Key words
sarcopenia - radiological screening - body composition analysis - quantitative imaging
- segmentation
Definition
The term “sarcopenia” is composed of the two Greek words “sarx meaning flesh” and
“penia meaning loss”. Sarcopenia refers to age-related progressive and generalized
loss of muscle mass and strength. The course of this primary aging process can be
intensified by comorbidities and physical inactivity [1]. Sarcopenia can also be present in children and adolescents, e. g., as part of a
tumor disease like hepatoblastoma, long-term steroid therapy, muscular dystrophy,
or chronic liver disease [2]
[3]
[4]. [Table 1] provides a list of potential risk factors for sarcopenia. In addition to functional
limitations, sarcopenia often causes an increase in trauma/falls and resulting injuries
that can further limit quality of life. Approximately 20 % of 70-year-olds and about
50 % of 75-year-olds are affected by sarcopenia [5]. With respect to gender distribution, the specified or estimated prevalence is higher
in men or in women depending on the definition and population [6]. There are already some studies with large case numbers addressing the prevalence,
risk factors, and screening of sarcopenia [7]
[8]
[9]
[10]. According to the “UK Biobank” study including 168 682 participants, pre-sarcopenic
men and sarcopenic women have an elevated risk of osteoporosis [10]. In their study, Soh and Won examined the relationship between sarcopenia and fall
risk in older Korean adults. They used data from the “Korean Frailty and Aging Cohort
Study” [11]. As a result of demographic changes in population structure, sarcopenia will play
an increasing role in the future.
Table 1
Risk factors for sarcopenia and their significance.
|
Age
|
Often begins in middle age and advances with increasing age
|
|
Gender
|
Depending on the classification that is used, the prevalence is higher in women or
men (reference Petermann-Rocha et al., Morley et al.)
|
|
Malnutrition
|
Especially decreased protein intake. The need for protein increases with age
|
|
Lack of physical activity
|
Often due to diseases like osteoarthritis/other types of degeneration or as a result
of tumor diseases. This results in a vicious cycle that further aggravates muscle
atrophy
|
|
Hormone deficiency
|
Particularly testosterone or estrogen deficiency
|
|
Inflammatory processes
|
Chronic inflammatory processes like in rheumatoid arthritis promote muscle atrophy
|
|
Tumor diseases
|
Often a consuming disease and side effects of treatment resulting in undernourishment/malnourishment
Due to longer hospitalization and/or surgical interventions, there is a risk of progressive
physical inactivity
|
|
Muscular dystrophy
|
Rare cause of progressive muscular weakness and decreased muscle mass due to genetic
mutations
|
In addition to sarcopenia, patients often also have an elevated fat mass. So called
“Sarcopenic obesity” is a special type of obesity. Muscle mass normally increases
in response to an increase in weight load. This adaptive mechanism can be disrupted
particularly in older people [12]. In addition, reduced energy consumption does not necessarily result in a decrease
in appetite [13]. Consequently, these patients have lower muscle mass and strength in relation to
the increased fat mass. There is a significant prevalence of sarcopenic obesity also
in children and adolescents. According to a review, the prevalence is 5.7 % to 69.7 %
in girls and 7.2 % to 81.3 % in boys. A connection with cardiometabolic events, severity
of non-alcoholic fatty liver disease, inflammation, and mental health has also already
been described [14].
The term “cachexia” must be differentiated from sarcopenia. The diagnostic criterion
for cachexia is weight loss (fat and muscle mass) of more than 5 % in the last 6 months
or of more than 2 % in people who are already underweight (body mass index [BMI] < 20 kg/m²)
or have sarcopenia. According to the consensus of an international group of experts,
sarcopenia in tumor patients is a fundamental part of cachexia and an important element
in the evaluation of tumor patients [15]
[16]. The difference between cachexia and sarcopenia in tumor diseases is that sarcopenia
as well as weight loss must be present in the cachexia [15].
Diagnosis
There are various screening methods to diagnose sarcopenia, which are listed, for
example, by the European Working Group on Sarcopenia in Older People (EWGSOP)
[17]. In general, these can be categorized as direct and indirect methods [18]. The indirect measurement methods include the detection of molecular and cellular
changes in the skeletal muscles, e. g. based on biomarkers. The negative effects of
sarcopenia can also be analyzed. This includes, for example, reduced quality of life,
increased fall risk, and an increased hospitalization rate [19].
Radiological methods
There are various radiological methods for diagnosing sarcopenia like dual X-ray absorptiometry
(DXA) or cross-sectional imaging methods like computed tomography (CT) and magnetic
resonance imaging (MRI). [Table 2] provides an overview. The most important methods are described in greater detail
below.
Table 2
Imaging techniques available to detect sarcopenia (modified from Mohamed Ali et al.
[6]). Additional presentation of advantages and limitations. DXA: dual-energy-X-rax
absorptiometry, CT: computed tomography, MRI: magnetic resonance imaging.
|
Imaging
|
Advantages
|
Limitations
|
|
DXA
|
|
|
|
CT
|
-
Quantification of the musculature and the degree of fatty changes
-
Can be used for the entire body or for individual muscles
-
Evaluations can also be performed in the case of examinations for other medical questions
|
|
|
MRI
|
|
-
Expensive and not widely available
-
Long examination times and technically challenging
-
No cutoff values for the diagnosis of sarcopenia
|
|
Ultrasound
|
|
|
Dual X-ray absorptiometry (DXA)
DXA is a radiodiagnostic method with comparably low radiation exposure and is the
most commonly used method for analyzing body composition [20]. This measurement method is based on the different absorption rates of low and high-energy
X-rays in mineralized tissue, fat, and soft tissue. [21].
With DXA, the following values can typically be determined: lean mass (LM), fat mass
(FM), bone mineral content (BMC), and areal bone mineral density (aBMD). The LM is
the measurement of all types of tissue that are neither fat nor bone. The sum of the
LM of the upper and lower extremities, known as the appendicular LM (ALM), is used
to determine the muscle mass. The correlation with body size is used to calculate
the ALM index (ALMI = ALM/body size2). The EWGSOP specifies an ALMI of < 6 kg/m2 for women and < 7 kg/m2 for men as cutoff values for reduced muscle mass [20]. [Fig. 1] shows examples of DXA images.
Fig. 1 Dual X-ray absorptiometry (DXA) of two female patients. A A 74-year-old obese female patient with an appendicular lean index (ALMI) of 4.24 kg/m2 (red arrow). According to the European Working Group on Sarcopenia in Older People
(EWGSOP), sarcopenia is, therefore, present (cut-off value for women < 6 kg/m2). B A 40-year-old female patient without sarcopenia with an ALMI of 8.44 kg/m2 (red arrow).
The limited comparability of the results from different manufacturers is a limitation
of DXA. Compared to CT and MRI, it provides only two-dimensional images. Moreover,
DXA does not allow a statement about the qualitative composition of muscle, e. g.
in relation to fat deposits.
Computed tomography
Using CT as the most widely available cross-sectional imaging method, skeletal muscle
in different regions of the body can be analyzed. In addition to the muscle area and
volume, the density values of individual muscles or muscle groups can be determined
[22]. The segmentation of the musculature and the determination of the cross-sectional
area (CSA) are typically performed at the level of L3 or L4. The psoas muscle alone
is analyzed at this level or the entire muscle area is analyzed at the corresponding
level [23]. The abdominal wall musculature, the psoas muscle, the autochthonous back muscles,
and the quadratus lumborum muscle are segmented for this purpose. In addition, the
subcutaneous and visceral fat tissue can also be quantified. In principle, muscle
segmentation is performed manually. Alternatively, masks with predefined Hounsfield
units (HU) can also be used. Goodpaster et al. defined a range between 0 and 100 HU
for the musculature, while Mitsiopoulos et al. used a density range of –29 to 150
HU [24]
[25]. A direct comparison of the two HU ranges showed significant differences in the
calculated muscle areas [26]. Due to the larger HU range, interstitial adipose tissue in the muscle tissue is
also taken into consideration. In contrast, when using the lower HU range, only the
adipose tissue-free muscle (ATFSM) is taken into consideration. In healthy young adults,
anatomical skeletal muscle is only slightly greater than the ATFSM. However, the IAT
increases with age, in the case of obesity, and also in the case of certain diseases,
e. g. muscular dystrophy [25]. For this reason, the range of –29 to 150 HU has become established.
Moreover, the direct comparison of the skeletal muscle areas calculated from non-contrast
and contrast-enhanced CT examinations resulted in some significant differences depending
on the HU range being used [26]. The range between –29 and 150 HU yielded the most reliable results also in this
study. Therefore, the HU range must be taken into consideration in the evaluation,
interpretation, and comparison of results.
The muscle areas calculated based on a CT scan correlate very well with the total
body muscle mass [27]
[28]. The ratio of muscle area to body size determines the skeletal muscle index (SMI)
(SMI = CSA/body size2). For the diagnosis of sarcopenia, the EWGSOP currently only defines cutoff values
for DXA and the bioelectric impedance analysis (BIA) [17]. Some studies also specify corresponding cutoff values for the SMI [23]
[29]
[30]. In tumor patients, the CT data of Prado et al. and Martin et al. were used most
frequently as a reference. The data of Prado et al. was based on the analysis of a
total of 2115 patients with solid tumors of the gastrointestinal tract or the respiratory
system [29]. In the study by Martin et al., the skeletal muscle index (SMI) of 1473 patients
with a malignancy of the lung or the gastrointestinal tract was determined [30]. There are already age- and gender-specific percentile curves for the total psoas
muscle area (tPMA) for children [31].
An advantage of these indices is that the muscle mass can be determined in all CT
examinations of the trunk in addition to the primary clinical question and the corresponding
calculations can be performed (secondary use of CT data). CT is suitable particularly
in tumor patients for diagnosing sarcopenia since it is often performed already during
diagnosis and subsequently in defined intervals for evaluating treatment response.
The muscles can only be analyzed at the indicated level (L3 and L4) in abdominal examinations.
For this reason, Derstine et al. evaluated examination of skeletal muscle from Th10
to L5 [32]. As a result, sarcopenia diagnosis can also be performed during chest CT. An additional
advantage of abdominal CT is that skeletal muscle as well as the abdominal fat distribution
are determined (subcutaneous and intra-abdominal fat tissue). These supplementary
measurements thus allow a body composition analysis based on DXA [33].
An example of an analysis of body composition based on routine CT is shown in [Fig. 2].
Fig. 2 Body composition analysis at L3 level of a 67-year-old patient with an AEG tumor
before A, C and after B, D neoadjuvant chemotherapy with a marked reduction of skeletal muscle area and subcutaneous
and visceral adipose tissue. Skeletal muscle area decreased from 123.40 cm2 to 65.06 cm2. Skeletal muscle index decreased from 36.85 cm2/m2 to 19.43 cm2/m2. Skeletal muscle area is red, subcutaneous adipose tissue is blue, and visceral adipose
tissue is green. Segmentation was performed semiautomatically with an own Python application
based on SimpleITK. For skeletal muscle, Hounsfield unit (HU) mask was between –5
and 135 HU and for adipose tissue between –190 and –30 HU.
One of the limitations of CT is the higher dose in comparison to DXA. According to
the German Commission on Radiological Protection, the applied effective dose of whole-body
DXA is 1–10 µSv [34]. The typical effective dose for CT of the abdomen and pelvis is approximately 11 mSv
[35]. However, this can vary based on individual factors, such as sex, age, and constitution.
This disadvantage is balanced out when acquisition is performed as part of examinations
regarding other medical questions, e. g. in routine staging examinations for tumor
patients as mentioned above.
Magnetic resonance imaging (MRI)
In addition to the advantage of excellent soft-tissue contrast, MRI can be used as
a radiation-free alternative. In addition to the targeted imaging of individual muscles
or muscle groups, whole-body examinations are also possible. Moreover, MRI allows
not only qualitative visualization, e. g. fatty infiltration or fibrosis, but also
quantitative determination of muscle mass and fat mass [36]. According to Pons et al., using MRI to determine the volumetry of muscles has proven
to be a valid and reliable method. In addition to manual segmentation, (semi-) automated
segmentation of certain muscles is also possible [37].
In addition to muscle segmentation, the Dixon method and MR spectroscopy (MRS) are
special techniques for the diagnosis of sarcopenia. Fatty infiltration of skeletal
muscle is an important factor in the limited mobility of patients with sarcopenia
[38]
[39]
[40]. The DIXON method can be used to determine fatty infiltration of skeletal muscle.
The various resonance frequencies of water and fat are used to analyze the percentages
of fat and water based on all proton signals. With the original DIXON sequence, two
echoes are acquired, one with water and fat in-phase (IP) and one with water and fat
opposed-phase (OP) [41]. The fat and water images can be generated by adding and subtracting the OP and
IP. Qualitative fat detection can be performed with the original DIXON method based
on the acquired images. In the clinical routine, this method is used, for example,
for the differentiation of adrenal tumors. The further development of the DIXON method
uses echoes at multiple time points and is referred to as the multi-echo (ME) method.
This approach allows direct water and fat quantification based on the calculation
of parametric maps. In clinical studies, the ME technique has already been used for
the quantification of liver fat [42]
[43]. In addition it allows analysis of fat distribution in the musculature via mapping
[44]
[45]. [Fig. 3] shows an example of fat quantification using the ME-Dixon method.
Fig. 3 Magnetic resonance imaging (MRI) with multi-echo (ME) Dixon technique. Calculated
water-only image A, fat-only image B, and results from the Dixon dataset to define fat fraction C with segmentation of the psoas muscle. Area of the right psoas muscle is 12.0 cm2 and 8.40 cm2 on the left side. Fat fraction is 7.5 % on the right side and 16.3 % on the left
side.
MRS is a further option for investigating sarcopenia [21]. In contrast to MR imaging, the result of spectroscopy measurement is not a cross-sectional
image but an intensity spectrum of frequency signals [46]. These volume-selective measurements make it possible to examine metabolic processes
in the human body, e. g., in skeletal muscles. Particularly with 31P-spectroscopy, the spectrum of phosphor metabolites and the changes in the concentrations
of these metabolites during muscular work can be analyzed [47], for example in patients with type II diabetes or peripheral arterial disease [48]
[49]
[50]. A relationship between changes in skeletal muscle metabolism and decreasing muscle
mass or changes in muscle function on MRS has also been described in connection with
sarcopenia. Additional studies are needed to confirm and further investigate these
results.
Reproducibility and intermodal concordance between MRI and CT in abdominal muscle
segmentation has already been shown in patients with renal cell carcinoma [51]. Examples of qualitative muscle changes are shown in [Fig. 4], [5].
Fig. 4 Magnetic resonance imaging (MRI) of a 45-year-old patient diagnosed with malignant
peripheral nerve sheath tumor (MPNST). Transverse T2 turbo spin echo (TSE) sequence
of the left femur 1 year after A and 6 years after B distal femoral amputation. There is a progressive reduction in the volume of the
thigh muscles and fat deposits, which is most pronounced in the semitendinosus muscle
(arrows).
Fig. 5 Magnetic resonance imaging (MRI) of a 36-year-old female patient diagnosed with Ewing’s
sarcoma. Transverse T2 turbo spin echo (TSE) sequence of the right thigh before A and 3 years after B resection. There is a reduction in the volume of the thigh muscles over time most
markedly of the adductor magnus muscle (arrows) and the gracilis muscle (arrowhead).
The musculature shows increasing streaky fat deposits.
In addition to cost and the sometimes limited availability, the long examination times
compared to other methods are limitations of MRI. Moreover, there are no absolute
cutoff values for the definition of sarcopenia. The method is normally used within
the framework of research at specialized centers.
Ultrasound
As an inexpensive, widely available, and radiation-free method, ultrasound represents
an alternative with good reproducibility [52]. Ticinesi et al. were able to determine the volume of the entire quadricep muscle
by determining the cross-section of the rectus femoris muscle. In addition, there
was very good correlation with MRI measurement [53]. The lack of standardization of ultrasound examination and the partly examiner-dependent
quality of implementation are limitations of the method. Moreover, if too much pressure
is applied to the ultrasound transducer, the muscle compartments can be overly compressed,
resulting in incorrectly small muscle volumes.
Non-radiological diagnostic methods
Non-radiological diagnostic methods
Further non-radiological diagnostic methods include bioelectric impedance analysis
(BIA), electromyography (EMG), determination of potassium level, and anthropometric
measurements, e. g., the circumference of the upper arm.
Artificial intelligence in sarcopenia diagnosis
Artificial intelligence in sarcopenia diagnosis
In some studies, muscle and fat tissue in CT and MRI datasets has already been analyzed
using artificial intelligence (AI) [9]
[54]
[55]
[56]
[57]. In their study including 1143 CT datasets, Nowak et al. used two neuronal networks
[57]. The CT scan to be analyzed at the level of L3 / L4 was selected with the first
neuronal network and the skeletal muscle and fat tissue were segmented with the second.
There was significant agreement between manual analysis and automatic analysis using
the two neuronal networks. Pickhardt et al. used an automated deep-learning approach
to analyze skeletal muscle at the level of the first and third lumbar vertebral body
from 9223 CT datasets [9]. The acquired data on sarcopenia and particularly on fatty infiltration of muscle
was comparable to clinical risk scores in the prediction of hip fractures and mortality.
As a result of an AI-based evaluation, the time expenditure for segmentation can be
reduced and the additionally acquired data can be included in the radiology report.
Radiomics analysis of skeletal muscle represents another approach to sarcopenia diagnosis
[58]. In one study radiomics was used to detect sarcopenia from the CT datasets of 247
patients with small-cell bronchial carcinoma. A machine learning model was used for
the analysis. However, additional studies are needed to further examine this promising
approach.
Clinical significance
As the world population ages, the frequency of sarcopenia will increase significantly,
with additional negative consequences. However, age-independent severe diseases like
malignancies, COPD, chronic heart or kidney diseases can cause a secondary loss of
muscle mass and strength. Early screening and intervention can significantly lower
costs for the public health care system.
Diverse clinical applications result from the described modality-based methods for
diagnosing sarcopenia. Some examples involving tumors and inflammatory/infectious
diseases are discussed in the following.
Tumor
Sarcopenia diagnosis is mainly used in tumorous diseases. Particularly patients with
malignancies often suffer from pronounced weight loss and the resulting consequences
due to the disease itself or treatment-associated side effects. In addition, major
surgeries and long hospital stays pose a risk in this patient population in particular.
In a group of patients with malignancies of the upper gastrointestinal tract, a prevalence
of sarcopenia of 11.5 % was seen with DXA based on the cutoff values according to
Suetta in a Danish reference collective [59] or 19.1 % compared to an Australian reference collective [60]. However, there was a significant discrepancy with respect to the cutoff values
for sarcopenia diagnosis based on CT in the same collective [61]. In detail, there was an average difference in the quantification of lean tissue
of 1.4 kg [61]. In particular, CT imaging is often used in this connection to determine body composition
since staging and follow-up examinations in patients with tumors are performed repeatedly
over the course of the disease so that the focus is on the secondary use of CT data.
In relation to the CT-based SMI for diagnosing sarcopenia, it was shown that particularly
the cutoff values of Martin et al. [30] and Prado et al. [29] were used in the available studies in oncology patients with a percentage of approx.
30 % and 45 %, respectively [62]. Patients with colorectal cancer were examined most frequently in the studies to
date, with a prevalence of low SMI values being seen in 46.0 % of cases (median percentage
of patients with a low SMI: 41.1 % in the curative cohort, 49.1 % in the cohort without
a curative approach) [62]. However, the highest prevalence of a low CT-based SMI was seen in patients with
esophageal and pulmonary cancer (49.8 % and 49.5 %, respectively), with the cohorts
with non-curative cancer being characterized by a higher prevalence comparable to
colorectal cancer [62]. For the individual tumor entities, there was a prevalence of low SMI values between
35 % and 50 %, which was approximately comparable among the individual tumor types,
cutoff values, and disease stages, so that reduced muscle mass or muscle quality based
on CT-based SMI values seems endemic among oncology patients [62]. It should be mentioned here that a reduced SMI – together with the slightly more
rarely used skeletal muscle density (SMD) – seems to have a negative effect on the
survival of oncology patients and thus has direct clinical relevance [63]
[64]
[65]. 38 studies with a total of 7843 patients with a diagnosis of a solid tumor were
included in a meta-analysis by Shachar et al. [63]. The tumor diseases most commonly examined in the studies were hepatocellular carcinoma
(n = 11), pancreaticobiliary tumors (n = 6), gastroesophageal tumors (n = 4), urothelial
carcinomas, renal cell carcinomas, and colorectal cancers (n = 3 in each case). In
all included studies, the SMI determined during diagnosis was a negative predictive
factor for survival. This was true for patients with and without metastases. In a
retrospective analysis in patients with pancreatic cancer undergoing first-line chemotherapy,
Kim et al. determined the SMI, SMD, and presence of sarcopenia [65]. A low SMI or SMD was a negative prognostic factor for survival. This effect was
even greater if both the SMI and SMD were low. Side effects of chemotherapy were also
observed more frequently in patients with a low SMI. In contrast, the broader use
of MRI-based methods for diagnosis or follow-up imaging of sarcopenia is still largely
absent. In a study including patients with various primary oncological diseases, there
was a strong positive intermodal correlation between the CSA and the paraspinal muscular
fatty infiltration according to CT and MRI [55]. Particularly the good correlation between the CS-MRI-based quantification of muscle
fat content and density values from CT imaging seems promising for being able to opportunistically
acquire comparatively valid markers from CT imaging that are not inferior to MRI [55]. DTI or MRS could allow further characterization of compartments affected by sarcopenia,
but these are not yet used on a representative basis particularly in patients with
tumor diseases.
Inflammation/infection
Another large patient group of interest for sarcopenia diagnosis includes inflammatory
and infectious diseases. These also promote catabolic metabolic reactions and result
in a decrease in muscle mass that is also intensified by long periods of inpatient
or outpatient bed rest [66]
[67].
Particularly in chronic inflammatory diseases like rheumatic diseases or in chronic
inflammatory bowel diseases, the correlation with sarcopenia has been shown usually
by DXA measurements in multiple studies like the systematic analysis by An et al.
[68].
With respect to inflammatory changes, Modesto et al. examined a healthy control group
and three patient groups with various stages of pancreatitis [69]. An initial episode of acute pancreatitis was differentiated from a recurrent acute
form of chronic pancreatitis. A modified psoas index was used in the analysis. The
group was able to show that the muscle volume of the psoas musculature is a suitable
biomarker to allow timely identification of the transition from recurrent acute to
chronic pancreatitis so that corresponding therapeutic measures can be initiated early.
However, there was a lack of a cutoff value for the absolute muscle volume so that
only comparative analyses between the patient groups were possible. In a further study,
a relationship between the loss of muscle mass and mortality with a month of hospitalization
was able to be shown in patients with necrotizing pancreatitis [70]. [Fig. 6] shows an example of a decrease in skeletal muscle in a patient with necrotizing
pancreatitis.
Fig. 6 Computed tomography (CT) of a 68-year-old patient with necrotizing pancreatitis at
the time of diagnosis A, C and four weeks later under therapy B, D. A significant reduction in subcutaneous (blue) and visceral (green) adipose tissue
can be seen. Skeletal muscle area (red) decreased from 179.32 cm2 to 162.69 cm2. Skeletal muscle index showed a moderate decrease from 53.24 cm2/m2 to 47.44 cm2/m2. Segmentation was performed semiautomatically with an own Python application based
on SimpleITK. For skeletal muscle, the Hounsfield unit (HU) mask was between –5 and
135 HU and for adipose tissue between –190 and –30 HU.
Multiple studies on sarcopenia have also recently been published with regard to the
novel disease COVID-19. Gualtieri et al. were able to use CT to document a decrease
in muscle mass during a stay in the ICU [71]. A prediction about the duration of hospital stay, intubation, and mortality in
COVID-19 patients can be made based on the pectoralis muscle area [72]. [Fig. 7], [8] show a decrease in skeletal muscle as a result of a severe COVID-19 infection.
Fig. 7 Computed tomography (CT) of a 44-year-old patient with COVID-19 pneumonia and severe
acute respiratory distress syndrome (ARDS) at baseline A, C and five weeks later B, D. The patient received intensive care treatment including invasive ventilation and
a veno-venous extracorporeal membrane oxygenation (ECMO). There is a decrease in skeletal
muscle area at the level of T12 from 99.38 cm2 to 87.13 cm2. Segmentation was performed semiautomatically with an own Python application based
on SimpleITK. For skeletal muscle, the Hounsfield unit (HU) mask was between –5 and
135 HU and for adipose tissue between –190 and –30 HU.
Fig. 8 Comparison of the area of the pectoralis major et minor muscle measured directly
above the aortic arch at baseline A and five weeks later B. The measured area of the pectoralis major et minor muscle showed a decrease from
20.14 cm2 to 14.95 cm2 on the right C, D and from 26.23 cm2 to 16.81 cm2 on the left E, F. Segmentation was performed semiautomatically with an own Python application based
on SimpleITK. For skeletal muscle, the Hounsfield unit (HU) mask was between –5 and
135 HU.
Complications, length of hospitalization, morbidity
Another interesting option for sarcopenia diagnosis is the possibility to better estimate
the risk of postoperative complications based on the perioperative determination of
muscle mass. This information can lead to better personalization of the indication
for surgery with more comprehensive physiotherapeutic preparation and workup if needed.
Jang et al. examined 284 patients prior to planned pancreatic surgery [73]. Preoperative determination of the muscle area standardized to body weight on abdominal
CT was able to show that highly significant and often feared postoperative pancreatic
fistulas requiring correspondingly long inpatient treatment and resulting in immobilization
are seen in sarcopenia patients. In contrast, other examined parameters such as the
preoperative diameter of the main pancreatic duct as is typically used for risk assessment
did not have any effect on the postoperative formation of fistulas. Moreover, a correlation
between the psoas index in older patients and in trauma patients and morbidity, duration
of hospitalization, and complication rate during inpatient care was able to be shown
[74]
[75].
Prophylaxis and treatment
Prophylaxis and treatment
Physical activity is the most important intervention in connection with prophylaxis
and the treatment of sarcopenia. Physical activity has a positive effect on muscle
mass, muscle strength, and physical function by mitigating age-related loss [76]
[77].
Although there is currently no specific treatment for sarcopenia, it is usually reversible.
The goal is to improve muscle mass, strength, and performance. Particularly in the
case of early intervention, the atrophy processes can be actively counteracted by
individualized physical training and proper diet [7]
[78]. Due to the targeted use of the muscles, age-appropriate and individualized progressive
strength and resistance training is particularly suitable for prevention and treatment
[79]
[80]. In addition to muscle strength, stamina can also be improved. Intensive daily activities
(e. g. housework and yard work) also help to optimize the musculature and quality
of life. To minimize the fall risk, training should also include balance training
[81]. To improve muscle protein synthesis, a personalized protein regimen with a sufficiently
high leucine percentage (e. g. present in whey protein) is recommended [82]
[83].
Conclusion/outlook
Due to changing demographics, sarcopenia as a chronic disease will become increasingly
important. Sarcopenia is associated with a negative effect on the course of diverse
diseases frequently in combination with an increased hospitalization rate and morbidity
not only in older patients. For numerous tumor diseases, sarcopenia was able to be
identified as a negative prognostic factor. Detection and follow-up can be performed
with various radiological methods. CT plays an important role since it is often performed
in the framework of other medical questions, e. g., in the routine staging of tumor
patients. Data and information regarding sarcopenia can be acquired at the same time.
The increasing use of AI-based segmentation of the skeletal muscles can additionally
reduce the time expenditure. The results can be included in the radiology report on
a supplementary basis. The planning of individualized treatment and follow-up can
help to improve the course of the disease. The analysis of radiomics data regarding
the skeletal muscles in sarcopenia diagnosis was already examined in patients with
non-small-cell bronchial carcinoma [58]
[84]. Dual-energy CT techniques and photon-counting CT are additional possibilities for
sarcopenia diagnosis.
Established absolute reference values are a major requirement to be able to make generally
valid statements regarding sarcopenia diagnosis in the clinical routine. Some study
results on this topic are already available [23]
[29]
[30]
[31]. Large population studies, e. g. the UK Biobank study and the NAKO study [85]
[86], can provide information in this regard. Whole-body MRI examinations of approximately
30 000 participants were acquired in the NAKO study and can be used to establish reference
values or to analyze MR radiomics.
It will certainly not take much longer for sarcopenia diagnosis to become established
as a fixed variable in the therapeutic decision tree, at least in tumor patients.
In the future not only sarcopenia screening but also the early detection of risk factors
may become more important. For example, certain radiomics analyses could act as potential
biomarkers, particularly in tumor patients.
-
Sarcopenia is a primarily age-dependent syndrome that can manifest to a greater degree
in patients with malignant tumor diseases.
-
Negative effects on the course of the disease in tumor patients can be prevented by
early detection and individualized treatment.
-
In radiological diagnosis, computed tomography (CT) has the advantage of being able
to be used to acquire additional parameters regarding sarcopenia (opportunistic use).
-
Comprehensive use requires generally accepted and established reference values.
-
Data can be quickly evaluated and implemented in the radiology report with artificial
intelligence (AI).
-
Radiomics analysis, dual-energy CT, and photon-counting CT are further options for
sarcopenia diagnosis.
