We evaluated the accuracy of the needle tip representation by different imaging techniques for the guidance of facet infiltrations. For visualisation of the lumbar facet joints we used a high-field magnetic resonance tomograph (MRT) with a 2.0 Tesla field and 3.5 mm slice thickness, an open low-field magnetic resonance tomography (MRT) with an 0.064 Tesla field and 9 mm slice thickness, and IMATRON electron beam computed tomograph (EBCT) with a slice thickness of 6 mm, and a mobile C-arm fluoroscope. The study was performed on 4 human cadaveric lumber spine preparations, each of which had 8 facet joints. Under imaging control, special injection needles were placed as close as possible to the facet joint space. Following placement of he needle, all specimens were scanned with the electron beam tomograph using a slice thickness of 1.5 mm. The thin-slice study served as the gold standard. The distance between the tip of the needle and the facet joint was measured in all the images. Comparison of the different modalities with the gold standard revealed the following results: 1) median values of the absolute differences were 1.25 mm for high-field MRI, 1.35 mm for 6 mm EBCT, 2.05 mm for low-field MRI, and 2.30 mm for X-ray fluoroscopy. 2) While there was no statistically significant difference in the accuracy of tip localization between high-field MRI and 6" EBCT (p = 0.293), both systems were more precise than low-field MRI (p = 0.04) and X-ray fluoroscopy (p = 0.009). When choosing the best imaging technique, such additional factors as radiation, costs and time, must also be considered. Provided necessary radiological precautions are taken, and assuming careful pre-interventional planning, CT. EBCT and X-ray fluoroscopy are currently more effective than the expensive, time-consuming and costly magnetic resonance tomography.