Bone marrow content measured in radioimmune bone marrow scintigraphy: intra- and inter-observer variability

The aim of this study was to assess the possible quantification of vertebral residual bone marrow content relative to the bone marrow content of a non-irradiated vertebra. This method is based on the vertebral count activity, measured using radioimmune bone marrow scintigraphy. First, however, we had to evaluate intra- and inter-observer variability. In three patients who underwent radioimmune bone marrow scintigraphy, two independent observers measured the count density in 51 (15 lumbar and 36 thoracic) vertebrae using a manually drawn region of interest. To evaluate intra- and inter-observer variability, we calculated the means and standard deviations of the differences between measurements. Bland-Altman plots were drawn for all vertebrae as well as for three subgroups of vertebrae (the upper thoracic spine, D1-D6; the lower thoracic spine, D7-D12; and the lumber spine, L1-L5). For all vertebrae, the mean (+/- S.D.) difference, expressed as a percentage of the overall mean, was -0.44 +/- 3.3% for observer 1 and -0.3 +/- 2.1% for observer 2 for intra-observer variability; inter-observer variability varied from 0.55 +/- 3.9% to 1.28 +/- 3.7%. On the Bland-Altman plots, the data points were evenly distributed above and below the 0-line and the linear regression equations matched the line of equality almost perfectly. This pattern was observed for all the vertebrae as well as for the subgroups of vertebrae. In conclusion, our results show that the intra- and inter-observer variabilities are not great, confirming that this technique is simple and robust and can be used for further quantification of bone marrow content in the axial skeleton.