Harry L. Messmore

Synthetic pentasaccharides do not cause platelet activation by antiheparin-platelet factor 4 antibodies.

A synthetic selective inhibitor of factor Xa, the pentasaccharide SR90107A/Org31540 is in clinical develop ment for the prophylaxis of postsurgical deep vein thrombosis. Another synthetic pentasaccharide with even more sustained inhibition of factor Xa, SanOrg34006, has also been developed. Both of these agents were tested in comparison to unfraction ated heparin and a low molecular weight heparin (enoxaparin) for their relative platelet activation potential in heparin-induced thrombocytopenia assays. Sera from patients (n = 30) with heparin-induced thrombocytopenia were pooled and validated for heparin-dependent aggregation responses. Using heparin- platelet factor 4 Sepharose columns, antibodies to heparin- platelet factor 4 were purified from the same pool. The effects of heparin, enoxaparin, SR90107A/Org31540, and San Org34006 were evaluated in a platelet aggregation assay using platelet donors (n = 10). At comparable concentrations, hep arin and enoxaparin consistently produced platelet activation, whereas both pentasaccharides failed to produce a response at a concentration up to 100 μg/mL (∼50 μM). Similarly, in the 14C-serotonin release and flow cytometric assays, heparin and enoxaparin produced positive responses (n = 30), whereas the two pentasaccharides consistently failed to produce any effect. These observations suggest that the two pentasaccharides with highly selective anti-Xa activity are devoid of generating anti heparin-platelet factor 4 antibody, do not produce heparin- induced thrombocytopenic responses and may inhibit active heparin-induced thrombocytopenia antibody platelet activation.

Mechanisms of Venous and Arterial Thrombosis in Heparin-Induced Thrombocytopenia

Since the reports by Weismann and Tobin in 1958 and Roberts et al. in 1964 called attention to paradoxical thrombosis in patients treated with heparin, the thrombotic aspect of the heparin-induced thrombocytopenia syndrome (HIT) has been emphasized. Yet to this day, the mechanism of thrombosis associated with HIT (HITT) is unclear. It is important to understand the etiology of HITT because of its devastating clinical consequences. We believe one rational approach to understand the mechanism underlying HITTS is to invoke Virchows triad: stasis, vascular injury and a hypercoagulable state. A hypercoagulable state exists in all HIT patients due to platelet activation by heparin antibody binding. Thrombin generation from platelet microparticles and exposed platelet phospholipid, coupled with stasis (elderly bedridden or otherwise sedentary ill patients who comprise the majority of the HIT population), provide two risk factors that can lead to venous thrombosis. A hypercoagulable state coupled with endothelial cell dysfunction due to injury from heparin antibody, activated platelets, leukocytes, platelet microparticles, complement, atherosclerosis or medical intervention can lead to arterial thrombosis. Of patients with HIT, HITT occurs in about 25%, suggesting that a second set of patient specific risk factors, in addition to the generation of pathological heparin antibodies, determine whether HITT will develop. Interaction between activated platelets and other platelets, and with endothelial cells, leukocytes, neutrophils, monocytes and cytokines are areas of research that may provide more specific characterization of the hypercoagulable state and vascular damage. Nuances involving genetic variation in platelets, endothelial cells and immune function are also likely to be a major component of the observed variability of this disease spectrum. Virchows triad may explain the different manifestations of HITTS.

Rivaroxaban - an oral, direct Factor Xa inhibitor - has potential for the management of patients with heparin-induced thrombocytopenia

Rivaroxaban is an oral, direct activated Factor Xa (FXa) inhibitor in advanced clinical development for the prevention and treatment of thromboembolic disorders. Currently available anticoagulants include unfractionated heparin (UFH) and low molecular weight heparins (LMWHs); however, their use can be restricted by heparin‐induced thrombocytopenia (HIT). HIT is usually caused by the production of antibodies to a complex of heparin and platelet factor‐4 (PF4). This study was performed to evaluate, in vitro, the potential of rivaroxaban as an anticoagulant for the management of patients with HIT. UFH, the LMWH enoxaparin, fondaparinux and the direct thrombin inhibitor argatroban were tested to enable comparative analyses. Rivaroxaban did not cause platelet activation or aggregation in the presence of HIT antibodies, unlike UFH and enoxaparin, suggesting that rivaroxaban does not cross‐react with HIT antibodies. Furthermore, rivaroxaban did not cause the release of PF4 from platelets and did not interact with PF4, unlike UFH and enoxaparin. These findings suggest that rivaroxaban may be a suitable anticoagulant for the management of patients with HIT.

Activation of platelets by heparin-induced thrombocytopenia antibodies in the serotonin release assay is not dependent on the presence of heparin.

Summary.  The serotonin release assay (SRA) tests for antibodies responsible for heparin‐induced thrombocytopenia (HIT). By definition, SRA‐positive antibodies cause platelet serotonin release in vitro, in the presence of low concentrations of heparin, but not with excess heparin. Many SRA‐positive sera activate platelets in the presence of saline without drug, either as a result of residual heparin in the specimen, or because of intrinsic features of the HIT antibodies. The present experiments show that neither exhaustive heparinase treatment, nor chromatographic removal of heparin abrogates the spontaneous platelet activation caused by these HIT antibodies. This is the first study to systematically demonstrate that in vitro activity of HIT antibodies can be independent of heparin. In addition, T‐gel chromatography demonstrated differences among fractions of enzyme‐linked‐immunosorbent assay (ELISA)‐positive HIT antibodies within individual specimens. Certain ELISA‐positive fractions had SRA activity while others did not, and the SRA activity was not proportional to HIT antibody ELISA titer. These data suggest that antibodies formed as a result of heparin treatment are heterogeneous, and that some can contribute to the pathogenesis of HIT even when heparin is no longer present.

Biochemical and pharmacologic heterogeneity in low molecular weight heparins. Impact on the therapeutic profile.

Ever since the introduction of low molecular weight heparins (LMWHs) for clinical use, one of the major questions raised relates to product interchangeability and the differences between each of the individual LMWH preparations. Although differences between various commercially available products have been described in terms of molecular weight profile and biologic properties, very limited information on the direct comparison of individual products in a defined clinical setting is available at this time. European Pharmacopeia (EP) and the World Health Organization (WHO) have developed guidelines to characterize these agents in terms of molecular weight and biologic profiles. On a gravimetric basis, these potency assignments differ for anti-Xa and anti-IIa activities in terms of U potency per mg. The relative distribution of various molecular weight components has also been reported to vary. The oligosaccharide composition, microstructural differences in terms of specific sugars and the presence of unique structural features and the interaction with endogenous mediators such as antithrombin (AT) and heparin cofactor II (HC II) also differ. At equivalent anti-Xa levels, the amount of the anti-IIa activity and anticoagulant activity differs. Since the bioavailability and relative pharmacokinetics of the anti-Xa and anti-IIa effects are different, the specific pharmacodynamic effects of these drugs also differ. A large preclinical data base is now available on the differences between various LMWHs. However, only limited clinical data is available in the current literature. To date, the LMWHs have been primarily used for the management of post-surgical DVT. Only smaller dosages (30-40 mg or 2,500 to 4,000 anti-Xa U total dose) have been used. In these studies, because of the low dose and subcutaneous route of administration, the differences in clinical effects are rather small. Since LMWHs are now developed for therapeutic use, where relatively higher doses are used, these pharmacokinetic/pharmacodynamic differences will become more apparent. The reported differences in the clinical efficacy of LMWHs in such indications as unstable angina may be due to their pharmacologic properties and molecular composition. There are also major differences in the non-anticoagulant actions of these agents such as their ability to interact with growth factors and antithrombotic effects. Based on the available literature, it can be concluded that each product exhibits individuality.

Prevalence, isotype, and functionality of antiheparin-platelet factor 4 antibodies in patients treated with heparin and clinically suspected for heparin-induced thrombocytopenia. The pathogenic role of IgG.

Antibodies to heparin-platelet factor 4 (PF4) complexes have been observed in patient with heparin-induced thrombocytopenia (HIT) syndrome. These antibodies may be of various isotypes and differ with respect to their ability to activate platelets/endothelial cells. This study determined the isotypes and functionality of antiheparin-platelet factor 4 (AHPF4) antibodies in 111 patients treated with heparin and clinically suspected for HIT. In this patient population, 50% had detectable AHPF4 cumulative IgA, IgG, and IgM (determined by enzyme-linked immunosorbent assay, ELISA), but only 35% was positive when tested with the (14)C-serotonin release assay (SRA). Using antihuman Ig specific for different isotypes, we found that 50% of the 111 samples was positive for IgG, 45% for IgM, and 37% for IgA. In 50 normal human serum (NHS) samples, only two were positive for IgG, but 33 were positive for IgM, indicating a potential humoral response to the heparin-PF4 complex prior to heparin administration. Patients that were ELISA(+) for AHPF4 antibody titer were subdivided into SRA-positive (+) and SRA-negative (-) groups. The SRA(+) group had a mean ELISA optical density (OD) for AHPF4 IgA/IgG/IgM of 2.1, while the SRA(-) group had a mean OD of 0.8 (P<.001). The SRA(+) group had greater mean OD values for all three individual isotypes. Using flow cytometry, we determined the ability of different patient samples to activate platelets. Samples that contained IgG and were SRA(+) activated platelets (as measured by microparticle generation and P-selectin expression) in the presence of therapeutic concentrations of heparin. NHS and samples containing IgA and/or IgM that were SRA(-) were not able to produce microparticles nor were they able to increase expression of P-selectin. Together, these data indicate that IgG is the principal mediator of platelet activation in patients with HIT, with IgA and IgM playing a less significant role in the pathophysiology of this syndrome.

POTENTIAL USE OF RECOMBINANT HIRUDIN AS AN ANTICOAGULANT IN A CARDIOPULMONARY BYPASS MODEL

Recombinant (r) hirudin is a potent thrombin-specific inhibitor derived from the natural hirudin of the leech (Hirudo medicinalis). We have studied the efficacy of r-hirudin compared with heparin in a canine model of cardiopulmonary bypass operations. Two administration regimens were used for r-hirudin: group 1, 1.0 mg/kg intracardiac bolus then intravenous bolus at 30 minutes (n = 10); and group 2, 1.0 mg/kg intracardiac bolus with 1.25 +/- 0.04 mg.kg-1.h-1 intravenous infusion (n = 8). Group 3 was given an intracardiac bolus of heparin, 1.66 mg/kg (n = 9). Aspiration of blood from the chest cavity revealed no significant difference between the three groups. Measurement of fibrin deposits in the pump line filter revealed higher amounts in the r-hirudin groups (p = 0.02). Decreases in platelets, fibrinogen, and hematocrit due primarily to hemodilution were the same in each group. The bleeding time assay showed less prolongation for r-hirudin than for heparin (p less than 0.001). No antagonist for r-hirudin was used; however, due to its short half-life all coagulation variables returned to baseline within 30 minutes after cardiopulmonary bypass. Because r-hirudin lacks effect on platelets, is a poor immunogen, does not require a plasma cofactor, and may not require an antagonist, it may provide an alternative anticoagulant to heparin in cardiopulmonary bypass. Additional studies are, however, needed to optimize the dose and to evaluate other clinical aspects of r-hirudin.

Circadian interrelationships among levels of plasma fibrinogen, blood platelets, and serum interleukin-6.

Circadian (24 h) rhythms of fibrinogen, interleu kin-6 (IL-6), and platelet levels were studied in 11 males ages 46 to 72 years. Since there is a known circadian rhythm for fibrinogen and IL-6, we postulated that the peak level (acro phase) of fibrinogen would follow the acrophase of IL-6, based on the fact that IL-6 is the stimulus for fibrinogen production in the liver. Platelet levels were measured to show whether there was any correlation with the IL-6 acrophase because it has been reported that IL-6 affects megakaryocytes and platelets in dogs. We found that the acrophase for IL-6 occurred at 02:03 h and the acrophase for fibrinogen occurred at 09:16 h. Platelet counts peaked at 16:56 h. Thus, there was a positive correlation between IL-6 and fibrinogen acrophases and a negative corre lation of each with the acrophase for platelets. The positive linkage of IL-6 with fibrinogen in this study suggests that sup pression of IL-6 production would lower those peak fibrinogen levels that occur in the morning in association with arterial ischemic events. This could result in fewer arterial ischemic events, especially in the morning. Key Words: Circadian— Fibrinogen—Interleukin—6 (IL-6)—Platelets—Arterial ischemia.

Day-night variations in blood levels of nitric oxide, T-TFPI, and E-selectin

Circadian (8/24 hours) variations in serum nitric oxide (NO), total tissue factor pathway inhibitor (T-TFPI), and E-selectin levels were studied in healthy adults and in subjects with type II diabetes. We postulated a possibility a functional relationship between them because vascular endothelium is the primary site of their synthesis and functions. NO is released by the action of eNO synthase isoform and modulates physiologic responses (e.g., vascular dilation, relaxation, increasing blood flow, inhibition of platelet and white blood cell adhesion); T-TFPI, a coagulation inhibitor, is also released from endothelial cells, and is bound to plasma lipoproteins and to glycosaminoglycans; E-selectin is expressed on endothelial cells after activation by inflammatory cytokines (interleukin-1β and tumor necrosis factor-α) and elevated levels have been reported in a variety of pathologic conditions, including diabetes. We found that obese diabetic subjects had greater mean concentrations of NO and E-selectin than healthy men, 39.25 versus 12.71 μM and 81.51 versus 26.03 ng/mL, respectively. The T-TFPI levels were essentially similar in both groups of men, 47.10 versus 48.76 ng/mL. We observed that the time of peak concentrations of T-TFPI and E-selectin was similar to the timing of NO trough levels, suggesting a possible functional relationship. It may be hypothesized, therefore, that the higher concentrations of NO, unbalanced by increases in T-TFPI and E-selectin, may result in increased vascular wall uptake of lipoproteins in diabetic subjects, who are at greater risk than healthy men for developing diffuse atherosclerosis.

Simple scoring system for early management of heparin-induced thrombocytopenia.

This study was performed to develop a simple scoring system to aid in the early clinical management of patients suspected of heparin-induced thrombocytopenia (HIT) with regard to decisions for continued heparin therapy. The system was designed to arrive at low (0) or possible (1) probability scores without knowledge of laboratory test results (except platelet counts) to avoid delays. As the safest clinical approach is to discontinue heparin, intermediate and high scores were combined. Critically ill VA hospital patients (n = 100) with a ≥30% fall in platelet count were assessed by platelet aggregation (PA), 14C-serotonin release assay (14C-SRA), and GTI ELISA. In this population, 53% were scored 1 and of these 43% were positive by laboratory test. Emphasizing the decision to discontinue heparin, the clinical signs of HIT were paramount for the immediate determination of a diagnosis of HIT without dependence on a positive laboratory test.