Article Text

Download PDFPDF
Evaluation of Ultrasound-Assisted Thoracic Epidural Placement in Patients Undergoing Upper Abdominal and Thoracic Surgery: A Randomized, Double-Blind Study
  1. David B. Auyong, MD*,
  2. Lucy Hostetter, MD*,
  3. Stanley C. Yuan, MD*,
  4. April E. Slee, MS and
  5. Neil A. Hanson, MD*
  1. *Department of Anesthesiology, Virginia Mason Medical Center, Seattle, WA
  2. Axio Research, Seattle, WA
  1. correspondence: David B. Auyong, MD, Virginia Mason Medical Center, 1100 9th Ave, Seattle, Washington 98101, USA (e-mail: David.Auyong{at}


Background and Objectives The placement of thoracic epidurals can be technically challenging and requires a thorough understanding of neuraxial anatomy. Although ultrasound imaging of the thoracic spine has been described, no outcome studies on the use of this imaging have been performed. We evaluated whether preprocedural ultrasound of the thoracic spine would facilitate the process of epidural catheterization.

Methods Subjects undergoing thoracic or upper abdominal surgery with planned thoracic epidural placement at T10 or higher were enrolled in this randomized double-blind study. Subjects were allocated into 1 of 2 groups for preoperative epidural placement: ultrasound guidance (group US) or palpation (group Palp). Subjects randomized to group US had a preprocedural ultrasound examination to identify pertinent spinal anatomy and make appropriate marks on the skin identifying midline and interlaminar spaces for targeted Tuohy needle insertion. Subjects in group Palp had a skin marking performed by palpation alone. Using the skin markings, all epidurals were performed using a loss of resistance to saline technique. Block levels were assessed with ice and pain scores obtained by a blinded nurse in the postanesthesia care unit. The primary outcome was procedural time from needle insertion to loss of resistance in the epidural space.

Results Seventy subjects were recruited and completed the study protocol. The median time for epidural needle placement to achieve loss of resistance in group US and group Palp was 188.5 seconds (interquartile range [IQR], 79.0–515.0) and 242.0 seconds (IQR, 87.0–627.0), respectively (P = 0.188). Using ultrasound to mark the skin overlying the targeted epidural space took a median time of 85 seconds (IQR, 69–113) for group US and 35 seconds (IQR, 27–51) for group Palp (P < 0.001). The number of needle passes was not significantly different between the 2 groups (P = 0.31). The use of ultrasound assistance resulted in a decreased number of needle skin punctures to achieve loss of resistance (P = 0.005). Mean pain scores after surgery were lower in group US compared to group Palp: 3.0 versus 4.7, respectively (P = 0.015).

Conclusions This is the first randomized study to evaluate the efficacy of preprocedural ultrasound marking for placement of thoracic epidural catheters. We observed that preprocedural ultrasound did not significantly reduce the time required to identify the thoracic epidural space via loss of resistance.

Clinical Trials Registration: NCT02785055 (

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • David B. Auyong has received honoraria from SonoSite FujiFilm (educational lectures and funded research) and Halyard Health (educational lectures) but declares no conflicts related to the research presented here. The other authors also declare no conflicts of interest.

    Attribution: Virginia Mason Medical Center, Seattle, WA.

    This work received no funding.