Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Critical Care Explorations www.ccejournal.org 1
Critical Care
Explorations
Crit Care Expl 2020; 2:e0171
DOI: 10.1097/CCE.0000000000000171
Skeletal Muscle Composition Predicts
Outcome in Critically Ill Patients
Sven H. Loosen, MD1,2; Maximilian Schulze-Hagen, MD3; Tobias Püngel, MD1; Lukas Bündgens, MD1;
eresa Wirtz, MD1; Jakob N. Kather, MD1; Mihael Vucur, PhD4; Pia Paenholz, MD5; Münevver Demir, MD6;
Philipp Bruners, MD3; Christiane Kuhl, MD3; Christian Trautwein, MD1; Frank Tacke, MD, PhD6;
Tom Luedde, MD, PhD2,4; Alexander Koch, MD1; Christoph Roderburg, MD6
Objectives: Parameters of patients’ body composition have been sug-
gested as prognostic markers in several clinical conditions including
cancer and liver transplantation, but only limited data on its value in
critical illness exist to date. In this study, we aimed at evaluating a
potential prognostic value of the skeletal muscle mass and skeletal
muscle myosteatosis of critically ill patients at admission to the ICU.
Design: Exploratory observational cohort study.
Setting: An urban, academic medical institution.
Patients: One-hundred fifty-five patients treated for critical illness on
a medical ICU.
Interventions: None.
Measurements and Main Results: We used routine CT scans to
assess the patients’ individual body composition. The skeletal muscle
index as a surrogate for sarcopenia was defined as the total skel-
etal muscle area at the level of the third lumbar vertebra on axial CT
scan, normalized for the patient’s height. Myosteatosis was evalu-
ated by assessing the mean skeletal muscle attenuation measured
in Hounsfield unit at the same sectional plane. The skeletal muscle
index and mean skeletal muscle attenuation at admission to the ICU
were significantly higher in patients with long-term survival (180-day
or 1-year mortality), while both parameters were comparable between
short-term survivors and nonsurvivors (ICU mortality or 30-d mortal-
ity). Patients with a skeletal muscle index or mean skeletal muscle
attenuation below our established ideal cutoff values (74.95 mm2/
cm and 29 Hounsfield unit) showed a significantly reduced overall
survival. These findings were confirmed in univariate and multivariate
Cox regression analyses. Furthermore, myosteatosis significantly cor-
related with the time of mechanical ventilation, the duration of hospital
stay, and the presence of sepsis.
Conclusions: Our data suggest that sarcopenia and myosteatosis rep-
resent important prognostic factors in critically ill patients that can be
easily obtained from routine CT scans. Both parameters at admission to
the ICU yield important information on the patients’ long-term outcome
and might be used for early clinical decision-making in these patients.
Key Words: biomarker; cachexia; intensive care unit; prognosis;
sarcopenia
Despite intensive research eorts, prediction of long-term
prognosis in critically ill patients has remained poor. In
this context, the body composition was identied as a
potential factor determining the outcome of patients treated on
an ICU. Dierent authors have suggested a link between low skel-
etal muscle mass and patients´ ICU mortality (1–3). However, the
term “low skeletal muscle mass” is only poorly dened and oen
used synonymous to sarcopenia. e latter is dened as the “pro-
gressive loss of muscle mass and strength with a risk of adverse
outcomes such as disability, poor quality of life and death” by
the Special Interest Group of the European Sarcopenia Working
Group in 2010 (4). Since both muscle strength and muscle mass
are dicult to measure in clinical routine, many authors have
dened sarcopenia as a skeletal muscle area index below the 5th
percentile of the of a healthy control population (5). However,
using this denition, the role of sarcopenia as a prognostic marker
Copyright © 2020 The Authors. Published by Wolters Kluwer Health, Inc.
on behalf of the Society of Critical Care Medicine. This is an open-access
article distributed under the terms of the Creative Commons Attribution-Non
Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permis-
sible to download and share the work provided it is properly cited. The work
cannot be changed in any way or used commercially without permission from
the journal.
Original Clinical Report
1Department of Medicine III, University Hospital Rheinisch-Westfälische
Technische Hochschule Aachen, Aachen, Germany.
2Clinic for Gastroenterology, Hepatology and Infectious Diseases, University
Hospital Düsseldorf, Medical Faculty of Heinrich Heine University Düsseldorf,
Düsseldorf, Germany.
3Department of Diagnostic and Interventional Radiology, University Hospital
Rheinisch-Westfälische Technische Hochschule Aachen, Aachen, Germany.
4Division of Gastroenterology, Hepatology and Hepatobiliary Oncology,
University Hospital RWTH Aachen, Aachen, Germany.
5Department of Urology, University Hospital Cologne, Cologne, Germany.
6Department of Hepatology and Gastroenterology, Charité University Medicine
Berlin, Berlin, Germany.
Loosen et al
2 www.ccejournal.org 2020 • Volume 2 • e0171
for critically ill patients remains controversial (5), which might be
due to the fact that functional aspects such as muscle strength and
muscle composition are not properly reected. Recently, Faron et
al (6) showed the clinical potential of fat-free muscle area (FFMA)
in MRI scans to predict outcome in patients with colorectal can-
cer, suggesting that FFMA might be a new and easily accessible
prognostic biomarker for prognosis.
In this study, we aimed at analyzing whether sarcopenia,
assessed by the skeletal muscle index (L3SMI), and the mean skel-
etal muscle attenuation (MMA) as a surrogate for muscular fat
deposition, represent prognostic factors in critically ill patients
treated on a medical ICU.
PATIENT AND METHODS
Study Design and Patient Characteristics
A total of 155 patients who were admitted to the medical ICU at
the Department of Medicine III at University Hospital Rheinisch-
Westfälische Technische Hochschule (RWTH) Aachen between
2006 and 2015 were enrolled in this study (detailed patient char-
acteristics are given in Tab le 1). e study was approved by the
local ethics committee (EK 150/06) of the University Hospital
RWTH Aachen, Germany, and written informed consent was
obtained from every participant or authorized relatives in the case
of unconsciousness.
Assessment of Body Composition and Definition of the
Skeletal Muscle Index
We only used CT scans in venous phase with a slice thickness of
5 mm for this study. We included CT scans that were performed
upon ICU admission. e total skeletal muscle area, as well as the
MMA, were segmented at the center plane of the third lumbar ver-
tebra on axial CT scans. We used this approach as it was recently
shown that single-slice measurements of the skeletal muscle area
and adipose tissue at anatomical landmarks are strongly associated
with total compartment volumes and therefore provide compara-
bility of results (7). All parameters were assessed manually using
a semi-automatically segmentation tool, 3D slicer, an open source
soware platform for medical image informatics (8). e following
muscle groups were segmented: rectus abdominis, external and
internal obliques, transversus abdominis, quadratus lumborum,
as well as the psoas major and erector spinae (Supplemental Fig.
1, Supplemental Digital Content 1, http://links.lww.com/CCX/
A240; legend, Supplemental Digital Content 8, http://links.lww.
com/CCX/A247). All muscles were identied and quantied with
cuto values of –29 to 150 Hounseld units (HUs). e MMA was
automatically calculated by the soware. e L3SMI was further
normalized for the patients’ height as follows:
Skeletal muscle indexL3SMI (mm=�
skeletal�muscle�area
2
()
/cm)
��
at�third� LV mm
body�height (cm)
2
()
Statistical Analysis
Statistical analyses were performed as recently described (9)
(Supplemental Patients and Methods, Supplemental Digital
Content 2, http://links.lww.com/CCX/A241). A p value of less
than 0.05 was considered statistically signicant (*p < 0.05;
**p < 0.01; ***p < 0.001).
RESULTS
The Individual Body Composition Does Not Predict
Short-Term Mortality or ICU Survival in Critically Ill
Patients
A total of n = 155 patients who were admitted to the medical
ICU for treatment of critical illness were included in this study
(Table1). e median L3SMI that we used as a surrogate for sarco-
penia was 76.983 mm2/cm (range, 43.78–1,001.75 mm2/cm). e
median skeletal muscle attenuation (MMA) that we used a sur-
rogate parameter for muscular fat deposition was 30 HU (range,
–21 to 54 HU). We hypothesized that the patients’ individual body
composition might predict short-term mortality in ICU patients.
We therefore compared the L3SMI between patients who survived
the ICU stay and were admitted to standard care (ICU survivors)
and patients who deceased during the ICU stay (ICU nonsurvi-
vors). Interestingly, we did not see a signicant dierence between
these two groups (Supplemental Fig. 2, A and B, Supplemental
Digital Content 3, http://links.lww.com/CCX/A242; legend,
Supplemental Digital Content 8, http://links.lww.com/CCX/
A247). In a second step, we evaluated whether or not an increased
fat deposition in the skeletal muscle might be a negative predictor
for short-term outcomes. erefore, we determined the MMA of
the skeletal muscle area that was used to calculate the L3SMI as a
surrogate for fatty muscle depositions. However, when comparing
the MMA of patients who did or did not survive the ICU stay,
we did not observe a signicant dierence between ICU survivors
and nonsurvivors (Supplemental Fig. 2C, Supplemental Digital
Content 3, http://links.lww.com/CCX/A242; legend, Supplemental
Digital Content 8, http://links.lww.com/CCX/A247). In line, there
was no signicant association between the MMA and the 30-day
mortality (Supplemental Fig. 2D, Supplemental Digital Content
3, http://links.lww.com/CCX/A242; legend, Supplemental Digital
Content 8, http://links.lww.com/CCX/A247).
The Individual Body Composition Is Associated With
Long-Term Outcome of ICU Patients
Hypothesizing that an unfavorable body composition (low
amount of skeletal muscle, high amount of muscular fat deposi-
tion) might be associated with an impaired long-term survival
rather than predicting short-term mortality, we next compared
the L3SMI between patients who did or did not survive for 180
days and 1 year, respectively. Strikingly, we observed signicantly
higher L3SMI values in patients who survived for 180 days com-
pared to patients who deceased during this time (Fig. 1A). e
median L3SMI was 79.05 mm2/cm for the 180 days survivors and
72.65 mm2/cm for nonsurvivors. In line, the L3SMI was signi-
cantly associated with the 1-year mortality rate of ICU patients
as patients who survived for at least 1 year had signicantly
higher L3SMI values compared to patients who died within the
rst year aer ICU admission (79.81 vs 74.05 mm2/cm; Fig. 1B).
Subsequently, we compared the MMA between patients who
Original Clinical Report
Critical Care Explorations www.ccejournal.org 3
survived for 180 days or 1 year aer ICU admission and nonsurvi-
vors. Interestingly, we found signicantly lower MMA in patients
who died within the rst 180 days or rst year following ICU
admission (Fig.1, C and D) compared to patients who survived
this time period, suggesting that a higher amount of muscular
fat deposition (low MMA) also represents a negative predictor of
long-term ICU survival. In line, binary logistic-regression analysis
revealed both the L3SMI and the MMA as prognostic factors for
180-day mortality (odds ratio [ORL3SMI], 0.979; 0.961–0.997; p =
0.025 and ORHU, 0.964; 0.937–0.993; p = 0.014).
Finally, we compared the predictive power of the L3SMI and
the MMA regarding long-term survival with circulating predictive
markers that have previously been associated with ICU survival,
such as the C-reactive protein (CRP), procalcitonin, and creati-
nine using receiver operating characteristic curve analyses. Here,
we observed higher area under the curve (AUC) values for the
L3SMI and the MMA for both prediction of 180-day (AUCL3SMI:
0.654, AUCMMA: 0.657) and 1-year (AUCL3SMI: 0.654, AUCMMA:
0.657) mortality when compared to established predictive serum
parameters (Fig.1, E and F). Importantly, the predictive power for
long-term mortality was even higher when we combined param-
eters of body composition with established prognostic serum
markers such as the MMA and CRP levels, revealing a higher
AUC of 0.682 (Fig. 1G).
The L3SMI and Myosteatosis Are Predictors of Overall
Survival in ICU Patients
We subsequently evaluated if L3SMI and the MMA are also indic-
ative for the patients’ overall survival (OS). erefore, we subdi-
vided our cohort of patients into two groups with either high or
low L3SMI or MMA values (50th percentile cuto) and compared
the OS between groups using Kaplan-Meier curve analysis. is
analysis revealed a signicantly impaired OS for patients with a
L3SMI below 76.98 mm2/cm compared to patients with an L3SMI
value above this cuto (Fig. 2A). In line, patients showing a larger
amount of muscular fat deposition with an MMA below 31 HU
had a signicantly impaired OS compared to patients with an
MMA above 31 HU (Fig. 2B). When using ideal cuto values
(L3SMI: 74.95 mm2/cm and MMA: 29 HU), the L3SMI and the
MMA both showed a highly signicant discriminatory potential
for the identication of ICU patients with an unfavorable long-
term prognosis (Fig.2, C and D). e median OS both in the low
L3SMI and in low MMA group was only 8.4 weeks compared with
64.8 weeks in the L3SMI high and 62.1 weeks in the MMA high
group. In line, Cox regression analyses revealed a L3SMI (hazard
ratio [HR], 1.850; 1.259–2.719; p = 0.002) as well as an MMA (HR,
1.690; 1.167–2.446; p = 0.005) below these cuto values as negative
prognostic factors for OS. Importantly, multivariate Cox regres-
sion analysis showed that the prognostic potential of the L3SMI
was independent of various clinicopathological parameters such
as the patients’ age, markers of systemic inammation (leucocyte
count and CRP), and parameters of organ dysfunction (bilirubin,
hemoglobin) (Table 2). Importantly, the combinational use of
the L3SMI and the MMA (low L3SMI/low MMA vs high L3SMI/
high MMA) revealed an even better prognostic potential to pre-
dict OS in critically ill patients (Fig. 2E). Finally, the association
of the patients’ body composition with outcome is further sup-
ported by quartile analyses, showing, for example, that a gradually
decreasing skeletal muscle density leads to increasingly reduced
TABLE 1. Characteristics of Study Cohort
Parameter Study Cohort
ICU patients 155
Gender (%)
Male–female 60.6–39.4
Age, yr, median (range) 60 (21–88)
Body mass index, kg/m2, median (range) 25.21 (13.89–69.92)
Skeletal muscle index, mm2/cm,
median (range) 76.983 (43.78–1,001.75)
Hounsfield unit, median (range) 31 (–21 to 54)
ICU stay, d, median (range) 11 (1–147)
Hospital stay, d, median (range) 32 (3–199)
Mechanical ventilation, hr, median (range) 193.0 (0–3,499)
Sepsis (no–yes, %) 42.9–57.1
Cause of ICU admission (%)
Pulmonary sepsis 25.2
Abdominal sepsis 7.7
Urosepsis 5.2
Other sepsis 12.3
Liver cirrhosis 7.1
Acute liver failure 4.5
Liver transplantation 2.6
Gastrointestinal bleeding 7.7
Cardiopulmonary 13.5
Acute pancreatitis 5.2
Other 9.0
Medical conditions (no–yes, %)
Diabetes mellitus 75.7–24.3
Arterial hypertension 48.5–51.5
Chronic alcohol intake 71.4–28.6
Chronic obstructive pulmonary disease 83.6–16.4
Liver cirrhosis 76.1–23.9
Malignant disease 67.2–32.8
Survival rates (yes–no, %)
ICU survival 71.2–28.8
30-d survival 73.4–26.6
180-d survival 48.5–51.5
1-yr survival 39.6–60.4
Follow-up survival 25.2–74.8
Loosen et al
4 www.ccejournal.org 2020 • Volume 2 • e0171
OS (median OSMMA: 132.6 wk if MMA > 75th percentile [39 HU],
47.0 wk if MMA between 75th and 50th percentile [31–39 HU],
31.4 wk if MMA between 50th and 25th percentile [21–31 HU],
and 13.1 wk if MMA < 25th percentile [< 21 HU]).
Subsequently, we aimed at identifying potential underlying
factors which contributed to an impaired OS in patients with an
unfavorable body composition. As the time of mechanical ventila-
tion on the ICU is a well-established prognostic factor for OS (10),
we correlated the MMA with the duration of ventilation and
observed a signicant negative correlation between the HU and
the total hours of mechanical ventilation on the ICU (Fig. 3A).
us, a higher degree of myosteatosis (lower MMA) was associ-
ated with a longer necessity of mechanical ventilation, potentially
leading to an impaired OS. Besides the duration of mechani-
cal ventilation, we also found a signicant negative correlation
between the MMA and the duration of hospital stay, including the
stay on the ICU and standard care (Fig. 3B). Finally, we compared
the MMA in patients who did or did not fulll the criteria of sep-
sis, representing another well-established prognostic parameter in
ICU patients. Here, we observed signicantly lower MMA values
in septic patients compared with patients with nonseptic disease
(Fig. 3C). On the contrary, the L3SMI did not correlate with any
of these parameters (Fig.3 D–F).
Body Composition and Patient Characteristics
To evaluate a potential association between the patients’ body
composition and clinicopathological parameters, we compared
the L3SMI and the MMA between patients with dierent medi-
cal conditions (diabetes, arterial hypertension, chronic obstruc-
tive pulmonary disease [COPD], liver cirrhosis, chronic alcohol
intake, and malignant disease). While patients with diabetes mel-
litus showed signicantly higher L3SMI values than nondiabetic
patients, we did not observe a signicant dierence of the L3SMI
between patients with or without arterial hypertension, COPD,
liver cirrhosis, chronic alcohol intake, or malignant disease. e
L3SMI was also not signicantly altered when patients were strati-
ed with respect to their acute medical condition that led to ICU
admission (Supplemental Fig. 3 A–G, Supplemental Digital
Figure 1. The ICU patients’ body composition predicts long-term outcome. A, Patients with 180-d survival have significantly higher skeletal muscle index
(L3SMI) values compared to patients who deceased during this time. B, The L3SMI is significantly higher in patients who survive for at least one after ICU
admission. Patients who die within the first 180 d (C) or first year (D) following ICU admission have significantly lower mean skeletal muscle attenuation (MMA).
The area under the curve values for the L3SMI and Hounsfield unit (HU) for prediction of 180-d (E) and 1-yr (F) mortality are higher compared to established
predictive serum parameters. G, The prediction of 180-d mortality is highest when the MMA and C-reactive protein (CRP) levels are combined.
PCT = procalcitonin.
Original Clinical Report
Critical Care Explorations www.ccejournal.org 5
Content 4, http://links.lww.com/CCX/A243; legend, Supplemental
Digital Content 8, http://links.lww.com/CCX/A247). Similarly, we
observed signicantly lower MMA levels in patients with preexist-
ing diabetes mellitus. We also found higher grades of myosteatosis
(low MMA) in patients with arterial hypertension, while patients
with or without COPD, liver cirrhosis, chronic alcohol intake, or
malignant disease had similar MMA values. e MMA was also
not altered between the dierent disease etiologies that led to
ICU admission (Supplemental Fig. 4 A–G, Supplemental Digital
Content 5, http://links.lww.com/CCX/A244; legend, Supplemental
Digital Content 8, http://links.lww.com/CCX/A247).
Furthermore, we performed extensive correlation analysis
between the L3SMI as well as the MMA and various clinical labo-
ratory parameters of organ dysfunction (Supplemental Table 1,
Supplemental Digital Content 6, http://links.lww.com/CCX/A245).
Here, we found a signicant negative correlation between the MMA
and serum cholesterol levels (rS: –0.185; p = 0.040), arguing that
hypercholesterolemia is associated with a higher degree of fatty
muscle deposition and myosteatosis. Furthermore, we observed a
negative correlation between the MMA and the creatinine as well
as NT-pro brain natriuretic peptide (NT-proBNP). Circulating
interleukin (IL)–6 levels positively correlated with the L3SMI (rS: 0.182;
p = 0.047), which corroborates the nding that IL-6 is released by
skeletal muscle tissue during systemic stress response (11). In addi-
tion, the L3SMI negatively correlated with serum levels of alka-
line phosphatase (alkaline phosphatase [rS: –0.275; p = 0.001] and
NT-proBNP [rS: –0.242; p = 0.011]).
DISCUSSION
e body composition of an individual patient might be deter-
mined by very dierent techniques such as bioelectrical imped-
ance analysis or air displacement plethysmography, just to name a
few (12, 13). However, in case of critically ill patients oen receiv-
ing extensive hydration, the value of these methods to determine
skeletal muscle quantity and quality is hampered by consecu-
tive uid overload (13). Furthermore, in such patients, classical
functional tests for estimating body composition and physical
strength (body weight, waist circumference, body mass index,
ability to walk or physical activity, and handgrip) do not provide
reliable results (13). erefore, CT, available for almost all ICU
patients, represents the gold standard for assessing muscle mass
Figure 2. The body composition correlates with the patients’ overall survival (OS). A, ICU patients with a low skeletal muscle index (L3SMI) (below 50th
percentile) show a significantly reduced OS. B, A low mean skeletal muscle attenuation (MMA) (below 50th percentile) is associated with a significantly reduced
OS. When using the optimal prognostic cutoff values, both the L3SMI (C) and the mean skeletal Hounsfield unit (HU) (D) show a highly significant prediction of
OS. E, The combinational use of the L3SMI and the MMA (L3SMI low/MMA low vs L3SMI high/MMA high) reveals the highest prognostic potential.
Loosen et al
6 www.ccejournal.org 2020 • Volume 2 • e0171
and muscle quality in critically ill patients (13–15). So far, the role
of sarcopenia and myosteatosis as prognostic factors in critical
illness is unclear. Within this study, we quantied both the skel-
etal muscle mass and the amount of skeletal muscle fat deposi-
tion and evaluated the prognostic value of these parameters. At
the respective optimal cuto values that we established using a
recently described biometric soware (16), both a low L3SMI and
a low MMA turned out as powerful predictors of OS. Importantly,
when both markers were combined (e.g., L3SMI low/MMA low
vs L3SMI high/MMA high patients), the prognostic potential was
even further increased (Fig.2).
Sarcopenia is a common characteristic of elderly and mori-
bund patients. Muscle wasting can be triggered by manifold dis-
ease conditions, including disuse, denervation, fasting, cancer,
cardiac failure, and renal dysfunction (17). As all of these factors
are frequently found in critically ill patients and might themselves
limit patients’ prognosis, systematic research on the role of an
impaired muscle mass and strength bears a risk of important bias.
In our analysis, however, L3SMI/MMA values were comparable
within the dierent etiologies of critical illness (Supplemental Fig.
3, Supplemental Digital Content 4, http://links.lww.com/CCX/
A243; and Supplemental Fig. 4, Supplemental Digital Content 5,
http://links.lww.com/CCX/A244 [legend, Supplemental Digital
Content 8, http://links.lww.com/CCX/A247]). Furthermore, mul-
tivariate analysis including the patients’ age as well as parameters
of systemic inammation (leucocyte count and CRP) and organ
dysfunction (bilirubin and hemoglobin), identied the L3SMI as
an independent prognostic factor in critically ill patients (Table2).
Nevertheless, both L3SMI, as well as the MMA, were linked to
metabolic and/or cardiovascular diseases since diabetics dem-
onstrated higher L3SMI and lower MMA values and the L3SMI/
MMA were correlated with cholesterol serum concentrations and
NT-proBNP serum levels. While our data clearly argue for a role
of the L3SMI and the MMA as prognostic markers in patients
undergoing ICU treatment, it remains unclear if these factors
might also have a predictive value in ICU patients of dierent
disease etiologies (e.g., trauma patients) or ICU patients receiv-
ing dierent treatment modalities (e.g., extracorporeal membrane
oxygenation therapy). us, further studies are warranted to not
only conrm their prognostic role in the context of critical illness
but also to assess if the L3SMI and the MMA could support future
biomarker-driven clinical decision algorithms in the multimodal
treatment of critical illness. Importantly, our analysis only gave
information on the prognosis of these patients but had no pre-
dictive value, meaning that it is unclear if those patients with an
unfavorable prognosis in terms of their individual L3SMI/MMA
levels might have benetted to a greater extent from a specic
therapy or even represent candidates for active symptom control
who should not been admitted to the ICU. Furthermore, we only
evaluated L3SMI/HU values at ICU admission but longitudinal
measurements of L3SMI/MMA could also be important to answer
the question whether therapeutic interventions aecting body
composition might inuence patients’ outcome.
Our data are in line with previous studies demonstrating an asso-
ciation of sarcopenia and an impaired prognosis in patients with
nonalcoholic liver disease (18), liver cirrhosis (19), cardiovascular
(20), and lung diseases (21). Furthermore, it was demonstrated that
sarcopenia might predict outcome of patients undergoing therapeu-
tic interventions such as transcatheter arterial chemoembolization
(TACE) (22), liver transplantation (23), or colorectal cancer sur-
gery (24). Within our study, we not only analyzed the association
between sarcopenia and patient’s outcome but also of myosteatosis
and patients´ outcome. Low skeletal muscle radiodensity is related
to the accumulation of fat deposits within muscle (25, 26). It was
recently demonstrated that sarcopenia and myosteatosis are inde-
pendent abnormalities that represent two separate biological pro-
cesses and that at least in patients with pancreatic and periampullary
adenocarcinomas they frequently do not occur coincidentally (27).
In contrast, we observed that the L3SMI and the MMA signicantly
correlated with each other in patients with critical illness (r: 0.168;
TABLE 2. Univariate and Multivariate Cox Regression Analysis of the Skeletal Muscle Index
and Several Clinicopathological Parameters for the Prediction of Overall Survival
Parameter
Univariate Cox Regression Multivariate Cox Regression
pHazard Ratio (95% CI) pHazard Ratio (95% CI)
Skeletal muscle index (L3SMI) <
74.95 mm2/cm 0.002 1.850 (1.259–2.719) 0.008 1.762 (1.160–2.678)
Age < 0.001 1.035 (1.020–1.051) < 0.001 1.033 (1.016–1.049)
Sex 0.893 0.974 (0.660–1.437)
Body mass index 0.458 0.998 (0.992–1.004)
Leukocytes 0.014 1.031 (1.006–1.057) 0.081 1.023 (0.997–1.050)
Hemoglobin 0.121 0.994 (0.986–1.002) 0.477 0.997 (0.989–1.005)
Bilirubin 0.210 0.979 (0.946–1.012) 0.561 0.988 (0.951–1.028)
Procalcitonin 0.288 1.004 (0.996–1.012)
C-reactive protein 0.151 1.002 (0.999–1.004) 0.370 1.001 (0.999–1.004)
Creatinine 0.876 0.995 (0.929–1.065)
Original Clinical Report
Critical Care Explorations www.ccejournal.org 7
p = 0.044; Supplemental Fig. 5, Supplemental Digital Content 7,
http://links.lww.com/CCX/A246 [legend, Supplemental Digital
Content 8, http://links.lww.com/CCX/A247]). Furthermore, the
combinational use of the L3SMI and MMA was superior to either
marker alone (Fig.2), highlighting that sarcopenia and myosteato-
sis might not represent simple prognostic markers but might reect
specic aspects within dierent diseases.
e pathophysiological relation between sarcopenia/myosteatosis
and impaired patients’ prognosis is not fully clear at present. Stretch
et al (27) used microarray analysis to identify canonical pathways
deregulated in samples from patients with sarcopenia and myoste-
atosis. Interestingly, in patients with sarcopenia, most dierentially
regulated genes were part of the antigen presentation pathway, while
in patients with myosteatosis genes of the oxidative phosphoryla-
tion pathway were most regulated. Additionally, genes of the lipid
metabolism pathway, potentially contributing to lipid accumulation
(e.g., adiponectin receptor 2, apolipoprotein L1, apolipoprotein L2,
apolipoprotein O, and paraoxonase 3), were regulated in myosteato-
sis but not (or to a much lesser extent) in sarcopenia (27).
As the direct association between sarcopenia and outcome is gain-
ing increasing attention in dierent clinical conditions, a reliable,
cost-eective, and easily applicable method for the evaluation of sar-
copenia is essential. Although our method of sarcopenia assessment
is easily performed in patients undergoing CT scans for clinical indi-
cations, the mere determination of the patients’ skeletal muscle status
would not justify a CT scan due to radiation exposure and costs. In
this line of thinking, assessment of sar-
copenia using ultrasound might repre-
sent a noninvasive, radiation-free, and
inexpensive alternative. As such, ultra-
sound measurement of rectus femoris
muscle thickness has been suggested
as a quick screening test for sarcopenia
assessment (28). In addition, dier-
ent dynamic features of muscle qual-
ity such as microcirculation, perfusion
during rest and activity, as well as the
muscle elasticity could be assessed
using ultrasound techniques (29).
Our study was limited by some
points. First of all, the study was
conducted in an exploratory, single-
center design, including a total of
n = 155 patients treated a single ICU.
Although this design might reduce
the inter-hospital bias, our data
clearly need further validation from
larger multicenter studies. Second,
our study included more than 50%
of patients fullling the criteria for
sepsis, which might represent a con-
founder of results. In addition, the
heterogeneity regarding the underly-
ing disease etiology could also inu-
ence the results on OS in our study
cohort. Finally, our study cohort was
heterogeneous in terms of duration of ICU stay, duration of hos-
pital stay, and duration of mechanical ventilation. Although this
aspect argues for a rather general validity of results among dier-
ent ICU patients, larger control studies including further multi-
variate analyses on confounding parameters on OS are warranted
to fully elucidate the role of sarcopenia in better-dened cohorts
of ICU patients. Nevertheless, our study is the rst to demonstrate
a potential prognostic value of myosteatosis in the context of criti-
cally ill patients. Furthermore, our study for the rst time dem-
onstrates that simultaneous assessment and combined analysis of
dierent parameters of the patients’ individual body composition
in critically ill patients might be superior over a single marker for
estimating the long-term outcome of these ICU patients. Such
data might provide important information for early clinical deci-
sion-making on patients in emergency departments or ICUs.
Drs. Loosen, Schulze-Hagen, and Püngel shared first authorship. Drs. Tacke,
Luedde, Koch, and Roderburg shared last authorship.
Drs. Loosen, Tacke, Koch, and Roderburg designed the study. Dr. Schulze-
Hagen performed analysis of body composition. Drs. Loosen, Püngel,
Bündgens, and Wirtz performed data acquisition. Drs. Loosen and Püngel
performed statistical analysis. Dr. Loosen generated figures and tables. Drs.
Loosen and Roderburg drafted the article and performed its revision. Drs.
Wirtz, Vucur, Paffenholz, Bruners, Kuhl, Trautwein, Tacke, Luedde, and Koch
provided intellectual input. All authors approved the final version of the article.
Supplemental digital content is available for this article. Direct URL citations
appear in the HTML and PDF versions of this article on the journal’s website
(http://journals.lww.com/ccejournal).
Figure 3. Increased levels of myosteatosis are associated with sepsis and duration of mechanical ventilation.
A, There is a significant negative correlation between the mean skeletal muscle attenuation (MMA) and the
duration of mechanical ventilation. B, The duration of hospital stay negatively correlates with the degree of
myosteatosis (MMA). C, Patients who fulfill the criteria of sepsis have significantly lower MMA. D–F, The
skeletal muscle index (L3SMI) does not correlate with any of these parameters. HU = Hounsfield unit, n.s. = not
significant.
Loosen et al
8 www.ccejournal.org 2020 • Volume 2 • e0171
Dr. Luedde was funded from the European Research Council (ERC) under the
European Union’s Horizon 2020 research and innovation program through the
ERC Consolidator Grant Phase Control (Grant Agreement n° 771083). Dr.
Luedde was further supported by the German Cancer Aid (Deutsche Kreb-
shilfe 110043 and a Mildred-Scheel-Professorship), the German-Research-
Foundation (SFB-TRR57/P06 and LU 1360/3-1), the Ernst-Jung-Foundation
Hamburg, the Interdisciplinary Center of Clinical Research (IZKF) Aachen,
and a grant from the medical faculty of the Rheinisch-Westfälische Technische
Hochschule Aachen.
For information regarding this article, E-mail: christoph.roderburg@charite.de
REFERENCES
1. Jaitovich A, Khan MMHS, Itty R, et al: ICU Admission muscle and fat
mass, survival, and disability at discharge: A prospective cohort study.
Chest 2019; 155:322–330
2. Moisey LL, Mourtzakis M, Cotton BA, et al; Nutrition and Rehabilitation
Investigators Consortium (NUTRIC): Skeletal muscle predicts ventilator-
free days, ICU-free days, and mortality in elderly ICU patients. Crit Care
2013; 17:R206
3. Ji Y, Cheng B, Xu Z, et al: Impact of sarcopenic obesity on 30-day mortal-
ity in critically ill patients with intra-abdominal sepsis. J Crit Care 2018;
46:50–54
4. Fielding RA, Vellas B, Evans WJ, et al: Sarcopenia: An undiagnosed con-
dition in older adults. Current consensus denition: Prevalence, etiology,
and consequences. International working group on sarcopenia. J Am Med
Dir Assoc 2011; 12:249–256
5. van der Werf A, Langius JAE, de van der Schueren MAE, et al: Percentiles
for skeletal muscle index, area and radiation attenuation based on com-
puted tomography imaging in a healthy Caucasian population. Eur J Clin
Nutr 2018; 72:288–296
6. Faron A, Pieper CC, Schmeel FC, et al: Fat-free muscle area measured by
magnetic resonance imaging predicts overall survival of patients under-
going radioembolization of colorectal cancer liver metastases. Eur Radiol
2019; 29:4709–4717
7. Faron A, Luetkens JA, Schmeel FC, et al: Quantication of fat and skeletal
muscle tissue at abdominal computed tomography: Associations between
single-slice measurements and total compartment volumes. Abdom
Radiol (NY) 2019; 44:1907–1916
8. Fedorov A, Beichel R, Kalpathy-Cramer J, et al: 3D Slicer as an image
computing platform for the Quantitative Imaging Network. Magn Reson
Imaging 2012; 30:1323–1341
9. Loosen SH, Roderburg C, Kauertz KL, et al: Elevated levels of circulating
osteopontin are associated with a poor survival aer resection of cholan-
giocarcinoma. J Hepatol 2017; 67:749–757
10. Herer B: Outcomes of prolonged mechanical ventilation before and aer
implementation of a respiratory ICU. Respir Care 2020; 65:1011–1018
11. Welc SS, Clanton TL: e regulation of interleukin-6 implicates skeletal
muscle as an integrative stress sensor and endocrine organ. Exp Physiol
2013; 98:359–371
12. Eslamparast T, Montano-Loza AJ, Raman M, et al: Sarcopenic obe-
sity in cirrhosis-the conuence of 2 prognostic titans. Liver Int 2018;
38:1706–1717
13. Montano-Loza AJ, Meza-Junco J, Baracos VE, et al: Severe muscle deple-
tion predicts postoperative length of stay but is not associated with sur-
vival aer liver transplantation. Liver Transpl 2014; 20:640–648
14. Montano-Loza AJ, Angulo P, Meza-Junco J, et al: Sarcopenic obesity and
myosteatosis are associated with higher mortality in patients with cirrho-
sis. J Cachexia Sarcopenia Muscle 2016; 7:126
15. Merli M, Berzigotti A, Zelber-Sagi S, et al: EASL Clinical Practice
Guidelines on nutrition in chronic liver disease. J Hepatol 2019;
70:172–193
16. Budczies J, Klauschen F, Sinn BV, et al: Cuto Finder: A comprehensive
and straightforward Web application enabling rapid biomarker cuto
optimization. PLoS One 2012; 7:e51862
17. Schefold JC, Bierbrauer J, Weber-Carstens S: Intensive care unit-acquired
weakness (ICUAW) and muscle wasting in critically ill patients with
severe sepsis and septic shock. J Cachexia Sarcopenia Muscle 2010;
1:147–157
18. Koo BK, Kim D, Joo SK, et al: Sarcopenia is an independent risk factor
for non-alcoholic steatohepatitis and signicant brosis. J Hepatol 2017;
66:123–131
19. Buchard B, Boirie Y, Cassagnes L, et al: Assessment of malnutrition, sar-
copenia and frailty in patients with cirrhosis: Which tools should we use
in clinical practice? Nutrients 2020; 12:186
20. Santana NM, Mendes RML, Silva NFD, et al: Sarcopenia and sarcope-
nic obesity as prognostic predictors in hospitalized elderly patients with
acute myocardial infarction. Einstein (Sao Paulo) 2019; 17:eAO4632
21. Hirai K, Tanaka A, Homma T, et al: Comparison of three frailty models
and a sarcopenia model in elderly patients with chronic obstructive pul-
monary disease. Geriatr Gerontol Int 2019; 19:896–901
22. Loosen SH, Schulze-Hagen M, Bruners P, et al: Sarcopenia is a negative
prognostic factor in patients undergoing transarterial chemoemboliza-
tion (TACE) for hepatic malignancies. Cancers (Basel) 2019; 11:1503
23. Czigany Z, Kramp W, Bednarsch J, et al: Myosteatosis to predict inferior
perioperative outcome in patients undergoing orthotopic liver transplan-
tation. Am J Transplant 2020; 20:493–503
24. Nakanishi R, Oki E, Sasaki S, et al: Sarcopenia is an independent pre-
dictor of complications aer colorectal cancer surgery. Surg Today 2018;
48:151–157
25. Stephens NA, Skipworth RJ, Macdonald AJ, et al: Intramyocellular lipid
droplets increase with progression of cachexia in cancer patients. J
Cachexia Sarcopenia Muscle 2011; 2:111–117
26. Aubrey J, Esfandiari N, Baracos VE, et al: Measurement of skeletal muscle
radiation attenuation and basis of its biological variation. Acta Physiol
(Oxf) 2014; 210:489–497
27. Stretch C, Aubin JM, Mickiewicz B, et al: Sarcopenia and myosteatosis
are accompanied by distinct biological proles in patients with pancreatic
and periampullary adenocarcinomas. PLoS One 2018; 13:e0196235
28. Rustani K, Kundisova L, Capecchi PL, et al: Ultrasound measurement of
rectus femoris muscle thickness as a quick screening test for sarcopenia
assessment. Arch Gerontol Geriatr 2019; 83:151–154
29. Ivanoski S, Vasilevska Nikodinovska V: Future ultrasound biomarkers for
sarcopenia: Elastography, contrast-enhanced ultrasound, and speed of
sound ultrasound imaging. Semin Musculoskelet Radiol 2020; 24:194–200
