Twenty years' experience of immune thrombocytopenia and intracranial haemorrhage in the Nordic countries—A NOPHO study

The Nordic Society of Paediatric Haematology and Oncology (NOPHO) Platelet working group represents Denmark, Finland, Iceland, Norway and Sweden. We here present retrospectively collected data on nine children from seven departments in the NOPHO countries with immune thrombocytopenia (ITP) and intracranial haemorrhage (ICH) diagnosed between 1998 and 2018. The children were between 2 and 13 years old when diagnosed with ITP. They presented with classical symptoms and platelet count (PLT) <20 × 109/L. ICH occurred between 5 and 1202 days after diagnosis. Four of nine children survived. Table 1 describes bleeding, blood sample results and treatments when ICH occurred. ↓ bleeding PLT 7 Unwell Vomiting IVIG Platelet transfusion ↓ bleeding PLT 115 and 70, 1 and 8 h after transfusion Diagnosed with SLE. Later treated with prednisolone and mycophenolate. PLT during follow-up >100 × 109/L, no bleeding symptoms. Attends a special school, difficulties in mathematics (present before ICH) Follow-up 7 years, transferred to adult care. No additional bleeding Romiplostim 1, 2 and 4 mg/kg PRBC 8, 3 and 1 day before ICH Gestagen Menorrhagia Headache One-year follow-up with no neurological sequelae. 2 months 22 days after ICH PLT 28 × 109/L. There after increasing and normal (110 × 109/L) Follow-up 5 years, transferred to adult care. No additional bleeding Normalisation of vison within 2 months. Treated 2 weeks, 1, and 2 months after ICH with dexamethasone 40 mg/m2/day for 4 days. After the first treatment PLT 112 × 109/L. PLT > 50 × 109/L for 14 months. Later relapse of severe thrombocytopenia and started eltrombopag Follow-up 7 years, transferred to adult care. No additional bleeding At diagnosis, six of nine children received medical treatment with intravenous immunoglobulin (IVIG) without clear effect in all. Bleeding decreased and PLT increased in patient 2. For patient 3, 4, 6, 7 and 9, the treatments had limited response. For details about symptoms, treatment and course of disease before ICH, see the Table S1. Prior to ICH patient 1 had unchanged, limited bleeding symptoms. Patient 2 had grade 3 bleeding 6 days before ICH without receiving treatment. Patient 3 was unwell with grade 3 bleeding 4 days before ICH and received IVIG and pooled red blood cells (PRBC). Patient 4 had repeated epistaxis on the day ICH occurred. Patient 5 had grade 2 bleeding 7 days prior to ICH. She received IVIG and platelet transfusion. Patient 6 had mucosal bleeding from four sites the last 2 weeks before ICH and received her regular treatment with romiplostim and PRBC. Patient 7 had bruising, epistaxis and microscopic haematuria 3 days before ICH. Patient 8 experienced increased petechia, epistaxis and menorrhagia in the last 21 days before ICH. She had PRBC three times in the last 2 weeks before ICH. Patient 9 had oral bleeding, epistaxis, tongue haematoma and subconjunctival haemorrhage during the last 7 days before ICH. Conclusively, seven of nine children with spontaneous ICH experienced significant bleeding the days before ICH. Six of them had mucosal bleeding (epistaxis, haematuria, menorrhagia). Trauma is a well-known risk factor for ICH in ITP and patient 1 suffered a high-energy head trauma. Patient 2 had a coup contrecoup injury on CT scan, but no history of trauma. None of the children reported the use of platelet-inhibiting medication. PLT at ICH diagnosis was <10 × 109/L in eight of nine children. In the last days before ICH, haemoglobin decreased median 3 g/dl (range 1.1–6.3 g/dl) in six of nine children. Patients 2, 3, 4, 5 and 8 felt generally unwell, patient 9 felt tired. After diagnosis of ICH, eight of nine children received platelet transfusions, prednisolone and IVIG after ICH. Patients 3 and 6 underwent emergency craniotomy. Patients 3, 5 and 7 were splenectomized. Patient 2 received PRBC, patient 3 had IVIG the day before ICH. Patient 6 received tranexamic acid and coagulation factors. Patient 7 received prednisolone (2 mg/kg) daily for 2 weeks, thereafter weekly rituximab, and romiplostim for 4 weeks. Patient 9 received repeated platelet transfusions but progressed, and neurosurgical intervention was not possible. In total, four children died. Three of the five children, who survived ICH, received further treatment. Patient 1 received IVIG and prednisolone with limited response. Patient 5 was splenectomized day 5 after ICH followed by rituximab weekly (3 weeks). Patient 8 continued high-dose dexamethasone (40 mg/m2), 4 days, every fourth week, four times. The PLT normalised after the second treatment. Patients 1 and 5 have mild neurological sequelae. Patients 4, 7 and 8 have fully recovered. The overall limitation is that the case series is retrospective and based on patients known by the physicians in the NOPHO platelet working group. As a part of the national structured health care system in all the Nordic countries, certain conditions are only cared for in cooperation with experts from the university hospitals. ICH in a child with ITP is quite rare, and it is therefore highly unlikely, the experts affiliated with the NOPHO platelet working group would not have been consulted. However, we did not consult our national health registries to ensure we captured all children with ITP and ICH which may lead to an underestimation of the risk for ICH. The number of new ITP patients each year, that is, 4.8/100 000 in a total population of 120 345 800 children in the Nordic countries from 1998 to 2018. Thus, the estimated frequency of ICH in the Nordic countries among children with ITP is 0.2% from 1998 to 2018. Besides, the total number of ICH cases may be too low, the number of ITP patients at risk during the 20 years, may be too high an estimate, since few remain chronic ITP patients. On the other hand, the frequency does compare to the earlier reported frequencies of ICH in ITP children are between 0.1% and 0.5% over 10–20 years.1-3 In eight children, prior to ICH, we found increased bleeding, subjectively feeling unwell or decreasing haemoglobin. These observations may constitute prodromal symptom of severe bleeding as well as other bleeding than petechia and ecchymosis.1 However, the children may earlier have experienced similar bleeding symptoms not followed by ICH, since we do not have systematic clinical descriptions and laboratory data on the children from all encounters. Today, ITP in children is handled with a watch-and-wait approach. In different studies, the reported treatment frequencies were 57% (1998–2000),4 and as low as only 16% (2009).2 Covering the same timeframe, severe bleeding was reported unchanged between 3% and 8% by Grainger et al.,2 but later much higher by Neunert et al.3 They did note, that the definition of severe bleeding was not consistent across studies, but extensive mucosal bleeding was present in most studies. All the children reported here, received several medical treatments with increased bleeding severity, which is in concordance with the literature.5 In conclusion, ICH in children with ITP is rare with an estimated 0.2% frequency per year in the Nordic countries between 1998 and 2018. Despite a decrease in overall treatment, children with bleeding symptoms received significant medical treatment. The combination of malaise, increased bleeding and haemoglobin decline may indicate severe bleeding. For your contributions (in alphabetic order): H Hasle (Aarhus, Denmark), O G Jónsson (Reykjavik, Iceland), M Koskenvuo (Helsinki, Finland), T Mikkelsen (Aarhus, Denmark), A Mikko (Tampere, Finland), S Rosthøj (Aalborg, Denmark), S Ryhänen (Helsinki, Finland) and B Zeller (Oslo, Norway). All listed authors have no conflict of interest. Table S1 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.