Secondary Logo

Journal Logo

Original Article

Thoracic epidural puncture guided by an acoustic signal: clinical results

Lechner, T. J. M.*; van Wijk, M. G. F.*; Maas, A. J. J.; van Dorsten, F. R. C.*

Author Information
European Journal of Anaesthesiology: September 2004 - Volume 21 - Issue 9 - p 694-699
  • Free


There are many circumstances where thoracic epidural anaesthesia is indicated. However the technique is more difficult to perform than lumbar epidural placement. It is well known that there is a 'learning curve' before one reaches consistency in performing an epidural block [1]. With the thoracic approach, control of the needle is critical, since injury to the spinal cord is possible if the needle is advanced too far [2,3].

There is no standard method to describe the technique, nor is there a validated procedure available to confirm correct catheter placement. Thoracic epidural techniques are somewhat awkward and complication rates may be higher than for lumbar techniques [4,5]. Confirmation of correct catheter placement is especially important when epidural anaesthesia is combined with general or spinal anaesthesia. Studies by Ghia and co-workers, and Tsui and co-workers address this issue [6-8].

In two previous studies we have demonstrated the clinical results of the lumbar epidural puncture with the acoustic puncture assist device, and concluded that the puncture technique with the device in these studies proved to be safe and effective in the lumbar region [9,10].

Before starting a multi-centre randomized prospective trial we wanted to evaluate whether the device is also safe and efficient for use in the thoracic region, and whether it can be used as a practical tool to confirm correct catheter placement.


After approval by the local Institutional Ethics Committee and written, informed patient consent was obtained, 100 consecutive patients, ASA I-III, were included in this study.

Patients were scheduled for thoracic or major abdominal surgery under thoracic epidural anaesthesia combined with general anaesthesia. Patients with known coagulation disorders, hypersensitivity to amide local anaesthetics, skin lesions at the puncture site or neuromuscular disease were excluded from this study. Patient characteristics data (age, height and weight) and type of surgery were noted.

A schematic presentation of the experimental set-up used in this study is shown in Figure 1. The set-up was identical to the one used in our earlier study [10]. The epidural puncture was performed by one of six staff anaesthesiologists in our department. Each of them had previously performed more than 500 epidural procedures, both lumbar and thoracic.

Figure 1
Figure 1:
Experimental set-up for acoustically guided puncture of the epidural space. 1: infusion pump; 2a: polyethylene extension tube; 2b: polyvinylchloride extension tube; 3: three-way stopcock; 4: epidural needle; 5: pressure transducer; 6: pressure amplifier; 7: voltage controlled oscillator; 8: loudspeaker; 9: instrumentation recorder.

In accordance with standard hospital procedures, patients were connected to a monitoring device (AS3; Datex, Helsinki, Finland) to observe electrocardiogram, pulse oximetry and non-invasive blood pressure (BP). A large-bore intravenous (i.v.) cannula was inserted and a preload of a colloid solution (15 mL kg−1) was given.

Under aseptic conditions and with the patient in the sitting position, local infiltration of the skin with lidocaine 2% was performed at the level of the chosen thoracic vertebral interspace. An 18-G epidural needle (Braun Perican; Melsungen A.G., Melsungen, Germany) was inserted 1-2 cm via the midline approach.

After the stylet was withdrawn, the needle was flushed with saline, and the experimental device was attached to the needle via a 120-cm long polyvinyl chloride tubing. The infusion pump (rate set at 100 mL h−1), the sound amplifier and the instrumentation recorder of the experimental set-up were then activated. Kinking the extension tube, which led to a clearly audible increase in pitch tone and a clear rise in the pressure, tested the system.

Since the infusion rate was kept constant, the pressure in the system was solely caused by the density of the tissue in which the needle tip was at any given moment. Pressure changes resulted in corresponding variations in acoustic and visual signals; i.e. increasing pressure gave rise to a higher pitched tone and an upward deflection of the pressure gauge and vice versa. In this way the passage of the needle tip from tissues of high resistance (ligamentum flavum) to those of low resistance (epidural space) was made audible and visible.

Holding the epidural needle with both hands, the anaesthesiologist moved the needle in a continuous forward direction toward the epidural space. The moment the anaesthesiologist heard a distinct fall in pitch tone, preceded by an initial rise, needle advancement was stopped, and the pressure curve was checked. If the record confirmed the aural signal of a reduction in pressure, the apparatus was disconnected from the needle, and a 20-G epidural catheter (Braun Perifix) was inserted 3 cm cephalad.

It is the policy in our hospital not to insert an epidural catheter in an unconscious patient. In this study we did it once, at the specific request of a 15-yr-old patient scheduled for splenectomy and nephrectomy in the same session. The epidural puncture was performed after induction of anaesthesia.

In 10 patients we also connected the apparatus to the epidural catheter to record the pressure curve of the catheters we believed to be correctly placed with the pump running at 100 mL h−1. We recorded this pressure and disconnected the apparatus again.

After negative aspiration, a test dose of a local anaesthetic was administered through the catheter and, 3 min later, the complete dose was given. The choice of local anaesthetic and dose dependent on the personal preference of the practising anaesthesiologist.

Thirty minutes later, the sensory level (pinprick) and motor block (Bromage scale) were tested. The procedure was considered to be a clinical success if there was an adequate sensory block and no supplemental i.v. analgesics were needed during surgery, i.e. there were no unacceptable changes (>15%) in heart rate and BP, nor were there other signals of intraoperative stress.

Induction of anaesthesia consisted of a propofol infusion, an i.v. bolus injection of a muscle relaxant and an i.v. bolus injection of sufentanil 10 μg (Janssen-Cilag B.V.; 5000 LT Tilburg, The Netherlands) to facilitate tracheal intubation. Ventilation was controlled artificially and anaesthesia was maintained with propofol administered by infusion. Bolus injections of muscle relaxant were administered as required. Just before skin incision sufentanil 20 μg was administered via the epidural catheter and a continuous epidural infusion with a mixture of bupivacaine hydrochloride 2.5 mg mL−1 and sufentanil 1 μg mL−1 was started (3-4 mL h−1).

For each patient, the efficacy of the device in localizing the epidural space (good, normal, insufficient or worthless), the usefulness of the acoustic signal (indicative or misleading), the need for supplemental analgesics, the tactile perception of the ligamentum flavum and any observed complications were recorded.


In the studied population the median age was 63 yr (range 15-86), median height 173 cm (range 155-189) and median weight 73 kg (range 45-115). Of the procedures 42 were thoracic, 58 were major abdominal (of which 45 were general surgery), 10 urologic and 3 gynaecologic. Fourteen patients were classified as ASA I, 52 as II and, 34 as III.

In all patients the anaesthesiologist detected the epidural space by means of the acoustic signal and correct identification was confirmed by the pressure monitor. An example of such a recording is given in Figure 2. An epidural block with adequate analgesia was achieved in all patients which verified correct placement of the epidural catheter. None of the patients needed additional i.v. analgesics during surgery.

Figure 2
Figure 2:
Pressure recording produced by the experimental device showing the test signal (1), start of needle advancement (2), needle withdrawal because of needle contact to bone and start of repeated needle advancement (3), a pressure decrease at entering the epidural space (4) and a low-pressure plateau (4 and 5) due to the free flow of saline.

In two patients the anaesthesiologist had to perform a puncture at three different levels, because only bony structures were encountered on the first two attempts or the epidural catheter could not be inserted on the first two attempts.

In four patients, two punctures were needed because it was not possible to pass between the vertebrae on the first attempt in three of them. In the fourth patient, the catheter was withdrawn too far after initial intravascular placement (Fig. 3). Eventually in all patients the epidural space was located solely by acoustic guidance.

Figure 3
Figure 3:
Pressure recording produced by the experimental device showing an ongoing rise in pressure (1) after withdrawal of the catheter, indicating no free flow of saline, i.e. the tip of the catheter is no longer in the epidural space.

In one patient the epidural catheter was still in the epidural space after a 2-cm withdrawal, following intravascular placement (Fig. 4). Figure 4 is also representative of the recordings we made for the 10 patients to whom we connected the apparatus to the epidural catheter.

Figure 4
Figure 4:
Pressure recording produced by the experimental device showing a slow increase of the pressure caused by the small diameter of a 20-G catheter, followed by a plateau (1) due to free flow of saline, suggesting that the tip of the catheter is in the epidural space.

In the case of the patient in whom we performed the epidural puncture after induction of anaesthesia, we tested the block after operation. Catheter placement was uneventful.

In 50 patients the ligamentum flavum was never felt, but was detected by the acoustic signal.

In all patients the efficacy of the device was scored as 'good' and the value of the acoustic signal was classified as 'indicative'. Except for intravascular placement of the catheter in two patients, no other complications were recorded.


In two previous studies we chose to evaluate the efficacy of the acoustic device in the lumbar vertebral region in order to minimize the risk of damage to the spinal cord [9,10]. Since the results of these studies were positive, we were encouraged to extend the use of this new puncture technique. We chose the thoracic epidural puncture, which is normally restricted to experienced anaesthesiologists because of the possible serious complications.

All of the anaesthesiologists participating in this study were experienced, and during residency they were all trained to use the midline approach, advancing the needle in a constant forward direction; however they had not extensively used the acoustic device before. Nevertheless the only recorded complication in 100 consecutive patients was intravascular placement of the catheter in two patients.

This result suggests that the acoustically guided technique is safe for use in the thoracic region. None of the patients needed additional i.v. analgesics, meaning that they had an adequate epidural block. Therefore in all cases the fall in pitch tone was indicative, and the documented pressure a reliable indicator of correct identification of the epidural space (Fig. 2).

We believe, based on experience in our own practice, that the need to perform two or three punctures in six patients before reaching the epidural space would have been similar if we had used one of the conventional techniques.

The techniques most used for the thoracic epidural puncture are the 'hanging drop', and 'loss of resistance' techniques, either by median or para-median approach [11,12]. A disadvantage of the first is that the needle can be obstructed, in which case the drop of liquid no longer functions as a vital warning signal. In the thoracic region the passage between the spinous processes of the vertebrae is very steep, and often long. Therefore absolute needle control is required, which is not always possible when the loss of resistance technique is used, because only one hand is free to hold the needle, the other being needed to search for the loss of resistance.

With the acoustic device the anaesthesiologist can handle the epidural needle with both hands, and localize the epidural space by means of an acoustic signal. It is well known that the sense of hearing is better suited to detect small changes, than the sense of touch [13]. This fundamental concept was confirmed in the 50 patients in whom the anaesthesiologist did not feel the passage through the ligamentum flavum, but was warned by the change in pitch tone.

Also a supervising anaesthesiologist can hear the acoustic signal. Due to this, the acoustic guided puncture seems to be a useful tool in teaching procedures.

In contrast to the 'hanging drop' technique, the acoustic device can be used in both the thoracic and lumbar regions, and the constant flush prevents the needle from becoming obstructed.

The fact that an expected pressure trace was found in 10 correctly placed catheters is useful information. Although the cohort in this study is rather small, the trace might be used in future to confirm correct catheter placement.

Studies by Ghia and co-workers, and Tsui and co-workers address the issue of control of correct catheter placement [6-8]. In 10 patients Ghia connected the end of the epidural catheter, after presumed placement in the epidural space, to a disposable pressure transducer and injected 5 mL normal saline. He compared the pressure record of this bolus injection with localization by computed tomography (CT) after contrast injection through the catheter in the same patient. He concluded that there was a strong relationship between pressure recorded and CT, and therefore that the pressure changes can be used reliably to confirm correct catheter placement.

Tsui also perceived the need for a reliable method to confirm correct epidural catheter placement. He used electrical stimulation of a catheter especially constructed for this purpose. It is possible that the acoustic device used in the present study, when further developed from the present prototype, will be a simple and inexpensive way to identify and document correct position of thoracic epidural catheters.

None of the conscious patients complained of pain during the puncture procedure, which suggests that underlying neural structures were not traumatized. In this study every puncture was successful, including one performed on an unconscious patient. We are aware that practices and policies regarding catheter placement in unconscious or conscious patients vary from country to country [14]. Some experts believe that epidural insertion in an anaesthetized patient affords safer conditions and facilitates placement [15]. Since the puncture technique with the acoustic device proved to be accurate in this study, and the procedure can be documented, it would be interesting to conduct a study in which the device is used with unconscious patients. Many patients would be grateful if this possibility was offered to them.

In this study we found no disadvantages of the technique. We realize, however, that experienced anaesthesiologists would possibly have had the same results if one of the conventional techniques had been used.

From this cohort study we conclude that thoracic epidural puncture using the acoustic device is safe, simple and reliable. Potential advantages of the technique include greater stability in needle handling, which is of special importance in the thoracic region; continuous needle flush, which prevents obstruction; and confirmation and documentation of correct identification of the epidural space and subsequent catheter placement. The auditory signal is a useful tool in training situations, because the signal is audible to trainer and trainee alike.

The results of this study encourage us to proceed with further a randomized multi-centre trial with well-defined end-points in order to compare the acoustic device with conventional techniques.


The authors thank Ms. D. Brashear for her assistance in preparing the manuscript. The general manager of the operating room gave written permission to use a disposable pressure transducer for each procedure. There were no further sources of financial support. The authors hold a patent on the acoustic puncture assist device.


1. 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 Pain Med 1996; 21: 182-190.
2. Kasai T, Yaegashi 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: 65-67.
3. Horlocker TT, Wedel DJ. Neurological complications of spinal and epidural anesthesia. Reg Anesth Pain Med 2000; 25: 83-98.
4. McLeod GA, Davies HTO, Munnoch N, Bannister J, Macrae W. Postoperative pain relief using thoracic epidural analgesia: outstanding success and disappointing failures. Anaesthesia 2001; 56: 75-81.
5. Burstal R, Wegener F, Hayes C, Lantry G. Epidural analgesia. Prospective audit of 1062 patients. Anaesth Intens Care 1998; 26: 165-172.
6. Ghia JN, Arora SK, Castillo M, Mukherji SK. Confirmation of location of epidural catheters by epidural pressure waveform and computed tomography cathetergram. Reg Anesth Pain Med 2001; 26: 337-341.
7. Tsui BC, Gupta S, Finucane B. Confirmation of epidural catheter placement using nerve stimulation. Can J Anaesth 1998; 45: 640-644.
8. Tsui BC, Guenther C, Emery D, Finucane B. Determining epidural catheter location using nerve stimulation with radiological confirmation. Reg Anesth Pain Med 2000; 25: 306-309.
9. Lechner TJM, van Wijk MGF, Maas AJJ. Clinical results with a new acoustic device to identify the epidural space. Anaesthesia 2002; 57: 768-772.
10. Lechner TJM, van Wijk MGF, Maas AJJ, et al. Clinical results with the acoustic puncture assist device, a new acoustic device to identify the epidural space. Anesth Analg 2003; 96: 1183-1187.
11. Guttierez A. Anesthesia metamerica peridural. Rev Cir Buenos Aires 1932; 12: 665-685.
12. Dogliotti AM. A new method of block anaesthesia. Segmental peridural spinal anaesthesia. Am J Surg 1933; 20: 107-118.
13. Blamey PJ, Cowan RS, Alcantara JI, Whitford LA, Clark GM. Speech perception using combinations of auditory, visual, and tactile information. J Rehabil Res Dev 1989; 26: 15-24.
14. Romer HC, Russell GN. A survey of the practice of thoracic epidural analgesia in the United Kingdom. Anaesthesia 1998; 53: 1016-1022.
15. Minzter BH, Johnson RF, Grimm BJ. The practice of thoracic epidural analgesia: a survey of academic medical centers in the United States. Anesth Analg 2002; 95: 472-475.


© 2004 European Academy of Anaesthesiology