To the Editor, Recent correspondence in the Journal highlighted the usefulness of a porcine carcass phantom for teaching and learning ultrasound-guided paravertebral techniques because of the notable similarity of spinal and paravertebral structures in pigs and humans. We have previously described a method of deodorizing and curing a porcine model for lumbar sonoanatomy. We have now used the porcine model to assess the feasibility of a singleoperator ultrasound-guided epidural technique. The phantom was a T12-L5 section of pork spine with associated musculature cured in 66% alcoholic gel (Aqium Gel, Ego Pharmaceuticals Pty Ltd, Australia). The Sir Charles Gairdner Group Human Research Ethics Committee approved the project (No. 2010-179), and all participants gave written informed consent. Ten anesthesiologists were recruited who were experienced in ultrasound-guided regional anesthesia but were non-expert in spinal sonography. They were shown a technique of paramedian epidural insertion using ultrasound assistance (Sonosite M-Turbo, 13-8 MHz linear array, Bothel, WA, USA) with a 17G echogenic Tuohy needle (Pajunk TuohySono, Geisingen, Germany) and an Epidrum device (Exmoor Innovations, UK) that provides visual confirmation of loss of resistance (LOR) when the air-filled diaphragm deflates. The Tuohy-Epidrum-syringe assembly was introduced in-plane from the caudal end of the transducer, and the needle was advanced until the tip approached the ligamentum flavum. At this point, the Epidrum device was inflated (Figure). Deflation occurred as the needle tip entered the epidural space, and the Epidrum and syringe were removed. The primary endpoint was successful threading of the catheter and appearance in the epidural space at the cephalad end of the phantom. Each anesthesiologist performed the task twice. Their attempts were recorded and timed by one of the authors (S.B.). Participants made a subjective assessment of the technique using a structured questionnaire. Secondary endpoints included time taken from needle first on view to visual LOR and percentage of time the needle was in view. The catheter was threaded successfully in 19 out of 20 attempts. One failure occurred due to premature Epidrum deflation prior to the needle entering the epidural space. This was clearly imaged but protocol dictated an attempt at catheter insertion. Mean (standard deviation) time to LOR was 51 (27) sec. The needle was visible 90 (10)% of the time corresponding with 90% of participants assessing visibility as excellent on a three-point rating scale. The task was evaluated as easy or intermediate by 60% of participants. While the other 40% of participants performed equally well, they expressed their natural caution over a new technique that others have labelled as ‘‘advanced’’. All participants would consider performing the technique on a patient, which suggests that phantom practice allowed participants to gain confidence that could result in changes to clinical practice. The echogenic TuohySono overcame the needle visibility difficulties at steep insertion angles that have previously been highlighted. Stiffness of the Tuohy allowed the needle rather than the transducer to be manipulated to preserve an in-plane view. The stationary transducer maintained an optimal image of the target at all times. The Epidrum made a single-operator Parts of the data have been presented as a poster at the Australian New Zealand College of Anaesthetists Annual Scientific Meeting in Hong Kong, May 2011.