A prototype axial ultrasound needle guide to ... - Wiley Online Library

Anaesthesia 2014, 69, 746–751

doi:10.1111/anae.12675

Original Article A prototype axial ultrasound needle guide to reduce epidural bone contact G.-S. Chen,1 Y.-C. Chang,2 Y. Chang3 and J.-S. Cheng4 1 Assistant Investigator, 4 Research Assistant, Institute of Biomedical Engineering and Nanomedicine, National Health Research Institutes, Zhunan, Taiwan 2 Masters Student, 3 Professor, Institute of Biomedical Engineering, National Yang-Ming University, Taipei, Taiwan

Summary We have developed an ultrasound probe through the centre of which an epidural needle can pass, intended to reduce the rate of contact between bone and needle during epidural insertion. We tested the ability of this probe to identify the lumbar interspace, using A-mode ultrasound, in a submerged plastic model, a porcine phantom and five human volunteers. In the plastic model, the minimum echo representing the interspace was only 8.8% of the maximum echo from the ‘bone’. In the porcine model, the echo variations between the interspace and L3 were up to 48% and the needle was safely inserted into the interspace without bone contact under guidance. The human study also showed that the maximum bone echo was at least three times stronger than the interspace echo. Axial ultrasound guidance, with the needle passing through the probe, offers a method for reducing bone contact during epidural insertion. .................................................................................................................................................................

Correspondence to: G.-S. Chen Email: [email protected] Accepted: 6 March 2014

Introduction Pain and failed epidural catheterisation can be caused by needle contact with vertebrae. Ultrasound imaging can reduce vertebral contact, bleeding and failure during epidural catheterisation, both in children and obese adults [1, 2, 4, 5]. Trauma and failure to catheterise the epidural space are reduced from about 7% with palpation to about 1% under ultrasound guidance, accompanied by a reduction in the number of attempts and needle redirections [3]. More precise identification of the epidural space might increase these benefits further. A two-wavelength fibreoptic-guided insertion technique has been developed [6]. Optical fibres built into the hollow stylet of a Tuohy needle enabled clear distinction between the ligamentum flavum and the epidural space. In 746

addition, a novel epidural needle with an embedded high-frequency ultrasonic transducer allowed indentification of the porcine epidural space using A-mode signals from the ligamentum flavum and the dura mater [7]. Ultrasonographic imaging and epidural needle insertion were initially performed by two people [1], an inconvenient and time-consuming process. One person alone can insert an epidural needle and catheter under ultrasound guidance, although fatigue can compromise the technique [8]. An off-axial needle guide, with the needle tilted relative to the probe, facilitates the procedure [9]. We have developed an ultrasound probe through the centre of which the epidural needle passes, with the intention of reducing the rate of vertebral contact. © 2014 The Association of Anaesthetists of Great Britain and Ireland

Chen et al. | Prototype ultrasound needle guide to reduce epidural bone contact

Anaesthesia 2014, 69, 746–751

Methods Figure 1 shows the 2-mm diameter hole through the centre of the 10-mm diameter prototype ultrasound probe, that accommodates a 17-G epidural needle (Arrow; Teleflex Incorporated, Limerick, PA, USA). The piezoelectric transducer was made of ceramic PZT-5A. Table 1 lists the specifications of the probe [10], all of which comply with the US Food and Drug Administration’s safety guidance [11]. The interface between bone and soft tissue is distinguished with Amode ultrasound [12]. The acoustic impedances of cortical bone and soft tissue are 7.38 and 1.63 MRayl, respectively [13]. As a result, the amplitude of the returned echo from the bone is much higher than that from the non-bony tissue. We tested whether we could distinguish between high-density vertebrae and low-density spinal cord in a plastic model of the lumbar spine. We used a custommade system to position the probe precisely along the lumbar model (Fig. 2), which permits up–down (ventrodorsal) and forward–backward (craniocaudal) movement of the probe in increments of 0.25 and 0.5 mm,

Figure 2 The custom-made positioning system for phantom experiments, consisting of a water tank, a pair of linear guide rails for forward–backward movement, a platform and rotation mechanism for up– down motion, an arc mechanism for rotation of the probe, and a probe holder. respectively. We used an ultrasound pulser-receiver in all experiments (5072PR; Olympus, Waltham, MA, USA). Echo signals were displayed on an oscilloscope (TDS 3032B; Tektronix, Beaverton, OR, USA) and saved in the computer for processing. All of the measurements were performed in pure degassed water. Distance was estimated by calculating the product of the sound velocity in water (1500 m.s 1) and the ultrasound travelling time from the front surface of the probe to the reflector. We compared ultrasonographic distances with the average of five calliper measurements. Figure 3 shows the sagittal view of the lumbar porcine phantom. We recorded signals from four

Figure 1 The prototype ultrasound probe and circuit. The epidural needle is passed through the hole in the centre of the probe. One port of the matching circuit is connected to the probe and the other port is linked to the pulser/receiver. Table 1 Specifications of the ultrasound probe.

Centre frequency; MHz 3 dB bandwidth; MHz Pulse width; ls Maximum negative pressure; MPa Mechanical index Spatial peak pulse average intensity; W.cm 2 Spatial peak temporal average intensity; mW.cm

2

© 2014 The Association of Anaesthetists of Great Britain and Ireland

2.02 0.82 1.16 0.43 0.29 0.15 1.30

Figure 3 Cross-sectional view of the pig phantom. The echo signals were: 1, skin-fat; 2, fat-fat; 3, fat-ligament; 4, ligament-bone or ligament-muscle. 747

Anaesthesia 2014, 69, 746–751


interfaces: the skin and fat; two layers of fat; fat and ligament; and ligament and bone (spinous process) or ligament and muscle. We used aqueous gel as the acoustic medium between the probe and porcine skin. We recorded signals as the probe was moved in a craniocaudal direction. We amplified echo signals by 20 dB. The operator inserted the epidural needle at five levels when the ligament-spinous process signal decreased and that between ligament and muscle appeared. Depths were estimated as for the plastic model, but with an estimated sound velocity of 1540 m.s 1. The Institutional Review Board of National Health Research Institutes approved the human study. Informed consent was obtained from all five male volunteers. The subject sat on a chair, slightly bent forward. Ultrasound echo signals were recorded between L3 and L5, amplified by 30 dB, with aqueous gel as the acoustic medium. All measurements were carried out without palpation and visual cues. We did not insert any epidural needles.

not cover the interspace between the L3 and L4 processes (Fig. 4a, depth 13 mm, peak amplitude 0.54 V). Subsequently, echoes from the L3 process decreased with caudad movement as those from the spinal cord increased (Fig. 4b, depth 44 mm, 0.11 V). The echo signals increased in amplitude to 0.39 V in the interspace, without significant echo signals from the L3 spinous process or inferior articular process (Fig. 4c). Finally, the probe was moved to the superior segment of the L4 spinous process (Fig. 4d, depth 38 mm, 0.18 V). Figure 5 illustrates results from the porcine lumbar phantom. The operator identified the L3–4 interspace by the disappearance of the signal between ligament and bone and by the appearance of the signal between ligament and muscle (Fig. 5b). The distances between interfaces estimated by ultrasound at L3 were 5.5, 11.4, 11.7 and 4.3 mm, respectively, which were within 10% of the mean (SD) distances measured by calliper: 3.8 (0.2); 10.6 (0.4); 15.2 (0.8); and 3.7 (0.3) mm, respectively. The caudad limit of the interspace was detected by reappearance of the signal for the interspace between ligament and bone (Fig. 5c). The operator then moved the probe rostrally until the echo signal between ligament and muscle recurred. The epidural needle was then inserted: in all five experiments, the needle passed without contacting bone (Fig. 6).

Results Figure 4 illustrates the echo signal moving caudad in a plastic model from L3 to L4. We detected a strong echo signal from the inferior segment of the L3 spinous process when the projected area of the probe did

(a)

(b)

(c)

(d)

Figure 4 The echo signals from the plastic model as a function of depth: (a) at L3; (b) between L3 and the interspace; (c) the interspace; (d) L4. 748



Anaesthesia 2014, 69, 746–751

(a)

Figure 6 An epidural needle inserted under ultrasound guidance into the pig phantom without bone contact.

(b)

five human studies. The peak amplitude of 2.5 V over L3 (Fig. 7a) decreased to 1.39 V with caudad movement (Fig. 7b). The amplitude decreased further to 0.90 V directly over the interspace (Fig. 7c), which increased to 2.72 V over L4 (Fig. 7d). Figure 8 shows the profile of echo amplitude between L3 and L5: the interspace nadir corresponds to sites for needle insertion.

Discussion

(c)

Figure 5 The echo signals from the pig phantom: (a) at L3; (b) the interspace; (c) L4. The echo signals were: 1, skin-fat; 2, fat-fat; 3, fat-ligament; 4, ligament-bone or ligament-muscle. The five male volunteers were 23–30 years old with body mass indices 19–29 kg.m 2. Figure 7 illustrates movement of the probe from L3 to L4 in one out of the © 2014 The Association of Anaesthetists of Great Britain and Ireland

In the phantom study, the echo signal did not require amplification because the ultrasonic attenuation coefficient of water is only 0.00025 np.cm 1 at 1 MHz. In the porcine model and human studies, the signal was amplified because of attenuation by tissue. There is only one subcutaneous fat layer in humans compared with two in the pig. When subjects bent their backs, the supraspinous ligament became very thin: it was difficult to distinguish the interface between ligament and fat from the interface between ligament and bone. However, the L3 and L4 echoes were both about three times stronger than the interspace echo (Fig. 6). The variation in signal amplitude was sufficient to identify the interspace from spinous processes. Our probe was only able to differentiate bony structures from other structures, whilst commercially 749

Anaesthesia 2014, 69, 746–751


(a)

(b)

(c)

(d)

Figure 7 The echo signals from human subjects: (a) L3; (b) between L3 and the interspace; (c) the interspace; (d) L4. the National Health Research Institutes (Project ME101-PP-10), Taiwan.

Competing interests No competing interests declared.

References Figure 8 The profile of signal amplitudes L3–5: the interspace nadir corresponds to sites for needle insertion. In this case, the maximum echo occurred at L3 (100%), and the minimum-echo signals were 9% of the maximum. The needle insertion can be performed at the minimum-echo locations. available 2D and 3D ultrasound systems can distinguish spinous processes, laminae, facet joints and transverse processes. A more powerful, higherfrequency ultrasound signal would improve the resolution of our probe, which might be further improved by geometrical focusing. Alternatively, it could be combined with a fibreoptic-guided technique [7] or high-frequency ultrasound probe needle [8]. In conclusion, we have developed a prototype ultrasound system that might reduce bone contact during neuraxial blockade.

Acknowledgements The study was supported by the National Science Council (Project NSC100-2221-E-400-002-MY3) and 750

€senberg A, et al. Epidural catheter 1. Willschke H, Marhofer P, Bo placement in children: comparing a novel approach using ultrasound guidance and a standard loss-of-resistance technique. British Journal of Anaesthesia 2006; 97: 200–7. 2. Gnaho A, Cirodde A, Lemarec C, Chazalon P, Jost D, Gentili ME. Ultrasound guided epidural anesthesia versus standard loss of resistance technique in obese patients. European Journal of Pain Supplements 2011; 5: 102. 3. Shaikh F, Brzezinski J, Alexander S, et al. Ultrasound imaging for lumbar punctures and epidural catheterisations: systematic review and meta-analysis. British Medical Journal 2013; 346: f1720. 4. Vallejo MC, Phelps AL, Singh S, Orebaugh SL, Sah N. Ultrasound decreases the failed labor epidural rate in resident trainees. International Journal of Obstetric Anesthesia 2010; 19: 373–8. 5. Marhofer P, Harrop-Griffiths W, Willschke H, Kirchmair L. Fifteen years of ultrasound guidance in regional anaesthesia: part 2-recent developments in block techniques. British Journal of Anaesthesia 2010; 104: 673–83. 6. Ting CK, Tsou MY, Chen PT, et al. A new technique to assist epidural needle placement: fiberoptic-guided insertion using two wavelengths. Anesthesiology 2010; 112: 1128–35. 7. Chiang HK, Zhou Q, Mandell MS, et al. Eyes in the needle: novel epidural needle with embedded high-frequency ultrasound transducer – epidural access in porcine model. Anesthesiology 2011; 114: 1320–4. 8. Karmakar MK, Li X, Ho AM, Kwok WH, Chui PT. Real-time ultrasound-guided paramedian epidural access: evaluation of a novel in-plane technique. British Journal of Anaesthesia 2009; 102: 845–54. © 2014 The Association of Anaesthetists of Great Britain and Ireland

Chen et al. | Prototype ultrasound needle guide to reduce epidural bone contact 9. Tran D, Kamani AA, Al-Attas E, Lessoway VA, Massey S, Rohling RN. Single-operator real-time ultrasound-guidance to aim and insert a lumbar epidural needle. Canadian Journal of Anesthesia 2010; 57: 313–21. 10. Phipps N. Acoustic Intensity Measurement System: Application in Localized Drug Delivery Electrical Engineering. San Jose, CA: San Jose State University, 2010. 11. Phillips R, Harris G. FDA Guidance Information for Manufacturers Seeking Marketing Clearance of Diagnostic Ultrasound


Anaesthesia 2014, 69, 746–751

Systems and Transducers. Rockville, MD: Center for Devices and Radiological Health/US FDA, 2008. 12. Shung KK. Gray-scale ultrasonic imaging. In: Diagnostic Ultrasound: Imaging and Blood Flow Measurements. Boca Raton, FL: Taylor & Francis Group, 2006: 79–88. 13. Culjat MO, Goldenberg D, Tewari P, Singh RS. A review of tissue substitutes for ultrasound imaging. Ultrasound in Medicine and Biology 2010; 36: 861–73.

751