Management of heparin-induced thrombocytopenia (HIT) during cardiac or vascular surgery

INTRODUCTION — Heparin-induced thrombocytopenia (HIT) is a rare, potentially life-threatening reaction to heparin in which a patient develops antibodies to a heparin-platelet factor 4 complex that induce platelet activation, thrombocytopenia, and a greatly increased risk for venous and arterial thrombosis.

The diagnosis and management of HIT can be challenging, especially in patients who require systemic anticoagulation for cardiac or vascular surgery. Such patients often have thrombocytopenia due to other causes, may have clinically insignificant antibodies detectable on a HIT immunoassay, or may be in the midst of an evolving immune response. If urgent/emergency surgical intervention is necessary, full evaluation of the presence of HIT may not be possible in the preoperative period. Intraoperative management of HIT is also challenging in such patients, particularly if complete avoidance of exposure to heparin is not feasible. Available evidence to guide management in this setting is very limited.

This topic presents an approach to the perioperative management of individuals with suspected HIT, a previous diagnosis of HIT, or known active HIT who require elective or emergency surgery that typically includes heparin administration (eg, vascular surgery, cardiac surgery with or without cardiopulmonary bypass [CPB]).

Related subjects are presented separately:

●Diagnosis of HIT, including use of the 4 Ts score – (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia".)

●Management of HIT, including available non-heparin anticoagulants – (See "Management of heparin-induced thrombocytopenia".)

●Preoperative considerations for cardiac and vascular surgery – (See "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Heparin-induced thrombocytopenia (HIT)'.)

●Complications of CPB – (See "Intraoperative problems after cardiopulmonary bypass".)

●Other causes of thrombocytopenia in hospitalized patients – (See "Diagnostic approach to thrombocytopenia in adults" and "Drug-induced immune thrombocytopenia".)

OVERVIEW AND SCOPE OF THE PROBLEM

Prevalence, mechanism, and typical findings — HIT is an adverse drug reaction that can occur following exposure to heparin of any type (any unfractionated heparin [UFH] or low molecular weight [LMW] heparin), independent of the dose, administration route, or duration of exposure. Risk factors for HIT include surgery, female sex, and older age, as well as administration of UFH rather than LMW heparin and therapeutic rather than prophylactic heparin dosing. However, there is no dose of heparin that is too low to cause HIT, including heparin flushes or heparin-bonded catheters.

Many of the risk factors for HIT (surgery, use of UFH, use of higher heparin doses) are present in cardiac and vascular surgical patients. As a result, the prevalence of HIT is as high as 1 to 3 percent in these patients, which is higher than most other populations. The following studies illustrate the range of findings:

●In a series of over 10,000 patients who underwent cardiac surgery with or without cardiopulmonary bypass (CPB), 153 (1.5 percent) were tested for HIT antibodies due to clinical suspicion based on platelet counts that had decreased by more than 50 percent or were below 50,000/microL [1]. HIT was diagnosed with a positive functional assay (heparin-induced platelet activation [HIPA]) in 21 of these patients (14 percent of those with suspected HIT; 0.2 percent of the entire cohort). Postoperative thromboembolic complications occurred in 10 patients with a positive HIPA result (48 percent), compared with 2 of the 132 individuals with a negative test (1.5 percent).

●In a series of 329 patients who required more than seven days in the intensive care unit (ICU) following cardiac surgery with CPB, 70 developed thrombocytopenia (21 percent) [2]. Testing for HIT was performed in approximately half of the patients, revealing only six positive results based on a functional assay (8.6 percent of those with thrombocytopenia; only 1.8 percent of the entire group). Among 2242 patients with shorter ICU stays who were also analyzed in this study, only three developed symptomatic HIT (0.1 percent).

●In a series of 306 patients who received a mechanical circulatory support (MCS) device, 4.2 percent had confirmed HIT [3].

The mechanism of HIT involves formation of an immunoglobulin G (IgG) autoantibody directed against endogenous platelet factor 4 (PF4) in a complex with heparin. There is no IgM phase [4]. These antibodies are referred to as "HIT antibodies" or "anti-PF4/heparin antibodies" or simply "PF4/heparin antibodies" [5-7]. They coat and activate platelets, leading to thrombocytopenia and thrombosis:

●Thrombocytopenia – Thrombocytopenia occurs due to removal of IgG-coated platelets by macrophages in the liver or spleen, similar to other drug-induced thrombocytopenias. Typically, the platelet count decreases to <50 percent of the patient's baseline within 5 to 10 days of initiating heparin, consistent with the time for IgG autoantibody production. A rapid drop in the platelet count can occur within one day of heparin exposure in an individual who received heparin in the previous one to three months and developed anti-PF4/heparin antibodies [8].

●Thrombosis – The mechanism of thrombosis is incompletely understood and is thought to include a number of procoagulant effects. Thromboembolic events may be venous or arterial. Thrombosis occurs in as many as 50 percent of patients who receive no treatment; these data are discussed separately. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'Thrombosis'.)

Additional details regarding the clinical presentation and pathophysiology of HIT as well as studies from which the above prevalences are derived are presented separately. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'Pathophysiology'.)

In rare cases, HIT-type antibodies may develop in the absence of heparin exposure (or may persist for a prolonged period following heparin discontinuation). These antibodies can cause platelet activation, thrombocytopenia, and thrombosis (so-called "autoimmune HIT" or "refractory HIT"). (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'Terminology and HIT variants' and "Management of heparin-induced thrombocytopenia", section on 'Autoimmune (heparin-independent) HIT variants'.)

In contrast to HIT, some individuals develop thrombocytopenia within one to two days after heparin administration that is mild, transient, and not associated with platelet activation or thrombosis (sometimes termed "HIT type I"). In the absence of clinical information, distinguishing this entity from clinically significant HIT (ie, "HIT type II") is initially challenging, but diagnosis is usually straightforward once it becomes clear that thrombocytopenia is not severe or persistent or a negative immunoassay result is confirmed. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'Terminology and HIT variants' and 'True HIT versus clinically insignificant anti-PF4/heparin antibodies' below.)

True HIT versus clinically insignificant anti-PF4/heparin antibodies — Not all antibodies to PF4/heparin cause HIT. In some cases, antibodies with this specificity may be detected through routine screening programs (which are not recommended) or when testing is sent in an individual with thrombocytopenia or thrombosis caused by conditions other than HIT. Diagnosis of HIT requires a clinical syndrome of thrombocytopenia (or marked decrease in platelet count that remains within the normal range) together with anti-PF4/heparin antibodies that are capable of causing platelet activation. IgG-specific immunoassays are becoming more widely available, and many "false positives" can be attributed to nonspecific IgM antibodies seen with a polyclonal assay but absent on an IgG-specific assay [9].

There are many immunoassay platforms available, both polyclonal (IgG, IgM, and IgA) and IgG-specific, complicating recommendations based on optical densities (ODs). Also, cutoffs for determining that an assay result is positive or negative differ among different immunoassay platforms. The definitions of low, intermediate, and high OD results should be determined and published by each local laboratory and are often accompanied by a measured heparin specificity or a heparin confirmatory step. Heparin specificity and a higher OD increase the likelihood of a positive functional assay, but evaluation is complex and expert opinion is helpful [10].

Our approach to evaluating clinically significant HIT and the importance of presumptive treatment while awaiting testing are outlined below. (See 'Confirm the diagnosis and determine PF4/heparin antibody status' below.)

Other causes of thrombocytopenia and/or thrombosis — Other potential causes of thrombocytopenia or thrombosis in patients undergoing cardiac or vascular surgery include the following [2]:

●Thrombocytopenia

•Surgical blood loss, with hemodilution due to red blood cell transfusion and/or volume replacement

•Consumption of platelets in the extracorporeal circuit

•Medications such as vancomycin that cause drug-induced thrombocytopenia (table 1)

•Infection and/or sepsis

•Disseminated intravascular coagulation (DIC)

Additional details and an approach to the evaluation of thrombocytopenia are included in separate topic reviews. (See "Diagnostic approach to thrombocytopenia in adults" and "Drug-induced immune thrombocytopenia".)

●Thrombosis

•The surgical procedure itself

•Underlying cardiac or vascular disease

•Indwelling vascular catheters or other devices

•Postoperative infection

Additional details and an approach to the evaluation of thrombosis are included in separate topic reviews. (See "Overview of the causes of venous thrombosis" and "Stroke: Etiology, classification, and epidemiology" and "Overview of intestinal ischemia in adults".)

MANAGEMENT APPROACH

General principles — Management of HIT in individuals undergoing cardiac or vascular surgery can be challenging because avoidance of heparin is a key principle of HIT management, yet use of heparin is a key component of surgical procedures requiring intraoperative anticoagulation due to its favorable properties (eg, parenteral administration, easy monitoring, rapid reversibility with protamine) and extensive clinical experience with its use. Management goals in patients with HIT are to reduce the likelihood of potentially life-threatening thromboembolic complications related to platelet factor 4 (PF4)/heparin antibodies, while at the same time enabling performance of the surgical procedure as safely as possible.

In patients with active HIT, heparin should be scrupulously avoided (by delaying surgery altogether or by using an alternate anticoagulant) because heparin administration in the presence of these antibodies is likely to activate platelets and greatly increase thrombosis risk. If possible, heparin should also be avoided for patients who had HIT in the past due to the risk of creating an anamnestic response that causes HIT to recur. Thus, it is important to establish whether the original diagnosis of HIT was accurate and to determine whether PF4/heparin antibodies are present. (See 'Confirm the diagnosis and determine PF4/heparin antibody status' below and 'Assess the urgency of surgery' below and 'Avoid unnecessary heparin exposure' below.)

If administration of heparin is absolutely necessary, its effects should be minimized, as illustrated in the algorithm (algorithm 1):

●If HIT antibodies are present, an intervention to reduce the antibody effect should be used, such as intravenous immune globulin (IVIG) or plasma exchange. (See 'HIT antibodies present' below.)

●If HIT antibodies are absent, the duration of heparin exposure should be limited to the day of surgery only; subsequent anticoagulation can be achieved with a non-heparin agent. (See 'HIT antibodies absent' below.)

Anticoagulation with a non-heparin agent can be given despite thrombocytopenia because the platelet count is expected to increase rapidly once heparin exposure is stopped.

Antifibrinolytic agents (tranexamic acid, epsilon aminocaproic acid) are routinely used in cardiac surgery. Evidence is lacking regarding use of these agents in individuals with HIT; we are not aware of a reason to avoid them. Details of their use are discussed separately. (See "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Antifibrinolytic administration'.)

Postoperative monitoring of the platelet count is appropriate. In some cases, the antibody status is also assayed. (See 'Postoperative platelet count monitoring' below and "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'HIT antibody testing'.)

Confirm the diagnosis and determine PF4/heparin antibody status

Prior diagnosis of HIT — Patients who carry the diagnosis of HIT should be presumed to have had HIT until proven otherwise (algorithm 1). However, some individuals may be labeled as having had HIT when in fact they did not. A typical example of a patient who may be mislabeled is an individual with a low pretest probability and/or a low titer immunoassay (eg, enzyme-linked immunosorbent assay [ELISA]) that was not confirmed by definitive functional testing for HIT antibodies. This issue is further confounded by the fact that some laboratories report ELISA results without giving the optical density (OD) for this result. In such cases, evaluation of the likelihood of HIT based on the OD may not be possible without contacting the laboratory directly. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'HIT antibody testing'.)

Review of the medical record and the results of prior laboratory testing is necessary to confirm a valid diagnosis. Involvement of the consulting hematologist or laboratory medicine specialist is very helpful when there is any confusion or disagreement regarding the prior diagnosis. (See 'Consult hematology or laboratory medicine' below.)

In individuals with a previous diagnosis of HIT (confirmed or strongly suspected), PF4/heparin antibody testing is used to determine whether the antibodies are detectable or have declined to an undetectable level (algorithm 1). The rationale is that if the antibodies remain present, heparin exposure is likely to precipitate rapid-onset clinical HIT, with thrombocytopenia and possibly life-threatening thrombosis. In contrast, if antibodies are not detectable, heparin exposure may produce an anamnestic response, but patients often have a window of one or more days of heparin use before triggering clinical HIT. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'Timing'.)

Suspected current HIT — Patients suspected to have clinical HIT before or during surgery should be evaluated as well as treated presumptively for HIT until results of the laboratory evaluation are available (algorithm 1).

Our approach to evaluation and presumptive treatment includes a close review of the history, medications, physical examination, and platelet counts to determine the likelihood (pretest probability) of HIT, which is estimated by calculating the 4 Ts score (calculator 1). The score takes into account four variables, each of which is given 0, 1, or 2 points (see "Clinical presentation and diagnosis of heparin-induced thrombocytopenia"):

•Platelet count (degree of Thrombocytopenia)

•Timing of platelet count drop relative to initial heparin exposure

•Presence of Thrombosis

•Possibility of oTher causes of thrombocytopenia

Further management depends on this estimate:

●For those with a low likelihood (low probability, 0 to 3 points on the 4 Ts score), or if HIT has been excluded, we look for other causes of thrombocytopenia (see 'Other causes of thrombocytopenia and/or thrombosis' above). We do not perform routine PF4/heparin antibody testing in individuals who are unlikely to have HIT (ie, individuals with normal [or stable] platelet counts who do not have thrombosis, or those with a low pretest probability of HIT based on the 4 Ts score). However, it should be noted that the 4 Ts score was not developed in the setting of cardiac surgery, which may affect predictive ability; the time course for any decrease in platelet count may be equally helpful [11-13]. Evaluation for HIT may be repeated if clinical circumstances change or if the pretest probability increases.

●For those with an intermediate (4 to 5 points) or high (6 to 8 points) probability 4 Ts score, we take the following precautions:

•Stop all heparin exposure. Sources of heparin are listed in the table (table 2).

•Administer a non-heparin anticoagulant. Options and their advantages and disadvantages are listed in the table (table 3).

•If the individual is receiving warfarin, stop the warfarin and reverse its effect with vitamin K to reduce the risk of venous limb gangrene. Warfarin may be restarted once therapeutic anticoagulation with another non-heparin agent has been achieved. The rationale and supporting evidence are discussed in more detail separately. (See "Management of heparin-induced thrombocytopenia" and "Management of heparin-induced thrombocytopenia", section on 'Hold or reverse warfarin'.)

•Test for PF4/heparin antibodies. Typically, an immunoassay such as an ELISA is done first. If the optical density (OD) is very low in this test, a diagnosis of HIT is ruled out; conversely, HIT is very likely if the OD is very high. Specific OD values and their implications are listed in the table (table 4). For patients with uncertain diagnoses (ie, intermediate OD values), gold-standard functional testing is necessary using a serotonin release assay (SRA) or a heparin-induced platelet activation (HIPA) assay. Functional assays often have a longer turnaround time, particularly if these tests are sent to outside laboratories. Suspicion of an evolving antibody titer (eg, due to a strong clinical suspicion of HIT with a very low ELISA OD value early in the clinical course) may prompt a repeat ELISA.

Notably, the antiplatelet agent ticagrelor may interfere with some functional assays for HIT, causing a false-negative result. As explained in a separate topic, an alternative functional assay should be performed, or if one is not available, the 4 Ts score and optical density on the HIT immunoassay may be used to guide diagnosis. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'Patients receiving ticagrelor'.)

●Once HIT has been diagnosed, we continue anticoagulation with a non-heparin agent (intravenous initially, then oral once platelet count has recovered) for three to six months. Decisions regarding the duration of anticoagulation with a non-heparin agent depend on whether the patient has had a thromboembolic event.

Further details regarding this approach and a general overview of HIT management are illustrated in the algorithm (algorithm 2) and discussed in detail separately. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia" and "Management of heparin-induced thrombocytopenia".)

Assess the urgency of surgery — Delaying surgery may be an appropriate option for individuals with functionally relevant PF4/heparin antibodies if the prior episode of HIT is still resolving (algorithm 1). Once antibodies become undetectable, thrombosis risk is reduced, although specific data regarding risk are limited [14].

Delay of surgery until clinical HIT has resolved and PF4/heparin antibodies are negative, when possible, is consistent with expert opinion, a 2012 guideline from the American College of Chest Physicians (ACCP), and a 2018 guideline from the American Society of Hematology (ASH) [5,15]. In some cases, however, delaying surgery is not feasible.

Decision-making regarding risks and benefits of delaying surgery for a specific patient necessitates consultation with the surgeon to determine urgency, as well as input from the anesthesiologist and consulting hematologist or laboratory medicine expert regarding options for non-heparin therapy and laboratory monitoring.

Avoid unnecessary heparin exposure — Exposure to heparin should be scrupulously avoided, including heparin flushes, heparin-bonded catheters, and other sources of heparin, as listed in the table (table 2). Even small amounts of heparin (eg, heparin flushes) can cause severe consequences, including life-threatening thrombosis, by creating the "HIT antigen" (heparin bound to PF4) on platelet surfaces, especially in individuals who have already developed PF4/heparin antibodies from prior heparin exposure.

If anticoagulation is absolutely necessary for a given procedure, it may be in the patient's best interest to receive intraoperative heparin [16]. Small case series have not shown increases in thromboembolic events when this practice of limiting heparin exposure to the surgical procedure is used [17]. The need for additional interventions to reduce the likelihood of platelet activation by heparin depends on whether HIT antibodies are present, as discussed below, although the level of evidence is low. (See 'Specific patient populations' below.)

If postoperative anticoagulation is needed (eg, for venous thromboembolism [VTE] prophylaxis, cardiac or vascular indications, or active HIT), a non-heparin anticoagulant should be used. Options include direct thrombin inhibitors (eg, argatroban, bivalirudin, dabigatran), direct factor Xa inhibitors (eg, apixaban, edoxaban, rivaroxaban), the indirect factor Xa inhibitor fondaparinux, or danaparoid (not available in the United States), which inhibits thrombin and factor Xa (table 2). The choice among these depends on renal and hepatic function; the likelihood of needing rapid reversal; and institutional factors such as cost, agent availability, or local preferences, as discussed separately. (See "Management of heparin-induced thrombocytopenia", section on 'Anticoagulation'.)

The diagnosis of HIT and the need to avoid unnecessary heparin exposure should be communicated to all clinicians involved in the patient's care. For these reasons, we place signs at the patient's bedside and in the medical record alerting all clinicians of the critical importance of avoiding any heparin exposure.

Monitor platelet count — The best parameter for assessment of clinical HIT is the platelet count. Clinically significant PF4/heparin antibodies cause platelets to be sequestered by deposition in the microvasculature, whereas clinically insignificant antibodies do not affect the platelet count.

We monitor the platelet count according to the ACCP guidelines, which take into account the baseline risk of HIT in different patient populations and with different heparin products and dosing (table 5) [5].

Consult hematology or laboratory medicine — The hematologist or laboratory medicine consultant can assist with diagnosis, laboratory testing, risk assessment, and/or management of HIT in the preoperative, intraoperative, and/or postoperative periods.

Ideally, input is sought as early as possible to allow sufficient time for testing that might be necessary. Many hospital laboratories do not perform functional testing for PF4/heparin antibodies, and send-out testing may have a turnaround time of several days. In some cases, the hematologist or laboratory medicine consultant can facilitate more rapid turnaround.

SPECIFIC PATIENT POPULATIONS

Urgent procedures (no time to do preoperative HIT testing) — In emergency cases in which cardiac or vascular surgery cannot be delayed, there may not be sufficient time to determine whether platelet factor 4 (PF4)/heparin antibodies are present. These individuals are assessed for the likelihood of HIT, and in those with a high pretest probability of HIT, treated as if they have HIT until definitive testing is available (algorithm 1).

If the patient has a history of HIT or suspected HIT, the platelet count is initially used as a surrogate for testing for active HIT, and the interval since the previous episode of clinical HIT is used as a surrogate for estimating the likelihood of circulating PF4/heparin antibodies. Urgent consultation among the surgeon, anesthesiologist, and consulting hematology or laboratory medicine HIT expert is advised.

●If thrombocytopenia or an episode of clinical HIT occurred within approximately the previous three months (or approximately 100 days), pathologic anti-PF4/heparin antibodies are likely still detectable in the patient's plasma [8]. The safest approach is treatment as if the patient has active HIT until definitive laboratory testing is available, as discussed below. (See 'HIT antibodies present' below.)

●If the platelet count is normal and there is a remote history of HIT (eg, years ago), the likelihood that the patient has clinically significant PF4/heparin antibodies is extremely low [18]. In such cases, it is often reasonable to allow intraoperative exposure to heparin, with postoperative thromboprophylaxis using a non-heparin agent if indicated. The probability of a positive functional test for PF4/heparin antibodies is estimated to be <5 percent by three months and virtually zero by 12 months following an episode of clinical HIT, although exceptions have occurred [8,18]. If the timing of a prior suspected episode of HIT is unclear, decision-making depends on the clinical judgment of the surgeon, anesthesiologist, and available consultants with expertise in managing HIT. (See 'HIT antibodies absent' below.)

Elective surgery

Active/current HIT — Patients with active HIT (or a diagnosis of HIT within the preceding three or so months) are in an extremely prothrombotic state because they are likely to have circulating anti-PF4/heparin antibodies that can activate platelets and can cause life-threatening venous and/or arterial thrombosis. Thus, we make every possible attempt to avoid use of heparin and to delay the surgical procedure in these individuals (algorithm 1). If possible, we delay surgery until PF4/heparin antibodies become undetectable, which typically takes approximately three months (up to 100 days); if this is not possible, we use a non-heparin anticoagulant [8,18].

Similar to patients with recent acute venous or arterial thromboembolism due to another etiology, we continue the non-heparin anticoagulant until the risk of recurrent thrombosis has returned to baseline.

Immunoassay positive for PF4/heparin antibodies but no history of HIT — As noted above, we do not advocate testing for PF4/heparin antibodies in the absence of a clinical suspicion for HIT, as it is very common to find antibodies on a HIT immunoassay that have no clinical significance [19]. This may cause unnecessary delays or inappropriate use of heparin alternatives. Other concerns related to immunoassays are noted above. (See 'True HIT versus clinically insignificant anti-PF4/heparin antibodies' above.)

However, if such antibody testing has been performed, the choice of heparin versus an alternative anticoagulant should be based on the clinical need for reversibility and risk of bleeding. If heparin is used, platelet counts should be subsequently monitored. If a functional assay is negative, retesting is prompted by clinical suspicion of HIT to determine whether there is an evolving antibody response.

History of HIT — The risk of recurrent HIT after heparin re-exposure is estimated to be approximately 2 to 5 percent [18]. Therefore, it is prudent to recheck an immunoassay if a history of HIT is reported, especially if the episode was accompanied by thrombosis. Platelet counts should be monitored if the patient is undergoing a surgical procedure that necessarily involves heparin exposure. Use of a non-heparin anticoagulant during surgery is an alternative strategy.

HIT antibodies present — Typically, PF4/heparin antibodies are present for approximately 100 days (approximately three months) but may persist for longer [18]. Intraoperative use of heparin in individuals with HIT antibodies has the potential to promote platelet activation and can precipitate rapid-onset HIT. Therefore, management typically involves selection of one of the following options (algorithm 1):

●Delay surgery – Functional assays will determine platelet-activating properties of PF4/heparin antibodies. Elective surgery can usually be delayed until functional assays are negative or antibodies are no longer detectable. The latter typically requires an interval of approximately three months (up to 100 days) after the initiation of treatment for the HIT episode [18]. Further management once the antibody testing becomes negative is discussed below. (See 'HIT antibodies absent' below.)

●Perform plasmapheresis prior to heparinization – Plasmapheresis may be used to remove heparin/PF4 antibodies [20]. A case series described 11 cardiac surgery patients with acute or subacute HIT who were managed with intraoperative plasmapheresis followed by administration of intraoperative heparin [21]. A single plasmapheresis session reduced the anti-PF4/heparin antibody titer by approximately 50 to 80 percent and reduced antibody activity [22]. None of the patients with reduced titers developed clinical HIT. It is prudent to track platelet count and antibody testing postoperatively and to use a heparin alternative if anticoagulation is required in the postoperative period. There is no evidence that the lower levels of direct thrombin inhibitors needed for medical rather than surgical anticoagulation increase the risk of bleeding when compared with heparin.

●Give intravenous immune globulin prior to heparinization – High-dose intravenous immune globulin (IVIG) may be administered preoperatively to reduce the activity of heparin/PF4 antibodies [20]. In one case report, a patient with a history of HIT and persistent platelet activating antibodies (positive serotonin release assay [SRA]) required urgent femoral-popliteal revascularization for severe limb ischemia [23]. This patient was treated with preoperative IVIG (90 grams daily for two days). The patient tolerated the surgical procedure well without a drop in platelet count or thrombotic complications. In vitro testing demonstrated that addition of IVIG blocked heparin-dependent platelet activation by the patient's serum in a dose-dependent manner (ie, SRA became transiently negative), consistent with observations of in vivo protection from platelet activation.

●Use a non-heparin anticoagulant instead of heparin – Surgical procedures may be performed using a non-heparin anticoagulant such as bivalirudin or argatroban instead of heparin. Both of these drugs are direct thrombin inhibitors with relatively short half-lives. In one survey, bivalirudin was used as an alternative anticoagulant by 75 percent of the 304 clinician respondents (anesthesiologists and perfusionists) [24].

Drawbacks for use of bivalirudin or argatroban include risks for excessive bleeding due to inability to reverse their anticoagulant effects [25]. Argatroban is monitored by the activated partial thromboplastin time (aPTT). There are no standardized methods to monitor the degree of anticoagulation with bivalirudin, although doubling of the activated whole blood clotting time (ACT) or aPTT would suggest an anticoagulant effect. Furthermore, at the conclusion of a surgical procedure, residual unmetabolized anticoagulant can lead to significant bleeding and the need for transfusions.

In addition, there are risks of clotting of stagnant blood due to inadequate anticoagulant effect. Thrombus may form within the cardiopulmonary bypass (CPB) circuit, cardioplegia line, ventricular assist devices, or within the chest itself during cardiac surgical procedures (for bivalirudin, due to cleavage by thrombin) [25-28]. Thus, if perfusion to the patient is interrupted during CPB, blood circulation within the CPB circuit itself must be maintained to avoid this potentially catastrophic complication, typically by using an arteriovenous bridge [29].

•Bivalirudin – Protocols for use of bivalirudin during cardiac surgery with or without CPB have been established (table 6) [25,30-32]. Bivalirudin is metabolized renally and is suitable for individuals with hepatic insufficiency. It has a relatively short half-life of approximately 25 minutes in patients with normal renal function due to its metabolism by blood proteases. Thus, all anticoagulant effects are typically resolved within approximately two hours of administration of the last dose of bivalirudin. This property is useful to achieve hemostasis in the postbypass and postoperative periods. However, perioperative deterioration of renal function reduces clearance, intraoperative hypothermia reduces metabolism, and administration of a large total dose (eg, during a long complex case) may prolong duration of action. (See "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Heparin-induced thrombocytopenia (HIT)'.)

Studies have described successful use of bivalirudin rather than heparin during cardiac surgical procedures with or without CPB in individuals with HIT [25]. In patients without HIT, one trial with random assignment of individuals to undergo CPB using either bivalirudin or unfractionated heparin reported similar outcomes and complication rates for each anticoagulant [30].

•Argatroban – There are very limited data for use of argatroban during cardiac surgery requiring CPB [25,28,33-35]. Catastrophic intraoperative bleeding has been reported in some patients [27,28,36]. Argatroban is metabolized hepatically with a plasma half-life of approximately 40 to 50 minutes and may be suitable for individuals with renal insufficiency. However, its half-life is markedly increased in older or critically ill patients and in those with hepatic or renal dysfunction [37,38]. In nonsurgical and postoperative settings, use of argatroban appears reasonable [25,39].

●Coadminister an anti-platelet agent with heparin – The glycoprotein IIb/IIIa (GPIIb/IIIa) receptor antagonist tirofiban has been administered 10 minutes before planned heparin administration in patients with HIT [40]. After an initial bolus of 10 mcg/kg, tirofiban is administered as a continuous infusion of 0.15 mcg/kg per minute, with discontinuation of the infusion one hour before the conclusion of CPB. Similarly, the P2Y₁₂ inhibitor cangrelor has been administered as an initial bolus of 30 mcg/kg approximately 10 minutes before heparin administration, followed by a continuous infusion at 4.0 mcg/kg per minute, with discontinuation of the infusion at the end of CPB or at the time of protamine administration [41,42].

●Coadminister epoprostenol with heparin – Case reports have described successful use of intravenous epoprostenol (also known as prostacyclin or prostaglandin I₂ [PGI₂]), administered together with heparin immediately prior to or during the procedure, to prevent platelet activation by the PF4/heparin antibodies [43,44]. Since intravenous epoprostenol causes vasodilation, a vasopressor infusion (eg, norepinephrine at 0.05 to 0.1 mcg/kg per minute) is typically necessary to maintain adequate systemic blood pressure. Epoprostenol is discontinued after weaning from CPB and administration of protamine to neutralize heparin anticoagulation. This discontinuation should be gradual, in increments of 5 nanograms/kg per minute every five minutes.

All of these approaches may be reasonable options for managing individuals with active HIT antibodies, although the level of evidence is not high and there are no compelling reasons to select one over the other. In general, the choice depends on institutional expertise, experience, and availability. In some centers, plasmapheresis or IVIG is preferred because these therapies allow the cardiac surgeon to use heparin per their normal protocol rather than managing an unfamiliar anticoagulant during CPB [18]. IVIG is less invasive than plasmapheresis. However, plasmapheresis may have a longer-lasting effect on antibody titer and activity, and is often selected for patients who may be intolerant of postoperative anticoagulation. In other cases, concerns about catastrophic thrombosis with heparin may lead to a decision to use a non-heparin agent, typically bivalirudin, particularly if bleeding risk is low (eg, off-pump or non-complex on-pump cardiac surgical procedures, vascular surgical procedures). In all cases, close communication among the surgeon, anesthesiologist, and consulting hematology or laboratory medicine expert is advised throughout the perioperative period. (See 'Consult hematology or laboratory medicine' above.)

HIT antibodies absent — If PF4/heparin antibodies are undetectable on a functional assay, use of intraoperative heparin for a short duration is reasonable in a patient who had a previous history of HIT and now requires a surgical procedure for which heparin is indicated [14,23].

For such individuals, a brief exposure to heparin on the day of surgery may cause clinically significant PF4/heparin antibodies to be produced within 5 to 10 days by an anamnestic response, but heparin will no longer be present by the time the antibodies develop and clinical HIT is less likely to develop. Antibodies do not invariably recur [8]; however, delayed-onset HIT remains possible. Thus, daily platelet count monitoring, high clinical suspicion, and follow-up with hematology are prudent [45].

A 2016 series described good outcomes in 20 individuals with planned re-exposure to heparin after an episode of clinical HIT; only one developed recurrent HIT in the postoperative period [18].

●Of these 20 patients, 10 had a positive immunoassay (presumed to be waning antibody levels) but a negative functional assay at the time of baseline testing. Of these 10, seven had at least a doubling of the optical density (OD) on the immunoassay, while five developed a positive functional assay (SRA). None developed recurrent HIT during surgery or in the postoperative period.

●Of the 10 individuals who had a negative immunoassay at baseline, six developed a positive immunoassay, and of these, four developed a positive functional assay (SRA). One developed recurrent clinical HIT.

In such patients, it is critically important to limit heparin exposure to the surgical procedure only, especially avoiding any heparin exposure in the preoperative period. All postoperative anticoagulation (eg, for thromboprophylaxis or other indications) should be done with a non-heparin agent. Heparin flushes and other sources of heparin (table 2) should be scrupulously avoided.

POSTOPERATIVE PLATELET COUNT MONITORING — Some patients develop clinical HIT in the postoperative period (approximately 1 to 3 percent after cardiac surgery) [5,29,46-48]. Thus, clinicians should be vigilant for signs of possible HIT in any patient who has received heparin. If HIT is suspected, initial evaluation should focus on the parameters assessed in the 4 Ts score (calculator 1) (see 'Suspected current HIT' above) in conjunction with antibody testing.

There are risks for failing to diagnose (and treat) clinical HIT when it is present, including potentially catastrophic thrombosis. However, unnecessarily treating a patient for HIT also incurs risks, including prolonging hospitalization or unnecessary anticoagulation with increased bleeding risk.

Diagnosis of HIT can be challenging after cardiac surgery or vascular surgical procedures, particularly if cardiopulmonary bypass (CPB) was employed, due to these factors:

●Thrombocytopenia is common in such patients, with platelet count decreases of approximately 40 to 50 percent after CPB, typically occurring during the first 72 postoperative hours. In part, this is due to the prolonged contact of platelets with the artificial surface of an extracorporeal circuit [11,46,49]. Other potential causes for transiently decreased platelet counts include drug-induced thrombocytopenia and post-transfusion purpura. (See "Drug-induced immune thrombocytopenia" and "Immunologic transfusion reactions", section on 'Post-transfusion purpura'.)

●Clinically unimportant platelet factor 4 (PF4)/heparin antibodies often coexist with thrombocytopenia, but are typically causally unrelated [29]. The incidence of heparin-induced antibodies to PF4/heparin is as high as 25 to 70 percent by immunoassay after CPB, and 4 to 20 percent by platelet activation assay [47,50-53]. However, most patients with a positive immunoassay but negative functional assay for PF4/heparin antibodies do not develop clinical HIT [18,29]. A positive antibody test alone does not increase thromboembolic risk after cardiac surgery, and postoperative screening is not recommended [19].

Because the risk of HIT is higher in individuals with a history of HIT, we monitor the patient and the platelet count in the postoperative period.

●Typically, we obtain a platelet count daily. Platelet counts during the peak period of HIT (day 5 through 10 after heparin exposure) are the most critical. If the patient is discharged before hospital day 10, it is prudent to continue checking platelet counts in the outpatient setting every other day or every three days until day 10, if possible [11]. We consider a secondary fall in the platelet count of at least 50 percent that begins between the fifth and tenth postoperative day, and without other apparent causes for the thrombocytopenia, to be highly concerning for HIT [11,12,46,54]. We make a presumptive diagnosis of HIT in these patients and administer a non-heparin anticoagulant (table 3). (See "Management of heparin-induced thrombocytopenia".)

●Daily physical examination for symptoms and signs of thrombosis or thrombotic sequelae. After discharge, the patient is counseled regarding potential symptoms of thromboembolic complications, as delayed HIT has rarely been reported beyond the typical 10-day window.

●Laboratory testing for PF4/heparin antibodies is only done if HIT is suspected based on a secondary fall in platelet count (ie, we do not perform testing for anti-heparin/PF4 antibodies if surgery is uneventful and the platelet count does not change).

INSTRUCTIONS AFTER DISCHARGE — Individuals with a history of HIT should be educated about the condition and the importance of future avoidance of heparin, and this information should be conveyed clearly in the medical record. Use of a wallet card (form 1) or medical alert bracelet may be helpful for some individuals. Additional details are available in a separate topic. (See "Management of heparin-induced thrombocytopenia", section on 'Subsequent management'.)

If an individual with a history of HIT requires anticoagulation following discharge (eg, for venous thromboembolism [VTE] prophylaxis or treatment), a non-heparin anticoagulant should be used. However, we do not reinitiate anticoagulation specifically for HIT in individuals re-exposed to heparin, unless they develop recurrent thrombocytopenia due to a recurrence of HIT.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Anticoagulation" and "Society guideline links: Management of cardiopulmonary bypass" and "Society guideline links: Heparin-induced thrombocytopenia (HIT)".)

SUMMARY AND RECOMMENDATIONS

●Definition – Heparin-induced thrombocytopenia (HIT) is an adverse drug reaction to unfractionated heparin (UFH) or low molecular weight (LMW) heparin, independent of the dose, administration route, or duration of exposure. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia".)

Many risk factors for HIT are present in cardiac and vascular surgical patients; the prevalence of HIT in these individuals is approximately 1 to 3 percent. Risk is higher in individuals with a prior history of HIT. However, there are numerous other potential reasons for thrombocytopenia and/or thrombosis, and not all antibodies to platelet factor 4 (PF4)/heparin cause HIT. (See 'Overview and scope of the problem' above.)

●Evaluation – In individuals with a previous diagnosis of HIT, PF4/heparin antibody testing is used to determine whether the antibodies are detectable or have declined to an undetectable level (algorithm 1). Patients with suspected clinical HIT at the time of surgery should be evaluated using the 4 Ts score (calculator 1) and treated presumptively for HIT until laboratory results are available. (See 'Confirm the diagnosis and determine PF4/heparin antibody status' above.)

●Management – Patients with active HIT or HIT within approximately the preceding three months are in an extremely prothrombotic state and at risk for life-threatening venous and/or arterial thrombosis. We make every possible attempt to avoid heparin or delay surgery in these individuals. Early involvement of the hematologist and/or laboratory medicine specialist is appropriate to facilitate risk assessment, rapid turnaround of laboratory testing, and management decisions, as outlined in the algorithm (algorithm 1). (See 'General principles' above and 'Confirm the diagnosis and determine PF4/heparin antibody status' above and 'Assess the urgency of surgery' above.)

•Avoid heparin – Unnecessary heparin exposure (flushes) are avoided. (See 'Avoid unnecessary heparin exposure' above.)

•Delay surgery – If possible, surgery may be delayed until PF4/heparin antibodies are undetectable. If emergency surgery cannot be delayed, management decisions are based on the platelet count and, if HIT was present before, the interval since the previous HIT episode. (See 'Elective surgery' above.)

•Options if surgery cannot be delayed – If surgery cannot be delayed, options include non-heparin anticoagulation, plasmapheresis, administration of intravenous immune globulin (IVIG), or administration of epoprostenol. Data are extremely limited. Decisions are individualized. (See 'Urgent procedures (no time to do preoperative HIT testing)' above and "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Heparin-induced thrombocytopenia (HIT)'.)

•Monitor platelet count – We monitor the platelet count regularly, including the postoperative period; platelet counts from day 5 through 10 after heparin exposure are most critical. If the patient is discharged before hospital day 10, it is prudent to continue checking platelet counts after discharge. Postoperative testing for PF4/heparin antibodies is only done if HIT is suspected based on a secondary fall in platelet count. Patients should be educated about postoperative considerations and should have information available to present to other clinicians should the need arise. (See 'Postoperative platelet count monitoring' above and 'Instructions after discharge' above.)