelets in vitro (Barradas et al., 1987). These properties could lead to greater platelet activation in vivo and, consequently, greater potential for PF4 release. Moreover, the bovine heparin chains would be expected to better form the large multimolecular complexes that compose the target antigen for HIT antibodies.
Lot-to-lot variability within heparin of a particular animal source could also contribute to variable frequency of HIT. Stead and coworkers (1984) reported a striking cluster of six patients with pulmonary embolism complicating HIT identified within a few weeks at one institution. A particular lot of bovine lung heparin in use in the operating room was linked to these events: patient serum-induced platelet aggregation occurred in the presence of this particular lot of heparin, but not when other lots of bovine lung heparin from the same manufacturer were used.
Was this article helpful?