Needle Heating During Interventional Magnetic Resonance Imaging at 1.5- and 3.0-T Field Strengths

OBJECTIVES: The aim of this study was to test the hypothesis that clinically used magnetic resonance (MR)-conditional needles of varying lengths, orientations, locations, and pulse sequences can result in excessive heating during MR imaging (MRI)-guided interventions that can be minimized to physiological ranges with proper selection of the needle length, needle position, and modification of pulse sequence parameters. MATERIALS AND METHODS: We simulated a clinical interventional MRI setting with 2 standard American Society for Testing and Materials F2182-11A phantoms and measured temperatures with fiber optic sensors. Temperature profiles were monitored for commercial 10, 15 and 20 cm MR-conditional cobalt-chromium needles in clinically relevant perpendicular, 45-degree oblique, and parallel orientations relative to the static magnetic field (B0) and center, right off-center, and left off-center needle tip locations in the z = 0 plane. Clinically available interventional MRI pulse sequences including turbo spin echo (TSE), fast TSE, slice encoding for metal artifact correction, compressed sensing slice encoding for metal artifact correction, half-Fourier acquisition single-shot TSE (HASTE), HASTE inversion recovery, fluoroscopic steady-state gradient echo (3.0 T only), fast low-angle shot gradient echo, and volumetric interpolated breath-hold examination gradient echo pulse sequences were tested at 1.5 and 3.0 T field strengths. Acquired temperature data were analyzed using Friedman and Wilcoxon signed-rank tests with Bonferroni correction. RESULTS: After 5-minute of continuous MRI, less than 2.5°C heating occurred when needles were oriented perpendicular and 45-degree oblique to B0, regardless of field strengths. Higher temperature rises capable of causing permanent tissue damage were observed when needles were oriented in parallel to B0 (1.5 T: 22°C with 20 cm needles, 3.0 T: 8°C with 10 and 15 cm needles) using higher radiofrequency energy pulse sequences, such as TSE and HASTE. Left off-center location, parallel orientation, and needle lengths close to half of the radiofrequency pulse wavelength were positively associated with higher temperature rises. CONCLUSIONS: Under the herein used experimental conditions, clinically used MR-conditional needles can heat to supraphysiologic temperatures during prolonged MRI at 1.5 and 3.0 T field strengths; however, the temperature rise can be balanced to physiological ranges with proper selection of needle length, needle orientation, and pulse sequence parameters. Caution must be exercised when using different MRI systems, as results may not directly translate.