Thrombocytopenia Caused by Immunologic Platelet Destruction

Meghan S. Liel

Michael Recht

David C. Calverley

Immune thrombocytopenia (ITP) occurs when platelets undergo premature destruction as a result of autoantibody or immune complex deposition on their membranes. Although this disorder was previously known as idiopathic thrombocytopenic purpura, it is now correctly termed ITP because this nomenclature more clearly reflects the immune-mediated mechanism of the disease.¹ In this chapter, both primary and secondary types of ITP are discussed (Table 47.1). Human immunodeficiencyvirus-related autoimmune thrombocytopenia, which is also in major part a result of the deposition of autoantibody or immune complexes, or both, on the platelet surface, is discussed in Chapter 64.

The diagnosis of ITP is primarily a diagnosis of exclusion, because currently available clinical assays for platelet-associated antibodies or serum antiplatelet antibodies/immune complexes are neither specific nor sensitive enough for routine clinical use. These disorders are characterized by peripheral thrombocytopenia (confirmed by examination of the peripheral smear), with a normal or increased number of megakaryocytes present on bone marrow examination, and absence of splenomegaly. Those patients who have no identifiable underlying cause, which might include infections, collagen vascular diseases, lymphoproliferative disorders (chronic lymphocytic leukemia or lymphoma), or drugs, are diagnosed as primary ITP. In some instances, ITP may be the presenting manifestation of an underlying disease, and additional manifestations appear weeks to months later.

PRIMARY IMMUNE THROMBOCYTOPENIA

Primary ITP refers to thrombocytopenia in which apparent exogenous etiologic factors are lacking and in which diseases known to be associated with secondary thrombocytopenia have been excluded. This syndrome has been recently reviewed.², ³, ⁴, ⁵, ⁶

Acute ITP and chronic ITP differ in incidence, prognosis, and therapy (Table 47.2). These differences illustrate the wide spectrum of disorders that by definition are included in the syndrome, but many clinicians have long believed that acute ITP and chronic ITP are fundamentally different disorders.

TABLE 47.1 IMMUNE THROMBOCYTOPENIA

Primary

Secondary

Infections

Collagen vascular diseases

Lymphoproliferative disorders

Solid tumors

Drugs

Miscellaneous

Incidence

The annual incidence of ITP in the United States is estimated to be 1.6/10,000.⁷ Acute ITP, defined as thrombocytopenia occurring for <6 months and usually resolving spontaneously, most often affects children and young adults. The incidence peaks in the winter and spring, following the incidence of viral infections.⁸, ⁹ Acute ITP is most common between 2 and 6 years of age. Approximately 7% to 28% of children with acute ITP develop the chronic variety.⁹, ¹⁰, ¹¹ Chronic ITP, lasting >6 months and requiring therapy to improve the thrombocytopenia, occurs most commonly in adults. In chronic ITP in adults, the median age is usually 40 to 45 years,¹², ¹³ although in one large series of patients, 74% of 934 cases were younger than age 40 (range, 16 to 87 years of age).¹⁴ The ratio of females to males is nearly 1:1 in acute ITP⁹, ¹⁰, ¹¹, ¹⁵ and 2 to 3:1 in chronic ITP.¹³, ¹⁴

Pathophysiology

The syndrome of ITP is caused by platelet-specific autoantibodies that bind to autologous platelets, which are then rapidly cleared from the circulation by the mononuclear phagocyte system via macrophage Fcγ receptors predominantly in the spleen and liver¹⁶ (Fig. 47.1). The ITP antibody does not fix complement in vitro when tested by the usual techniques, but activation of components of complement on the platelet surface may be demonstrated.¹⁷, ¹⁸

A compensatory increase in platelet production takes place in most patients in response to the autoantibody-mediated platelet destruction described above. In others, however, platelet production appears to be impaired as a result of either intramedullary destruction of antibody-coated platelets by marrow macrophages or the inhibition of megakaryopoiesis. Autoantibodies from ITP patients have been shown to inhibit production of megakaryocytes in vitro, and megakaryocyte apoptosis has also been observed in this setting.¹⁹, ²⁰, ²¹ Turnover studies have shown platelet production to be reduced or inappropriately normal in around two thirds of ITP patients.²², ²³, ²⁴ In one study, megakaryocytes were small in patients with anti-gpIb/V/IX autoantibodies and increased in size and cytoplasmic area in patients with anti-gpIIb/IIIa autoantibodies,
suggesting that the anti-gpIIb/IIIa autoantibody impairs platelet production.²⁵ Megakaryocyte colony formation (colony-forming unit megakaryocyte) is increased in acute ITP.²⁶, ²⁷ In chronic ITP, decreased megakaryocyte colony formation has been reported.²⁸

TABLE 47.2 FEATURES OF ACUTE AND CHRONIC IMMUNE THROMBOCYTOPENIA

Feature	Acute ITP	Chronic ITP
Peak age of incidence	Children, 2-6 y	Adults, 20-40 y
Sex predilection	None	3:1 female to male
Antecedent infection	Common 1-3 wk before	Unusual
Onset of bleeding	Abrupt	Insidious
Hemorrhagic bullae in mouth	Present in severe cases	Usually absent
Platelet count	<20,000/µl	30,000-80,000/µl
Eosinophilia and lymphocytosis	Common	Rare
Duration	2-6 wk; rarely longer	Months or years
Spontaneous remissions	Occur in 80% of cases	Uncommon

FIGURE 47.1. Antiplatelet antibody-induced destruction of platelets (P) in chronic immune thrombocytopenia. Immunoglobulin binds to platelet-associated antigen, resulting in phagocytosis by macrophages (M). Antiplatelet antibody-coated platelets bind to macrophages through macrophage Fc receptors (FcR). •, platelet membrane antigen; γ, platelet autoantibody.

Consistent with findings of inappropriately low platelet production in ITP, plasma levels of endogenous thrombopoietin have not been found to be elevated in ITP.²⁹, ³⁰ This differs from impaired platelet production settings such as following chemotherapy.²⁹, ³⁰, ³¹ This may be a result of clearance of thrombopoietin by the megakaryocyte and platelet mass.³²

Platelet Antibodies

In 1951, Harrington and colleagues first reported that the infusion of plasma from patients with ITP predictably induced thrombocytopenia in normal recipients.³³ Shulman and colleagues then demonstrated that the responsible factor was an immunoglobulin (Ig) of the IgG class that was species-specific and could be removed from serum by absorption to and elution from normal human platelets.³⁴ In addition, the platelet-depressing factor produced effects in vivo that were quantitatively and qualitatively similar to those produced by known platelet antibodies. In 1982, van Leeuwen first identified platelet membrane glycoprotein IIb/IIIa as a dominant antigen by demonstrating that the autoantibodies eluted from ITP platelets bound to normal platelets but not to platelets from patients with Glanzmann thrombasthenia.³⁵ Increased quantities of IgG have been demonstrated on the platelet surface in ITP, and the rate of platelet destruction is proportional to levels of such platelet-associated Ig.³⁶, ³⁷ Autoantibodies are readily found in plasma or platelet eluate in patients with active disease, but are infrequently found in patients in remission.³⁸, ³⁹ Disappearance of the antibodies correlates with the appearance of normal platelet counts.³⁸

The antiplatelet antibodies and platelet antigens involved in ITP have been extensively studied (Table 47.3). Antiplatelet autoantibodies bind to many of the major platelet membrane glycoproteins through the Fab portion of the molecule⁴⁰, ⁴¹ (Fig. 47.2). Platelet gpIIb/IIIa was the first platelet antigen detected, and microtiter assays using platelet monoclonal antibodies to gpIb and gpIIb/IIIa later demonstrated that platelet autoantibodies bind to both major platelet membrane glycoproteins.⁴², ⁴³, ⁴⁴ Serum autoantibodies can react with IIb or IIIa or the intact IIb/IIIa complex.⁴⁴, ⁴⁵, ⁴⁶, ⁴⁷, ⁴⁸, ⁴⁹, ⁵⁰, ⁵¹, ⁵², ⁵³, ⁵⁴, ⁵⁵ Platelet autoantibody binding to gpIbα has been reported, but data indicate that the majority of Ib/V/IX autoantibodies are directed to the complex.⁵⁶, ⁵⁷ Some autoantibodies react with gpIV and α₂β₁, although the plasma from these patients usually also contains autoantibodies reacting with one of the other two major platelet membrane antigens.⁵⁷ Serum antibodies to P-selectin (CD62P) and α_vβ₃ have been detected as well, but their clinical significance is unknown.⁵⁸, ⁵⁹

TABLE 47.3 CHARACTERISTICS OF PLATELET AUTOANTIBODIES IN PRIMARY IMMUNE THROMBOCYTOPENIA

			Comments
Ig subtype		IgG, IgA, IgM^a	IgG and IgA have equal frequency
Antigen specificity		gpIIb/IIIa, Ib/V/IX	Most common antigens
		gpIa/IIa, IV	Never only antigen if positive
		Granule membrane protein-140	One patient reported
		Glycosphingolipid	Rare; unclear pathogenetic importance
		Cardiolipin	Common; ?pathogenetic importance
Presence of antibody
	Plasma	85% of patients
	Platelet eluate	75% of patients
	Complement fixation by antibody	Rare	Unresolved importance
	Circulating immune complexes	Rare	Unresolved importance
gp, glycoprotein; Ig, immunoglobulin.
^aIgM is never present as the only antiplatelet Ig in an individual patient.

Antibodies in ITP sera have also been demonstrated to bind to glycosphingolipids⁶⁰, ⁶¹ and cardiolipin.⁶², ⁶³, ⁶⁴, ⁶⁵ Although two studies identified antiphospholipid antibodies (lupus anticoagulant activity or anticardiolipin antibodies) in 46% and 38%, respectively, of ITP patients at diagnosis, there was little clinical evidence that they played a role in the pathogenesis of the disease or affected outcome.⁶⁶, ⁶⁷

The presence of antibodies against multiple antigens is seen in most patients.⁶⁸ Once destruction of platelets within antigenpresenting cells occurs, this generates a series of neoantigens that in turn results in sufficient antibody to cause thrombocytopenia. This phenomenon is termed epitope spread.⁶⁹ Plasma autoantibodies and autoantibodies eluted from platelets in the same patient may have slight differences in antigen specificities within a membrane glycoprotein complex.⁴⁹ There is evidence from two studies that the specific β₃ antigen epitope in any given patient to which anti-gpIIIa autoantibodies are directed may influence the clinical presentation and course of the disease.⁴⁹, ⁷⁰ Whether these findings can be generalized to other ITP autoantibody epitopes is unknown.

Autoantibodies bind to platelets and cause thrombocytopenia primarily by shortening platelet survival. However, rare autoantibodies have also been reported that bind to glycoproteins and activate platelets.⁷¹, ⁷², ⁷³, ⁷⁴, ⁷⁵ Additionally, one patient with an anti-gpIIIa antibody was reported to have developed an antibodyrelated defect in aggregation and adhesion.⁷⁶

The incidence of serum autoantibodies to platelet gpIIb/IIIa is the same in the acute and chronic forms of childhood ITP (68% vs. 62%, respectively).⁷⁷ The presence of anti-gpIIb/IIIa, therefore, does not predict which children will develop the chronic form of the disease, and, in fact, these data provide evidence that the mechanism may be the same in both acute and chronic forms of ITP.

The role of cell-mediated immunity in ITP remains uncertain, although data from patients with ITP suggest that T lymphocytes
demonstrate phenotypic and functional abnormalities. Platelet reactive T-cell clones can be identified from the peripheral blood of adults and peripheral blood and spleens of children with chronic ITP, suggesting that autoreactive peripheral T lymphocytes may mediate or participate in disease pathophysiology.⁵³, ⁷⁸, ⁷⁹ Adults with ITP often have increased numbers of HLA-DR⁺ T cells, increased numbers of soluble interleukin-2 receptors, and a cytokine profile suggesting the activation of precursor and mature helper T-cells.⁷⁹ CD3⁺ T cells from ITP patients in one DNA microarray study had increased expression of genes involved in cell-mediated cytotoxicity and, in addition, cytotoxic cell-mediated lysis of autologous platelets was shown in active ITP.⁸⁰

FIGURE 47.2. Types of antibody-mediated platelet destruction. A: Platelet autoantibodies bind to variable external and internal platelet epitopes. B: Quinine/quinidine-dependent antibodies. The antibody target is a complex of drug and glycoprotein (GP) (usually gpIb/V/IX or gpIIb/IIIa). C: Heparin-dependent antibodies. The antigen-antibody complex (target: platelet factor 4 [PF4]/heparin) activates platelets by the binding of immunoglobulin G (IgG) to Fc γRIIA on platelets. D: Platelet alloantibodies bind to platelet tertiary conformational epitopes on the platelet membrane. (Modified from Kelton, reference 40.)

Platelet Survival

Platelet survival, measured using ⁵¹chromium- or ¹¹¹indium-labeled platelets, is shortened in ITP, and the survival time can range from 2 to 3 days to a matter of minutes.²², ²³, ²⁴, ⁸⁰, ⁸¹, ⁸², ⁸³, ⁸⁴, ⁸⁵, ⁸⁶, ⁸⁷, ⁸⁸, ⁸⁹ Using ¹¹¹indium-labeled autologous platelets, some investigators found an inverse correlation between venous platelet count and platelet survival,⁸⁶ whereas others have not.⁸⁷ These study differences may relate to the types of patients included, as patients with mild to moderate thrombocytopenia have a longer measured platelet survival than patients with severe thrombocytopenia.

Splenic sequestration and destruction account for the shortened survival in most patients,⁹⁰, ⁹¹ but the liver and the reticuloendothelial cells of the bone marrow can play a major role in the sequestration of antibody-coated platelets, especially in patients with very low platelet counts or continued thrombocytopenia after splenectomy.²², ²³, ⁸⁶ Patients with severe thrombocytopenia have been found to have elevated levels of macrophage colony-stimulating factor⁹² and increased in vitro monocyte-platelet rosette formation.⁹³ The spleen has also been implicated as a site of antibody formation.⁹⁴, ⁹⁵ In a pathologic study of 83 spleens that had been removed after patients did not respond to steroid therapy, investigators found splenic weights of <300 g, prominent secondary lymphoid follicles (28%), foamy macrophages (67%), and megakaryocytic extramedullary hematopoiesis (60%).⁹⁶ These pathologic changes reflect the two major pathogenic roles of the spleen: Antiplatelet antibody production and macrophage-mediated platelet destruction.

Clinical Picture

Immune Thrombocytopenia in Children

In children with acute ITP, the onset of the disorder usually is sudden (Table 47.2). A history of infection preceding the onset of bleeding has been documented repeatedly.⁹⁷, ⁹⁸ In one series, such infections were noted within 3 weeks of the onset of ITP in 84% of cases.⁸, ⁹ Varicella zoster virus and Epstein-Barr virus are the most frequently identifiable viruses, although nonspecific unidentified viral infections predominate.⁹⁹ Acute ITP may also occur after vaccination.¹⁰⁰ Although it is most common between 2 and 5 years of age, ITP occurs in all pediatric age groups.¹⁰¹ The severity of ITP in infants is similar to that in older children, but compared to older children, a relatively small percentage of infants develop chronic ITP. A male preponderance was observed in the infant population in two studies.¹⁰¹, ¹⁰² Even though thrombocytopenia is likely to be severe, the bleeding manifestations of acute ITP in children usually are mild,¹⁰³ and intracranial hemorrhage occurs in <1% of patients. The rare adult with the acute form of the disorder, however, may suffer hemorrhage and a more fulminant course. Acute ITP in children usually is self-limited; spontaneous remissions occur in as many as 90% of patients.⁸, ⁹ The duration of the disease ranges from a few days to a few months, with an average of 4 to 6 weeks.¹⁰⁴ The favorable prognosis of ITP in children reflects the preponderance of the acute form of the disease in this age group. Children with thrombocytopenia of >6 months duration are classified as chronic ITP, although spontaneous remissions may still occur in an occasional child after 6 months.¹⁰⁵, ¹⁰⁶ Fever of mild degree has been reported, and the spleen tip may be palpable in up to 10% of patients, but this is believed to be the same incidence as is seen in normal children.¹⁰⁷ The spleen is usually of normal weight in those patients who proceed to splenectomy.⁹⁷

Immune Thrombocytopenia in Adults

In adults, the onset of the chronic form of the disorder usually is insidious (Table 47.2). A long history of hemorrhagic symptoms of mild to moderate severity is often described by the patient, but antecedent infections or fever are uncommon. Patients with chronic ITP usually have a fluctuating clinical course. Episodes of bleeding may last days or weeks and may be intermittent or even cyclic. Spontaneous remissions are very uncommon in adults, with an estimated occurrence of <5%.¹³, ¹⁴, ¹⁰⁸, ¹⁰⁹, ¹¹⁰ Most spontaneous remissions occur early; however, remissions have been described after 6 months in a small number of patients.¹³ Relapses in some cases are associated with vaccination.¹¹¹ Often the clinical course is surprisingly benign.

Bleeding Manifestations

The hemorrhagic manifestations of ITP are of the purpuric type. Patients with only ecchymoses and petechiae have “dry” purpura; those with mucous membrane bleeding in addition to skin manifestations have “wet” purpura.¹¹² Platelet counts are usually lower and the complication rates higher in those with wet purpura. In a series of 712 patients reported by the Israeli ITP study group, 82% of all patients had bleeding limited to the skin, although 43% of adult women reported menometrorrhagia.⁹⁸

In general, the severity and frequency of hemorrhagic manifestations correlate with the platelet count (Fig. 47.3).¹¹³ Bleeding after trauma without spontaneous hemorrhage is usual in mildly affected patients with platelet counts >50,000/µl. Thrombocytopenia associated with counts between 10,000 and 50,000/µl results in spontaneous hemorrhagic manifestations of varying severity, such as ecchymoses and petechiae. Patients with platelet counts <10,000/µl are at risk for serious morbidity and mortality from bleeding, although the mortality rate is actually quite low.¹¹⁴ Patients who have an increased risk of bleeding include those with a history of bleeding, those with additional bleeding diatheses, and patients >60 years of age.¹², ¹¹⁴ Older patients have also been reported to have an increased incidence of major, life-threatening bleeding.¹², ¹¹⁵, ¹¹⁶

Skin and Mucous Membranes

Spontaneous bleeding into the skin in the form of petechiae is characteristic. These lesions are minute, red to purple hemorrhages that range in size from that of a pinpoint to that of a pinhead (Fig. 47.4). They are flat, do not blanch with pressure, and appear and regress, often in crops, over a period of days. They are most conspicuous in areas of vascular stasis, such as the areas below tourniquet sites, the dependent portions of the body (especially around the ankles), and areas subjected to constriction from belts or stockings, as well as on skin surfaces over bony prominences. The presence of petechiae on the face and neck is unusual, except as the result of coughing.

Ecchymoses may develop on any skin surface. In ITP, they are seldom associated with subcutaneous hematomas and infrequently spread or dissect into deeper or adjacent structures. Large, purple, superficial ecchymoses may be seen, particularly on the back and thighs. Circular ecchymoses often surround even atraumatic venipuncture sites, but external bleeding from such sites is uncommon. Hemorrhagic vesicles or bullae may be seen inside the mouth and on other mucous surfaces. The bullae probably are the result of severe acute thrombocytopenia rather than a specific feature of any particular pathogenetic form.

FIGURE 47.3. Bleeding manifestations in relation to platelet count in patients with primary immune thrombocytopenia. Bleeding manifestations (or duration) are graded from 0 to 4, as follows: 0, no bleeding; 1, minimal, resulting from trauma; 2, spontaneous, but self-limited; 3, spontaneous, requiring special attention (e.g., nasal packs); and 4, massive uncontrolled or poorly controlled. (From Lacey and Penner, reference 113.)

FIGURE 47.4. Petechiae. Pinpoint, nonblanching erythematous capillary bleeding sites are most common in dependent body areas or pressure points.

Gingival bleeding and epistaxis are common. The latter usually responds for a time to conservative measures, such as nasal packing or tamponade, often to recur intermittently. Epistaxis may originate from lesions resembling petechiae in the nasal mucosa. Such lesions also may be found in the mucous membranes of the throat and mouth, sometimes in the absence of cutaneous hemorrhage. In many patients, discrete bleeding points cannot be identified.

The genitourinary tract is a frequent site of bleeding. Menorrhagia may be the only symptom of ITP and may appear for the first time at puberty. Hematuria also is a common symptom, the blood coming from the kidneys, the bladder, or the urethra, although bleeding into the kidney parenchyma is rare. Gastrointestinal bleeding is usually manifested by melena or, less often, by hematemesis.

Central Nervous System

Intracranial hemorrhage is the most serious complication of ITP. Fortunately, it is rare, affecting 1% to 2% or less of patients with severe thrombocytopenia.⁹, ¹¹⁷ The hemorrhages usually are subarachnoid, often are multiple, and vary in size from petechiae to large extravasations of blood. Numerous small hemorrhages often are seen in the retina; subconjunctival hemorrhage may also occur.

Bleeding after Trauma

Excessive bleeding often follows tooth extractions, tonsillectomy, or other operations or injuries and may first suggest the diagnosis of ITP. In contrast to the hereditary coagulation disorders, such traumatic bleeding is seldom voluminous or rapid. Slow persistent oozing may occur after trivial cuts, razor nicks, and scratches. Delayed bleeding and spontaneous hemarthrosis, which are characteristic of the hereditary coagulation disorders, are extremely rare in ITP.

Laboratory Findings

Blood

The mean platelet count of patients at the time of diagnosis of ITP is 25 to 30,000 and severe thrombocytopenia (<10,000) is frequently seen.¹¹⁷, ¹¹⁸ Abnormalities in platelet size and morphologic appearance are common. The platelets often are abnormally large (3 to 4 µm in diameter) and reveal more than normal variation in size and shape. Abnormally small platelets and platelet fragments (“microparticles”) also are evident and may represent the equivalent of microspherocytes and schistocytes.¹¹⁹, ¹²⁰, ¹²¹, ¹²² Although megakaryocyte fragments may be apparent in routine blood smears, quantitative studies reveal subnormal numbers of these fragments.¹²³

Estimates of mean platelet volume (MPV) and the extent of platelet size heterogeneity (platelet distribution width) by means of automated particle counters may, if present, provide useful information in the evaluation of patients with ITP.¹²⁴ The presence of numerous megathrombocytes results in high MPV values.¹²⁵ Platelet distribution width also is increased, presumably reflecting an abnormal degree of platelet anisocytosis.¹²⁶ The exact mechanism underlying such megathrombocytosis is still uncertain, but it may be the result of accelerated platelet production in response to platelet destruction. When MPV is increased in patients with ITP, it is typically inversely correlated in a nonlinear manner with the platelet count. In contrast, low MPV values have been reported in association with big spleen syndromes¹²⁶ and some myeloproliferative disorders, after chemotherapy with cytotoxic drugs, and in patients with septic thrombocytopenia.¹²⁷

Significant abnormalities in the other blood counts should prompt a thorough evaluation for other causes of thrombocytopenia as these findings are unusual in ITP. Anemia, if present, is proportional to the extent of blood loss and is usually normocytic. If bleeding has been severe and long-standing, iron deficiency anemia may occur. Occasionally, recent severe hemorrhage may produce reticulocytosis and moderate macrocytosis. Antiplatelet antibodies in patients with ITP do not usually cross-react with erythrocytes, although erythrocyte fragmentation, presumably the result of weak complement activation, may occur.¹²¹ Patients may also have a positive direct Coombs test and autoimmune hemolytic anemia; the combination is known as Evans syndrome.¹²⁸, ¹²⁹

The total leukocyte count and the differential count usually are normal. Eosinophilia has been noted, particularly in children, but this finding is by no means consistent. Lymphocytosis with abnormal cells resembling those characteristic of infectious mononucleosis also has been reported.¹³⁰, ¹³¹

Tests of hemostasis and blood coagulation reveal only changes attributable to thrombocytopenia, such as a prolonged bleeding time. The results of tests of blood coagulation, including prothrombin time, partial thromboplastin time, and fibrinogen, are normal in patients with uncomplicated thrombocytopenia. Slight increases in the levels of fibrinogen degradation products have been demonstrated in the plasma of some patients with ITP.¹³² Plasma levels of glycocalicin, a portion of platelet membrane gpIb, may be high in patients with ITP and other forms of platelet destruction. As noted previously, concentrations of thrombopoietin are not significantly increased in ITP.

Bone Marrow

ITP causes no characteristic bone marrow changes; therefore, bone marrow examination should not be routine. The American Society of Hematology (ASH) guidelines for management of ITP recommend against bone marrow biopsy in both children and adults with history, physical exam, CBC, and peripheral smear typical of ITP.⁵ However, bone marrow aspiration may be helpful in the differential diagnosis of ITP in patients who have atypical findings that may suggest some of these other etiologies.

In patients with ITP, alterations in the bone marrow are usually limited to the megakaryocytes, although normoblastic hyperplasia may develop as a result of blood loss. The leukocytes are essentially normal with the exception of occasional eosinophilia.¹³³

Megakaryocytes usually are increased in size¹³⁴ and are increased or normal in number,¹³⁵, ¹³⁶ the numbers correlating roughly with the MPV. Morphologic abnormalities of these giant cells are present in most patients with ITP. “Smooth” forms with single nuclei, scanty cytoplasm, and relatively few granules are common. Presumably, they represent the results of markedly
accelerated platelet production and the presence of many young forms.¹³⁷, ¹³⁸ The changes just summarized are similar to those found in most forms of thrombocytopenia caused by accelerated platelet destruction and are not characteristic or diagnostic of ITP.

Antiplatelet Antibodies

Primary ITP is a diagnosis of exclusion and relies on clinical impression. A number of different types of antiplatelet antibody tests have been developed and reported through the years.³⁶, ¹³⁹, ¹⁴⁰, ¹⁴¹, ¹⁴², ¹⁴³, ¹⁴⁴, ¹⁴⁵, ¹⁴⁶ Most of these tests were quite cumbersome and therefore never became available for routine testing. These tests measured different types of Ig, including serum antiplatelet antibodies, platelet-associated surface Ig, or total platelet Ig and are now generally regarded as unreliable.¹⁴⁷ The platelet Ig is released, along with other α-granule proteins, such as platelet factor 4 and β-thromboglobulin, during platelet activation and secretion. It is presumed that some of these released proteins bind to the platelet surface. These observations make it difficult to use either the platelet-associated IgG assays or the total platelet Ig assays for the diagnosis of ITP.¹⁴⁸

In a more recent generation of antiplatelet antibody assays, monoclonal antibodies for the specific platelet membrane glycoproteins that are implicated in ITP are used in antigen-capture-type assays (also called glycoprotein immobilization assays).¹⁴⁹ Studies have identified a specificity of 78% to 93%.¹⁵⁰, ¹⁵¹ However, the sensitivity (49% to 66%) is not high enough to exclude the presence of ITP if the test is negative.⁶⁸, ¹⁵¹, ¹⁵² None of these tests is in routine clinical use, and experts disagree on their role in the diagnosis of ITP. Future direction might include the use of flow cytometry in the diagnosis and follow-up of autoimmune thrombocytopenia.¹⁵³

Differential Diagnosis

The initial step in the evaluation of a thrombocytopenic patient is inspection of the peripheral blood smear to confirm the decreased platelet count.¹⁵⁴ Thrombocytopenia may be produced artifactually by clumping of the platelets in the blood sample caused by ethylenediaminetetra-acetic acid-associated platelet agglutinins (pseudothrombocytopenia),¹⁵⁵, ¹⁵⁶ or the platelets may be unavailable for counting because they are bound in rosette formation to the surface of white blood cells in the venous blood sample (“platelet satellitism”).¹⁵⁷, ¹⁵⁸

The diagnosis of ITP is usually a diagnosis of exclusion based on a demonstration of peripheral thrombocytopenia, with a history, physical examination, and complete blood count that do not suggest another cause for the thrombocytopenia. Splenomegaly suggests that the thrombocytopenia may be a result of hypersplenism related to the presence of a separate underlying disease associated with splenic enlargement.

The initial manifestations of acute leukemia, myelodysplastic syndrome,¹⁵⁹, ¹⁶⁰ myelophthisic processes, and aplastic anemia may mimic ITP. These other types of underlying hematologic disorders are suggested by anemia out of proportion to blood loss and by changes in the leukocytes not attributable to either hemorrhage or complicating infection. Although not routinely recommended for the diagnosis of ITP, bone marrow aspiration may be helpful in the differential diagnosis of ITP in patients who have atypical findings that may suggest some of these other etiologies.

The presence of schistocytes in the blood smear suggests that thrombocytopenia may be associated with a microangiopathic process (see Chapter 48). In thrombotic thrombocytopenic purpura (TTP) or hemolytic-uremic syndrome (HUS), thrombocytopenia is associated with laboratory manifestations of hemolysis, including elevated levels of lactate dehydrogenase and indirect bilirubin.¹⁶¹ Patients may also have transient, multifocal neurologic signs or symptoms, renal insufficiency, and/or fever in TTP, or renal insufficiency alone in HUS. After a diagnosis of ITP, the next essential step is to distinguish between primary ITP purpura and secondary forms of ITP purpura, such as human immunodeficiency virus, hepatitis C, or Helicobacter pylori infections; collagen vascular diseases such as systemic lupus erythematosus (SLE); lymphoproliferative disorders such as chronic lymphocytic leukemia; and drug ingestion. The ASH guidelines recommend considering testing all adults with a new diagnosis of ITP for hepatitis C and HIV as treatment plans differ in these settings.⁵ The importance of careful inquiry regarding drug ingestion or exposure to toxic substances cannot be overemphasized, because thrombocytopenia attributable to drugs or toxins often is indistinguishable from ITP.¹⁶² The development of thrombocytopenia in an adult, in particular, should arouse suspicion of a pharmacologic etiology, because many of the drugs associated with thrombocytopenia are used more often by adults than by children. It is also essential to eliminate the possibility that the thrombocytopenia is secondary to heparin administration. ITP may be produced by heparin administered in any dose and by any route of administration, including heparin-bonded catheters.¹⁶³ Finally, the antiphospholipid antibody syndrome is a disorder that may be associated with thrombocytopenia. In the usual case, this presentation may be associated with thromboembolic manifestations, anticardiolipin antibodies, and coagulation inhibitors of the lupus type (see Chapter 54).

Treatment of Primary Immune Thrombocytopenia

Many of the treatment recommendations for ITP are based on expert opinion rather than high-level evidence from randomized controlled trials. The ASH has published a practice guideline on ITP.⁵, ¹⁶⁴, ¹⁶⁵ The reader is referred to this practice guideline for specific questions regarding the treatment of patients with ITP.

Children

Childhood ITP is usually benign and self-limited. Therefore, treatment is not required for most patients. Treatment is reserved for patients with severe thrombocytopenia (<20,000/µl) and bleeding, or patients who remain thrombocytopenic for >6 months, that is, those with chronic ITP (Table 47.4).¹⁶⁶ Treatment guidelines recommend observation for all children with no or mild bleeding, regardless of platelet count.⁵ Although the greatest
fear in the acute form is intracranial hemorrhage, several large studies show that even with low platelet counts (<30,000/µl) life-threatening bleeding and intracranial hemorrhage are rare (<0.5%).¹⁶⁶, ¹⁶⁷

TABLE 47.4 RECOMMENDATIONS FOR INITIAL TREATMENT OF IMMUNE THROMBOCYTOPENIA (IVIG) PATIENTS WITH PLATELET COUNTS <20,000-30,000/µl^a

	Children	Adults
Asymptomatic	None	Steroids (preferred) or IVIG
Minor purpura Mucosal membrane bleeding that may require clinical intervention	None IVIG or steroids	Steroids (preferred) or IVIG) IVIG and/or steroids
Severe, life-threatening bleeding	Steroids and IVIG Hospitalization Consider platelet transfusion and other measures	Steroids and IVIG Hospitalization Consider platelet transfusion and other measures
^aThe current ASH guidelines recommend treatment for adults with platelet counts of <30,000/µl.

Treatment guidelines for pediatric ITP recommend first-line treatment with either a single dose of intravenous immunoglobulin (IVIG) (0.8 to 1 g/kg) or a short course of steroids.⁵ Several multicenter randomized trials have been performed in high-risk patients with acute ITP to define whether treatment is associated with a prompt increase in platelet counts. These clinical trials demonstrated that treatment with either oral prednisone or IVIG was associated with a more rapid rise in platelet count to >20,000/µl than either no therapy or treatment with anti-D.¹⁶⁸, ¹⁶⁹ Only IVIG shortened the time to reach a platelet count >50,000/µl. More recently, children with platelet counts <10,000/µl or counts of 10,000 to 29,000/µl and mucosal bleeding were studied in a prospective randomized clinical trial, and IVIG raised platelet counts faster than three corticosteroid regimens.¹⁷⁰ Therefore, IVIG should be preferentially used if a rapid rise in platelet count is desired.

Treatment of children who develop persistent ITP or are unresponsive to initial treatment is evolving. Splenectomy remains a standard given high response rates; however, many experts recommend delaying splenectomy for at least 12 months given the frequency of spontaneous remission.⁵ Rituximab treatment has been used in children and adolescents in a small study comprised of 36 patients with chronic ITP.¹⁷¹ In contrast to observations in adults, the median time to response was 1 week (range 1 to 7 weeks), and the overall response rate was 31%. Outcome was not associated with age, prior pharmacologic response, prior splenectomy, duration of disease, screening platelet count, refractoriness, or IgM reduction. Therefore, rituximab may be considered as an alternative to splenectomy in children with chronic ITP or in patients with persistent bleeding despite first-line treatments.⁵ High-dose dexamethasone (0.6 mg/kg/day) may also be considered in patients who are unresponsive to initial therapy based on small studies showing a 25% response rate in refractory patients.¹⁷²

FIGURE 47.5. Therapy of adult immune thrombocytopenia (ITP). (1) Minimal emergency therapy includes intravenous (IV) methylprednisolone and intravenous immunoglobulin (IVIG). Intravenous anti-D and platelet transfusions may be given as needed. All three modalities given prior to transfusions may help preserve platelet longevity in the circulation and repeated or continuous platelet transfusions may be required in urgent situations. (2) Initial treatment of ITP typically consists of steroids (prednisone or dexamethasone) with the goal of attaining a platelet count of >30,000/µl and cessation of bleeding. IVIG or anti-D may be used if steroids are contraindicated or the patient has persistent thrombocytopenia despite steroids. (3) Thrombocytopenia recurs in most adults as corticosteroids are tapered. Treatment options for refractory ITP include splenectomy, thrombopoietin mimetics, and rituximab. (Modified from Cines and Bussel, reference 459.)

Adults

Patients with chronic ITP may have mild thrombocytopenia that can be followed without treatment. The incidence of bleeding is correlated with the platelet count; therefore, patients with platelet counts >50,000/µl rarely have spontaneous bleeding and may require treatment only if extensive operative procedures are planned. Patients with platelet counts <20,000 to 30,000/µl or significant mucosal membrane bleeding with platelet counts <50,000/µl are usually treated⁵ (Table 47.4 and Fig. 47.5).

No prospective studies on long-term prognosis of ITP after treatment are reported. However, it has been reported that most adult patients have a good response to treatment (without necessarily returning to normal platelet counts) and have no excess mortality when compared to the general population.¹⁷³ A small group of patients who had severe thrombocytopenia after 2 years of primary and secondary therapies had a mortality risk of 4.2 (95% confidence interval, 1.7 to 10.0) resulting from both bleeding and infectious complications related to therapy.

ITP is uncommon in elderly patients; only 30% of patients in reported series are >45 years of age.¹³³, ¹⁷⁴ However, these patients may be more refractory to therapy¹¹⁵ and appear to have a higher incidence of hemorrhagic complications than younger patients.¹² Guthrie and colleagues reported a 52% incidence of life-threatening or fatal bleeding in their series of 40 patients
older than age 45 years.¹¹⁵ This risk of fatal bleeding in patients with platelet counts that are chronically <30,000/µl is estimated at 0.4% per year for patients <40 years of age and 13.0% per year for patients >60 years of age.¹⁷⁵

First-line Treatment

Initial treatment for adult patients with newly diagnosed ITP is typically steroids. Other options that increase the platelet count rapidly include IVIG and anti-D. The most recent ASH guideline recommends use of these alternative first-line therapies in patients with contraindications to steroids.⁵ Patients with life-threatening bleeding may require parenteral glucocorticoids and IVIG followed by platelet transfusions,¹⁷⁶ plasmapheresis,¹⁷⁷ or even emergency splenectomy as first-line treatment. Standard ITP treatments and their dosing schedules are shown in Table 47.5.

Steroids

Steroids are the conventional first-line therapy for adult ITP. Dameshek first reported his experience with prednisone therapy in 1958,¹⁷⁸ when 30 consecutive patients with acute (N = 11) or chronic (N = 19) ITP were treated with 20 to 150 mg/day of prednisone. Twenty-two of the patients demonstrated an increase in platelet count to normal after an average interval of 22 days and then were maintained on 2.5 to 15.0 mg/day of prednisone. In eight patients, prednisone was discontinued without relapse.

Numerous retrospective studies of steroid treatment in both children and adults with ITP have been reported.¹³, ¹⁴, ⁹⁷, ⁹⁸, ¹⁰⁴, ¹⁰⁸, ¹⁰⁹, ¹¹⁰, ¹³³, ¹⁷⁴, ¹⁷⁹, ¹⁸⁰, ¹⁸¹, ¹⁸² The criteria for inclusion of patients in the individual studies and the criteria for response to treatment vary significantly among the reports; therefore, it is difficult to combine the data accurately. However, some useful observations can be made. Complete (CR) and partial (>50,000/µl) responses (PR) in patients treated with prednisone (usually, 1 mg/kg/day as starting dose) average 65% to 85%, but sustained responses after discontinuation of the drug occur in only 25% or less of patients.¹⁸³, ¹⁸⁴ Platelet counts usually increase within 1 week in responding patients and have usually reached peak values by 2 to 4 weeks. Patients who have not had any response by 4 weeks are unlikely to respond to prednisone and therefore should be considered for other forms of treatment. No pretreatment patient characteristics have predicted a patient’s response to steroids.

TABLE 47.5 THERAPEUTIC AGENTS AND THEIR DOSING SCHEDULES

Agent	Dose and Schedule
Anti-D immunoglobulin	50-75 µg/kg IV, repeated at 3-week intervals as indicated
Cyclophosphamide	150 mg daily for up to 8 wk
Colchicine	200 mg daily for up to 4 wk
Dexamethasone	40 mg daily for 4 d, repeated every 14 d for 4 cycles
Danazol	400 mg twice daily for 1 mo or longer
Eltrombopag	50 mg daily. Must be continued indefinitely.
IVIG	1 g/kg IV for 1-2 d, repeated every 2 to 4 wk as indicated
Prednisone	1 mg/kg daily for up to 28 d, then taper to lowest dose possible
Rituximab	375 mg/m² IV weekly for 4 doses
Romiplostim	1-10 mcg/kg subcutaneous injection weekly. Start at 1 mcg/kg and titrate based on platelet count. Must be continued indefinitely.
Vincristine	2 mg at 5- to 7-day intervals for 2 or more doses
Vinblastin	7.5 mg at 5- to 7-day intervals for 3 or more doses
Modified from Narang et al.⁴⁶⁰

Several prospective randomized steroid treatment trials in patients with ITP have been reported. Bellucci and colleagues randomized patients between low (0.25 mg/kg/day) or high (1.0 mg/kg/day) prednisone for 3 weeks, with taper and discontinuation by the end of the fourth week.¹⁸⁵ If bleeding continued, patients could be increased from low dose to high dose, or a second 4-week course could be given. CR, defined as a platelet count >100,000/µl for at least 6 months, was seen in 74% of children and 41% of adults. CRs or PRs occurred in 83% of children and 59% of adults. No significant differences were seen between low- and high-dose regimens in either age group. Mazzucconi and colleagues randomized patients between prednisone 0.5 mg/kg/day and 1.5 mg/kg/day.¹⁸⁶ The response rates in adults were not significantly different between patients treated with low- versus high-dose steroids: 30% and 34% CR, respectively. In children, however, the rates were 64% for low-dose versus 81% for high-dose prednisone. Therefore, there is evidence to support the use of lower doses of steroids in adults than have conventionally been used at the beginning of treatment. However, based on the opinion of an expert panel convened by the ASH, high-dose prednisone (1 mg/kg/day) is recommended as appropriate initial treatment in ITP patients with platelets <30,000/µl, including asymptomatic patients.⁵, ¹⁶⁴, ¹⁶⁵

Another prospective randomized, controlled trial compared prednisone (1 mg/kg/day) to IVIG (400 mg/kg/day for 4 days), or to both in a small number of patients.¹⁸⁷ A platelet count >50,000/µl was achieved in 82%, 54%, and 92% of patients, respectively. The median times to peak platelet counts were 8.5, 7.0, and 7.0 days. These authors concluded that there was no advantage for IVIG over conventional corticosteroid treatment. In another recent randomized control trial, platelet counts >50,000/µl were achieved more quickly in patients receiving IVIG than in those receiving intravenous methylprednisolone, but long-term outcomes were the same.¹⁸⁸

There are also several studies that have evaluated the role of high-dose dexamethasone in the initial treatment of ITP. This protocol was initially developed because of concern regarding long-term side effects of daily prednisone use. Initial trials showed that using 40 mg of dexamethasone daily for 4 consecutive days results in an initial response in 85% of patients.¹⁸⁹, ¹⁹⁰ Additional work has shown that repeating dexamethasone bursts may provide long-term disease control in some patients. Mazzucconi and colleagues treated newly diagnosed patients with ITP with dexamethasone 40 mg daily for 4 days, repeating these cycles every 14 days for 4 cycles.¹⁹¹ A total of 85% of patients responded with 65% achieving a CR. Relapse-free survival at 15 months was 81% indicating that many of these patients may be long-term responders.

There are conflicting opinions regarding the choice of prednisone versus dexamethasone for first-line treatment of ITP. The writers of the 2011 ASH guidelines favor longer-term prednisone over dexamethasone⁵ but other experts in the field think that either is reasonable.¹⁹²

Only gold members can continue reading. Log In or Register to continue