Ultrasound-Guided Epidural Catheter Insertion in Children

Rapp, Hans-Jürgen MD; Folger, A MD; Grau, T MD, PhD

Section Editor(s): Davis, Peter J.

doi: 10.1213/01.ANE.0000156579.11254.D1
Pediatric Anesthesia: Research Report

Epidural catheters (EC) are often used in pediatric patients for intraoperative and postoperative pain relief. The small anatomical structures and catheter insertion under general anesthesia make it more difficult to perform EC and to prevent damage. In this study we investigated the use of ultrasound (US) in detecting neuraxial structures during insertion and placement of EC in children. ASA I–II children scheduled for elective surgery under combined general and epidural anesthesia were studied. Patients received balanced anesthesia using sevoflurane, opioids and rocuronium. Before EC insertion US examination in a lateral position was done to visualize and identify neuraxial structures. Quality of visualization and site and depth of structures were recorded. Using a sterile kit to hold the US probe in position and enable the visualization of the neuraxial structures, an epidural cannula was inserted, using the loss of resistance technique, as the EC passed under US control to the desired level. Of 25 children, 23 were evaluated. Epidural space, ligamentum flavum, and dural structures were clearly identified and the depth to skin level estimated in all patients. Loss of resistance was visualized in all patients with a lumbar epidural approach. Correlation of US measured depth and depth of loss of resistance was 0.88. In eight of 23 patients EC could be visualized during insertion and in 11 others it could be visualized with additional US planes. US is an excellent tool to identify neuraxial structures in both infants and children. The size and the incomplete ossification of the vertebra allow exact visualization and localization of the depth of the epidural space, the loss of resistance, and all relevant neuraxial structures.

IMPLICATIONS: Epidural catheters in children are mostly inserted under sedation or general anesthesia. This study showed that the use of ultrasound could help visualize all relevant neuraxial structures and their site and depth from the skin.

*Department of Anesthesiology and Intensive Care, University Hospital Mannheim and †Department of Anesthesiology University Hospital Heidelberg, Faculties of the Ruprecht-Karls-University, Heidelberg, Germany

Accepted for publication January 7, 2005.

Address for correspondence: Hans-Jürgen Rapp, MD, Department of Anesthesiology and Intensive Care Medicine, University Hospital Mannheim gGmbH, Ruprecht-Karls-University Heidelberg, Theodor-Kutzer-Ufer 1–3, 68167 Mannheim, Germany. Address e-mail to h-j.rapp@urz.uni-heidelberg.de.

Article Outline

Epidural analgesia provides excellent intraoperative and postoperative analgesia in both adults and children. Consequently epidural catheters (EC) often are used for intra and postoperative pain relief after major surgery in children (1,2). The size of anatomical structures, however, sometimes makes it difficult to perform EC in infants and small children and the risk to benefit question arises. Several publications about difficulties or damage are described in the literature (3–5).

Cork et al. (6) first described the possibility of visualizing neuraxial structures by ultrasound (US), but until recently US was not used in inserting and visualizing EC. Technical improvements in US enable digital depiction at high resolution, which makes its clinical use feasible (7–10).

The aim of this study was to investigate the possibility of using US imaging to detect neuraxial structures before and during placement of EC in infants and children and to visualize the site and location of the catheter or catheter tip.

Back to Top | Article Outline


After ethics committee approval and parental written consent was obtained, children scheduled for elective major surgery were enrolled. All patients were of ASA physical status I–II undergoing elective surgery and requiring general anesthesia with combined epidural analgesia for intraoperative and postoperative pain relief. Patients known or suspected to have neurological diseases or anatomical malformation of centroaxial structures were excluded from the study.

Patients older than 6 mo were premedicated with midazolam 0.4 mg/kg rectally. Anesthesia was induced using 5–7 mg kg1- thiopentone, and manual ventilation was performed with oxygen (Fio2 1.0) and sevoflurane (1.0 vol. % end-tidal) via facemask. After tracheal intubation anesthesia was maintained with sevoflurane (up to 1.5–2 vol. %), along with fentanyl and rocuronium. Standard monitoring included precordial stethoscope, electrocardiogram, pulse oximetry, noninvasive arterial blood pressure, esophageal temperature, capnography, and end-tidal concentrations of volatile anesthetics.

Before surgery, in right- or left-sided lateral position, US examination was performed of the sacral, lumbar, and thoracic region at levels S-2, L-4, and Th-6 to identify the relation of the tissues, as well as tissue visibility and depth. A 7-MHz probe (patients 1 to 15) and an 8.2–11 MHz multifrequency US probe (patients # 16 through 23) were used (General Electric Logic 400 US machine with 7 and 8.2–11 MHz probes; GE Medical Systems Milwaukee, WI).

The spinal column can be penetrated in part by US. The space between the spine processes and the vertebral discs can be visualized as an US-suited window. Other parts such as the spine process or the pedicles of the vertebra, depending on the individual grade of ossification and the corresponding density, produce highly echogenic and bright signals such that other and more distant imaging information is diminished. These US penetrable or unpenetrable parts are cited as US window and US shadow, respectively. The relation of the “window to shadow” decreases from sacral to thoracic regions. Therefore the sacral and lumbar region are best suited for applying US. Neuraxial structures are tissues of different echogenicity. Hyperechogenic tissues (tendons, the dura mater or bony structures) reflect an intense signal and appear as bright on the US screen. Iso-echogenic tissues are of identical density and appear gray whereas hypoechogenic tissues (fluids) reflect as a weak signal and appear black on the US screen.

Depth from skin to neuraxial structures was measured in cm using the relevant “frozen image” and the measuring device of the US machine. The distance from skin level to the ventral side of the ligamentum flavum was estimated and presumed as depth of the loss of resistance (LOR). Longitudinal median and longitudinal paramedian and transversal levels were visualized and stored on SVHS video tapes, paper prints, and magnetic optical storage devices. The visibility of the different structures was described on a four-point scale as follows: 1 = excellent and clearly visible, 2 = visible, 3 = recognizable, and 4 = unrecognizable.

The EC (B. Braun Pediatric Epidural Kit, B. Braun Melsungen, Germany) was inserted for intraoperative and postoperative analgesia using a LOR technique with saline injection. Simultaneous with epidural puncture, a second person using sterile conditions (GE sterile US Toolkit, GE Medical Systems) positioned the US probe at the level of the catheter insertion using a paramedian position. The EC was inserted under US control to visualize catheter tip and position in the longitudinal paramedian plane. After insertion and subcutaneous tunnelling of the catheter, a sterile dressing was placed and the catheter was fixed on the patient's back and left upper arm. To localize the catheter tip at the end of the procedure, the US probe was placed once more onto the sterile foil to determine the final position.

All values are expressed as mean ± sd and range. The level of statistical significance was set to P ≤ 0.05. Statistical analysis was done with Microsoft Excel® 97 or Biostatistics Primer (McGraw-Hill, New York, NY) for Bland-Altman analysis (11,12).

Back to Top | Article Outline


US-guided insertion of EC, site and depth of neuraxial structures were studied in 25 pediatric patients; because of technical problems only 23 were evaluated (seven female and 16 male, with an age between 5 mo and 10 yr). Patient demographics are shown in Table 1.

Catheter insertion was performed at levels L4-5 to T6-7. LOR was reached at depths of 10–28 mm. US-estimated distance from skin to the ligamentum flavum was between 9 and 23 mm (Table 2). The data of US-estimated distance between skin and ligamentum flavum and the corresponding depth of the clinically estimated LOR are listed in Table 2. A correlation of 0.88 could be estimated between US depth of the ligamentum flavum and the measured LOR. The two compared methods showed a high conformity with an accuracy of 2.02 ± 2.03 mm (Fig 1).

At all levels neuraxial structures were visible in a characteristic manner. Ligamentum flavum, dura mater, intrathecal space, spinal cord, nerve roots, and nerve fibers could be clearly distinguished. The dura appeared as bright and highly echogenic and the subarachnoid space filled with cerebrospinal fluid appeared as echo-free and black. The spinal cord with low echogenicity showed a double line at thoracic level and the conus medullaris and cauda equina showed high bilateral structures with high echogenicity, brightness, and pulsatile character. The different layers were evaluated on a 4-point scale (1 = excellent and clearly visible, 2 = visible, 3 = recognizable, and 4 = unrecognizable). Ligamentum flavum and dural cover were identified as 1 or 2. US images of the different levels are shown in Figures 3 through 5. Figure 3 (sacral), Figure 4 (lumbar), and Figure 5 (thoracic) show typical US images in the longitudinal and transversal plane. The area between the spine processes, which is penetrable by US and suited for US application compared with the area of spine processes and pedicles where no US imaging is possible, decreased with increasing spinal level. The ratio of visible to non-visible segments was 2:1 at the sacral level, 1:1 at the lumbar level, and 1:2 at the thoracic level.

LOR could be seen as widening of the epidural space followed by a ventral movement of the dural cover and a compression of the dural sac. (Figures 6 through 8). This could be seen in all patients except the two patients with thoracic insertion in which the application of US was stopped because of technical problems. Catheters were inserted to a final epidural length of 3.5 to 6 cm. In 8 of 23 patients the catheter could be identified and visualized immediately during insertion and threading. In 11 other patients the forward movement and location of the catheter in the epidural space could be visualized using additional US planes; thus in 19 of 23 patients catheter visibility was possible (Table 2 and Fig. 2). An intrathecal position could be excluded by the visibility of the correct epidural puncture and saline injection followed by ventral movement of the dura (Figs. 6 and 7). After the catheter was inserted, the final catheter position was determined in 12 patients using a paramedian plane. In 6 patients the catheter position could be determined when more than 2 US planes were used, whereas in 5 patients the catheter position could not be determined.

Back to Top | Article Outline


This study shows that US is a valuable and reliable tool during insertion of EC in children. Relevant neuraxial structures are visible and thus the depth of the epidural space and the expected distance to LOR can be estimated before insertion. The Bland-Altman analysis (11,12), a statistical method for assessing agreement between two methods of clinical measurement, resulted in a high concordance of the US-measured depth of the ligamentum flavum and the depth in which the LOR occurred.

The study enhances the classical method using LOR by adding visibility of the anatomical structures. This US method was described approximately 20 years ago (6), but recent technical improvements were necessary for clinical use in this field. The paramedian approach and midline puncture enable real-time needle placement (10,13). The technical improvements involving linear digital data transfer up to 15 MHz enable one to visualize neuraxial structures in infants.

In all study patients the ligamentum flavum, the dural structures, and the epidural space, including the location and depth, could be identified. Additionally, in the patients with lumbar approach to insertion of the EC the catheter was visible in 8 and partially visible in another 11 patients.

In pregnant women, studies have demonstrated that US use decreased the number of attempts to puncture the epidural space (13,14). US has also been used for regional anesthesia of peripheral nerves (15,16). The literature on the use of US in pediatric anesthesia is limited (9,15). This present study demonstrates the utility of US for pediatric epidural catheter placement at the lumbar and lower thoracic levels; however, there are some limitations (17,18). In infants younger than 1 year of age the visibility of EC was not possible in all patients. Also, its use and reliability in upper thoracic EC have not been evaluated.

Back to Top | Article Outline


US is a very valuable tool for inserting EC in anesthetized children. Relevant anatomical structures and their location and the corresponding depth can be identified. Additional injection of medications into the epidural space can be visualized. The addition of this real-time procedure makes the insertion of EC in children easier.

Back to Top | Article Outline


1. Bosenberg A. Pediatric regional anesthesia update. Paediatr Anaesth 2004;14:398–402.
2. Williams DG, Howard RF. Epidural analgesia in children: a survey of current opinions and practices amongst UK paediatric anaesthetists. Paediatr Anaesth 2003;13:769–76.
3. Kasai T, Yaegashi K, Hirose M, Tanaka Y. Spinal cord injury in a child caused by an accidental dural puncture with a single-shot thoracic epidural needle. Anesth Analg 2003;96:65–7.
4. Rose JB. Spinal cord injury in a child after single-shot epidural anesthesia. Anesth Analg 2003;96:3–6.
5. Horlocker TT, Abel MD, Messick JM Jr, Schroeder DR. Small risk of serious neurologic complications related to lumbar epidural catheter placement in anesthetized patients. Anesth Analg 2003;96:1547–52.
6. Cork RC, Kryc JJ, Vaughan RW. Ultrasonic localization of the lumbar epidural space. Anesthesiology 1980;52:513–6.
7. Grau T, Leipold RW, Conradi R, et al. Efficacy of ultrasound imaging in obstetric epidural anesthesia. J Clin Anesth 2002;14:169–75.
8. Kapral S, Marhofer P, Grau T. Ultrasound in local anaesthesia. Part I: technical developments and background [in German]. Anaesthesist 2002;51:931–7.
9. Chawathe MS, Jones RM, Gildersleve CD, et al. Detection of epidural catheters with ultrasound in children. Paediatr Anaesth 2003;13:681–4.
10. Grau T, Leipold RW, Horter J, et al. Paramedian access to the epidural space: the optimum window for ultrasound imaging. J Clin Anesth 2001;13:213–7.
11. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;8476:307–10.
12. Bland JM, Altman DG. Comparing methods of measurement: why plotting difference against standards method is misleading. Lancet 1995;346:1085–7.
13. Grau T, Leipold RW, Horter J, et al. The lumbar epidural space in pregnancy: visualization by ultrasonography. Br J Anaesth 2001;86:798–804.
14. Grau T, Leipold RW, Conradi R, et al. Ultrasound imaging facilitates localization of the epidural space during combined spinal and epidural anesthesia. Reg Anesth Pain Med 2001;26:64–7.
15. Retzl G, Kapral S, Greher M, Mauritz W. Ultrasonographic findings of the axillary part of the brachial plexus. Anesth Analg 2001;92:1271–75.
16. Perlas A, Chan VW, Simons M. Brachial plexus examination and localization using ultrasound and electrical stimulation: a volunteer study. Anesthesiology 2003;99:429–35.
17. Dick EA, de Bruyn R. Ultrasound of the spinal cord in children: its role. Eur Radiol 2003;13:552–62.
18. Tsui BC, Wagner A, Cave D, Kearney R. Thoracic and lumbar epidural analgesia via the caudal approach using electrical stimulation guidance in pediatric patients. Anesthesiology 2004;100:683–9.
© 2005 International Anesthesia Research Society