Abstract
PURPOSE: To investigate how unselected bone marrow (BM) biopsy examinations could help in the diagnosis of clinically unknown nonhematological tumors. METHODS: 10,112 plastic-embedded BM biopsy sections were retrospectively analyzed. In the analysis we focused on the following aspects: (1) the frequency of BM involvement arising from clinically unknown nonhematological malignancies, (2) the clinical indication for BM biopsy examination in cases with tumor BM metastasis, (3) the primary sites of the metastatic tumors, and (4) the advantage of plastic-embedded biopsy sections over paraffin-embedded samples and the complementarity of biopsy with aspiration smears. RESULTS: Of the 10,112 BM samples analyzed, 101 (1.0%) were interpreted as being nonhematological tumor metastases. In cases with metastatic tumors, the nonhemocyte-related changes, such as skeletal pain (25%) and bone destruction (5%), were considerably higher than in those without metastasis (3.7 and 0.32%, respectively; P < 0.001). Primary lesions were documented antemortem in 50 of the 101 biopsy-positive cases (49.5%); the most frequent being in the lung, gastric tract, and breast. Using this assay, a higher incidence of metastatic tumors was detected when compared with previously reported paraffin-embedded samples. The frequency of metastatic tumors based on aspiration smears when the positive results for biopsy sections were used as a reference was 74.3%. All the 101 sections with metastatic tumors showed various degrees of myelofibrosis. CONCLUSIONS: We concluded that the routine BM biopsy examination is helpful in detecting insidious metastatic nonhematological malignancies. Skeletal pain and bone destruction are critical indications for susceptible patients to undergo BM examination. Plastic embedding of biopsy sections appears to be more sensitive than the paraffin embedding of samples and is an excellent complementation to isolated aspiration smears.