Initial bone marrow aspiration in childhood idiopathic thrombocytopenia: Decision analysis

Purpose: Bone marrow aspiration (BMA) is routinely performed before startin g steroid therapy in children with idiopathic thrombocytopenia, primarily t o rule out leukemia. Methods: A decision tree for the initial management of a child older than a ge 6 months, presenting with idiopathic thrombocytopenia, without blasts on the peripheral smear was construe ted. The three strategies are: 1) initia l BMA in all patients; 2) initial BMA only in patients at high risk; and 3) empiric therapy for all patients without initial BMA. High-risk criteria i nclude any of: platelet count >50 x 10(9)/L; hemoglobin <100g/L (age younge r than 12 months) or <110g/L (age older than 12 months); white blood cell c ount <5 x 10(9)/L (younger than 6 years) or <4 x 10(9)/L (older than 6 year s); or absolute neutrophil count <1.5 x 10(9)/L (younger than 6 years) or < 2 x 10(9)/L (older than 6 years). The results are expressed as quality-adju sted life years (QALYs), a measure that estimates the overall life expectan cy in years for patients receiving a particular treatment strategy, correct ed for the patient's quality of life. Results: The base case results are: 1) BMA all = 69.649 QALYs; 2) high-risk BMA = 69.652 QALYs; and 3) empiric therapy = 69.644 QALYs. These results i ndicate a three-way toss-up because there is less than a 4-day quality-adju sted difference (0.01) between strategies. Conclusion: This study indicates that the initial BMA does not significantl y change the overall QALYs of a child presenting with thrombocytopenia and, consequently, is not mandatory in every patient before starting steroids.