Color flow Doppler ultrasonography can distinguish caudal epidural injection from intrathecal injection

Color flow Doppler ultrasonography has been used to confirm caudal epidural injection, but its ability to detect accidental intrathecal injection is unknown. We hypothesized that, when using color flow Doppler, the injection of fluid into the epidural space would result in turbulent flow which would appear as a burst of color while intrathecal injection would show an absence of a color flow Doppler signal.

Two groups of pediatric patients (up to 6 years of age) were prospectively enrolled for this observational study during a 2-month period. One group (group E) consisted of patients suitable for elective surgery using caudal epidural analgesia, and the other (group I) included patients receiving lumbar puncture for intrathecal chemotherapeutic injection. After induction of general anesthesia and placement of the patient in the lateral position, an 8 MHz curved array probe (Sonosite TITAN, Bothell, WA) was applied to obtain a transverse image of the lumbar region (L1-L3). Real-time images using color flow Doppler were obtained and recorded during initial injections of 2 consecutive (20 seconds apart) aliquots of 0.1 mL/kg medication of local anesthetic (0.25% bupivacaine) or chemotherapy drugs (mixture of methotrexate, cytarabine, and hydrocortisone) at a rate of 0.5 to 1.0 mL/s. After obtaining the study images, the rest of the medication was injected in standard fashion. A blinded anesthesiologist later evaluated the recorded images to determine a positive or negative result (positive = presence of turbulence as illustrated by a medley of color; negative = no turbulence or color). Sensitivity, specificity, and positive and negative predictive values were calculated for those patients who had successful analgesia (group E) and intrathecal (group I) injections.

Forty recorded images from 41 patients (group E, n = 21; group I, n = 20) were included in the analysis. The observed sensitivity, specificity, positive predictive value, and negative predictive values were all 100%. The lower 95% confidence limits were 0.832.

In the context of this study, color flow Doppler could differentiate epidural from intrathecal injection into the caudal space of children up to 6 years of age using a 0.1 mL/kg injection volume and injection rate of 0.5 to 1.0 mL/s.