A comparative study of MIB-1 staining indices of gliomas measured by NIH Image analysis program and conventional...

To evaluate the NIH Image measured MIB-1 staining index (SI) as a useful tool for the determination of proliferative activity of gliomas, MIB-1 immunohistochemistry was performed on paraffin sections to estimate the proliferative activity of 50 gliomas. Five to eight RGB images at 200 magnifications were acquired per sample in a Macintosh computer using a RD-175 digital camera and Adope Photoshop program, followed by conversion of the RGB image to B (blue subtracted) and R (red subtracted) images and finally automatic cell counting of more than 1000 cells per sample by the NIH Image analysis program. The mean MIB-1 staining indices by both NIH count and visually assessed (manual) count respectively were 12.7% +/- 9.9% (mean +/- standard deviation) and 12.6% +/- 9.9% in all gliomas (n = 50), 2.9% +/- 0.6% and 2.3% +/- 0.6% in grade I and II astrocytomas (n = 9), 10.7% +/- 8.9% and 11.1% +/- 9.0% in grade III gliomas (n = 17), 16.1% +/- 9.1% and 16.3% +/- 9.1% in glioblastoma multiformes (n = 20), 26.0% +/- 4.9% and 24.5% +/- 2.6% in other glial tumors (n = 4). These paired values were assumed to be identical, since the p value obtained by paired t-test in all gliomas was 0.905. We conclude that the MIB-1 staining index measured by NIH Image is reliable and universal.