• Andrew D Franklin, Amanda N Lorinc, Matthew S Shotwell, Elton B Greene, and Curtis A Wushensky.
• From the Departments of *Anesthesiology, †Biostatistics, and ‡Radiology, Vanderbilt University Medical Center, Nashville, TN.
• Reg Anesth Pain Med. 2015 May 1;40(3):245-8.

Background And ObjectivesEpidural catheters placed for perioperative analgesia in young children confer clinical benefits but are technically challenging to insert. Approximations of the skin to epidural space depth in this population are limited to direct needle measurement and ultrasonography. Magnetic resonance imaging (MRI) is the most comprehensive imaging modality of the spine. This study aims to produce a more clinically useful formula from MRI data to estimate pediatric epidural depth.MethodsSeventy children with normal lumbar spine MR images were enrolled. After determination of epidural depth, linear regression was used to estimate a weight-based formula. Analysis of variance and bootstrap methods were used to evaluate this formula against 4 commonly cited formulae. The quality of predictions was evaluated using the mean absolute prediction error.ResultsThe estimated weight-based formula as derived by MRI data is given by: skin to epidural depth (mm) = 9.00 + 0.62 * weight in kilograms. The mean absolute prediction error was 2.56 mm (95% confidence interval [95% CI], 2.12-3.04) for the new formula. Additional derived formulae are skin to dorsal dura depth (mm) = 13.52 + 0.71 * weight in kilograms (mean absolute prediction error, 2.48 mm; 95% CI, 2.00-3.03) and skin to ventral dural depth (mm) = 23.08 + 0.86 * weight in kilograms (mean absolute prediction error, 2.50 mm; 95% CI, 2.04-3.06).ConclusionsWe provide the first predictive formulae, based on MRI data, for pediatric epidural depth estimation.

33 keywords...

hide…