Age-related skeletal muscle decline is similar in HIV-infected and uninfected individuals.
ABSTRACT Skeletal muscle (SM) mass decreases with advanced age and with disease in HIV infection. It is unknown whether age-related muscle loss is accelerated in the current era of antiretroviral therapy and which factors might contribute to muscle loss among HIV-infected adults. We hypothesized that muscle mass would be lower and decline faster in HIV-infected adults than in similar-aged controls.
Whole-body (1)H-magnetic resonance imaging was used to quantify regional and total SM in 399 HIV-infected and 204 control men and women at baseline and 5 years later. Multivariable regression identified associated factors.
At baseline and Year 5, total SM was lower in HIV-infected than control men. HIV-infected women were similar to control women at both time points. After adjusting for demographics, lifestyle factors, and total adipose tissue, HIV infection was associated with lower Year 5 SM in men and higher SM in women compared with controls. Average overall 5-year change in total SM was small and age related, but rate of change was similar in HIV-infected and control men and women. CD4 count and efavirenz use in HIV-infected participants were associated with increasing SM, whereas age and stavudine use were associated with decreasing SM.
Muscle mass was lower in HIV-infected men compared with controls, whereas HIV-infected women had slightly higher SM than control women after multivariable adjustment. We found evidence against substantially faster SM decline in HIV infected versus similar-aged controls. SM gain was associated with increasing CD4 count, whereas stavudine use may contribute to SM loss.
- [Show abstract] [Hide abstract]
ABSTRACT: To compare the effect that initiating different antiretroviral therapy (ART) regimens has on weight, BMI, and lean body mass (LBM) and explore how changes in body composition are associated with bone mineral density (BMD). A5224s was a sub-study of A5202, a prospective trial of 1857 ART-naive participants randomized to blinded abacavir-lamivudine (ABC/3TC) or tenofovir DF-emtricitabine (TDF/FTC) with open-label efavirenz (EFV) or atazanavir-ritonavir (ATV/r). All participants underwent dual-energy absorptiometry (DXA) and abdominal computed tomography for body composition. Analyses used two-sample t-tests and linear regression. A5224s included 269 participants: 85% men, 47% white non-Hispanic, median age 38 years, HIV-1 RNA 4.6 log10 copies/ml, and CD4 cell count 233 cells/μl. Overall, significant gains occurred in weight, BMI, and LBM at 96 weeks postrandomization (all P < 0.001). Assignment to ATV/r (vs. EFV) resulted in significantly greater weight (mean difference 3.35 kg) and BMI gain (0.88 kg/m; both P = 0.02), but not LBM (0.67 kg; P = 0.15), whereas ABC/3TC and TDF/FTC were not significantly different (P ≥ 0.10). In multivariable analysis, only lower baseline CD4 cell count and higher HIV-1 RNA were associated with greater increase in weight, BMI, or LBM. In multivariable analyses, increased LBM was associated with an increased hip BMD. ABC/3TC vs. TDF/FTC did not differ in change in weight, BMI, or LBM; ATV/r vs. EFV resulted in greater weight and BMI gain but not LBM. A positive association between increased LBM and increased hip BMD should be further investigated through prospective interventional studies to verify the impact of increased LBM on hip BMD.AIDS (London, England) 08/2013; 27(13):2069-79. · 6.56 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Motor impairment is highly prevalent in HIV-infected patients. Here, we assess associations between peripheral muscular deficits as evaluated by the 5 sit-to-stand test (5STS) and structural integrity of the motor system at a central level. Eighty-six HIV-infected patients receiving combination antiretroviral therapy and with no major cerebral events, underwent an MRI scan and the 5STS. Out of 86 participants, forty presented a score greater than two standard deviations above mean normative scores calculated for the 5STS and were therefore considered as motor-impaired. MRI-structural cerebral parameters were compared to the unimpaired participants. Fractional Anisotropy (FA), Axial Diffusivity (AD) and Radial Diffusivity (RD), reflecting microstructural integrity, were extracted from Diffusion-Tensor MRI. Global and regional cerebral volumes or thicknesses were extracted from 3D-T1 morphological MRI. Whereas the two groups did not differ for any HIV variables, voxel-wise analysis revealed that motor-impaired participants present low FA values in various cortico-motor tracts and low AD in left cortico-spinal tract. However, they did not present reduced volumes or thicknesses of the precentral cortices compared to unimpaired participants. The absence of alterations in cortical regions holding motor-neurons might argue against neurodegenerative process as an explanation of White Matter (WM) disorganization.PLoS ONE 07/2013; 8(7):66810-. · 3.53 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: In HIV-infected persons, osteoporosis is common and has a multifactorial etiology including traditional risk factors, such as smoking and low body weight, as well as direct effects of HIV infection and antiretroviral therapy. Multiple studies indicate that HIV-infected persons are at increased risk of low bone mass as compared to the general population. Emerging data suggest that the increased prevalence of reduced bone mass in HIV infection predisposes patients to an increased risk of fracture. This review discusses the epidemiology of low bone mass and fracture in HIV-infected persons, addresses the multiple causes of reduced bone mineral density in HIV infection, and offers recommendations on screening HIV-infected persons for bone loss.Clinical Reviews in Bone and Mineral Metabolism 12/2012; 10(4).
Journal of Gerontology: MEDICAL SCIENCES
Cite journal as: J Gerontol A Biol Sci Med Sci. 2011 March;66A(3):332–340
Published by Oxford University Press on behalf of the Gerontological Society of America 2011.
Advance Access published on February 10, 2011
of non–AIDS-related mortality and morbidity (1,2). It has
been reported that many of these conditions appear to occur at
an earlier age in HIV-infected persons (3). Skeletal muscle
(SM) mass is known to decrease both with advanced age (sar-
copenia) and with disease (cachexia), including HIV infection
(4). Unintentional weight loss and SM cachexia are associ-
ated with increased mortality risk (5–8). Even in the modern
highly active antiretroviral therapy era, wasting and weight
loss remain common in some HIV-infected patients (7).
ESPITE the introduction of highly active antiretroviral
therapy, HIV-infected persons remain at increased risk
It is unknown whether the rate of age-related muscle loss
is accelerated in HIV infection and which factors might
contribute to muscle loss among HIV-infected adults.
In uninfected participants, in the cross-sectional National
Health and Nutrition Examination Survey (NHANES) study,
dual-energy X-ray absorptiometry (DXA)-measured fat-free
mass appeared to remain stable or increase until age 40–45
years and was lower in older ages in both men and women (9).
In an uninfected cohort recruited at age 46–80 years, mus-
cle mass declined at a rate of 1%–2% per year (10). In adults
aged 70–79 years, Goodpaster and colleagues (11) reported
Age-Related Skeletal Muscle Decline Is Similar in
HIV-Infected and Uninfected Individuals
Kevin E. Yarasheski,1,* Rebecca Scherzer,2,3,* Donald P. Kotler,4 Adrian S. Dobs,5 Phyllis C. Tien,2,3
Cora E. Lewis,6 Richard A. Kronmal,7 Steven B. Heymsfield,8 Peter Bacchetti,9 and Carl Grunfeld,2,3
for the Study of Fat Redistribution and Metabolic Change in HIV Infection (FRAM)
1Department of Medicine, Washington University, St. Louis, Missouri.
2Department of Medicine, University of California, San Francisco.
3Metabolism Section, Veterans Affairs Medical Center, San Francisco, California.
4Division of Gastroenterology, St. Luke’s-Roosevelt Hospital Center, New York, New York.
5Division of Endocrinology, Johns Hopkins University, Baltimore, Maryland.
6Division of Preventive Medicine, University of Alabama at Birmingham.
7Department of Biostatistics, University of Washington, Seattle.
8Pennington Biomedical Research Center, Baton Rouge, Louisiana.
9Department of Epidemiology and Biostatistics, University of California, San Francisco.
*These contributed equally to this work.
Address correspondence to Carl Grunfeld, MD, PhD, University of California, Veterans Affairs Medical Center, Metabolism Section 111F, 4150 Clement
Street, San Francisco, CA 94121. Email: email@example.com
Background. Skeletal muscle (SM) mass decreases with advanced age and with disease in HIV infection. It is un-
known whether age-related muscle loss is accelerated in the current era of antiretroviral therapy and which factors might
contribute to muscle loss among HIV-infected adults. We hypothesized that muscle mass would be lower and decline
faster in HIV-infected adults than in similar-aged controls.
Methods. Whole-body 1H-magnetic resonance imaging was used to quantify regional and total SM in 399
HIV-infected and 204 control men and women at baseline and 5 years later. Multivariable regression identified
Results. At baseline and Year 5, total SM was lower in HIV-infected than control men. HIV-infected women were
similar to control women at both time points. After adjusting for demographics, lifestyle factors, and total adipose tissue,
HIV infection was associated with lower Year 5 SM in men and higher SM in women compared with controls. Average
overall 5-year change in total SM was small and age related, but rate of change was similar in HIV-infected and control
men and women. CD4 count and efavirenz use in HIV-infected participants were associated with increasing SM, whereas
age and stavudine use were associated with decreasing SM.
Conclusions. Muscle mass was lower in HIV-infected men compared with controls, whereas HIV-infected women
had slightly higher SM than control women after multivariable adjustment. We found evidence against substantially faster
SM decline in HIV infected versus similar-aged controls. SM gain was associated with increasing CD4 count, whereas
stavudine use may contribute to SM loss.
Key Words: Sarcopenia—Lipoatrophy—Fat redistribution—Body composition.
Received August 2, 2010; Accepted November 11, 2010
Decision Editor: Luigi Ferrucci, MD, PhD
SKELETAL MUSCLE IN HIV INFECTION
a loss of leg lean mass of approximately 1% per year. Another
recent study of men older than 50 years (8) found a loss of
92 g (0.16%) per year in total lean mass. Physical activity
was controlled for in some (8,10) but not all (9,11) studies.
Few longitudinal studies of lean mass in HIV infection
have been conducted in the recent highly active antiretrovi-
ral therapy era, and most were small. In one report, 23 clin-
ically stable HIV-infected adults (17 men and 6 women)
were studied before and 24 months after initiation of dual
nucleoside reverse transcriptase inhibitor–based therapy;
lean mass was relatively stable by DXA (0.5 kg decline,
p = .37) (12). Another 2-year study of 23 HIV-infected men
(17 on highly active antiretroviral therapy) and 26 healthy
controls found that lean mass tended to increase in HIV-in-
fected (1.3 kg, p = .13) and was stable in controls (13). A
study of 152 HIV-infected men and women found overall
increases in DXA-measured trunk lean mass of 0.9% per
year, with little change in extremity lean mass (14). A study
of 101 HIV-infected men with lipodystrophy found
DXA-measured lean mass stable over 4 years (15). Because
both lean and fat mass increase when body weight in-
creases(16–18), potential interactions between lean and fat
need to be considered when analyzing and interpreting
longitudinal alterations in lean mass.
No large nationally representative study has compared
SM changes over several years in a population of HIV-in-
fected and control participants. The aim of this analysis was
to determine the rate of SM change and factors associated
with SM after 5 years of follow-up in the Study of Fat
Redistribution and Metabolic Change in HIV infection
(FRAM), a large, nationally representative multi-ethnic
cohort of HIV-infected and control men and women, in
which regional adipose tissue (AT) distribution and changes
in AT over time have been studied (19–21). We hypothe-
sized that muscle mass would be lower and decline faster in
HIV-infected adults than in similar-aged controls.
The FRAM study was initially designed to evaluate the
prevalence and correlates of changes in fat distribution,
insulin resistance, and dyslipidemia in a representative
sample of HIV-infected participants and HIV-seronegative
controls in the United States. The methods used in the
FRAM study have been described in detail previously (22).
HIV-infected participants were recruited from 16 HIV or
infectious disease clinics or cohorts. HIV diagnosis was de-
termined at each individual clinical site. Control partici-
pants were recruited from two centers from the Coronary
Artery Risk Development in Young Adults (CARDIA)
study (23). CARDIA participants were originally recruited
as a sample of healthy 18- to 30-year-old Caucasian and
African American men and women from four cities in 1985
–1986 for a longitudinal study of cardiovascular risk fac-
tors, with population-based recruitment in three cities and
recruitment from the membership of a prepaid health care
program in the fourth city. Body mass index (BMI) in the
CARDIA population and within each group is very similar
to NHANES. FRAM recruited CARDIA participants en-
rolled in an ancillary study, the Visceral Fat and Metabolic
Rate in Young Adults (VIM) Study (24). The VIM ancillary
study recruited participants from two of the four CARDIA
centers in 1995–1996. The VIM ancillary study enrolled
approximately 100 CARDIA participants from each of
the race and gender groups with BMI distributed equally
above and below race- and gender-specific medians of the
population-based CARDIA study.
FRAM Year 5 retention outcomes for participants ini-
tially enrolled have been reported (2). At the second exam
(Year 5), 261 HIV-infected participants and 14 controls
could not be recontacted, 128 HIV-infected participants and
6 controls were deceased, and 213 HIV-infected partici-
pants and 36 controls declined to enroll. Compared with
those who were either alive or had unknown vital status at
Year 5, those who died were older, more often African
American, and had a higher prevalence of smoking, detect-
able HIVRNA, hepatitis C, history of AIDS, and more
kidney disease and inflammation (25).
The Year 5 follow-up exam included 581 HIV-infected
participants and 241 controls with baseline measures.
Three control participants were excluded because Year 5
review indicated that they were now taking antiretroviral
(ARV) medications. HIV and control participants were
excluded if they had contraindications to MRI, such as
metal implants, claustrophobia, or weight greater than
136 kg and height greater than 196 cm as per specifications
of the scanner manufacturers. We report here on the
subset of 399 HIV-infected participants and 204 controls
who had SM measured at both time points. Because a
greater percentage of HIV-infected participants did not
have measured MRI and were more likely to be deceased
by the second exam, we adjusted analyses as described
below to address the concern of selection bias. Institu-
tional review boards at all sites approved the protocols for
both FRAM exams.
Magnetic Resonance Imaging
Whole-body magnetic resonance imaging (MRI) was
performed to quantify regional and total SM and AT vol-
umes, as described in detail previously (11,19,22,26). All
scans were read by the same analyst at the Obesity Research
Center, St. Luke’s-Roosevelt Hospital, New York, NY.
Imaging techniques and anatomical sites (based on bone
landmarks) were identical between HIV-infected and con-
trol participants. In the baseline FRAM exam, SM volume
included intermuscular AT. At the Year 5 exam, SM and
intermuscular AT (27) were measured separately. We
YARASHESKI ET AL.
therefore added intermuscular AT to SM at Year 5 to enable
proper comparison of baseline and Year 5 SM measures.
Anatomic sites considered in this analysis were: legs,
lower trunk (abdomen and back), upper trunk (chest and
back), and arms. MRI reproducibility averaged 0.7% for
SM and 1.1% for AT (26). Volumes were converted to
mass, assuming a density of 0.92 kg/L for fat and 1.04
kg/L for muscle (26). SM and AT measures were adjusted
for height and frame size (measured by elbow breadth), as
described below. We did not adjust to BMI, as BMI is
influenced by the phenomenon being studied: quantity of
Height and weight were measured by standardized proto-
cols. Frame size (28) was assessed by measuring right
elbow breadth with a bicondylar Vernier caliper (±0.1 cm).
Standardized questionnaires were used to determine demo-
graphic characteristics; medical history; HIV risk factors;
and use of alcohol, tobacco, and illicit drugs (22,23).
Research associates interviewed participants and reviewed
medical charts regarding ARV medication use. A diagnosis
of AIDS was made by history of opportunistic infection or
CD4 count <200 c/mL.
Hepatitis C RNA testing was performed on frozen sera
using the Bayer Versant 3.0 branched DNA (bDNA) assay
(Leverkusen, Germany) in the entire cohort. CD4 lympho-
cyte count and percent, HIV RNA level in HIV-infected par-
ticipants, and other blood specimens were analyzed in a
single centralized laboratory (Covance, Indianapolis, IN).
Insulin resistance was calculated using the homeostasis
model assessment from fasting glucose (milligrams per
deciliter) and insulin (microunits per milliliter) concentra-
tions as: (insulin × glucose)/405. Cystatin C was measured
in previously frozen sera stored at −70°C, using a particle-
enhanced immunonephelometric assay (BNII nephelometer;
Dade Behring Inc., Deerfield, IL).
Analyses that compared characteristics of HIV-infected
participants with controls excluded HIV-infected individ-
uals with recent opportunistic infection (within a month)
and were restricted to those between the ages of 33 and 45
at baseline (n = 240) because the control population did not
include participants outside this age range. In a sensitivity
analysis, we compared SM measures for all HIV-infected
participants (19–76 years at baseline) with controls. This
analysis explored the possibility that SM mass within the
narrow age range was not representative of the extended age
range in HIV-infected cohort.
SM levels in Figure 1 were calculated using least squares
means from a linear regression model with dependent vari-
able of total SM, containing terms for HIV status, gender,
height, frame size, and quadratic terms for height and frame
size. Least squares means are within-group means adjusted
for other effects in the model and are also known as popula-
tion marginal means (29).
SM levels were compared between HIV-infected and
control groups using t tests. SM changes from baseline to
the 5-year follow-up exam were compared using a paired t
test within HIV-infected and control groups separately.
Spread in 5-year SM change between HIV-infected and
control participants was compared using Levene’s test for
equality of variance.
We analyzed SM using multivariable linear regression
with robust standard errors (30,31) with Year 5 SM or SM
change as the dependent variable. All analyses were
adjusted for height, frame size, and quadratic terms for
height and frame size, except where indicated otherwise.
Separate models were constructed, controlling sequentially
for (a) HIV status and demographic factors, (b) lifestyle fac-
tors, and (c) AT, because subcutaneous AT changes are
prevalent in HIV infection and differ between men and
women (19–21). To ensure that models were not overfit, we
built parsimonious models using a backward stepwise pro-
cedure. We also examined associations of homeostasis
model assessment and cystatin C with SM in exploratory
analyses. Age, gender, and race were included in every
model. Interactions of HIV status, gender, ethnicity, and age
with SM were assessed and included if they reached statisti-
cal significance. The linearity assumption was tested for
continuous measures by adding quadratic terms to the mod-
els (p > .4 for all linear models of change in SM shown) and
by examining generalized additive models (32). We also
analyzed relative percentage change in SM, defined as
log(follow-up SM) minus log(baseline SM), which is less
skewed than percentage change and which downweights
exceptionally large values of SM.
Candidate lifestyle factors assessed that might affect SM
included physical activity, tobacco use, alcohol use, ade-
quate food intake, and illicit drug use. We considered base-
line, Year 5, and changes from baseline as candidates in our
models. Candidate HIV-specific factors (tested only for
HIV models) included AIDS diagnosis, reported HIV dura-
tion, HIV RNA level (log 10), current and nadir CD4 count
(log 2), hepatitis C infection (by RNA), days since last
opportunistic infection, recent opportunistic infection status
(last 100 days), and HIV risk factors. In multivariable mod-
els controlling for the above factors, we evaluated ever use,
duration on, and duration off each individual ARV drug and
ARV class as previously defined (19).
Multiple imputation utilizing the Markov chain Monte
Carlo method for arbitrary missing data was used to impute
missing covariate values (33). HIV-infected participants
were missing MRI more often than controls due in part to
higher rates of death and loss to follow-up between exams,
as described elsewhere (2). To mitigate potential selection
bias, we therefore adjusted estimates using an inverse
probability weighting approach (34) by modeling each
SKELETAL MUSCLE IN HIV INFECTION
participant’s probability of having non-missing SM using
logistic regression analysis. The inverse of this probability
was used as a weight (applied to participants with measured
SM) in multivariable regression analyses.
All analyses were conducted using the SAS system, ver-
sion 9.2 (SAS Institute, Inc., Cary, NC).
MRI-measured SM was available on 603 participants
whose demographic and baseline characteristics are pre-
sented in Table 1. Within the age range of controls (33–45
years), HIV-infected and control participants were similar
in age, height, and percentage of Caucasians and African
Americans, but HIV-infected participants were more often
men (67% vs 51%) through the design of the controls. At
baseline, control participants weighed more and had higher
Comparison of Baseline and Year 5 Muscle Mass in
HIV-Infected and Control Participants
Mean SM at baseline and Year 5 in HIV-infected and
control participants, restricted to the age range of controls
(age 33–45 years at baseline) and adjusted as described in
the Methods section, are shown in Figure 1. HIV-infected
men had lower mean total SM compared with control men
at both baseline (29.0 vs 31.4 kg, p < .0001) and Year 5
(28.8 vs 31.6 kg, p < .0001). By comparison, HIV-infected
women and control women had similar mean total SM at
baseline (26.3 vs 25.5 kg, p = .23) and Year 5 (26.5 vs 26.2 kg,
p = .64).
Because HIV infection remained associated with lower
SM at Year 5 in men, we examined models of Year 5 total
SM to determine whether differences between HIV-infected
and control participants remained after multivariable
adjustment (Table 2). After multivariable adjustment for
demographic factors (age, sex, and race), HIV-infected men
averaged −3.3 kg lower SM compared with control men
(p < .0001), but SM was similar in HIV-infected and control
Table 1. Baseline Participant Characteristics by HIV Status and Age Group
HIV+ (AR, OI excluded) Control HIV+ (all)
Baseline age (y)
Current CD4 (cells/mL)
HIV RNA (1,000/mL)
Detectable HIV RNA
History of AIDS by OI/CD4
Hepatitis C infection
Notes: Includes participants with SM measured at both baseline and Year 5 exam. Data are presented as median (IQR) or numbers (percent). AR = age restricted;
BMI = body mass index; IQR = interquartile range; OI = opportunistic infection; SM = skeletal muscle.
***p < .0001; **p < .001 for HIV+ versus control comparison within overlapping age group.
Figure 1. Comparison of total skeletal muscle at baseline and year 5 by HIV
status and gender. Solid symbol = HIV+ patients and open symbol = control.
Skeletal muscle is adjusted for height and frame size as described in the
Methods section. Age restricted and opportunistic infection excluded. HIV ver-
sus control t test of HIV versus control at baseline (FRAM1) and Year 5
YARASHESKI ET AL.
women (only 0.23 kg higher in HIV-infected women,
p = .68). The HIV association in men was somewhat attenu-
ated after adjusting for demographic and lifestyle factors
(−2.8 kg, p < .0001). Further adjustment for total AT attenu-
ated the HIV association in men (to −1.3 kg, p = .020). In
women, HIV infection was associated with 1.2 kg higher
SM (p = .022) in fully adjusted models that included AT.
The test for HIV by gender interaction was statistically sig-
nificant in all models. Similar results were found for
regional SM (data not shown). Results were also similar in
separate sensitivity analyses: (a) without multiple imputa-
tion or adjustment for selection bias and (b) using the full
age range in the HIV sample (19–76 years at baseline).
Comparison of Changes in Muscle Mass of HIV-Infected
and Control Participants
The distribution of change in total SM between baseline
and Year 5 is shown in Figure 2. Because similar distribu-
tions for SM change were seen in men and women (data not
shown), analysis of change was pooled. Overall, the aver-
age change in absolute SM for HIV-infected and control
participants was small (−0.015 vs 0.42 kg, p = .20). How-
ever, there was a broader distribution of SM change within
the HIV-infected cohort than in the controls (SD = 4.3 vs
2.0, p = .0004).
We also sought to determine whether there was evidence
for accelerated SM loss in HIV-infected participants. Base-
line age was negatively associated with SM change in both
HIV-infected and control participants (Figure 3), meaning
that on average younger participants experienced SM gains
and older participants experienced SM losses. We therefore
modeled the difference in 5-year SM change between HIV-
infected and control participants, with separate estimates for
different ages (Table 3). After multivariable adjustment for
demographics, the expected 5-year SM changes from base-
line for individuals aged 35, 40, and 45 years at baseline
were small, and none of the differences between HIV and
controls in expected SM change reached statistical signifi-
cance. Estimated 5-year SM change for a 35-year-old
participant was +0.75 kg (p = .020) in controls and +0.28 kg
(p = .34) in HIV-infected participants. Expected SM change
for a 40-year-old participant was +0.19 kg (p = .39) in con-
trols and −0.14 kg (p = .61) in HIV participants. Expected
SM change for a 45-year-old participant was −0.36 kg (p =
.27) in controls and −0.55 kg (p = .055) in HIV participants.
After further multivariable adjustment for physical activity,
smoking, and change from baseline in total AT, the differ-
ences in SM between HIV and control narrowed and de-
creased in statistical significance. Expected 5-year SM change
in 35-year-old HIV-infected participants was −0.19 kg lower
than controls (p = .64), −0.064 kg lower at age 40 years (p =
.84), and +0.067 kg higher at age 45 years (p = .86).
In a sensitivity analysis, we restricted the age in the
multivariable model analysis to HIV-infected and control
participants in the same age range; we also found little
Table 2. Multivariable Linear Regression Analysis of Year 5 Muscle Mass by HIV Status
Combined Men and Women Men Women
(n = 240)
(n = 204)
(n = 161)
(n = 105)
(n = 79)
(n = 99)
Total SM (kg), M ± SD*
Estimate for HIV+ vs control (95% CI)
Adjusted for demographics†
Demographic + lifestyle†
(Demographic, lifestyle, AT)†
28.1 ± 4.528.8 ± 4.828.8 ± 4.631.6 ± 4.8 26.5 ± 5.2 26.2 ± 4.7
−1.79 (−2.64 to −0.95)
−1.18 (−2.08 to −0.28)
−3.34 (−4.50 to −2.18)
−2.78 (−3.97 to −1.59)
0.23 (−0.87 to 1.32)
1.07 (−0.037 to 2.18)
−0.24 (−1.09 to 0.61) .58 −1.33 (−2.46 to −0.21)
1.19 (0.18 to 2.20)
Notes: Results above are age restricted and OI excluded. Outcome is raw SM; model controls for height, frame size, and quadratic terms. Covariates were selected
from Year 5 exam. Bolded p-values denote statistical significance. AT = adipose tissue; CI = confidence interval; OI = opportunistic infection; SM = skeletal muscle.
* Mean ± SD for SM is adjusted for height, frame size, and quadratic terms, as in Figure 1.
† HIV by gender interaction: p < .001 for all models.
Figure 2. Distribution of change from baseline in total skeletal muscle (SM)
by HIV status. Solid line = HIV+ patients (n = 399), broken line = control (n =
204). Quantity plotted is: 5 × (raw change in total SM)/(years between exams).
Age restricted and opportunistic infection excluded. Distribution of change is
broader in HIV compared with control: p = .0004, using Levene’s test for ho-
mogeneity of variance. Patterns are similar in men and women.
SKELETAL MUSCLE IN HIV INFECTION
difference between HIV and control (age 35: +0.73 kg,
p = .17; age 40: +0.13 kg, p = .73; and age 45: −0.46 kg,
p = .41). Finally, we also analyzed relative percentage
change in SM from baseline. We again found little differ-
ence between HIV and control (age 35: −1.3%, p = .53; age
40: −0.77%, p = .67; and age 45: −0.26%, p = .90).
Factors Associated With Muscle Mass in HIV-Infected
We used multivariable analysis to identify factors that
were independently associated with total SM at Year 5
(Supplemental Table 1). We adjusted for demographics,
lifestyle factors, HIV-related factors, and AT. After adjust-
ment, age and female gender were associated with lower
total SM, whereas higher current CD4 and indinavir expo-
sure were associated with higher SM. Total AT was also
strongly associated with higher SM. Results were similar
when stratified by gender, although the indinavir associa-
tion was stronger in men.
Factors Associated With Change in Muscle Mass in
We used multivariable analysis to identify factors that
were independently associated with 5-year change in total
SM (Table 4). Overall, the average change in SM was nearly
zero (Figure 2), although there was a broad distribution of
change. We adjusted for demographics, lifestyle factors,
HIV-related factors, and AT. After adjustment, aging was
associated with decreased change in SM, whereas physical
activity and increases in CD4 were associated with
increased change in SM. Among those with no stavudine
exposure, the average change in SM was +0.18 kg, but each
year of exposure was associated with a larger SM loss
(−0.22 kg/year, p = .0061). By contrast, longer exposure to
efavirenz was associated with smaller SM loss (+0.15 kg/
year, p = .047). Increases in total AT were strongly associ-
ated with increased change in SM. There were no statisti-
cally significant interactions between age and demographics,
lifestyle factors, AT, or HIV-specific factors.
In this study of MRI-measured total body SM mass, we
found that age-associated SM changes occurred at a similar
rate over 5 years in HIV-infected adults as it did in similar-
aged HIV-negative controls. Even after multivariable ad-
justment, estimates for SM change were similar between
younger (35–40 years) as well as older (40–45 years) HIV-
infected participants and controls. These data suggest that
over 5 years, the typical rate of age-associated SM loss is
not greater or accelerated in 33- to 45-year-old HIV-infected
adults in the era of modern ARV therapy.
There were differences between men and women in the
association of HIV infection with SM levels at both exams.
After controlling for age, race, lifestyle factors, and adipos-
ity, HIV-infected men had less total SM than control men at
baseline, and their SM remained lower than in controls after
5 years. It is likely that some of the HIV effect on SM may
Table 3. Estimated 5-Year Change in Total SM (kg) by HIV Status (not age restricted)
Baseline AgeAge 35 Age 40Age 45
Estimate (95% CI)
p Value Estimate (95% CI)
p ValueEstimate (95% CI)
Control (n = 204)
HIV+ (n = 399)
HIV vs control (demographic adjusted)*
HIV vs control (fully adjusted)†
+0.75 (0.12 to 1.38)
+0.28 (−0.29 to 0.85)
−0.47 (−1.27 to 0.33)
−0.19 (−1.00 to 0.61)
+0.19 (−0.24 to 0.62)
−0.14 (−0.67 to 0.39)
−0.33 (−0.93 to 0.28)
−0.064 (−0.67 to 0.54)
−0.36 (−1.00 to 0.28)
−0.55 (−1.11 to 0.0052)
−0.19 (−0.97 to 0.59)
0.067 (−0.66 to 0.79)
Notes: Estimated 5-year change for those aged 35, 40, and 45 years at baseline from pooled not age-restricted HIV versus control models. Estimates are from
linear models not models stratified by age. CI = confidence interval; SM = skeletal muscle.
Outcome: 5 × (raw change in total SM)/(years between exams).
* Model is SM change = HIV + demographics (age, race, and gender).
† Model is SM change = HIV + demographics + lifestyle (Year 5) + change in AT.
Figure 3. Association of age with change from baseline in total skeletal
muscle by HIV status in men and women. Solid symbols/lines are HIV+ pa-
tients (n = 399) and open symbols and dashed lines are controls (n = 204).
Quantity plotted is: 5 × (raw change in total SM)/(years between exams). Nine
outliers (change more than ±10 kg) were excluded to enhance visibility of re-
gression lines. Outliers participated in calculation of regression lines. SM =
YARASHESKI ET AL.
be mediated by lifestyle factors and especially the decreased
adiposity of lipoatrophy. However, despite the significant
attenuation of the HIV effect in men when controlling for
lifestyle and especially AT, there is an additional effect of
HIV in men, beyond what these factors reflect. By contrast,
HIV-infected women, who start with greater AT, appeared
to have greater total SM at Year 5 compared with control
women after multivariable adjustment, with little attenua-
tion when adjusting for AT. These findings are consistent
with the previously published hypothesis that the greater
AT mass at baseline in women protects against subsequent
loss of SM (35–37).
Regardless, comparisons by sex indicated that over 5
years, the average change in SM in HIV-infected men and
women was similar to control men and women over the ex-
amined control age range. In HIV-infected participants, we
found that an increase in CD4 count over the 5 years of
study was a strong predictor of more gain/less loss in SM
(and conversely CD4 decrease was associated with more
loss/less gain in SM). This might have been expected be-
cause low CD4+ T-cell count and high plasma HIV viremia
are associated with a lower muscle protein synthesis rate,
and ARV therapy–induced improvements in immune and
virologic status increase muscle protein synthesis and re-
duce some aspects of muscle proteolysis (38,39), although
the precise molecular level regulator is not clear.
We found that higher CD4 count and indinavir exposure
were associated with greater Year 5 SM, even after control-
ling for total AT. The indinavir finding appears to contradict
in vitro and rodent findings, where indinavir exposure
acutely reduced SM protein synthesis and impaired protein
translation initiation and efficiency in cultured C2C12 myo-
cytes and in rats (40). Although the positive association of
CD4 count with SM is not novel (4,14,41), we find no other
reports in the literature of a positive relationship of indina-
vir with muscle mass. We cannot rule out confounding fac-
tors in those who were able to continue longer use of
indinavir. Of note, unlike stavudine and efavirenz, indinavir
did not show an independent association with SM change.
After adjustment, exposure to stavudine was associated
with more loss in SM, whereas efavirenz was associated
with smaller losses in SM. The association of stavudine use
with lower SM even after multivariable adjustment for AT
raises the question of other toxic effects of stavudine, per-
haps on nerve or muscle. Laboratory animals (42,43) and
HIV-seronegative adults (44) exposed to a short course of
stavudine experience sustained adverse effects on SM mito-
chondrial DNA copy number, biogenesis, and function, but
their relationship to muscle protein mass has not been re-
ported. We are unaware of any data on the effect of efavi-
renz on SM. In contrast, the Nutrition For Healthy Living
Study found little association of stavudine with lean body
mass changes (14) and did not report on efavirenz use.
There are some limitations to our study. At baseline,
the HIV-infected participants spanned a wider age range
(19–76 years) than the controls (33–45 years). This limits
our ability to compare the estimated rates of SM change in
HIV+ versus control in older participants. This was an
observational study, so we cannot infer a causal link
between stavudine, efavirenz, or indinavir use with SM
amounts or changes. The findings suggest against SM
declining substantially faster in HIV-infected adults than in
similar-aged controls. The broad confidence intervals in
Table 2 leave open some possibility that clinically relevant
SM loss may be accelerated in HIV-infected adults. How-
ever, the distribution of SM change was much broader in
HIV-infected men and women compared with controls,
even in age-restricted analyses. These data suggest that
greater weight loss does occur in a subset of HIV-infected
participants, and the multivariable analysis supports the
concept that weight change is related to CD4 count. An
additional limitation is that the SM measure at baseline
included intermuscular AT, whereas at Year 5 SM and inter-
muscular AT were measured separately. We dealt with this
Table 4. Multivariable Linear Regression Analysis of Change in Total SM (kg)* Over 5 Years in All HIV-Infected Participants (n = 399)
Model 1: Demographics
Adjusted R2 = 0.041
Model 2: Demographics,
Lifestyle, and HIV Related
Adjusted R2 = 0.11
Model 3: Demographics,
Lifestyle, and AT
Adjusted R2 = 0.10
Model 4: Demographic,
Lifestyle, HIV Related, and AT
Adjusted R2 = 0.16
Estimate (95% CI)
p ValueEstimate (95% CI)
p ValueEstimate (95% CI)
p ValueEstimate (95% CI)
Female vs male
African American vs white
Other vs white
Age (per decade)
Physical activity: Q2 vs Q1
Physical activity: Q3 vs Q1
Physical activity: Q4 vs Q1
Change in CD4 (doubling)
Efavirenz (per y)
Stavudine (per y)
Change in total AT (kg)
−0.15 (−0.99 to 0.70)
−0.57 (−1.52 to 0.38)
−0.87 (−1.85 to 0.12)
−0.81 (−1.19 to −0.42)
−0.25 (−1.34 to 0.83)
0.18 (−1.12 to 1.48)
1.08 (0.14 to 2.02)
−0.29 (−1.12 to 0.55)
−0.74 (−1.69 to 0.21)
−1.17 (−2.14 to −0.20)
−1.00 (−1.39 to −0.62)
−0.22(−1.27 to 0.83)
0.26(−0.99 to 1.50)
0.86(−0.079 to 1.80)
0.83 (0.47 to 1.18)
0.13(−0.018 to 0.29)
−0.19 (−0.35 to −0.035)
−0.12 (−0.91 to 0.67)
−0.29 (−1.18 to 0.60)
−0.72 (−1.74 to 0.29)
−0.78 (−1.13 to −0.42)
0.029 (−1.02 to 1.08)
0.48 (−0.68 to 1.64)
1.16 (0.25 to 2.07)
−0.13 (−0.92 to 0.65)
−0.34 (−1.21 to 0.54)
−0.93 (−1.88 to 0.023)
−0.91 (−1.26 to −0.56)
0.00 (−1.03 to 1.03)
0.57 (−0.55 to 1.70)
1.01 (0.092 to 1.93)
0.60 (0.22 to 0.98)
0.15 (0.0018 to 0.30)
−0.22 (−0.37 to −0.061)
0.14 (0.063 to 0.22)
0.15 (0.071 to 0.23)
Notes: Model fit was not improved by controlling for height and frame size. Grayed-out and italicized factors did not enter the model, but are shown added back
to the selected model. Bolded p-values denote statistical significance. AT = adipose tissue; SM = skeletal muscle.
* Outcome: 5 × (raw change in total SM)/(years between exams). Covariates were selected from Year 5 exam, except for change in CD4 and AT.
SKELETAL MUSCLE IN HIV INFECTION
by adding intermuscular AT to SM to make the Year 5 mea-
surement comparable to baseline. Although we controlled
for physical activity, we were unable to control for muscle
strength or quality. Previous studies have found that muscle
function may be more important than mass in predicting
morbidity and mortality (11,45,46). We also did not assess
for the frailty-related phenotype, which is increased in HIV
infection (47); the frailty-related phenotype is strongly as-
sociated with low CD4 count, consistent with our finding of
a positive association of change in CD4 count with change
in SM. Finally, there may have been incomplete or inade-
quate control for factors that may confound or explain the
association between HIV infection and SM. For example,
we did not measure gonadal steroid levels, which decline
with age and are lower in HIV infection.
A major strength of our study is the comparison of SM
change over 5 years directly measured using MRI in HIV-
infected and control adults. This allowed us to account for
normal SM changes with aging. The controls were enrolled
in the VIM substudy (24) of the CARDIA cohort, where the
average BMI is similar to that of the nationally representa-
tive sample of NHANES. A further strength of our study is
the ability to adjust for AT amount and changes, which in-
In conclusion, 5 years after the first exam in the FRAM
study, average change in SM was similar in HIV-infected
and control participants. HIV-infected men had lower SM
compared with control men, even after controlling for de-
mographics, lifestyle factors, and AT. HIV-infected women
had similar or slightly higher SM than control women. We
found suggestive evidence against substantially acceler-
ated SM loss in HIV infection. HIV-infected participants
were more prone to gain as well as lose SM compared with
controls. Increased CD4 count was associated with more
SM gain/less loss and decreased CD4 count with more
loss/less gain. Given that an increase in CD4 count is an
indication of the effectiveness of ARV therapy and a de-
crease in CD4 an indication of failure of ARV therapy, our
data support the concept that effective ARV therapy has an
important impact on changes in SM. As HIV muscle loss
and wasting 5% or more have been associated with mor-
bidity and mortality (8), the long-term consequences of the
wide spectrum of SM loss or gain found here need addi-
This work was supported by National Institutes of Health (NIH) grants
RO1-DK57508, HL74814, HL 53359, K23 AI66943, and UL1 RR024131;
NIH GCRC grants M01-RR00036, RR00051, RR00052, RR00054,
RR00083, RR0636, and RR00865; the Albert L. and Janet A. Schultz Sup-
porting Foundation; and with resources and the use of facilities of the Vet-
erans Affairs Medical Centers of, Atlanta, District of Columbia, New York,
and San Francisco. The funding agency had no role in the collection or
analysis of the data. Clinicaltrials.gov ID: NCT0033144
Conflict of Interest
All authors received funding from some of the supporting grants.
Supplementary material can be found at: http://biomed.gerontologyjournals
Role of the Funder
The funder played no role in the conduct of the study, collection of the
data, management of the study, analysis of data, interpretation of the data,
or preparation of the manuscript. A representative of the funding agent
participated in planning the protocol. As part of the standard operating
procedures of CARDIA, the manuscript was reviewed at the NHLBI, but
no revisions were requested.
Sites and Investigators
University Hospitals of Cleveland (Barbara Gripshover, MD); Tufts
University (Abby Shevitz, MD (deceased) and Christine Wanke, MD);
Stanford University (Andrew Zolopa, MD); University of Alabama at Bir-
mingham (Michael Saag, MD); John Hopkins University (Joseph Cofran-
cesco, MD, and Adrian Dobs, MD); University of Colorado Heath Sciences
Center (Lisa Kosmiski, MD, and Constance Benson, MD); University of
North Carolina at Chapel Hill (David Wohl, MD, and Charles van der
Horst, MD*); University of California at San Diego (Daniel Lee, MD, and
W. Christopher Mathews, MD*); Washington University (E. Turner Over-
ton, MD, and William Powderly, MD); VA Medical Center, Atlanta (David
Rimland, MD); University of California at Los Angeles (Judith Currier,
MD); VA Medical Center, New York (Michael Simberkoff, MD); VA Med-
ical Center, Washington DC (Cynthia Gibert, MD); St Luke’s-Roosevelt
Hospital Center (Donald Kotler, MD, and Ellen Engelson, PhD); Kaiser
Permanente, Oakland (Stephen Sidney, MD); University of Alabama at
Birmingham (Cora E. Lewis, MD).
Fram 2 Data Coordinating Center
University of Washington, Seattle (Richard A. Kronmal, PhD, Mary
Louise Biggs, PhD, J. A. Christopher Delaney, Ph.D., and John Pearce).
Image Reading Centers
St Luke’s-Roosevelt Hospital Center (Steven Heymsfield, MD, Jack
Wang, MS, and Mark Punyanitya). Tufts New England Medical Center,
Boston (Daniel H. O’Leary, MD, Joseph Polak, MD, Anita P. Harrington).
Office of the Principal Investigator
University of California, San Francisco, Veterans Affairs Medical Cen-
ter and the Northern California Institute for Research and Development
(Carl Grunfeld, MD, PhD, Phyllis Tien, MD, Peter Bacchetti, PhD,
Michael Shlipak, MD, Rebecca Scherzer, PhD, Mae Pang, RN, MSN,
Heather Southwell, MS, RD)
1. Palella FJ Jr., Baker RK, Moorman AC, et al. Mortality in the highly
active antiretroviral therapy era: changing causes of death and disease
in the HIV outpatient study. J Acquir Immune Defic Syndr. 2006;43:
2. Cockerham L, Scherzer R, Zolopa A, et al. Association of HIV infection,
demographic and cardiovascular risk factors with all-cause mortality in
the recent HAART era. J Acquir Immune Defic Syndr. 2010;53:102–106.
3. Deeks SG. Immune dysfunction, inflammation, and accelerated aging
in patients on antiretroviral therapy. Top HIV Med. 2009;17:118–123.
4. Smit E, Skolasky RL, Dobs AS, et al. Changes in the incidence and
predictors of wasting syndrome related to human immunodeficiency
virus infection, 1987-1999. Am J Epidemiol. 2002;156:211–218.
5. Evans WJ. Skeletal muscle loss: cachexia, sarcopenia, and inactivity.
Am J Clin Nutr. 2010;91:1123S–1127S.
6. Grunfeld C, Feingold KR. Metabolic disturbances and wasting in the ac-
quired immunodeficiency syndrome. N Engl J Med. 1992;327:329–337.
7. Tang AM, Jacobson DL, Spiegelman D, Knox TA, Wanke C. Increas-
ing risk of 5% or greater unintentional weight loss in a cohort of HIV-
infected patients, 1995 to 2003. J Acquir Immune Defic Syndr. 2005;
YARASHESKI ET AL.
8. Szulc P, Munoz F, Marchand F, Chapurlat R, Delmas PD. Rapid loss
of appendicular skeletal muscle mass is associated with higher all-
cause mortality in older men: the prospective MINOS study. Am J Clin
9. Li C, Ford ES, Zhao G, Balluz LS, Giles WH. Estimates of body co-
mposition with dual-energy X-ray absorptiometry in adults. Am J Clin
10. Hughes VA, Frontera WR, Roubenoff R, Evans WJ, Singh MA. Lon-
gitudinal changes in body composition in older men and women: role
of body weight change and physical activity. Am J Clin Nutr.
11. Goodpaster BH, Park SW, Harris TB, et al. The loss of skeletal muscle
strength, mass, and quality in older adults: the health, aging and
body composition study. J Gerontol A Biol Sci Med Sci. 2006;61:
12. Tsekes G, Chrysos G, Douskas G, et al. Body composition changes in
protease inhibitor-naive HIV-infected patients treated with two nucle-
oside reverse transcriptase inhibitors. HIV Med. 2002;3:85–90.
13. Bolland MJ, Grey AB, Horne AM, et al. Bone mineral density remains
stable in HAART-treated HIV-infected men over 2 years. Clin Endo-
crinol (Oxf). 2007;67:270–275.
14. McDermott AY, Terrin N, Wanke C, Skinner S, Tchetgen E, Shevitz
AH. CD4+ cell count, viral load, and highly active antiretroviral
therapy use are independent predictors of body composition altera-
tions in HIV-infected adults: a longitudinal study. Clin Infect Dis.
15. Degris E, Delpierre C, Sommet A, et al. Longitudinal study of body
composition of 101 HIV men with lipodystrophy: DXA criteria for
lipodystrophy evolution. J Clin Densitom. 2010;13(2):237–244.
16. Thomas DM, Das SK, Levine JA, et al. New fat free mass—fat mass
model for use in physiological energy balance equations. Nutr Metab.
2010; In press.
17. Heymsfield SB, Scherzer R, Pietrobelli A, Lewis CE, Grunfeld C.
Body mass index as a phenotypic expression of adiposity: quantitative
contribution of muscularity in a population-based sample. Int J Obes
18. Westerterp KR, Meijer GA, Kester AD, Wouters L, ten Hoor F.
Fat-free mass as a function of fat mass and habitual activity level. Int
J Sports Med. 1992;13:163–166.
19. Bacchetti P, Gripshover B, Grunfeld C, et al. Fat distribution in men
with HIV infection. J Acquir Immune Defic Syndr. 2005;40:121–131.
20. Bacchetti P, Cofrancesco J, Heymsfield S, et al. Fat distribution in
women with HIV infection. J Acquir Immune Defic Syndr. 2006;42:
21. Grunfeld C, Saag M, Cofrancesco J, et al. Regional adipose tissue
measured by MRI over five years in HIV-infected and control subjects
indicates persistence of HIV-associated lipoatrophy. AIDS. 2010;24:
22. Tien PC, Benson C, Zolopa AR, Sidney S, Osmond D, Grunfeld C.
The study of fat redistribution and metabolic change in HIV infection
(FRAM): methods, design, and sample characteristics. Am J Epide-
23. Friedman GD, Cutter GR, Donahue RP, et al. CARDIA: study design,
recruitment, and some characteristics of the examined subjects. J Clin
24. Hill JO, Sidney S, Lewis CE, Tolan K, Scherzinger AL, Stamm ER.
Racial differences in amounts of visceral adipose tissue in young
adults: the CARDIA (Coronary Artery Risk Development in Young
Adults) study. Am J Clin Nutr. 1999;69:381–387.
25. Choi A, Scherzer R, Bacchetti P, et al. Cystatin C, albuminuria, and
5-year all-cause mortality in HIV-infected persons. Am J Kidney Dis.
26. Gallagher D, Belmonte D, Deurenberg P, et al. Organ-tissue mass
measurement allows modeling of REE and metabolically active tissue
mass. Am J Physiol. 1998;275:E249–E258.
27. Scherzer R, Shen W, Heymsfield SB, et al. Intermuscular adipose tis-
sue and metabolic associations in HIV infection. Obesity (Silver
Spring). 2010; In press.
28. Frisancho AR, Flegel PN. Elbow breadth as a measure of frame size
for US males and females. Am J Clin Nutr. 1983;37:311–314.
29. Searle SR, Speed FM, Milliken GA. Populations marginal means in
the linear model: an alternative to least squares means. Am Stat. 1980;
30. Huber P. The behavior of maximum likelihood estimates under non-
standard conditions. In Proceedings of the Fifth Berkeley Symposium
on Mathematical Statistics and Probability; 1967; Berkeley, CA:
University of California Press. 221–223.
31. White H. A heteroskedasticity-consistent covariance matrix estimator
and a direct test for heteroskedasticity. Econometrica. 1980;48:817–830.
32. Hastie T, Tibshirani R. Generalized Additive Models. New York:
Chapman & Hall; 1990.
33. Schafer JL. Multiple imputation: a primer. Stat Methods Med Res.
34. Robins JM, Finkelstein DM. Correcting for noncompliance and depen-
dent censoring in an AIDS Clinical Trial with inverse probability of cen-
soring weighted (IPCW) log-rank tests. Biometrics. 2000;56:779–788.
35. Mulligan K, Tai VW, Schambelan M. Cross-sectional and longitudinal
evaluation of body composition in men with HIV infection. J Acquir
Immune Defic Syndr Hum Retrovirol. 1997;15:43–48.
36. Kotler DP, Wang J, Pierson RN. Body composition studies in patients
with the acquired immunodeficiency syndrome. Am J Clin Nutr.
37. Grinspoon S, Corcoran C, Miller K, et al. Body composition and en-
docrine function in women with acquired immunodeficiency syn-
drome wasting. J Clin Endocrinol Metab. 1997;82:1332–1337.
38. Yarasheski KE, Zachwieja JJ, Gischler J, Crowley J, Horgan MM,
Powderly WG. Increased plasma gln and Leu Ra and inappropriately
low muscle protein synthesis rate in AIDS wasting. Am J Physiol.
39. Yarasheski KE, Smith SR, Powderly WG. Reducing plasma HIV RNA
improves muscle amino acid metabolism. Am J Physiol Endocrinol
40. Hong-Brown LQ, Pruznak AM, Frost RA, Vary TC, Lang CH. Indina-
vir alters regulators of protein anabolism and catabolism in skeletal
muscle. Am J Physiol Endocrinol Metab. 2005;289:E382–E390.
41. Chantry CJ, Cervia JS, Hughes MD, et al. Predictors of growth and
body composition in HIV-infected children beginning or changing an-
tiretroviral therapy. HIV Med. 2010.
42. Note R, Maisonneuve C, Letteron P, et al. Mitochondrial and meta-
bolic effects of nucleoside reverse transcriptase inhibitors (NRTIs) in
mice receiving one of five single- and three dual-NRTI treatments.
Antimicrob Agents Chemother. 2003;47:3384–3392.
43. Divi RL, Leonard SL, Walker BL, et al. Erythrocebus patas monkey
offspring exposed perinatally to NRTIs sustain skeletal muscle mito-
chondrial compromise at birth and at 1 year of age. Toxicol Sci.
44. Fleischman A, Johnsen S, Systrom DM, et al. Effects of a nucleoside
reverse transcriptase inhibitor, stavudine, on glucose disposal and mi-
tochondrial function in muscle of healthy adults. Am J Physiol Endo-
crinol Metab. 2007;292:E1666–E1673.
45. Newman AB, Kupelian V, Visser M, et al. Strength, but not muscle
mass, is associated with mortality in the health, aging and body compo-
sition study cohort. J Gerontol A Biol Sci Med Sci. 2006;61:72–77.
46. Swallow EB, Reyes D, Hopkinson NS, et al. Quadriceps strength pre-
dicts mortality in patients with moderate to severe chronic obstructive
pulmonary disease. Thorax. 2007;62:115–120.
47. Desquilbet L, Margolick JB, Fried LP, et al. Relationship between a
frailty-related phenotype and progressive deterioration of the immune
system in HIV-infected men. J Acquir Immune Defic Syndr. 2009;50: