To the Editor:— The recent article of Perlas et al. 1attests to the capability of ultrasound to image neural structures. These authors elegantly demonstrate, as previously reported, that spatial compounding of ultrasound beams allows excellent definition of nerves 2,3and tissue planes, 4both critical to the success of peripheral nerve block procedures. Their landmark study also identifies fundamental limitations of the commonly practiced nerve stimulation techniques.
The authors’ own approach, however, is not without risk. Perlas et al. propose to advance the needle in a lateral to medial direction above the clavicle to the brachial plexus, a mesiad direction first proscribed by Winnie. 5With this approach, the needle heads toward the spinal cord, lung, and major vessels, thereby risking injury to these vital structures. This small volunteer study is underpowered to address these safety concerns at a statistical level.
Other technical issues in the article also deserve comment. When scanned perpendicular to the long axis with a linear-array transducer, tendons seem more echogenic than nerves. 6However, the echogenicity of nerves and tendons varies with the angle of the scan head, a phenomenon known as anisotropy . 7,8Tendons, which have few nonspecular reflectors, are much more highly anisotropic than nerves. 9Therefore, whether a tendon is more echogenic than a nerve depends on the angle of the scan head. Although compound imaging reduces these anisotropic effects, 4the same principles apply.
Another important issue is what constitutes contact of nerve and needle (lack of ability to detect separation). Resolution of ultrasound images is limited, with lateral resolution typically worse than axial resolution. This gives rise to the needle blur pattern, the extent of which also depends on the acoustic background and image quality controls. 10Even with optimal alignment of the ultrasound beam and needle to identify the needle tip, the “contact” thus depends on the angle of approach and the complexity of the surrounding tissues. To complicate matters further, small structures may be interposed between nerve and needle but unresolvable from background with ultrasound imaging. The study therefore begs the question of single-injection block success rates with sonographic evidence of contact in the absence of nerve-stimulated muscle contraction.
Ultrasound-guided procedures are not without complications. Although compound imaging improves needle tip visibility, 11even in skilled hands it is difficult to keep the needle tip within the plane of imaging during an entire procedure. Needle visibility issues are especially important with thin flexible needles, such as the 22-gauge, 2-inch needles used by these investigators, which may bend partially out of the plane of imaging. The lateral to medial approach to brachial plexus blocks above the clavicle invokes potential risks to vital structures in the event of poor needle tip visibility. We are therefore concerned that the approach of Perlas et al. may be a step in the wrong direction.
University of California, San Francisco, San Francisco General Hospital, San Francisco, California. email@example.com