PF329 INTRACRANIAL HEMORRHAGE IN IMMUNE THROMBOCYTOPENIA: A SINGLE‐CENTRED STUDY OF 36 PATIENTS

Background: Intracranial hemorrhage (ICH) is a devastating complication of immune thrombocytopenia (ITP). However, given the low incidence of ICH and the adverse effects of prophylactic management, individualized prevention based on risk assessment is necessary. Several case‐control studies have been conducted to define the risk factors of ICH in either children or adults with primary ITP, and a preliminary algorithm for identification of high‐risk patients based on combinations of several risk factors has been developed (Blood, 2009), but a detailed conclusion remains uncertain. Aims: To further determine the predisposing features associated with ICH in primary ITP in a case‐control retrospective study and to develop preliminary algorithms for identification of high‐risk patients. Methods: A total of 36 patients suffering from primary ITP complicated with ICH were identified retrospectively from 1990 to the present. For each case, we enrolled the next 2 hospitalized patients with ITP who presented with a platelet count less than 50,000/μL as controls. The collected data included patient characteristics, events preceding ICH, platelet counts, ITP durations, bleeding symptoms and the Khellaf bleeding score (Haematologica, 2005). The utility of some algorithms based on combinations of the identified predisposing features (hematuria and Khellaf bleeding score ≥ 6) and vital characteristics (age >60 years and PLT count <10 000/μL) of the patients were evaluated for ICH risk assessment. The data were analyzed with the 2‐tailed Fisher's exact test and nonparametric Mann‐Whitney test. Results: In total, 36 cases and 72 controls were included. The comparison between the cases and controls was illustrated. Head trauma and ITP diagnosed within 3 months were associated with ICH (P = 0.035 and 0.001, respectively). Notably, a significant difference was observed between the enrollment ages of the two groups (P = 0.038), which might be partially ascribed to the higher proportion of patients under 18 years of age among the controls (P = 0.042). The 12 cases with ICH as the presenting feature were excluded from the bleeding symptoms and bleeding score analyses. Urinary bleeding was associated with ICH (P = 0.001), as was well documented previously (Blood, 2009). No significant differences were found in visceral bleeding, but visceral bleeding at more than 1 location was associated with ICH (P = 0.005). We calculated the highest Khellaf bleeding score in the medical records for each patient, and a bleeding score ≥ 6 was identified as a predisposing feature. Combining the 2 factors (age over 60 years and a Khellaf bleeding score ≥ 6) identified 19 (79%) of the 24 patients with ICH, whereas the combination of an age over 60 years and hematuria excluded 39 (81%) of the 48 controls. Fifteen (63%) of the cases presented with at least 2 of the 4 items (listed in the Methods) versus 18 (37%) of the controls. Summary/Conclusion: Head trauma, hematuria, visceral bleeding at more than 1 location and a Khellaf bleeding score ≥ 6 were identified as predisposing features of ICH in primary ITP. Combinations of several predisposing factors that might be applied to management algorithms for ICH prevention were also presented.