A technique of measuring the precision of an MR-guided stereotaxic installation using anatomic specimens.

PURPOSE To develop a method for direct measurement, using anatomic specimens, of the precision of MR-guided stereotaxic location and to describe its application to a 1.5-T MR unit with a Leksell stereotaxic frame. METHODS Small pieces of gelfoam (1 X 1 X 1 mm), soaked in gadopentetate dimeglumine, were stereotaxically introduced into formalin-fixed human heads using a Leksell D (three experiments) or G (nine experiments) stereotaxic frame. The head and the frame were then introduced into a 1.5-T MR unit. The target coordinates (as set on the stereotaxic frame by one investigator) were then compared with the MR-determined stereotaxic coordinates (calculated independently by another investigator). The imprecisions Ex, Ey, and Ez in each direction were defined as the differences between the calculated and the chosen coordinates. RESULTS Regarding the three targets studied with the D frame, mean imprecision Ex was 1.08 +/- 0.50 mm (mean +/- SEM), Ey 0.83 +/- 0.58 mm, and Ez 0.75 +/- 0.25 mm. For the nine targets studied with the G frame, Ex was 0.48 +/- 0.17 mm, Ey 0.69 +/- 0.14 mm, and Ez 0.82 +/- 0.13 mm. Statistical analysis of the results showed no significant difference between Ex or Ey and half the size of a pixel, indicating that, in the axial plane, stereotaxic MR precision is limited only by pixel size. A statistically significant difference was observed in the coronal plane between Ez and half the size of a pixel, but it must be stressed that Ez remained smaller than 1 mm. CONCLUSION MR-guided stereotaxic location can be used with confidence for most diagnostic, functional, and therapeutic procedures.