Is there a place for ultrasound in diagnosing sarcopenia?

Various ultrasound parameters may be used to assess muscle mass and quality. The ultrasound parameters used in the research and clinical settings of sarcopenia management are muscle thickness (MT), muscle cross-sectional area (CSA), echo intensity (EI), muscle fiber pennation angle (PA) and muscle fiber length (FL). Additional parameters used in ultrasound muscle assessment include muscle volume (MV), muscle stiffness assessed through elastography, muscle contraction potential, and assessment of muscle microcirculation with contrast-enhanced ultrasound (CEUS).^39,40 A summary of the most commonly utilised ultrasound parameters is shown in Table 1.

Although an increasing number of studies incorporate various ultrasound parameters in muscle assessment, the most commonly measured parameters to assess muscle quantity and muscle quality remain muscle thickness and echo intensity, respectively. Due to their accessibility, size and location, upper and lower leg muscle groups, namely gastrocnemius and quadriceps femoris (particularly rectus femoris) are most widely measured. Additionally, most studies have shown that ultrasound parameters of these muscles are superior compared to other muscle groups regarding validity of the method to detect sarcopenia.^9,41 The vast majority of studies have utilised linear transducer probes for muscle assessment, as is recommended by the European Geriatric Medicine Society (EuGMS) SARCUS (SARCopenia through UltraSound) working group, since linear transducer probes are more adapted to assess muscle anatomy. However, some studies have shown comparable reliability and validity of both curved and linear probes for select ultrasound parameters and muscle groups.^42,43

The most commonly used methods for assessment of muscle mass (quantity) both in clinical practice and research settings are magnetic resonance imaging (MRI), computed tomography (CT), dual-energy X-ray absorptiometry (DXA) and bioelectrical impedance analysis (BIA).

CT examination provides an accurate and reliable assessment of muscle mass and represents the gold standard for non-invasive muscle mass assessment. However, due to the associated high dose of radiation, it is not suitable for neither everyday clinical use in sarcopenia management, nor can the use of CT be justified for purely research purposes. DXA and BIA are more widely available, yet both methods present with certain limitations regarding either accuracy, cost-effectiveness and/or accessibility.

The following sections present an overview of ultrasound parameters, used in muscle quantity assessment, and of research on the agreement between these parameters and reference methods in sarcopenia diagnosis. Ultrasound parameters that have been used in muscle quantity assessment studies are muscle thickness, cross-sectional area and muscle volume.

Muscle thickness is perhaps the most widely studied muscle ultrasound parameter in sarcopenia research. Muscle thickness represents the distance between the superficial and deep muscle fascia, while some authors define it as the distance between the bone-muscle interface and the adipose tissue-muscle interface.⁴⁴ Muscle thickness is considered a reliable parameter for quantitative ultrasound muscle assessment⁹, which can be easily and quickly measured.

Several studies have confirmed good reliability and validity of ultrasound-measured muscle thickness compared to reference imaging methods (DXA, CT, MRI) and direct cadaver measurements.^9,45-50 One study concluded that measurements of gastrocnemius medialis thickness obtained by ultrasound are reliable and correlate well with DXA-derived appendicular lean muscle mass and muscle performance in older individuals.²¹

In a study on patients with cirrhosis and sarcopenic obesity, Dhariwal et al. demonstrated that ultrasound-measured MT of thigh and forearm muscles demonstrate high diagnostic accuracy for sarcopenia and correlate well with computed tomography-determined skeletal muscle index (SMI) in these patients.⁵¹

In a recent comprehensive meta-analysis, Fu et al. concluded that ultrasound muscle parameters showed low to moderate diagnostic accuracy for sarcopenia, whereby the accuracy of the method depended on the parameters analysed, the muscles examined, the reference standards used, and the patient population included in the study. The authors of the analysed studies used different diagnostic criteria for sarcopenia, while almost all included studies used either DXA or BIA as the reference method for determining body composition. The MT of the gastrocnemius, rectus femoris and tibialis anterior muscles showed the highest, albeit moderate diagnostic accuracy for sarcopenia. The authors also reported cutoff values of ultrasound measurements of different muscle groups for sarcopenia diagnosis.⁴¹ Similarly, a meta-analysis conducted by Zhao et al. showed that the MT of the rectus femoris and gastrocnemius muscles showed the highest, yet also moderate, diagnostic value for low muscle mass or sarcopenia.⁵² A scoping review of six studies by Staempfli et al. also reported that the MT of rectus femoris showed the highest validity for sarcopenia diagnosis.⁵³ Similarly, in a systematic review of six studies, Nies et al. concluded that US examination of the rectus femoris muscle is a promising method to aid in the diagnosis of sarcopenia in various clinical populations.⁵⁴

A meta-analysis by Li et al., which included 9 studies using ultrasound, reported a high pooled correlation coefficient between MT of upper and lower limb muscles and DXA, demonstrating an acceptable diagnostic accuracy for sarcopenia.⁵⁵

Despite promising results, further research is required to determine the general utility of US-measured muscle thickness in predicting sarcopenia. Some authors have suggested correcting the muscle thickness according to body mass or body mass index (BMI), since body weight may influence muscle thickness through increase of local adipose deposits.⁵⁶ Regarding muscle volume, several prediction equations have been proposed based on measurable muscle ultrasound parameters and muscle volume determined by MRI.^57,58 However, more studies are needed to correlate muscle thickness to total muscle volume of individual muscles using the proposed equations.

The cross-sectional area (CSA) of the muscle depends on the number and size of individual muscle fibers and is usually determined at the point of the largest muscle diameter. It is important to distinguish between anatomical and physiological cross-sectional muscle area. The anatomical cross-sectional area (ACSA) is the area of the muscle perpendicular to its longitudinal axis, while the physiological cross-sectional area (PCSA) is the area of the muscle perpendicular to the course of its muscle fibers (ie, muscle volume divided by fascicle length). Anatomical cross-sectional area (ACSA) and physiological cross-sectional area (PCSA) are equivalent in non-pennate muscles.; however, in pennate muscles, they differ.⁵⁹ Muscle strength is more closely related to PCSA than to ACSA because PCSA represents the maximum number of potential actin-myosin cross bridges that can be activated in parallel during contraction. Therefore, when assessing muscle strength, relying solely on ACSA measurements is not recommended.^59,60 Muscle strength can be inferred from the cross-sectional area, as it correlates with muscle volume.⁶¹ Although an indirect assessment of muscle strength, it can prove useful in patients who are incapable of active muscle contraction.⁶²

In a study involving a small group of healthy elderly individuals, Reeves et al. reported a strong correlation between the muscle CSA of the vastus lateralis muscle measured by ultrasound and that determined by MRI. The authors reported an intraclass correlation coefficients of 0.998 for the reliability of ultrasound and 0.999 for its validity when compared to MRI.³³

Seymour et al. discovered a 25% reduction of rectus femoris CSA in patients with chronic obstructive pulmonary disease (COPD) compared to healthy controls. Additionally, they found a significant, though moderate, correlation between the CSA of the rectus femoris measured via ultrasound and the fat-free mass (FFM) derived from bioelectrical impedance analysis (BIA) in the control group (COPD patients).⁶³

A meta-analysis conducted by Zhao et al. showed a moderate diagnostic accuracy of rectus femoris CSA for low muscle mass determined by DXA or BIA.⁵² Similarly, in a more recent meta-analysis, Fu et al. showed that the CSA of rectus femoris and biceps brachii muscles showed a moderate diagnostic accuracy for sarcopenia diagnosis.⁴¹ In a study on 313 geriatric outpatients, Ozturk et al. concluded that rectus femoris CSA may accurately predict sarcopenia in these patients (AUC 0.766 and 0.773 for women and men, respectively).⁶⁴

A nationwide multicentre study on 991 hospitalised patients at risk for malnutrition demonstrated a significant positive correlation between the rectus femoris CSA and BIA-derived body cell mass as well as with handgrip strength, and a significant negative correlation with the Timed Up and Go test. Additionally, cutoff points of ultrasound measurements were determined for probable, confirmed, and severe sarcopenia.⁶⁵ For several muscles/muscle groups, it might prove difficult to measure cross-sectional area with conventional ultrasound methods and standard linear probes. Extended field-of-view modes may be used to facilitate imaging in these cases.

As mentioned above, no consensus has been reached on the definition of the term muscle quality, which is used to describe both changes in muscle specific strength as well as the composition of muscle tissue. However, myosteatosis is the most commonly used indicator for muscle quality in both clinical practice and research setting. The term myosteatosis describes the pathological fatty infiltration of muscle tissue and is, independent from sarcopenia, negatively associated with survival and other adverse treatment outcomes in various patient populations.^66–69 Muscle biopsy is considered the gold standard for assessing the degree of myosteatosis; however, due to its invasive nature and potential for complications, it is not used in routine clinical practice. CT examination is the most accurate non-invasive method for assessing myosteatosis, in which tissue attenuation is measured and expressed in Hounsfield units (HU).⁷⁰

Several other novel indicators for muscle quality have been proposed, including BIA-derived phase angle. With the continued development and refinement of methods for assessing muscle tissue quality, these indicators are expected to become increasingly important in the diagnosis and management of sarcopenia.⁸

Ultrasound parameters used in the assessment of muscle quality are pennation angle, fascicle length and echo intensity. Additionally, muscle stiffness, determined by sonoelastography, contraction potential as well as assessing microcirculation using contrast-enhanced ultrasound may provide further information on muscle quality.

The following sections provide an overview of ultrasound parameters and techniques for muscle quality assessment as well as available data on agreement with reference methods.

Information regarding muscle composition may be obtained by measuring the echo intensity (EI) of the muscle, also termed muscle echogenicity.⁷¹ Increased EI is an indicator of muscle degeneration, which presents as an increase in the proportion of intramuscular fatty and connective tissue.⁷² Echo intensity is most often determined by analysing the intensity of image points of the US image using grey scale analysis. A histogram function is used in the analysis, which is enabled by several image processing programs. This type of quantitative grey scale analysis has been shown to be more accurate than visual subjective US image assessment.⁷³ In a study on 40 young and older adults, Watanabe et al. demonstrated a moderate association between the echo intensity of anterior thigh muscles and CT-determined muscle attenuation of these muscles, signifying that echo intensity at least partially reflected the degree of intramuscular fatty infiltration.⁷⁴

Based on several studies, echo intensity of lower limb muscles appears to be useful in sarcopenia detection. Isaka et al. found an association between echo intensity of the tibialis anterior and gastrocnemius medialis muscles and the presence of BIA-determined sarcopenia in older individuals.⁷⁵ Similarly, Yamada et al. demonstrated the ability of echo intensity to differentiate between sarcopenia and normal/presarcopenia groups, diagnosed using BIA, in older men and women.⁷⁶ In a study on 78 patients with rheumatoid arthritis, Yoshida et al. reported a superior discriminatory performance of combined EI and CSA measurements of biceps brachii and rectus femoris muscles in sarcopenia diagnosis, compared to EI and CSA measurements alone.⁷⁷

Concerns have been raised regarding low interand intra-rater reliability of echogenicity measurements. Strasser et al. reported low intraclass correlation coefficients for EI, particularly in the older experimental group.⁷⁸ However, several other studies have showed very good intraclass correlation coefficients (≥ 0.800) and very small standard errors of measurement (≤ 7.26 %).^57,79 Valera-Calero et al. demonstrated good to excellent intra-(ICC 0.800–0.989) and inter-rater (ICC 0.841–0.948) reliability of EI measurements of cervical multifidus and short rotator muscles in healthy volunteers.⁸⁰

The method of measuring EI has several limitations. The EI of muscles in the elderly population and in certain patient populations is significantly higher than in younger people, which must be considered in the final assessment.⁸¹ The assessment may also be influenced by various external factors, e.g. ultrasound probe parameters, probe tilt, patient rest duration, participant positioning, and the patient’s hydration status.⁶² Additionally, inconsistent methodological approaches used across studies measuring EI make comparing results challenging.

Fukumoto et al. studied the influence of focus depth of US images on the depth-dependent attenuation of EI and the relationship between EI and MRI-measured intramuscular adipose tissues (IntraMAT). The correlation between EI and IntraMAT was found to be stronger when the focus was kept in the middle of the rectus femoris or vastus intermedius muscle compared to the top of the image.⁸²

Girts et al. determined that higher US image gain significantly increased EI values of the vastus lateralis muscle, whereas EI values were stable between depths of 4.0 and 6.0 cm.⁸³ However, Paris et al. also showed the importance of maintaining consistent depth across all ultrasound measurements.⁸⁴ Scafoglieri et al. demonstrated that even after standardising for gain, depth, and frequency, the EI values still differed significantly across other additional US settings.⁸⁵

The effect of subcutaneous fat on EI measurement has also been the subject of numerous studies, with some conflicting results. Young et al. proposed a correction factor for in vivo subcutaneous fat thickness on EI measurements, which has been used in several subsequent studies.^86–88 Müller et al. found that increasing exogenous fat thickness between the probe and the region of interest (ROI) resulted in a decrease in EI of the tibialis muscle.⁸⁹ Contrarily, Palmer et al. found that raw EI correlated better than corrected EI measurements with physical performance in healthy older women.⁹⁰

The exact influence of different US system settings and subcutaneous fat on EI measurements remains unclear. However, it seems imperative to use standardised settings as well as to assess the need to correct for subcutaneous fat thickness. Furthermore, more advanced image processing techniques, commonly referred to as texture analysis, have been suggested to potentially overcome some of the limitations linked to muscle echo intensity measurements.⁹¹

Muscle architecture can be described by the angle at which muscle fibers are connected to the fascia (pennation angle), as well as the length of the muscle fibers (fascicle length), both of which can be measured by US. The pennation angle was defined as the angle of insertion of muscle fascicles into the deep aponeurosis, while fascicle length was defined as the length of the fascicular path between the insertions of the fascicle into the superficial and deep aponeuroses.^59,92

Muscle architecture plays an important role in muscle force generation and is related to muscle function.⁹³ In sarcopenia, due to the smaller number of consecutive sarcomeres, the length of muscle fibers is reduced, while the pennation angle becomes smaller.⁹² Both parameters are related to a decrease in the maximum force and speed of muscle fiber contraction in sarcopenic patients.^94,95

The measurements of muscle architecture parameters are highly dependent on correct measurement techniques. Joint position, muscle contraction during measurement, probe placement on the skin as well as probe orientation relative to the muscle fiber course may influence results.^96,97 In cases where the fascicle extended beyond the boundaries of the acquired ultrasound image, the length of the missing portion can be estimated by linearly extrapolating both the observed fascicular trajectory and the aponeurosis.^59,92,98 Concerning pennation angle measurement, significant variability may occur regardless of the measurement site, potentially due to differing levels of myosteatosis and/or fibrosis within the same muscle.⁹⁹

Regarding calculation of fascicle length, several authors^57,100 used the following formula: FL( mm)=MT( mm)*sin(PA)−1,$${\rm{FL}}({\rm{mm}}) = {\rm{MT}}({\rm{mm}})*\sin ({\rm{PA}}) - 1,$$ where FL = fascicle length, mm = millimetre, MT = muscle thickness and PA = pennation angle.

One study reported the intra-rater correlation coefficient for the measurement of the pennation angle of the gastrocnemius medialis muscle ranging from 0.738 to 0.820.²¹ Several studies have shown good reproducibility of measurements in young individuals¹⁰¹, whereas Strasser et al. reported lower reproducibility of pennation angle measurements of the quadriceps muscle.⁷⁸

The utility of muscle pennation angle measurement to differentiate sarcopenic vs. non-sarcopenic adults remains unclear. Similarly, while the fascicle length tends to shorten with advancing age, its capability of differentiating patients with and without sarcopenia is also questionable. In a study carried on 100 elderly community-dwellers, Kuyumcu et al. concluded that both sarcopenic and nonsarcopenic subjects had similar pennation angles of the gastrocnemius muscle, while muscle thickness and fascicle length values were significantly lower in patients with sarcopenia.¹⁰² In a study on 57 healthy elderly individuals, Alvarez et al. found no correlation between gastrocnemius pennation angle and DXA measurements, concluding that pennation angle is not a suitable measurement in sarcopenia diagnosis.²¹ However, in a study on 279 elderly and 60 younger controls, Narici et al. demonstrated a significant correlation between DXA-derived skeletal muscle index (SMI) and the ratio between FL and MT of the vastus intermedius muscle (termed ultrasound sarcopenia index, USI).¹⁰³

As a measurement of muscle quality, the use of pennation angle in combination with other ultrasound parameters might provide a more comprehensive assessment of overall muscle health in sarcopenic individuals. The use of pennation angle in sarcopenia research remains relatively new, and further research is needed to validate its utility. Ultrasound measurement of fascicle length and pennation angle of pennate muscles requires standardized protocols of assessment and specific training of operators. Adequate reproducibility might be achieved once these criteria are met.

The microcirculation of the skeletal muscle is the primary and most important site for capillary-tissue exchange of nutrients, oxygen, and hormones, particularly during exercise.¹⁰⁴ CEUS is used to assess the degree of muscle vascularization. Reduced blood flow to the muscles, caused by microvascular damage and decreased nitric oxide production, has been cited as a significant factor in the development of sarcopenia.^105,106 Mitchell et al. have noted that the SonoVue contrast medium can be used to demonstrate a reduced circulatory response in muscle tissue to a nutritional stimulus, which may indirectly indicate a reduction in vascularization.¹⁰⁷ Further research on patients with sarcopenia is needed to assess the utility of this method in a clinical setting, which may be diminished due to the need for using a contrast agent.^15,59

Elastography is based on the change of the muscle’s biomechanical properties due to the increased content of fibrous and adipose tissue as well as glycated products.¹⁰⁸ By measuring the change in muscle stiffness, information regarding muscle function (strength, power, range of motion) can be obtained.³⁹ Shear wave sonoelastography, in contrast to strain mode, seems to have emerged as the dominant and superior modality for assessing muscle stiffness.^109,110 Alfuraih et al. suggested that changes in muscle stiffness, detected by elastography, might be correlated to muscle weakness.¹¹¹

Due to conflicting research results, it remains unclear whether muscle stiffness increases or decreases with age.^112–114 Janczyk et al. conducted a systematic review to explore the potential of sonoelastography as a reliable method for assessing sarcopenia in older adults. Their findings indicate that the passive elastic constant was significantly higher in sarcopenic individuals compared to healthy subjects after passive stretching. Among the ten studies reviewed, four reported increased muscle stiffness in older adults, two reported decreased stiffness, and four found no significant differences. Ultimately, the authors could not draw definitive conclusions about the usefulness of elastography in assessing sarcopenia.¹¹⁵ Bastijns et al. propose that differences in passive torque, activity levels and the ultrasound probe axis may account for changes in shear values in different age groups.¹¹⁰ Standardisation of measurement protocols is paramount for effective comparison of studies using elastography in sarcopenia assessment.

As of now, there is no universally accepted standardized approach for conducting muscle ultrasonography in clinical practice. In 2018, The European Geriatric Medicine Society (EuGMS) SARCUS (SARCopenia through UltraSound) working group published their first recommendations on the standardization of the use of ultrasound for muscle assessment. In 2021, SARCUS proposed an updated consensus protocol for using ultrasound in muscle assessment, including measurement of muscle thickness, cross-sectional area, fascicle length, pennation angle and echogenicity.³⁹

Both the EWSGOP2 and the SARCUS working groups emphasize that ultrasound has proven to be an accurate and reliable technique of muscle mass measurement in different populations, while showing high repeatability. However, standardization of measurement methods is paramount to perform extensive and comparative studies.^8,39 The following paragraphs highlight the important aspects regarding standardisation of measurement protocols.

Due to its size, accessibility and comfort for the patient, US assessment of muscle mass is most often performed at the anterior compartment of the thigh. However, standardized anatomical landmarks and measuring points have now been proposed for 39 muscles/muscle groups.³⁹ Standardization of ultrasound measurement points is vital due to the absence of definitive data regarding the extent of heterogeneity in ultrasound parameters across the muscle bulk.¹¹⁶

According to recommendations, muscles should be assessed in a relaxed state. Patients should refrain from any physical activity for at least thirty minutes and should remain in a lying position for the last five minutes prior to examination.^39,117 This is important since physical activity or changing the body’s position may affect the fluid distribution and in turn the measured parameters.¹¹⁷

Studies have shown that the position of the patient as well as head of bed elevation during ultrasound assessment can significantly affect measurements, therefore using a standardized technique regarding positioning is vital during the examination.¹¹⁸ For follow up measurements, the same position should be used as during the first exam.

Depending on the muscle/muscle group being measured, patients should be examined in either supine or prone position. In clinical practice, some patients might find it difficult to lie in a prone position. Therefore, a sitting position, with knees and ankles bent in 90°, may be used to examine certain muscle groups, e.g. lower leg muscles and muscles of the head and neck.³⁹

Muscle fiber pennation angle and fascicle length should be measured while keeping the transducer probe perpendicular to the longitudinal axis of the muscle. Cross-sectional area and echo intensity should be determined from images obtained by the probe being placed parallel to the longitudinal muscle axis. Muscle thickness can in practice be measured using either probe placement. Some parameters, such as muscle thickness, can easily be measured during the examination, while others (e.g. echo intensity) will need to be additionally analysed post examination. The US images can be analysed using various open-source scientific image processing programs, e.g. ImageJ (https://imagej.nih.gov/ij).

A high-frequency linear ultrasound probe (5–10 MHz) is usually recommended for muscle assessment. A minimum transducer length of 5 cm seems advisable to visualise as much tissue as possible, particularly in subjects with a larger muscle bulk. Standard B-mode should be applied to visualize the different muscle components. A copious amount of ultrasound gel should be used to avoid excessive pressure on the muscle, which could affect the measurements. It is advised to keep the probe as perpendicular to the skin surface as possible.

Ultrasound system settings can be set to have the best possible view of the muscle that is to be assessed. However, since these settings may significantly affect measurements of certain parameters (particularly echo intensity), settings should be standardised for study purposes as well as for follow up examinations in the clinical setting, particularly for specific muscles/muscle groups. This is particularly important for follow up examinations and in research settings, to produce reliable and comparable results.

There is no consensus on whether the muscles of the dominant or non-dominant side of the patients are to be assessed. Studies have reported measurements from both dominant and nondominant sides, while other studies did not report this information. Since it is not clear whether the measurement side has clinical relevance, it is paramount for future research purposes and analysis to clearly indicate which side was assessed.

The accuracy of ultrasound parameters in assessing muscle function has been a subject of numerous studies. Assessment of muscle function presents one of the cornerstones of sarcopenia diagnosis. While muscle strength can be used as a strong indicator of sarcopenia, the deterioration of physical performance may serve as an indicator of disease severity. Therefore, determining the reliability of ultrasound parameters in assessing muscle function may provide further opportunity to incorporate ultrasound in sarcopenia diagnosis.

A significant association between ultrasound-measured forearm muscle thickness and hand grip strength was demonstrated in both young and elderly volunteers.^119,120 Other authors found a significant correlation between hand grip strength and ultrasound measurements of lower limb muscles as well.¹²¹ Seymour et al. demonstrated a positive correlation between rectus femoris cross-sectional area and knee extensor strength in COPD patients.⁶³

Ismail et al. found that ultrasound measurements of muscle quality were more strongly correlated with muscle strength than DXA-determined muscle mass.¹²² Thomaes et al. found significant correlation between maximal quadriceps muscle strength and ultrasound-measured muscle thickness of rectus femoris in patients with coronary artery disease.³²

In a study on 26 young and 26 older individuals, Strasser et al. reported a highly significant correlation between ultrasound-measured muscle thickness of quadriceps femoris muscles and maximal knee extensor strength. Muscle echogenicity was significantly higher in the older group. While no correlation was found between quadriceps muscle echogenicity and knee extensor strength in the older group, a negative correlation was found in the young cohort. Similarly, a correlation between pennation angle and knee extensor strength was found only in the vastus intermedius muscle in young individuals.⁷⁸

Conversely, Fukumoto et al. found that ultrasound-measured quadriceps femoris echo-intensity in a group of 92 older healthy Japanese women was negatively correlated with muscle thickness and knee extensor isometric strength, while a positive correlation between muscle thickness and knee extensor strength was demonstrated.¹²³ Similarly, Watanabe et al. demonstrated an inverse association between echo intensity of the quadriceps muscle and knee extension strength in elderly men.¹²⁴ Wilheim et al. also reported significant negative correlations between echo intensity of the quadriceps femoris muscles and both knee extension strength as well as the 30 second sit-to stand test in 50 healthy men.¹²⁵ Hirasawa et al. also found an association between echo intensity of the vastus lateralis muscle and knee extension strength in patients with type 2 diabetes.¹²⁶

The clinical correlations between site-specific muscle loss and physical performance remains poorly understood. A study on older communitydwelling women conducted by Abe et al. showed that an age-related loss of adductor/quadriceps muscles may be associated with a decrease in performance of more difficult tasks, such as zig-zag walking, yet is not significantly correlated with gait speed.¹²⁷ Similarly, Madden et al. found no correlation between vastus medialis muscle thickness and subject gait speed.¹²⁸ On the other hand, Mateos-Angulo et al. showed a significant negative correlation between echo intensity of lower limb muscles and gait speed as well as short physical performance battery test in older adults.¹²⁹ Osawa et al. concluded that echo intensity, but not muscle thickness, was associated with physical activity and functional mobility scores in the very elderly.¹³⁰

In a scoping review, Kitagawa et al. found poor to moderate associations between muscle echo intensity measurements of various muscles and functional performance tests. The authors concluded that the accurate effect size and causal inferences between muscle echo intensity and functional performance remained unclear.¹³¹ A meta-analysis of twenty-eight studies by Yuan et al. demonstrated moderate to strong correlations between echo intensity, muscle thickness and cross-sectional area with muscle strength. The authors also reported no significant association between ultrasound parameters and gait speed or Timed Up and Go test. Weak correlation was reported between echo intensity and muscle thickness with the sit-to-stand test.¹³² In another meta-analysis, Oranchuk et al. demonstrated the strongest correlation between quadriceps femoris echo intensity and knee extensor strength.¹³³

As of now, it is not yet clear whether any ultrasound parameter can be used as a strong indicator for physical performance. However, based on current research, qualitative parameters, such as echo intensity, might indeed be proven to be superior in this regard.

Despite having numerous advantages compared to other diagnostic methods, the use of ultrasound in muscle assessment still presents with several limitations. A large proportion of research in the field of US muscle assessment has been done on healthy populations, which does not necessarily reflect the dynamics of muscle parameter changes in different patient populations. Additionally, ultrasound parameters regarding muscle quality and quantity do not have clearly defined reference values. To establish this, large-scale longitudinal studies both on healthy people and patient cohorts should be done, which would enable satisfactory differentiation between normal and sarcopenic states. Third, muscle quantity and quality values could be subject to change, particularly due to systemic inflammation, muscle damage or changes in fluid balance, vascular permeability or glycogen levels.^134,135 Additionally, the exact spreading pattern and evolution of the different architectural components throughout the muscle is still unclear, as well as the exact influence of pre-investigation physical activity and patient position on the accuracy of measurements. Also, examiners require a certain level of experience and/or training. Intraand inter-rater reliability has been a subject of concern, particularly for muscle quality parameters (e.g. echo intensity), while large muscles, such as the quadriceps, are quite easily quantitatively assessed. The influence of ultrasound system settings on measurements should also be considered.

Furthermore, it is still unclear which muscle group parameters correlate best with overall muscle quantity and quality. While a lot of research has been focused on large muscle groups, smaller muscles have received relatively little attention yet may be of equal interest due to their specific functions.³⁹ Additionally, while most of the research focused on the diagnostic utility of ultrasound parameters of superficial muscles, some studies have included measurements of deeper muscles, most commonly vastus intermedius and the soleus muscle. The depth of muscle tissue being assessed may influence the accuracy of measurements, particularly of muscle quality (e.g. echo intensity). Therefore, several authors have encouraged the use of a correction factor considering subcutaneous fat thickness, although results using this method have been conflicting. In obese patients with a deeper layer of subcutaneous fat, the ultrasound image depth and focus may need to be adapted to improve visualisation, and the use of correction factors should be considered. At the same time, it is paramount to establish standardisation of both ultrasound system settings and methodology of measurements.

Furthermore, the validity of ultrasound-derived prediction equations for the estimation of muscle mass in older adults is also yet to be established, due to the lack of definitive results from studies.⁹ Therefore, it is vital to harmonize ultrasound measurement protocols and establish cutoff values for sarcopenia diagnosis, which would also provide more reliable metadata in the future.