Severe bleeding events in adults and children with primary immune thrombocytopenia: a systematic review

BACKGROUND: The burden of severe bleeding in adults and children with immune thrombocytopenia (ITP) has not been established. OBJECTIVES: To describe the frequency and severity of bleeding events in patients with ITP, and the methods used to measure bleeding in ITP studies. PATIENTS/METHODS: We performed a systematic review of all prospective ITP studies that enrolled 20 or more patients. Two reviewers searched Medline, Embase, CINAHL and the Cochrane registry up to May 2014. Overall weighted proportions were estimated using a random effects model. Measurement properties of bleeding assessment tools were evaluated. RESULTS: We identified 118 studies that reported bleeding (n = 10 908 patients). Weighted proportions for intracerebral hemorrhage (ICH) were 1.4% for adults (95% confidence interval [CI], 0.9-2.1%) and 0.4% for children (95% CI, 0.2-0.7%; P < 0.01), most of whom had chronic ITP. The weighted proportion for severe (non-ICH) bleeding was 9.6% for adults (95% CI, 4.1-17.1%) and 20.2% for children (95% CI, 10.0-32.9%; P < 0.01) with newly-diagnosed or chronic ITP. Methods of reporting and definitions of severe bleeding were highly variable in primary studies. Two bleeding assessment tools (Buchanan 2002 for children; Page 2007 for adults) demonstrated adequate inter-rater reliability and validity in independent assessments. CONCLUSIONS: ICH was more common in adults and tended to occur during chronic ITP; other severe bleeds were more common in children and occurred at all stages of disease. Reporting of non-ICH bleeding was variable across studies. Further attention to ITP-specific bleeding measurement in clinical trials is needed to improve standardization of this important outcome for patients.