The Effect of Different Injection Rates on the Spread of Spinal Anesthesia

Abstract

To the Editor: The article by Horlocker et al. [1], which concludes that in spinal anesthesia with a 22-gauge Whitacre needle, a fast injection (0.5 mL/s) of 0.3% bupivacaine at room temperature results in a higher maximum sensory level than slow injection (0.02 mL/s), is an interesting study that deserves more elaboration. It is true, as they state, that previous studies that have failed to document a difference have probably used a "slow" rate of injection that was too rapid, resulting in turbulent flow and mixing. They are to be commended for achieving such a slow infusion rate with an electric syringe pump, but they fail to state the type of pump used. Most syringe pumps actually produce pulsatile flow with consequent accelerations and decelerations in flow rate, although the mean flow rate is steady. This effect produces quite marked turbulent flow patterns. If they did use a constant rate infusion device, then the slow flow rate (< 3 mL/min) used would have produced a smooth, nonturbulent stream through a 22-gauge Whitacre needle [2]. We have just completed a study looking at the distribution patterns of isothermal, isobaric methylene blue dye after injection at various flow rates through atraumatic spinal needles (Serpell MG, Gray WM, personal communication). With an open fluid tank, we have found that at high flow rates the injectate is rapidly dispersed and diluted. Conversely, at low flow rates, the injectate can travel for a similar or greater distance, but as a concentrated stream, before finally gradually dispersing by diffusion. When the experiment is repeated and a flat perspex plate is placed perpendicular to the needle, we found that with slow rates the injectate adhered to the plate and continued to travel further along in its original direction before the momentum ran out. The fast rates produced a turbulent pattern that whirled back from the surface of the plate and the distribution was less directional, with the dye spreading laterally and backwards to the initial direction. Our model is not intended to simulate that of the subarachnoid space. There will be many other factors that affect the final distribution of local anesthetics in the clinical setting. These factors include the cylindrical dimension and presence of the cord and nerve roots, pressure changes transmitted by the circulation, respiration and body movements, and also thermal and baricity differences of the injectate. Our in vitro observations, however, support findings of Horlocker et al. that with a fast injection rate there would be more lateral (cephaled and caudad) spread when the Whitacre needle orifice is point perpendicular to the longitudinal axis of the cord. Also, it would suggest a more directional spread with slow injection rates if the needle orifice were pointed along the longitudinal axis. We are currently undertaking a clinical study to assess whether there is a difference between fast and slow injection rates through a 25-gauge Whitacre needle with the orifice pointing cephalad. Preliminary findings do in fact suggest that there is a higher cephalad spread with the slow rate. M. G. Serpell, MB, ChB, FRCA W. M. Gray, BSc, PhD Department of Anaesthesia, Western Infirmary, Glasgow G11 6NT, Scotland