Abstract 1069

Poster Board I-91

Oversulfated Chondroitin Sulfate (OSCS) has been reported to be the main contaminant responsible for the adverse reactions and deaths associated with the use of heparin in 2007/2008. However, limited information is available on its origin and biologic actions. OSCS can be produced utilizing porcine, bovine, squid, and shark chondroitin sulfate (CS). Other mammalian and marine sources can also be used to produce OSCS. To test the hypothesis that OSCS from different origins may vary in their molecular profile and biological actions, starting material from porcine origin (PCS) was compared with OSCS produced from bovine (BOSCS), porcine (POSCS), squid (SqOSCS), and shark (SOSCS) chondroitin sulfate. In addition, OSCS was isolated from several contaminated batches of heparins and two low molecular weight heparin (LMWH) preparations. Some of the contaminated unfractionated heparins were also found to contain significant amounts of high molecular weight dermatan sulfate. The molecular profile studies were carried out using HPLC and NMR. The biologic studies included anticoagulant profile, contact activation by measuring generation of kallikrein from prekallikrein, SERPIN binding profile, and platelet activation studies including HIT antibody screening and C14 serotonin release. Noncontaminated unfractionated heparin was used as a reference material. In comparison to UFH (MW 17.2 kDa), the starting material CS exhibited a relatively high molecular weight (41.4 kDa). The other oversulfated agents exhibited a molecular weight range of 17-25.4 kDa. The hypersulfated dermatan sulfate (HDS) exhibited a molecular weight of 41.6 kDa. The NMR spectra of each of these agents exhibited the characteristic structural signals. While the CS and HDS exhibited relatively weaker anticoagulant activity (<5 USP units/μg), the OSCS from different species exhibited assay and concentration dependent anticoagulant effects in the PT, aPTT, Heptest, and TT (25-40 U/mg). Unlike heparin none of the OSCS preparations showed any interactions with antithrombin but exhibited varying degrees of interactions with heparin cofactor II. The contaminants isolated from different batches of UFH did not exhibit a comparable molecular profile to heparin, with a range of 12.8- 21.1 kDa. Moreover, their anticoagulant responses and interactions with SERPINs considerably varied. In the contact activation studies all of the contaminants exhibited a relatively higher activation of prekallikrein to kallikrein. The contaminants isolated from LMWHs were found to be somewhat lower in molecular weight in comparison to contaminants isolated from UFH preparations. However, their biologic profile was similar or stronger than those isolated from UFH. In the HIT antibody screening, PCS and HDS produced weak responses whereas the OSCS produced stronger activation of platelets. The contaminants isolated from heparins also produced stronger HIT antibody mediated aggregation of platelets. Similar results were obtained in the serotonin release assays. These results clearly demonstrate that the contaminants isolated from recalled heparins are heterogeneous in nature and may be responsible for the variations noted in the patients treated with these contaminated batches of heparin.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.