In January, the National Institute for Health and Clinical Excellence (NICE) issued guidance on the use of ultrasound to facilitate catheterisation of the epidural space. In the published information, available to the public, the use of ultrasound compared with the ‘traditional landmark approach’ is quoted as being associated with fewer attempts, a decreased incidence of dural puncture and post procedure headache, and overall greater patient satisfaction [1]. These conclusions, which were based on obstetric and paediatric populations, and also on those requiring corrective back surgery, do not provide a balanced approach to the actual findings of the studies quoted. In two small studies involving neonates and children, there were no sound clinical end points that reached statistical significance [2, 3]. Further, we have reservations regarding the conclusions reached from the obstetric studies. In the three studies [4–6] where ultrasound was compared to a standard landmark technique (all cited from the same main author) there is no compelling statistical end point that favours the use of ultrasound. The decreased incidence of dural puncture quoted by NICE is based on two cases in the 150 control group (landmark) compared with one in the prepuncture ultrasound group. There are no statistical data for this [4]. In addition, whilst there were more cases of severe post-procedure headache in the control group (10% versus 2.7%) this actually grossly exceeds the number of dural punctures. Further, whilst ‘patient satisfaction with epidural space procedure’ was deemed to be higher in the ultrasound group, overall ‘patient satisfaction with general treatment’ was better in the control group. In pregnant women with a body mass index over 33 kg.m2, scoliosis, kyphosis, hyperlordosis or where there was anticipated difficulty in epidural insertion, fewer puncture attempts were required when ultrasound was used [5]. However, there were no recognised dural punctures in either group, and no statistical difference in failure rates or headaches. In a randomised control trial of 30 pregnant women comparing ultrasound and the landmark technique for the insertion of a combined spinal-epidural for Caesarean section, there was a significant reduction in the number of attempts using imaging. However, the success rate without ultrasound was unusually low, with 60% requiring more than one attempt [6]. We conducted a local email survey of anaesthetists in Wessex in light of these NICE guidelines. Of the 151 respondents, a 40% response rate, the majority being consultants, 97% have never used ultrasound to image the epidural space, although in the same geographical district it is common to use ultrasound both for regional anaesthesia and vascular access. Over 60% of respondents had received formal ultrasound teaching for the techniques that they used. In contrast, more than 90% were not aware of any courses that taught ultrasound techniques to identify the epidural space. Overwhelmingly, they concluded that without formal teaching, they would not find ultrasound guided epidural catheter insertion a useful technique. Ultrasound is an invaluable tool for anaesthetists both in vascular access and regional anaesthesia, its benefit in facilitating epidural catheter insertion is not demonstrated in the current literature referenced by the NICE guidelines. We agree that it has the potential to be a useful adjunct as a prepuncture method in those patients with difficult anatomical landmarks (obesity and back anomalies [7]). This will primarily be to obtain the level and estimated depth of the epidural space (accepting that some of the accuracy will be lost because of the probe compression, local anaesthetic infiltration and discrepancies in needle angle). In conclusion, there will need to be more evidence before most anaesthetists will be convinced of the usefulness of ultrasound for the placement of an epidural. The required infrastructure will need to be available before the NICE guidance can be adopted.