Abstract

Although severe deficiency in ADAMTS13 activity (<5%) can result in the development of thrombotic thrombocytopenic purpura (TTP), a less marked reduction in ADAMTS13 occurs in the acute phase reaction of sepsis and DIC, and may promote thrombosis in various disease states. Heparin induced thrombocytopenia (HIT) shares many important similarities with TTP. Both are autoimmune prothrombotic disorders characterised by pathogenic antibodies, marked endothelial cell activation, thrombocytopenia, and the development of thrombosis. ADAMTS13 parameters in HIT have not been well characterised and there is no consensus on the severity of the reduction in ADAMTS13 activity. We hypothesize that a pronounced reduction in ADAMTS13 activity can occur in HIT and investigated potential mechanisms underlying ADAMTS13 depletion. We evaluated ADAMTS13 activity, anti-ADAMTS13 IgG and VWF:Ag in double-spun citrated plasma from 20 patients (median age 64, Male:Female = 14:6). 15 were HIT positive (of whom 10 were cardiothoracic cases) and 5 were suspected HIT thrombocytopenic cases, but HIT negative. HIT antibodies were measured by ELISA (either: PF4 enhanced, GTI Diagnostics; or Asserachrom HPIA, Diagnostica Stago) and platelet aggregation (heparin dependence was demonstrated using a known responder as the platelet donor). ADAMTS13 activity was determined by a modified residual collagen binding assay, and anti-ADAMTS13 IgG by an in-house ELISA using recombinant ADAMTS13-coated microtitre plates. VWF:Ag was measured using an immunoturbidimetric assay (Dade Behring). The median HIT antibody concentration in the 15 HIT cases was 1.57 OD using PF4-enhanced ELISA (range 0.91–2.44), and 143% using HPIA (range 69–181%). The median platelet nadir was 57×109/L (range 6–132×109/L). ADAMTS13 activity (normal range (NR): 66–126%) was reduced in 15/15 (100%) HIT positive patients (median 17%, range 0–55%), but also in 4/5 (80%) HIT negative thrombocytopenic patients (median 31%, range 18–76%). There was no statistically significant difference between these two groups (p=0.12). Furthermore, the ADAMTS13 activity of the 10 cardiothoracic patients was not significantly different from the remaining HIT positive cases. Anti-ADAMTS13 IgG was detected in 3/15 HIT positive patients (with IgG levels of 12%, 28% and 69% [NR <4%]) and 1/5 HIT negative cases (IgG of 21%). VWF:Ag was elevated (>200 iu/dL) in 18/20 with a median VWF:Ag of 297 iu/dL (range 165–606 iu/dL). These observations show that ADAMTS13 activity is often low in HIT patients, and can be markedly reduced in some cases. However, ADAMTS13 IgG autoantibodies were only detected in 3 cases, suggesting that autoantibody production is not the predominant mechanism for reduced ADAMTS13 function. The high VWF:Ag observed in HIT is an acute phase reactant and represents excess ADAMTS13 substrate, which drives the process of VWF proteolysis and causes circulating ADAMTS13 depletion. This may ultimately promote the prothrombotic milieu seen in HIT.

Author notes

Disclosure: No relevant conflicts of interest to declare.