JUERD WIJNTJES
JUERD WIJNTJES Why, when and how MUSCLE ULTRASOUND IN CLINICAL PRACTICE
Muscle ultrasound in clinical practice © Juerd Wijntjes, 2025 ISBN: 978-94-6522-382-7 All rights reserved. No part of this thesis may be reproduced or transmitted in any form or by any means without written permission of the author. Cover design and layout: © evelienjagtman.com Printing: Ridderprint | www.ridderprint.nl The research presented in this thesis was conducted within the Clinical Neuromuscular Imaging Group at the Department of Neurology, Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Center, Nijmegen, the Netherlands. Printing of this thesis was financially supported by the Dutch Society for Clinical Neurophysiology (NVKNF).
Proefschrift ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen op gezag van de rector magnificus prof. dr. J.M. Sanders, volgens besluit van het college voor promoties in het openbaar te verdedigen op donderdag 18 december 2025 om 12:30 uur precies door Juerd Wijntjes geboren op 24 mei 1984 te Hoorn Why, when and how MUSCLE ULTRASOUND IN CLINICAL PRACTICE
Promotoren: Prof. dr. N. van Alfen Prof. dr. B.G.M. van Engelen Prof. dr. C.L. de Korte Copromotor: Dr. J. Doorduin Manuscriptcommissie: Prof. dr. M.A.A.P. Willemsen Dr. S. Lassche (Zuyderland) Prof. dr. R.A.J. Nievelstein (UMC Utrecht)
Table of contents Chapter 1 General introduction and thesis outline 7 Chapter 2 Detection of neuromuscular disease Educational section 35 Part 1 - Why Chapter 3 Muscle ultrasound: present state and future opportunities Muscle & Nerve. 2021;63:455–466 59 Chapter 4 Diagnostic accuracy of gray scale muscle ultrasound screening for pediatric neuromuscular disease Muscle & Nerve. 2021;64:50–58 89 Part 2 - When Chapter 5 Visual versus quantitative analysis of muscle ultrasound in neuromuscular disease Muscle & Nerve. 2022;66:253–261 117 Chapter 6 Comparison of muscle ultrasound and needle electromyography findings in neuromuscular disorders Muscle & Nerve. 2024;69:148–156. 145 Part 3 - How Chapter 7 Improving Heckmatt muscle ultrasound grading scale through Rasch analysis Neuromuscular Disorders. 2024;42:14-21 169 Chapter 8 Short-term educational value of online neuromuscular ultrasound courses Muscle & Nerve.2023;67:63–68 193 Chapter 9 Summary, general discussion and future perspectives 209 Appendices Nederlandse samenvatting 231 Research data management 237 List of publications 241 Dankwoord - Acknowledgements 247 About the author 255 Where 259
CHAPTER 1 General introduction and thesis outline
General introduction and thesis outline | 9 General introduction This thesis focuses on the application and interpretation of muscle ultrasound in clinical practice. Muscle ultrasound is a user- and patient-friendly method primarily used in diagnosing patients with a potential neuromuscular disease. This introduction systematically outlines the tissue and technique explored in this thesis: muscle and ultrasound. Given that the author’s focus is both on education and science, this introduction has an educational character where possible. For this reason, an educational section on the diagnostic methods (other than muscle ultrasound) that are used in the analysis of patients with suspected neuromuscular diseases, has been added as Chapter 2, following this introduction.
10 | Chapter 1 Muscle The human body consists of various types of muscle tissue. In this thesis, we focus on skeletal muscles, hereafter simply referred to as “muscles”, that play a crucial role in moving the body skeleton, maintaining posture and interacting with the physical world. Muscles are surrounded by a thick layer of connective tissue called fascia. Within this fascia, the muscle can move freely without affecting the surrounding tissues. Superficial muscles are typically located just beneath the skin and a layer of subcutaneous fat. Tendons at the proximal and distal end of a muscle connect the muscle to the bone1. Muscles are composed of muscle fibers, each consisting of connected muscle cells called myoblasts, with cell nuclei situated at the periphery. The sarcomere, the smallest structural unit of a muscle fiber, houses myofilaments composed of two contractile proteins: F-actin and myosin. Sarcomeres align in parallel rows, forming myofibrils. Connective tissue called endomysium surrounds groups of myofibrils, shaping individual muscle fibers. Additional connective tissue layers, known as perimysium, encircle groups of muscle fibers, creating muscle fascicles. Finally, these fascicles are enveloped by another layer of connective tissue called epimysium, collectively forming the skeletal muscle1. See figure 1. Beneath the endomysium, the muscle fibers are enveloped by a membrane called the sarcolemma. From the sarcolemma, small tubular structures called T-tubules form, running around individual myofibrils. Alongside the T-tubules, the individual myofibrils are surrounded by the sarcoplasmic reticulum. This is a cellular structure that stores calcium. Releasing calcium into the myofibril causes the two contracting proteins, F-actin and myosin, to bind together and contract2. This process is also known as the cross-bridge cycle. In this cycle, adenosine triphosphate (ATP) binds to myosin, and the breakdown of ATP to adenosine diphosphate (ADP) ultimately leads to the binding of myosin to actin, causing myosin and actin to slide past each other over a distance of approximately 11 nanometers. The structure of muscles, including their shape and fiber direction, varies from one muscle to another and depends on the level of strength and the type of movement it performs. Some muscles have parallel muscle fibers aligned with the tendon, while others have a pennate or bipennate structure with diagonal muscle fibers in relation to the tendon. This diversity results in more than 650 different muscles in the body, each having a unique morphology1. See figure 2.
General introduction and thesis outline | 11 Figure 1: the structure of a skeletal muscle showing the various components from muscle fascicle to actin / myosin.. Depending on age and body composition, further changes occur in the muscles of healthy individuals. From around the age of 50 years, there is an increase in the amount of connective tissue in muscles at the expense of healthy muscle fibers, a condition known as sarcopenia3,4. When the body mass index (BMI) increases, the amount of fatty tissue within the muscles also increases5,6. Nerves that control the muscles (motor neurons) originate in the spinal cord. Each motor neuron controls a group of muscle fibers within a muscle through innervation by its axons, forming a motor unit. Most of these axons are surrounded by myelin, which facilitates faster nerve conduction. The junction from the nerve to the sarcolemma of a muscle fiber is called the neuromuscular junction. The nerve can cause the muscle to contract by releasing acetylcholine in this junction, ultimately leading to calcium release by the sarcolemma. See figure 3. A motor unit can consist of type I fibers, primarily responsible for endurance, or type II fibers, mainly involved in rapid strength delivery7. The two types of muscle fibers are distributed across the muscle in a checkerboard pattern8. See also figure 9.
12 | Chapter 1 Figure 2: impression of different muscles of the human body.
General introduction and thesis outline | 13 Figure 3: activation of muscle contraction by the nerve. The release of acetylcholine at the neuromuscular junction triggers calcium release from the sarcolemma, causing the contraction of actin and myosin.
14 | Chapter 1 Diseased Muscle Diseases affecting muscles and/or nerves (neuromuscular diseases) can both lead to structural abnormalities in the muscles. In muscle diseases (myopathies), abnormalities occur and accumulate in the muscle fibers. Fibers may die (necrosis), leading to tissue inflammation and fibrosis (formation of connective tissue, i.e. scar) and fatty infiltration. Additionally, muscle fibers attempt to regenerate. Regenerating muscle fibers are thinner (atrophic) and contain a nucleus positioned in the middle rather than excentric in the muscle fiber because of active DNA transcription needed for the recovery process7,9. There are hundreds of different myopathies10. This number increases annually due to the discovery of new genetic causes11. The nature and extent of the abnormalities in muscle fibers varies between these different myopathies. In muscular dystrophy, pronounced necrosis of muscle fibers can be observed. Additionally, signs of regeneration are evident. As the disease progresses, the muscle’s ability to recover diminishes, and the entire muscle undergoes extensive fibrosis and fatty replacement7,12. Inflammatory myopathies are characterized by an (auto-)inflammatory response against muscle fibers. Inflammatory cells will surround the muscle fibers. Necrosis can be observed, and specific cells are clearing remnants of muscle fibers through phagocytosis. Regenerating, atrophic muscle fibers are also present, and fibrosis can occur when regenerating fails7. Metabolic and mitochondrial myopathies are characterized by defects in essential proteins (enzymes) or cell structures responsible for energy production (mitochondria). These defects pose challenges in activating (metabolism) the muscle. Structural abnormalities in the muscle are much less common. Depending on the underlying defect, characteristic abnormalities may be observed, such as the accumulation of fat droplets in muscle fibers when there is a disturbance in fat metabolism. Another example is the accumulation of mitochondria in the case of a mitochondrial defect7,13,14.
General introduction and thesis outline | 15 Diseases of the nerve (neurogenic diseases) could also result in muscle fiber pathology. When a nerve (cell) is lost, the corresponding muscle fibers of the motor unit will no longer be activated by the nerve (denervation). These muscle fibers will atrophy, and in case of persistent denervation, they undergo necrosis. Another possibility is that a muscle fiber can be reactivated by another healthy nerve from a neighboring motor unit (reinnervation). An interesting phenomenon is that in case of neurogenic disease, the reinnervated muscle fiber adopts the same muscle fiber type as the reactivating motor unit (type grouping). This results in the loss of the normal checkerboard pattern, with groups of type I or type II fibers arranged closely together. Additionally, the motor unit becomes larger. If this motor unit, in turn, is also denervated, it leads to a large area with atrophic muscle fibers (group atrophy). See figure 9A and B. Ultimately, this process may result in only a few large motor units surrounded by denervated muscle consisting of fibrosis and fat replacement,15–17.
16 | Chapter 1 Ultrasound One speaks of ultrasound when the frequency of sound waves is above the audible range for the human ear, i.e. over 20,000 cycles per second (20 kHz). For medical imaging, sound waves in the range of 1 to 3 million Hz are used (1 – 30 MHz). When ultrasound encounters a transition between two materials with a different acoustic impedance (see table 1), a portion of the sound waves will reflect as an “echo.” Acoustic impedance is a measure of tissue stiffness and depends on tissue density and the speed of sound. An ultrasound device can both generate ultrasound waves and capture the “echo”, resulting in an ultrasound image. However, most of the echoes are produced by very small structures (smaller than the wavelength i.e. >50 micrometer (µm)) of which a huge number are present in the body (cells, cell nuclei, muscle fibers etc). The body is full of these small structures resulting in a wealth of reflections that are superpositioned creating a reflection pattern called speckle. The latter is the dominant pattern present in an echogram. Ultrasound can be generated using a probe containing a transducer array, which consists of hundreds of small piezoelectric elements. Piezoelectric elements can expand when subjected to electrical current. Under mechanical pressure, they release electrical energy. This dual capability allows them to generate sound waves on one hand and receive sound waves on the other. Table 1: Acoustic impedance of different types of tissue. Speed of sound (c). Acoustic impedance (Z) The unit used for acoustic impedance is a rayl. Tissue Density (kg/m3) c (m/s) Z (rayls) Air 1.2 330 3.96 * 102 Fat 924 1450 1.34 * 106 Water 1000 1480 1.48 * 106 Muscle 1068 1600 1.71 * 106 Skull bone 1912 4080 7.80 * 106 Properties of ultrasound Sound requires a medium to propagate, unlike, for example, light or X-rays. Tissue can be considered to be an elastic medium. A mechanical inward movement (compression) puts pressure on the tissue, as a piston compressing a spring (see figure 4) 18. When this pressure locally diminishes (rarefaction), the spring will stretch, while the compressed part of the spring moves further. When
General introduction and thesis outline | 17 compression and rarefaction are cyclically applied by the piezoelectric elements, sound waves can be generated. Depending on the vibration frequency of this cycle, the wavelength of the sound waves will vary. See also figure 418. Figure 4: ultrasound propagation through elastic tissue, compared to a piston attached to a spring. High-frequency expansion and contraction of piezoelectric elements generate ultrasound, which propagates as a longitudinal wave through tissue by compressing and rarefaction of tissue particles. Figure adapted from Buschberg et al. 18. Interaction of ultrasound with tissue Ultrasound will be partly or fully reflected when it encounters a transition between two different types of tissues that each have a different acoustic impedance. The greater the difference, the more of the sound waves will reflect. Sound waves can interact with tissue in several ways. The most relevant to this thesis are the following interactions of sound waves with muscle tissue: reflection, deflection, refraction, scattering, and attenuation. From an ultrasonographic perspective, muscle fibers have a distinctive characteristic: they are disproportionately much longer compared to their cross-sectional diameter. The ultrasound characteristics of muscle tissue will therefore differ in the longitudinal and axial directions. Viewed longitudinally, reflection occurs as sound waves reflect off the elongated surface of the muscle fascicles, see figure 5A. This results in the typical pennate pattern of a muscle. To maximize the amount of reflections, the probe should
18 | Chapter 1 be held as perpendicular to the tissue as possible. For muscle tissue, this means scanning at a 90-degree angle to the underlying bone or major fascial structure. When the probe is not kept at this angle, sound waves will hit the tissue obliquely and will be partly transmitted (refraction) and reflected, but not back to the probe. In neuromuscular ultrasound literature this is referred to as deflection. As a result, the amount of sound waves that will return to the ultrasound probe to be visualized on the ultrasound image will decrease, resulting in a darker appearance of the ultrasound image. Tilting the probe only 2-degrees from the perpendicular position can decrease the measured grayscale level up to 10%19. See figure 5B. Viewed transversely, ultrasound of the individual muscle fibers will result in scattering. Scattering occurs when a sound wave encounters structures with different acoustic impedance and a size equal to or smaller than the wavelength. Scattering results in the reflection of ultrasound waves in all directions. Consequently, most of the scattered soundwaves do not return to the probe and hence do not show up on the screen as image information. Using a frequency of 10 MHz, the wavelength is approximately 150 µm. Normal muscle fibers are around 40-50 µm in diameter, and atrophic ones will likely be smaller. This scattering effect will increase as more microstructures with different acoustic impedance like lipid droplets and connective tissue accumulate in the muscle fibers in the case of a diseased muscle. The reflections will interact with each other causing enhancing or diminishing of the signal intensities. This results in a “speckled” or “grainy” image due to the various intensities of the soundwaves received by the transducer. See figure 5C. In muscle tissue both reflection and scattering occur due to the wide variation in the diameters of the structures that make up the muscle, each also surrounded by their own connective tissue layers. Attenuation is the loss of signal intensity as the sound wave penetrates deeper into the tissue. This is caused by scattering and absorption of the ultrasound waves. Absorption occurs as the tissue itself absorbs energy from the signal. A severely affected muscle resulting in a lot of scattering will thus show pronounced attenuation. Attenuation increases exponentially as the ultrasound wave penetrates deeper into the tissue. See figure 5D.
General introduction and thesis outline | 19 Figure 5: schematic figure of the interaction of ultrasound with muscle tissue. A (reflection): sound waves reflect back to the probe. B (deflection): sound waves are reflected away from the probe. C (scattering): reflection of sound waves in all directions encountering structures with a size smaller than the wavelength. D (attenuation): loss of signal intensity in the deeper layers of the tissue caused by absorption and scattering of the ultrasound waves.
20 | Chapter 1 Image data acquisition and processing For data collection and eventual imaging, multiple hardware components are required. The beamformer controls the generation and detection of sound waves. It directs the transmitter, which regulates the transmitted energy. By allowing more voltage to the piezoelectric elements, the intensity of the transmitted ultrasound waves increases. The intensity that is generated has to be below a certain limit to keep ultrasound imaging safe. To receive sound waves (the echo), the beamformer pauses the transmitter. The time between the pulses is known as the pulse repetition frequency, usually ranging between 100-300 µs. When the piezoelectric elements receive sound waves, these are converted into electrical signals. The later a soundwave is received, the deeper the system assumes the reflector must have been as it interprets a time delay as a longer distance traveled by the sound wave. Since the speed of sound in soft biological tissues is on average 1540 m/s with a variation between 1480 and 1650 m/s, the travel time can be directly converted into distance. Through an analog-to-digital converter the analog electric signal will be digitalized for further processing.. All information is then summed as a function of time (depth) and forwarded to the receiver, which creates a 2D grayscale image from it. The signal can undergo further processing. With time gain compensation (TGC), the signal can be amplified based on return time (i.e. depth). Digital amplification of the entire signal, also referred to as gain, is also possible. With dynamic range or compression, the range of shades of gray to be displayed on the monitor can be determined. Increasing the dynamic range would result in more information about different gray values, resulting in a brighter image. Decreasing the dynamic range would result in more contrast (black and white). With noise rejection, stochastic signals can be filtered out to prevent unwanted noise on the screen. Ultimately, the grayscale values sent to the screen will be optimized through device dependent software post processing to generate a final image on the display18. Ultrasound beam formation After leaving the transducer, the individual ultrasound waves are transmitted in such a way by the individual piezoelectric elements that they will converge into a point at a certain depth. As the beam penetrates further into the tissue, the combination of transmitted waves will diverge, resulting in a spread of the acoustic energy. This transition point, where the beam is narrowest, is called the focal zone. At this point, the lateral resolution is the highest. See figure 6. Higher frequency probes produce narrower beams and therefore have better lateral resolution. A linear ultrasound
General introduction and thesis outline | 21 probe with a frequency range of 5-17 MHz will have a lateral resolution around 0.030.1 mm. Resolution is a measure of the ability with which two objects perpendicular to the ultrasound beam can be distinguished. The frequency also strongly influences attenuation. A frequency of 10 MHz results in ten times more attenuation than at 1 MHz. Consequently, the frequency is always a trade-off between resolution and penetration18,20. The focal zone, or focus, can be adjusted in depth by modifying the timing of emitting sound waves by the piezoelectric elements. A superficial focus is created by allowing the outer elements of the probe to emit soundwaves earlier than the inner ones. The deepest focus is created by firing all elements almost simultaneously18. Figure 6: ultrasound beam, focal zone and lateral resolution. Influence of ultrasound post processing on the level of echogenicity For this thesis, it is important to emphasize that many factors influence the final grayscale levels displayed in the ultrasound image on the screen. The user can influence echogenicity by tilting the angle of the transmitter on the tissue, causing more or less reflection, deflection, or scattering. The effect of the insonation angle on echogenicity is illustrated in figure 5A and B. The ultrasound frequency used, affects grayscale levels as frequency determines the degree of scattering and attenuation. The grayscale level will also depend on depth as attenuation and absorption of the sound waves increase with depth (“everything turns dark in the deep”). Additionally, the user can influence the grayscale level of the image by adjusting gain, time gain compensation, dynamic range, and noise rejection. Finally, there is proprietary software-based manipulation of the grayscale levels to obtain “nice” looking images, which varies for the different manufacturers and ultrasound devices.
22 | Chapter 1 Muscle ultrasound Ultrasound of normal muscle To recap, ultrasound waves are reflected when encountering a transition between two different types of tissues with different acoustic impedance. When ultrasound waves travel through healthy muscle tissue, relatively few tissue transitions and hence few reflections occur. Therefore, normal muscle has a relatively low echogenicity (black appearance). The transition from muscle tissue to connective tissue results in the reflection of sound waves and, consequently, higher echogenicity (white appearance). This reveals the linear or feathered muscle fiber structure in the longitudinal direction, while a speckled aspect emerges in the transverse scanning direction. This latter aspect is also referred to as a starry night appearance21. In a healthy muscle, a significant amount of ultrasound passes through, making structures such as an underlying bone easily recognizable as the sound waves will reflect at the muscleto-bone transition. To have sufficient dynamic range to visually detect abnormalities in muscle, it is best to choose a preset in which muscle tissue appears dark and the surrounding fascia structures and/or underlying bone white. Ultrasound of diseased muscle tissue Structural abnormalities in diseased muscle, such as fat replacement and fibrosis lead to an increase in the amount of reflections and scattering of ultrasound waves. This subsequently leads to an increase in echogenicity and, therefore, a whiter appearance of the muscle on the ultrasound screen. See figure 7. It is important to recognize that physiological changes too can result in higher echogenicity, such as the increase in fibrous tissue in the context of aging (sarcopenia) or the stacking of fat in muscle in case of a higher body mass index. Additionally, due to structural differences between the individual muscles, echogenicity will vary for each different muscle throughout our body. For instance, the echogenicity of a healthy dorsiflexor muscle of the foot (tibialis anterior muscle) is inherently higher than that of a thigh muscle such as the rectus femoris21,22. Visual analysis The most straightforward way to analyze muscle ultrasound is through visual analysis, i.e. looking at the image. A highly abnormal muscle is easily recognizable as a white appearing muscle beneath the rather black appearing overlying layer of subcutaneous fat, where the normal structure of the muscle is lost, and all the sound is completely reflected in the upper layer of the muscle (attenuation). As a result, an underlying structure such as a bone may sometimes no longer be visible. See figure 7.
General introduction and thesis outline | 23 Figure 7: muscle ultrasound images of a rectus femoris muscle in a healthy (left) individual and an individual with known Duchenne muscular dystrophy (right). For semi-quantitative visual analysis, an ordinal scale - as developed by Heckmatt et al. - can be used23. Grade 1 represents a normal muscle, while Grade 4 corresponds to a severely abnormal muscle. See also table 2. Table 2: Heckmatt grading scale. Heckmatt grade Criteria 1 normal muscle structure and echogenicity 2 increased muscle echogenicity with a distinct bone echo 3 marked increased muscle echogenicity with diminished bone echo 4 strongly increased muscle echogenicity with total loss of bone echo The visual interpretation of an abnormal versus a physiologically normal echogenicity of a muscle depends on the experience of the observer. This limits the sensitivity of visual analysis in distinguishing between a normal and diseased muscle to 70%24. When using the Heckmatt grading, the reported sensitivity is slightly higher, up to 76%25.
24 | Chapter 1 Heckmatt grading, once used for a single thigh muscle, is now applied to muscles throughout the body, often without an underlying bone, complicating the use of the original criteria. Better understanding the scale's clinimetrical properties and application across different muscles would improve the use of visual muscle ultrasound. Quantitative analysis With quantified grayscale analysis, the sensitivity for distinguishing between a normal and diseased muscle increases to 92%26. In the version of this technique used at the Radboudumc, the mean grayscale value is calculated from a manually selected region of interest in the muscle ultrasound image. This value can be compared to a reference value for that specific muscle. This reference value is adjusted for age, height, sex, and weight using multiple logistic regression analysis. After correction, a Z-score can be calculated, representing the number of standard deviations that the grayscale value deviates from the reference value22. Z − score (SD) = − A Z-score of > 2 (i.e. > 95th percentile) is considered abnormal. The examination can be performed by a technician trained to conduct the study in a standardized manner. Utilizing anatomical landmarks, measurements should always be taken at the same location, ensuring correct probe positioning directly over the muscle with an exact 90-degree angle of the probe to the underlying fascia or bone, avoiding exerting pressure on the muscle. The ultrasound device settings chosen will influence the echogenicity. This means that system and setting-specific reference values will need to be acquired for each different ultrasound device version, probe and preset, which is a labor-intensive and therefore costly process for most centers. The agreement between the visual and quantitative analysis of muscle ultrasound is not fully established. Gaining a deeper understanding of when these two techniques align, and when they do not, would provide valuable insights into determining the optimal use of each method.
General introduction and thesis outline | 25 Figure 8: illustrated histological findings in disease progression of a muscular dystrophy (A). Muscle ultrasound findings in a healthy subject and patients with facioscapulohumeral dystrophy (FSHD) in different stadia of disease progression (B). Figure inspired by thesis of S. Vincenten.
26 | Chapter 1 Disease specific findings In muscular dystrophy, there is a progressive fat replacement and fibrosis throughout the muscle, resulting in a homogenously white appearance of the muscle and loss of its normal texture, commonly referred to as a ground glass appearance. See figure 8B for muscle ultrasound examples of patients with facioscapulohumeral dystrophy (FSHD) with increasing disease severity. In advanced muscular dystrophy, the upper layer of the muscle already reflects a significant number of sound waves (attenuation). This results in a muscle ultrasound image with a very bright top muscle layer with a dark area underneath. In neurogenic diseases a “moth-eaten” appearance can be seen in longerstanding progressive neurogenic diseases for example slowly progressive motor neuron diseases such as spinal muscular dystrophy. In this case, round black areas (representing vital muscle tissue innervated by remaining individual motor units) are surrounded by echogenic, permanently denervated fibrotic muscle tissue. The moth-eaten pattern is the ultrasound correlate of type grouping. See figure 9C. Implementation, training and certification Muscle ultrasound has proven to be a valuable diagnostic tool gaining increasing interest in clinical practice over the last decades. An important challenge for implementing muscle ultrasound in clinical practice is obtaining access to dedicated ultrasound equipment. Often, ultrasound devices are available in acute care and radiology departments, and sharing arrangements with other departments can pose financial and logistic challenges. The advent of affordable handheld ultrasound devices and combined ultrasound-EMG equipment provides a possible solution to this issue. Another important challenge is the incorporation of this technique into standard training programs for medical specialists. A consensus-based guideline, authored by multiple experts in the field of muscle ultrasound, emphasized that visual muscle ultrasound assessment and grading should be considered a fundamental skill in the implementation of the technique27. The scarcity of experts in performing and interpreting muscle ultrasound makes widespread education on the technique challenging. Organizations such as the IFCN Society of Neuromuscular Imaging (https://www.ifcn.info/get-involved/ special-interest-groups/ifcn-society-of-neuromuscular-imaging) play a crucial role in educating medical specialists worldwide by organizing annual courses and conference meetings.
General introduction and thesis outline | 27 Figure 9: process of denervation and reinnervation in neurogenic disease (A). Correlating illustrated histological findings with increasing disease progression (B). Muscle ultrasound findings in a healthy subject and patients with different types of neurogenic disease in different degrees of axonal damage / denervation (C).
28 | Chapter 1 Aims and outline of the thesis With the development of quantified grayscale analysis, muscle ultrasound has become a well-validated, sensitive, easily applicable, and highly patient-friendly diagnostic technique. In this form it has routinely become part of our clinical practice over the past 10-20 years. At the department of clinical neurophysiology in the Radboudumc, neuromuscular ultrasound is increasingly used when screening for a neuromuscular disease (n = 1500 per year; increase of approximately 20% over the last 5 years) and EMG is decreasingly used (n = 2000 per year with a decrease of approximately 20% over the last 5 years). Most of the research on muscle ultrasound has focused on how to implement this technique as an outcome measure in clinical trials. The downside of this focus is that it creates a knowledge gap regarding research on the screening and diagnostic value of muscle ultrasound. In this thesis, my aim is to address this gap and critically appraise why, when and how we use muscle ultrasound in clinical practice.
General introduction and thesis outline | 29 Why In chapter 3 we will provide a comprehensive understanding of why we should use muscle ultrasound and focus on the current state and future opportunities of the technique. Since the introduction of muscle ultrasound there have been significant changes in the diagnostic approach to patients suspected of having a neuromuscular disease. With the advent of mass genetic testing panels (see Chapter 2), most classical inherited neuromuscular diseases can be directly genetically confirmed. This prompts the question why we still use muscle ultrasound to screen for patients with neuromuscular disease. Therefore, in Chapter 4, we will reexamine the diagnostic value of muscle ultrasound in children suspected of having a neuromuscular disease in the era of whole exome sequencing.
30 | Chapter 1 When Muscle ultrasound analysis is based on visual and/or quantitative methods. Visual analysis has the advantage that it can be performed using just simple visual assessment or using a semi-quantitative grading scale. This makes the technique straightforward to implement but also dependent on the experience of the observer. The quantitative method has a better diagnostic value but is not widely available due to the need for device specific reference values. To answer the question when we preferably use one and/or the other, we want to better understand the relationship between the visual and quantitative muscle ultrasound techniques. In Chapter 5, we therefore will investigate the relationship between findings of visual and quantitative muscle ultrasound analysis. Muscle ultrasound is a technique often used in addition to needle electromyography (see also Chapter 2). However, the relationship between abnormal findings in muscles examined with both ultrasound and needle electromyography has not yet been clearly delineated. When do both techniques align, and when do they not? For this reason we will examine the relationship between findings of needle electromyography and muscle ultrasound findings in different types of neuromuscular disorders in Chapter 6.
General introduction and thesis outline | 31 How Semiquantitative assessment of muscle ultrasound can be performed by using the 4-point Heckmatt scale. The scale was originally designed for assessing the quadriceps muscle, but in clinical practice it is used for assessing muscles in every region of the body and a lot is still unknown about the clinimetrical performance of this scale. Are we using the scale correctly and if not, how should we use it? Therefore, we will analyze the measurement properties of the Heckmatt scale in Chapter 7. The demand for accessible education on muscle ultrasound is growing, both for those seeking to learn and those aiming to teach. Since experts in this field are not available everywhere, the question arises: how can we properly train our clinicians? One of the solutions could be online education. For this reason, we will investigate the educational value of online education on neuromuscular ultrasound in Chapter 8.
32 | Chapter 1 References 1. Friedrich P. Sabotta Atlas of Anatomy, Volume 1. In: Sabotta Atlas of Anatomy. Vol Volume 1. 17th ed. 2. Walter F. Boron ELB. Cellular Physiology of Skeletal, Cardiac, and Smooth Muscle. In: Medical Physiology. 3rd ed. 2017 3. Ticinesi A, Meschi T, Narici M V., Lauretani F, Maggio M. Muscle Ultrasound and Sarcopenia in Older Individuals: A Clinical Perspective. J Am Med Dir Assoc: 2017;18:290–300. https:// doi.org/10.1016/j.jamda.2016.11.013. 4. Frontera WR, Ochala J. Skeletal Muscle: A Brief Review of Structure and Function. Behav Genet: 2015;45:183–195. https://doi.org/10.1007/s00223-014-9915-y. 5. Kovanlikaya A, Mittelman SD, Ward A, Geffner ME, Dorey F, Gilsanz V. Obesity and fat quantification in lean tissues using three-point Dixon MR imaging. Pediatr Radiol: 2005;35:601–607. https://doi.org/10.1007/s00247-005-1413-y. 6. Nijboer-Oosterveld J, Van Alfen N, Pillen S. New normal values for quantitative muscle ultrasound: Obesity increases muscle echo intensity. Muscle Nerve: 2011;43:142–143. https://doi.org/10.1002/mus.21866. 7. David S. Strayer JESER. Skeletal Muscle and peripheral nervous system. In: Rubin’s Pathology: Mechanisms of Human Disease. 8th ed. 2020 8. Sundaram C, S. M. Approach to the Interpretation of Muscle Biopsy. In: Muscle Biopsy. InTech; 2012 https://doi.org/10.5772/39034. 9. Joe AWB, Yi L, Natarajan A, Le Grand F, So L, Wang J, et al. Muscle injury activates resident fibro/adipogenic progenitors that facilitate myogenesis. Nat Cell Biol: 2010;12:153–163. https://doi.org/10.1038/ncb2015. 10. Lord Walton of Detchant, Rowland LP, McLeod JG. Classification of neuromuscular disorders. J Neurol Sci: 1994;124:109–130. https://doi.org/10.1016/0022-510X(94)90192-9. 11. Benarroch L, Bonne G, Rivier F, Hamroun D. The 2023 version of the gene table of neuromuscular disorders (nuclear genome). Neuromuscular Disorders: 2023;33:76–117. https://doi.org/10.1016/j.nmd.2022.12.002. 12. Cardone N, Taglietti V, Baratto S, Kefi K, Periou B, Gitiaux C, et al. Myopathologic trajectory in Duchenne muscular dystrophy (DMD) reveals lack of regeneration due to senescence in satellite cells. Acta Neuropathol Commun: 2023;11. https://doi.org/10.1186/ s40478-023-01657-z. 13. Lilleker JB, Keh YS, Roncaroli F, Sharma R, Roberts M. Metabolic myopathies: A practical approach. Pract Neurol: 2018;18:14–26. https://doi.org/10.1136/practneurol-2017-001708. 14. Pfeffer G, Chinnery PF. Diagnosis and treatment of mitochondrial myopathies. Ann Med: 2013;45:4–16. https://doi.org/10.3109/07853890.2011.605389. 15. van Baalen A. Muscle fibre type grouping in high resolution ultrasound. Arch Dis Child: 2005;90:1189–1189. https://doi.org/10.1136/adc.2005.083261. 16. Breiner A. Denervation. In: Encyclopedia of the Neurological Sciences. Elsevier; 2014. p. 971–972 https://doi.org/10.1016/B978-0-12-385157-4.00655-2. 17. Zhao W-P, Kawaguchi Y, Matsui H, Kanamori M, Kimura T. Histochemistry and Morphology of the Multifidus Muscle in Lumbar Disc Herniation. Spine (Phila Pa 1976): 2000;25:2191–2199. https://doi.org/10.1097/00007632-200009010-00009.
General introduction and thesis outline | 33 18. Bushberg J, Seibert A, Leidholdt Jr E, Boone J. Ultrasound. In: The Essential Physics of Medical Imaging. 3rd ed. 2013 https://doi.org/10.1118/1.4811156. 19. Dankel SJ, Abe T, Bell ZW, Jessee MB, Buckner SL, Mattocks KT, et al. The Impact of Ultrasound Probe Tilt on Muscle Thickness and Echo-Intensity: A Cross-Sectional Study. Journal of Clinical Densitometry: 2020;23:630–638. https://doi.org/10.1016/j. jocd.2018.10.003. 20. Allisy-Roberts P, Williams J. Farr’s Physics for Medical Imaging. Elsevier, 2008. https:// doi.org/10.1016/C2009-0-34335-4. 21. Pillen S, Arts IMP, Zwarts MJ. Muscle ultrasound in neuromuscular disorders. Muscle Nerve: 2008;37:679–693. https://doi.org/10.1002/mus.21015. 22. Scholten RR, Pillen S, Verrips A, Zwarts MJ. Quantitative ultrasonography of skeletal muscles in children: Normal values. Muscle Nerve: 2003;27:693–698. https://doi. org/10.1002/mus.10384. 23. Heckmatt JZ, Pier N, Dubowitz V. Real-time ultrasound imaging of muscles. Muscle Nerve: 1988;11:56–65. https://doi.org/10.1002/mus.880110110. 24. Pillen S, van Keimpema M, Nievelstein RAJ, Verrips A, van Kruijsbergen-Raijmann W, Zwarts MJ. Skeletal muscle ultrasonography: Visual versus quantitative evaluation. Ultrasound Med Biol: 2006;32:1315–1321. https://doi.org/10.1016/j.ultrasmedbio.2006.05.028. 25. Brandsma R, Verbeek RJ, Maurits NM, van der Hoeven JH, Brouwer OF, den Dunnen WFA, et al. Visual screening of muscle ultrasound images in children. Ultrasound Med Biol: 2014;40:2345–2351. https://doi.org/10.1016/j.ultrasmedbio.2014.03.027. 26. Rahmani N, Mohseni-Bandpei MA, Vameghi R, Salavati M, Abdollahi I. Application of Ultrasonography in the Assessment of Skeletal Muscles in Children with and without Neuromuscular Disorders: A Systematic Review. Ultrasound Med Biol: 2015;41:2275–2283. https://doi.org/10.1016/j.ultrasmedbio.2015.04.027. 27. Tawfik EA, Cartwright MS, Grimm A, Boon AJ, Kerasnoudis A, Preston DC, et al. Guidelines for neuromuscular ultrasound training. Muscle Nerve: 2019;60:361–366. https://doi.org/10.1002/mus.26642.
CHAPTER 2 Detection of neuromuscular disease Educational section
Detection of neuromuscular disease – Educational section | 37 Clinical evaluation Detecting a neuromuscular disease starts with the medical history and physical examination. Inquiries should be made about signs of weakness such as abnormal walking, difficulty standing up, or even falling. Information about the onset, duration, and progression of symptoms can help in making an accurate diagnosis. Delays in motor development or the presence of neuromuscular diseases in the family can be clues to the existence of a hereditary condition. Observation is an important part of the physical examination. For instance, lack of facial expression may be observed in cases of facial muscle weakness (myopathic facies). If there is weakness in the shoulder muscles, the shoulders may appear forward, and when the arm is moved, the shoulder blade can detach from the shoulder (winging scapula). Hip girdle weakness may cause the patient to struggle when moving from lying to standing, relying on their arms for support on the knees (Gower’s sign) and they can develop a waddling gait. Attention should also be given to certain physical deformities. Weakness in the paraspinal muscles can lead to a curve in the spinal column (scoliosis). Pronounced high-arched feet may result from congenital or progressive weakness of the foot muscles. During the examination of the muscles, attention should be paid to the wasting of the muscle bulk (atrophy). Atrophy often indicates a neurogenic cause, although it can also occur in severe myopathies. Additionally, one should look for brief spontaneous movements in the muscles (fasciculations), which also suggest a neurogenic pathology. The strength of individual muscles is assessed using the Medical Research Council (MRC) scale where grade 0 indicates no contraction (paralysis) and grade 5 represents maximum contraction (normal strength), or with dynamometry that measures strength in Newtons. The pattern of identified weakness can provide direction for the correct diagnosis. For example, in some myopathies, there may be specific involvement of the muscles at the level of the hip/shoulder girdle. Tendon reflexes are often normal in mild myopathies and might decrease as weakness increases. In neurogenic diseases, reflexes are often decreased or even absent, except when there is also upper motor neuron involvement such as in amyotrophic lateral sclerosis (ALS). Sensory examination remains normal in the case of a myopathy or conditions affecting only motor neurons. In peripheral nerve disorders, sensory examination often also shows abnormalities1.
38 | Chapter 2 Laboratory investigations Muscle tissue breakdown results in the release of the enzyme creatine kinase (CK) into the bloodstream. An elevated CK (at least 2 times higher than the normal value) may indicate the presence of a myopathy 2. A significantly high CK level can occur in certain muscle diseases such as Duchenne muscular dystrophy. However, a normal CK does not rule out a myopathy. A normal CK can for example be found in metabolic myopathies, where the energy metabolism in the muscle is disrupted. In this case, end products of energy metabolism, such as lactate, can be disturbed3. In neurogenic disorders with denervation the CK level can be slightly increased too4. In the case of a suspected inflammatory myopathy one can test for myositis specific auto-antibodies5 .
Detection of neuromuscular disease – Educational section | 39 Genetic tests A hereditary cause can now be identified for an increasing number of myopathies. In patients with a high suspicion of such a hereditary cause (young age, family members with a known myopathy), genetic material (genes) can be examined using a blood sample. Genetic material consists of deoxyribonucleic acid (DNA). Abnormalities in the DNA that lead to a disease are called mutations. If a mutation is known in a family, or there is a clinically strong suspicion of a myopathy with a known mutation, targeted investigation into this DNA abnormality can be conducted6. If a mutation is not known, the entire coding DNA (exome) can be checked for mutations known in myopathies. This is called whole exome sequencing (WES). It is important to realize that not all known mutations are found with WES7,8. For example, mitochondria have their own DNA material. When there is a suspicion of an inherited mitochondrial myopathy, mitochondrial DNA from muscle tissue should preferably be examined because mitochondrial DNA can vary by tissue type9. The approach to diagnosing neuromuscular diseases has evolved significantly with the advent of genetic testing. The latest studies on the diagnostic value of muscle ultrasound were conducted prior to this diagnostic evolution. Consequently, the patient population referred for muscle ultrasound screening has changed, making a reevaluation of its diagnostic value desirable.
40 | Chapter 2 Electrodiagnostic studies Electrodiagnostic studies include nerve conduction studies and needle electromyography that can be used to identify a problem within the nerve, muscle, or neuromuscular junction. Nerve conduction studies are performed by stimulating a peripheral nerve with an electrical pulse. The function of sensory axons can be tested by measuring the electrical signal running along the sensory axons in the region where they provide sensation to the skin resulting in a sensory nerve action potential (SNAP). A decreased or absent SNAP amplitude indicates loss or dysfunction of the sensory nerves. Stimulated motor axons will reach a muscle and cause the muscle fibers to contract. The electrical signal that can be recorded when the muscle fibers are contracting is called the compound muscle action potential (CMAP). The amplitude and shape of this CMAP can be used to detect abnormalities in the motor axons or muscle. The speed at which the electrical potential spreads along the nerve can be measured if the length of the segmental spread is known and is referred to as the conduction velocity. In certain (often treatable) nerve diseases that cause damage to the myelin, the electrical potential will progress very slowly along the nerve resulting in abnormal low conduction velocities10. A neuromuscular junction disorder can be detected by rapid sequential electric stimulation called repetitive nerve stimulation. In neuromuscular junction disorders, the signal reaching muscle fibers is disrupted. In cases of postsynaptic pathology (e.g. myasthenia gravis), reduced uptake of acetylcholine results in suboptimal or no muscle fiber activation, causing a gradual decrease in CMAP amplitude, known as decrement. During repetitive nerve stimulation, the amount of acetylcholine released into the synaptic cleft decreases with each stimulation. In a normal situation, during this decrease in acetylcholine release, muscle fibers remain activated and the CMAP amplitude does not decrease. However, in neuromuscular junction disorders, this decrease in acetylcholine release does result in suboptimal muscle fiber activation, and hence, decrement. In presynaptic pathology (e.g. Lambert Eaton myasthenic syndrome), reduced release of acetylcholine results in suboptimal muscle fiber activation and hence a lower CMAP amplitude. In these cases, high-frequency stimulation can optimize this disturbed acetylcholine release, activating more muscle fibers resulting in an increased CMAP amplitude, a phenomenon known as increment.
Detection of neuromuscular disease – Educational section | 41 When performing needle electromyography, the electrical signal of individual motor units can be examined using a needle containing two integrated electrodes in the core and the shaft (a concentric needle). This allows the study of individual motor unit action potentials (MUAPs). In a healthy muscle, all muscle fibers of a motor unit will contract almost simultaneously, resulting in a MUAP with approximately two phases (see figure 1). When a muscle contracts increasingly, an increasing number of motor units will be activated. This increase in the number of active motor units during contraction is also referred to as recruitment. Additionally, in a muscle at rest, it can be observed whether muscle fibers spontaneously emit signals (spontaneous muscle fiber activity). In a healthy muscle, there should be no spontaneous muscle fiber activity in the muscle at rest. With needle EMG, the muscle can be examined for myopathic or neurogenic abnormalities. In the case of myopathy, the reduced number of muscle fibers will result in shorter duration and lower amplitude of the MUAP (see figure 1). Due to the often large variability in diameter of the muscle fibers left, there is increased polyphasia due to the different conduction velocities across the muscle fibers. With advanced disease, the decreased number of muscle fibers will lead to immediate activation of fibers from multiple motor units even when low force is required (increased recruitment). In case of a neurogenic disease, degenerated neurons are no longer stimulating the corresponding muscle fibers of that neuron. With ongoing disease activity, the number of motor units decreases. Denervated muscle fibers emit signals that induce healthy nerves to sprout new branches, reinnervating/adopting these fibers. The increased number of muscle fibers adopted by a single nerve will result in higher amplitude and longer duration of the MUAPs. Since the conduction velocity over new nerve sprouts is initially very slow, it will take longer to activate the reinnervated muscle fibers. The muscle fibers of one motor unit therefore will not contract simultaneously, leading to polyphasia (see figure 1). The decreased number of motor units results in fewer active motor units available during contraction. This is referred to as reduced recruitment. A denervated muscle fiber will exhibit spontaneous activity at rest. Spontaneous depolarization of the muscle fiber can either result in positive sharp waves and fibrillation potentials11. Although these findings are classically associated with neurogenic diseases, local damage to muscle fibers in myopathy can also lead to spontaneous muscle fiber activity, particularly in the form of fibrillation potentials, when parts of the muscle fiber are disconnected from the endplate zone.
www.ridderprint.nlRkJQdWJsaXNoZXIy MTk4NDMw