BACKGROUND: Maldistribution of intrathecal local anesthetic has recently been implicated as a contributor to neurotoxic injury. In vitro modeling can be used to understand the distribution of anesthetic agents within the subarachnoid space. We describe an in vitro modeling technique that uses digital video image processing and its application to catheter injection of local anesthetic. METHODS: A clear plastic model of the subarachnoid space, including a simulated spinal cord and cauda equina, was filled with lactated Ringer's solution. Phthalocyanine blue dye of known concentration was injected into the model through small-bore (28-G) and large-bore (18-G) catheters. Injections were performed at a variety of controlled rates and sacral catheter positions, and the propagation of dye throughout the model was recorded on videotape, digitized by computer, and converted to a two-dimensional image of dye concentration. A subset of data was compared with results obtained from spectrophotometric analysis. RESULTS: There was a strong correlation (r = 0.98) between data obtained with analysis by digital video image processing and those obtained spectrophotometrically. Catheter size, catheter angle, and injection rate significantly influenced the distribution and peak concentration of simulated anesthetic. No major differences in distribution or peak concentration were observed with the two types of 28-G catheters. CONCLUSIONS: The digital video image processing technique can be used to quantify anesthetic distribution rapidly within a model of the subarachnoid space without disturbing the distribution. The current results demonstrate a strong dependence of anesthetic distribution on catheter angle, catheter size, and injection rate. Comparisons between 28-G catheters suggest that the difference in reported incidence of cauda equina syndrome associated with different 28-G catheters cannot be explained on the basis of differences in anesthetic distribution.
BACKGROUND: Maldistribution of intrathecal local anesthetic has recently been implicated as a contributor to neurotoxic injury. In vitro modeling can be used to understand the distribution of anesthetic agents within the subarachnoid space. We describe an in vitro modeling technique that uses digital video image processing and its application to catheter injection of local anesthetic. METHODS: A clear plastic model of the subarachnoid space, including a simulated spinal cord and cauda equina, was filled with lactated Ringer's solution. Phthalocyanine blue dye of known concentration was injected into the model through small-bore (28-G) and large-bore (18-G) catheters. Injections were performed at a variety of controlled rates and sacral catheter positions, and the propagation of dye throughout the model was recorded on videotape, digitized by computer, and converted to a two-dimensional image of dye concentration. A subset of data was compared with results obtained from spectrophotometric analysis. RESULTS: There was a strong correlation (r = 0.98) between data obtained with analysis by digital video image processing and those obtained spectrophotometrically. Catheter size, catheter angle, and injection rate significantly influenced the distribution and peak concentration of simulated anesthetic. No major differences in distribution or peak concentration were observed with the two types of 28-G catheters. CONCLUSIONS: The digital video image processing technique can be used to quantify anesthetic distribution rapidly within a model of the subarachnoid space without disturbing the distribution. The current results demonstrate a strong dependence of anesthetic distribution on catheter angle, catheter size, and injection rate. Comparisons between 28-G catheters suggest that the difference in reported incidence of cauda equina syndrome associated with different 28-G catheters cannot be explained on the basis of differences in anesthetic distribution.
Authors: Per Thomas Haga; Giulia Pizzichelli; Mikael Mortensen; Miroslav Kuchta; Soroush Heidari Pahlavian; Edoardo Sinibaldi; Bryn A Martin; Kent-Andre Mardal Journal: PLoS One Date: 2017-03-15 Impact factor: 3.240