• danaparoid;
  • direct thrombin inhibitors;
  • fondaparinux;
  • heparin-induced thrombocytopenia;
  • immune response;
  • PF4/heparin complexes


  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References

Summary.  The current major problem with HIT is its overdiagnosis. This concept follows from the HIT central paradigm: HIT is caused by a subset of antibodies against platelet factor 4 (PF4)/heparin complexes that have strong platelet-activating properties. Prospective studies show that only a minority of sera containing such antibodies exhibit platelet-activating properties. Ironically, the earliest tests for HIT – platelet activation assays – remain today the most diagnostically useful, particularly the washed platelet assays. But the wider application of PF4-dependent immunoassays, and their much greater sensitivity for the larger subset of non-platelet-activating (and non-HIT-inducing) antibodies, has resulted in HIT overdiagnosis in many centres. Studies of anti-PF4/heparin immunization in diverse clinical situations have provided insights into the factors that influence the HIT immune response. Besides the conundrum of anticoagulant-induced thrombosis (including its potentiation of coumarin-induced microthrombosis), HIT evinces numerous other paradoxes: (i) it is a platelet-activating disorder with venous thrombosis as its predominant clinical manifestation; (ii) ‘delayed-onset’ (or ‘autoimmune’) HIT can lead to dramatic worsening of HIT-associated thrombosis despite cessation of heparin; (iii) partial thromboplastin time (PTT) monitoring of direct thrombin inhibitor treatment – and confounding of PTT monitoring by HIT-associated consumptive coagulopathy – infers that the worst subset of HIT patients may fail this therapeutic approach; (iv) the highly sulfated pentasaccharide anticoagulant, fondaparinux, can (rarely) cause HIT yet appears to be an effective treatment for this disorder; and (v) the transience of the HIT immune response means that many patients with previous HIT can safely receive future heparin.


  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References

Since its first description in 1973 by Rhodes, Dixon, and Silver [1], immune heparin-induced thrombocytopenia (HIT) has continued to fascinate clinicians and scientists alike. Its peculiar and enigmatic nature led many at first to doubt its existence; even those who accepted HIT wondered whether it really could cause thrombosis [2]. After all, that an anticoagulant causes thrombosis seems as counter-intuitive as anything one might find in clinical medicine.

Only a little over a decade ago, three investigators from three continents – Beng H. Chong, Andreas Greinacher, and myself – generated a consensus statement [3] articulating what can now be regarded as the ‘central paradigm’ of HIT: that it comprises a clinicopathologic syndrome requiring (i) one or more clinical events (primarily thrombocytopenia); and (ii) laboratory evidence for a heparin-dependent immunoglobulin (usually IgG) using a sensitive and specific assay [3]. Moreover, we stated that this pathologic IgG activates platelets via their Fc receptors [4,5]. The term proposed for this syndrome, ‘heparin-induced thrombocytopenia’, pointed to the central role of heparin in initiating this syndrome.

Viewing the topic now 13 years later, the essential elements of this central paradigm remain, but there is now a growing realization that those patients who form pathogenic platelet-activating antibodies – and thereby develop HIT – comprise only a relatively small subset among the heparin-exposed patients who develop an anti-PF4/heparin immune response [6–10]. Within the past 10–20 years, recognition of HIT has evolved from gross underdiagnosis to wild overdiagnosis. In essence, the widespread detection of anti-PF4/heparin antibodies by commercially-available PF4-dependent immunoassays has promoted an overdiagnosis phenomenon [11].

This review begins with the central paradigm of HIT, and examines the evidence that detecting platelet-activating antibodies is most useful for diagnosis. The second section will deal with five key paradoxes of HIT, and their clinical relevance.

HIT central paradigm

  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References

The HIT central paradigm in 2011:

HIT is caused by a subset of antibodies of IgG class that produce strong activation of platelets via their FcγIIa (IgG) receptors; these antibodies recognize large multimolecular complexes of PF4 bound to heparin, and are already detectable in patient serum/plasma at the beginning of the HIT-associated platelet count decline. In (almost) all cases, patients have undergone a proximate immunizing exposure to heparin or certain other PF4-binding polyanions.

Platelet activation assays using platelet-rich plasma

The platelet-activating characteristics of HIT were evident at its discovery. Dr. Glen Rhodes was a medical student at Duke University when he found that plasma from a thrombocytopenic patient with recurrent pulmonary embolism despite heparin treatment caused platelets (within citrated plasma) to aggregate in the presence of heparin (Fig. 1).


Figure 1.  HIT antibody-induced platelet aggregates from an experiment in the early 1970’s. Magnification approximately 250–500×. Photomicrograph kindly provided by Dr. Glen R. Rhodes, Palmyra, VA.

Download figure to PowerPoint

Studies using platelet-rich plasma are not used so often anymore for diagnosis of HIT, primarily because of the issue of lower sensitivity [12].

Washed platelet activation assays

Dave Sheridan – working in the laboratory of John Kelton at McMaster University – wanted to preserve precious HIT sera and so adapted a microtiter assay of ‘washed platelets’ from a method developed by Fraser Mustard [2], and the platelet serotonin-release assay (SRA) was born [13]. The characteristic profile of platelet activation at 0.1–0.3 U mL−1 heparin and inhibition by high heparin (100 U mL−1) presaged the discovery of the multimolecular and stoichiometrically precise relationship between PF4 and heparin: at very high heparin concentrations, the stoichiometry is wrong for antigen formation.

The microtiter format of the assay allowed assessment of parallel samples and controls in numbers high enough for quality control, as well as to study the biological activity of sera from HIT patients.

This enabled my first research paper on HIT, exploring heterogeneity in the platelet activation responsiveness to HIT sera among different platelet donors (in a hierarchical fashion), and the recognition that there are ‘strong’ and ‘weak’ HIT sera. Thus, the idea evolved that selecting platelet donors based upon reactivity, and utilizing both weak and strong positive HIT serum controls, ensure that HIT sera can be reliably detected [14].

The ‘dichotomizing’ SRA

Figure 2 illustrates the ‘dichotomizing’ nature of the SRA, where most sera yield clear-cut results, either negative or strong positive, with relatively few samples in-between [15]. And because HIT sera also react in a hierarchical fashion – the inclusion of weak-positive controls means that one has an easy quality control step to ensure the assay is sensitive enough to detect patients with HIT [14].


Figure 2.  Dichotomization of results of serotonin-release assay (SRA) testing for HIT antibodies (n = 405 patients tested). The data are shown as deciles of mean percent serotonin-release (at 0.1 and 0.3 IU mL−1 unfractionated heparin). The conventional cut-off defining a positive SRA test result is 20%; however, the Author generally uses 50% as the cut-off (assuming all controls react as expected, including weak-positive control serum), as this better discriminates between HIT and non-HIT thrombocytopenia. Only approximately 10% of patients investigated for HIT achieved a positive test result at this cut-off. Overall, > 97% of patients in this data set tested either clearly negative (< 20% serotonin-release) or strongly-positive (≥ 80% serotonin-release). Reprinted, with permission, from [15].

Download figure to PowerPoint

Utility of the SRA

The power of the SRA became evident when it was applied to a clinical trial of unfractionated heparin (UFH) vs. low-molecular-weight heparin (LMWH) therapy [16]. We had available for study 387 patient sera, and in blinded studies found 20 sera that tested positive in the SRA (at the threshold of > 50% serotonin-release). Six of these patients met our definition of HIT (platelet count fall to < 150 × 109 L−1 that began after day 5 of heparin treatment). This compared with only 2 of 367 patients who were SRA– (both patients had clear non-HIT explanations for their thrombocytopenia). The odds ratio (OR) for developing thrombocytopenia (< 150 × 109 L−1) among SRA+ patients in this study was very high: 78.2 (95%confidence interval, 12.0–818.8); P < 0.001.

Later, in the section, ‘Fondaparinux: can it cause HIT? does it treat HIT?’, we will review a more recent study [17] that further shows the much greater ability of the SRA compared with the PF4-dependent enzyme-immunoassay (EIA) to distinguish between clinically-relevant and irrelevant antibodies.

Refining the definition of ‘thrombocytopenia’ in HIT

Some of the non-thrombocytopenic SRA+ patients did develop substantial platelet count declines that did not, however, fall to < 150 × 109 L−1. Working with a biostatistician, Professor Rob Roberts, we used SRA+ status as the ‘gold standard’ and examined various definitions of thrombocytopenia: we found that a > 50% fall in the platelet count (beginning at the peak platelet count after postoperative day 4) correlated even better with SRA+ status [18]. Thus, the OR for SRA+ status to predict > 50% platelet count fall was even higher: 103.1 (95%CI, 23.6–500); P < 0.001) [19]. This concept of a relative platelet count fall is especially relevant to HIT because surgery is an important risk factor for HIT, and thus HIT can complicate postoperative thrombocytosis.

Another washed platelet activation assay for HIT

In parallel, in Europe, Andreas Greinacher experimented with platelet washing, and found he could do away with the requirement for radioactive serotonin: after all, the platelets could be seen to aggregate macroscopically, and so a semi-quantitative assay based upon visual inspection of washed platelet aggregation – the heparin-induced platelet activation (HIPA) assay – was born [20].

It is interesting to consider the different approaches to platelet activation testing for HIT in North America vs. continental Europe. In the former, pedigree platelet donors are utilized; in the latter, transfusion centres use random volunteers who donate blood; to compensate for the risk of unsuitable (non-reactive) platelet donors, the HIPA is usually performed with four different donors, with two or three positives (depending on the centre) required to define a positive test result. A few centres perform platelet aggregation testing using platelets in citrated plasma; although diagnostic specificity is good, approximately 75%–80% of HIT samples fail to yield positive results (suboptimal diagnostic sensitivity) [12].

PF4-dependent antigen assays

The seminal discovery by Jean Amiral et al. [21] – soon confirmed by others [22–24] – that the major antigen of HIT was PF4 bound to heparin pointed to a new way forward. Surely, by knowing the specific target of the platelet-activating antibody, diagnosis of HIT would greatly improve. But, this goal has proven elusive.

What was found when the new PF4-dependent EIAs were applied to the archived study material [7–9,16,18]? Basically, the old platelet activation assay had already detected all of the HIT cases, which the new EIAs also identified. However, the new assays also detected anti-PF4/heparin antibodies among numerous non-HIT patients. Figure 3 show these data presented as an ‘iceberg’ [7,8,25]. These and other studies of heparin-exposed patients show that the majority of anti-PF4/heparin antibodies are not platelet-activating [6–10].


Figure 3.  Iceberg model of HIT. The top panel depicts a generic ‘iceberg’, illustrating the interrelationship of clinical HIT with formation of anti-PF4/heparin antibodies detected by the platelet serotonin-release assay, an in-house EIA-IgG, and a commercial polyspecific EIA (from GTI Diagnostics, Inc. Waukesha, WI, USA) that detects antibodies of all three major immunoglobulin classes (IgG, IgA, and IgM) against PF4/polyanion (EIA-GAM). The lower panel illustrates the event rates for clinical HIT (HIT), including the subgroup with HIT-T, in relation to the frequencies of antibody formation, in three clinical settings: UFH or LMWH thromboprophylaxis during orthopedic surgery (data from a randomized controlled trial) and UFH thromboprophylaxis postcardiac surgery (prospective cohort study). + indicates positive test. Abbreviations: EIA-GAM, enzyme-immunoassay that detects IgG, IgA, and IgM antibodies against PF4/polyvinylsulfonate complexes; EIA-IgG, enzyme-immunoassay that detects IgG antibodies against PF4/heparin complexes; HIT, heparin-induced thrombocytopenia; HIT-T, HIT-associated thrombosis; LMWH, low-molecular-weight heparin; SRA, serotonin-release assay; UFH, unfractionated heparin. Reprinted with permission from [25].

Download figure to PowerPoint

Why is the EIA so frequently positive in the absence of HIT? One reason is the frequent presence of non-platelet-activating IgA and IgM class antibodies [26,27]. Another is the presence of non-platelet-activating IgG [28]. Also, antibodies sometimes are detected against PF4 alone [29] and, perhaps, impurities within the target antigen.

Improvements in the EIA  There are ways that the operating characteristics of the EIA can be improved: (i) consider the magnitude of a positive reaction, expressed in optical density (OD) values [15,30]; (ii) use IgG-specific assays [8,10,31,32]; (iii) use the high heparin inhibition step (‘confirmatory maneuver’) to increase specificity [33] – although this can be problematic at high ODs [34–36]. But, none of these improvements enhance the EIAs to the diagnostic power of the dichotomizing washed platelet activation assays.

Timeline of the HIT immune response

Antigen assays – due to their high sensitivity – allow for the characterization of the immune response to PF4/heparin complexes. Serial plasma samples were available from two orthopedic surgery thromboprophylaxis trials – one of which had several patients who developed HIT – which permitted studies of anti-PF4/heparin seroconversion, including its relationship to development of HIT [37,38]. These studies showed that there is a characteristic ‘timeline’ to the anti-PF4/heparin (and corresponding HIT) immune response:

  • 1
     HIT antibodies are generated rapidly (detectability, median day 4) and without IgM class precedence, suggesting HIT is an atypical – perhaps secondary – immune reaction;
  • 2
     There is a specific sequence of events: antibody generation (day 4) followed by the onset of the platelet count fall (median, day 6) followed by a > 50% fall (median, day 8) followed by thrombosis (median, day 10) (Fig. 4);
  • 3
     The timeline indicates that HIT antibodies are readily detectable within patient serum/plasma at the onset of the platelet count fall attributable to HIT; and
  • 4
     HIT exhibits a ‘point immunization’ phenomenon: most seroconversion events occur in a very narrow window after an immunizing exposure to heparin.

Figure 4.  Characteristic timeline of HIT: anti-PF4/heparin antibodies (by EIA) per postoperative day in 12 patients with HIT and 36 seropositive non-HIT control patients. (A) Mean (±SEM) optical density (OD) of anti-PF4/heparin antibodies detected using commercial immunoassay (EIA-GAM) from GTI Diagnostics Inc. that detects antibodies of all three immunoglobulin classes (IgG, IgA, IgM). HIT patients are indicated by bsl00001, and seropositive non-HIT controls by □. On each day beginning on postoperative day 6, there is a significant difference in the mean of the OD levels between the patients with HIT and the seropositive non-HIT controls (P < 0.05 by nonpaired t test). At the top of the figure, summary data for 12 HIT patient profiles are shown for four key events (first day of antibody detection, beginning of HIT-related platelet count fall, platelet count fall ≥ 50%, and thrombotic event), summarized as median (small red squares within rectangles), IQR (rectangles), and range (ends of thin black lines). (B) Mean (±SEM) OD values of anti-PF4/heparin antibodies detected using an in-house immunoassay (EIA-Ig) that detects antibodies of the individual immunoglobulin classes, IgG (red circles), IgA (green triangles), and IgM (blue inverted triangles) for HIT (solid symbols) and non-HIT (open symbols). On each postoperative day beginning on day 5, there is a significant difference in the mean of the OD units for the IgG immunoassay between the patients with HIT and the seropositive non-HIT controls (**< 0.005 for days 6–10; *< 0.05 for days 5, 11, and 12). In addition, among the 34 non-HIT controls who tested positive for IgG antibodies, mean (±SEM) maximum OD values for the EIA-IgG were significantly greater in the eight patients who tested positive in the serotonin-release assay (SRA) compared with the 26 patients who tested negative in the SRA (1.30 ± 0.15 vs. 0.96 ± 0.07 units; = 0.025). Among the 20 patients who tested positive in the SRA, mean (±SEM) maximum OD values for the EIA-IgG showed a trend to higher levels in the 12 patients with clinical HIT, compared with the eight seropositive non-HIT controls (1.63 ± 0.09 vs. 1.30 ± 0.15 units; = 0.059). EIA indicates enzyme immunoassay; and HIT, heparin-induced thrombocytopenia. Reprinted, with permission, from [37].

Download figure to PowerPoint

Clinical sample investigation (CSI) hematology  Blood samples often remain within a laboratory for some days before they are discarded. Retrieving such samples prior to disposal is a useful tool for solving timing of seroconversion problems in HIT [39]. These observational studies have also corroborated the finding that the antibodies are readily detectable at the beginning of the platelet count fall indicating HIT [39,40], including patients with rapid-onset HIT [41]. These studies showed that rapid-onset HIT is not an anamnestic immune response (i.e., rapid formation of antibodies due to immune memory that resulted from a previous heparin exposure), but rather simply due to antibodies already being present (because of a recent heparin exposure) when the heparin is readministered [41].

HIT Overdiagnosis

Overdiagnosis of HIT is a real issue: there is risk of complications from non-heparin anticoagulants and from inappropriate withholding of heparin or other interventions during acute thrombocytopenia [42]. Further, management issues are complicated at the time of subsequent readmission to hospital when heparin therapy might otherwise be indicated.

How often do patients who are investigated for HIT really have HIT?  Based upon positive washed platelet activation assay status, in Greifswald, it is only approximately 6% [43]; in Hamilton, it is approximately 10%–12% [15]. The corresponding frequency of a positive EIA is approximately two to three times higher, approximately 15%–25%, respectively. Ironically, the frequency of anti-PF4/heparin antibody detectability is even higher when one performs a prospective ‘serosurveillance study’: here the frequencies are 25%–30% for postorthopedic surgery patients receiving UFH [7–9] and 50%–75% for postcardiac surgery patients [7,25,34]. Thus, there is a very high frequency of non-pathogenic antibody formation if sensitive EIAs are used to study blood samples taken more than a week after perioperative heparin exposure.

Overdiagnosis of HIT: assuming that EIA+ status equals HIT

How do EIA+/SRA− patients compare with EIA+/SRA+ patients when testing is performed because of clinically-suspected HIT? In one such study of 100 consultations for thrombocytopenia, there were 16 EIA+/SRA+ patients, 16 EIA+/SRA− patients, and 68 EIA−/SRA− patients [11]. Only the EIA+/SRA+ patients had a high frequency of thrombosis (11/16 = 69%). The other two groups had a low frequency of thrombosis (approximately 7% each), but a relatively high frequency of mortality (indeed, numerically higher than the HIT patients). This infers that EIA+/SRA− status does not indicate HIT (and risk for HIT-associated thrombosis), but rather the presence of a non-HIT thrombocytopenic disorder. In this regard, the higher mortality reflects the known high risk of mortality in thrombocytopenic patients [44].

The perils of a repeat assay for HIT

HIT occurs most often in postoperative patients, a population that invariably has early, transient postoperative platelet count declines. Since HIT usually occurs in an otherwise ‘well’ postoperative patient, the HIT-associated platelet count decline characteristically evinces a new platelet count fall that begins after resolution of the immediately preceding perioperative decline. Given that serological assays for HIT antibodies are already positive at the very beginning of this HIT-associated platelet count decline, this indicates that if a thrombocytopenic patient is found to be HIT assay-negative using a sensitive EIA or washed platelet activation assay, this should be considered as strong evidence against HIT as the explanation for the thrombocytopenia. Thus, if a repeat test is ordered on a routine basis 2 or 3 days later – as opposed to it being repeated because there is a new platelet count fall and/or a thrombotic event – then any positive test that is obtained will most likely simply be a non-HIT seroconversion event. This is one of the causes for HIT overdiagnosis at some centres. Indeed, as shown in a study from Greifswald, an early-onset and persisting thrombocytopenia in a post-cardiac surgery patient population is almost always not HIT, even when subsequent antibody assays are positive [45]. In contrast, when a new thrombocytopenic process is superimposed on a preexisting thrombocytopenia, and the patient tests positive for platelet-activating antibodies, then this could well indicate the presence of HIT [46]. Given the high mortality of patients with early postoperative thrombocytopenia (for non-HIT reasons), not recognizing this pitfall of routine repeat serology can lead to wrong attribution of patient morbidity and mortality to HIT.

Risk factors for HIT and for anti-PF4/heparin immunization

  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References

Space permits only a brief discussion of the various factors that appear to influence anti-PF4/heparin immunization and/or risk for HIT. These include: (i) type of heparin (UFH > LMWH > fondaparinux [16–18,25,47,48]; (ii) patient setting (surgical > medical > obstetrical; pediatric) [25,47]; (iii) extent of trauma (major > minor) [49]; and (iv) sex (female > male) [47]. In addition, there are several risk factors for anti-PF4/heparin immunization in which an association with risk of HIT per se possibly can be inferred [50]: (i) type of orthopedic surgery (elective knee > elective hip); (ii) timing of first-dose LMWH prophylaxis among elective hip replacement patients (post- > pre-operative); (iii) timing of first-dose heparin prophylaxis among trauma hip replacement patients (pre- > post-operative). How could one explain these various observations?

Point immunization

HIT appears to exhibit a ‘point immunization’ in which heparin exposure during a critical time period – intraoperative heparin (cardiac and vascular surgery) and early postoperative heparin (postoperative antithrombotic prophylaxis). Several considerations support this viewpoint. First, timing of onset is typically 5–10 days after starting heparin [41] even when patients continue heparin for > 10 days [51]. Second, PF4 is a key component of the antigen, and PF4 release is associated with platelet activation either induced by heparin [52] or that accompanies surgery [53]. Third, pre- (vs. post-) operative administration of LMWH appears to have different effects on anti-PF4/heparin immunization, depending on whether the situation is elective or post-trauma hip replacement surgery [50].


A model of optimal PF4/heparin ratios (‘stoichiometry’) could explain some of these features, including the role of a perioperative point immunization when stoichiometrically optimal concentrations of heparin:PF4 are most likely to occur [50,54,55]. For example, this model could explain why preoperative first-dose administration of LMWH was associated with a reduced immunization rate in elective hip replacement surgery: sequestration of PF4 in a non-immunizing setting by preoperative LMWH could make less PF4 available during the critical early postoperative period when point immunization usually occurs. In contrast, the higher immunization rate associated with pre- (vs. post-) operative first-dose LMWH in the setting of trauma hip surgery could reflect a ‘double-dose’ effect, where the initial LMWH injection is placed between two periods of PF4 release (first with trauma, then with surgery). Figure 5A presents a model of various factors that could influence immunization risk; Fig. 5B shows factors that influence subsequent ‘breakthough’ of HIT among immunized patients.


Figure 5.  Factors influencing immunogenicity and breakthrough of postoperative HIT. (A) Immunogenicity. ‘Point immunization’ occurs when PF4 release (increased with surgery) coincides with heparin (UFH, LMWH, or fondaparinux) administration, for example, first postoperative dose. Factors that enhance immunization risk (red +) include: inflammation, major (vs. minor) trauma, elective knee (vs. hip) surgery, pre- (vs. post-) operative first-dose LMWH (for trauma hip surgery) and high body-mass index (BMI); some of these (e.g., high BMI) increase the probability of achieving optimal stoichiometric ratios between PF4 and heparin. Factors that decrease immunization (green −) include pre- (vs. post-) operative first-dose LMWH (for elective hip surgery) and low BMI; these could result in suboptimal PF4:hep ratios (e.g., sequestration of PF4 when LMWH is given preoperatively). Immunization is most frequent with UFH and less with LMWH and fondaparinux. (B) Breakthrough. HIT-IgG begin to form a median of 4 days after beginning heparin, followed by platelet count fall (thrombocytopenia) and thrombosis. Antibody levels peak at approximately days 10–14, and begin to wane thereafter. ‘Heparin-dependent’ HIT-IgG are pathogenic when postoperative heparin is given, with potential to cause platelet activation and thereby HIT (‘cross-reactivity’) as follows: UFH > LMWH >> fondaparinux. Platelet-dependent factors probably also explain potential for ‘breakthrough’ of HIT, including FcγIIa receptor numbers (genotype?), platelet PF4 quantity and binding, and postoperative platelet hyperreactivity. Generation of procoagulant platelet-derived microparticles helps to explain risk of venous thrombosis with HIT. For unknown reasons, female sex is also a risk factor for HIT (not shown). The concept of pathogenicity of ‘heparin-independent’ HIT-IgG is discussed subsequently (see Fig. 6).

Download figure to PowerPoint

To test our model, we predicted that high body-mass index (BMI) individuals receiving postoperative enoxaparin would exhibit a higher frequency of immunization compared with low BMI individuals (rationale: the stoichiometrically optimal concentrations of enoxaparin in relation to expected postoperative PF4 levels were rather high in relation to usual blood concentrations [54], and thus large body weight individuals – who would be expected to have lower enoxaparin blood concentrations for fixed dosing – should have had higher immunization rates). The data supported this prediction: the immunization rate with LMWH was significantly greater for the highest BMI quartile than for the lowest quartile: relative risk = 3.11 (P = 0.026) [50].

‘Spontaneous’ HIT

Ironically, although almost all cases of HIT involve a proximate exposure to immunizing heparin (or some other PF4-binding polyanion), it now seems plausible that some patients develop clinical illness that is identical to HIT on both clinical and serological grounds, but in whom no proximate heparin exposure can be identified [56–59]. Although this syndrome has been called ‘spontaneous’ HIT, it is far from spontaneous, as it seems to require a preceding inflammatory or infectious trigger, albeit one unaccompanied by exposure to heparin. Several cases had preceding orthopedic surgery (with postoperative warfarin anticoagulation). Presumably, PF4 tetramers are somehow rendered immunogenic, either through exposure to a polyanionic bacterial surface [60,61] or to chondroitin sulfates released during orthopedic surgery or unknown mechanisms. Japanese investigators also reported that some patients had positive tests for anti-PF4/heparin antibodies at time of presentation of myocardial infarction [62].

HIT paradoxes

  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References

Besides the fundamental conundrum of anticoagulant-induced thrombosis, there are numerous other HIT paradoxes, five of which will be considered here.

The procoagulant platelet response and risk of coumarin necrosis

The earliest reports of HIT emphasized its association with arterial thrombosis [2], a concept that fit with its platelet-activating nature. However, in the 1990’s, it was recognized that HIT was more often associated with venous thrombosis [16,63]: indeed, the OR between HIT and venous thrombosis ranges from 20 to 40 [64]. This association could reflect the strong capacity of HIT antibodies to induce a platelet procoagulant response, including the formation of procoagulant microparticles [65–67]. Other factors include ‘pancellular’ activation that additionally involves monocytes [68] and endothelium [24,69]. Extreme hypercoagulability in HIT – as shown by very high thrombin-antithrombin complex levels – helps to explain its strong association with coumarin necrosis, particularly the variant known as venous limb gangrene [70,71]. This dual adverse drug reaction – macro and microvascular thrombosis leading to limb loss even in a patient with normal arteries, and occurring because of two distinct prothrombotic events caused by two different classes of anticoagulants – represents perhaps one of the greatest paradoxes in all of clinical medicine [72]. Inhibition of activated protein C generation by HIT antibodies could also be a relevant pathogenic factor [73].

Autoimmune HIT and the futility of stopping heparin

Approximately 12 years ago, Kelton and I [74] and Rice et al. [75] reported a syndrome named ‘delayed-onset HIT’. This was characterized by the onset of HIT that began several days after stopping heparin (Fig. 6). We found that serum from such patients reacted more strongly than control HIT sera both in the EIA and in the SRA. We also found that sera from patients with delayed-onset HIT could activate platelets even in the absence of pharmacologic heparin.


Figure 6.  Conceptual framework of HIT: focus on heparin-independent platelet activation and delayed-onset (‘autoimmune’) HIT. The upper panel shows the timeline of HIT antibody (HIT-Ab) formation, as judged by optical density units in an anti-PF4/polyanion EIA; the middle panel illustrates a platelet count decline in the absence of heparin (or with small amounts of heparin, e.g., ‘flushes’) indicating ‘delayed-onset’ HIT, with intensification of HIT-associated hypercoagulability from day 7 to 14, especially after stopping heparin. The lower panel compares different classes of anticoagulant for expected effects on HIT-associated hypercoagulability. Reprinted, with modifications, with permission from [80].

Download figure to PowerPoint

In recent years, it has become evident that the majority of HIT sera exhibit some degree of heparin-independent platelet activation [76,77]. In a recent study, we showed that this property correlated with slower platelet count recovery [78].

These observations also extend the scope of the delayed-onset HIT phenomenon. The strict definition of this syndrome is the onset of thrombocytopenia well after the last exposure to heparin. However, the presence of antibodies with heparin-independent activation is not dependent on how long heparin is continued. We have observed that progressive declines in platelet count after stopping heparin are also associated with these ‘autoimmune-like’ antibodies (Fig. 7).


Figure 7.  HIT-associated consumptive coagulopathy: confounding of PTT monitoring. The patient developed HIT beginning 6 days after starting dalteparin thromboprophylaxis post-hip fracture surgery. On day 11, the patient required admission to the intensive care unit for severe hypotension secondary to adrenal failure (flat ACTH stimulation test; CT imaging: bilateral adrenal infarction). HIT-associated bilateral lower-limb DVT was treated with argatroban begun at 1 mcg kg−1 min−1. Despite stopping heparin, the platelet count fell over 4 days from 80 to 18 × 109 L−1 (nadir). Five PTT measurements pre-argatroban were all elevated (range, 38–41 s; normal 22–35 s). Post-initiating argatroban, persisting supratherapeutic PTT values resulted in multiple argatroban dose reductions and interruptions; when HIT was most intense (as judged by platelet count nadir), argatroban dosing ranged from 0 to only 0.1 mcg kg−1 min−1. The laboratory and clinical profile is consistent with intense HIT-associated consumptive coagulopathy, with confounding of PTT monitoring of argatroban therapy and associated underdosing despite supratherapeutic PTT values. Abbreviations: Arg, argatroban; bid, twice-daily; DVT, deep vein thrombosis; F, female; HIT, heparin-induced thrombocytopenia; ICU, intensive care unit; INR, international normalized ratio; OD, once-daily; PTT, partial thromboplastin time; sc, subcutaneous; U, units; ULN, upper limit of normal. Reprinted with permission, from [78].

Download figure to PowerPoint

HIT-associated consumptive coagulopathy

Approximately 10%–20% of patients with HIT evince overt (decompensated) disseminated intravascular coagulation (DIC), with one or more of absolute (or relative) hypofibrinogenemia, elevated international normalized ratio (INR), and elevated partial thromboplastin time (PTT) [51], perhaps as a result of procoagulant changes to platelets [65,66], monocytes [68], and/or endothelium [69]. Among the most widely used alternative non-heparin anticoagulants to treat HIT are the direct thrombin inhibitors (DTIs), lepirudin and argatroban. Usually, a global coagulation assay, the PTT, is used to monitor these agents.

Perils of PTT-monitored DTI therapy  What happens when the PTT is used to monitor a patient with HIT-associated DIC? This can lead to inappropriate dose interruptions and/or underdosing [78–80]: this suggests another paradox, that the most severely affected HIT patients could be ‘untreatable’ by DTIs (at least, when monitored by the PTT; assays that measure DTI levels more reliably are described [81,82] but used in few centres). Figure 7 illustrates a patient with baseline (pre-argatroban) PTT elevation with persisting supratherapeutic PTT values despite drastic reduction (and intermittent cessation) of argatroban [78]. Other similar cases have been reported [79,80]. These observations may help to explain why limb loss rates in the DTI trials were not clearly lower than in the control groups; it may be that a subset of HIT patients with severe HIT-associated hypercoagulability fail DTI therapy.

This problem also raises the possibility that anticoagulants that are not monitored by the PTT – e.g., the Xa-inhibiting agents, danaparoid [83,84] and fondaparinux [85–87] – may be more appropriate management options for HIT (Fig. 6) [80].

Fondaparinux: does it cause HIT? can it treat HIT?

An interesting paradox relates to fondaparinux therapy [88]: this synthetic sulfated AT-dependent pentasaccharide inhibitor of factor Xa is associated with formation of anti-PF4/heparin antibodies (at least as frequent as seen with LMWH [48,50,89]), and, rarely, clinical HIT itself [90–92]. However, the probability of fondaparinux exacerbating HIT due to cross-reactivity of antibodies, although reported [93,94], is unlikely (probably < 1%). Indeed, in the Matisse venous thromboembolism trials that compared a ‘heparin’ (either enoxaparin or UFH) with fondaparinux, it was shown that fondaparinux significantly reduced the risk of precipitating rapid-onset HIT among the small subset of patients who were enrolled but who had platelet-activating antibodies (Table 1) [17].

Table 1.   HIT antibodies and clinical outcomes (> 50% platelet count fall) in relation to anticoagulant received (UFH or enoxaparin vs. fondaparinux) and antibody status (EIA+/SRA+ vs. EIA+/SRA− and vs. EIA−/SRA−)
 Patients with > 50% platelet count fall during study drug administrationP value: effect of treatment group
UFH or enoxaparinFondaparinux
  1. EIA, enzyme-immunoassay; SRA, serotonin-release assay (platelet activation assay); UFH, unfractionated heparin. *Three received unfractionated heparin, one enoxaparin. Eight received unfractionated heparin, seven enoxaparin. Includes one patient without platelet-activating antibodies (EIA+/SRA−) treated with fondaparinux whose platelet count fell by > 50% but recovered 2 days later while still receiving fondaparinux. §Fifteen received unfractionated heparin, 12 enoxaparin. Includes one patient without antibodies of either type treated with fondaparinux whose platelet count fell by 50% upon initiation of an antibiotic for pneumonia and red cell transfusion for anemia. Reprinted, with permission, from [17].

EIA+/SRA+ (n = 14)4/4*0/10P = 0.001
EIA+/SRA− (n = 33)0/151/18P = 1.00
P value: effect of SRA result (+ or −) on thrombocytopenia within a treatment groupP < 0.001P = 1.00
EIA−/SRA− (nested controls, n = 47)0/27§1/20P = 0.42
P value: effect of EIA result (+ or −) within a treatment group; EIA−/SRA−
 vs. EIA+/SRA+P < 0.001P = 1.00
 vs. EIA+/SRA−P = 1.00P = 1.00

This Matisse trial substudy also supports the utility of the washed platelet activation assay in discerning between clinically-relevant and irrelevant antibodies: among patients with strong positive EIAs for PF4-dependent antibodies (> 1.0 units), the risk of precipitating thrombocytopenia among the heparin-treated patients was seen exclusively in those patients with platelet-activating antibodies (P < 0.001) (Table 1). Overall, only 14 (11%) of the 127 EIA+ patients (> 0.40 OD units) entered into the Matisse trials had SRA+ status, again pointing to the substantial risk of overdiagnosis of HIT had systematic anti-PF4/heparin antibodies been sought for clinical decision-making.

Furthermore, observational studies that have examined five or more patients with putative HIT (and supported by a positive test for HIT antibodies [n = 36]) have shown a high success rate with fondaparinux, and with a low bleeding risk [85–87]. These data are supported by our own recent experience in Hamilton [95] treating 12 patients with SRA+ HIT (7 [58%] with HIT-associated thrombosis) with fondaparinux. Indeed, pooling these data totaling 48 patients with putative HIT – with none developing new thrombosis with fondaparinux treatment – suggests an upper 95% confidence limit of 7.4% for new thrombosis; these data suggest that fondaparinux is likely to be at least similarly effective as DTIs to treat HIT (for review of results of DTI therapy for HIT [96]). Given that fondaparinux is proven effective for treatment of venous thromboembolism and acute coronary syndrome, this favorable experience suggests that it is an appropriate agent for managing patients with HIT (including HIT-associated thrombosis), especially when one considers that most patients with suspected HIT do not have this diagnosis. However, given the seemingly insurmountable hurdles to perform the necessary studies, it seems unlikely that fondaparinux will receive regulatory approval for HIT.

Transience of HIT antibodies: safety of repeat heparin exposure

The final paradox to consider is the unusual transience of HIT antibodies. Following an episode of HIT, the antibodies can become non-detectable within days or a few weeks [41], including even when heparin is continued [38]. Further, we and others have observed that re-exposure to heparin is usually uneventful in such patients, allowing for deliberate re-exposure in situations such as cardiac or vascular surgery [41,96]. Indeed, two groups have even shown that long-term readministration of heparin is feasible, such as for the provision of hemodialysis [97,98]. However, even if the risk of HIT upon re-exposure is no greater than in a heparin-naïve patient, recurrence of HIT remains a possibility [99].

Final thoughts

  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References

The central paradigm of HIT infers that wider implementation of washed platelet activation assays will lead to a greater appreciation that ‘true’ HIT represents but a subset of patients with PF4/heparin-reactive antibodies. This will reduce HIT overdiagnosis. The hierarchical nature of HIT serum-induced platelet activation means that the inclusion of weak-positive control sera will help ensure quality control, so that a negative washed platelet activation assay does indeed rule out the diagnosis of HIT irrespective of the EIA results.

Given that ‘true’ HIT represents a small subset among patients investigated for this disorder, simpler therapeutic approaches might also facilitate management strategies. These include: (i) continuation of heparin in low probability situations; (ii) prophylactic (low-dose) alternative anticoagulants (e.g., subcutaneous fondaparinux, danaparoid, or r-hirudin) in low- or intermediate probability situations; and (iii) therapeutic-dose anticoagulation restricted to patients with proven thrombosis, high probability for HIT, or convincing laboratory support for HIT (especially a positive washed platelet activation assay).


  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References

The author thanks Jo-Ann I. Sheppard for preparation of the Figures. Several of the studies described in this report were supported by the Heart and Stroke Foundation of Ontario (operating grants T2967, B3763, T4502, T5207, T6157, and T6950).

Disclosure of conflicts of interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References

The author has received lecture honoraria from GlaxoSmithKline, Pfizer Canada, and Sanofi-Aventis, has provided consulting services to, and/or has received research funding from, Canyon Pharmaceuticals, GTI Diagnostics Inc, GlaxoSmithKline, and Paringenix, and has provided expert witness testimony relating to heparin-induced thrombocytopenia.


  1. Top of page
  2. Abstract
  3. Introduction
  4. HIT central paradigm
  5. Risk factors for HIT and for anti-PF4/heparin immunization
  6. HIT paradoxes
  7. Final thoughts
  8. Acknowledgements
  9. Disclosure of conflicts of interest
  10. References
  • 1
    Rhodes GR, Dixon RH, Silver D. Heparin induced thrombocytopenia with thrombotic and hemorrhagic manifestations. Surg Gynecol Obstet 1973; 136: 40916.
  • 2
    Warkentin TE. History of heparin-induced thrombocytopenia. In: WarkentinTE, GreinacherA, eds. Heparin-Induced Thrombocytopenia, 4th edn. New York: Informa Healthcare USA, 2007: 119.
  • 3
    Warkentin TE, Chong BH, Greinacher A. Heparin-induced thrombocytopenia: towards consensus. Thromb Haemost 1998; 79: 17.
  • 4
    Kelton JG, Sheridan D, Santos A, Smith J, Steeves K, Smith C, Brown C, Murphy WG. Heparin-induced thrombocytopenia: laboratory studies. Blood 1988; 72: 92530.
  • 5
    Chong BH, Fawaz I, Chestermann CN, Berndt MC. Heparin-induced thrombocytopenia: mechanism of interaction of the heparin-dependent antibody with platelets. Br J Haematol 1989; 73: 23540.
  • 6
    Amiral J, Peynaud-Debayle E, Wolf M, Bridey F, Vissac AM, Meyer D. Generation of antibodies to heparin-PF4 complexes without thrombocytopenia in patients treated with unfractionated or low-molecular-weight heparin. Am J Hematol 1996; 52: 905.
  • 7
    Warkentin TE, Sheppard JI, Horsewood P, Simpson PJ, Moore JC, Kelton JG. Impact of the patient population on the risk for heparin-induced thrombocytopenia. Blood 2000; 96: 17038.
  • 8
    Warkentin TE, Sheppard JI, Moore JC, Moore KM, Sigouin CS, Kelton JG. Laboratory testing for the antibodies that cause heparin-induced thrombocytopenia: how much class do we need? J Lab Clin Med 2005; 146: 3416.
  • 9
    Greinacher A, Eichler P, Lietz T, Warkentin TE. Replacement of unfractionated heparin by low-molecular-weight heparin for postorthopedic surgery antithrombotic prophylaxis lowers the overall risk of symptomatic thrombosis because of a lower frequency of heparin-induced thrombocytopenia. Blood 2005; 106: 29212.
  • 10
    Bakchoul T, Giptner A, Najaoui A, Bein G, Santoso S, Sachs UJH. Prospective evaluation of immunoassays for the diagnosis of heparin-induced thrombocytopenia. J Thromb Haemost 2009; 7: 12605.
  • 11
    Lo GK, Sigouin CS, Warkentin TE. What is the potential for overdiagnosis of heparin-induced thrombocytopenia? Am J Hematol 2007; 82: 103743.
  • 12
    Greinacher A, Amiral J, Dummel V, Vissac A, Kiefel V, Mueller-Eckhardt C. Laboratory diagnosis of heparin-associated thrombocytopenia and comparison of platelet aggregation test, heparin-induced platelet activation test, and platelet factor 4/heparin enzyme-linked immunosorbent assay. Transfusion 1994; 34: 3815.
  • 13
    Sheridan D, Carter C, Kelton JG. A diagnostic test for heparin-induced thrombocytopenia. Blood 1986; 67: 2730.
  • 14
    Warkentin TE, Hayward CPM, Smith CA, Kelly PM, Kelton JG. Determinants of donor platelet variability when testing for heparin-induced thrombocytopenia. J Lab Clin Med 1992; 120: 3719.
  • 15
    Warkentin TE, Sheppard JI, Moore JC, Sigouin CS, Kelton JG. Quantitative interpretation of optical density measurements using PF4-dependent enzyme-immunoassays. J Thromb Haemost 2008; 6: 130412.
  • 16
    Warkentin TE, Levine MN, Hirsh J, Horsewood P, Roberts RS, Gent M, Kelton JG. Heparin-induced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N Engl J Med 1995; 332: 13305.
  • 17
    Warkentin TE, Davidson BL, Büller HR, Gallus A, Gent M, Lensing AWA, Piovella F, Prins MH, Segers AEM, Kelton JG. Prevalence and risk of pre-existing heparin-induced thrombocytopenia antibodies in patients with acute venous thromboembolism. Chest 2011. [Epub ahead of print 10 March 2011].
  • 18
    Warkentin TE, Roberts RS, Hirsh J, Kelton JG. An improved definition of immune heparin-induced thrombocytopenia in postoperative orthopedic patients. Arch Intern Med 2003; 163: 251824.
  • 19
    Warkentin TE, Cook RJ. Impact of laboratory testing for heparin-induced antibodies: using Bayes’ rule to prevent overdiagnosis of heparin-induced thrombocytopenia. [Bedeutung von Laboruntersuchungen von Heparin-induzierten Antikörpern: Einsatz des Bayes Wahrscheinlichkeitstheorems zur Prävention der Überdiagnose einer Heparin-induzierten Thrombozytopenie]. J Lab Med 2011; 35: 4554.
  • 20
    Greinacher A, Michels I, Kiefel V, Mueller-Eckhardt C. A rapid and sensitive test for diagnosing heparin-associated thrombocytopenia. Thromb Haemost 1991; 66: 7346.
  • 21
    Amiral J, Bridey F, Dreyfus M, Vissoc AM, Fressinaud E, Wolf M, Meyer D. Platelet factor 4 complexed to heparin is the target for antibodies generated in heparin-induced thrombocytopenia. Thromb Haemost 1992; 68: 956.
  • 22
    Kelton JG, Smith JW, Warkentin TE, Hayward CPM, Denomme GA, Horsewood P. Immunoglobulin G from patients with heparin-induced thrombocytopenia binds to a complex of heparin and platelet factor 4. Blood 1994; 83: 32329.
  • 23
    Greinacher A, Potzsch B, Amiral J, Dummel V, Eichner A, Mueller-Eckhardt C. Heparin-associated thrombocytopenia: isolation of the antibody and characterization of a multimolecular PF4-heparin complex as the major antigen. Thromb Haemost 1994; 71: 24751.
  • 24
    Visentin GP, Ford SE, Scott JP, Aster RH. Antibodies from patients with heparin induced thrombocytopenia/thrombosis are specific for platelet factor 4 complexed with heparin or bound to endothelial cells. J Clin Invest 1994; 93: 818.
  • 25
    Lee DH, Warkentin TE. Frequency of heparin-induced thrombocytopenia. In: WarkentinTE, GreinacherA, eds. Heparin-Induced Thrombocytopenia, 4th edn. New York: Informa Healthcare USA, 2007: 67116.
  • 26
    Amiral J, Wolf M, Fischer AM, Boyer-Neumann C, Vissac AM, Meyer D. Pathogenicity of IgA and/or IgM antibodies to heparin-PF4 complexes in patients with heparin-induced thrombocytopenia. Br J Haematol 1996; 92: 9549.
  • 27
    Suh JS, Malik MI, Aster RH, Visentin GP. Characterization of the humoral immune response in heparin-induced thrombocytopenia. Am J Hematol 1997; 54: 196201.
  • 28
    Amiral J, Pouplard C, Vissac AM, Walenga JM, Jeske W, Gruel Y. Affinity purification of heparin-dependent antibodies to platelet factor 4 developed in heparin-induced thrombocytopenia: biological characteristics and effects on platelet activation. Br J Haematol 2000; 109: 33641.
  • 29
    Pauzner R, Greinacher A, Selleng K, Althaus K, Shenkman B, Seligsohn U. False-positive tests for heparin induced thrombocytopenia in patients with antiphospholipid syndrome and systemic lupus erythematosus. J Thromb Haemost 2009; 7: 10704.
  • 30
    Greinacher A, Ittermann T, Bagemühl J, Althaus K, Fürll B, Selleng S, Lubenow N, Schellong S, Sheppard JI, Warkentin TE. Heparin-induced thrombocytopenia: towards standardization of platelet factor 4/heparin antigen tests. J Thromb Haemost 2010; 8: 202531.
  • 31
    Lindhoff-Last E, Gerdsen F, Ackermann H, Bauersachs R. Determination of heparin-platelet factor 4-IgG antibodies improves diagnosis of heparin-induced thrombocytopenia. Br J Haematol 2001; 113: 88690.
  • 32
    Warkentin TE, Sheppard JI, Moore JC, Kelton JG. The use of well-characterized sera for the assessment of new diagnostic enzyme-immunoassays for the diagnosis of heparin-induced thrombocytopenia. J Thromb Haemost 2010; 8: 2168.
  • 33
    Whitlatch NL, Kong DF, Metjian AD, Arepally GM, Ortel TL. Validation of the high-dose heparin confirmatory step for the diagnosis of heparin-induced thrombocytopenia. Blood 2010; 116: 17616.
  • 34
    Warkentin TE, Sheppard JI. No significant improvement in diagnostic specificity of an anti-PF4/polyanion immunoassay with use of high heparin confirmatory procedure. J Thromb Haemost 2006; 4: 2812.
  • 35
    Bakchoul T, Giptner A, Bein G, Santoso S, Sachs UJH. Performance characteristics of two commercially available IgG-specific immunoassays in the assessment of heparin-induced thrombocytopenia (HIT). Thromb Res 2011; 127: 3458.
  • 36
    Althaus K, Strobel U, Warkentin TE, Greinacher A. Combined use of the high heparin step and optical density to optimize diagnostic sensitivity and specificity of an anti-PF4/heparin enzyme-immunoassay. Thromb Res 2011. [Epub ahead of print 25 May 2011].
  • 37
    Warkentin TE, Sheppard JI, Moore JC, Cook RJ, Kelton JG. Studies of the immune response in heparin-induced thrombocytopenia. Blood 2009; 113: 49639.
  • 38
    Greinacher A, Kohlmann T, Strobel U, Sheppard JI, Warkentin TE. The temporal profile of the anti-PF4/heparin immune response. Blood 2009; 113: 49706.
  • 39
    Warkentin TE, Sheppard JI. Clinical sample investigation (CSI) hematology: pinpointing the precise onset of heparin-induced thrombocytopenia (HIT). J Thromb Haemost 2007; 5: 6367.
  • 40
    Linkins LA, Warkentin TE. The approach to heparin-induced thrombocytopenia. Semin Respir Crit Care Med 2008; 29: 6674.
  • 41
    Warkentin TE, Kelton JG. Temporal aspects of heparin-induced thrombocytopenia. N Engl J Med 2001; 344: 128692.
  • 42
    Smythe MA, Warkentin TE, Woodhouse AL, Zakalik D. Venous limb gangrene and fatal hemorrhage: adverse consequences of HIT “overdiagnosis” in a patient with antiphospholipid syndrome. Am J Hematol 2011; 86: 18891.
  • 43
    Greinacher A, Juhl D, Strobel U, Wessel A, Lubenow N, Selleng K, Eichler P, Warkentin TE. Heparin-induced thrombocytopenia: a prospective study on the incidence, platelet-activating capacity and clinical significance of anti-PF4/heparin antibodies of the IgG, IgM, and IgA classes. J Thromb Haemost 2007; 5: 166673.
  • 44
    Hui P, Cook DJ, Lim W, Fraser GA, Arnold DM. The frequency and clinical significance of thrombocytopenia complicating critical illness: a systematic review. Chest 2011; 139: 2718.
  • 45
    Selleng S, Malowsky B, Strobel U, Wessel A, Ittermann T, Wollert HG, Warkentin TE, Greinacher A. Early-onset and persisting thrombocytopenia in post-cardiac surgery patients is rarely due to heparin-induced thrombocytopenia even when antibody tests are positive. J Thromb Haemost 2010; 8: 306.
  • 46
    Selleng S, Selleng K, Wollert HG, Muellejans B, Lietz T, Warkentin TE, Greinacher A. Heparin-induced thrombocytopenia in patients requiring prolonged intensive care unit treatment after cardiopulmonary bypass. J Thromb Haemost 2008; 6: 42835.
  • 47
    Warkentin TE, Sheppard JI, Sigouin CS, Kohlmann T, Eichler P, Greinacher A. Gender imbalance and risk factor interactions in heparin-induced thrombocytopenia. Blood 2005; 108: 293741.
  • 48
    Warkentin TE, Cook RJ, Marder VJ, Sheppard JI, Moore JC, Eriksson BI, Greinacher A, Kelton JG. Anti-platelet factor 4/heparin antibodies in orthopedic surgery patients receiving antithrombotic prophylaxis with fondaparinux or enoxaparin. Blood 2005; 106: 37916.
  • 49
    Lubenow N, Hinz P, Thomaschewski S, Lietz T, Vogler M, Ladwig A, Jünger M, Nauck M, Schellong S, Wander K, Engel G, Ekkernkamp A, Greinacher A. The severity of trauma determines the immune response to PF4/heparin and the frequency of heparin-induced thrombocytopenia. Blood 2010; 115: 1797803.
  • 50
    Warkentin TE, Cook RJ, Marder VJ, Greinacher A. Anti-PF4/heparin antibody formation post-orthopedic surgery thromboprophylaxis: the role of non-drug risk factors and evidence for a stoichiometry-based model of immunization. J Thromb Haemost 2010; 8: 50412.
  • 51
    Warkentin TE. Clinical picture of heparin-induced thrombocytopenia. In: WarkentinTE, GreinacherA, eds. Heparin-Induced Thrombocytopenia, 4th edn. New York: Informa Healthcare USA, 2007: 2166.
  • 52
    Dawes J, Pumphrey CW, McLaren KM, Prowse CV, Pepper DS. The in vivo release of human platelet factor 4 by heparin. Thromb Res 1982; 27: 6576.
  • 53
    Chesterman CN, McGready JR, Doyle DJ, Morgan FJ. Plasma factors of platelet factor 4 measured by radioimmunoassay. Br J Haematol 1978; 40: 489500.
  • 54
    Greinacher A, Alban S, Omer-Adam MA, Weitschies W, Warkentin TE. Heparin-induced thrombocytopenia: a stoichiometry-based model to explain the differing immunogenicities of unfractionated heparin, low-molecular-weight heparin, and fondaparinux in different clinical settings. Thromb Res 2008; 122: 21120.
  • 55
    Rauova L, Zhai L, Kowalska MA, Arepally GM, Cines DB, Poncz M. Role of platelet surface PF4 antigenic complexes in heparin-induced thrombocytopenia pathogenesis: diagnostic and therapeutic implications. Blood 2006; 107: 234653.
  • 56
    Warkentin TE, Makris M, Jay RM, Kelton JG. A spontaneous prothrombotic disorder resembling heparin-induced thrombocytopenia. Am J Med 2008; 121: 6326.
  • 57
    Jay RM, Warkentin TE. Fatal heparin-induced thrombocytopenia (HIT) during warfarin thromboprophylaxis following orthopedic surgery: another example of ‘spontaneous’ HIT? J Thromb Haemost 2008; 6: 1598600.
  • 58
    Pruthi RK, Daniels PR, Nambudiri GS, Warkentin TE. Heparin-induced thrombocytopenia (HIT) during postoperative warfarin thromboprophylaxis: a second example of postorthopedic surgery ‘spontaneous’ HIT? J Thromb Haemost 2009; 7: 499501.
  • 59
    Mallik A, Carlson KB, DeSancho MT. A patient with ‘spontaneous’ heparin-induced thrombocytopenia and thrombosis after undergoing knee replacement. Blood Coagul Fibrinolysis 2011; 22: 735.
  • 60
    Krauel K, Pötschke C, Weber C, Kessler W, Fürll B, Ittermann T, Maier S, Hammerschmidt S, Bröker BM, Greinacher A. Platelet factor 4 binds to bacteria – inducing antibodies cross-reacting with the major antigen in heparin-induced thrombocytopenia. Blood 2011; 117: 13708.
  • 61
    Greinacher A, Holtfreter B, Krauel K, Gätke D, Weber C, Itterman T, Hammerschmidt S, Kocher T. Association of natural anti-platelet factor 4/heparin antibodies with periodontal disease. Blood. [Epub ahead of print 9 June 2011].
  • 62
    Matsuo T, Tomaru T, Kario K, Hirokawa T, HIT Research Group of Japan. Incidence of heparin-PF4 complex antibody formation and heparin-induced thrombocytopenia in acute coronary syndrome. Thromb Res 2005; 115: 47581.
  • 63
    Warkentin TE, Kelton JG. A 14-year study of heparin-induced thrombocytopenia. Am J Med 1996; 101: 5027.
  • 64
    Warkentin TE. Management of heparin-induced thrombocytopenia: a critical comparison of lepirudin and argatroban. Thromb Res 2003; 110: 7382.
  • 65
    Warkentin TE, Hayward CPM, Boshkov LK, Santos AV, Sheppard JI, Bode AP, Kelton JG. Sera from patients with heparin-induced thrombocytopenia generate platelet-derived microparticles with procoagulant activity: an explanation for the thrombotic complications of heparin-induced thrombocytopenia. Blood 1994; 84: 36919.
  • 66
    Warkentin TE, Sheppard JI. Generation of platelet-derived microparticles and procoagulant activity by heparin-induced thrombocytopenia IgG/serum and other IgG platelet agonists: a comparison with standard platelet agonists. Platelets 1999; 10: 31926.
  • 67
    Tardy-Poncet B, Piot M, Chapelle C, France G, Campos L, Garraud O, Decousus H, Mismetti P, Tardy B. Thrombin generation and heparin-induced thrombocytopenia. J Thromb Haemost 2009; 7: 147481.
  • 68
    Rauova L, Hirsch JD, Greene TK, Zhai L, Hayes VM, Kowalska MA, Cines DB, Poncz M. Monocyte-bound PF4 in the pathogenesis of heparin-induced thrombocytopenia. Blood 2010; 116: 502131.
  • 69
    Cines DB, Tomaski A, Tannenbaum S. Immune endothelial-cell injury in heparin-associated thrombocytopenia. N Engl J Med 1987; 316: 5819.
  • 70
    Warkentin TE, Elavathil LJ, Hayward CPM, Johnston MA, Russett JI, Kelton JG. The pathogenesis of venous limb gangrene associated with heparin-induced thrombocytopenia. Ann Intern Med 1997; 127: 80412.
  • 71
    Srinivasan AF, Rice L, Bartholomew JR, Rangaswamy C, La Perna L, Thompson JE, Murphy S, Baker KR. Warfarin-induced skin necrosis and venous limb gangrene in the setting of heparin-induced thrombocytopenia. Arch Intern Med 2004; 164: 6670.
  • 72
    Warkentin TE. Heparin-induced thrombocytopenia: IgG-mediated platelet activation, platelet microparticle generation, and altered procoagulant/anticoagulant balance in the pathogenesis of thrombosis and venous limb gangrene complicating heparin-induced thrombocytopenia. Transfus Med Rev 1996; 10: 24958.
  • 73
    Kowalska MA, Rauova L, Hayes V, Cines DB, Bougie DW, Aster RH, Krishnaswamy S, Poncz M. Heparin-induced thrombocytopenia antibodies inhibit PF4-dependent enhancement of activated protein C formation by binding to antigenic complexes formed with the chondroitin sulfate side-chain of thrombomodulin. Blood 2010; 116 (Suppl.): 318 (Abstract).
  • 74
    Warkentin TE, Kelton JG. Delayed-onset heparin-induced thrombocytopenia and thrombosis. Ann Intern Med 2001; 135: 5026.
  • 75
    Rice L, Rice L, Attisha WK, Drexler A, Francis JL. Delayed-onset heparin-induced thrombocytopenia. Ann Intern Med 2002; 136: 2105.
  • 76
    Prechel MM, McDonald MK, Jeske WP, Messmore HL, Walenga JM. Activation of platelets by heparin-induced thrombocytopenia antibodies in the serotonin release assay is not dependent on the presence of heparin. J Thromb Haemost 2005; 3: 216875.
  • 77
    Socher I, Kroll H, Jorks S, Santoso S, Sachs UJ. Heparin-independent activation of platelets by heparin-induced thrombocytopenia antibodies: a common occurrence. J Thromb Haemost 2008; 6: 197200.
  • 78
    Linkins LA, Warkentin TE. Heparin-induced thrombocytopenia: real world issues. Semin Thromb Hemost 2011; in press.
  • 79
    Greinacher A, Warkentin TE. The direct thrombin inhibitor hirudin. Thromb Haemost 2008; 99: 81929.
  • 80
    Warkentin TE. Agents for the treatment of heparin-induced thrombocytopenia. Hematol Oncol Clin North Am 2010; 24: 75575.
  • 81
    Salmela B, Joutsi-Korhonen L, Saarela E, Lassila R. Comparison of monitoring methods for lepirudin: impact of warfarin and lupus anticoagulant. Thromb Res 2010; 125: 53844.
  • 82
    Love JE, Ferrell C, Chandler WL. Monitoring direct thrombin inhibitors with a plasma diluted thrombin time. Thromb Haemost 2007; 98: 23442.
  • 83
    Chong BH, Gallus AS, Cade JF, Magnani H, Manoharan A, Oldmeadow M, Arthur C, Rickard K, Gallo J, Lloyd J, Seshadri P, Chesterman CN. Prospective randomised open-label comparison of danaparoid with dextran 70 in the treatment of heparin-induced thrombocytopenia with thrombosis. A clinical outcome study. Thromb Haemost 2001; 86: 11705.
  • 84
    Lubenow N, Warkentin TE, Greinacher A, Wessel A, Sloane DA, Krahn EL, Magnani HN. Results of a systematic evaluation of treatment outcomes for heparin-induced thrombocytopenia in patients receiving danaparoid, ancrod, and/or coumarin explain the rapid shift in clinical practice during the 1990s. Thromb Res 2006; 117: 50715.
  • 85
    Kuo KHM, Kovacs MJ. Successful treatment of heparin induced thrombocytopenia (HIT) with fondaparinux. Thromb Haemost 2005; 93: 9991000.
  • 86
    Lobo B, Finch C, Howard A, Minhas S. Fondaparinux for the treatment of patients with acute heparin-induced thrombocytopenia. Thromb Haemost 2008; 99: 20814.
  • 87
    Grouzi E, Kyriakou E, Panagou I, Spiliotopoulou I. Fondaparinux for the treatment of acute heparin-induced thrombocytopenia: a single-center experience. Clin Appl Thromb Hemost 2010; 16: 6637.
  • 88
    Warkentin TE. Fondaparinux: does it cause HIT? can it treat HIT? Expert Rev Hematol 2010; 3: 56781.
  • 89
    Motokawa S, Torigoshi T, Maeda Y, Maeda K, Jiuchi Y, Yamaguchi T, Someya S, Shindo H, Migita K. IgG-class anti-PF4/heparin antibodies and symptomatic DVT in orthopedic surgery patients receiving different anti-thromboembolic prophylaxis therapeutics. BMC Musculoskelet Disord 2011; 12: 22 [Epub ahead of print].
  • 90
    Warkentin TE, Maurer BT, Aster RH. Heparin-induced thrombocytopenia associated with fondaparinux. N Engl J Med 2007; 356: 26534.
  • 91
    Rota E, Bazzan M, Fantino G. Fondaparinux-related thrombocytopenia in a previous low-molecular-weight heparin (LMWH)-induced heparin-induced thrombocytopenia (HIT). Thromb Haemost 2008; 99: 77981.
  • 92
    Salem M, Elrefai S, Shrit MA, Warkentin TE. Fondaparinux thromboprophylaxis-associated heparin-induced thrombocytopenia syndrome complicated by arterial thrombotic stroke. Thromb Haemost 2010; 104: 10712.
  • 93
    Savi P, Chong BH, Greinacher A, Gruel Y, Kelton JG, Warkentin TE, Eichler P, Meuleman D, Petitou M, Herault JP, Cariou R, Herbert JM. Effect of fondaparinux on platelet activation in the presence of heparin-dependent antibodies: a blinded comparative multicenter study with unfractionated heparin. Blood 2005; 105: 13944.
  • 94
    Warkentin TE, Lim W. Can heparin-induced thrombocytopenia be associated with fondaparinux use? Reply to a rebuttal. J Thromb Haemost 2008; 6: 12436.
  • 95
    Warkentin TE, Pai M, Schulman S, Spyropoulos A, Eikelboom JW. Fondaparinux treatment of acute heparin-induced thrombocytopenia (HIT) confirmed by a positive serotonin-release assay (SRA+). J Thromb Haemost 2011; in press (Abstract).
  • 96
    Warkentin TE, Greinacher A, Koster A, Lincoff AM. Treatment and prevention of heparin-induced thrombocytopenia. American College of Chest Physicians evidence-based clinical practice guidelines (8th edition). Chest 2008; 6 (Suppl.): 340S80S.
  • 97
    Hartman V, Malbrain M, Daelemans R, Meersman P, Zachée P. Pseudo-pulmonary embolism as a sign of acute heparin-induced thrombocytopenia in hemodialysis patients: safety of resuming heparin after disappearance of HIT antibodies. Nephron Clin Pract 2006; 104: c1438.
  • 98
    Wanaka K, Matsuo T, Matsuo M, Kaneko C, Miyashita K, Asada R, Matsushima H, Nakajima Y. Re-exposure to heparin in uremic patients requiring hemodialysis with heparin-induced thrombocytopenia. J Thromb Haemost 2010; 8: 6168.
  • 99
    Gruel Y, Lang M, Darnige L, Pacouret G, Dreyfus X, Leroy J, Charbonnier B. Fatal effect of reexposure to heparin after previous heparin-associated thrombocytopenia and thrombosis. Lancet 1990; 336: 10778.
  • 100
    Pötzsch B, Klövekorn WP, Madlener K. Use of heparin during cardiopulmonary bypass in patients with a history of heparin-induced thrombocytopenia. N Engl J Med 2000; 343: 515.