Naji and colleagues’ observations based on a review of a series of anatomically normal MRI scans may provide a useful basis to encourage continued refinement of neuraxial blockade in anaesthetic practise [1]. We agree that this is a difficult and important area of investigation. The authors outlined some of the practical limitations of their study including the position of patients for anaesthesia compared with the supine position in which MRI scanning would have been conducted, and described the difficulties of accurate clinical identification of which lumbar interspace is being used. 

We would like to make the following comments on their paper. Firstly, the study reviewed MRI scans performed on both male and female patients. The authors do not say whether the changes in shape of the theca were affected by the sex of the patient. The authors highlight reports of a failure rate of up to 16% of the spinal component when employing a combined spinal-epidural technique and imply that failure of the spinal component is a 'common problem' in their practice. They do not state whether their own practice uses the midline approach to the epidural space, paramedian approach or a combination of both.

We subscribe to the view that spinals and epidurals (in obstetric patients at least) should be performed at the lowest possible interspace and speculate that the triangular shape of the theca at the L4-5 interspace demonstrated in the authors paper may prove advantageous when employing a CSE using the paramedian approach. This is because the spinal needle will approach the dura from a more perpendicular direction compared with the midline approach, which may be quite oblique if the dural sac is triangular at that level. The study concludes that the anatomical shape of the theca at the L4-5 interspace makes puncture of the dura by the spinal needle less likely. This may also suggest that the L4-5 interspace will reduce the chance of accidental dural puncture by the Tuohy needle if the midline approach is utilised.

J.L. Westbrook

Nuffield Department of Anaesthesia

John Radcliffe Hospital

Headington

Oxford

OX3 9DU

Reference

1. Naji M, Williams M, Hourihan MD, Collis RE. Shape of the thecal sac: L3/4 interspace compared with L4/5. Anaesthesia 2009; 64: 39-42.