MR-guided lumbar sympathicolysis

The aim of this study was to demonstrate the feasibility of MR-guided lumbar sympathicolysis (LSL) in a non-selected patient population. One hundred one MR-guided LSL procedures were performed in 89 patients according to Haaga's technique using a horizontally open clinical MR system (0.2 T) and non-ferromagnetic 20-G cannulas (neurolysis, n=93; blockade, n=8). Only gradient-recalled sequences in either single or multislice mode [fast imaging with steady-state precession (FISP) and fast low-angle shot] were applied for anatomical survey and needle guiding. Bupivacaine injection was monitored with MR fluoroscopically. Fluid distribution was subsequently documented in a CT scan in 65 patients. Ninety-one LSL procedures could be successfully completed. Ten patients were not treated using MR due to patient inconvenience, severe motion artifacts (n=4 each), excessive spondylophytes, and retroperitoneal hematoma (n=1 each). One case of ureteral necrosis occurred. Motion artifacts were rated less severe in single-slice FISP sequences and in obese patients. An average of 3.48 sequence measurements were required for definitive needle placement. Average table time was 32.3 min. An MR-guided LSL is feasible and can be performed with acceptable safety and time effort. It can be recommended for repeated sympathetic blockades in younger patients to avoid cumulative irradiation associated with CT guidance.