Association of Platelet Count and Mean Platelet Volume with Severity of Bleeding Events in Patients with Immune Thrombocytopenia

Introduction: Immune thrombocytopenia (ITP) is an autoimmune disorder of antibody-mediated destruction of platelets. In clinical practice, severe thrombocytopenia with a platelet count < 20 x 109/L is associated with significant bleeding events. Immature platelets, which are more prevalent in ITP, are more reactive to thrombin and better at promoting hemostasis, suggesting a lower platelet threshold for severe bleeding in ITP compared to other causes of thrombocytopenia. We aimed to define the platelet thresholds and mean platelet volume (MPV) associated with different grades of bleeding severity in ITP. Methods: This retrospective chart review included adult patients who were admitted in the Brown University Health system with the diagnosis of ITP from January 2016 to June 2023. Outcomes included severity of bleeding events, platelet count, MPV, and treatment patterns. Exclusion criteria included any patients with concern for secondary causes of thrombocytopenia (reactive, infectious, drug-induced) and those with confounding factors like underlying hematologic or oncologic malignancy, chronic liver disease, HIV, connective tissue/autoimmune disease, or a separate primary bleeding disorder. Bleeding event severity was defined by the World Health Organization (WHO) Bleeding Scale: Grade 0 (G0) as no bleeding, G1 as petechial bleeding, G2 as mild blood loss, G3 as gross blood loss requiring transfusion, and G4 as debilitating/fatal blood loss including intracranial and retinal bleeds. Results: Out of 129 eligible patient encounters, 35% (n=45) had G0, 49% (n=64) had G1-2, and 15% (n=20) had G3-4 bleeding events. The median age was 70 years, with 50% being female. The median MPV was 10.6 for G0, 10.6 for G1-2, and 10.1 for G3-4 events. The median platelet count was 39 x 109/L (interquartile range [IQR] 10-81) for G0, 9 x 109/L (IQR 5-22) for G1-2, and 14 x 109/L (IQR 10-45) for G3-4 events. The mean Charlson Comorbidity Index (CCI) was significantly different between G1-2 (4.0) and G3-4 (5.0) groups (p = 0.03). Following first-line treatment (steroids and/or intravenous immunoglobulin), median platelets were 72 x 109/L (IQR 30-156) for G0, 102 x 109/L (IQR 31-147) for G1-2, and 91 x 109/L (IQR 55-119) for G3-4 bleeding events. There was a significant difference in mean platelet counts between G3-4 bleeding events and G0 events (p = 0.03) but this became non-significant after first-line treatment. Conclusion: In patients with ITP, G3-4 bleeding events were associated with significantly lower platelet counts compared to G0 events, indicating a clinically relevant platelet threshold. Patients with G3-4 bleeding events also had a higher mean CCI risk score than those with G1-2 events, thus underscoring another factor aside from platelet count in determining bleeding risk. G3-4 bleeding events were overall less frequent compared to the more common minor G1-2 events. However, there was no significant association between MPV and bleeding event severity. Finally, this study reinforced that bleeding events of any grade (G1-4) in ITP occurred at lower platelet thresholds than those expected in other causes of thrombocytopenia.