[Optimization of the image intensifier-assisted technique of lumbar sympathetic block. Computed tomographic simulation of a paravertebral puncture access].

The paravertebral approach is the most common technique for local anaesthetic and neurolytic lumbar sympathetic blocks. However, guidelines on the site of needle insertion differ. As there have been several case reports on accidental trauma to the ureter and the kidney, this study was undertaken to evaluate the site of paravertebral needle insertion and the fluoroscopic landmarks for lumbar sympathetic blocks by computed tomographic puncture simulation. METHODS. CT scans of 73 patients at the level of L2-4 were analysed with respect to the location of kidneys, the vertebral insertion of the diaphragm, the distance of the sympathetic trunk from the ventral and lateral border of the lumbar vertebrae, the paravertebral distance of a tangent from the sympathetic trunk through the kidney surface and the frequency of inadvertent puncture of major organs by different paravertebral approaches. For needle insertion distances of 6, 8, 10 and (at the level of L4) 12 cm lateral to the midline were simulated. RESULTS. Simulating a paravertebral approach of 6 cm resulted in perforation of the right (or left) kidney only at the level of L2 in 1.4% (2.8%) of cases. The incidence of accidental renal puncture due to a paravertebral approach of 8 cm was 26.0%, 4.1% and 0 (26.0%, 2.7% and 0) at the levels L2, L3 and L4, respectively. A more lateral insertion of needles 10 cm from the spinous process increased the frequency of anticipated renal puncture to 57.5%, 19.2% and 1.4% (65.8%, 26.0% and 1.4%); in addition, perforation of liver parenchyma was detected in two cases. Although the lower pole of the kidney reached the mid-vertebral level of L4 in only 23.3 (15.1)% of cases, a lateral approach 12 cm from the midline still showed a 8.2 (4.1)% incidence of kidney perforation and inadvertent trauma to the intestine in two cases. The mean distance from the sympathetic trunk to the ventral border of the lumbar vertebra (in simulation of a lateral fluoroscopic view) was 0.80, 0.66 and 0.59 cm, analogous measurements to the lateral border averaged 0.37, 0.43 and 0.50 cm at L2, L3 and L4, respectively. At the level of L2 the medial insertion of the diaphragm was identified in 45% of patients in close anatomical relationship to the psoas fascia. CONCLUSION. In order to reduce the risk of accidental trauma to major organs the paravertebral distance of insertion of the needles from the midline should not exceed 6, 7 and 10 cm for lumbar sympathetic blocks at the levels of L2, L3 and L4, respectively. However, a paravertebral approach of less than 6 cm may cause a lateral and ventral deviation of the needle from the sympathetic chain. Under fluoroscopy a correct needle position is obtained at an average distance of 0.5-0.8 cm dorsal to the anterior vertebral border, advancing the needle to the ventral border may cause an accidental puncture of the vena cava in more than 20% of patients undergoing nerve block of the right sympathetic chain. Furthermore, at the level of L2 inadvertent placement of the needle tip within the vertebral insertion of the diaphragm must be considered as a reason for atypical spread of contrast medium.