J Appl Physiol 114: 559–565, 2013.
First published January 3, 2013; doi:10.1152/japplphysiol.01042.2012.
Sarcopenia and a physiologically low respiratory quotient in patients with
cirrhosis: a prospective controlled study
Cathy Glass,1 Peggy Hipskind,1 Cynthia Tsien,2 Steven K. Malin,3 Takhar Kasumov,2 Shetal N. Shah,4
John P. Kirwan,2,3,5 and Srinivasan Dasarathy2,3
1
Department of Nutrition Therapy, Digestive Disease Institute, Cleveland Clinic, Cleveland, Ohio; 2Department of
Gastroenterology and Hepatology, Digestive Disease Institute, Cleveland Clinic, Cleveland, Ohio; 3Department of
Pathobiology, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio; 4Department of Diagnostic Radiology, Imaging
Institute, Cleveland Clinic, Cleveland, Ohio; and 5Metabolic Translational Research Center, Endocrinology and Metabolism
Institute, Cleveland Clinic, Cleveland, Ohio
Submitted 23 August 2012; accepted in final form 27 December 2012
hospitalized cirrhosis patients; respiratory quotient; outcome; metabolism
RESPIRATORY QUOTIENT CALCULATED from gas exchange on indirect calorimetry is used to study the relative contribution of
different substrates to energy metabolism (21, 43). RQ is
defined as V̇CO2 divided by V̇O2 (9). Low respiratory quotient
(RQ) has been consistently reported in patients with cirrhosis
but its clinical or physiological relevance has not been evaluated in the context of skeletal muscle loss or sarcopenia (21,
Address for reprint requests and other correspondence: S. Dasarathy, NE4
208, Lerner Research Institute, Cleveland Clinic Foundation, Cleveland, OH
44195 (e-mail: [email protected]).
http://www.jappl.org
34, 40, 41). Low RQ in patients with cirrhosis has been
interpreted as evidence of accelerated starvation (21, 46, 51).
Because starvation is accompanied by impaired skeletal muscle
protein synthesis and increased proteolysis, these could potentially contribute to the reduction in skeletal muscle mass or
sarcopenia in these patients (8, 10, 11). During states of
starvation, there is a shift in contribution of different substrates
to energy generation (3, 32). In cirrhosis, reduction in RQ
occurs early and is more severe compared with healthy subjects
(33, 51). Low RQ in patients with cirrhosis is believed to be a
consequence of altered energy metabolism with a rapid switch
in substrates from carbohydrate to fatty acid oxidation and
gluconeogenesis from amino acids (2, 4, 19). Theoretical and
experimental evidence using calorimetry has shown that the
RQ from the three major substrates for energy are different (9).
The lowest physiological RQ is considered to be 0.69 based on
the assumption that the entire energy cost is derived from the
oxidation of fatty acids. It has also been suggested that measured RQs less than 0.65 are due to methodological errors (18).
There are few reports on measured RQs less than 0.69 (6). In
the liver, ketogenesis, which is an incomplete oxidation of fatty
acids, will decrease the RQ and gluconeogenesis, which incorporates carbon that can potentially lower the RQ to values
much lower than 0.69 (45). Reduced hepatic RQ due to
accelerated ketogenesis and gluconeogenesis is counterbalanced by the complete oxidation of these metabolites in nonhepatic tissue, resulting in whole body RQ that may be lower
than the predicted RQ based on substrate oxidation.
Cirrhosis is characterized by reduced hepatic glycogen content (2, 17, 36) and insulin resistance (2, 20, 30, 42). Both of
these metabolic responses exaggerate ketogenesis and gluconeogenesis that can lower the hepatic RQ to values below 0.69
(2, 4, 36). Whole body RQ, however, is the net of hepatic and
nonhepatic RQ. When ketones and glucose are completely
oxidized in the periphery, primarily in skeletal muscle, this RQ
is 1.0, thereby increasing the whole body RQ (15, 32). However, when the skeletal muscle mass is reduced as occurs in
cirrhosis (35), it is likely to result in incomplete oxidation of
ketones. This will increase the relative contribution of the very
low hepatic RQ to net whole body RQ. This can potentially
explain abnormally low whole body RQs that have previously
been considered to be due to methodological errors (18).
Previous reports on patients with cirrhosis to study energy
metabolism have been done only in stable, outpatient populations (21, 34, 40, 43, 48). Hospitalized patients with cirrhosis
have a greater severity of illness and are therefore likely to be
8750-7587/13 Copyright © 2013 the American Physiological Society
559
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
Glass C, Hipskind P, Tsien C, Malin SK, Kasumov T, Shah SN,
Kirwan JP, Dasarathy S. Sarcopenia and a physiologically low
respiratory quotient in patients with cirrhosis: a prospective controlled
study. J Appl Physiol 114: 559 –565, 2013. First published January 3,
2013; doi:10.1152/japplphysiol.01042.2012.—Patients with cirrhosis
have increased gluconeogenesis and fatty acid oxidation that may
contribute to a low respiratory quotient (RQ), and this may be linked
to sarcopenia and metabolic decompensation when these patients are
hospitalized. Therefore, we conducted a prospective study to measure
RQ and its impact on skeletal muscle mass, survival, and related
complications in hospitalized cirrhotic patients. Fasting RQ and resting energy expenditure (REE) were determined by indirect calorimetry in cirrhotic patients (n ⫽ 25), and age, sex, and weight-matched
healthy controls (n ⫽ 25). Abdominal muscle area was quantified by
computed tomography scanning. In cirrhotic patients we also examined the impact of RQ on mortality, repeat hospitalizations, and liver
transplantation. Mean RQ in patients with cirrhosis (0.63 ⫾ 0.05) was
significantly lower (P ⬍ 0.0001) than healthy matched controls (0.84 ⫾
0.06). Psoas muscle area in cirrhosis (24.0 ⫾ 6.6 cm2) was significantly (P ⬍ 0.001) lower than in controls (35.9 ⫾ 9.5 cm2). RQ
correlated with the reduction in psoas muscle area (r2 ⫽ 0.41; P ⫽
0.01). However, in patients with cirrhosis a reduced RQ did not
predict short-term survival or risk of developing complications. When
REE was normalized to psoas area, energy expenditure was significantly higher (P ⬍ 0.001) in patients with cirrhosis (66.7 ⫾ 17.8
kcal/cm2) compared with controls (47.7 ⫾ 7.9 kcal/cm2). We conclude that hospitalized patients with cirrhosis have RQs well below
the traditional lowest physiological value of 0.69, and this metabolic
state is accompanied by reduced skeletal muscle area. Although low
RQ does not predict short-term mortality in these patients, it may
reflect a decompensated metabolic state that requires careful nutritional management with appropriate consideration for preservation of
skeletal muscle mass.
560
Abnormally Low Respiratory Quotient in Cirrhosis
metabolically decompensated in terms of substrate contribution
to energy production.
We hypothesized that hospitalized patients with cirrhosis are
more likely to have significantly lower RQs than have been
previously reported. We also hypothesized that the reduction in
RQ would be accompanied by loss of skeletal muscle. A
prospective study was performed to quantify RQ in these
patients and relate the outcome to abdominal muscle mass
measured by CT scanning using our previously established
methodology (52).
METHODS
Glass C et al.
sured REE exceeded the predicted values by over 20% of that
predicted by the Harris Benedict equation and hypometabolic when
the measured REE was reduced by 20% of the predicted value (27).
The predictive values were generated using both actual measured
body weight as well as the estimated ideal body weight using the
Hamwi formula (13).
Body weight. Anthropometric measurements (whole body weight,
height) were obtained in a standard hospital gown on a calibrated
scale and a wall-mounted stadiometer. Body mass index (BMI) was
calculated as body mass (kg) divided by height (m2).
Measurement of muscle area. Appendicular muscle areas were
measured on a Leonardo Workstation using Oncocare (Seimens
Healthcare, Erlangen, Germany) as previously described (52). In
brief, unenhanced axial CT scans at the mid fourth lumbar vertebral
level were used to determine the cross-sectional area and mean
attenuation (in Hounsfield units) of the major and minor psoas and
paraspinal muscles on both sides as well as the total abdominal wall
muscles. Subcutaneous and visceral adipose tissue mass were
quantified. All CT scans had been performed as part of clinical care
of these patients. Use of single slice CT images to quantify muscle
and fat area as a reflection of whole body measure of skeletal
muscle and adipose tissue has been previously reported by others
and is considered a standard method to quantify skeletal muscle
and fat mass (23, 47).
All studies were approved by the Institutional Review Board at the
Cleveland Clinic, and written informed consent was obtained from all
subjects.
Statistical analyses. Qualitative variables were compared using the
Chi square test and quantitative variables compared using the Student’s t-test. Correlations were determined using the Pearson’s correlation coefficient. Patients were followed up after the study till they
died or the date of last follow-up. Multivariate analysis was performed
to examine the relation between RQ and psoas area and visceral fat
mass. All values were expressed as mean ⫾ standard deviation unless
otherwise specified. A P value ⬍0.05 was considered statistically
significant. On the basis of previous data of an anticipated difference
of at least 35% in measured RQ between patients with cirrhosis and
controls, an ␣ of 5% and a power of 80%, we required at least 11
subjects in each study group. All analyses were performed using SPSS
20.0 (IBM, Armonk, NY).
RESULTS
Clinical characteristics. Clinical and demographic features
of cirrhotic patients and controls are shown in Table 1. The
reasons for hospitalization were infection (n ⫽ 6), encephalopathy (n ⫽ 8), renal failure (n ⫽ 3), ascites (n ⫽ 3), and other
causes (n ⫽ 5). Ten of these patients had well-controlled
diabetes mellitus (HbA1C 6.4 ⫾ 1.4 g/dl). As expected, the
biochemical parameters (serum bilirubin, albumin, and coagu-
Table 1. Clinical and demographic features
Number
Age, yr
Height, in.
Weight, kg
BMI, kg/m2
MELD score
Child-Pugh score
REE, kcal/day
VO2, ml/min
Predicted REE (Harris Benedict), kcal/day
Cirrhosis Patients
Controls
25
56.6 ⫾ 8.0 (43–79)
68.5 ⫾ 4.2 (60–76)
89.0 ⫾ 20.9 (58–139)
29.6 ⫾ 7.7 (18–56)
17.4 ⫾ 4.4 (11–28)
9.0 ⫾ 1.5 (6–12)
1,525.6 ⫾ 305.3 (930–2,190)
233.1 ⫾ 45.5 (159–319)
1,711.6 ⫾ 293.9 (1,262–2,394)
25
54.3 ⫾ 12.9 (32–75)
67.2 ⫾ 3.4 (59.1–74)
92.4 ⫾ 19.6 (46.7–131.6)
31.4 ⫾ 4.8 (18.2–38.3)
1,571.2 ⫾ 278.3 (1,003–2,386)
224.2 ⫾ 41 (146.2–347)
1,751 ⫾ 289.2 (1,153–2,381)
Values are means ⫾ SD. Figures in parentheses are the range of values. BMI, body mass index; MELD, model for end stage liver disease; REE, resting energy
expenditure; RQ, respiratory quotient.
J Appl Physiol • doi:10.1152/japplphysiol.01042.2012 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
Subjects. Consecutive adult hospitalized cirrhotic patients (n ⫽ 25)
were included in this prospective study. Some of the clinical and
demographic details of the patients with cirrhosis have been published
elsewhere because they were part of an ongoing prospective study on
energy metabolism in cirrhosis (12). All patients were on a standard
hospital diet with a 2 g sodium but no protein restriction. No other
dietary alterations were made to the patients’ meals. No patient
received enteral or parenteral nutrition. A simultaneous cohort of
healthy subjects who were age, sex, and body mass index matched (n ⫽ 25)
formed the control population. All free living controls were allowed to
eat their usual diet. Because it was difficult to match the dietary intake
of hospitalized patients with cirrhosis with that of outpatient, healthy
controls, we ensured that the duration of fasting before quantifying
RQ and REE was similar. Patients with uncontrolled diabetes mellitus, renal failure on dialysis, medications or supplements that affected
skeletal muscle mass, advanced malignancies, end stage organ failure
other than cirrhosis were excluded. Cirrhotic patients were followed
up after discharge to determine their outcomes. After an overnight
fast, blood glucose, serum amino transferases, blood urea nitrogen,
serum creatinine, albumin, and bilirubin were obtained on the day of
the study in the clinical core laboratory. Child-Pugh and the model for
end stage liver disease (MELD scores) were calculated for the cirrhotic subjects (16). Urinary ketones were measured in the first
morning sample in all subjects.
Indirect calorimetry. The V̇O2 and V̇CO2 were quantified using an
open canopy indirect calorimeter (Vmax Encore, Viasys, Yorba Linda,
CA) in the clinical research unit as previously described (12). In brief,
patients were allowed to rest for 10 min prior to the determination of
substrate oxidation and energy expenditure. Breath samples were
collected for a total of 15 min, and only the last 5 min were used for
analysis. The calorimeter was calibrated prior to each study and was
carefully checked for any leaks. All cirrhotic patients also had breath
samples analyzed by a hand held respiratory calorimeter (MedGem,
Microlife, Golden, CO) either preceding or following the indirect
calorimetry. Resting energy expenditure (REE) was also calculated
for all study participants using the Harris Benedict equations (14).
Cirrhotic patients were classified as hypermetabolic when the mea-
•
Abnormally Low Respiratory Quotient in Cirrhosis
Table 2. Biochemical characteristics of hospitalized patients
with cirrhosis
Cirrhosis Patients
•
Table 3. Characteristics of patients with cirrhosis with very
low respiratory quotient
Controls
Serum bilirubin, g/dl
6.1 ⫾ 8.0 (0.2–36.7)
0.46 ⫾ 0.12 (0.2–0.7)*
Serum albumin, g/dl
2.8 ⫾ 0.8 (1.4–4.2)
4.6 ⫾ 1.2 (3.8–10.1)*
INR
1.5 ⫾ 0.7 (1–3.8)
Serum creatinine, mg/dl 1.21 ⫾ 0.72 (0.46–3.42) 0.9 ⫾ 0.2 (0.63–1.6)*
Blood urea nitrogen,
23.68 ⫾ 14.3 (8–65)
14.7 ⫾ 3.1 (9–23)*
mg/dl
Serum chloride, mg/dl
103.1 ⫾ 6.0 (93–113)
Arterial bicarbonate,
20.3 ⫾ 4.9 (11–27)
mg/dl
Hemoglobin A1c, g/dl
5.3 ⫾ 2.0 (0–9.8)
Values are means ⫾ SD. Figures in parentheses are the range of values.
Blans indicated values not obtained for the healthy population. INR, International normalized ratio. *P ⬍ 0.01.
Fig. 1. Respiratory quotient in hospitalized patients with cirrhosis (n ⫽ 25) was
significantly lower than that in matched healthy control subjects (n ⫽ 25).
***P ⬍ 0.01.
Number of patients
Age, years
Height, in.
Weight, kg
BMI, kg/m2
MELD score
Child- Pugh score
RQ
REE, kcal/day
VO2, ml/min
Predicted REE (Harris Benedict),
kcal/day
Duration of follow up, mo
Serum bilirubin, mg/dl
Serum albumin, g/dl
INR
Serum creatinine, mg/dl
Blood urea nitrogen, mg/dl
Low RQ (⬍0.68)
High RQ (ⱖ0.68)
20
56.2 ⫾ 8.2
68.8 ⫾ 4.3
92.2 ⫾ 21.8
30.5 ⫾ 8.1
18.2 ⫾ 4.5
9.0 ⫾ 1.6
0.61 ⫾ 0.04
1536.5 ⫾ 311
239.5 ⫾ 48.4
1,715.9 ⫾ 276.6
5
58.2 ⫾ 7.6
67.4 ⫾ 4.0
76.4 ⫾ 10.4
26.2 ⫾ 5.3
17.2 ⫾ 4.5
9.2 ⫾ 1.1
0.71 ⫾ 0.02
1482 ⫾ 311
207.6 ⫾ 17
1,694.6 ⫾ 392.9
6.1 ⫾ 3.3
5.8 ⫾ 8.3
2.7 ⫾ 0.7
1.6 ⫾ 0.8
1.3 ⫾ 0.8
25.1 ⫾ 15.5
7.8 ⫾ 2.1
7.6 ⫾ 7.4
2.2 ⫾ 0.8
1.2 ⫾ 0.5
1.0 ⫾ 0.0
18.0 ⫾ 6.0
Values are means ⫾ SD.
(RQ ⫽ 0.58 ⫾ 0.03) compared with the remaining 20 in whom
urinary ketones were absent (RQ ⫽ 0.64 ⫾ 0.05).
The REE in patients with cirrhosis (1,525.6 ⫾ 305.3 kcal/
day) was similar to that in controls (1,571.2 ⫾ 278.3 kcal/day).
REE estimated by the Harris Benedict equation in patients with
cirrhosis using either the actual measured weight (1,711.6 ⫾
293.9) or the predicted ideal body weight (1,510.2 ⫾ 223.4)
showed no significant difference (P ⬎ 0.1). By using the
conventional methods to classify patients with cirrhosis based
on REE, there were 3 (12%) hypermetabolic, 7 (28%) hypometabolic, and 15 (60%) eumetabolic patients with cirrhosis.
Skeletal muscle in cirrhosis. Abdominal muscle area was
significantly lower in patients with cirrhosis compared with
controls (Table 4). Skeletal muscle density was also significantly lower in patients with cirrhosis than controls but did
not correlate with RQ. BMI did not correlate with RQ either
in controls or in patients with cirrhosis. However, psoas muscle
area correlated (r2 ⫽ 0.69; P ⬍ 0.001) highly with the RQ in
the study subjects (Fig. 2). Patients with cirrhosis had lower
muscle area and lower RQ compared with controls. However,
in the cirrhotic patients alone, RQ did not correlate with muscle
area.
REE showed a positive correlation with psoas muscle area in
both groups (Fig. 3). Despite no difference in mean REE in the
two groups, mean REE normalized to psoas area was significantly higher (P ⫽ 0.002) in patients with cirrhosis compared
with controls (Fig. 4). These relations were maintained when
the other groups of muscle were used (data not shown).
Fat mass in cirrhosis. Both visceral and subcutaneous fat
mass were significantly lower in patients with cirrhosis compared with controls (Table 4; P ⬍ 0.05). Both visceral fat mass
(r2 ⫽ 0.39; P ⫽ 0.01) and subcutaneous fat mass (r2 ⫽ 0.46;
P ⫽ 0.01) correlated significantly with the RQ (Figs. 5 and 6).
There was no significant correlation between fat mass (visceral
or subcutaneous) and the measured or predicted REE. Multivariate analysis with RQ as the dependent variable showed a
significant relation with psoas muscle area (P ⫽ 0.038) and a
trend effect for visceral fat mass (P ⫽ 0.058).
J Appl Physiol • doi:10.1152/japplphysiol.01042.2012 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
lation profile) in patients with cirrhosis were abnormal because
all of them were hospitalized and decompensated (12). Trace
urinary ketones were present in 5 (20%) patients with cirrhosis.
The laboratory tests in the healthy controls were within normal
limits (Table 2).
Energy metabolism. Cirrhotic subjects had significantly
lower (P ⬍ 0.01) RQ compared with healthy age, BMI and sex
matched healthy controls (Fig. 1). Of the 25 patients with
cirrhosis, 20 patients had an RQ less than 0.68. Although this
group was considered to be metabolically decompensated, their
Child-Pugh and MELD scores were not significantly different
from those with RQ values greater than 0.68 (Table 3). Similarly their renal functions were also not different. Methodological errors contributing to nonphysiological RQs in patients
with cirrhosis were excluded by comparing the V̇O2 measured
by the metabolic cart with that obtained from the HHRC.
Because V̇O2 measured using both the metabolic cart and the
HHRC were nearly identical (233.1 ⫾ 45.5 vs. 220.2 ⫾ 44.0
ml/min)(12), it suggests that our RQ measurements were accurate and reliable. Furthermore, in a contemporaneous cohort
of healthy subjects using the same metabolic cart, the RQs
obtained were within the normal expected range. Laboratory
values of severity of liver disease, including serum bilirubin,
albumin, prothrombin time, creatinine, and blood urea nitrogen, did not correlate with measured RQ or reduction in RQ in
patients with cirrhosis. The mean RQ was significantly lower
(P ⫽ 0.02) in the five patients with trace urinary ketones
561
Glass C et al.
562
Abnormally Low Respiratory Quotient in Cirrhosis
•
Glass C et al.
Table 4. Muscle and adipose tissue characteristics in cirrhotic and control subjects
Cirrhosis Patients
2
Psoas muscle area, cm
Paraspinal muscle area, cm2
Abdominal wall muscle area, cm2
Psoas muscle density in HU
Paraspinal density in HU
Abdominal wall density in HU
Visceral fat mass, mm3
Subcutaneous fat mass, mm3
24.0 ⫾ 6.6 (12.39–38.61)
55.5 ⫾ 9.7 (33.1–72.7)
74.5 ⫾ 20.7 (36–129.5)
27.8 ⫾ 13.5 (⫺8.2–46)
7.1 ⫾ 18.7 (⫺23.2–38.4)
14.9 ⫾ 12.6 (⫺15–34.7)
36.4 ⫾ 20.8
57.7 ⫾ 28.9
Controls
35.9 ⫾ 9.5 (17.3–54.4)†
59.7 ⫾ 9.6 (45.1–79.0)†
91.2 ⫾ 20.4 (53.7–108.5)†
39.0 ⫾ 6.1 (32.6–47.7†
5.4 ⫾ 11.7 (⫺1.1–40.6)†
7.6 ⫾ 11.0 (⫺5.5–29.2)†
56.6 ⫾ 20.5†
79.7 ⫾ 24.2*
Values are means ⫾ SD. Figures in parentheses are the range of values. HU, Hounsfield units. *P ⬍ 0.05; †P ⬍ 0.01.
DISCUSSION
The novel finding in this study is that very low whole body
RQs occur in hospitalized, decompensated, cirrhotic patients.
Skeletal muscle mass, measured as psoas and other abdominal
muscle areas, was significantly lower in patients with cirrhosis
than controls. RQ correlated highly and directly with the
skeletal muscle area. Although resting energy expenditure in
patients with cirrhosis was similar to that in controls, when
normalized for muscle area, patients with cirrhosis had much
higher REE than controls. An additional observation in this
study was that the reduction in RQ was also accompanied by
lower visceral and subcutaneous fat mass compared with controls. The degree of reduction in RQ did not predict either
in-hospital or short-term adverse outcomes. Finally, worsening
Fig. 2. Scatterplot showing that respiratory quotient is highly directly correlated with psoas muscle area (used as a measure of skeletal muscle mass).
r2 ⫽ 0.71; P ⬍ 0.001.
severity of liver disease measured by Child’s score or MELD
score was not associated with greater decline in RQ.
Low RQ in cirrhosis has been reported by a number of
authors in the past and is believed to be due to an altered
energy metabolism characterized by a switch to fatty acids and
gluconeogenic substrates (31, 33, 39, 46). These metabolic
alterations are considered to be the consequence of both hepatocellular dysfunction and portacaval shunting (31). Because
hepatic glycogen content is decreased in cirrhosis (2, 17, 33),
there is a greater need for increased noncarbohydrate-derived
substrate oxidation to meet the bioenergetic demands of these
patients. To accommodate the increased energy needs, fatty
acid oxidation and gluconeogenesis (requiring the carbon skeleton from muscle protein derived amino acids) are increased in
these patients (21, 26, 28, 29, 36). These responses are accompanied by a reduction in measured RQ (21). Consistent with
these anticipated metabolic responses, we observed much
lower RQ than has been predicted (RQ values of 0.69 or
higher) based on estimated contributions of different substrates
to the oxidative metabolism (9, 18). Previous studies have
reported increased gluconeogenesis and ketogenesis in patients
with cirrhosis and this may be the reason for these unexpected
values in this study (2, 31, 33, 36, 39, 50). The trace urinary
ketones in patients with very low RQ support our interpretation
that the very low RQ is partially accounted for by increased
ketogenesis in these patients. Although we cannot address the
exact mechanisms responsible for these very low RQ values in
the present study, patients with the trace ketones were not
diabetic and all our patients with diabetes had good glycemic
control. Because gluconeogenesis was not quantified, its contribution to the reduction in RQ cannot be determined in these
Fig. 3. Scatterplot showing that resting energy expenditure was significantly
correlated with psoas muscle area in cirrhotic and control subjects. r2 ⫽ 0.19;
P ⬍ 0.05.
J Appl Physiol • doi:10.1152/japplphysiol.01042.2012 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
Outcomes. Of the 25 patients with cirrhosis included in the
study, 11 (44%) died and 7 (28%) underwent liver transplantation over a period of 10.9 ⫾ 5.2 mo of follow up. Twentythree patients were readmitted (3.7 ⫾ 2.7 hospitalizations)
during follow up. RQ did not correlate with short-term outpatient survival in these patients with cirrhosis (Table 5). Readmission rates and postdischarge development of other complications, including renal failure, GI bleed, and encephalopathy,
also did not correlate with the degree of reduction in RQ.
Metabolic status based on REE (hyper or hypometabolic) and
REE normalized to psoas area were also similar in patients
with cirrhosis who survived and those who died.
Abnormally Low Respiratory Quotient in Cirrhosis
•
563
Glass C et al.
Fig. 4. Resting energy expenditure normalized to muscle area was significantly
higher in patients with cirrhosis than controls. ***P ⬍ 0.01.
Fig. 5. Scatterplot showing that respiratory quotient is directly correlated with
visceral fat mass (on CT). r2 ⫽ 0.39; P ⫽ 0.01.
Fig. 6. Scatterplot showing that respiratory quotient is directly correlated with
subcutaneous fat mass (on CT). r2 ⫽ 0.46; P ⫽ 0.01.
with cirrhosis may therefore be a response to the skeletal
muscle depletion along with increased hepatic gluconeogenesis
and ketogenesis. This is similar to the altered skeletal muscle
ketone metabolism described in cardiac failure (15), another
common condition characterized by sarcopenia. Skeletal muscle attenuation was lower in patients with cirrhosis compared
with controls that was similar to our previous report (52).
However, skeletal muscle density measured by CT was not
predictive of outcome and its physiological significance in
cirrhosis needs to be evaluated.
Quantification of urinary nitrogen helps in determining the
contribution of different substrates to whole body metabolism
(21). In the present study, urinary nitrogen was not quantified,
but our direct measurements of muscle area show that the
lower RQ is accompanied by lower muscle area. Another
interpretation of our observations is that skeletal muscle deTable 5. Predictors of mortality
Number
Age, yr
Height, in.
Weight, kg
BMI, kg/m2
MELD score
Child-Pugh score
RQ
REE, kcal/day
VO2, ml/min
Predicted REE (Harris Benedict),
kcal/day
Duration of follow up, mo
Bilirubin, mg/dl
Albumin, g/dl
INR
Serum creatinine, mg/dl
BUN, mg/dl
Psoas muscle area, cm2
Paraspinal muscle area, cm2
Abdominal wall muscle area, cm2
Visceral fat mass, mm3
Subcutaneous fat mass, mm3
Alive
Dead
17
57.2 ⫾ 8
68.5 ⫾ 4.6
91.4 ⫾ 24.6
30.5 ⫾ 9.1
16.4 ⫾ 3.9
8.6 ⫾ 1.5
0.63 ⫾ 0.05
1,546.5 ⫾ 304.3
231.8 ⫾ 48.5
1,712.2 ⫾ 286.5
8
55.4 ⫾ 8.3
68.5 ⫾ 3.3
84.1 ⫾ 8.6
27.9 ⫾ 3.1
19.5 ⫾ 5.0
9.9 ⫾ 1.2*
0.63 ⫾ 0.06
1,481.3 ⫾ 323.4
235.9 ⫾ 41.4
1,710.5 ⫾ 329.4
6.9 ⫾ 2.2
3.1 ⫾ 3.0
2.7 ⫾ 0.85
1.5 ⫾ 0.8
1.3 ⫾ 0.6
24.2 ⫾ 14.5
23.8 ⫾ 7.3
56.4 ⫾ 9.4
77.7 ⫾ 22.0
35.3 ⫾ 22.3
44.1 ⫾ 23.5
Values are means ⫾ SD. *P ⬍ 0.05; †P ⬍ 0.01.
J Appl Physiol • doi:10.1152/japplphysiol.01042.2012 • www.jappl.org
5.4 ⫾ 3.8
12.7 ⫾ 11.4†
2.4 ⫾ 0.5
1.6 ⫾ 0.5
1.1 ⫾ 1.0
22.6 ⫾ 14.8
24.4 ⫾ 5.4
53.7 ⫾ 10.2
67.5 ⫾ 16.7
38.7 ⫾ 18.6
67.9 ⫾ 29.8
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
patients. Our data do, however, suggest that detailed metabolic
studies using isotopic tracers in decompensated patients with
cirrhosis are required to examine the mechanisms of these
metabolic responses.
Whole body RQ is determined by the contribution of the
metabolism of individual organs (24). It has been suggested
that the liver contributes to 20 –30% of whole body energy
expenditure at rest (24). Because hepatic substrate utilization is
altered in cirrhosis due to depletion of glycogen stored and
increased fatty acid oxidation, the hepatic RQ is anticipated to
be lower than the whole body RQ. The contribution of hepatic
RQ to whole body RQ is also determined by the relative
contribution of other organs like the skeletal muscle (24).
Because skeletal muscle completely oxidizes ketones and is not
gluconeogenic (37, 38), its RQ is higher than that of the liver.
If the skeletal muscle mass was unaltered, then its contribution
to whole body RQ would be increased in patients with cirrhosis, resulting in higher RQs than are observed in the present
study. However, skeletal muscle area was significantly lower in
patients with cirrhosis than controls, and this observation is
similar to that previously reported by us and others (7, 22, 52).
As a consequence, the contribution of skeletal muscle metabolism to whole body RQ is likely to result in a reduction in
whole body RQs. Very low RQ in this population of patients
564
Abnormally Low Respiratory Quotient in Cirrhosis
Glass C et al.
as it was in the patients with cirrhosis alone. Interestingly, in
the present study, we did not observe a difference in muscle
area between survivors and nonsurvivors. This is different
from that previously reported in nonhospitalized, free living
patients with cirrhosis (22). This may be due to differences in
the metabolic and other clinical characteristics of the patients
in the present study.
In conclusion, we demonstrate for the first time that low RQs
below 0.69 occur in advanced liver disease and do not necessarily predict mortality. The direct and significant relation
between RQ and muscle area suggests that altered skeletal
muscle protein turnover contributes to this metabolic response.
In vivo studies of individual organ contributions specifically
focusing on skeletal muscle contribution to substrate utilization
patterns in cirrhosis may provide the basis for specific interventions to reverse the reduced muscle mass in these patients.
GRANTS
This work was supported in part by the National Institutes of Health,
National Center for Research Resources, CTSA 1UL1RR024989, Cleveland,
OH; S.D. was supported by NIH RO1 DK083414 and NIH UO1 DK 061732;
J.P.K. was supported in part by NIH RO1 AG12834; S.K.M. was supported in
part by T32DK007319.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors.
AUTHOR CONTRIBUTIONS
Author contributions: C.G., P.H., T.K., J.P.K., and S.D. conception and
design of research; C.G., P.H., C.T., S.K.M., S.N.S., J.P.K., and S.D. performed experiments; C.G., P.H., C.T., S.K.M., T.K., S.N.S., J.P.K., and S.D.
analyzed data; C.G., P.H., S.K.M., T.K., S.N.S., J.P.K., and S.D. interpreted
results of experiments; C.G., S.K.M., J.P.K., and S.D. prepared figures; C.G.,
P.H., S.K.M., T.K., S.N.S., J.P.K., and S.D. drafted manuscript; C.G., P.H.,
C.T., S.K.M., T.K., S.N.S., J.P.K., and S.D. edited and revised manuscript;
C.G., P.H., C.T., S.K.M., T.K., S.N.S., J.P.K., and S.D. approved final version
of manuscript.
REFERENCES
1. Bosy-Westphal A, Kossel E, Goele K, Later W, Hitze B, Settler U,
Heller M, Gluer CC, Heymsfield SB, Muller MJ. Contribution of
individual organ mass loss to weight loss-associated decline in resting
energy expenditure. Am J Clin Nutr 90: 993–1001, 2009.
2. Bugianesi E, Kalhan S, Burkett E, Marchesini G, McCullough A.
Quantification of gluconeogenesis in cirrhosis: response to glucagon.
Gastroenterology 115: 1530 –1540, 1998.
3. Cahill GF Jr. Fuel metabolism in starvation. Annu Rev Nutr 26: 1–22,
2006.
4. Changani KK, Jalan R, Cox IJ, Ala-Korpela M, Bhakoo K, TaylorRobinson SD, Bell JD. Evidence for altered hepatic gluconeogenesis in
patients with cirrhosis using in vivo 31-phosphorus magnetic resonance
spectroscopy. Gut 49: 557–564, 2001.
5. Creutzberg EC, Schols AM, Bothmer-Quaedvlieg FC, Wouters EF.
Prevalence of an elevated resting energy expenditure in patients with
chronic obstructive pulmonary disease in relation to body composition and
lung function. Eur J Clin Nutr 52: 396 –401, 1998.
6. Curtis DE, Bradfield RB. Long-term energy intake and expenditure of
obese housewives. Am J Clin Nutr 24: 1410 –1417, 1971.
7. Englesbe MJ, Patel SP, He K, Lynch RJ, Schaubel DE, Harbaugh C,
Holcombe SA, Wang SC, Segev DL, Sonnenday CJ. Sarcopenia and
mortality after liver transplantation. J Am Coll Surg 211: 271–278, 2010.
8. Essen P, McNurlan MA, Wernerman J, Milne E, Vinnars E, Garlick
PJ. Short-term starvation decreases skeletal muscle protein synthesis rate
in man. Clin Physiol 12: 287–299, 1992.
9. Ferrannini E. The theoretical bases of indirect calorimetry: a review.
Metabolism 37: 287–301, 1988.
10. Garlick PJ, Millward DJ, James WP. The diurnal response of muscle
and liver protein synthesis in vivo in meal-fed rats. Biochem J 136:
935–945, 1973.
J Appl Physiol • doi:10.1152/japplphysiol.01042.2012 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
rived amino acids to gluconeogenesis contribute to sarcopenia
in these patients. However, it is possible that the low RQ may
not be directly responsible for the reduction in muscle mass,
rather, it may be due to the metabolic perturbations that result
in a low RQ.
Consistent with the contribution of increased fatty acid
oxidation to the reduction in RQ in cirrhosis, we observed a
reduction in both visceral and subcutaneous fat mass measured
on imaging. These observations are similar to those we have
reported earlier in a separate cohort of patients with cirrhosis
who had lower visceral fat mass (52). This is also consistent
with the trace ketonuria seen in these patients. Our observations, however, only show an association between the low RQ
and fat mass and more detailed direct metabolic kinetic studies
using tracers will be necessary to demonstrate that the low fat
mass in cirrhosis is accompanied by increased fatty acid
oxidation and consequent low RQ (19). It was interesting that
on multivariate analysis, psoas muscle area had a significant
independent effect on RQ, whereas fat mass showed only a
trend to altering RQ. Our studies also do not permit us to
examine the differential contribution of increased gluconeogenesis, ketogenesis, and fatty acid oxidation to the lower RQ
and such studies are necessary to design targeted interventions
to reverse these metabolic abnormalities in cirrhosis. In the
present studies, we used CT images at L4 vertebra as estimates
of whole body fat and muscle mass because these have been
shown by others to be a valid measure of whole body fat and
muscle mass (23, 47). Although REE in this group of patients
with cirrhosis was similar to that in controls and only 12%
could be considered hypermetabolic, patients with cirrhosis
had a significantly higher mean REE normalized to muscle
area. Because skeletal muscle is the largest metabolically
active organ, normalizing for muscle area showed that these
patients appear to have a higher metabolic demand due to the
contribution from other organs (1, 25). Loss of muscle mass
with hypermetabolism has been reported in chronic renal
failure, chronic obstructive lung disease, and congestive heart
failure (5, 44, 49, 53, 54). However, the relation between RQ
and body composition in these patients has not been studied in
detail. Our data suggest that similar metabolic studies to
quantify the contribution of substrate dysmetabolism may
provide potential targets and outcome measures to study the
response to interventions to reverse sarcopenia.
Such a severe derangement in energy metabolism is expected to result in adverse outcomes. RQ values ⬍0.85 has
been previously reported to increase mortality over a long-term
period (48). Our studies, however, did not demonstrate such an
adverse effect on survival during the shorter period of follow
up in patients with cirrhosis. Because all the patients with
cirrhosis in our study had RQs ⬍0.85, the ability of RQ to
predict outcome may have been lost. This may reflect a
threshold effect where, after a certain degree of metabolic
perturbation has occurred, the outcome is determined by the
whole body compensatory processes that may not necessarily
alter the measured RQ. This may also explain the absence of
more severe muscle loss in patients with RQ ⬍0.68 compared
with those with an RQ ⬎0.68 because the majority of our
patients had a very low RQ. The relation between RQ and
muscle area, however, became evident when this was examined in both patients with cirrhosis and control subjects because the RQ was no longer restricted to a very narrow range
•
Abnormally Low Respiratory Quotient in Cirrhosis
Glass C et al.
565
32. Owen OE, Reichard GA Jr, Patel MS, Boden G. Energy metabolism in
feasting and fasting. Adv Exp Med Biol 111: 169 –188, 1979.
33. Owen OE, Reichle FA, Mozzoli MA, Kreulen T, Patel MS, Elfenbein
IB, Golsorkhi M, Chang KH, Rao NS, Sue HS, Boden G. Hepatic, gut,
and renal substrate flux rates in patients with hepatic cirrhosis. J Clin
Invest 68: 240 –252, 1981.
34. Peng S, Plank LD, McCall JL, Gillanders LK, McIlroy K, Gane EJ.
Body composition, muscle function, and energy expenditure in patients
with liver cirrhosis: a comprehensive study. Am J Clin Nutr 85: 1257–
1266, 2007.
35. Periyalwar P, Dasarathy S. Malnutrition in cirrhosis: contribution and
consequences of sarcopenia on metabolic and clinical responses. Clin
Liver Dis 16: 95–131, 2012.
36. Petersen KF, Krssak M, Navarro V, Chandramouli V, Hundal R,
Schumann WC, Landau BR, Shulman GI. Contributions of net hepatic
glycogenolysis and gluconeogenesis to glucose production in cirrhosis.
Am J Physiol Endocrinol Metab 276: E529 –E535, 1999.
37. Previs SF, Brunengraber DZ, Brunengraber H. Is there glucose production outside of the liver and kidney? Annu Rev Nutr 29: 43–57, 2009.
38. Previs SF, Fernandez CA, Yang D, Soloviev MV, David F, Brunengraber H. Limitations of the mass isotopomer distribution analysis of
glucose to study gluconeogenesis. Substrate cycling between glycerol and
triose phosphates in liver. J Biol Chem 270: 19806 –19815, 1995.
39. Reichle FA, Owen OE, Golsorkhi M, Kreulen T. Hepatic metabolism in
patients with alcoholic cirrhosis. Surgery 84: 33–36, 1978.
40. Riggio O, Angeloni S, Ciuffa L, Nicolini G, Attili AF, Albanese C,
Merli M. Malnutrition is not related to alterations in energy balance in
patients with stable liver cirrhosis. Clin Nutr 22: 553–559, 2003.
41. Riggio O, Merli M, Capocaccia L. Macronutrient oxidation in liver
disease. Ital J Gastroenterol 25: 272–275, 1993.
42. Riggio O, Merli M, Leonetti F, Giovannetti P, Foniciello M, Folino S,
Tamburrano G, Capocaccia L. Impaired nonoxidative glucose metabolism in patients with liver cirrhosis: effects of two insulin doses. Metabolism 46: 840 –843, 1997.
43. Riggio O, Merli M, Romiti A, Pinto G, Fanella R, Attili AF, Capocaccia L. Early postprandial energy expenditure and macronutrient use
after a mixed meal in cirrhotic patients. J Parenter Enteral Nutr 16:
445–450, 1992.
44. Roubenoff R, Roubenoff RA, Cannon JG, Kehayias JJ, Zhuang H,
Dawson-Hughes B, Dinarello CA, Rosenberg IH. Rheumatoid cachexia:
cytokine-driven hypermetabolism accompanying reduced body cell mass
in chronic inflammation. J Clin Invest 93: 2379 –2386, 1994.
45. Schutz Y, Ravussin E. Respiratory quotients lower than 0.70 in ketogenic
diets. Am J Clin Nutr 33: 1317–1319, 1980.
46. Scolapio JS, Bowen J, Stoner G, Tarrosa V. Substrate oxidation in
patients with cirrhosis: comparison with other nutritional markers. J
Parenter Enteral Nutr 24: 150 –153, 2000.
47. Shen W, Punyanitya M, Wang Z, Gallagher D, St-Onge MP, Albu J,
Heymsfield SB, Heshka S. Total body skeletal muscle and adipose tissue
volumes: estimation from a single abdominal cross-sectional image. J
Appl Physiol 97: 2333–2338, 2004.
48. Tajika M, Kato M, Mohri H, Miwa Y, Kato T, Ohnishi H, Moriwaki
H. Prognostic value of energy metabolism in patients with viral liver
cirrhosis. Nutrition 18: 229 –234, 2002.
49. Toth MJ, Matthews DE. Whole-body protein metabolism in chronic
heart failure: relationship to anabolic and catabolic hormones. J Parenter
Enteral Nutr 30: 194 –201, 2006.
50. Trevisani F, Bernardi M, Gasbarrini G. Gluconeogenesis from nitrogen
precursors and hyperammonemia of cirrhosis: is there a cause-effect
relationship? Hepatology 9: 343–345, 1989.
51. Tsien CD, McCullough AJ, Dasarathy S. Late evening snack: exploiting
a period of anabolic opportunity in cirrhosis. J Gastroenterol Hepatol 27:
430 –441, 2012.
52. Tsien CD, Shah S, McCullough A, Dasarathy S. Reversal of sarcopenia
predicts survival after transjugular intrahepatic portosystemic stent. Eur J
Gastroenterol Hepatol 25: 85–93, 2013.
53. Wang AY, Sea MM, Tang N, Sanderson JE, Lui SF, Li PK, Woo J.
Resting energy expenditure and subsequent mortality risk in peritoneal
dialysis patients. J Am Soc Nephrol 15: 3134 –3143, 2004.
54. Yoneda T, Yoshikawa M, Fu A, Tsukaguchi K, Okamoto Y, Takenaka
H. Plasma levels of amino acids and hypermetabolism in patients with
chronic obstructive pulmonary disease. Nutrition 17: 95–99, 2001.
J Appl Physiol • doi:10.1152/japplphysiol.01042.2012 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
11. Garlick PJ, Millward DJ, James WP, Waterlow JC. The effect of
protein deprivation and starvation on the rate of protein synthesis in tissues
of the rat. Biochim Biophys Acta 414: 71–84, 1975.
12. Glass C, Hipskind P, Cole D, Lopez R, Dasarathy S. Handheld calorimeter is a valid instrument to quantify resting energy expenditure in
hospitalized cirrhotic patients: a prospective study. Nutr Clin Pract 27:
677–688, 2012.
13. Hamwi GJ. Therapy (changing dietary concepts). In: Diabetes Mellitus:
Diagnosis and Treatment, vol. 1, edited by Danowski TS. NY: American
Diabetes Association, 1964, p. 73–78.
14. Hipskind P, Glass C, Charlton D, Nowak D, Dasarathy S. Do handheld
calorimeters have a role in assessment of nutrition needs in hospitalized
patients? A systematic review of literature. Nutr Clin Pract 26: 426 –433,
2011.
15. Janardhan A, Chen J, Crawford PA. Altered systemic ketone body
metabolism in advanced heart failure. Tex Heart Inst J 38: 533–538, 2011.
16. Kamath PS, Wiesner RH, Malinchoc M, Kremers W, Therneau TM,
Kosberg CL, D’Amico G, Dickson ER, Kim WR. A model to predict
survival in patients with end-stage liver disease. Hepatology 33: 464 –470,
2001.
17. Krahenbuhl S, Weber FL Jr., Brass EP. Decreased hepatic glycogen
content and accelerated response to starvation in rats with carbon tetrachloride-induced cirrhosis. Hepatology 14: 1189 –1195, 1991.
18. McClave SA, Lowen CC, Kleber MJ, McConnell JW, Jung LY,
Goldsmith LJ. Clinical use of the respiratory quotient obtained from
indirect calorimetry. J Parenter Enteral Nutr 27: 21–26, 2003.
19. Merli M, Eriksson LS, Hagenfeldt L, Wahren J. Splanchnic and leg
exchange of free fatty acids in patients with liver cirrhosis. J Hepatol 3:
348 –355, 1986.
20. Merli M, Leonetti F, Riggio O, Giaccari A, Romiti A, Sbraccia P,
Tamburrano G. Resistance to insulin suppression of plasma free fatty
acids in liver cirrhosis. J Endocrinol Invest 13: 787–795, 1990.
21. Merli M, Riggio O, Romiti A, Ariosto F, Mango L, Pinto G, Savioli M,
Capocaccia L. Basal energy production rate and substrate use in stable
cirrhotic patients. Hepatology 12: 106 –112, 1990.
22. Montano-Loza AJ, Meza-Junco J, Prado CM, Lieffers JR, Baracos
VE, Bain VG, Sawyer MB. Muscle wasting is associated with mortality
in patients with cirrhosis. Clin Gastroenterol Hepatol 10: 166 –73, 173,
2012.
23. Mourtzakis M, Prado CM, Lieffers JR, Reiman T, McCargar LJ,
Baracos VE. A practical and precise approach to quantification of body
composition in cancer patients using computed tomography images acquired during routine care. Appl Physiol Nutr Metab 33: 997–1006, 2008.
24. Muller MJ. Hepatic energy and substrate metabolism: a possible metabolic basis for early nutritional support in cirrhotic patients. Nutrition 14:
30 –38, 1998.
25. Muller MJ, Bosy-Westphal A, Kutzner D, Heller M. Metabolically
active components of fat free mass (FFM) and resting energy expenditure
(REE) in humans. Forum Nutr 56: 301–303, 2003.
26. Muller MJ, Bottcher J, Selberg O. Energy expenditure and substrate
metabolism in liver cirrhosis. Int J Obes Relat Metab Disord 17, Suppl 3:
S102–S106, 1993.
27. Muller MJ, Bottcher J, Selberg O, Weselmann S, Boker KH,
Schwarze M, von zur MA, Manns MP. Hypermetabolism in clinically
stable patients with liver cirrhosis. Am J Clin Nutr 69: 1194 –1201, 1999.
28. Muller MJ, Fenk A, Lautz HU, Selberg O, Canzler H, Balks HJ, von
zur MA, Schmidt E, Schmidt FW. Energy expenditure and substrate
metabolism in ethanol-induced liver cirrhosis. Am J Physiol Endocrinol
Metab 260: E338 –E344, 1991.
29. Muller MJ, Lautz HU, Plogmann B, Burger M, Korber J, Schmidt
FW. Energy expenditure and substrate oxidation in patients with cirrhosis:
the impact of cause, clinical staging and nutritional state. Hepatology 15:
782–794, 1992.
30. Nielsen MF, Caumo A, Aagaard NK, Chandramouli V, Schumann
WC, Landau BR, Schmitz O, Vilstrup H. Contribution of defects in
glucose uptake to carbohydrate intolerance in liver cirrhosis: assessment
during physiological glucose and insulin concentrations. Am J Physiol
Gastrointest Liver Physiol 288: G1135–G1143, 2005.
31. Owen OE, Mozzoli MA, Reichle FA, Kreulen TH, Owen RS, Boden G,
Polansky M. Hepatic and renal metabolism before and after portasystemic
shunts in patients with cirrhosis. J Clin Invest 76: 1209 –1217, 1985.
•