High-frequency Ultrasonography of the Digital and Palmar Nerve Branches of the Hand in Peripheral Nerve Diseases

Summary

The present protocol describes high-frequency neuromuscular ultrasonography of the digital and palmar branches of the median and ulnar nerve, which can aid in localizing peripheral nerve diseases and be adapted to evaluate digital nerve injuries.

Abstract

Peripheral nerve ultrasound is a well-established imaging technique to evaluate certain peripheral nerve pathologies. However, there is a poor correlation between ultrasound abnormalities of peripheral nerves and electrodiagnostic or clinical evidence of axonal loss. This is a significant limitation of peripheral nerve ultrasound, as many peripheral nerve diseases encountered in clinical settings are related to axonal loss. Furthermore, clinical and electrodiagnostic evidence of axonal loss directly correlates with disability in all peripheral nerve diseases. However, due to the floor effects often encountered in electrodiagnostic studies, these correlations, as well as definitive diagnoses, are often challenging. Thus, imaging techniques that correlate with axonal loss are essential for expanding the utility of peripheral nerve ultrasound as a potential biomarker for peripheral nerve diseases. With new technological advancements and the ever-increasing imaging capabilities of high-frequency ultrasound, the palmar and digital nerve branches of the hand can be imaged with exceptionally high resolution even using point-of-care ultrasound devices. Their superficial and distal-most anatomic locations are ideal for evaluating polyneuropathies, as these branches degenerate earliest during axonal loss. However, no studies have systematically evaluated these nerve branches to determine if they can be reproducibly measured with ultrasound. The current protocol was adapted for the systematic assessment of cross-sectional areas of the median and ulnar nerves in the palmar surface and digits of the hand. This protocol provides reference data for a subset of nerves that demonstrate high intraclass correlation coefficients between three separate ultrasonographers. Finally, as a proof of concept and to demonstrate the clinical applications of this protocol, representative data from individuals with genetically confirmed inherited polyneuropathies are compared with established normative data to examine cross-sectional area differences.

Introduction

The expansion of clinical ultrasound to evaluate peripheral nerves and muscles has substantially improved the ability to diagnose neuromuscular disorders¹. Over the past 2 decades, ultrasound has emerged as a tool to directly image anatomical changes in the neuromuscular system, which correlate with pathological processes. Ultrasound is most commonly combined with clinical history and examination to provide further detail or support electrodiagnostic studies, which are considered a gold standard equivalent for diagnosing peripheral nerve disease². In some cases of focal neuropathies such as carpal tunnel syndrome, ultrasound can be used in lieu of electrodiagnostic results with high sensitivity and specificity³. Due to its low cost, its ability to be performed at the bedside, and its non-invasive properties, ultrasound is the preferred imaging modality for the neuromuscular system for many clinicians^4,5.

Peripheral nerve ultrasound has been explicitly proven invaluable for the localization of peripheral nerve diseases caused by abnormalities in myelin, such as chronic immune demyelinating polyneuropathies (CIDP)^6,7 and Charcot-Marie-Tooth disease type 1A (CMT1A)^7,8. In these diseases, the focal or diffuse cross-sectional area enlargements of nerves in the upper and lower extremities are well described. However, cross-sectional area enlargement is not specific to demyelinating diseases, as it has also been described in axonal polyneuropathies, albeit sparsely⁸. However, the cross-sectional enlargement in axonal diseases is significantly less robust and is non-uniform throughout the nerve. Due to these challenges, the utility of ultrasound in axonal neuropathies is limited.

Most peripheral nerve ultrasound studies have focused on imaging relatively proximal nerve locations primarily due to the larger nerve size, which makes identification more straightforward. However, the distal-most branches of peripheral nerves degenerate earliest in a Wallerian-like fashion during axonal loss in polyneuropathies^9,10. Due to their smaller diameters, the imaging resolution has been a limiting factor for the reproducible imaging of these nerve branches. Recently, transducer resolution has improved sustainably due to more rapid and seamless image compounding techniques. Now, structures of approximately 500 µm can be routinely imaged with point-of-care ultrasound, and structures with sizes as low as 30 µm can be imaged using ultra-high frequency systems¹¹. Thus, it is conceivable that the distal nerve branches in the feet and hands could be reliably evaluated with point-of-care ultrasound.

The palmar and digital nerve branches of the hand are the distal-most branches of the median, radial, and ulnar nerves. The palmar branches carry motor nerves (median and ulnar only) to the interossei muscles, in addition to the digital sensory nerves¹². In cadaver studies, the palmar and digital branches measure between 0.8 mm and 2.1 mm in diameter¹³, which is well within the detection range for high-frequency ultrasound transducers. In addition, their superficial location allows for high- and ultra-high frequency imaging because of the minimal frequency loss through connective tissues or muscles. However, no studies have established normative cross-sectional areas of the digital or palmar nerve branches of the hand using ultrasound, which are necessary to allow for clinical and research studies. Therefore, the present protocol evaluates the palmar and digital nerve branches in hand.

Technical considerations
The principles of neuromuscular-focused ultrasound must be reviewed as a foundation before starting this protocol¹⁴. In addition, several specific considerations were made for the current protocol. A transducer with a small footprint and frequency above 15 MHz is recommended, given the hand's natural contours. A 10-22 MHz transducer, with a footprint of 8 mm x 13 mm (see Table of Materials), was used with a compatible digital ultrasound system.

The next considerations are imaging depth and focal zones. In all the present imaging studies, the palmar and digital nerves were less than 0.35 cm in depth. Thus, using a consistent depth of 1 cm is recommended for reproducibility. Furthermore, improved imaging can be achieved at this depth by placing two focal zones at the device's maximal heights.

Consistent image adjustments (frequency, gain, and grey maps) are strongly recommended. With the minimal superficial tissues overlying and surrounding the nerves, adjustments in these parameters during imaging will not improve the imaging resolution or quality. Given the small diameters of these nerves, utilizing secondary image analysis software such as ImageJ^15,16 for cross-sectional area measurements is recommended.

Protocol

All experiments in this study were performed in compliance with the Wayne State University and Detroit Medical Center Institutional Review Boards (IRB) under an approved protocol for the natural history of individuals with peripheral neuropathies. Informed consent was obtained from all human participants.

1. Instrumental setup

1. Enter the patient's name or other identifiers as needed to organize the captured images.
2. Sanitize the ultrasound probe (see Table of Materials) and apply ample ultrasound gel to the head of the transducer before imaging.
3. Set the capture frequency to the highest frequency allowed by the ultrasound machine that does not require further image compounding or contrast.
   NOTE: For the present study, 20 MHz was used.
4. Set the total imaging or screen depth to 1 cm, with two focal positions as superficial as possible.
   ​NOTE: These depth and focal locations allow imaging of all the palmar and digital divisions of the median and ulnar nerves in the hand. Once determined by the examiner's preference, a consistent gain and grey map is recommended for the reproducibility of the measurements.

2. Patient preparation

NOTE: The patient inclusion criterion for inherited peripheral neuropathy was a confirmed mutation in a gene known to cause inherited peripheral neuropathy (please see the representative results section), with no exclusion criteria. For controls, the inclusion criterion was a normal result from electrodiagnostic testing of the upper extremities. Control exclusion criteria included a history or current diagnosis of diabetes mellitus, thyroid dysfunction, any known vitamin abnormalities, previous bariatric surgery, metabolic syndrome, a body mass index greater than 29, a history of traumatic nerve injury, or a history of large or small fiber peripheral neuropathy.

1. Place the patient in a supine position with the elbow extended and the wrist supinated such that the dorsal surface of the forearm is resting comfortably on the table.
   NOTE: Before positioning, examine the hand and arm that will be evaluated as any wounds, rashes, or skin irritation near or over the areas to be imaged are relative contraindications and may preclude ultrasound imaging.
2. Extend the patient's wrist and metacarpals so that the fingernails contact the table's surface. Allow the thumb to be slightly adducted and flexed for comfort.

3. Ultrasound examination

NOTE: There are four common palmar branches of the median nerve and two common palmar branches of the ulnar nerve¹³. Each digit has a medial and lateral digital branch, with digits 1-3 being purely median innervated and digit 5 being purely ulnar innervated. Digit 4 is dually innervated by the median nerve on the lateral surface and the ulnar nerve on the medial surface. This protocol concentrates on imaging the median common palmar nerve to digit 2, the lateral digital branch to digit 2, the ulnar common palmar branch to digit 5, and the medial branch to digit 5.

1. Start by identifying the median common palmar branch to digit 2 using the lateral transverse flexor palmar crease as the landmark (Figure 1).
   NOTE: The median common palmar branch to digit 2 and the ulnar common palmar branch to digit 5 are significantly larger than the accompanying blood vessels when compared to the digital nerves. Thus, identifying these nerves first can assist with identification in more distal branches, which are more similar in cross-sectional area to the blood vessels.
2. Place the transducer proximal to the radial flexor palmar crease (Figure 1).
   NOTE: In the present study, individuals without a radial flexor palmar crease were not encountered. However, in the case of an absent palmar crease, place the transducer 2 cm proximal to the base of the second metacarpal.
3. Hold the transducer perpendicular to the expected course of the nerve with as little pressure as possible while still ensuring complete contact with the transducer and skin.
   NOTE: Given the superficial location of the palmar and digital branches, they are susceptible to pressure distortions from the transducer.
4. Adjust the angle of the transducer such that the smallest cross-sectional area with a uniform epineurium is identified. This technique lessens the likelihood of out-of-plane imaging, which can alter results.
5. Optimize the image by making gentle back and forth movements of the transducer to minimize anisotropy of the nerve.
   NOTE: Anistrophy refers to the variance in the reflected ultrasound waveforms based on the angle of the transducer¹⁷. Adjusting the transmitted wave angle creates reflective waveforms that are out of plane and not received by the ultrasound transducer, leading to hypoechogenicity (or an increase in the black signal) of a structure. Nerves have relatively low anisotropy, meaning significant angle changes are required to create hypoechogenicity of the nerve. In comparison, muscles and tendons have relatively high anisotropy. By performing small back and forth motions with the transducer, the angular limits in which anisotropy arises can help identify the proper plane for nerve imaging¹⁸.
6. If possible, utilize Power Doppler imaging (PDI, see Table of Materials) to identify any blood vessels nearby (perform at all locations).
7. Once optimized, capture the image at this location. Mark the nerve within the ultrasound system prior to saving so that the nerve can be located for further measurement. Label the image with the nerve name, location, and side.
8. Next, image the lateral digital branch of digit 2 at the metacarpophalangeal joint by advancing the transducer distally toward the end of the digit 2. Then, stop the transducer just distal and lateral to the center of the flexor crease of the second metacarpophalangeal joint.
9. Once optimized, capture the image at this location. Mark the nerve within the ultrasound system before saving so that the nerve can be located for further measurement. Label the image with the nerve name, location, and side.
10. To evaluate for focal pathologies or non-uniform cross-sectional enlargements of the palmar or digital branches to digit 2, advance the transducer distally toward the end of the digit 2.
    NOTE: Digital nerves can be adequately visualized within 1.5-2 cm proximal to the distal interphalangeal joint (DIP).
11. Then, from the most distal point from which the branch can be visualized, follow the nerve proximally to the common median nerve just distal to the carpal tunnel.
12. Next, image the ulnar common palmar branch to digit 5 by identifying the ulnar transverse palmar crease (Figure 1).
13. Place the transducer perpendicular to the expected course of the nerve and adjust the imaging as described in steps 3.3.-3.6.
14. Once optimized, capture the image at this location. Mark the nerve within the ultrasound system prior to saving so that the nerve can be located for further measurement. Label the image with the nerve name, location, and side.
15. Next, image the medial branch at the metacarpophalangeal (MCP) joint by advancing the transducer toward the end of the digit 5, stopping just distal to the flexor crease of the MCP.
16. Optimize the image as previously mentioned in steps 3.3.-3.6.
17. Once optimized, capture the image at this location. Mark the nerve within the ultrasound system prior to saving so that the nerve can be located for further measurement. Label the image with the nerve name, location, and side.
18. Evaluate for focal or segmental cross-sectional abnormalities along digit 5. By identifying the distal-most location, visualize the nerve and scan proximally back to Guyon's canal.
    NOTE: Guyon's canal, also known as the ulnar canal or tunnel, is located on the medial aspect of the hand/wrist where the ulnar nerve and artery pass. The borders of Guyon's canal include the superficial carpal ligament superiorly, the flexor retinaculum and hypothenar muscles inferiorly, the pisiform and pisohamate ligament medially, and the hook of hamate laterally¹⁹.
19. Perform all measurements on both sides.
    NOTE: For the present study, for individuals with a BMI of less than 33, imaging of the entire course of the median and ulnar nerve back into the brachial plexus was possible with the transducer used here. Although this protocol only focuses on a limited number of nerves, in the case of focal or traumatic neuropathies (for which the evaluation of nerves is not shown in this protocol), it is recommended to use the distal heads of the metacarpals as a starting point for tracing and evaluating other digital and palmar nerves.
20. Save all the images and export them to a mass storage device. If using ImageJ, export the images as .jpg files.

4. Measuring cross-sectional area

NOTE: ImageJ, an open-source image processing software (see Table of Materials), was utilized for the present study, and the steps below are adapted for this software.

1. Open ImageJ software.
2. Select File in the program interface and click on Open. Navigate to the directory where the ultrasound images are stored.
3. Select the .jpg associated with the nerve to be measured.
4. Set the scale for the image by selecting the line tool and using the scale bar on the original ultrasound image to draw a 1 cm straight line. Click on Analyze and choose Set Scale.
   NOTE: The distance in pixels is automatically calculated from the 1 cm line and is auto-populated in the first box. To output measurements in millimeters squared (mm²), change the known distance to 10. Close the set scale box.
5. Use the freehand selection tool to outline the nerve at the border of the epineurium and surrounding tissues (Figure 2, yellow line).
   NOTE: The zoom function can be used in ImageJ to determine the exact pixels separating the epineurium from the surrounding structures for more precise measurements.
6. Measure the cross-sectional area by clicking on Analyze and choosing Measure.

Results

For the normative data, 20 individuals were selected with normal electrophysiology results, no neurological complaints, past medical history of or current diabetes mellitus, thyroid dysfunction, vitamin abnormalities, metabolic syndrome, carpal or cubital tunnel syndrome, exposure to chemotherapeutics, or severe hand trauma, and who had not been pregnant within the last 1 year (Table 1). Given the small subset, we did not stratify our data by age, gender, weight, or height, all of which are known to affe...

Discussion

The present protocol describes high-frequency ultrasound of the hand's digital and palmar nerve branches. This study was designed to test the hypothesis that cross-sectional area enlargement in distal nerve branches correlates with axonal loss. Extensive multicenter natural history studies of individuals with different subsets of axonal diseases will be needed to resolve this hypothesis. In addition to its potential research benefits, this protocol can also be applied clinically to localize peripheral nerve complaint...

Materials

Name	Company	Catalog Number	Comments
10-22mHz Transducer	General Electric Health Care	H48062AB	Small foot print transducer
ImageJ	NIH	N/A	https://imagej.nih.gov/ij/
Logiq eR8 Ultrasound Beam Former	General Electric Health Care	H48522AS	This is the beamformer and image processor which includes Power Doppler Imaging
Ultrasound Gel	Parker Labratories	44873	Standard ultrasonoic gel, non sterile