Monitoring Intraneural Needle Injection

Our objective to accurately localize the needletip during peripheral nerve block is primarily safety driven because we realize that unintentional intraneural needle injection can result in nerve injury.1,2 An intraneural injection, whether intrafascicular or extrafascicular (interfascicular), always indicates a violation of the epineurium boundary. Although the mechanisms of nerve injury are many and not fully understood, clinical manifestation of neurologic deficit is thought to be more likely to occur when trauma is inflicted on nerve fascicles.3 The threat of an intrafascicular needle puncture is raised especially in nerves with a high nerve fascicle to connective tissue ratio, for example, proximal nerve root.4 Although an extrafascicular intraneural injection may have no neurologic sequelae,5,6 we must recognize that current methods of needle guidance, whether nerve stimulation or ultrasound, cannot distinguish between a hazardous intraneural injection and one that is safe. Electrical stimulation is an improved method of nerve localization over paresthesia,7 but its accuracy in detecting needle-to-nerve contact remains suboptimal, as shown in both animal8,9 and human studies.10 A threshold stimulating current, 0.2 mA ≤ intensity ≤ 0.5 mA, is often considered the optimal extraneural stimulating end point for peripheral nerve block.11–13 However, human study data indicate otherwise and reveal the following: (1) there is a wide interindividual variability in the intensity of stimulating current required to evoke a motor response7; (2) different nerves (e.g., radial and ulnar nerves at the elbow) in a given individual may have different stimulation thresholds10; (3) the same nerve in a given individual may have different stimulation thresholds at different body locations (e.g., median nerve in the axilla versus wrist) due to different tissue-specific electrical impedances14; and (4) higher stimulation thresholds are observed in patients with neuropathy (e.g., diabetes).6 Most importantly, a threshold stimulating current >0.5 mA does not always exclude intraneural needle placement.12 For example, the reported intraneural stimulation threshold of the popliteal sciatic nerve ranges from 0.35 to 1.2 mA (mean, 0.58 mA) in humans12 and 0.08 to 1.8 mA in pigs, with 12.5% requiring current ≥0.8 mA.9 Nerve tissue architecture can influence the stimulation threshold by virtue of the proportion of connective tissue within the nerve4,15 and around it.16 Furthermore, a threshold stimulating current of 0.2 mA ≤ intensity ≤ 0.5 mA cannot exclude intraneural needle placement. Sala Blanch et al.13 observed similar stimulating current thresholds when the needle was positioned intraneurally and extraneurally (mean, 0.33 vs 0.35 mA, respectively) in human sciatic nerve. In this issue of Anesthesia & Analgesia, Wiesmann et al.17 went one step further and demonstrated that even a stimulating current intensity <0.2 mA could not discern an extraneural from an intraneural needle insertion. By applying electrical stimulation of varying pulse durations (0.1, 0.3, and 1 milliseconds) to the terminal branches of the brachial plexus in an open dissection pig model, the authors found that the minimal stimulating current was indistinguishable between needle position inside a nerve and in direct nerve contact (a mean intensity of 0.12 mA in both cases, with 0.1-millisecond pulse duration). However, the stimulating current was higher (mean, 0.28 mA) when the needle was 1 mm away from the nerve. The findings of this study are in contradistinction to past pig8,18 and human6 data in that the stimulating current required for both intraneural and direct extraneural stimulation of the brachial plexus was significantly and consistently lower (<0.2 mA vs >0.5 mA). The authors attribute this observation to differences in experimental conditions among studies and the lack of hyperpolarization “stimulation block.” Ultrasound, another nerve localization method, allows needle and nerve visualization, and thus presumably increases the sensitivity of detecting needle-to-nerve contact.7 Past cadaver and animal studies have demonstrated the usefulness of ultrasound in intraneural needle detection, but this happens often after initiation of an intraneural injection, resulting in sonographic evidence of nerve expansion and a change in nerve echogenicity.13,19 Clinical incidents of unintentional intraneural needle placement are attributed to poor needle tracking skills and failure to identify the needletip in real time.20,21 Realizing these limitations, it may be prudent to accept an injection end point (satisfactory local anesthetic spread around the nerve) rather than to always seek a needle position end point (direct needle-to-nerve contact) during ultrasound-guided regional anesthesia. Applying the hydrodissection technique with repeated small volume injections to open the path for needle advancement and to achieve a perineural injection without direct needle-to-nerve contact may improve safety. Recent clinical outcome data also suggest enhanced benefits when nerve stimulation and ultrasound are combined.22,23 Injection pressure monitoring is another potential safeguard against intraneural injection. Data extracted from animal models suggest that injection pressure is higher with intrafascicular compared with extrafascicular injection and that high pressure (>20–25 psi) results in histologic and functional nerve damage.24–26 Because the anesthesiologist’s ability to perceive injection pressure during peripheral nerve block is highly variable, the proposition of using pressure monitoring devices to keep the injection pressure to <15 psi is attractive. However, the fact that low pressure has been observed during intrafascicular injection (<11 psi) in 42%25 to 60%26 of cases underscores the need for more vigorous studies to further examine the sensitivity, specificity, and limitations of this monitoring tool. Peak injection pressure may also vary as a function of nerve size, nerve composition (connective tissue to nerve fascicle ratio), needle size, injection rate, injection volume, and perhaps animal species. For example, the peak pressure for intrafascicular injection (mean ± SD) was 10.9 ± 3.6 psi for rabbit sciatic nerve,24 29.7 ± 7.4 psi for dog sciatic nerve,26 and 48.9 ± 10.2 psi for human cadaver brachial plexus nerve root,27 measured under different experimental conditions. In addition, peak pressure is achieved often seconds after initiation of an injection (not immediate) and is likely influenced by the volume and rate of injection. The initial high resistance to injection encountered during the isostatic phase of pressure testing (before injection)24–27 is often due to needle–fascial sheath contact and not intraneural needle placement, as visualized under ultrasound. We believe that the endeavor to monitor and prevent intraneural needle trauma and neurologic injury is still a work in progress. The search for a consistently effective monitor continues as evidenced by the development of other novel technologies, for example, electrical impedance monitoring,28 high-definition imaging,29 electromagnetic needle tracking,30 acoustic radiation force impulse imaging,31 and tissue sensing technology.32 At the present time, perhaps the strategy of depositing local anesthetic around the nerve without direct needle-to-nerve contact, whenever possible, is the best safeguard against nerve trauma while waiting for the “perfect” monitor. DISCLOSURES Name: Faraj W. Abdallah, MD. Contribution: This author helped write the manuscript. Attestation: This author approved the manuscript. Conflicts of Interest: The author has no conflicts of interest to declare. Name: Vincent W. S. Chan, MD, FRCPC. Contribution: This author helped write the manuscript. Attestation: This author approved the manuscript. Conflicts of Interest: Dr. Vincent Chan receives equipment support for research from BK Medical, Philips Medical Systems, SonoSite, and Ultrasonix. This manuscript was handled by: Terese T. Horlocker, MD.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.