Drug-induced thrombocytopenia.
Abstract: * Drug-induced thrombocytopenia was first described in the 19th century, yet our understanding of its pathogenesis continues to evolve. The list of drugs implicated in drug-induced thrombocytopenia is extensive and growing. Many, if not most, of these medications induce thrombocytopenia by immune mechanisms. Because the degree of thrombocytopenia can put patients at risk for serious bleeding, a prompt diagnosis is key to clinical management. The laboratory approach to diagnosing drug-induced thrombocytopenia is 2-pronged. First, nondrug causes of thrombocytopenia must be ruled out. Second, testing for drug-dependent platelet antibodies, available at specialized reference laboratories, often can identify the offending medication, although usually not in time for initial clinical management. Once a medication is suspected of causing thrombocytopenia, it must be discontinued promptly, and the patient should be monitored closely. Thrombocytopenia generally resolves quickly after offending medication withdrawal, and the prognosis of drug-induced thrombocytopenia is then excellent.

(Arch Pathol Lab Med. 2009;133:309-314)
Article Type: Report
Subject: Antibodies (Testing)
Antibodies (Physiological aspects)
Viral antibodies (Testing)
Viral antibodies (Physiological aspects)
Thrombocytopenia (Diagnosis)
Thrombocytopenia (Risk factors)
Thrombocytopenia (Care and treatment)
Patients (Care and treatment)
Patients (Management)
Drugs (Adverse and side effects)
Drugs (Research)
Drugs (Health aspects)
Authors: Kenney, Barton
Stack, Gary
Pub Date: 02/01/2009
Publication: Name: Archives of Pathology & Laboratory Medicine Publisher: College of American Pathologists Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2009 College of American Pathologists ISSN: 1543-2165
Issue: Date: Feb, 2009 Source Volume: 133 Source Issue: 2
Topic: Event Code: 330 Product information; 200 Management dynamics; 310 Science & research Computer Subject: Company business management
Accession Number: 230151968
Full Text: Drug-induced thrombocytopenia (DIT) is an increasingly common cause of isolated thrombocytopenia, a fact that should not be surprising considering today's ever-expanding pharmacopeia. (1) Drug-induced thrombocytopenia was first recognized, albeit by clinical manifestations rather than by laboratory studies, more than 140 years ago by Vipan, (2) who noted the onset of purpura in patients that he treated with quinine. Even today, quinine remains a well-described cause of DIT. (3,4) However, hundreds of additional drugs have since been implicated in DIT. (3,5,6) In part due to the difficulty of definitively diagnosing DIT, the true incidence of the condition is not known. (3,5) It is estimated that up to 25% of critically ill patients develop DIT and that the overall incidence of DIT may be at least 10 cases per million population per year. (1,3) Fortunately, DIT affects only a small proportion of patients exposed to any one of the culprit medications. (3)

In recent times, much attention has been brought to the topic due to the increasingly frequent identification of hepar-ininduced thrombocytopenia (HIT) in hospitalized patients. However, HIT represents only one variant of DIT, with diagnostic and clinical implications quite different from those associated with other medications. In fact, the pathophysiologic mechanisms of DIT are diverse but can be split into 2 major categories: (1) decreased platelet production via marrow suppression and (2) peripheral platelet clearance, usually by one of several possible immune mechanisms. (1,7,8)

Drug-induced thrombocytopenia presents several diagnostic and management challenges. First, it can be difficult to definitively prove that a decrease in platelet count is drug-related, especially in the absence of sufficiently frequent platelet counts to allow for a clearly defined temporal relationship between drug administration and thrombocytopenia. As a result, DIT may be overlooked clinically, particularly because the differential diagnosis of thrombocytopenia can be extensive, especially in the critically ill patient. (3) Some patients are also exposed simultaneously to many medications that could be implicated in DIT and perhaps to the transfusion of multiple blood products, thus further clouding the clinical and immunohematologic picture. Another challenge lies in the absence of rapid and reliable laboratory assays to confirm the diagnosis of DIT. As a result, DIT frequently remains a clinical diagnosis of exclusion and correlation. Clinical management can also be problematic. When other causes of thrombocytopenia, such as disseminated intravascular coagulation, thrombotic thrombocytopenic purpura, and posttransfusion purpura (PTP) have been ruled out, careful assessment of the medication list is essential, and prompt discontinuation of any offending agent is necessary. Deciding which drug(s) to discontinue, and if that drug can indeed be withdrawn safely, often represents a difficult clinical quandary.

CLINICAL FEATURES

Rapid identification of a thrombocytopenia-inducing medication can be quite important, as life-threatening bleeding may occur due to DIT. In general, DIT is manifested clinically by petechiae, bruising, and epistaxis. Bleeding can, however, become more severe and take the form of gastrointestinal or genitourinary mucosal bleeding (so-called wet purpura) or even intracranial or pulmonary hemorrhage. (3,7,9) An extensive review of the literature on all forms of DIT demonstrated a 9% rate of major bleeding in reported cases, with a small but not insignificant mortality rate. An additional 67% of patients demonstrated bleeding that was considered less than severe, a finding that may nonetheless contribute to morbidity in the critically ill patient.5 Heparin-induced thrombocytopenia represents a different clinical entity. The immune mechanism behind HIT initially results in platelet activation before platelet consumption. Consequently, thrombosis, rather than hemorrhage, is a complication seen in up to 50% of patients with HIT. (7,10) For this reason, HIT is often categorized separately from other causes of DIT and is frequently excluded from investigations of this topic.

Most patients with DIT experience moderate to severe thrombocytopenia (platelet count less than 50 000/[micro]L), with the majority reaching nadir levels below 20 000/ [micro]L. (1,11) However, in some cases the platelet count is profoundly reduced, sometimes as low as 1000/[micro]L. (9) This is one reason that DIT is a serious condition that requires prompt attention. A drop in platelet count to a level at high risk for spontaneous hemorrhage may occur rapidly. This is in contrast to the usual course of HIT, which is characterized by a mean platelet nadir of approximately 55 000/[micro]L, and counts below 20 000/[micro]L in fewer than 10% of affected patients. (7,10,11)

PATHOGENESIS AND DIFFERENTIAL DIAGNOSIS

The differential diagnosis of thrombocytopenia is extensive and includes a wide variety of immune and nonimmune causes, as well as drug- and nondrug-related causes. The following offers a brief review of nondrug etiologies, followed by a more detailed summary of the pathogenesis of DIT.

Pseudothrombocytopenia

Before considering the diagnosis of DIT, it is necessary to rule out spurious, in vitro causes of a low platelet count. A look at the peripheral blood smear is warranted to confirm the count and to rule out pseudothrombocytopenia due to platelet clumping or satellitism. Platelet agglutination is particularly common when blood is drawn in tubes using EDTA as an anticoagulant, leading to falsely decreased automated platelet counts. (7,12,13) Some patients possess naturally occurring antibodies against GPIIb/IIIa, which develop significant avidity for their target in the presence of EDTA, leading to platelet agglutination. (13) Satellitism, consisting of platelet rosetting around neutrophils, is also associated with EDTA and is believed to be mediated by similar immune mechanisms. This issue can be avoided with the use of citrate tubes for blood samples used for the platelet count.

Nondrug, Nonimmune Thrombocytopenia

Nondrug causes of true thrombocytopenia also must be ruled out. Generally speaking, the etiology of thrombocytopenia can be categorized by decreased platelet production, increased platelet destruction, or platelet sequestration. 8 In slowly evolving cases of thrombocytopenia, marrow aplasia or suppression due to myelodysplastic or metastatic disease must be considered. In more rapidly developing cases, causes of nonimmune platelet destruction must be investigated, such as disseminated intravascular coagulation and thrombotic thrombocytopenic purpura/ hemolytic uremic syndrome. (7-14) In the absence of clear clinical and supporting laboratory criteria for these conditions, and in the presence of a temporal association with drug exposure, DIT becomes more likely.

Nondrug, Immune Thrombocytopenia

Idiopathic thrombocytopenic purpura results from the idiopathic production of platelet autoantibodies that mediate abnormally rapid clearance of platelets and thrombocytopenia. Idiopathic thrombocytopenic purpura differs from DIT in that the pathogenic antibodies are neither drug-induced nor drug-dependent for reactivity. (8) Although one form of DIT results from drug-induced platelet autoantibody formation, prototypically in response to gold salts or procainamide (see later), the platelet-directed autoantibodies of idiopathic thrombocytopenic purpura have no known inciting stimulus. (3) Idiopathic thrombocytopenic purpura should be high on the list of differential diagnoses for isolated thrombocytopenia, a fact that can make its distinction from DIT all the more difficult.

Posttransfusion purpura is a drug-independent form of alloimmune, rather than autoimmune, thrombocytopenia in which patients make antibodies to platelet antigens that they lack as a result of exposure from pregnancy or transfusion. (8) The most common antigen involved in this process is human platelet antigen 1. On reexposure to the same platelet antigen on subsequent transfusion, these antibodies mediate enhanced clearance of both the transfused and endogenous platelets, resulting in significant thrombocytopenia. PTP is clinically distinguished from DIT because of its temporal association with transfusion, rather than with medication use. The clinical distinction can be difficult to make in patients receiving both transfusion and medications associated with DIT. Testing for PTP is available and consists of 2 possible approaches. The first, and most direct test, is to identify anti-human platelet antigen 1 antibody in the patient's serum. Alternatively, the patient can be antigen-typed or genotyped to confirm a lack of human platelet antigen 1 expression, a finding that would support the possibility of alloimmunization against human platelet antigen 1, especially considering the extremely high prevalence of this antigen.

Nonimmune DIT

Platelet production is dependent on adequate marrow function and a quantitatively and qualitatively sufficient megakaryocyte population. Many antineoplastic medications induce marrow suppression, typically in all hematopoietic cell lines but occasionally in only the megakaryocyte lineage. (1,7) Typically, a dose-dependent decrease in the production of circulating platelets is observed. The most commonly cited drugs in this category represent myeloablative chemotherapeutic compounds, such as alkylating agents, antimetabolites, and cytotoxic drugs, although certain antiviral agents, tolbutamide, and the thiazide diuretics have also been implicated. (15-17)

The time course of DIT related to marrow suppression is generally slow, often for several weeks, reflecting the time required to deplete the megakaryocyte population (Figure 1). (15) Thrombocytopenia is usually an anticipated consequence of therapy with myelosuppressive agents, and as such the diagnosis of associated thrombocytopenia is generally neither difficult nor surprising to the clinical team. (17,18) Therefore, this form of nonimmune DIT should not pose a diagnostic challenge.

Immune DIT

The more common and more problematic form of DIT to diagnose is that mediated by immune processes. Platelets appear to be affected by immune-mediated, drug-dependent destruction more than other marrow-derived cell types. (19) The mean delay in onset of immune-mediated thrombocytopenia is reported to be on the order of 1 to 2 weeks following patients' exposures to an offending, immunogenic drug, although the delay in individual patients ranges from hours to several years. (3,5,11) This variability undoubtedly is determined by whether the patient (1) has preexisting antibodies from a prior exposure to the drug, (2) needs to mount an amnestic response because a previously made antibody has fallen below clinically significant levels, or (3) has never before been exposed to the drug and must undergo primary alloimmunization. Anamnestic responses require on the order of 3 to 10 days, whereas primary alloimmunization requires at least 2 to 3 weeks (Figure 1).

[FIGURE 1 OMITTED]

Two explanations have been proposed for the rare cases of DIT that occur within 24 hours. (9,19) The first, as noted previously, is that the patient has been previously exposed to the offending drug (even if undocumented) and that the acute drop represents a response to preexisting alloantibodies. An explanation for patients with rapid-onset DIT, but no prior drug exposure, is that they have preexisting, "naturally occurring" antibodies. Some researchers have proposed that drug-dependent antibodies could derive from a collection of naturally occurring antibodies with weak affinity for self-antigens, which become more avid for platelet membrane antigens in the setting of exposure to a specific drug. (3,14,19,20) This theory has yet to be fully investigated.

The proposed mechanisms by which some medications induce platelet-specific antibodies or enhance antibody binding to platelets are varied, and only some are well understood. (1,3,11,14,19,21) The most commonly cited mechanisms include hapten-dependent antibody formation, drug-glycoprotein complex antibody formation, autoantibody formation, ligand-induced binding site creation, drug-specific antibody formation, and immune complexmediated antibody formation (Table 1). These mechanisms are described in more detail later.

Perhaps the first to be outlined, as a result of penicillinassociated immune hemolytic anemia, was the hapten-dependent mechanism. (22) This process involves covalent binding of the offending medication to a platelet membrane glycoprotein resulting in a new antigenic structure, that is, a neoepitope. (3,21) A specific immune response is mounted to the hapten-membrane glycoprotein complex.

Another mechanism involves the noncovalent interaction between a drug and a platelet membrane glycoprotein. In this category, antibodies react with the resulting drug-glycoprotein complex. The prototypical medication in this category is quinine, although this process may also be seen with quinidine, nonsteroidal anti-inflammatory drugs, and some sulfonamide antibiotics. (23) The presence of the drug is essential for adequate avidity of antibody for platelet antigen. (3,21) In the absence of the drug, antibody will not bind to the platelet surface, and thrombocytopenia will not occur. Whether the drug itself induces the production of antibody or whether the noncovalent complex of drug and platelet glycoprotein is immunogenic is not entirely understood. (1)

Rarely, antigenic drugs can induce the production of autoantibodies that bind to platelet surface glycoproteins. As true autoantibodies, they can bind platelet antigens in the absence of the drug, and the resultant thrombocytopenia can persist even after the drug is withdrawn. The prototypical medication in this category is gold, with a smaller proportion of cases attributed to procainamide. (3)

Several antiplatelet agents are known to result in DIT. One drug class that can do this is the GPIIb/IIIa inhibitors, represented by eptifibatide and tirofiban, which induce an immune response to platelets via formation of a ligand-induced binding site. These drugs bind to the GPIIb/IIIa complex on the platelet surface, leading to exposure of an immunogenic neoepitope (that the drug is not part of), with resultant antibody formation and platelet destruction in the presence of the drug. (24) The other class of antiplatelet agents is represented by abciximab, a biologic agent consisting of human-murine chimeric Fab fragments specific for platelet GPIIIa. It is believed that some patients possess antibodies, some of which may be naturally occurring, which recognize the murine component of the chimeric fragments. (25) It is well documented that thrombocytopenia may develop on first exposure to abciximab, supporting this concept. (3) Because the anti-bodies are formed against the drug itself, and the platelets are destroyed as bystanders due to drug binding, this process is referred to as drug-specific antibody formation.

The final mechanism is that prototypic of heparin and the heparin-like drugs. In the circulation, heparin binds to soluble or platelet surface-bound platelet factor 4 forming an antigenic structure. Resulting antibodies bind to the heparin-platelet factor 4 complex via their Fab portions, and the Fc portions of these antibodies bind to receptors on the platelet surface resulting in platelet activation. (10) This mechanism is distinct from others involved in DIT, as the end result is platelet activation. Also in contrast to other forms of immune-mediated DIT, the Fc portion of the antibody interacts with platelet surface Fc receptors rather than the Fab portion interacting against surface epitopes. (11)

LABORATORY FINDINGS

A laboratory-based algorithm for diagnosis of DIT is shown in Figure 2. Laboratory testing should first be geared toward ruling out nondrug causes of thrombocytopenia. As described previously, the possibility of pseudothrombocytopenia must be addressed first. Subsequently, testing should be initiated for nonimmune platelet-consumptive states such as disseminated intravascular coagulation and thrombotic thrombocytopenic purpura/ hemolytic uremic syndrome. (7) Although these conditions frequently result in multiple hematologic abnormalities at the laboratory level, DIT in contrast generally presents with isolated thrombocytopenia. Therefore, assessment of the blood smear for schistocytes is warranted, along with coagulation studies, fibrinogen levels, and assays for fibrin split products and D-dimer. Additional studies to identify hemolysis, such as lactate dehydrogenase, haptoglobin, and bilirubin, are also indicated. If abnormalities are detected in these laboratory parameters, the differential diagnosis shifts toward nondrug causes of thrombocytopenia; if no abnormalities are detected, DIT becomes more likely. It should be noted, however, that idiopathic thrombocytopenic purpura can also present with isolated thrombocytopenia; therefore, careful clinical investigation is necessary to supplement available laboratory data. As mentioned previously, HIT is often categorized separately from other causes of DIT. Because laboratory testing for heparin-induced platelet antibodies is readily available, an attempt is often made to rule out HIT prior to pursuing more elaborate assays for DIT in patients taking heparin in addition to other DIT-associated drugs (note that the algorithm presented in Figure 2 assumes prior laboratory/ clinical exclusion of HIT or that the patient has had no exposure to unfractionated or low molecular weight heparin). Currently available enzyme immunoassays for antibodies against heparin-platelet factor 4 complex allow for sensitive detection of HIT, and the gold standard serotonin-release assay may also be used under some circumstances.

These assays may be applied in combination, and clinical correlation is always required to correctly diagnose or exclude HIT by these methods.

[FIGURE 2 OMITTED]

Bone marrow biopsy may be performed, especially in slowly evolving cases or those not clearly temporally associated with drug exposure. In general, examination can assess for marrow aplasia, myelodysplasia, or replacement by a neoplastic process. Additionally, the megakaryocyte population can be assessed for quantity and size characteristics. Bone marrow subjected to therapy with cytotoxic drugs will frequently be hypoplastic, and megakaryocytes will be decreased in number. (8) In cases of peripheral destruction, assuming enough time has passed for a compensatory reaction, megakaryocyte numbers will be increased well above baseline, and large forms may be seen. (15) Concurrent examination of the peripheral blood smear and the use of automated mean platelet volume measurements may also aid in the distinction between marrow suppression and peripheral platelet destruction, as the latter condition is generally associated with an increase in circulating platelet size. The reticulated platelet assay can provide similar information, as the reticulated platelet count tends to vary proportionately with marrow function, and very high counts can be seen in the setting of peripheral platelet destruction.

For investigating suspected cases of immune-mediated thrombocytopenia, immunoassays are available that can detect platelet-specific antibodies in the serum. These tests are useful in diagnosing idiopathic thrombocytopenic purpura and the autoimmune form of DIT but do not typically detect the drug-dependent antibodies of DIT. More specific reference testing is available for drug-dependent platelet antibodies using enzyme immunoassay or flow cytometry technology. These assays use patient serum and normal donor platelets. In the enzyme immunoassay assay, patient serum is incubated with intact platelets as a target both in the presence and in the absence of the drug in question. Enzyme-labeled anti-immunoglobulin (Ig) G is then added and detected. The flow cytometry assay works similarly, only the anti-IgG is labeled with a fluorescent tag and detected via flow cytometric methods. In both assays, positivity is defined by the emergence of a signal when the drug is present under in vitro conditions, with a lack of signal when the drug is absent. (1,9,26) These assays are currently available only at selected reference laboratories but may become more readily available in the future. Unfortunately, because of the time required to conduct testing for drug-dependent platelet antibodies, results are sometimes not available in time to guide initial clinical management. However, the turnaround time for results can be as short as several days in some cases, which may aid in clinical decision making.

Limitations of these assays for drug-dependent platelet antibodies include poor aqueous solubility of some drugs, which may make it difficult to maintain adequate drug concentrations in vitro. Additionally, the degree to which drug metabolites may contribute to the process of DIT in vivo is unclear. Drug-dependent platelet antibody assays are currently conducted using the parent drugs only, with no attempt to test metabolites. (1,11) One final limitation is that a patient's unique platelet antigenic expression may play an important physiologic role. The use of random platelets in the previously mentioned assays instead of the patient's own platelets may lead to false-negative results, if for example the relevant antigen is less highly expressed on the assay platelets. This has led to the suggestion of using autologous platelets under some testing circumstances. (1)

CURRENT TREATMENT AND PROGNOSIS

One of the difficulties inherent both in the diagnosis of DIT and in the specific identification of the offendingmedication is the frequent treatment of patients with multiple possible medication culprits. (15) The obligation exists to carefully review the medication needs of the patient and to determine where medication substitution should be made to discontinue the offending medication. If at all feasible, any and all possible offending medications should be stopped immediately, and they should be substituted with similarly acting medications that are not immunologically cross-reactive. (3,15) With myelosuppressive medications, thrombocytopenia is expected and managed with platelet transfusion. However, in immune-mediated DIT it is imperative that the offending drug be discontinued as rapidly as possible, as the thrombocytopenia can be severe, acute, and perhaps most importantly reversible. Because the mechanism in immune cases is drug-dependent, and antibody binding is in part linked to the presence of drug, withdrawal of the medication will typically result in resolution of the thrombocytopenia. The time frame for platelet count recovery to greater than 100 000/[micro]L is generally within 1 to 10 days, with a median of about 7 days. (1,3) However, the pharmacokinetics of drug clearance and the effects of altered renal or liver function may induce some variability. Once the drug is withdrawn, the prognosis of DIT is generally excellent.

Determination of the culprit medication can be simple if the patient is receiving only one medication. However, if the patient is receiving multiple medications, identification of the offending agent becomes more problematic and requires that the effect of individual drugs be determined in isolation. This can be accomplished with one-atatime subtraction or substitution approaches. One drug is discontinued or substituted for at a time to see if the platelet count rebounds. Medications are stopped or substituted in order of their likelihood of causing DIT, guided by compilations of the most commonly implicated medications and their levels of evidence (Table 2). (5,6) This oneata-time approach, as opposed to stopping or substituting for all medications at once, poses risks to the patient, because the culprit medication may not be immediately discontinued.

It is less advisable in cases of severe thrombocytopenia and requires that the platelet count be monitored frequently to detect further drops in a timely fashion.

However, a correct, initial "educated guess" can minimize the disruption to the patient's drug therapy. A risk analysis should be undertaken for the individual patient, weighing the risk of thrombocytopenia against the risk of discontinuing a particular medication. In some cases it may be difficult or impossible to stop one or more medications because they may be life-sustaining in the critically ill patient, for example, antibiotics in sepsis.

Other approaches, usually supportive or supplementary, have been attempted, including administration of intravenous immune globulin and/or steroids and use of plasmapheresis. (1) However, the utility of these approaches is unclear, and withdrawal of the offending drug remains paramount. Platelet transfusion during DIT is often used, particularly when platelet counts fall to levels of high risk for spontaneous hemorrhage. However, patients may initially be refractory to platelet transfusion due to the continued presence of antiplatelet antibodies and residual drug in the serum. Moreover, HIT and PTP should be ruled out, because platelet transfusion in these conditions may add "fuel to the fire." In the case of HIT, platelet transfusion may result in thrombotic complications, and in PTP it could theoretically prolong the duration of the thrombocytopenia.

CONCLUSION

Drug-induced thrombocytopenia is a serious condition that should be considered part of the differential diagnosis in a patient with thrombocytopenia who is receiving medication. Drug-induced thrombocytopenia can be difficult to definitively diagnose, particularly in its immune-mediated forms. The diagnosis is still largely made by exclusion of other causes of thrombocytopenia and by correlation of the timing of thrombocytopenia with the administration of an offending medication. Specialized testing for drug-dependent platelet antibodies is becoming increasingly available for confirmation of the identity of an offending medication, although results are not always available in a timely enough manner to guide initial diagnosis and treatment. Careful analysis of the totality of clinical data and laboratory findings can lead to correct diagnosis and ultimately to appropriate patient care.

Accepted for publication June 16, 2008.

References

(1.) Visentin GP, Liu CY. Drug-induced thrombocytopenia. Hematol Oncol Clin North Am. 2007;21:685-696, vi.

(2.) Vipan WH. Quinine as a cause of purpura. Lancet. 1865;2:37.

(3.) Aster RH, Bougie DW. Drug-induced immune thrombocytopenia. N Engl J Med. 2007;357:580-587.

(4.) Bougie DW, Wilker PR, Aster RH. Patients with quinine-induced immune thrombocytopenia have both "drug-dependent" and "drug-specific" antibodies. Blood. 2006;108:922-927.

(5.) George JN, Raskob GE, Shah SR, et al. Drug-induced thrombocytopenia: a systematic review of published case reports. Ann Intern Med. 1998;129:886-890.

(6.) Rizvi MA, Kojouri K, George JN. Drug-induced thrombocytopenia: an updated systematic review. Ann Intern Med. 2001;134:346.

(7.) Patnode NM, Gandhi PJ. Drug-induced thrombocytopenia in the coronary care unit. J Thromb Thrombolysis. 2000;10:155-167.

(8.) Hillman RS, Ault KA, Rinder HM. Hematology in Clinical Practice. New York, NY: McGraw Hill; 2005.

(9.) Von Drygalski A, Curtis BR, Bougie DW, et al. Vancomycin-induced immune thrombocytopenia. N Engl J Med. 2007;356:904-910.

(10.) Hassan Y, Awaisu A, Aziz NA, Aziz NH, Ismail O. Heparin-induced thrombocytopenia and recent advances in its therapy. J Clin Pharm Ther. 2007;32:535544.

(11.) Greinacher A, Eichler P, Lubenow N, Kiefel V. Drug-induced and drugdependent immune thrombocytopenias. Rev Clin Exp Hematol. 2001;5(3):166200.

(12.) Lau LG, Chng WJ, Liu TC. Transfusion medicine illustrated: unnecessary transfusions due to pseudothrombocytopenia. Transfusion. 2004;44:801.

(13.) Braester A. Pseudothrombocytopenia as a pitfall in the treatment of essential thrombocythemia. Eur J Haematol. 2003;70:251-252.

(14.) Rizvi MA, Shah SR, Raskob GE, George JN. Drug-induced thrombocytopenia. Curr Opin Hematol. 1999;6:349-353.

(15.) Wazny LD, Ariano RE. Evaluation and management of drug-induced thrombocytopenia in the acutely ill patient. Pharmacotherapy. 2000;20:292-307.

(16.) Kowdley KV. Hematologic side effects of interferon and ribavirin therapy. J Clin Gastroenterol. 2005;39(suppl 1):S3-S8.

(17.) Lee GR, Bithell TC, Poerster J, Athens J, Lukens J. Wintrobe's Clinical Hematology. Philadelphia, Pa: Lea & Febiger; 2003.

(18.) Carey PJ. Drug-induced myelosuppression: diagnosis and management. Drug Saf. 2003;26:691-706.

(19.) Aster RH. Drug-induced immune cytopenias. Toxicology. 2005;209(2): 149-153.

(20.) Pillai S. Two lymphoid roads diverge: but does antigen bade B cells to take the road less traveled? Immunity. 2005;23:242-244.

(21.) Burgess JK. Molecular mechanisms of drug-induced thrombocytopenia. Curr Opin Hematol. 2001;8:294-298.

(22.) Murphy MF, Riordan T, Minchinton RM, et al. Demonstration of an immunemediated mechanism of penicillin-induced neutropenia and thrombocytopenia. Br J Haematol. 1983;55:155-160.

(23.) Bougie D, Aster R. Immune thrombocytopenia resulting from sensitivity to metabolites of naproxen and acetaminophen. Blood. 2001;97:3846-3850.

(24.) Aster RH. Immune thrombocytopenia caused by glycoprotein IIb/IIIa inhibitors. Chest. 2005;127(suppl 2):S53-S59.

(25.) Curtis BR, Divgi A, Garritty M, Aster RH. Delayed thrombocytopenia after treatment with abciximab: a distinct clinical entity associated with the immune response to the drug. J Thromb Haemost. 2004;2:985-992.

(26.) Grossjohann B, Eichler P, Greinacher A, Santoso S, Kroll H. Ceftriaxone causes drug-induced immune thrombocytopenia and hemolytic anemia: characterization of targets on platelets and red blood cells. Transfusion. 2004;44: 1033-1040.

Barton Kenney, MD; Gary Stack, MD, PhD

From the Department of Laboratory Medicine, Yale University School of Medicine (Drs Kenney and Stack) and the Department of Pathology and Laboratory Medicine, West Haven Veterans Affairs Hospital (Dr Stack), New Haven, Conn.

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: Barton Kenney, MD, Department of Laboratory Medicine, Yale University School of Medicine, 333 Cedar St, New Haven, CT 06512 (e-mail: barton.kenney@yale.edu).
Table 1. Mechanisms of Immune-Mediated Drug-Induced Thrombocytopenia

Mechanism             Description

Hapten-dependent      Drug (hapten) binds covalently to platelet
                      membrane glycoprotein producing a
                      neoepitope recognized by antibody
Drug-glycoprotein     Drug interacts noncovalently with platelet
complex (qui-         membrane glycoprotein; antibody binds to
nine-type)            drug-glycoprotein complex
Ligand-induced        Drug binds to platelet GPIIb/IIIa complex
binding site (fi-     inducing conformational change elsewhere
ban-type)             and formation of a neoepitope recognized
                      by antibody
Drug-specific anti-   Drug consists of chimeric Fab fragments
body                  against GPIIIa with a murine component
                      that is recognized by antibody
Autoantibody          Drug induces an autoantibody that reacts
                      with a platelet surface glycoprotein in
                      the absence of the drug
Immune complex        Drug reacts with platelet factor 4 to
                      produce an antigenic complex against
                      which antibodies react; resulting immune
                      complexes bind to platelet
                      Fc receptors resulting in platelet
                      activation

                      Clinical       Special Laboratory
Mechanism             Consequence    Testing *

Hapten-dependent      Hemorrhage     Drug-dependent
                                     platelet antibody
                                     assay
Drug-glycoprotein     Hemorrhage     Drug-dependent
complex (qui-                        platelet antibody
nine-type)                           assay
Ligand-induced        Hemorrhage     Drug-dependent
binding site (fi-                    platelet antibody
ban-type)                            assay

Drug-specific anti-   Hemorrhage     Drug-dependent
body                                 platelet antibody
                                     assay
Autoantibody          Hemorrhage     Anti-platelet anti-
                                     body assay
                                     (nonspecific)
Immune complex        Thrombosis     Heparin-platelet
                                     factor 4 anti-
                                     body assay

Mechanism             Prototype Drugs

Hapten-dependent      Penicillin, cephalo-
                      sporins

Drug-glycoprotein     Quinine, quinidine,
complex (qui-         NSAIDs([dagger]), sulfon-
nine-type)            amides
Ligand-induced        Eptifibatide,
binding site (fi-     tirofiban, lotrafiban
ban-type)

Drug-specific anti-   Abciximab
body

Autoantibody          Gold salts,
                      procainamide

Immune complex        Unfractionated hepa-
                      rin, low-molecular-

                      weight heparins

* See text for details regarding testing modalities.

([dagger]) NSAIDs indicates nonsteroidal anti-inflammatory drugs.

Table 2. Most Commonly Reported Medications
Implicated in Drug-Induced Thrombocytopenia *

Medication ([dagger])           Drug Class

Gold salts                      Antirheumatic
Abciximab                       Anti-GPIIb/IIIa
                                Fab fragment
Cimetidine                      H2-receptor blocker
Rifampin                        Antimicrobial
Quinidine                       Cinchona alkaloid
Para-aminosalicylic acid        Antimicrobial
Carbamazepine                   Anticonvulsant
Acyclovir                       Antiviral
Phenytoin                       Anticonvulsant
Valproate                       Anticonvulsant
Quinine                         Cinchona alkaloid
Prednisone                      Corticosteroid
Trimethoprim-sulfamethoxazole   Antimicrobial
Cephalosporins                  Antimicrobial
Eptifibatide                    GPIIb/IIIa inhibitor
Interferon                      Antiviral
Hydrochlorothiazide             Antihypertensive
Lotrafiban                      GPIIb/IIIa inhibitor
Procainamide                    Antiarrhythmic
Sulfasalazine                   Anti-inflammatory
Suramin                         Growth factor
                                receptor blocker

* The data are based on a count of published case reports of definite
and probable drug-induced thrombocytopenia as compiled by George
et al, (5) Rizvi et al, (6) and the University of Oklahoma, Norman,
drug induced thrombocytopenia Web site. Available at:
http://moon.ouhsc.edu/jgeorge/DITP.html. Accessed March 31, 2008.

([dagger]) The number of reported cases is not necessarily proportional
to the actual frequency of drug-induced thrombocytopenia attributable
to a particular drug.
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.