Functional localizers for improved DLPFC targeting: a comparison against standard rTMS targets

Repetitive TMS (rTMS) has emerged as a powerful tool in treatment of patients diagnosed with treatment-resistant depression when applied over DLPFC. Still, there is considerable variability in treatment response across patients. Inaccurate targeting is likely one of the most prominent causes for the modest response rates observed in TMS therapy. Two commonly used targeting approaches for TMS positioning in clinical setups are the “5cm” method and the F3 electrode target in the 10-20 EEG system. The TMS coil can, however, also be placed on targets via MRI-guided TMS neuronavigation, either specified by standardized MNI coordinates or individual fMRI-based functional localizers. In this study, we assess the spatial localizations of DLPFC targets across seven targeting approaches frequently used and compare them against an DLPFC target defined individually by a functional localizer based on facial emotion processing. Fifteen healthy volunteers (7f/8m) were examined using 7T ultra-high field functional MRI using the emotion discrimination task (EDT). This well-established task has been demonstrated to result in increased activation of the DLPFC (Sladky et al., 2013) as participants ere asked to match the emotional expressions. Distance variability was calculated as Euclidian distance between each subject’s activation maximum within the mask defined by group DLPFC cluster and seven common TMS targets. MNI coordinates of common TMS targets were defined based on previous research (Fox et al., 2012). We observed high inter-subject variability in the location of DLPFC activation peaks, with distances of up to 23.2 mm. Our results show that those TMS targets with higher stimulation efficacy (Fox et al., 2012) show lower mean distances (Rusjan: 16.79mm, Paus/Cho: 18.02mm) than targets with low efficacy (EEG-F3: 27.93mm, 5cm: 40.59mm). We conclude that in applications where individual fMRI-based functional localizers are not available, TMS targets in MNI space should be used for optimized TMS treatment. Figure 2Boxplot of distance of single subject activation maxima to common targets. Distances to functional TMS targets (Rusjan, Paus/Cho) and BA46 that have higher stimulation efficacy (Fox et al, 2012) show lower medians than distance to targets with lower efficacy (average 5cm, EEG F3).2 View Large Image Figure Viewer Download Hi-res image