Heparin-induced thrombocytopenia (HIT) is a relatively common prothrombotic adverse drug reaction of unusual pathogenesis that features platelet-activating immunoglobulin G antibodies. The HIT immune response is remarkably transient, with heparin-dependent antibodies no longer detectable 40 to 100 days (median) after an episode of HIT, depending on the assay performed. Moreover, the minimum interval from an immunizing heparin exposure to the development of HIT is 5 days irrespective of the patient’s previous heparin exposure status or history of HIT. This means that short-term heparin reexposure can be safely performed if platelet-activating antibodies are no longer detectable at reexposure baseline and is recommended when heparin is the clear anticoagulant of choice, such as for cardiac or vascular surgery. The risk of recurrent HIT 1 to 2 weeks after heparin reexposure is ∼2% to 5% and is attributable to formation of delayed-onset (or autoimmune-like) HIT antibodies that activate platelets even in the absence of pharmacologic heparin. Some studies suggest that longer-term heparin reexposure (eg, for chronic hemodialysis) may also be reasonable. However, for other antithrombotic indications that involve patients with a history of HIT (eg, treatment of venous thromboembolism or acute coronary syndrome), preference should be given to non-heparin agents such as fondaparinux, danaparoid, argatroban, bivalirudin, or one of the new direct-acting oral anticoagulants as appropriate.

Heparin-induced thrombocytopenia, or HIT, represents an antibody-mediated adverse drug reaction characterized by platelet-activating immunoglobulin G (IgG) that recognize platelet factor 4 (PF4)/heparin complexes.1,2  Because HIT is a clinical-pathological disorder, diagnosis requires integrating clinical and laboratory features.3,4  HIT is a profound hypercoagulability state5  strongly associated with thrombosis.6,7  Moreover, HIT is relatively common: although it occurs in only ∼0.2% of hospitalized patients undergoing any heparin exposure,8  it is more common in certain high-risk patient populations. For example, the frequency of HIT is ∼5% in postorthopedic surgery patients receiving unfractionated heparin (UFH) for 10 to 14 days.9,10  In cardiac surgery patients who are given intraoperative UFH (for cardiopulmonary bypass [CPB]) and who receive postoperative UFH thromboprophylaxis, the frequency is ∼1% to 2%.11-15  The ubiquity of heparin and the relatively common occurrence of HIT explain the increasing numbers of patients with a history of HIT.

Several issues are relevant when considering anticoagulant management of a patient with a history of HIT. First, there are 3 conditions—cardiac surgery, vascular surgery, hemodialysis—for which UFH is the clear anticoagulant of choice. Second, there are different laboratory tests to detect anti-PF4/heparin antibodies, for which implications regarding antibody pathogenicity vary, particularly for immunoassays versus platelet activation assays.16  Third, anti-PF4/heparin antibodies are among the most transient in clinical medicine17 : a patient with HIT will generally have antibodies detectable by both PF4-dependent immunoassay and platelet activation assay at the time of acute HIT, but these antibodies can be difficult to detect as little as a few weeks or months later. Fourth, a minimum of 5 days is required to generate, or to regenerate, pathogenic platelet-activating antibodies.17,18  Thus, for certain patient populations, deliberate and planned reexposure to heparin has a rational basis. Indeed, this strategy was pioneered by Pötzsch et al,19  who described unremarkable outcomes of heparin reexposure for CPB in 10 patients with previous HIT who were antibody negative at reexposure (and who remained antibody negative 10 days after reexposure).

For many clinical situations, alternative non-heparin anticoagulation is well established (eg, treatment and/or prevention of venous thromboembolism or management of acute coronary syndrome). Thus, one can usually identify 1 or more appropriate options among the large panoply of agents, such as direct oral anticoagulants, either factor Xa (rivaroxaban, apixaban, edoxaban) or thrombin (dabigatran) inhibiting, or a parenteral agent (anti-Xa [fondaparinux, danaparoid] or anti-IIa [bivalirudin, argatroban]). For these clinical settings, we do not discuss UFH and low molecular weight heparin (LMWH) as management options for patients with a previous history of HIT.

The HIT immune response does not conform to the classic, adaptive immune response taught in the first-year medical school curriculum. There, the primary immune response, triggered by a first exposure to a foreign substance (antigen), triggers an early (ie, beginning about 5 to 7 days after exposure) IgM-predominant and transient response.20  The secondary immune response, after reexposure to the same antigen, begins earlier, is IgG predominant, and typically results in long-lasting antibody detectability.

In the case of HIT, any immunizing exposure to heparin, whether triggered by the first exposure to heparin or any subsequent reexposure, leads to a relatively rapid immune response, with detectability of anti-PF4/heparin antibodies occurring a median of 4 days after exposure21,22  and onset of HIT occurring at least 1 day later (ie, occurrence within the characteristic day 5 to 10 window of HIT, in which day 0 indicates the presumed immunizing heparin exposure).17,21  There does not seem to be any tendency to an earlier, more rapid immune response, including in patients who have had previous heparin exposures17  or even a previous history of HIT.18,23-25  (Although there is a syndrome named rapid-onset HIT, it does not reflect a new immune response but is simply the triggering of an acute platelet count decrease in a patient who already has circulating HIT antibodies associated with a recent exposure to heparin.17,26 )

A 51-year-old male, who was the index case in the original report of congenital erythroblastic multinuclearity associated with a positive acidified serum test27  (and was listed as patient 8 in our report18  describing outcomes of heparin reexposure in patients with a history of HIT), developed severe HIT (platelet count nadir, 8 × 109/L) and was successfully treated with therapeutic-dose danaparoid (Figure 1A, upper panel). Tests for HIT antibodies were strongly positive (Figure 1A, lower panel). Three years later, he developed acute pulmonary edema secondary to flail mitral valve, requiring urgent cardiac surgery. Hematology consultation was requested because of the history of HIT. There was no time to perform repeat HIT antibody testing before cardiac surgery. What treatment was recommended?

Figure 1

UFH reexposure for urgent cardiac surgery in a patient with a history of HIT. Patient 8 in Table 1. (A) Summary of the clinical (upper panel) and serological (lower panel) features of the patient’s previous episode of HIT (1999). The patient had unusually severe HIT but ultimately did well with therapeutic-dose danaparoid sodium (no new or progressive thrombosis and no ischemic limb loss). The patient tested strongly positive in the SRA, EIA-IgG, and EIA-IgA, and weakly positive in the EIA-IgM tests. (B) Summary of the clinical (upper panel) and serological (lower panel) features of the patient’s UFH reexposure to permit urgent cardiac surgery (2002). After reexposure to UFH, the patient had early-onset and persistent thrombocytopenia (platelet count nadir, 29 × 109/L on POD 6), but this was not related to HIT (the SRA tested negative on POD6). The prolonged period of postoperative thrombocytopenia was probably also unrelated to subsequent formation of platelet-activating antibodies, given that the platelet count recovery occurred as the patient was developing a strongly positive SRA, and given that the buffer control (ie, at 0 U/mL UFH) reactivity in the SRA remained negative throughout the postoperative period and also remained negative in the presence of danaparoid (data indicating absence of danaparoid cross-reactivity not shown). IV, intravenous.

Figure 1

UFH reexposure for urgent cardiac surgery in a patient with a history of HIT. Patient 8 in Table 1. (A) Summary of the clinical (upper panel) and serological (lower panel) features of the patient’s previous episode of HIT (1999). The patient had unusually severe HIT but ultimately did well with therapeutic-dose danaparoid sodium (no new or progressive thrombosis and no ischemic limb loss). The patient tested strongly positive in the SRA, EIA-IgG, and EIA-IgA, and weakly positive in the EIA-IgM tests. (B) Summary of the clinical (upper panel) and serological (lower panel) features of the patient’s UFH reexposure to permit urgent cardiac surgery (2002). After reexposure to UFH, the patient had early-onset and persistent thrombocytopenia (platelet count nadir, 29 × 109/L on POD 6), but this was not related to HIT (the SRA tested negative on POD6). The prolonged period of postoperative thrombocytopenia was probably also unrelated to subsequent formation of platelet-activating antibodies, given that the platelet count recovery occurred as the patient was developing a strongly positive SRA, and given that the buffer control (ie, at 0 U/mL UFH) reactivity in the SRA remained negative throughout the postoperative period and also remained negative in the presence of danaparoid (data indicating absence of danaparoid cross-reactivity not shown). IV, intravenous.

Transience of HIT antibodies (seroreversion) is one of the remarkable features of HIT. The time to nondetectability of HIT antibodies after an episode of HIT ranges from 40 to 100 days, depending on the assay performed (serotonin release assay [SRA] and enzyme-linked immunosorbent assay [EIA]-IgG, respectively).17  There is considerable biological variability around these point estimates: some patients develop negative or weakly positive results within 1 or 2 weeks after HIT. Indeed, platelet counts can recover despite continued heparin administration in patients with acute HIT,22,28  with waning of HIT antibody levels despite ongoing administration of heparin.22 

On the basis of the known natural history of HIT antibody seroreversion and the need for urgent cardiac surgery 3 years after HIT, I (T.E.W.) viewed as negligible the possibility of clinically significant HIT antibodies still being present. I therefore recommended the usual intraoperative anticoagulation with UFH, with postoperative thromboprophylaxis with danaparoid sodium,29  a mixture of anticoagulant glycosaminoglycans (predominantly heparan sulfate) approved in many jurisdictions (although not in the United States) for treatment of HIT. Figure 1B (upper panel) shows the postoperative clinical course, which was characterized by early postoperative pulmonary edema and/or pneumonia and (non-HIT-related) early-onset and persisting thrombocytopenia, with platelet count recovery 2 weeks later. Figure 1A-B (lower panels) compare the HIT antibody test results at the time of previous HIT (strongly positive SRA, EIA-IgG, and EIA-IgA with a weak positive EIA-IgM), with the results following reexposure. Despite repeat seroconversion (by EIA-IgG, EIA-IgA, and SRA), there was no new platelet count decrease that occurred in association with this seroconversion, and serum-induced platelet activation at 0 U/mL UFH was negligible. The patient ultimately did well with no postoperative thrombosis.

A 56-year-old female had a previous history of HIT (without associated thrombosis) that complicated the use of therapeutic-dose dalteparin for new-onset atrial fibrillation; HIT was diagnosed on the basis of a 57.4% platelet count decrease to 77 × 109/L (nadir) and no other plausible explanation (4Ts score, 6 points). The patient tested strongly positive by polyspecific EIA (>4.0 optical density [OD] units) and had a positive SRA. The patient had mitral valve regurgitation, and valve repair was recommended. Subsequent repeat testing for HIT antibodies using another blood sample obtained 24 weeks after HIT diagnosis showed that the SRA was now negative, but the EIA-IgG remained positive (0.85 OD units). What treatment was recommended?

EIA-positive/SRA-negative patients who receive heparin do not develop acute HIT.9,14,16,30  Accordingly, we18  and others31  have used heparin with good results in patients with a previous history of HIT who tested negative in a sensitive (washed platelet) activation test, such as the SRA or heparin-induced platelet activation (HIPA) test but who continued to test EIA positive.

Table 1 presents our updated experience with deliberate planned UFH reexposures for cardiac or vascular surgery in patients with a previous history of HIT and summarizes our previous experience,18  with the addition of 2 more recent cases (including case 2 presented above).32  Of the 20 patient reexposures summarized, 10 featured a positive EIA-IgG at reexposure baseline. None of these 10 EIA-positive patients developed apparent HIT, either at the time of UFH reexposure or during the later postoperative period.

Table 1

Serological and clinical outcomes of 20 heparin reexposures performed for cardiac or vascular surgery

PatientAge (y)SexInterval (wk)UFH rechallengePostoperative ACSRAEIASeroconversion byRecurrent HIT
IgGIgAIgMSRAEIA
EIA-IgG negative at heparin reexposure baseline (n = 10)      Pre Post Pre Post Pre Post Pre Post    
 3 60 132 CPB DS − +++ − +++ − ++ − Yes Yes No 
 4 48 Vasc DS − − − − − − − − No No No 
 5 60 CPB Nil − − − − − − − − No No No 
 6 65 307 Vasc DS − − − − − − − − No No No 
 8 51 180 CPB DS − +++ − +++ − ++ − Yes Yes No 
 10 73 37 CPB DS − ++ − +++ − ++ − − Yes Yes No 
 12 74 62 Vasc DS − − − − − − − − No No No 
 17 68 597 CPB Fx − +++ − +++ − − − +* Yes Yes Yes 
 18a 48 414 Vasc Fx − − − − − − − No Yes No 
 18b —  +132 Vasc Fx − − − − − − − No Yes No 
EIA-IgG positive at heparin reexposure baseline (n = 10)                 
 9 77 21 Vasc Clopid/asp − − − − − − No No No 
 11 72 47 Vasc Clopid/asp − +++ +* +++* − − − − Yes Yes No 
 13 48 22 CPB DS − − ++ ++ − +* − − No Yes No 
 14 66 22 Vasc Nil − +++ +++ − +++ − − Yes Yes No 
 15 57 515 CPB DS − +++ ++ +++ − +* ++* Yes Yes No 
 16 61 422 Vasc DS − +++ +++ − − − Yes Yes No 
 19 62 166 CPB Warf − − ++ ++ − − − − No No No 
 20 54 20 CPB Fx − − ++ +++ − − − No Yes No 
 21 76 12 CPB Fx PLEX preoperative − − ++ ++ − − − − No No No 
 22 56 28 CPB DS − +++ +++ − − − Yes Yes No 
PatientAge (y)SexInterval (wk)UFH rechallengePostoperative ACSRAEIASeroconversion byRecurrent HIT
IgGIgAIgMSRAEIA
EIA-IgG negative at heparin reexposure baseline (n = 10)      Pre Post Pre Post Pre Post Pre Post    
 3 60 132 CPB DS − +++ − +++ − ++ − Yes Yes No 
 4 48 Vasc DS − − − − − − − − No No No 
 5 60 CPB Nil − − − − − − − − No No No 
 6 65 307 Vasc DS − − − − − − − − No No No 
 8 51 180 CPB DS − +++ − +++ − ++ − Yes Yes No 
 10 73 37 CPB DS − ++ − +++ − ++ − − Yes Yes No 
 12 74 62 Vasc DS − − − − − − − − No No No 
 17 68 597 CPB Fx − +++ − +++ − − − +* Yes Yes Yes 
 18a 48 414 Vasc Fx − − − − − − − No Yes No 
 18b —  +132 Vasc Fx − − − − − − − No Yes No 
EIA-IgG positive at heparin reexposure baseline (n = 10)                 
 9 77 21 Vasc Clopid/asp − − − − − − No No No 
 11 72 47 Vasc Clopid/asp − +++ +* +++* − − − − Yes Yes No 
 13 48 22 CPB DS − − ++ ++ − +* − − No Yes No 
 14 66 22 Vasc Nil − +++ +++ − +++ − − Yes Yes No 
 15 57 515 CPB DS − +++ ++ +++ − +* ++* Yes Yes No 
 16 61 422 Vasc DS − +++ +++ − − − Yes Yes No 
 19 62 166 CPB Warf − − ++ ++ − − − − No No No 
 20 54 20 CPB Fx − − ++ +++ − − − No Yes No 
 21 76 12 CPB Fx PLEX preoperative − − ++ ++ − − − − No No No 
 22 56 28 CPB DS − +++ +++ − − − Yes Yes No 

Patient age is at the previous episode of HIT. Patient numbers are from Warkentin and Sheppard18  with patient 21 reported subsequently32  and patient 22 representing the most recent heparin exposure (December 2015) discussed in detail in this article (case 2). Note that for patient 11, the positive EIA-IgG at reexposure baseline was not inhibited >30% by high heparin. Note also that the data shown for patient 10 differs from that presented previously,18  because repeat studies showed that the baseline EIA reactivity before reexposure was consistently negative upon retesting. Strength of assay results: EIA: −, negative; +, weakly positive (0.40 [or 0.45] to 0.99 OD units; ++, moderately positive (1.00 to 1.99 OD units); +++, strongly positive (≥2.00 OD units). SRA: −, negative (<20% serotonin release); +, weakly positive (20.0% to 49.9% release); ++, moderately positive (50.0% to 79.9% release); +++, strongly positive (≥80% release).

AC, anticoagulation; Clopid/asp, clopidogrel/aspirin; DS, danaparoid sodium; F, female; Fx, fondaparinux; M, male; PLEX preoperative, plasma exchange administered preoperatively (pre-heparin reexposure); Vasc, vascular surgery; warf, warfarin.

*

Positive anti-PF4/heparin EIA was not inhibited by >30% in the presence of high heparin.

Any/all EIA seroconversions for that patient were not inhibited >30% by high heparin.

Table 1 indicates that a positive EIA at reexposure baseline is not associated with an appreciably increased risk of subsequent seroconversion to a positive SRA or more strongly positive EIA (defined as >100% increase in OD). Of the 10 patients with a baseline negative EIA, 4 patients (40%) developed subsequent SRA seroconversion, and 6 patients (60%) developed EIA seroconversion. In contrast, SRA and EIA seroconversion occurred in 5 (50%) of 10 and 7 (70%) of 10, respectively, of the patients who tested EIA positive at reexposure baseline.

As will be discussed later (see “What is the risk of recurrent HIT with heparin reexposure for cardiac or vascular surgery?”), in our 20-patient case series, we observed only 1 case of recurrent HIT, but ironically this occurred in 1 of the 10 patients who tested negative by EIA for anti-PF4/heparin antibodies at preoperative reexposure baseline.

I (J.A.M.A.) recommended the usual intraoperative anticoagulation with UFH and postoperative thromboprophylaxis with danaparoid for the patient (patient 22 in Table 1). The mitral valve repair proceeded uneventfully. Figure 2A shows the postoperative platelet count changes. Although there were no platelet count changes suggestive of HIT (eg, a >30% platelet count decrease after postoperative day 4 [POD4]), the patient did develop atrial fibrillation on POD14. The patient was switched from subcutaneous danaparoid to oral apixaban. No thrombosis was evident.

Figure 2

UFH reexposure for cardiac surgery in a patient with history of HIT who is now SRA negative but EIA-IgG positive. Patient 22 in Table 1. (A) Clinical summary. The patient had HIT in May 2015. UFH reexposure to permit mitral valve repair was performed in November 2015. The patient received postoperative anticoagulation with danaparoid sodium. The postoperative course was unremarkable except for atrial fibrillation (POD14), which was treated with subcutaneous danaparoid followed by oral apixaban (continued at discharge). (B) Serological summary (including seroconversion after reexposure). The patient tested strongly positive for HIT antibodies at HIT diagnosis, including EIA-IgG/IgA/IgM >4.00 OD units (PF4 Enhanced, Immucor GTI Diagnostics, Waukesha, WI), EIA-IgG (McMaster Platelet Immunology Laboratory) 2.67 OD units (90% inhibition by high heparin), and a positive SRA (peak serotonin release, 92% at 0.1 U/mL UFH, with 34% serotonin release at buffer control). At 24-week follow-up (day −29), the SRA tested negative (<10% serotonin release), but the EIA-IgG remained positive at 0.85 OD units (68% inhibition by high heparin). After reexposure to UFH, the patient developed repeat strongly positive seroconversion by SRA and EIA-IgG and weakly positive seroconversion by EIA-IgA. The buffer control reactivity in the SRA remained negative throughout the postoperative period. Later testing for danaparoid cross-reactivity (POD12 blood sample) showed a peak of 79% serotonin release in the presence of danaparoid (panel C). (C) Positive testing for danaparoid cross-reactivity. POD12 serum was later tested for danaparoid cross-reactivity; patient serum-induced serotonin-release at buffer control (ie, at 0 U/mL UFH) was weakly positive (21% release), which increased to a peak of 79% serotonin release at 0.3 anti-factor Xa U/mL (final) danaparoid, thus confirming in vitro cross-reactivity.

Figure 2

UFH reexposure for cardiac surgery in a patient with history of HIT who is now SRA negative but EIA-IgG positive. Patient 22 in Table 1. (A) Clinical summary. The patient had HIT in May 2015. UFH reexposure to permit mitral valve repair was performed in November 2015. The patient received postoperative anticoagulation with danaparoid sodium. The postoperative course was unremarkable except for atrial fibrillation (POD14), which was treated with subcutaneous danaparoid followed by oral apixaban (continued at discharge). (B) Serological summary (including seroconversion after reexposure). The patient tested strongly positive for HIT antibodies at HIT diagnosis, including EIA-IgG/IgA/IgM >4.00 OD units (PF4 Enhanced, Immucor GTI Diagnostics, Waukesha, WI), EIA-IgG (McMaster Platelet Immunology Laboratory) 2.67 OD units (90% inhibition by high heparin), and a positive SRA (peak serotonin release, 92% at 0.1 U/mL UFH, with 34% serotonin release at buffer control). At 24-week follow-up (day −29), the SRA tested negative (<10% serotonin release), but the EIA-IgG remained positive at 0.85 OD units (68% inhibition by high heparin). After reexposure to UFH, the patient developed repeat strongly positive seroconversion by SRA and EIA-IgG and weakly positive seroconversion by EIA-IgA. The buffer control reactivity in the SRA remained negative throughout the postoperative period. Later testing for danaparoid cross-reactivity (POD12 blood sample) showed a peak of 79% serotonin release in the presence of danaparoid (panel C). (C) Positive testing for danaparoid cross-reactivity. POD12 serum was later tested for danaparoid cross-reactivity; patient serum-induced serotonin-release at buffer control (ie, at 0 U/mL UFH) was weakly positive (21% release), which increased to a peak of 79% serotonin release at 0.3 anti-factor Xa U/mL (final) danaparoid, thus confirming in vitro cross-reactivity.

Figure 2B depicts the results of testing serial blood samples for HIT antibodies, which showed EIA-IgG and SRA seroconversion by POD7, followed by EIA-IgA seroconversion on POD10. Peak SRA reactivity was observed on POD13, with 89% serotonin release seen at 0.3 U/mL. Despite the strong SRA seroconversion, the patient’s clinical course was judged successful, without recurrence of HIT. Of note, cross-reactivity for danaparoid was shown using a POD12 sample (Figure 2C), but the clinical significance of this result is uncertain.

A 76-year-old woman, listed as patient 21 in Table 1, developed HIT while receiving UFH for medical thromboprophylaxis and required cardiac surgery 12 weeks later to remove renal carcinoma invading the inferior vena cava and right atrium. Her EIA-IgG remained strongly positive (2.58 OD units) as did her SRA (99% serotonin release at 0.3 IU/mL UFH). What management options could be considered?

Three different approaches have been described to manage a patient known to have a positive washed platelet activation assay (either the SRA or HIPA test) at the time of cardiac or vascular surgery and in whom further watchful waiting in anticipation of a negative assay is judged infeasible because of urgent/emergent need for surgery: (1) intraoperative anticoagulation with a non-heparin anticoagulant, especially bivalirudin33,34 ; (2) intraoperative anticoagulation with UFH combined with a platelet inhibitor such as iloprost (prostacyclin analog)35 ; and (3) therapeutic plasma exchange (TPE) to convert the patient from seropositive to seronegative32  or to reduce EIA titers.36  Although all 3 approaches have their merits, my (T.E.W.) approach has been to permit the surgeons to use UFH (their usual practice) to avoid the inherent complications of learning to use a nonstandard anticoagulant for a complex procedure such as CPB, usually in an urgent/emergent setting. Moreover, using bivalirudin for CPB has risks (eg, proteolysis of bivalirudin within stagnant blood, which results in clotting within the surgical field, clamped native vessels and/or grafts, or within the CPB tubing and/or reservoir37,38 ) as does iloprost (eg, severe hypotension38 ).

Thus, for case 3, where clinical circumstances did not allow waiting for the SRA to become negative, I (T.E.W.) used TPE to convert the patient from SRA-positive to SRA-negative status.32  Interestingly, although 4 plasma exchanges achieved SRA-negative status, the patient continued to have a moderately positive EIA-IgG (Figure 3). The patient underwent cardiac surgery without developing clinical evidence of acute HIT; in addition, the patient did not show evidence of postoperative seroconversion.

Figure 3

Serial SRA and IgG-specific anti-PF4/heparin EIA test results in relation to 4 therapeutic plasma exchange sessions performed on 4 consecutive days (last therapeutic plasma exchange performed 2 days before cardiac surgery using heparin). This is patient 21 in Table 1. Reprinted from Warkentin et al32  with permission.

Figure 3

Serial SRA and IgG-specific anti-PF4/heparin EIA test results in relation to 4 therapeutic plasma exchange sessions performed on 4 consecutive days (last therapeutic plasma exchange performed 2 days before cardiac surgery using heparin). This is patient 21 in Table 1. Reprinted from Warkentin et al32  with permission.

To confirm that TPE-induced seroreversion to a negative SRA would always precede EIA seroreversion, we performed serial dilutions of 15 different strongly positive HIT sera and showed that seroreversion to a negative SRA always occurred before an EIA became negative.32  Thus, performing serial TPE to the end point of a negative EIA would mean that several additional plasma exchanges would be performed versus the minimum required to achieve a negative SRA.

Welsby and colleagues36  used intraoperative TPE to decrease anti-PF4/heparin antibody levels in patients with recent HIT who were given UFH for heart transplantation despite being EIA positive (but platelet aggregation test negative) at surgery. None of the patients developed recurrent HIT using this management strategy.

Table 1 shows that recurrent HIT developed in only 1 patient (patient 17), for a frequency of 5% (1 of 20) in our updated case series. Postoperative heparin was not given to any of the 20 patients; rather, anticoagulation was administered with danaparoid (n = 10), fondaparinux (n = 5), warfarin (n = 1), clopidogrel/aspirin (n = 2), or no postoperative anticoagulation (n = 2). The lack of postoperative heparin might suggest that the risk of recurrent HIT should be very low, perhaps negligible, given that the patient would not be receiving heparin when a recurrent immune response might be forming. What then explains the recurrence of HIT in patient 17?

Figure 4 summarizes the clinical and serological features of patient 17 who developed recurrent HIT. He developed an unexpected 86.5% platelet count decrease (from 148 to 20) that began on POD7 while receiving fondaparinux thromboprophylaxis; the patient was shown to have an asymptomatic lower-limb deep vein thrombosis by ultrasound (4Ts score, 8 points). Of note, the patient formed strong HIT antibodies that activated platelets in vitro even in the absence of pharmacologic heparin (Figure 4, open circles). Moreover, there was no cross-reactivity with fondaparinux. Thus, this patient had the serological features of delayed-onset HIT39,40  (also called autoimmune HIT41 ). However, this patient was unusual in another respect: both of his episodes of HIT (in 1998 and 2009) were characterized by the phenomenon of persisting HIT (ie, both HIT episodes lasted for more than 4 weeks before platelet count recovery, which is a rare but recognized course of HIT).42,43  In addition, the patient developed a generalized maculopapular rash (not a reported feature of HIT). Thus, it is difficult to generalize the outcome of this most unusual patient case. Nevertheless, our overall observed frequency of HIT of 5% is in keeping with a recent literature review by Dhakal et al.44  They reviewed 141 reexposure episodes in 136 patients who had a history of HIT and identified 3 patients with recurrent HIT for an overall frequency of 2.1% (95% CI, 0.73% to 6.07%).

Figure 4

Patient with recurrent HIT after UFH reexposure for elective heart surgery. Patient 17 in Table 1. (A) First episode of HIT (1998) complicated by symptomatic deep vein thrombosis (DVT) and pulmonary embolism (PE). (B) Second episode of HIT after intraoperative UFH rechallenge (2009). The patient developed HIT on POD7, with associated asymptomatic DVT shown by venous ultrasound (US). (C) Timing of SRA and EIA seroconversion after UFH reexposure. Both the EIA-IgG and SRA became positive on POD6. IgM seroconversion occurred on POD7. Serotonin release at buffer control (ie, at 0 U/mL UFH) was very strong (>90% from POD10 to POD13), which explains why the patient developed HIT despite not receiving postoperative heparin. (D) Assessment of UFH- and fondaparinux-dependent platelet activation in the presence of patient serum at various dilutions. Strong serum-induced platelet activation (>80% serotonin release) was observed at 0 U/mL UFH (ie, buffer control), using neat and 1/8 diluted serum; strong heparin-dependent platelet activation was shown by the increase in serotonin release at 0.3 U/mL UFH compared with buffer control at higher dilutions of serum (1/16, 1/32, 1/64, 1/128). The absence of fondaparinux-dependent platelet activation argues against fondaparinux cross-reactivity as an explanation for the patient’s persisting thrombocytopenia. Fonda, fondaparinux. Reprinted from Warkentin et al18  with permission.

Figure 4

Patient with recurrent HIT after UFH reexposure for elective heart surgery. Patient 17 in Table 1. (A) First episode of HIT (1998) complicated by symptomatic deep vein thrombosis (DVT) and pulmonary embolism (PE). (B) Second episode of HIT after intraoperative UFH rechallenge (2009). The patient developed HIT on POD7, with associated asymptomatic DVT shown by venous ultrasound (US). (C) Timing of SRA and EIA seroconversion after UFH reexposure. Both the EIA-IgG and SRA became positive on POD6. IgM seroconversion occurred on POD7. Serotonin release at buffer control (ie, at 0 U/mL UFH) was very strong (>90% from POD10 to POD13), which explains why the patient developed HIT despite not receiving postoperative heparin. (D) Assessment of UFH- and fondaparinux-dependent platelet activation in the presence of patient serum at various dilutions. Strong serum-induced platelet activation (>80% serotonin release) was observed at 0 U/mL UFH (ie, buffer control), using neat and 1/8 diluted serum; strong heparin-dependent platelet activation was shown by the increase in serotonin release at 0.3 U/mL UFH compared with buffer control at higher dilutions of serum (1/16, 1/32, 1/64, 1/128). The absence of fondaparinux-dependent platelet activation argues against fondaparinux cross-reactivity as an explanation for the patient’s persisting thrombocytopenia. Fonda, fondaparinux. Reprinted from Warkentin et al18  with permission.

Table 1 shows that the frequency of EIA seroconversion in our case series was 13 (65%) of 20, a rate similar to that in other populations after cardiac surgery.11,13,14,45,46  However, 9 (69.2%) of the 13 EIA-IgG seroconverting patients also showed SRA seroconversion, a proportion much higher than expected (∼5% to 20%).14,45,46  This suggests that patients with a history of HIT are more likely to develop platelet-activating antibodies within their anti-PF4/heparin response and thus to develop HIT if they receive postoperative heparin (not recommended), or if they develop strongly reactive autoimmune-like platelet-activating antibodies (substantial serotonin release at buffer control), a phenomenon we discuss further.

Table 2, which summarizes the positive SRA results among the 13 SRA-seroconverting patients, shows that patient 17, who developed recurrent HIT, exhibited the strongest serum-induced serotonin release at buffer control (ie, 0 U/mL UFH). This in vitro phenomenon is correlated with clinical features of various types of HIT such as delayed-onset HIT,39,40  persisting HIT,43  fondaparinux-induced HIT,47  and spontaneous HIT syndrome,48  and we believe this serological profile explains the potential for recurrent HIT beginning 5 to 10 days after heparin reexposure despite no further heparin being given. These antibodies prompt platelet activation via (non-heparin) platelet-associated glycosaminoglycans.49 

Table 2

Details of SRA seroconversion in 13 heparin reexposures performed for cardiac or vascular surgery

PatientSerotonin-release (%) at concentrations of UFHPlatelet count decrease >30% or thrombosis on or after POD5
00.10.3100
37 59 83 50 No 
17 84 No 
10 11 73 No 
11 75 89 No 
14 50 100 17 No 
15 54 89 96 24 No 
16 47 88 No 
17 95 100 98 HIT: 86% platelet decrease from 148 to 20 beginning on POD7 (Figure 4
22 4 (79*82 89 No; however, atrial fibrillation developed on POD14, and in vitro cross-reactivity (79% serotonin release) with danaparoid 0.3 anti-Xa U/mL was shown (Figure 4
PatientSerotonin-release (%) at concentrations of UFHPlatelet count decrease >30% or thrombosis on or after POD5
00.10.3100
37 59 83 50 No 
17 84 No 
10 11 73 No 
11 75 89 No 
14 50 100 17 No 
15 54 89 96 24 No 
16 47 88 No 
17 95 100 98 HIT: 86% platelet decrease from 148 to 20 beginning on POD7 (Figure 4
22 4 (79*82 89 No; however, atrial fibrillation developed on POD14, and in vitro cross-reactivity (79% serotonin release) with danaparoid 0.3 anti-Xa U/mL was shown (Figure 4
*

In vitro cross-reactivity (79% serotonin release) with danaparoid 0.3 anti-Xa U/mL was demonstrated (Figure 2).

In addition to the high frequency of a positive SRA, there are other noteworthy features of the recurrent anti-PF4/heparin immune response. For example, the timing of antibody formation is no different than in the primary HIT-related immune response.18  For the patients shown in Table 1 for whom evidence of an anti-PF4/heparin immune response was documented after heparin reexposure, the earliest day of detectability of antibodies was POD5 (patient 18), with a median time to a positive SRA observed on POD8 (range, POD6 to POD11), for EIA-IgG on POD7 (range, POD6 to POD9), EIA-IgA POD9 (range, POD7 to POD10), and EIA-IgM POD8 (range, POD5 to POD9). This timing of seroconversion is no sooner than that reported in other patients exhibiting anti-PF4/heparin seroconversion after heparin exposure, either with or without overt HIT.17,21,22  Interestingly, available data regarding antibody isotypes (IgG, IgA, IgM) for the initial episode of HIT, as well as for the subsequent repeat seroconversion after UFH reexposure, never showed formation of a new isotype that had not been present during the time of previous HIT, although the converse was not true (ie, in 3 of 8 patients, an isotype detected at previous HIT was not present at repeat seroconversion).

Given the risk of recurrent HIT in patients who undergo planned reexposure to UFH, even when postoperative heparin is not administered, we recommend platelet count monitoring once per day (or once every other day), not only during the first 4 postoperative days (when risk of recurrent HIT is biologically implausible), but especially during the POD5-to-POD10 window during which recurrent HIT could occur (Table 3). We also recommend collecting and storing serial postoperative blood samples to permit serological investigations, if needed, to determine whether there has been any occurrence of unexpected thrombocytopenia (which is not usually performed in real time). We would treat any platelet count decrease >30% that occurs during the day 5 to day 14 time period as presumptive HIT, as was done for patient 17 (Figure 4). This patient was treated promptly for HIT (increase in fondaparinux from prophylactic to therapeutic dose) and the ultimate clinical course was favorable, with recovery of thrombocytopenia and lack of progression of subclinical deep vein thrombosis.

Table 3

Approach to planned reexposure to heparin for cardiac or vascular surgery, or for hemodialysis, in a patient with a previous history of HIT

ConsiderationsComment
Pre-UFH reexposure considerations  
 1. Does the patient have a strong indication for heparin, such as UFH for cardiac or vascular surgery, or UFH or LMWH for hemodialysis? Numerous non-heparin anticoagulants are available for most other anticoagulant treatment or prophylaxis indications 
 2. Did the previous episode of HIT occur more than 3 to 12 months ago? The probability of a positive SRA is <5% by 3 months and negligible by 12 months, thus allowing for heparin reexposure in emergency settings (when there is no time for repeat testing) 
 3. Do the serological studies indicate that the patient can receive UFH or LMWH safely? EIA-negative or EIA-positive/SRA-negative (or HIPA-negative) status usually indicates that heparin administration is safe 
Recommended intra- and postoperative management (cardiac or vascular surgery)  
 1. Give UFH (and protamine, if indicated) per usual dosing and practice There is negligible risk of triggering acute HIT if platelet-activating antibodies are not present 
 2. Postoperative anticoagulation, whether prophylactic or therapeutic dose, should be given with a non-heparin anticoagulant Fondaparinux and danaparoid are viewed as nonheparin anticoagulants, although there is a low risk of clinically significant cross-reactivity with these drugs 
 3. Perform platelet counts daily as inpatient, every 2 to 3 days as outpatient, until at least postoperative day 10 Platelet count decrease in the first 4 postoperative days does not indicate HIT; however, any new decrease (>30%) in platelet count that begins in the postoperative day 5 to 10 window should be presumed HIT unless proven otherwise 
Hemodialysis  
 1. Do the serological studies indicate that the patient can receive UFH or LMWH safely? EIA-negative or EIA-positive/SRA-negative (or HIPA-negative) status indicates that heparin administration is safe 
 2. Switch to UFH (or LMWH) for hemodialysis per usual practice at dialysis center The intention (in the absence of HIT recurrence) is to continue heparin indefinitely 
 3. Obtain pre- and postdialysis platelet counts and monitor for systemic (anaphylactoid) reactions (until approximately day 14 after the resumption of heparin) Recurrent HIT could occur during the day 5 to 14 window after resuming heparin for hemodialysis; thus, monitoring for HIT recurrence during the day 5 to 14 time period is recommended 
ConsiderationsComment
Pre-UFH reexposure considerations  
 1. Does the patient have a strong indication for heparin, such as UFH for cardiac or vascular surgery, or UFH or LMWH for hemodialysis? Numerous non-heparin anticoagulants are available for most other anticoagulant treatment or prophylaxis indications 
 2. Did the previous episode of HIT occur more than 3 to 12 months ago? The probability of a positive SRA is <5% by 3 months and negligible by 12 months, thus allowing for heparin reexposure in emergency settings (when there is no time for repeat testing) 
 3. Do the serological studies indicate that the patient can receive UFH or LMWH safely? EIA-negative or EIA-positive/SRA-negative (or HIPA-negative) status usually indicates that heparin administration is safe 
Recommended intra- and postoperative management (cardiac or vascular surgery)  
 1. Give UFH (and protamine, if indicated) per usual dosing and practice There is negligible risk of triggering acute HIT if platelet-activating antibodies are not present 
 2. Postoperative anticoagulation, whether prophylactic or therapeutic dose, should be given with a non-heparin anticoagulant Fondaparinux and danaparoid are viewed as nonheparin anticoagulants, although there is a low risk of clinically significant cross-reactivity with these drugs 
 3. Perform platelet counts daily as inpatient, every 2 to 3 days as outpatient, until at least postoperative day 10 Platelet count decrease in the first 4 postoperative days does not indicate HIT; however, any new decrease (>30%) in platelet count that begins in the postoperative day 5 to 10 window should be presumed HIT unless proven otherwise 
Hemodialysis  
 1. Do the serological studies indicate that the patient can receive UFH or LMWH safely? EIA-negative or EIA-positive/SRA-negative (or HIPA-negative) status indicates that heparin administration is safe 
 2. Switch to UFH (or LMWH) for hemodialysis per usual practice at dialysis center The intention (in the absence of HIT recurrence) is to continue heparin indefinitely 
 3. Obtain pre- and postdialysis platelet counts and monitor for systemic (anaphylactoid) reactions (until approximately day 14 after the resumption of heparin) Recurrent HIT could occur during the day 5 to 14 window after resuming heparin for hemodialysis; thus, monitoring for HIT recurrence during the day 5 to 14 time period is recommended 

The above 3 illustrative cases were real. But how would we treat the following (theoretical) case? Consider a patient with a history of definite previous HIT with no detectable HIT antibodies who now requires cardiac surgery for replacement of a mechanical heart valve. The patient is receiving chronic warfarin therapy. Classic bridging with UFH or LMWH is problematic because of the risk of retriggering HIT antibodies before surgery.

Our plan would be to hold warfarin for 2 days before surgery and give prothrombin complex concentrates immediately before surgery, along with a small dose of intravenous vitamin K. This would avoid the need for preoperative heparin bridging. Prothrombin complex concentrates contain heparin,50  so we would avoid giving them a few days before surgery.) The cardiac surgery would then proceed with usual heparinization and protamine reversal. During the postoperative period, once hemostasis was apparent, we would recommend that intravenous UFH be given, starting at a low dose (eg, 500 U/h), with gradual increase over the next 3 to 4 days. Although this would seem to break the rule about avoiding UFH other than for surgery, in this case, UFH is the preferred anticoagulant for mechanical prostheses, and it would (in theory) be safe for the first 96 hours during the early stages of warfarin therapy. Then, in the evening of the fourth day, we would begin dosing with 5 mg fondaparinux and then give 7.5 mg the morning of the fifth day, with the plan being to continue fondaparinux once per day while continuing warfarin overlap (an alternative anticoagulant option would be danaparoid, if available). Platelet count monitoring for HIT recurrence would be performed until at least POD10.

This article has focused on managing the patient with a previous history of HIT who requires cardiac or vascular surgery. But what about the other clinical setting for which UFH is the anticoagulant agent of choice—hemodialysis? Hemodialysis and other forms of renal replacement therapy have an important distinguishing feature from cardiac/vascular surgery, namely that they are not a one-off exposure but rather a therapy that will require continuous or intermittent heparin exposure for a period of at least several days, if not weeks, months, or indefinitely. Thus, acute hemodialysis and other renal replacement therapy will generally require anticoagulation with a non-heparin anticoagulant (eg, argatroban,51  danaparoid,52  bivalirudin,53  or fondaparinux54-56 ), which at least in theory can be continued indefinitely. However, the expense and inconvenience of using an alternative non-heparin anticoagulant for intermittent hemodialysis 3 times per week on an indefinite basis has led some investigators57-59  to resume UFH anticoagulation after HIT antibody seroreversion.

Hartman et al57  reported no instances of HIT recurrence in 3 patients who resumed anticoagulation with LMWH agent nadroparin; the patients had undergone hemodialysis with lepirudin after HIT was diagnosed because of heparin-induced anaphylactoid reactions during hemodialysis, which led to recognition of HIT. More recently, Wanaka et al58,59  reported unremarkable resumption of UFH anticoagulation for hemodialysis in 14 patients after antibody seroreversion; however, HIT recurred in their fifteenth patient who underwent UFH reexposure for hemodialysis.59  This 69-year-old male developed recurrent thrombocytopenia and circuit clotting on the fifth hemodialysis session (ie, day 10) after resumption of UFH, which is consistent with our own experience that HIT recurrence (from negative SRA baseline) does not occur more quickly than what is usually seen with typical-onset HIT. On the basis of this experience, it would seem reasonable to consider resumption of UFH in a patient with previous HIT who is undergoing chronic hemodialysis, although this would require a thorough discussion of the risks and benefits with the patient. If UFH were to be resumed, it would be prudent to measure pre- and postdialysis platelet counts and to collect pre- and postdialysis blood samples for testing for HIT antibodies for at least 2 weeks after resumption of UFH or LMWH anticoagulation. Although the risk of recurrent HIT is likely to be low (∼5% frequency), vigilance for HIT recurrence is key. On theoretical grounds, it might be reasonable to infuse UFH slowly (over 30 minutes) from days 5 to 14 after reinitiating UFH rather than as a bolus because of the potential for an anaphylactoid reaction with bolus UFH administration.60  We do not have any personal experience with resuming UFH or LMWH in this clinical setting. A further consideration is that later recurrence of HIT is possible if intercurrent surgery or another acute event resets the HIT immunological clock.42 

Subacute HIT is defined as a patient who has detectable platelet-activating HIT antibodies but who has a normal platelet count.31,38  Such a clinical situation will usually occur for a relatively short period of time (several weeks) after recovery from an episode of HIT, or in a patient who has formed (platelet-activating) HIT antibodies after UFH exposure but in whom HIT might not have occurred (perhaps because postoperative heparin was not being given). What are the potentially adverse outcomes if heparin (UFH or LMWH) is administered to such a patient?

It is clear that administration of heparin in therapeutic doses (eg, 5000 U bolus of UFH17,60 ; 12 000 U injection of LMWH dalteparin61 ) can result in abrupt platelet count decrease (rapid-onset HIT), sometimes with a life-threatening anaphylactoid reaction.59,60  Although rare, acute intraoperative thrombosis has also been reported when UFH (∼5000 U) has been administered to a patient undergoing vascular surgery.62  (Interestingly, patient cases have been reported in which UFH to permit emergency thromboembolectomy did not result in adverse consequences.63 ) To our knowledge, there are no reports of patients who have experienced adverse consequences of UFH administered for CPB in which the explanation was unrecognized presence of (platelet-activating) HIT antibodies at time of cardiac surgery. In fact, a case was reported64  in which a patient had an uneventful intraoperative course, and it was subsequently proven that the patient was in the early stages of acute HIT when UFH was administered for CPB. This suggests that the very large doses of UFH routinely given during cardiac surgery might be relatively protective against adverse consequences of HIT antibodies. Supportive data include observations in patients with a ventricular assist device who have acute or recent HIT and who underwent heart transplantation with UFH irrespective of their anti-PF4/heparin antibody status; although the authors found that antibody-positive patients had lower postoperative platelet counts compared with controls, none developed intraoperative thrombosis.65  HIT is a disorder characterized by many paradoxes, and it is not inconceivable that the risk of catastrophic consequences of high doses of UFH administered during cardiac surgery in a patient with HIT antibodies are much lower than one might surmise.

Table 3 summarizes concepts regarding heparin reexposure and our recommendations for patients with a previous history of HIT.

The authors acknowledge the support of the McMaster Platelet Immunology Laboratory in performing the testing for HIT antibodies, both PF4-dependent immunoassays and the SRA washed platelet assay, to identify the serological patterns that underlie the HIT and non-HIT-associated anti-PF4/heparin immune responses, both serconversion and seroreversion, especially the work of Jo-Ann I. Sheppard (who also prepared the figures) and Jane C. Moore. T.E.W. thanks John G. Kelton and Andreas Greinacher for their ongoing support.

Contribution: T.E.W. wrote the initial draft, contributed cases 1 and 3, and summarized the data for the first 2 tables; J.A.M.A. contributed case 2; and both authors revised the manuscript.

Conflict-of-interest disclosure: T.E.W. has received lecture honoraria from Pfizer Canada and Instrumentation Laboratory, royalties from Taylor & Francis (Informa), and consulting fees and research funding from W.L. Gore and Instrumentation Laboratory and has provided expert witness testimony relating to HIT. The remaining author declares no competing financial interests.

Correspondence: Theodore E. Warkentin, Hamilton General Hospital, Room 1-270B, 237 Barton St East, Hamilton, ON L8L 2X2, Canada; e-mail: twarken@mcmaster.ca.

1
Greinacher
 
A
Pötzsch
 
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
, vol. 
71
 
2
(pg. 
247
-
251
)
2
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
, vol. 
83
 
11
(pg. 
3232
-
3239
)
3
Warkentin
 
TE
Greinacher
 
A
Gruel
 
Y
Aster
 
RH
Chong
 
BH
scientific and standardization committee of the international society on thrombosis and haemostasis
Laboratory testing for heparin-induced thrombocytopenia: a conceptual framework and implications for diagnosis.
J Thromb Haemost
2011
, vol. 
9
 
12
(pg. 
2498
-
2500
)
4
Cuker
 
A
Clinical and laboratory diagnosis of heparin-induced thrombocytopenia: an integrated approach.
Semin Thromb Hemost
2014
, vol. 
40
 
1
(pg. 
106
-
114
)
5
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
, vol. 
127
 
9
(pg. 
804
-
812
)
6
Warkentin
 
TE
HITlights: a career perspective on heparin-induced thrombocytopenia.
Am J Hematol
2012
, vol. 
87
 
Suppl 1
(pg. 
S92
-
S99
)
7
Greinacher
 
A
CLINICAL PRACTICE. Heparin-Induced Thrombocytopenia.
N Engl J Med
2015
, vol. 
373
 
3
(pg. 
252
-
261
)
8
Smythe
 
MA
Koerber
 
JM
Mattson
 
JC
The incidence of recognized heparin-induced thrombocytopenia in a large, tertiary care teaching hospital.
Chest
2007
, vol. 
131
 
6
(pg. 
1644
-
1649
)
9
Warkentin
 
TE
Roberts
 
RS
Hirsh
 
J
Kelton
 
JG
An improved definition of immune heparin-induced thrombocytopenia in postoperative orthopedic patients.
Arch Intern Med
2003
, vol. 
163
 
20
(pg. 
2518
-
2524
)
10
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
, vol. 
106
 
8
(pg. 
2921
-
2922
)
11
Warkentin
 
TE
Sheppard
 
JA
Horsewood
 
P
Simpson
 
PJ
Moore
 
JC
Kelton
 
JG
Impact of the patient population on the risk for heparin-induced thrombocytopenia.
Blood
2000
, vol. 
96
 
5
(pg. 
1703
-
1708
)
12
Warkentin
 
TE
Greinacher
 
A
Heparin-induced thrombocytopenia and cardiac surgery.
Ann Thorac Surg
2003
, vol. 
76
 
6
(pg. 
2121
-
2131
)
13
Pouplard
 
C
May
 
MA
Iochmann
 
S
, et al. 
Antibodies to platelet factor 4-heparin after cardiopulmonary bypass in patients anticoagulated with unfractionated heparin or a low-molecular-weight heparin : clinical implications for heparin-induced thrombocytopenia.
Circulation
1999
, vol. 
99
 
19
(pg. 
2530
-
2536
)
14
Pouplard
 
C
May
 
MA
Regina
 
S
Marchand
 
M
Fusciardi
 
J
Gruel
 
Y
Changes in platelet count after cardiac surgery can effectively predict the development of pathogenic heparin-dependent antibodies.
Br J Haematol
2005
, vol. 
128
 
6
(pg. 
837
-
841
)
15
Warkentin
 
TE
Heparin-induced thrombocytopenia in critically ill patients.
Semin Thromb Hemost
2015
, vol. 
41
 
1
(pg. 
49
-
60
)
16
Warkentin
 
TE
 
How I diagnose and manage HIT. Hematology Am Soc Hematol Educ Program. 2011;2011:143-149
17
Warkentin
 
TE
Kelton
 
JG
Temporal aspects of heparin-induced thrombocytopenia.
N Engl J Med
2001
, vol. 
344
 
17
(pg. 
1286
-
1292
)
18
Warkentin
 
TE
Sheppard
 
JA
Serological investigation of patients with a previous history of heparin-induced thrombocytopenia who are reexposed to heparin.
Blood
2014
, vol. 
123
 
16
(pg. 
2485
-
2493
)
19
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
, vol. 
343
 
7
pg. 
515
 
20
Mahmoudi
 
M
Vaccination.
In: Immunology Made Ridiculously Simple
2014
Miami, FL
MedMaster Inc.
(pg. 
54
-
61
)
21
Warkentin
 
TE
Sheppard
 
JA
Moore
 
JC
Cook
 
RJ
Kelton
 
JG
Studies of the immune response in heparin-induced thrombocytopenia.
Blood
2009
, vol. 
113
 
20
(pg. 
4963
-
4969
)
22
Greinacher
 
A
Kohlmann
 
T
Strobel
 
U
Sheppard
 
JA
Warkentin
 
TE
The temporal profile of the anti-PF4/heparin immune response.
Blood
2009
, vol. 
113
 
20
(pg. 
4970
-
4976
)
23
Gruel
 
Y
Lang
 
M
Darnige
 
L
, et al. 
Fatal effect of re-exposure to heparin after previous heparin-associated thrombocytopenia and thrombosis.
Lancet
1990
, vol. 
336
 
8722
(pg. 
1077
-
1078
)
24
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
, vol. 
99
 
4
(pg. 
779
-
781
)
25
Warkentin
 
TE
Sheppard
 
JA
Manheim
 
JC
HIT complicating fondaparinux prophylaxis: fondaparinux-dependent platelet activation as a marker for fondaparinux-induced HIT.
Thromb Haemost
2014
, vol. 
112
 
6
(pg. 
1319
-
1322
)
26
Lubenow
 
N
Kempf
 
R
Eichner
 
A
Eichler
 
P
Carlsson
 
LE
Greinacher
 
A
Heparin-induced thrombocytopenia: temporal pattern of thrombocytopenia in relation to initial use or reexposure to heparin.
Chest
2002
, vol. 
122
 
1
(pg. 
37
-
42
)
27
Crookston
 
JH
Crookston
 
MC
Burnie
 
KL
, et al. 
Hereditary erythroblastic multinuclearity associated with a positive acidified-serum test: a type of congenital dyserythropoietic anaemia.
Br J Haematol
1969
, vol. 
17
 
1
(pg. 
11
-
26
)
28
Warkentin
 
TE
Sheppard
 
JA
Heels-Ansdell
 
D
, et al. 
Canadian Critical Care Trials Group; Australian and New Zealand Intensive Care Society Clinical Trials Group
Heparin-induced thrombocytopenia in medical surgical critical illness.
Chest
2013
, vol. 
144
 
3
(pg. 
848
-
858
)
29
Chong
 
BH
Magnani
 
HN
Warkentin
 
TE
Greinacher
 
A
Danaparoid for the treatment of heparin-induced thrombocytopenia.
Heparin-Induced Thrombocytopenia
2013
5th ed
Boca Raton, FL
CRC Press
(pg. 
466
-
488
)
30
Warkentin
 
TE
Davidson
 
BL
Büller
 
HR
, et al. 
Prevalence and risk of preexisting heparin-induced thrombocytopenia antibodies in patients with acute VTE.
Chest
2011
, vol. 
140
 
2
(pg. 
366
-
373
)
31
Selleng
 
S
Haneya
 
A
Hirt
 
S
Selleng
 
K
Schmid
 
C
Greinacher
 
A
Management of anticoagulation in patients with subacute heparin-induced thrombocytopenia scheduled for heart transplantation.
Blood
2008
, vol. 
112
 
10
(pg. 
4024
-
4027
)
32
Warkentin
 
TE
Sheppard
 
JA
Chu
 
FV
Kapoor
 
A
Crowther
 
MA
Gangji
 
A
Plasma exchange to remove HIT antibodies: dissociation between enzyme-immunoassay and platelet activation test reactivities.
Blood
2015
, vol. 
125
 
1
(pg. 
195
-
198
)
33
Koster
 
A
Dyke
 
CM
Aldea
 
G
, et al. 
Bivalirudin during cardiopulmonary bypass in patients with previous or acute heparin-induced thrombocytopenia and heparin antibodies: results of the CHOOSE-ON trial.
Ann Thorac Surg
2007
, vol. 
83
 
2
(pg. 
572
-
577
)
34
Czosnowski
 
QA
Finks
 
SW
Rogers
 
KC
Bivalirudin for patients with heparin-induced thrombocytopenia undergoing cardiovascular surgery.
Ann Pharmacother
2008
, vol. 
42
 
9
(pg. 
1304
-
1309
)
35
Palatianos
 
G
Michalis
 
A
Alivizatos
 
P
, et al. 
Perioperative use of iloprost in cardiac surgery patients diagnosed with heparin-induced thrombocytopenia-reactive antibodies or with true HIT (HIT-reactive antibodies plus thrombocytopenia): An 11-year experience.
Am J Hematol
2015
, vol. 
90
 
7
(pg. 
608
-
617
)
36
Welsby
 
IJ
Um
 
J
Milano
 
CA
Ortel
 
TL
Arepally
 
G
Plasmapheresis and heparin reexposure as a management strategy for cardiac surgical patients with heparin-induced thrombocytopenia.
Anesth Analg
2010
, vol. 
110
 
1
(pg. 
30
-
35
)
37
Warkentin
 
TE
Koster
 
A
Bivalirudin: a review.
Expert Opin Pharmacother
2005
, vol. 
6
 
8
(pg. 
1349
-
1371
)
38
Koster
 
A
Selleng
 
S
Warkentin
 
TE
Greinacher
 
A
Management of intraoperative anticoagulation in patients with heparin-induced thrombocytopenia undergoing cardiovascular surgery.
Heparin-Induced Thrombocytopenia
2013
5th ed
Boca Raton, FL
CRC Press
(pg. 
550
-
572
)
39
Warkentin
 
TE
Kelton
 
JG
Delayed-onset heparin-induced thrombocytopenia and thrombosis.
Ann Intern Med
2001
, vol. 
135
 
7
(pg. 
502
-
506
)
40
Warkentin
 
TE
Agents for the treatment of heparin-induced thrombocytopenia.
Hematol Oncol Clin North Am
2010
, vol. 
24
 
4
(pg. 
755
-
775, ix
)
41
Tvito
 
A
Bakchoul
 
T
Rowe
 
JM
Greinacher
 
A
Ganzel
 
C
Severe and persistent heparin-induced thrombocytopenia despite fondaparinux treatment.
Am J Hematol
2015
, vol. 
90
 
7
(pg. 
675
-
678
)
42
Warkentin
 
TE
Warkentin
 
TE
Greinacher
 
A
Clinical picture of heparin-induced thrombocytopenia.
Heparin-Induced Thrombocytopenia
2013
5th ed
Boca Raton, FL
CRC Press
(pg. 
24
-
76
)
43
Kopolovic
 
I
Warkentin
 
TE
Progressive thrombocytopenia after cardiac surgery in a 67-year-old man.
CMAJ
2014
, vol. 
186
 
12
(pg. 
929
-
933
)
44
Dhakal
 
P
Giri
 
S
Pathak
 
R
Bhatt
 
VR
Heparin reexposure in patients with a history of heparin-induced thrombocytopenia.
Clin Appl Thromb Hemost
2015
, vol. 
21
 
7
(pg. 
626
-
631
)
45
Warkentin
 
TE
Sheppard
 
JI
Sun
 
JCJ
Jung
 
H
Eikelboom
 
JW
Anti-PF4/heparin antibodies and venous graft occlusion in postcoronary artery bypass surgery patients randomized to postoperative unfractionated heparin or fondaparinux thromboprophylaxis.
J Thromb Haemost
2013
, vol. 
11
 
2
(pg. 
253
-
260
)
46
Bauer
 
TL
Arepally
 
G
Konkle
 
BA
, et al. 
Prevalence of heparin-associated antibodies without thrombosis in patients undergoing cardiopulmonary bypass surgery.
Circulation
1997
, vol. 
95
 
5
(pg. 
1242
-
1246
)
47
Warkentin
 
TE
Chakraborty
 
AK
Sheppard
 
JA
Griffin
 
DK
The serological profile of fondaparinux-associated heparin-induced thrombocytopenia syndrome.
Thromb Haemost
2012
, vol. 
108
 
2
(pg. 
394
-
396
)
48
Warkentin
 
TE
Basciano
 
PA
Knopman
 
J
Bernstein
 
RA
Spontaneous heparin-induced thrombocytopenia syndrome: 2 new cases and a proposal for defining this disorder.
Blood
2014
, vol. 
123
 
23
(pg. 
3651
-
3654
)
49
Padmanabhan
 
A
Jones
 
CG
Bougie
 
DW
, et al. 
Heparin-independent, PF4-dependent binding of HIT antibodies to platelets: implications for HIT pathogenesis.
Blood
2015
, vol. 
125
 
1
(pg. 
155
-
161
)
50
Greinacher
 
A
Michels
 
I
Mueller-Eckhardt
 
C
Heparin-associated thrombocytopenia: the antibody is not heparin specific.
Thromb Haemost
1992
, vol. 
67
 
5
(pg. 
545
-
549
)
51
Murray
 
PT
Reddy
 
BV
Grossman
 
EJ
, et al. 
A prospective comparison of three argatroban treatment regimens during hemodialysis in end-stage renal disease.
Kidney Int
2004
, vol. 
66
 
6
(pg. 
2446
-
2453
)
52
Magnani
 
HN
A review of 122 published outcomes of danaparoid anticoagulation for intermittent haemodialysis.
Thromb Res
2010
, vol. 
125
 
4
(pg. 
e171
-
e176
)
53
Al-Ali
 
FS
Elsayed
 
M
Khalifa
 
S
, et al. 
Successful use of a bivalirudin treatment protocol to prevent extracorporeal thrombosis in ambulatory hemodialysis patients with heparin-induced thrombocytopenia.
Hemodial Int
2015
, vol. 
20
 
2
(pg. 
204
-
207
)
54
Kalicki
 
RM
Aregger
 
F
Alberio
 
L
Lämmle
 
B
Frey
 
FJ
Uehlinger
 
DE
Use of the pentasaccharide fondaparinux as an anticoagulant during haemodialysis.
Thromb Haemost
2007
, vol. 
98
 
6
(pg. 
1200
-
1207
)
55
Speeckaert
 
MM
Devreese
 
KM
Vanholder
 
RC
Dhondt
 
A
Fondaparinux as an alternative to vitamin K antagonists in haemodialysis patients.
Nephrol Dial Transplant
2013
, vol. 
28
 
12
(pg. 
3090
-
3095
)
56
Mahieu
 
E
Claes
 
K
Jacquemin
 
M
, et al. 
Anticoagulation with fondaparinux for hemodiafiltration in patients with heparin-induced thrombocytopenia: dose-finding study and safety evaluation.
Artif Organs
2013
, vol. 
37
 
5
(pg. 
482
-
487
)
57
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
, vol. 
104
 
4
(pg. 
c143
-
c148
)
58
Wanaka
 
K
Matsuo
 
T
Matsuo
 
M
, et al. 
Re-exposure to heparin in uremic patients requiring hemodialysis with heparin-induced thrombocytopenia.
J Thromb Haemost
2010
, vol. 
8
 (pg. 
616
-
618
)
59
Wanaka
 
K
Matsushima
 
H
Kaneko
 
C
, et al. 
Recurrence of heparin-induced thrombocytopenia induced by the re-administration of heparin after negative conversion of HIT antibodies in a patient with hemodialysis [abstract].
J Thromb Haemost
2013
, vol. 
11
 
Suppl 2
pg. 
1095
  
Abstract PO 195
60
Warkentin
 
TE
Greinacher
 
A
Heparin-induced anaphylactic and anaphylactoid reactions: two distinct but overlapping syndromes.
Expert Opin Drug Saf
2009
, vol. 
8
 
2
(pg. 
129
-
144
)
61
Hillis
 
C
Warkentin
 
TE
Taha
 
K
Eikelboom
 
JW
Chills and limb pain following administration of low-molecular-weight heparin for treatment of acute venous thromboembolism.
Am J Hematol
2011
, vol. 
86
 
7
(pg. 
603
-
606
)
62
Warkentin
 
TE
Heparin-induced thrombocytopenia and the anesthesiologist.
Can J Anaesth
2002
, vol. 
49
 
6 Suppl
(pg. 
S36
-
S49
)
63
Warkentin
 
TE
Pai
 
M
Cook
 
RJ
Intraoperative anticoagulation and limb amputations in patients with immune heparin-induced thrombocytopenia who require vascular surgery.
J Thromb Haemost
2012
, vol. 
10
 
1
(pg. 
148
-
150
)
64
Warkentin
 
TE
Sheppard
 
JI
Clinical sample investigation (CSI) hematology: pinpointing the precise onset of heparin-induced thrombocytopenia (HIT).
J Thromb Haemost
2007
, vol. 
5
 
3
(pg. 
636
-
637
)
65
Schroder
 
JN
Daneshmand
 
MA
Villamizar
 
NR
, et al. 
Heparin-induced thrombocytopenia in left ventricular assist device bridge-to-transplant patients.
Ann Thorac Surg
2007
, vol. 
84
 
3
(pg. 
841
-
845, discussion 845-846
)