Background and Aims There is insufficient evidence on which ultrasound (US) view can predict epidural depth for midline epidural procedure in children with scoliosis. We hypothesized that the US estimated distance from the skin to the epidural space (US-ED) in the paramedian sagittal oblique (PSO) plane is comparable with the US-ED in the TM plane to predict actual epidural depth.

Methods The institutional review board of the Severance Hospital has been granted (IRB no. 4-2021-0266). 55 patients being placed in a flexed left-sided position, US-EDs was measured in the bilateral PSO and TM plane at the L2/3 interspace. During the midline epidural puncture using the loss-of- resistance technique to air, the needle depth from the skin to the epidural space was sought (table 1). Correlation between the US-EDs and the needle depth was investigated with Pearson’s correlation coefficient (PCC), Concordance Correlation Coefficient (CCC). The graded visibility of posterior dura complex was compared.

Results PCC and CCC between the US-EDs and the needle depth were excellent in all planes. Amongst all US-EDs, the longer value of the US-ED in the PSO taken from both sides showed highest PCC and CCC value (table 2). The ‘good’ visibility is significantly higher in the PSO view than in the TM view (72.7% vs. 38.2%, P-value <0.001).

Conclusions PSO and TM planes are both interchangeably feasible to predict the needle depth in pediatric patients with lumbar scoliosis. However, the longer of the two US-EDs in the bilateral PSO view is more reliable than US-ED in the TM view with better visualization.

Ethics Committee Approval