An international group of ITP defined the platelet count to be less than 100 × 10⁹/L as the threshold for diagnosis [14]. A nationwide study in Japan analyzed 7774 patients diagnosed according to this threshold and reported that the mean platelet count was 22.0 × 10⁹/L in the total ITP patients (23.7 in females and 19.3 in males), 12.8 × 10⁹/L in patients less than 15 years old, 27.2 × 10⁹/L at 15–49 years old, and 21.1 × 10⁹/L in patients more than 50 years old [3]. Sixty-two percent of child patients had a platelet count of less than 10 × 10⁹/L. This percent was significantly higher than that (41.2%) in adult patients. Both an international multicenter (Intercontinental Cooperative ITP Study Group, ICIS) and UK studies had the similar results to a Japanese study although they included patients with platelet counts greater than 100 × 10⁹/L [2, 15, 16].

The most common bleeding symptom of ITP is purpura [3, 16]. Purpura and epistaxis occur more frequently in children than in adults (Table 1). Gastrointestinal bleeding is much less common: 1.11 and 3.9% of entire ITP patients in France and Japan, respectively [3, 4]. There was no significant difference in the frequency of gastrointestinal bleeding between child and adult ITP [3]. However, among adult ITP, there was a linear increasing relation between age and gastrointestinal bleeding [4]. Intracranial hemorrhage (ICH) is the most devastating bleeding event, and its incidence is 0.37–0.6% [3, 4]. Adult patients have more frequent ICH than children, and the older patients, especially over 60 years old, have an exponential increase in ICH. The role of risk factors for adult ICH including hypertension and diabetes mellitus in the onset of ICH remains to be determined. In child ICH, an epidemiologic study in the United States with a nationwide survey of childhood ITP with ICH showed that the estimated incidence of ICH was 0.19–0.78% and that platelet counts were less than 20 × 10⁹/L in 90% and less than 10 × 10⁹/L in 75% of children with ICH [17]. The ICIS study also reported the similar relationship between platelet counts and ICH in both child and adult patients [16]. However, the US study has revealed that severe thrombocytopenia was not sufficient for child ICH, but head trauma and bleeding symptoms beyond purpura, especially hematuria, were linked to ICH [17]. Such epidemiologic findings may be useful for selection of ITP patients with high risk of ICH.

Until recently, thrombosis was not considered as a part of ITP manifestations. Recent several studies suggest that ITP may have paradoxically an increased risk for thromboembolism. A US study with 3131 ITP patients from a database of US health insurance reported that the adjusted incidence rate ratio (IRR) of any vascular, venous, and arterial thromboembolism as compared with the age- and gender-matched reference cohort was 1.70 (95% CI, 1.41–2.05), 2.89 (95% CI, 1.33–6.29), and 1.58 (95% CI, 1.29–1.94), respectively [18]. A Danish population-based study with 391 adult patients from the Danish National Patients Registry showed that the IRR of venous thrombosis as compared with the reference cohort was 2.65 (95% CI, 1.27–5.50) [19]. Another Danish population study with 379 patients showed that the IRR of arterial thrombosis was 1.32 (95% CI, 0.88–1.98), indicating no significant difference [20]. A UK study with 1070 patients from GPRD also reported no significant difference in the IRR of arterial thrombosis, but significantly increased IRR of venous thrombosis [21]. A recent Scandinavian population-based cohort study with 1821 adult patients found that the occurrence of both arterial and venous thromboembolism increased with age and increasing comorbidity burden [22]. Splenectomized ITP patients had the increased risk of both arterial and venous thrombosis compared with non-splenectomized patients with hazard ratios of 3.2 (95% CI, 1.2–8.6) and 4.1 (95% CI, 1.1–15.7), respectively, with an overall hazard ratio of thrombosis in splenectomized of 3.5 (95% CI, 1.6–7.6) [23]. For ages >60 years, more than two classical risk factors for thrombosis at diagnosis and steroid use were independently associated with an increased risk of thrombosis. No independent investigations have explored the thrombotic risk associated with thrombopoietin receptor agonists although several clinical trials showed a certain rate of thrombotic events in ITP patients who received thrombopoietin receptor agonists [24]. This rate did not increase over time. Taken together, there seems to be a slight increase in venous thrombosis in chronic ITP, except for splenectomized and elderly patients [23, 24]. Although such thrombophilia may not be worthy to demand clinical intervention [23], the stratification of a high-risk group would be of value in the establishment of thrombosis-associated subgroup of ITP and choice of treatment.

The treatment strategy differs between child and adult ITP patients. In child ITP, the most common prescription is intravenous immunoglobulin [3, 16]. This is likely because high-dose immunoglobulin therapy results in a more rapid increase in platelet counts than other therapies when it is administered for management of bleeding symptoms at the acute onset of child ITP. In adult ITP, the most common prescription is corticosteroid that is recommended as a first-line therapy for adult ITP by the international consensus report and the American Society of Hematology guideline [8, 25]. The eradication of Helicobacter pylori in adult patients is performed more frequently in Japan than in western countries [3]. The rate of splenectomy was 29% of adult patients in the clinical practice of two decades ago [26], and it has been dramatically decreased to only 6% in recent years [2, 27], because patients are increasingly reluctant to undergo splenectomy and new drugs including anti-CD20 monoclonal antibody and thrombopoietin receptor agonists have been developed [28]. The frequency of each treatment will be altered over time, and an epidemiologic approach to thrombopoietin receptor agonists will have a considerable impact on the management and the burden of this disease.