Definition, diagnosis and treatment of immune thrombocytopenic purpura

In this issue, Arnold and colleagues document by a systematic review of controlled studies that eradication of H. pylori infection increases the platelet count in patients with immune thrombocytopenic purpura (ITP).1 This observation, together with data from a previous systematic review,2 requires that we address two questions. 1) What is the role of H. pylori infection in the pathogenesis of ITP? 2) What is the role of eradication of H. pylori infection in the management of ITP? These questions require a discussion of the definition, diagnosis, and current treatment of ITP. This discussion focuses on adults, as H. pylori infection is rare in children and two current systematic reviews have only identified studies of ITP in adults.1,2

Definition and diagnosis of immune thrombocytopenic purpura

ITP is defined as isolated thrombocytopenia with no clinically apparent associated conditions or other causes of thrombocytopenia.3 This definition provides the basis for the initial patient evaluation. No specific criteria establish the diagnosis of ITP; the diagnosis relies on the exclusion of alternative disorders, such as the examples listed in Table 1. Exclusion of recognized alternative etiologies of thrombocytopenia was the basis for the original name for ITP, idiopathic thrombocytopenic purpura.3 A current perspective has proposed the name primary immune thrombocytopenia for ITP, to distinguish ITP from identifiable alternative secondary etiologies.4

Should H. pylori infection be considered an alternative disorder, the same as HIV and hepatitis C infections? Does the diagnosis of H. pylori infection exclude the diagnosis of ITP? The current status of H. pylori infection, as indicated in Table 1, is uncertain. H. pylori infection may be an alternative, additional, or incidental unrelated disorder; all three possibilities are suggested by different studies. It is the inconsistency among current studies which has prevented broad acceptance of the association of H. pylori infection with ITP.

There are striking geographical disparities of both the frequency of H. pylori infection among patients with ITP and the frequency of platelet count responses following eradication of H. pylori infection, and these two frequencies correlate with each other.1,2 In countries with a high prevalence of H. pylori infection and high platelet count response rates following eradication, such as Japan (where most studies of H. pylori eradication in ITP have been performed) and Italy, testing for H. pylori infection has been recommended as a standard diagnostic procedure in adults with suspected ITP and eradication therapy is recommended as the initial treatment in H. pylori-infected patients.5,6 These studies suggest that thrombocytopenia associated with H. pylori infection may be an alternative disorder, similar to the thrombocytopenia associated with HIV and hepatitis C infections. In contrast, only three studies have been reported from the United States and Canada and they have all reported lower frequencies of both H. pylori infection and platelet count response rates following eradication treatment.7–9 These data suggest that H. pylori infection may be merely an incidental observation, not excluding the diagnosis of ITP and not clinically important for management. In between these alternative interpretations are studies reporting partial platelet count responses, suggesting that H. pylori infection may contribute to the thrombocytopenia in ITP but is not the sole cause.2 An explanation for the different platelet count response rates following H. pylori eradication may be the presence of different genotypes of H. pylori in different geographical regions.6 For example, most H. pylori strains in Japan express the product of the cytotoxin-associated gene A (CagA); the frequency of CagA-positive strains of H. pylori in Western countries is lower.2

Table 1.Evaluation of patients with suspected immune thrombocytopenic purpura: exclusion of alternative etiologies of thrombocytopenia.

Management of immune thrombocytopenic purpura

Ideally, treatments for ITP should be effective, safe, tolerable, and inexpensive. Detection and eradication of H. pylori infection is safe, tolerable, and inexpensive. However, because effectiveness has not been consistent, there has not been broad acceptance of detection and eradication of H. pylori infection as part of the routine evaluation and management of patients with ITP. Figure 1 presents an oversimplified algorithm of current management, with H. pylori eradication remaining outside the boxes of the routine sequence of treatments.

The goal of treatment is to achieve a safe platelet count to prevent serious bleeding; a normal platelet count may provide confidence that the ITP has resolved, but is not the primary goal. Therefore, the initial management decision is to distinguish patients who need only observation from patients who require treatment. Multiple large cohort studies have supported the use of a platelet count of 30×10/L as a practical distinguishing definition (for example 10). Although management decisions are traditionally based on the platelet count, these decisions must also consider the presence and severity of bleeding symptoms. When treatment is appropriate, corticosteroids have been the established initial agents for 50 years. Although prednisone (or prednisolone) at a daily dose of 1 mg/kg is commonly used, other regimens, such as a high dose of dexamethasone (40 mg) daily for four days may be more effective.11 The principal problem of corticosteroids, from the patients’ perspective, is that the treatment may become worse than the ITP itself, since side effects are often intolerable12 and bleeding symptoms are often minimal. Therefore, it is appropriate to use corticosteroids for only a limited time.

At each successive step in the management of ITP, treatments may be more effective but also may have greater risk. Therefore, the indications for continued treatment must be more stringent and continued treatment must be a shared decision between the physician and patient. Following failure of initial corticosteroid treatment to achieve a durable, safe platelet count, splenectomy has been the traditional second treatment for many decades. More than two-thirds of patients have durable responses.13 However, even with current laparoscopic procedures, risks of splenectomy remain, with clinically important complications in 10% of patients and death in 0.1%.13 Longterm risks following splenectomy, such as sepsis and thrombotic disorders, are a rare but important concern. Recently rituximab has become an alternative second treatment for patients with ITP, although the rate and durability of responses appear to be less than with splenectomy.14 A recent randomized clinical trial documented that a combination of four days of dexamethasone plus four weekly infusions of rituximab resulted in sustained safe platelet counts over 50×10/L at six months in 63% of patients, a significantly greater response than with dexamethasone alone (36% response).15 This combination may be an appropriate second treatment for patients with ITP. Although this clinical trial was performed in previously untreated patients,15 rituximab is not appropriate for initial treatment of patients with ITP.

Patients who require further treatment after splenectomy are uncommon, representing less than 10% of all patients who are initially diagnosed with ITP.16 Observation without treatment has been preferred for patients who are asymptomatic in spite of severe thrombocytopenia because further immunosuppressive treatments create greater risk for opportunistic infections, and one large case series documented that deaths from treatment-related infection exceeded deaths from ITP-related bleeding.10 In 2008, two thrombopoietin (TPO)-mimetic agents, romiplostim and eltrombopag, were approved by the US FDA for treatment of chronic ITP in patients who have had an insufficient response to corticosteroids, immunoglobulins, or splenectomy; romiplostim was also approved by the EMEA for a more restricted indication, splenectomized adults who are refractory to other treatments. The TPO-mimetic agents do not modify the course of the disease; they only support the platelet count and must be continually administered.17 Therefore, the most appropriate use of these agents is for patients who have severe and symptomatic thrombocytopenia following failure of treatments to achieve a durable, safe platelet count. For these patients, continuous treatment with TPO-mimetic agents is a remarkable and potentially life-saving measure. Not only can the platelet count be sustained at a safe level in approximately 85% of these patients,18 but side effects of steroids and other immunosuppressive agents are avoided. These agents may not be appropriate for treatment earlier in the sequence of ITP management because of the requirement for continuous treatment and the expense.

Figure 1.Sequence of management of patients with ITP – 2009. An oversimplified approach to management is outlined as three consecutive steps. Initial corticosteroid treatment should be limited in duration to avoid intolerable side effects. For patients who fail to respond to corticosteroid treatment, the indications for further treatment become more stringent, as the treatments involve more risk. Both splenectomy and rituximab are frequently used as second treatment options, with one following the other if there is no response. TPO-mimetic agents may become established as third treatment options, because of the risks of infection with more intensive and prolonged use of immunosuppressive agents. At this time, eradication of H. pylori infection is “outside the boxes” of conventional treatment. If it is demonstrated to be effective in certain groups of patients with suspected ITP, it could become a first option treatment because of its safety, tolerability, and low cost.

Role of H. pylori eradication in the management of immune thrombocytopenic purpura

Although current treatment options are more effective than even just one year ago,17 there remains substantial room for improvement. Current treatments may be effective, but they are all either poorly tolerated, have important risks, or are extremely expensive. Because detection and eradication of H. pylori infection are well tolerated, safe, and inexpensive, it has been proposed as a routine step in the initial evaluation and management of patients with ITP.2 This may be appropriate now in locations where the frequency of H. pylori infection is high. Although it may also be reasonable to consider H. pylori eradication in patients with ITP in all countries, enthusiasm has been limited by the low frequency of H. pylori infection and the low platelet count response rates to H. pylori eradication in countries such as the United States and Canada.

The most important results of the current reviews of H. pylori infection and ITP1,2 will be to increase awareness and stimulate further investigation of the relationship between H. pylori infection and ITP. The inconsistencies emphasized by these reviews2 provide a perfect clinical research question for international collaborations, with the goal to resolve the geographical disparities and to determine if H. pylori detection and eradication are appropriate initial measures for all patients with ITP, or only for some patients with ITP, or perhaps for all patients with ITP but only in some geographical regions. With greater understanding, H. pylori detection and eradication may then be moved “into the boxes” of appropriate management for patients with ITP.

Footnotes

• James N. George, MD, is the George Lynn Cross Research Professor of Medicine at the University of Oklahoma Health Sciences Center.
• Conflicts of interest Dr. George has served as a consultant Amgen, Inc., Thousand Oaks, CA and has received research support from Amgen for the development of romiplostim.