BACKGROUND: The diagnosis of immune thrombocytopenia (ITP) is one of exclusion. Although guidelines recommend against routine bone marrow examination (BME) at time of ITP diagnosis, the role of BME in relapsed/refractory ITP is unclear.
OBJECTIVES: To examine the frequency and predictors of BME in relapsed/refractory ITP.
PATIENTS/METHODS: This multicenter retrospective cohort study included adults with ITP who received second-line therapy in Alberta, Canada from 2012 to 2019. We calculated the frequency of BME and rate of abnormal marrow findings. Logistic regression was performed to assess predictors of BME and predictors of bone marrow pathology.
RESULTS: Of 324 patients with presumed ITP, 181 (56%) underwent BME. We observed a marked decline in the rates of BME among patients >60 years over the past decade, but not in patients younger than age 60 years. On multivariable logistic regression, older age (adjusted OR [aOR] 1.03, p = .0001), anemia (aOR 2.5, p = .01), splenomegaly (aOR 3.2, p = .01), splenectomy (aOR 2.4, p = .02), and lack of splenectomy response (aOR 3.4, p = .04) were significant predictors of BME. Abnormal marrow findings were found in eight (2% of overall cohort; 4% of BME): four myelodysplastic syndrome, one aplastic anemia, one chronic lymphocytic leukemia, one metastatic cancer, and one megaloblastic anemia. Seven (88%) underwent BME for bicytopenias/pancytopenias. Macrocytosis (aOR 9.6, p = .03) and rural residence (aOR 6.7, p = .02) were independent predictors of abnormal bone marrow findings.
CONCLUSIONS: Although routine BME is frequently performed in relapsed/refractory ITP, abnormal findings are rare. Future prospective studies are needed to help identify a subgroup of relapsed/refractory ITP who may benefit from BME.