Background and Objectives Advancement of catheters from the caudal to the thoracic level is an alternative to thoracic epidural anesthesia in infants and younger children; however, contamination of the insertion site may occur. This study examined the feasibility of the midline modified Taylor approach (L₅-S₁) for the advancement of epidural catheters to the thoracic level in infants.

Methods After Institutional Review Board (IRB) approval and parental consent, the L₅-S₁ interspace of infants 3 months to 2 years old was entered with an 18-gauge Crawford needle using the saline loss of resistance technique. A 20-gauge catheter with stylet (Abbott; North Chicago, IL) was then advanced the distance from the L₅-S₁ interspace to the desired thoracic level. If resistance was encountered, the catheter was withdrawn 1 to 2 cm, rotated along its long axis, and readvanced. The stylet was left in place, and a radiograph of the thoracolumbar spine was taken. The stylet was then removed, and the catheter was secured, tested, and dosed.

Results Sixteen infants (mean age, 14.4 ± 5.7 months and mean weight, 9.3 ± 1.4 kg) were studied. Fifteen of 16 catheters were inserted the full length planned. Fourteen of 16 catheters were straight (1 had a single bend, and 1 had multiple loops). Mean discrepancy between level desired and obtained was −1.7 ± 1.7 segments (median, −1.75). Discrepancy did not correlate with either desired level or length inserted, but did decrease with experience.

Conclusions The midline modified Taylor approach allows access to the thoracic epidural space via catheter advancement, while being below the terminus of the spinal cord and less likely to suffer contamination than the caudal approach.