We investigated safety issues and potential experimental confounds when performing functional magnetic resonance imaging (fMRI) investigations in human subjects with fully implanted, active, deep brain stimulation (DBS) systems. Measurements of temperature and induced voltage were performed in an in vitro arrangement simulating bilateral DBS during magnetic resonance imaging (MRI) using head transmit coils in both 1.5 and 3.0 T MRI systems. For MRI sequences typical of an fMRI study with coil-averaged specific absorption rates (SARs) less than 0.4 W/kg, no MRI-induced temperature change greater than the measurement sensitivity (0.1 degrees C) was detected at 1.5 T, and at 3 T temperature elevations were less than 0.5 degrees C, i.e. within safe limits. For the purposes of demonstration, MRI pulse sequences with SARs of 1.45 W/kg and 2.34 W/kg (at 1.5 T and 3 T, respectively) were prescribed and elicited temperature increases (>1 degrees C) greater than those considered safe for human subjects. Temperature increases were independent of the presence or absence of active stimulator pulsing. At both field strengths during echo planar MRI, the perturbations of DBS equipment performance were sufficiently slight, and temperature increases sufficiently low to suggest that thermal or electromagnetically mediated experimental confounds to fMRI with DBS are unlikely. We conclude that fMRI studies performed in subjects with subcutaneously implanted DBS units can be both safe and free from DBS-specific experimental confounds. Furthermore, fMRI in subjects with fully implanted rather than externalized DBS stimulator units may offer a significant safety advantage. Further studies are required to determine the safety of MRI with DBS for other MRI systems, transmit coil configurations and DBS arrangements.