Heparin-Induced Thrombocytopenia: Is it Under-diagnosed?

At an industry-sponsored symposium held October 23, 2002, in conjunction with the American College of Clinical Pharmacy’s Annual Meeting, three speakers addressed challenges in diagnosing and managing heparin-induced thrombocytopenia.

This program was supported by an unrestricted educational grant from GlaxoSmithKline Oncology.

Heparin-Induced Thrombocytopenia (HIT): Its Diagnosis and Treatment

According to L. Midori Kondo, PharmD, clinical instructor, University of Washington School of Pharmacy, Seattle, heparin-induced thrombocytopenia (HIT) involves a complicated pathophysiology. The condition occurs when heparin and platelet factor IV (PF4) join with IgG (the most common immune globulin) to create a multi-molecular complex that binds to the FC receptor on the platelet. Various events follow, including activation and aggregation of platelets (caused by a change in the IIb/IIIa ), release of procoagulant micro particles from the platelets, and the formation of more PF4. Repetition of the cycle results in a “thrombin storm.” (Figure 1)

There are two types of heparin-induced thrombocytopenia. Nonimmune-mediated HIT occurs in 10–30% of patients receiving heparin therapy. Symptoms (a slight decrease in platelets, with no clinical consequences of thrombosis) appear within one to four days after therapy begins. Management is through careful observation. Immune-mediated HIT, on the other hand, occurs in 1–3% of patients taking heparin. A moderate-to-severe drop in platelets occurs, with antibody formation. Because it takes time for the immune system to respond to heparin and PF4, thrombocytopenia does not form for about 5–14 days. Severe life and limb-threatening thrombosis can result in 30–75% of patients. Management calls for cessation of heparin and commencement of alternate coagulant therapy.

Diagnosis of HIT involves a two-step process. While a positive serologic test provides confirmation, initial clinical assessment is most important. According to Dr. Kondo, HIT should be suspected when a patient experiences a drop in platelet count or develops unexpected thrombosis soon after, or during heparin exposure. Clinical determination of the timing of heparin exposure is key. Onset can be as early as one to two days in patients previously exposed to heparin over the past three months. Thrombotic complications may or may not be present.

The College of American Pathologists calls for platelet count monitoring every other day, with comparisons made to baseline levels. Monitoring should commence on day four in heparin naïve patients and on day one for patients with prior heparin exposure. A study by Warkentin and colleagues demonstrated a wide range of platelet counts in patients with HIT, with a mean count of 60,000 per liter (Semin Hematol 1998;35(4):9-16). An inventory of other current medications should also be determined, to rule out other causes of thrombocytopenia, such as H2 antagonists and some antibiotics.

Dr. Kondo explained that the classic clinical presentation of HIT (a drop in platelets or an unexpected thrombosis that occurs four days after the commencement of heparin therapy) represents just one scenario. Alternative possibilities should be anticipated by emergency room personnel to avoid treatment errors. Delayed thrombosis can occur after cessation of heparin (9–30 days after hospital discharge), in patients whose previous heparin exposure was uncomplicated. Because thrombocytopenia is rare in these patients, a CBC is not the optimal diagnostic tool. Rather, ER clinical personnel should obtain a history of previous heparin administration to avoid the dangerous consequences of prescribing additional heparin to these patients.

In another clinical scenario, patients develop unexplained resistance to anticoagulation when a build-up of excess PF4 neutralizes the benefits of heparin therapy, and results in the development of new or larger clots. An additional possibility to watch for is the occurrence of an unexpected acute systemic event within 5–30 minutes of an IV heparin bolus.

Dr. Kondo explained that confirmation of an HIT diagnosis involves testing for heparin-induced antibodies. Tests are recommended in patients whose platelet count drops below 100,000 or to 50% of baseline; or in patients who have developed either new clots or progression of an existing clot. Two types of tests (functional and antigen) are available. Functional tests exploit the ability of the heparin-PF4-IgG complex to activate and measure platelets. The inconvenience of using washed platelets (for improved sensitivity) and the variability of platelet activation limit the utility of this method. The ELISA antigen test searches for the heparin-PF4-IgG complex and the associated antigen to determine positive results. The incidence of occasional false positives (when heparin alone or another antigen induces an immune response) further underscores the importance of clinical observations in making accurate diagnoses.

Dr. Kondo explained that while an “ideal” HIT test currently does not exist, in theory it would be functional, highly sensitive, highly specific, technically simple with rapid turnaround time, and predictive of which patients were at risk for developing thrombosis.

Though clinical assessments indicate that thrombocytopenia recovers soon after heparin withdrawal, with a median recovery time of four days, the risk of life and limb thrombosis remains without the initiation of alternative anticoagulation. In the absence of foolproof testing to predict thrombosis, Dr. Kondo recommended treating patients who raise clinical suspicion of HIT with alternative anticoagulation. Alternative anticoagulants include direct thrombin inhibitors (with no heparin antibody cross-reactivity), such as argatroban, lepirudin (both FDA approved for HIT) and bivalirudin (not approved for use as an alternative anticoagulant in HIT). The pharmacokinetic profiles of each drug are slightly different (half-life and routes of elimination) which may offer advantages to specific populations. Care must be taken with the use of argatroban (which is metabolized by the liver) in patients with hepatic dysfunction. Conversely, lepirudin presents challenges in renal dysfunction due to renal elimination.

Though warfarin can also be considered an alternative anticoagulant, patients must first receive therapy to prevent further extension of their clot. Transfer to oral warfarin therapy in HIT should begin with a transition phase where concurrent dosing of an alternative agent is administered to achieve adequate anticoagulation. Adequate recovery (demonstrated via platelet count monitoring) must be achieved before shifting to warfarin alone. Continual follow-up monitoring (via phone calls) is needed to ensure that thrombosis does not develop. According to Dr. Kondo, most patients transitioning to warfarin therapy have thrombosis. Warfarin therapy is controversial for patients who just have isolated thrombocytopenia.

In summary, Dr. Kondo emphasized the importance of early diagnosis of HIT through clinical assessment (including physical exams and patient histories), confirmed by laboratory assessment of platelets. “You want to stop every molecule of heparin and start alternative anticoagulation as soon as you expect HIT, to prevent further thrombotic complications,” she advised.

Considerations in the Management of HIT Patients Undergoing PCI

“ More than 700,000 people in the U.S. underwent PCI procedures in 1999,” said Dr. Maureen Smythe, PharmD, FCCP, professor, pharmacy practice, Wayne State University, Detroit, MI. Yet studies indicate that ischemic complications can still occur when unfractionated heparin (UFH) is used as a standard anticoagulant for patients undergoing PCI.

“The most significant limitation to heparin is its inability to have a significant effect on clot bound thrombin, so thrombin generation and platelet aggregation can occur locally in the area of the clot,” said Dr. Smythe. In addition, heparin can cause antithrombin activity to fall, which may be associated with the rebound hypercoagulable state after the discontinuation of UFH. Other problems include the stimulation of platelet aggregation and the incidence of HIT (Sem Thromb Hemostasis 1999; suppl 1:57-60).

According to Dr. Smythe, thrombin, which has procoagulant and anti-fibrinolytic activities and helps in clot formation, can be considered “the most important enzyme in control of homeostasis.” However, it can also be viewed as “one of the culprits in causing some of the acute complications of PCI.” As a result, attempts have been made to evaluate a new class of agents, “direct thrombin inhibitors” (DTIs) for PCI patients. In fact, current Level 2A recommendations from the ACCP call for the use of DTIs as an alternate to UFH in patients with known or suspected HIT undergoing PCI.

One potential advantage of DTIs is their lack of binding to plasma proteins, resulting in predictable anti-coagulant response. Moreover, since direct thrombin is not neutralized by PF4, DTIs maintain anticoagulant activity locally within a clot. Additionally, DTIs can inhibit both fluid phase and clot bound thrombin, resulting in greater attenuation of thrombus growth. Of the three available DTIs, only lepirudin is irreversible. The other two agents (argatroban and bivalirudin) have received FDA indications for PCI.

Several studies have been conducted to assess the efficacy of alternative anticoagulants in PCI patients. These patients face increased risk as they are already in a hypercoagulable state when facing the endovascular disruption associated with PCI. Results from a meta-analysis comparing UFH to DTIs in patients undergoing PCI indicated that DTIs were associated with a significant reduction in the composite endpoint of death and MI (evaluated at the end of treatment and at 7 and 30 days), fueled by a reduction in MI at all time points. DTIs were also associated with a significant reduction in the incidence of major bleeding, with no overall difference in incidence of intracranial bleeding. Research findings demonstrate that DTIs seem to provide more significant results (as measured by risk reduction in death and MI) in PCI patients than in acute coronary syndrome patients, though both patient groups benefit from these agents (Lancet 2002;359(9303):294- 302).

According to Dr. Smythe, “Argatroban is definitely the DTI with the most experience in HIT patients undergoing PCI, and it has been able to demonstrate comparable safety and efficacy.” Argatroban use in patients undergoing PCI is associated with several benefits. Like all DTIs, argatroban is associated with a lack of interaction with the HIT antibody and a significant impact on clot bound thrombin. Additional advantages include a flat dose response curve, short half-life, rapid therapeutic effect, and ease of monitoring with ACT.

Three recent clinical trials evaluated argatroban as an anticoagulant specifically in HIT patients with CAD undergoing PCI. The primary efficacy endpoint (subjective assessment of satisfactory outcome) was achieved in 95% of the 91 patients undergoing their first PCI, and in all of the 21 patients who underwent a repeat PCI. Subjective assessment of adequate coagulation demonstrated 98% efficacy in the initial group and 100% in the “repeat” group.

Moreover, secondary success (defined as absence of death, emergent bypass or Q-wave MI) was achieved in 98% of the “initial” group and 100% of the “repeat” group. Comparisons with historical data indicate that argatroban compares favorably to results obtained with UFH. Another secondary endpoint, angiographic success, was also achieved in 98% of the “initial” group and all patients in the “repeat” group. In both groups, the mean post-procedure stenosis was 11%. Major bleeding only occurred in 1.8% of patients. Com-parisons with data from the EPILOG trial demonstrate that argatroban compares favorably to UFH in terms of major bleeding.

A significant disadvantage associated with DTIs as a class of agents for HIT patients undergoing PCI is the absence of an available reversal agent. Moreover, both argatroban and bivalirudin are considerably more expensive than UFH.

Risk Reduction Strategies for the Prevention of Heparin-Induced Thrombocytopenia (HIT)

“ Bleeding from over-dosing is the number one complication of heparin therapy,” said Jerry Siegel, RPh, FASHP, senior director, pharmaceutical services, The Ohio State University Medical Center (OSUMC). Other risks include thrombosis (if heparin doses are too low), resistance, antibody induction, HIT, and thrombosis syndrome (HITTS).

Mr. Siegel reported that patient data from OSUMC indicated a 1–2 % incidence of HIT. Clinical manifestations include thrombocytopenia and thrombotic syndrome, generally with venous (versus arterial) clots, and skin reactions. Use of low weight heparins do not always provide a means of reducing risk of HIT, according to Mr. Siegel.

Mr. Siegel reviewed various means of reducing risk of HIT incidence. Innovations in delivery mechanisms, including programmable “smart” pumps have improved the safety of heparin delivery in recent years. According to Mr. Siegel, standardizing heparin concentrations (25,000 units/250 ml = 100 units/ml) is “probably the most important thing you can do from a pharmacy perspective,” to reduce the bleeding risk. “Going to weight-based dosing standardized formulas will decrease a lot of your standard heparin-related issues, but it won’t prevent HITTS,” Mr. Siegel said. Currently, OSUMC has a comprehensive computer order entry system, with clinical pharmacist monitoring. This system helps to standardize the ordering process, eliminate calculation errors, force weight-based dosing, and adjust automatically for obese patients.

In addition, computer ordering enables the launching of a hospital-wide “rules engine.” At OSUMC, orders for heparin trigger automatic requests for platelet counts every three days. However, clinical interpretation is still necessary since the “engine” does not automatically monitor platelets, discontinue counts when heparin is discontinued, or order HIT assays. To further enhance the efficacy of the system, a proposed “new rules engine” will monitor platelets for absolute count and issue a warning at designated levels. In addition, a Delta Tracking program is being considered at the hospital, to track rate of change of platelets with the goal of potentially identifying high risk patients.

Mr. Siegel proposed that computers will help hospital personnel consider clinical questions such as whether heparin is overused, whether the occurrence of HITTS is under-reported, and can the monitoring or dosing of heparin impact the occurrence rate of HIT.

The Safety Solutions program instituted by Mr. Siegel at OSUMC last January is another attempt to reduce drug-induced risks. Monthly “news-like” flyers are electronically sent to educated healthcare professionals throughout the hospital. “I’m trying to prevent things that we know will go wrong before they go wrong,” Mr. Siegel said.

A study conducted by Mr. Siegel and colleague, Anya Rockwell, evaluated the use of argatroban (an FDA-approved alternative to heparin for treatment of thrombosis in HIT) at OSUMC. Examination of patient records demonstrated that argatroban was a valuable alternative form of intravenous anticoagulation to heparin, but that baseline liver function tests were a necessary preliminary step to prevent over-anticoagulation. The research also demonstrated that argatroban should be used with caution in patients with liver insufficiency, or those taking herbal or alternative medications, based on the likelihood of increased bleeding and increased INR from an argatroban/herbal interaction. Moreover, given the impact on INR of an interaction of argatroban and warfarin, it is important to monitor INR as patients transfer over to warfarin.

In conclusion, Mr. Siegel outlined several strategies for detecting and preventing HIT. He advised that patients be asked their medication history and/or previous adverse experiences with heparin, and that they be screened for recent surgery, heparin use, and use of herbal or alternative medications. Moreover, he recommended that platelet counts be monitored every third day during heparin treatment, both for absolute counts and rate of change. When necessary, HIT assays should be requested, and when positive, heparin treatment should be discontinued and replaced with argatroban, if the clinical picture is consistent with HIT.

Remaining questions concern the duration of activity of heparin antibodies, the appropriateness of heparin dosing to patients with a history of HITTS, and the possible use of sub-clinical drops and platelet counts as valuable surrogate markers.