Aim: The aim of this study was to determine the effect of the applied method of image investigation (CT or MRI) on the following parameters of the neuronavigation (NN) assisted procedure: "number of attempts for registration", "duration of registration" and "registration accuracy".
Patients and methods: A total of 195 patients with various cranial pathological lesions underwent neuronavigation-assisted surgery between March 2003 and December 2005 at the Clinic of Neurosurgery of St. I. Rilsky University Hospital, Medical University, Sofia. All of them were included in our study. CT based registration was realized in 81 patients of our series and MR based registration--in 114 patients. The patients were examined and followed up in a standardized manner. We conducted a prospective study on the effect of the type of image investigation (either CT or MRI) on the parameters registration accuracy, number of attempts for registration and duration of registration. Statistical analysis was performed using a one-factor non-parametric rank analysis (Friedman ANOVA) with a factor "type of image study" which had four sublevels: "MR I", "CT", MRI+MRI Angiography" and "MRI+CT". Multiple Hotelling F-contrasts were applied; the level of statistical significance was 95% (Statistica 6.0, 2001, Statsoft, Tulsa, USA).
Results: In the series, the best (the lowest digital expression) mean registration accuracy was demonstrated in MRI based neuronavigation (1.6 mm). Besides that, in MRI-NN the number of attempts for registration was smaller and the time necessary for registration was shorter in comparison with CT-NN. In the "MR I+CT" subgroup the mean values of the investigated parameters were lower in comparison with the other three subgroups. Nevertheless, the differences were not statistically significant because of the wide statistical dispersions of the obtained results and the small number of patients investigated in this subgroup. ANOVA did not show statistically significant difference between the number of attempts for registration, the duration of registration and the registration accuracy in CT and MR-based neuronavigation-assisted procedures.
Conclusion: We did not found any statistically significant differences between CT- and MRI-based neuronavigation with respect to the accuracy and reliability of the methods. The type of image study that should be used for neuronavigation depends on the specific characteristics of the corresponding pathological lesion.