Article Text
PDF
Abstract
Purpose A solution for real-time, ultrasound-guided, central neuraxial blockade placement remains elusive. A device that enables single-operator neuraxial placement while simultaneously visualising the spinal anatomy and needle trajectory may improve patient safety. We engineered a novel needle guide, the EpiGuide two dimensional (2D), and compared prepuncture insertion sites as located using the guide versus standard manual palpation.
Methods Interdisciplinary collaboration between engineers and obstetric anaesthesiologists and multiple iterative refinement led to the EpiGuide 2D, a prototype multichannel needle guide for ultrasound transducers. Following ethics committee approval, 22 healthy adult participants were recruited to undergo prepuncture lumbar epidural placement using the guide. The primary outcome was accuracy, defined as the percentage of successful placements of prepuncture needle insertion sites within the manually palpated intervertebral space. The secondary outcome was distance between the prepuncture insertion sites guided by the EpiGuide 2D versus by sites guided by manually palpation.
Results Mean (SD) body mass index of participants was 22.6 (2.1) kg/m2. The success rate of selecting a prepuncture site within the manually palpated intervertebral space using EpiGuide 2D was 95.5% (42 of 44 attempts). Of the failed sites, the mean (SD) distance to the intervertebral space was 1.3 (0.3) mm in the caudal direction. The mean (SD) distance between EpiGuide 2D sites and palpated sites was 3.6 (2.0) mm.
Conclusion The EpiGuide 2D, jointly developed between engineers and anaesthesiologists, was found to be as accurate as manual palpation in placing a prepuncture site within the intervertebral space.
- neuraxial blockade
- epidurals
- ultrasound
- needle guide
Statistics from Altmetric.com
Introduction
Real-time ultrasound guidance has led to improved accuracy and safety of central line insertion and peripheral nerve blockade in contemporary anaesthetic practice.1 2 In contrast, central neuraxial blockade such as lumbar epidural catheter placement remains a predominantly ‘blind’ procedure. Most anaesthesiologists still rely on manual palpation of anatomical landmarks for epidural placement, despite evidence individuals with high body mass index (BMI) or spinal deformities can be difficult to landmark with it and that ultrasound can increase accuracy and reduce failure rates of insertion attempts.3–7 Prior studies have reported reductions in failure rates from 7%–8% to 1%–2%.4 6
A key barrier to the routine adoption of lumbar ultrasonography for neuraxial procedures is the difficulty of incorporating real-time ultrasound-guided techniques by a solo operator. The technique used to confirm the epidural needle location requires one hand to stabilise the needle, while conventional loss-of-resistance technique is performed with the other hand. Thus, introducing a standard two-dimensional (2D) ultrasound transducer for real-time guidance often requires two operators, as demonstrated in Grau et al, which is not feasible in many clinical situations.8 3D ultrasound imaging systems with an attached needle guide developed to target this problem have proven to be feasible in an ex vivo model.9 10 However, 3D ultrasound transducers are expensive, and their slow frame rate significantly prolongs the epidural procedure, making it impractical for clinical adoption. Other novel solutions, such as a specialised syringe that avoids the need for the standard tactile-based loss-of-resistance technique, or the use of passive magnetic tracking needle guidance with sensors on the needle have not successfully translated into clinical practice.11 12 Elsharkawy et al have further summarised available technologies in this field.13
Given these challenges, a solution for single-operator, real-time ultrasound-guided lumbar epidural placement remains elusive. A device that could facilitate this using current 2D ultrasound technology that is widely available and used for many other procedures would have the potential to improve patient safety. With the above goal, through an interdisciplinary collaboration between engineers at the University of British Columbia and obstetric anaesthesiologists at BC Women’s Hospital, we designed and fabricated a novel needle guide, the EpiGuide 2D. Before we can conduct a clinical study to determine if the epidural needle could be successfully guided into the epidural space, we wish to establish whether the epidural needle guided by EpiGuide 2D would land in a prepuncture site that would normally be located using standard manual palpation technique.
Thus, our hypothesis is that the EpiGuide 2D using with a 2D ultrasound transducer would place prepuncture needle insertion sites within the L3–4 or L2–3 intervertebral spaces as determined by manual palpation. We evaluate this hypothesis in a prepuncture study with a cohort of low to medium BMI participants.
Methods
Development of EpiGuide 2D
The EpiGuide 2D was jointly developed through collaboration between engineers and obstetric anaesthesiologists. First, the obstetric anaesthesia team determined key features the design must facilitate, including single-operator workflow, midline needle insertion and real-time visualisation of the needle tip during insertion. Next, using the human spine model, the engineering team determined that a guide directly attached onto the 2D ultrasound transducer using a paramedian imaging technique would enable all key elements to be incorporated. Specifically, the guide would facilitate both the ultrasound transducer and a standard epidural needle to be held together with one hand. This frees the secondhand to hold the syringe and perform loss-of-resistance using a continuous pressure technique (figure 1). The EpiGuide 2D’s ability to visualise a needle in phantom and ex vivo porcine models has been thoroughly evaluated.14 Honigmann et al found that the EpiGuide 2D could provide needle visualisation in porcine tissue for the majority of epidural space depths found in humans.14 After multiple iterations refining the ergonomics of the guide, the current EpiGuide 2D prototype was created.
Left: An illustration of the EpiGuide 2D in use by a solo operator. The EpiGuide 2D and ultrasound apparatus are held in the left hand, fingers securing the needle, while the right hand performs loss-of-resistance. Right: an illustration of the markings made at the desired level. 2D, two dimensional; D, distance between the two sites; E, needle insertion site from EpiGuide 2D; I, intervertebral space; M, midline, P, needle insertion site from manual palpation.
Design of EpiGuide 2D
The EpiGuide 2D is a multichannel needle guide designed to enable a single-operator to use their existing workflow and equipment while achieving real-time visualisation. The current prototype is made for use with a C-3 curvilinear transducer (Clarius Mobile Health, Vancouver, British Columbia, Canada) (figure 2). Additionally, it features a support pad with an indentation for stabilising the device against the patient’s back and to aid in marking the patient’s midline (figure 2). The EpiGuide 2D used in the study was made from 3D printed plastic (Afinia 3D, Chanhassen, Minnesota, USA) and is designed for single use. For clinical use, medical grade 3D printed plastic could be used instead. The device could also be made in an injection mould, or out of stainless steel without affecting quality. Sterilisation may be achieved through autoclaving or ethylene oxide gas, both of which are common methods for plastic medical devices. The needle channels are stacked vertically perpendicular to the transducer (figure 2). The ultrasound transducer can be placed in a paramedian orientation to allow for midline needle insertion while maintaining an acoustic window for real-time visualisation of needle trajectory.
(A) Rendering of the EpiGuide 2D used in this study where the multiple channels are stacked vertically. (B) Rendering of the EpiGuide 2D with a notched stabilisation pad (red circle) to permit the operator to stabilise the device against the patient’s back and make markings relative to the ultrasound with ease. (C) A white 3D printed plastic version of the stacked channel EpiGuide 2D, fitted onto a curvilinear transducer with a sterile probe cover and a 17G 9 cm echogenic Tuohy epidural needle. 2D, two dimensional.
The EpiGuide 2D mounts onto the end of the ultrasound transducer and has grooves to hold and guide a standard 17G Tuohy epidural needle. Using one hand to hold the ultrasound transducer with the EpiGuide 2D attached to it, the operator can visualise the anterior complex. They can use the fingers of the same hand to brace the needle into the guide’s channel. The operator’s free hand is then able to perform conventional continuous pressure loss-of-resistance under direct, real-time visualisation with partial visibility of needle progression while also maintaining the tactile feedback of a standard needle and syringe (figure 1).
Technical adaptations to facilitate needle visualisation
Traditional in-plane ultrasound-guided needle insertions, where the long axis of the needle is visualised under imaging, are difficult to perform for lumbar epidurals. This is due to the geometry of the needle insertion (figure 3) and the spinal anatomy, as well as the need to perform the insertion using a paramedian approach which is a departure from standard practice for many anaesthesiologists. Conventional understanding is that a needle inserted from out-of-plane can only be visualised as a single dot in the 2D ultrasound image as it crosses the imaging plane and therefore provides limited guidance (figure 3). However, 2D ultrasound images have a finite elevational resolution, perpendicular to the axial and lateral dimensions, as well as side-lobe artefacts. As a result of this elevational dimension, as demonstrated in a prior study by Honigmann et al,14 the needle may be visualised in the 2D image over a range rather than as a single point using a predefined angle.
Two orthogonal views of 2D ultrasound images and needle geometry. (A) Expected geometry and visualisation (red dot) for out-of-plane insertions. (B) Actual geometry and visualisation (red line). (C) Potential range of depth for needle visualisation using multiple channels (red lines) as provided in the EpiGuide 2D. (D) Traditional in-plane needle insertion. 2D, two dimensional.
Therefore, the EpiGuide 2D can also provide partial visualisation of the needle, and its progression, over various depths using both conventional 2D ultrasound imaging and an out-of-plane insertion technique. Through the use of multiple channels with predefined angles, the EpiGuide 2D is able to facilitate partial needle visualisation that covers the entire depth of the ultrasound image (figure 3). Each channel provides a distinct but overlapping range in which the needle can be partially visualised, allowing real-time imaging of the needle tip as it reaches the target. Knowing the target epidural space depth from a prepuncture ultrasound scan, and the range of depths each channel can visualise, the operator can select the optimum channel for the procedure. For example, if the prepuncture epidural space depth is estimated to be at 5 cm, the middle channel, which provides a visual window optimised for the 5 cm depth, would be used.
Study protocol
Healthy participants aged 19–64 were included. Participants were excluded if they had a BMI greater than 35 kg/m2, known spinal deformities (eg, moderate to severe scoliosis), or were unable to provide informed written consent.
After informed written consent was obtained, 22 participants were placed in sitting position with backs flexed and lumbosacral regions exposed (figure 4). The sitting position helps to increase the size of the intervertebral space and is common in anaesthesia practice. An ultrasound scan was performed by an anaesthesiologist with over 5 years of experience performing epidurals. The scan was used to identify the L3–4 intervertebral level (figure 4). The ultrasound depth was adjusted accordingly to visualise the epidural space. With the Clarius transducer, frequency and gain settings were automatically determined for each participant. No actual needle insertion or loss-of-resistance was performed in this prepuncture study. At the L3–4 intervertebral level, the anaesthesiologist performed manual palpation to mark the length of the intervertebral space (I) and the proposed needle insertion site (P-site) (figure 1). Markings were transferred onto a transparent sheet using the technique previously described by Beigi et al, and subsequently removed from the skin.9
Top: Example positioning of participants in sitting position. Bottom: imaging with the EpiGuide 2D in a paramedian orientation with a pen insert for demonstration. 2D, two dimensional.
Next, the EpiGuide 2D and ultrasound transducer were placed at the same level to obtain a paramedian ultrasound image (figure 5). A paramedian imaging approach was chosen because prior experimentation using a transverse imaging approach found poor accuracy in the needle reaching the target space. When the transducer was placed in a transverse orientation, the small intervertebral space was consistently obstructed by the width of the ultrasound transducer. The anaesthesiologist altered the transducer position and angle to optimise the anterior complex image (figure 5). A blunt sham needle was then inserted into the EpiGuide 2D needle channel to indent an insertion site on the participant’s skin (E-site) (figure 4). The channel was selected based off the epidural depth identified during the initial ultrasound scan. The E-site was transferred to the same transparent sheet as previously. All markings at the L3–4 level were repeated at the L2–3 level. Finally, the midline was found at each level using the ultrasound (figure 4). Midline identified through ultrasound has been previously found not to be significantly different from palpation identified midline.15
Ultrasound image in the paramedian orientation. AC, anterior complex; AP, articular process; ITS, intrathecal space; PC, posterior complex; TP, transverse process.
Outcomes and statistical analysis
The primary outcome was accuracy, defined as the percentage of E-sites successfully placed on the skin within the palpated L3–4 or L2–3 intervertebral space with the anterior complex visible on ultrasound. The EpiGuide 2D was considered accurate if the percentage was ≥95%. Figure 1 schematically illustrates an intervertebral space with an E-site.
Several additional outcomes were measured. First, for the failed E-sites that were outside of the intervertebral space, the mean distance between the failed site and the nearest point of the measured intervertebral space was calculated. Then, the distance between the E-sites and the P-sites was compared. The mean distance was reported as the total distance as well as the distances in the caudal-cranial and lateral-medial axes (figure 1). Next, the distance from each type of needle insertion site, P-sites and E-sites, to the ultrasound-measured midline was calculated (figure 1). Statistical significance was determined using the Student’s t-test. The mean length of the intervertebral space across all participants was also measured to assess the size of the ‘target’ entry space. Intraoperator variability in manual palpation was calculated by repeating three P-sites at the L3–4 level on a single patient and calculating their distance from the average. As this was a pilot study to obtain measurements from a novel design, no formal sample size calculation was performed.
Results
A total of 22 participants were recruited, with 16 males and 6 females. The mean (SD) BMI was 22.6 (2.1) kg/m2. The E-sites were successfully placed within the measured intervertebral space in 42 out of 44 attempts (95.5% success rate). The two failed E-sites were a mean (SD) distance of 1.3 (0.3) mm in the caudal direction from the intervertebral space. In comparing the E-sites to P-sites, the magnitude of the mean (SD) distance was 3.6 (2.0) mm overall. The mean signed components of this error were 0.4 (3.2) mm caudally and 0.7 (2.3) mm medially. These distances are visualised in figure 6. In comparing the distance of the sites from the midline, the mean (SD) distance of E-sites to midline was 2.1 (1.2) mm and the P-sites to midline was 1.5 (1.0) mm; no significant difference was found (p>0.05). Intraoperator variability was found to be within 1.0 mm. The mean (SD) intervertebral space length was 13.8 (2.3) mm.
Plot of distances from EpiGuide 2D to manual palpation, across all trials including failed attempts. Black marker indicates reference point, representing manual palpation. Blue markers indicate successful attempts (attempts which fell within intervertebral space), while red markers indicate failed attempts. Intervertebral space boundary is not shown. Green asterisk indicates mean including failed attempts. x-axis error bars and y-axis represent SD in the medial-lateral and cranial-caudal axes, respectively. 2D, two dimensional.
Discussion
When used in the paramedian sagittal imaging plane in a prepuncture setting, the EpiGuide 2D mounted on a Clarius C-3 2D curvilinear ultrasound transducer was shown to produce a pre-puncture site that falls within the manually palpated intervertebral space in 95% of attempts and with minimal distance from manually palpated prepuncture sites. Although the EpiGuide 2D was designed to work specifically on a Clarius 2D curvilinear ultrasound transducer in this study, it is a simple modification to allow the guide to be compatible with other curvilinear or linear ultrasound transducers. This allows clinicians who currently perform lumbar ultrasonography to use their existing equipment. When held in a paramedian orientation, the EpiGuide 2D was able to guide needle insertion sites within the intervertebral space >95% of the time, with no significant difference compared with needle insertion sites identified by manual palpation in participants with BMI <35 kg/m2. The distances of the failed E-sites were considerably close to the intervertebral space, at a mean distance of 1.3 mm away. Multiple sources of potential error may account for this distance, such as the intraoperator variability which was 1.0 mm.
A potential barrier to the use of ultrasound imaging in this setting is the expertise required for image acquisition and interpretation. Pattern recognition software for ultrasound may assist in these stages to simplify the overall procedure. Several solutions with demonstrated potential for clinical translation have emerged. Hetherington et al demonstrated how machine learning may reduce intervertebral level identification errors.16 At the desired intervertebral level, Seligman et al was able to landmark identification software to ease spinal ultrasound interpretation.17 Pesteie et al localised and highlighted the epidural space for users, similarly using machine learning.18 Finally, Beigi et al was able to enhance the visibility of the needle, even in imperceptible conditions.19 Similar software could be created to automatically advise the user which needle channel to use during a prepuncture scan to make the EpiGuide 2D even more user friendly. These innovations may in turn may make ultrasound-guided needle insertions accessible to a wide array of anaesthesiologists, regardless of their experience in spinal ultrasound.
There were several limitations to this study. There are multiple determinants of target epidural space entry success, one of which includes prepuncture site identification. We acknowledge that accuracy in identifying a prepuncture site is different from accuracy in entering the target epidural space. As such, further translational research is necessary prior to clinical adoption. This includes assessing the EpiGuide 2D for epidural blockade success in phantom models, ex vivo and in vivo animal models, and eventually humans. Other factors that could influence successful placement, including patient positioning, patient BMI, needle visibility at larger epidural space depth. Quantification of needle visibility at a range of target depths with the EpiGuide 2D was previously performed in laboratory conditions on ex-vivo models and illustrated clinically acceptable error.14 This prepuncture study provides reassurance that parameters measured in any future clinical studies involving EpiGuide can be expected to be within clinically acceptable range. Another limitation of our study is that all the procedures in the experiment were performed by a single anaesthesiologist. A learning effect is possible and future validation studies should be performed with multiple clinicians. Moreover, ultrasonography is most useful in patients with high BMI and spinal abnormalities; these patients were excluded from this study and thus limit the generalisability of the findings.
In conclusion, through engineer-clinician collaboration, we developed a novel epidural needle guide and demonstrated that the use of it on a 2D ultrasound transducer resulted in a 95.5% success rate of selecting a prepuncture site within the intervertebral space defined by manual palpation. These results support further study of EpiGuide 2D in a formal clinical trial to determine whether the epidural needle could be successfully guided into the epidural space and the technical modifications that may be required. This is a promising step towards performing real-time, 2D ultrasound-guided lumbar epidural placement by a single operator.
Acknowledgments
The authors acknowledge funding from the Natural Sciences and Engineering Research Council of Canada (NSERC Discovery 2020-05061). The authors would like to thank James Taylor for assistance with the manuscript.
References
Footnotes
Twitter @antonchau1
Contributors RR originally conceived the idea for the EpiGuide. RR and PA were coinvestigators for funding. RS, CB, SH, AC and RR were responsible for prototype design. SH and RS were responsible for prototype development. CB and RS were responsible for participant recruitment and study execution. RS performed data analysis. All authors contributed to the writing of the manuscript and providing feedback throughout the process. RR was the overall project principal investigator.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests The authors declare that a US provisional patent (No. 62/666,260) has been submitted for this work. The authors declare no conflicts of interest.
Patient consent for publication Not required.
Ethics approval The study was approved by the University of British Columbia Research Ethics Board (H15-01310).
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available on reasonable request.