Real-time ultrasonic observation of combined spinal-epidural anaesthesia : European Journal of Anaesthesiology | EJA

Secondary Logo

Journal Logo

Original Article

Real-time ultrasonic observation of combined spinal-epidural anaesthesia

Grau, T.; Leipold, R. W.; Fatehi, S.; Martin, E.; Motsch, J.

Author Information
European Journal of Anaesthesiology: January 2004 - Volume 21 - Issue 1 - p 25-31
  • Free


Combined spinal and epidural analgesia (CSE) has a rapid onset and highly reliable regional blockade with the possibility of prolonging the duration of analgesia. In addition, CSE is an effective way to reduce the total drug dosage required for anaesthesia. The CSE technique is in widespread use, particularly for major orthopaedic surgery and in obstetrics.

The quality of the block is heavily dependent on accurate cannulation of the epidural space and on correct placement of the epidural catheter. Complications of CSE, such as insufficient analgesia, paraesthesia, neurotrauma, postdural puncture headache and infection, are strongly associated with the puncture technique. Identification of the epidural space has traditionally been achieved by a blind, external landmark-guided mode: the loss of resistance method. Problems such as axis deviation of the needle trajectory-subsequent missing of the dural sac - are often encountered. Furthermore, the handling of the needle may cause complications and spinal component failure after successful dural puncture [1]. As the feedback to the operator is merely tactile, excessive needle movement - and dislocation of the needle - may occur at various points during the CSE procedure, e.g. when the stylet of the spinal needle is removed, at the subsequent connection of the syringe or during the injection of the local anaesthetic.

To facilitate the CSE procedure and avoid needle dislocation, a monitoring facility is desirable. Tsui and colleagues proposed the use of nerve stimulators to confirm the position of the needle tip during the puncture procedure [2]. Based on experiences with ultrasound imaging of the spine before cannulation of the epidural space [3-6], the authors tried to establish a less invasive method for real-time needle monitoring. The depth of the epidural space was measured (i.e. the distance from the skin to the ligamentum flavum) using ultrasonography before the epidural block was attempted. During the puncture, the aim was to monitor the advancement of the needle in real time. The objective was to detect any differences during the processes of puncture, structure detection and the quality of the puncture through imaging techniques.


The local hospital Ethics Committee granted approval for the study and written informed consent was obtained from all patients studied. Thirty parturients scheduled for Caesarean section with analgesia provided by a combined spinal-epidural technique were prospectively randomized into three equal groups by a closed envelope technique. Because of the very extensive diagnostic procedures, especially in the ultrasound online group, the numbers of patients able to be studied were relatively small. In the ultrasound group, the appropriate lumbar segments were monitored before and during puncture of the epidural space.

A Logiq 400® (General Electric, Solingen, Germany) ultrasound system equipped with a 7.5 MHz linear array probe was used as the ultrasonic scanner. Distances and angles were measured by the built-in distance and angle program. Scanning was performed in transverse and longitudinal planes, and real-time monitoring was done in a longitudinal plane using a paramedian access.

All CSE punctures were performed by the same anaesthetist (T. G.; about 2000 epidural blocks) and the online punctures were assisted by a colleague. The puncture was performed using the adjustable Durasafe® set (Becton Dickinson, Heidelberg, Germany). A single-space, needle-through-needle CSE technique together with the standard loss of resistance to saline method was used in all patients. The Tuohy needle was inserted using the midline approach. No unconventional measures were taken in the control group. In the offline group, the ultrasound images were taken just before the puncture to find an improved needle trajectory and improve the puncture process with the knowledge of the optimum puncture point, puncture angle and puncture depth. Visibility was judged on a four-point scale of 0-3, with 0 = none, 1 = insufficient, 2 = sufficient and 3 = very good.

In the online ultrasound group, the patients were placed in a sitting position and the skin over the L3-4 vertebral interspace was disinfected and anaesthetized. Sterile sleeves and sterile ultrasound gel (General Electric) was used before the ultrasound images were taken for online documentation of the puncture. The optimal depth and needle trajectory for the cannulation of the epidural space were identified and cannulation was monitored in real time. In these cases, the transducer had to be fixed by a helper to ensure the quality of imaging and puncturing.

In all groups, the position of the spinal needle tip in the subarachnoid space was confirmed by dural puncture ('dural click') and backflow of cerebrospinal fluid. The injection of the drug into the theca was followed by the placement of the epidural catheter through the Tuohy needle. Plain bupivacaine 0.5% 1.5 mL was used initially, followed by plain bupivacaine 0.5% 7.5 mL and sufentanil 10 μg for the epidural analgesia through the catheter.

Every ventral advancement of the needle was considered as a 'puncture attempt' even if no further skin puncture was performed, i.e. all redirections of the needle, which were usually considered as part of one seeking attempt, were counted as separate attempts. This convention seemed necessary, as otherwise a successful puncture attempt could mean a 'hole-in-one' a few seconds after skin puncture as well as a lengthy procedure with a dozen relocations of the needle. If the epidural space could not be cannulated at the first puncture site, a neighbouring vertebral interspace was punctured. The number of puncture attempts was recorded, as well as the number of puncture sites.

The onset time of the epidural analgesia was measured from the beginning of the puncture to the first signs of blockade. The time to complete motor block (according to Bromage), the side-effects of sensory blockade, incomplete analgesia and the intraoperative need for additional analgesics were surveyed. Analgesic efficacy was assessed using visual analogue scores (VAS). The onset time of CSE, from injection to the first signs of blockade or to complete block, and the time for the preparation of the patient were measured. Patient preparation time included all necessary measures, such as that for the ultrasound examination. However, no other ultrasound-related actions, e.g. the start time of the ultrasonograph's computer, were included.

The depth of the epidural space (i.e. the distance from the skin to the ligamentum flavum), as seen on the ultrasound images, was compared with the depth measured by the loss of resistance method. The precision of the offline ultrasound examination was calculated. Patient satisfaction with analgesia during Caesarean section was assessed using the VAS pain score. Patients were asked for their satisfaction by using a six-point verbal score, where 1 = very good, 2 = good, 3 = average, 4 = sufficient, 5 = unsatisfactory and 6 = insufficient. All patients were interviewed for any side-effects from the CSE.

Data were analysed using the χ2-test with Yates' correction where appropriate; linear intergroup differences were analysed using a t-test. For statistical analysis, Excel 97® (Microsoft, Redmond, WA, USA), SPSS 7.5® (SPSS Inc., Chicago, IL, USA) and Primer Biostatistics 5.0® (McGraw-Hill Professional, New York, NY, USA) software were used. Data are the mean ± SD unless otherwise stated.


Thirty patients receiving CSE in the three different study groups were examined. In these groups, patient characteristics data such as age, body weight, height and gestational age were very similar (Table 1). The preparations took about 4 min in the control group. In the ultrasound on- and offline groups, the ultrasound scanning added 1-2 min at most to the preparation time. The elapsed time from the injection to the first effects or complete block showed no significant differences between the three groups. Similar distances were found between the skin surface and the ligamentum flavum in all groups. The 95% precision (Bland-Altman) for ultrasound-measured depth correlated to puncture depth was calculated as precision = 0.786 mm (mean 1.245 mm) and for the ultrasound measured angle precision was 3.221° (mean 2.4°) (Table 2).

Table 1:
Patient characteristics.
Table 2:
Qualities of regional anaesthesia.

The online group was monitored in real-time ultrasound. In this context, interesting imaging aspects were found: with a diameter < 2 mm, the Tuohy and spinal needles are very slim and usually have a rounded and polished surface. Nevertheless, both types of needles could be detected by the ultrasound scans: a gentle rotation of the longitudinal axis of the needle (so that the aperture of the Tuohy needle or of the eye of the Whitacre spinal needle was repositioned) moved its tip into the midline axis of the ultrasound so as to generate a distinct echo. Dural puncture was observed in all the patients and dural tenting was seen in nine patients. It was possible to survey the complete procedure of needle placement through the paramedian acoustic window (Figs 1 and 2). Spinal needle placement and the application of intrathecal anaesthetics could be detected very easily.

Figure 1:
Native paramedian ultrasound image with the Tuohy needle placed epidurally and the CSE spinal needle placed intrathecally.
Figure 2:
Localization of the needles and structures.

However, epidural catheter placement could not be monitored completely. The echo structure of the epidural catheter was very similar to that of the epidural space itself. Hence, there was little chance to see the catheter placement at the tip of the Tuohy needle, and the complete placement could not be demonstrated with the conventional catheter material. In the present set-up, the ultrasonic resolution was not effective enough for reliable discrimination. The epidural application of the local anaesthetics into the epidural space caused hardly detectable effects because there were often semiechogenic effects. There was no visible depot and only a small variation or movement in the visible structures could be found. In five cases, a constant flow of local anaesthetic solution into the epidural space could be observed during injection. The movement of the dura caused by respiration did not alter the injection. The continuity of the membrane structures remained intact.

Anatomical landmarks such as spinous processes, transverse processes and facet joints are impermeable to ultrasound waves and were identified easily. The ultrasound beam entered the spinal column through the supra- and interspinous ligament, it being the 'acoustic window' between the vertebrae. The ligamentum flavum was the first strongly echogenic structure on this trajectory. Moreover, the dura mater could be identified in all cases.

At the level of the L3-4 vertebral interspace, the distances between the skin surface and ligamentum flavum varied between 41 and 75 mm. With ultrasound imaging, the epidural space could be positively identified in all cases. In both ultrasound groups, a significant reduction in the number of necessary puncture attempts was found (P < 0.036). In the ultrasound online group, 10 attempts to access the epidural space were successful on the first try and in the first puncture level (compared with four in the control group) (P < 0.036) (Tables 3 and 4).

Table 3:
Puncture attempts performed with the Tuohy needle in the control and ultrasound offline and online groups.
Table 4:
Number of spinal needle manipulations until successful intraspinal (cerebrospinal fluid) access.

The number of different interspaces punctured before successful cannulation of the epidural space was only significantly lower in the ultrasound online group (P < 0.036). There was no need to change the puncture site in any of these cases, while a change of the puncture site was required five times using the standard technique (Table 5).

Table 5:
Numbers of first level successes with the Tuohy needle.

During the operation, maximum VAS pain score was 0.7 ± 0.94 in the control group, 0.3 ± 0.94 in the offline group and 0.3 ± 1.63 in the online ultrasound group (n.s.). Patient satisfaction with epidural analgesia was rated as 1.65 ± 0.74 versus 1.35 ± 0.57 in the offline group and as 1.2 ± 0.42 in the online ultrasound group (n.s.).

Asymmetrical blockade was observed in 10% of those in the control group, but not in any of those in the ultrasound groups; patchy anaesthesia was seen in 10 and 0% of patients, respectively. The asymmetric spread of local anaesthetic (one in the control group) and in the occurrence of headache or backache (two in the control group) did not occur in the ultrasound groups; the χ2-test results were not significant for these features.

The incidence of side-effects from CSE was very low in this study. One patient in the offline ultrasound group (compared with two patients in the control group) complained of mild headache after surgery. Four patients suffered from a light backache after operation (one in the ultrasound group, three in the control group); here, the χ2-test did not reveal any statistically significant differences between the two groups.


The accurate identification of the epidural space is a most vital part of the combined spinal epidural anaesthesia technique. CSE failure is often related to a faulty puncture site or axis deviation during needle advancement. The aim was to develop a reliable way to visualize the puncture process itself using ultrasonography. In this context, of special interest was the influence of the imaging techniques on the puncture process and the visualization of dural tenting and the real-time observation of needle advancement. Furthermore, of interest was the evaluation and validation of the visibility of the epidural structures under online puncture conditions.

A 95% precision rate (Bland-Altman) of precision = 0.7 mm was found when the offline method was correlated with the ultrasound method when measuring the depth of the epidural space from the puncture site. Earlier measurements of the depth of the epidural space [7-10] have suggested that the distance from the skin to the epidural space in the present study ranged from 39 to 95 mm. Apparently, the prepuncture ultrasound method reduced the area where the ligamentum flavum was to be expected by a factor of 30-100, which is an extremely good value compared with former findings [4].

Patient preparation time was only slightly altered by the use of ultrasonography in either the off- or online techniques. Purposefully, the time for logistical measures and general preparation of the ultrasonograph (e.g. the start-up time of the computer) was not included. If these measures delay the performance of a CSE, the problem is organizational and unrelated to the ultrasound technique itself.

A positive effect on the rate of puncture attempts by using ultrasound imaging was found, as was a reduced number of secondary anatomical levels required to place the Tuohy needle correctly in the epidural space. The number of manipulations with the spinal needle and with the imaging techniques was reduced significantly. The use of ultrasound imaging was obviously helpful in finding the ideal needle trajectory and to improve puncture conditions by demonstration of the relevant anatomy. With the ultrasound data in both groups, the puncture process could be easily managed under the aspect of the inter-individual variation of the structures. The online technique also helped in the visualization of the epidural manipulations of the Tuohy and spinal needles. Nevertheless, there are limitations to this technique because of reflections of the ultrasound beam on calcified structures (e.g. facet joints) and the epidural catheters themselves. The visualization of the distribution of epidural medication was limited by the isoechogenic imaging of the three variables: catheter density echo from the epidural space itself and the echo from the applied local anaesthetic solution.

Ultrasound examination of the spinal column before CSE puncture was found to be quite promising in an earlier study [4]. However, the value of prepuncture findings is limited. When the depth of the epidural space is depicted by ultrasound, the shortest trajectory is measured. However, this 'ideal' trajectory is seldom achieved during blind needle placement. Furthermore, the passage of the blunt Tuohy needle through tissues generates high pressure [11]. The ensuing tissue deformation accounts for a difference of a few millimetres in the depth of the epidural space measured by ultrasound and needle, respectively. The use of real-time ultrasound examination in the present study eliminated this concern.

Recent studies have shown that the combined spinal-epidural technique is gaining in popularity [12,13]. It is widely agreed that neuraxial anaesthesia techniques should only be used by experienced practitioners [12,14-18], as the combined technique introduces the potential side-effects of each method. So far, approximately 20-25 procedures are necessary before any improvement in the 'blind' techniques of spinal and epidural anaesthesia are seen [19,20]. As the visual feedback makes the identification of vital landmarks significantly easier, ultrasound control may even accelerate the first steps of the learning process. The highest incidence of repeated 'blind' needle probing is a problem for beginners. Every attempt carries the risk of infection, haematoma and neurological damage. As the number of cannulation attempts were reduced, ultrasonic guidance may help to enhance the safety of this 'multi-compartment technique' [12].

Real-time ultrasonography of the epidural space is associated with good operating conditions, high imaging quality and high maternal satisfaction levels as it facilitates puncture of the epidural space and does not cause any discomfort. The number of examinations in this study was relatively small. Although the primary aim was the evaluation of the clinical usefulness of online imaging, it was also found that the ultrasound online procedures showed a more extensive diagnostic preparation compared with the offline techniques. Online techniques were more cost-intensive since the ultrasound equipment (gel and probe) was used under sterile conditions. In clinical practice, the use of the online technique may require a helper to fix the ultrasound probe; this might count as a disadvantage.

As ultrasonic real-time surveillance facilitates the performance of CSE, it enhances its quality. It provides essential and reliable information about the localization of the epidural space, its preceding structures and the actual procedure of needle placement. The individual extent of dural tenting and the intrathecal puncture process can be observed so that spinal failure can be avoided. No other diagnostic method can provide such 'on the job' feedback to the operator. In the future, real-time ultrasound control may prove useful for the development and testing of puncture equipment.


1. Collis RE, Baxandall ML, Srikantharajah ID, Edge G, Kadim MY, Morgan BM. Combined spinal epidural (CSE) analgesia: technique, management, and outcome of 300 mothers. Int J Obst Anesth 1994; 3: 75-81.
2. Tsui BC, Gupta S, Finucane B. Determination of epidural catheter placement using nerve stimulation in obstetric patients. Reg Anesth 1999; 24: 17-23.
3. Grau T, Leipold RW, Conradi R, Martin E, Motsch J. Ultraschall und Periduralanästhesie: Technische Möglichkeiten und Grenzen einer diagnostischen Untersuchung des Periduralraums. Anaesthesist 2001; 50: 94-101.
4. Grau T, Leipold RW, Conradi R, Martin E, Motsch J. Ultrasound imaging facilitates localization of the epidural space during combined spinal-epidural anesthesia. Reg Anaesth 2001; 26: 66-67.
5. Grau T, Leipold RW, Conradi R, Martin E. Ultrasound control for presumed difficult epidural puncture. Acta Anaesthesiol Scand 2001; 45: 766-777.
6. Grau T, Leipold RW, Horter J, Conradi R, Martin EO, Motsch J. The paramedian access to the epidural space: the optimum window for ultrasound imaging. J Clin Anesth 2001; 13: 213-217.
7. Chen KP, Poon YY, Wong CH. The depth to the epidural space. Ma Tsui Hsueh Tsa Chi 1989; 27: 353-356.
8. Harrison GR, Clowes NW. The depth of the lumbar epidural space from the skin. Anaesthesia 1985; 40: 685-687.
9. Hollway TE, Telford RJ. Observations on deliberate dural puncture with a Tuohy needle: depth measurements. Anaesthesia 1991; 46: 722-724.
10. Sutton DN, Linter SPK. Depth of extradural space and dural puncture. Anaesthesia 1991; 46: 97-98.
11. Telford RJ, Hollway TE. Observations on deliberate dural puncture with a Tuohy needle: pressure measurements. Anaesthesia 1991; 46: 725-727.
12. Burnstein R, Buckland R, Pickett JA. A survey of epidural analgesia for labour in the United Kingdom. Anaesthesia 1999; 54: 634-640.
13. Cook TM. Combined spinal-epidural techniques. Anaesthesia 2000; 55: 42-64.
14. Renck H. Neurological complications of central nerve blocks. Acta Anaesthesiol Scand 1995; 39: 859-868.
15. Horlocker TT, McGregor DG, Matsushige DK, Schroeder DR, Besse JA. A retrospective review of 4767 consecutive spinal anesthetics: central nervous system complications. Perioperative Outcomes Group. Anesth Analg 1997; 84: 578-584.
16. Horlocker TT, McGregor DG, Matsushige DK, Chantigian RC, Schroeder DR, Besse JA. Neurologic complications of 603 consecutive continuous spinal anesthetics using macrocatheter and microcatheter techniques. Perioperative Outcomes Group. Anesth Analg 1997; 84: 1063-1070.
17. Standl T, Wilhelm S, Gnirke A, Huneke B. Recent developments in obstetric regional anesthesia. A review of experiences at the Eppendorf University Hospital. Anasthesiol Intensivmed Notfallmed Schmerzther 1999; 34: 648-658.
18. Albrecht DM. Zentrale und periphere Leitungsanästhesien - Reevaluierung und Differenzierung nötig. Anaesthesist 2000; 49: 493-494.
19. Kestin IG. A statistical approach to measuring the competence of anaesthetic trainees at practical procedures. Br J Anaesth 1995; 75: 805-809.
20. Kopacz DJ, Neal JM, Pollock JE. The regional anesthesia 'learning curve'. What is the minimum number of epidural and spinal blocks to reach consistency? Reg Anesth 1996; 21: 182-190.

ANAESTHESIA, CONDUCTION, anaesthesia epidural, anaesthesia spinal; DIAGNOSTIC IMAGING, ultrasonography; SPINAL CANAL, epidural space

© 2004 European Academy of Anaesthesiology