Contribution to diagnosis and treatment of bone marrow aspirate results in critically ill patients undergoing bone marrow aspiration: a retrospective study of 193 consecutive patients

Background The purpose of the work was to assess the contribution to diagnosis and/or treatment (CDT) of bone marrow aspiration (BMA) in the critically ill patient. Methods The retrospective study included 193 patients. On the basis of BMA findings, contribution to diagnosis was defined by one of four previously unestablished diagnoses (maturation arrest of granulocyte precursors, hemophagocytic lymphohistiocytosis, hematological malignancy, marrow infiltration with cancer cells) and to treatment as the initiation or withdrawal of a specific treatment including the decision to forgo life-sustaining treatment (DFLST). Results A CDT of BMA was observed in 40/193 patients (20.7%). BMA contributed to diagnosis in 37 cases (granulocyte precursor maturation arrest, N = 10; hemophagocytic lymphohistiocytosis, N = 12; hematological malignancy, N = 15) and to treatment in 14, including three DFLSTs. In multivariate analysis, the factors associated with a CDT were hematological malignancy, cancer or non-malignant hematological abnormality known on admission, indication for BMA excluding isolated thrombocytopenia, higher pre-BMA HScore (calculated prior to BMA), and higher SOFA score with or without platelet-count SOFA subscore. In the 160 patients without hematological malignancy or cancer known on admission, non-malignant hematological abnormality known on admission, indication for BMA excluding isolated thrombocytopenia, higher pre-BMA HScore, and higher SOFA score calculated with or without platelet-count SOFA subscore were independently associated with a CDT of BMA. Conclusion BMA can have a significant CDT in ICU patients with or without a known hematological malignancy or cancer on admission. An HScore calculated before BMA can be a valuable tool for predicting a CDT of BMA.