Content uploaded by Peter F Collins
Author content
All content in this area was uploaded by Peter F Collins on Oct 31, 2017
Content may be subject to copyright.
REVIEW
Nutritional support and functional capacity in chronic obstructive
pulmonary disease: A systematic review and meta-analysis
PETER F. COLLINS,1,2 MARINOS ELIA1AND REBECCA J. STRATTON1
1Faculty of Medicine, Institute of Human Nutrition, Southampton General Hospital, University of Southampton,
Southampton, UK, and 2Faculty of Health, School of Exercise and Nutrition Sciences, Queensland University of Technology,
Brisbane, Australia
ABSTRACT
Currently, there is confusion about the value of using
nutritional support to treat malnutrition and improve
functional outcomes in chronic obstructive pulmonary
disease (COPD). This systematic review and meta-
analysis of randomized, controlled trials (RCT) aimed
to clarify the effectiveness of nutritional support in
improving functional outcomes in COPD. A systematic
review identified 12 RCT (n=448) in stable COPD
patients investigating the effects of nutritional support
(dietary advice (1 RCT), oral nutritional supplements
(10 RCT), enteral tube feeding (1 RCT)) versus control
on functional outcomes. Meta-analysis of the changes
induced by intervention found that while respiratory
function (forced expiratory volume in 1 s, lung capac-
ity, blood gases) was unresponsive to nutritional
support, both inspiratory and expiratory muscle
strength (maximal inspiratory mouth pressure +3.86
standard error (SE) 1.89 cm H2O, P=0.041; maximal
expiratory mouth pressure +11.85 SE 5.54 cm H2O,
P=0.032) and handgrip strength (+1.35 SE 0.69 kg,
P=0.05) were significantly improved and associated
with weight gains of ⱖ2 kg. Nutritional support pro-
duced significant improvements in quality of life in
some trials, although meta-analysis was not possible. It
also led to improved exercise performance and
enhancement of exercise rehabilitation programmes.
This systematic review and meta-analysis demon-
strates that nutritional support in COPD results in sig-
nificant improvements in a number of clinically
relevant functional outcomes, complementing a previ-
ous review showing improvements in nutritional
intake and weight.
Key words: chronic obstructive pulmonary disease, functional
capacity, meta-analysis, nutritional support.
Abbreviations: BMI, body mass index; COPD, chronic
obstructive pulmonary disease; ETF, enteral tube feeding; FEV1,
forced expiratory volume in 1 s; HGS, handgrip strength; IBW,
ideal body weight; ONS, oral nutritional supplements; PE max,
maximal expiratory mouth pressure; PI max, maximal inspiratory
mouth pressure; QoL, quality of life; RCT, randomized, controlled
trial; SD, standard deviation; SE, standard error.
INTRODUCTION
Chronic obstructive pulmonary disease (COPD) is a
progressive multi-organ systemic disease and a major
cause of morbidity and disability in aging. While
COPD has its primary effects in the lungs, adverse
structural and functional changes also occur in the
tissues of the heart and skeletal muscle, leading to
individuals with COPD experiencing a range of dis-
abilities that impact on their well-being and ability to
perform daily activities. Reduced respiratory function
and a decline in fat-free mass result in reduced exer-
cise tolerance1and peripheral muscle weakness,2,3
both disabling features of COPD, which are associated
with a poorer quality of life (QoL). Fat-free mass
depletion (even if body mass index (BMI) is within the
ideal range)4,5 has recently been found to be a signifi-
cant independent predictor of disability even after
adjustment for disease severity.6
Disease-related malnutrition is a common problem
in individuals with COPD, with between 30% and 60%
of inpatients and 10% and 45% of outpatients said to
be at risk.7Malnourished COPD patients demonstrate
greater gas trapping, lower diffusing capacity and a
reduced exercise performance when compared with
heavier non-malnourished patients with a similar
severity of disease.8However, the exact causal links
between malnutrition and COPD are difficult to
establish. Malnutrition may be the consequence of
greater disease severity. Alternatively, malnutrition
may be responsible for the wasting of the muscles
involved in breathing, exacerbating the progressive
nature of COPD. Similarly, in chronic anorexia
nervosa, the loss of body weight includes substantial
loss of lung tissue, which develops emphysematous-
like changes.9In addition to the uncertainty about the
causal links between malnutrition and COPD, there
Correspondence: Marinos Elia, Institute of Human Nutrition,
Faculty of Medicine, University of Southampton, MP 113, South-
ampton General Hospital, Southampton SO16 6YD, UK. Email:
m.elia@soton.ac.uk
Conflict of Interest Statement: Rebecca Stratton, PhD, RD,
RNutr, is also an employee of Nutricia Ltd.
Received 15 November 2012; invited to revise 9 January 2013;
revised 4 February 2013; accepted 7 February 2013 (Associate
Editor: Paul Thomas).
bs_bs_banner
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
doi: 10.1111/resp.12070
are controversies about how effective nutritional
support is in this patient group. Previous reviews and
meta-analyses10–12 suggested that malnutrition fails to
respond to nutritional treatment in COPD finding no
significant improvements in anthropometric or func-
tional measures. These conclusions have been
challenged by several randomized, controlled trials
(RCT),13–16 and a recently published meta-analysis
concluded that nutritional support was able to signifi-
cantly increase nutritional intake (energy and
protein), which was associated with a significant
improvement in a variety of anthropometric meas-
ures.17 The contrast in conclusions between the
reviews has been largely attributed to methodological
differences in data analysis discussed at length in the
paper.17 In essence, the previous Cochrane Collabora-
tion review12 carried out cross-sectional analysis
between intervention and control groups but failed to
account for baseline variability. The other review
accounted for pre- and post-intervention variability
finding a number of significant within-group im-
provements to be masked by cross-sectional analy-
sis.17 Nevertheless, confusion remains over whether
the recent positive findings translate beyond nutri-
tional intake and body weight, and into functional
improvements. The aim of this current systematic
review is to establish whether nutritional support
results in significant improvements in functional
capacity and QoL in patients with COPD.
METHODS
Search strategy and identification of trials
The review was planned, conducted and reported
according to published guidelines.18–20 The same
methodological approach to that of the previous
review17 was used; however, an updated systematic
search of the literature was carried out in July 2012
(databases accessed up to 4 July 2012) in order to
identify any additional RCT investigating nutritional
support in COPD reporting functional outcomes
(Fig. 1). Potentially relevant studies were identified
by searching electronic databases. The databases
searched included PubMed (accessed 4 July 2012),
Web of Science (accessed 4 July 2012) and OVID
(accessed 4 July 2012). In order to identify the largest
number of trials, a broad search strategy was imple-
mented, although trials were restricted to English lan-
guage citations only. The search terms and mesh
headings used included: chronic obstructive pulmo-
nary disease, COPD, emphysema, weight, depletion,
diet*, nutrition*, supplement*, protein, carbohydrate,
calori*, feed*, malnutrit*, nourish*, sip, nutrition
intervention, nutrition support. These search terms
were also systematically combined in order to identify
trials. In addition to electronic database searching,
manual searching of previous reviews on nutritional
support in COPD as well as references of identified
trials was undertaken.
Studies were initially screened by reading the
abstract, and where a study could not be excluded,
the full article was reviewed. The assessment of trial
eligibility was done by two independent assessors
(P.F.C. and M.E.), with any disagreement discussed
prior to inclusion.
Inclusion and exclusion criteria
Studies were deemed eligible for inclusion in the
review if they conformed to the pre-determined
inclusion criteria. To investigate the overall efficacy of
nutritional support (food strategies (food fortifica-
tion, food snacks, dietary advice), oral nutritional
supplements (ONS) and enteral tube feeding (ETF),
the following inclusion criteria for trials were devised:
(i) randomized trials; (ii) intervention with food strat-
egies, ONS or ETF; (iii) duration of intervention >2
weeks; (iv) control group receiving placebo or no
dietary intervention (e.g. usual care, which could
include advice and encouragement to eat), but other-
wise the same treatment as the intervention group; (v)
stable patients with a diagnosis of COPD (not exacer-
bating); (vi) human studies only; and (vii) English lan-
guage only.
The intervention could provide either a proportion
or all of the daily nutritional requirements for energy,
protein and micronutrients, and where feeds were
used (e.g. ONS), these could be nutritionally complete
or incomplete. Studies using parenteral nutrition and
single nutrient interventions were excluded.
Figure 1 Study selection process.
Nutrition support in COPD 617
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
Data extraction
Functional outcome data sought included respiratory
function (forced expiratory volume in 1 s (FEV1)), res-
piratory muscle strength (maximal inspiratory mouth
pressure (PI max) and maximal expiratory mouth
pressure (PE max)), peripheral muscle strength
(maximum voluntary handgrip strength (HGS)), exer-
cise performance (walking distance), QoL and addi-
tional outcomes including immunological measures.
In trials where mean values were reported without
standard deviations (SD) or standard errors (SE), they
were calculated from reported P-values. In one trial
that assessed HGS,15 data reported in kilograms were
considered to be unrealistic and therefore assumed to
be in pounds. Whenever possible, data from indi-
vidual subjects were used to calculate the summary
values from specific studies.15,21 Graphical data were
also used to establish summary values either when
there were no other data reported or when reported
results were imprecise due to rounding.22
Quality assessment
The quality of included studies was assessed by one
researcher (P.F.C.) and independently verified by
another assessor (R.J.S.) using the most commonly
used scoring system (Jadad scoring system).23 The
Jadad scoring system comprises of three components
addressing whether a study is described as rand-
omized, whether it is double-blind and whether drop-
outs were accounted for. It then scores on the
appropriateness of the randomization and blinding.
The Jadad scoring system does not assess the sample
size of trials.
Synthesis of data and statistical analysis
Following the extraction of data from included trials,
where appropriate and feasible, the results of compa-
rable outcome measures were combined in order to
carry out random effects meta-analyses using Com-
prehensive Meta-analysis (Biostat, Inc., Englewood,
NJ, USA, version 2) (Table 1). The overall treatment
difference was considered statistically significant if
the P-value was <0.05. Analysis explored differences
between groups as well as within group changes.
The effect size was reported as the difference in
mean ⫾SE. Four studies reported values adjusted for
baseline data.16,22,24,25 Meta-analysis was subjected to
sensitivity analysis whenever imputation of missing
data was carried out (this involved imputation of SD
for the change for the control arm of one study).13
Meta-regression was undertaken to examine whether
the differences in functional outcomes between the
two arms of the studies were related to each of the
following moderators: duration of intervention, %
ideal body weight (IBW) and age.
RESULTS
A total of 49 studies were identified as potentially eli-
gible from the literature search;13–16,21,22,24–66 of these,
37 failed to achieve the inclusion criteria (Fig. 1).
Reasons for exclusion included an unsuitable study
design or review in 10 studies,28,37,38,40,46,56,58,61–63 limited
or no nutrition provided in 9 studies,32,36,39,41,43,49,51,64,65 5
non-randomized trials,27,35,45,55,59 5 studies involving
an unsuitable population,26,30,47,54,57 5 studies with no
control or placebo,31,33,35,53,60 and inadequate interven-
tion duration in 2 studies.29,42 Goris et al.,44 which
was included in the previous review, could not be
included as it reported no functional outcomes. The
review included 12 RCT involving 448 individuals with
COPD who were randomized into either a treatment
group (n=232) or a control group (n=216) (Table 2).
Seven studies were performed completely within the
outpatient setting,13,24,25,48,50,52,66 three in the inpatient
setting16,21,22 and two involving periods in both of these
settings.14,15 The study by Schols et al.16,22 is referred to
as two separate trials in order to distinguish between
patients who were considered to be adequately
nourished16 and undernourished22 (Table 2). All the
patients recruited to the earlier trials had a diagnosis
Table 1 Functional outcome measures from randomized, controlled trials included in the systematic review and
meta-analyses
Outcome measure
Systematic review Meta-analysis
No. studies
No. participants
treatment/control No. studies
No. participants
treatment/control
FEV110 102/105†2 43/40
PI max 8 153/124 5 91/86
PE max 6 81/59 4 58/48
HGS 5 87/90 4 77/79
Walk and shuttle tests 7 150/149 0 —
QoL and breathlessness 6 85/90 0 —
Immunological 3 24/20 0 —
ADL 1 23/18 0 —
†Numbers refer to only eight studies. An additional two studies involving 217 patients reported changes in FEV1without specifying
the number that completed the tests in each group.16,22 Numbers reported are for those subjects that completed the intervention phase.
ADL, activities of daily living; FEV1, forced expiratory volume in 1 s; HGS, handgrip strength; PE max, maximum expiratory pressure;
PI max, maximum inspiratory pressure; QoL, quality of life.
PF Collins et al.618
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
Table 2 Summary of the randomized, controlled trials included in the systematic review according to intervention
Study
Sample size
(treatment/
control)†
Characteristics/setting
(intervention vs
control)
Nutritional intervention
(type/prescribed
amount/duration) Control group Outcome measures
Study quality
(Jadad score)‡
ONS
DeLetter66
(thesis)
18/17 Malnourished
82.8% IBW
Outpatients
ONS (Pulmocare, 1.5 kcal/mL) 1 can/day
ONS target: +355 kcal/day and 15 g
protein/day
9 weeks
Usual diet FEV1, 6MWT 11000 (2)
Efthimiou
et al.13
7/7 Malnourished
79.5% versus 81.3% IBW
Outpatients
59.9 versus 64.1 years
ONS (Build Up, 1.13 kcal/mL)
ONS target: +640–1280 kcal/days and
36–72 g protein/day, encouragement
to eat provided to both groups
12 weeks
Usual diet (with
encouragement)
FEV1, PI max, PE max,
sternomastoid
strength, HGS, 6MWT,
breathlessness scale,
general well-being
10000 (1)
Knowles
et al.48
13/12 Nourished and malnourished
61–108% IBW
Outpatients
68 versus 70 years
ONS (Sustacal, 1 kcal/mL, 0.043 g
protein/kcal)
ONS target: To increase total EI by 50%
Weekly encouragement: 8 weeks
Usual diet FEV1, PI max, PE max,
lymphocyte count,
serum transferring
11000 (2)
Lewis
et al.50
10/11 Malnourished
86.3% versus 84.6 % IBW
Outpatients
65.1 versus 59.3 years
ONS (Isocal HCN, 2 kcal/mL)
ONS target: 500–1000 kcal/day and
19–38 g protein/day, encouragement
8 weeks
Usual diet FEV1, PI max, PE max,
HGS
10000 (1)
Otte
et al.52
13/15 Malnourished
77% versus 73% IBW
Outpatients
56.5 years
ONS (Novo, 1 kcal/mL)
ONS target: +400 kcal/day and20 g
protein/day, encouragement
13 weeks
Placebo (blinded)
(encouragement)
FEV1, 12MWT, well-being 10111 (4)
Fuenzalida
et al.14
5/4 Malnourished inpatients and
outpatients
78.5% IBW
62.4 years
ONS (Sustacal HC, 1 kcal/mL)
ONS target: Up to 1080 kcal/day and up
to 46 g protein/day
3 weeks inpatient +3 weeks
outpatient (6 weeks total)
Usual diet FEV1, Lymphocyte count,
T-helper/suppressor
cells
10000 (1)
Rogers
et al.15
15/12 Malnourished
78% versus 79% IBW
64 years
Outpatients (intervention group
admitted for first 4 weeks)
ONS (various, self-selected) tailored to
individual dietary habits and dietary
advice
ONS target: Intakes >1.7¥REE and
minimum 1.5 g protein/kg per day
15 weeks
Usual diet PI max, PE max, HGS,
12MWT,
breathlessness rating,
QoL
10000 (1)
Schols
et al.16
33/38 Nourished
102.4% IBW
inpatient
PR programme (not hospital)
mean age unclear
ONS (Mixture of Nutridrink, Protifar,
Fantomalt,Oil; seven mixtures of
different flavours; 2.1 kcal/mL)
ONS target: +420 kcal/day and 15 g
protein/day, encouragement to eat
regular meals
8 weeks
Usual diet (and
encouragement with
oral diet)
FEV1, PI max, 12MWT 10001 (2)
Nutrition support in COPD 619
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
Table 2 Continued
Study
Sample size
(treatment/
control)†
Characteristics/setting
(intervention vs
control)
Nutritional intervention
(type/prescribed
amount/duration) Control group Outcome measures
Study quality
(Jadad score)‡
Schols
et al.22
39/25 Malnourished
84.1% IBW
Inpatient
PR programme (not hospital)
Mean age unclear
ONS (Mixture of Nutridrink, Protifar,
Fantomalt, Oil; seven mixtures of
different flavours; 2.1 kcal/mL)
ONS target: +420 kcal/day and 15 g
protein/day, encouragement to eat
regular meals
8 weeks
Usual diet and
encouragement with
meals
FEV1, PI max, 12MWT 10001 (2)
Steiner
et al.24
42/43 Nourished/malnourished
~105% IBW
(23.9 vs 23.5 kg/m2)
Outpatients
PR programme
66 versus 68 years
ONS (Respifor, 1.5 kcal/mL)
ONS target: +570 kcal/day and 28 g
protein/day
7 weeks
Placebo (blinded) HGS, ISWT, ESWT, QoL 10111 (4)
ETF
Whittaker
et al.21
6/4 Malnourished
76% versus 82% IBW
Inpatients
71 versus 64 years
Nocturnal ETF (Isocal)
ETF target; feed delivered: at least
1000 kcal/day or 1.7¥REE whichever
greater and 34 g protein
(nasoduodenal/jejunal tube feeding)
16 days
Placebo
ETF (equivalent
volume providing
<100 kcal/night)
FEV1, PI max, PE max,
adductor pollicis
muscle function,
lymphocyte count,
transferrin
11110 (4)
DA, dietary
leaflet
plus
milk
powder
Weekes
et al.25
31/28 Malnourished
~88%IBW
(~19.8 kg/m2)
outpatients
69 years
Tailored DA +leaflet of
information +milk powder
DA target: 600 kcal/day (no specific
protein target)
6 months
Leaflet of information FEV1, PI max, PE max,
HGS, QoL, activities of
daily living
10001 (2)
†Sample size at baseline (this occasionally differed from the sample size associated with results over time).
‡The number in parenthesis represents the overall score. The five individual scores represent scores for description and appropriateness of randomization/blinding as well as any
description of withdrawals.
6MWT, 6-min walk test; 12MWT, 12-min walk test; DA, dietary advice (education); EI, energy intake; ESWT, endurance shuttle walk test; ETF, enteral tube feeding; FEV1, forced expiratory
volume in 1 s; HGS, handgrip strength; IBW, ideal body weight; ISWT, incremental shuttle walk test; ONS, oral nutritional supplements; PE max, maximum expiratory pressure; PI max,
maximum inspiratory pressure; PR programme, pulmonary rehabilitation programme; QoL, quality of life; REE, resting energy intake.
PF Collins et al.620
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
of COPD (FEV1/forced vital capacity <0.70), which was
severe67 (mean FEV1<50% predicted (stage III) in all
trials). The patients were in a stable condition free
from exacerbation. Studies involving patients with
acute exacerbations were excluded.
Out of the 12 trials included in the analyses, 10
(n=379; intervention 195 vs control 184) provided
nutritional support by ONS,13–16,22,24,44,48,50,52,66 mostly
ready-made, liquid supplements, some of which were
specifically formulated for use in patients with COPD.
One trial used nocturnal ETF (n6vs4),
21 and another
used tailored dietary advice delivered by a dietitian
and the provision of whole milk powder (n31 vs 28).25
No trials were found involving interventions of food
snacks or food fortification alone. The intervention
period ranged from 16 days21 to 6 months,25 with the
amount of nutritional support prescribed ranging
from 35566 to 1080 kcal/day.13
Eight studies (n8)13–15,21,25,50,52,66 principally in-
volved malnourished (‘depleted’) individuals (BMI
<20 kg/m2or%IBW<90%). Three trials included both
adequately nourished and undernourished patients
who participated in a rehabilitation exercise pro-
gramme,16,22,24 but subgroup analysis according to
nutritional status was undertaken to examine some of
the outcomes. The remaining study included both
undernourished and nourished subjects (Table 2) but
with a predominance of underweight individuals
(over half with <85% IBW48).
The most commonly reported outcome was
pulmonary function (FEV1) reported in 10 tri-
als,13,14,16,21,22,25,48,50,52,66 followed by respiratory muscle
strength (PI max and PE max). Breathlessness was
reported in five studies,13–15,25,52 three studies reported
QoL obtained using validated tools,15,24,25 and two
trials reported subjective feelings of well-being,13,52
which can also be regarded as measures of
QoL.
Quality of studies
The review identified three studies assessed to be of
high quality (ⱖ4)21,24,52 and nine of lesser quality (ⱕ2)
using the Jadad scoring system23 (Table 2).
Pulmonary function and respiratory
muscle strength
Respiratory function (FEV1) was assessed in 10 stud-
ies,13,14,16,21,22,25,48,50,52,66 9 of which13,14,16,21,22,25,48,50,52,66
provided separate information in intervention and
control groups. However, the results were presented
in different ways: two reported no significant differ-
ences in the change in FEV1over time,25,52 seven
reported no significant change in either group over
time,13,16,21,22,25,48,52 and two reported the mean values of
FEV1at the start and end of the study period, but
because they were virtually identical50 or very close to
each other66 within the control and the intervention
groups, it can be deduced that there were no signifi-
cant changes over time in either group and no signifi-
cant differences between groups. Indeed, there was
no evidence from any of the studies that the changes
in FEV1or changes in other measures of respiratory
function, such as forced vital capacity,13,21,25,48,50,52
FEV1/forced vital capacity,21,50 total lung capacity13,21,48
and blood gases48,50,52 differed between intervention
and control groups. Two studies reporting measured
FEV125 and percentage predicted FEV152 were meta-
analysed using standardized differences. Nutritional
support was not associated with any improvement in
FEV1(-0.213 SE 0.22 L, P=0.335).
PI max was reported in eight studies13,15,16,21,22,25,48,50
of ONS (n6), ETF (n1) and dietary advice (n1). Five
of these studies13,15,16,21,25 were amenable to meta-
analysis, four of which favoured nutritional support
(Fig. 2). The overall summary measure obtained using
random effects meta-analysis was significant in
Figure 2 Random effects meta-analysis
of five studies measuring changes in
maximal inspiratory mouth pressure (PI
max) (cm H2O) in nutritional intervention
(n=91) and control groups (n=86). The
forest plot shows the difference in the
mean changes between groups. CI, con-
fidence interval.
Nutrition support in COPD 621
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
favour of nutritional support (+4.04 SE 1.86 cm H2O,
P=0.030). The meta-analysis was undertaken assum-
ing that the SD of the change in the control group of
one of the studies13 was the same as that of the ONS
group. The latter was established from the combina-
tion of the reported mean change in PI max and a
P-value of 0.05 (but because the reported P-value was
<0.05, the calculated value is larger than the true
value). When a sensitivity analysis was carried out
assuming the SD of the change ranged from 75% to
125% of that in the ONS group, there was little change
in the overall point estimate (+4.11 SE 1.83 cm H2O,
P=0.024 and +3.98, SE 1.90 cm H2O, P=0.049,
respectively). The associated weight change in the
same five studies also favoured nutritional support
(+2.17 SE 0.44 kg; P<0.001). Meta-regression found
no significant relationship between PI max (cm H20)
and each of the following variables: duration of inter-
vention (weeks) (slope =-0.294, SE 0.250, P=0,239),
% IBW (slope =-0.124, SE 0.187, P=0.058) and age
(years) (slope =-0.867, SE 0.779, P=0.266); excluding
the study of Schols et al.,16,22 which did not report the
mean age of the patients who had the PI max tests.
Of the three studies that could not be included in
the meta-analysis, one found a significant increase
in PI max over time in the ONS group and not in
the control group,48 one reported PI max to be
unchanged,50 and the final study did not report
the relevant data needed for inclusion in the
meta-analysis.22
PE max was reported in six studies, four involving
ONS,13,15,21,25,48,50 one ETF21 and one using dietary
advice,25 but meta-analysis was only possible in four
of them13,15,21,25 (Fig. 3). This meta-analysis found that
nutritional support significantly improved PE max in
favour of the intervention group (+13.06 SE 5.81 cm
H2O, P=0.025), with all four studies favouring the
intervention group and two significant in their own
right (one involving ETF21 and the other involving
ONS15) (Fig. 3). The meta-analysis of PE max was
undertaken assuming that the SD of the change in the
control group of one of the studies13 was the same as
that in the ONS group.When a sensitivity analysis was
carried out assuming that the SD of the change
ranged from 75% to 125% of that of the ONS group,
there was virtually no change in the point estimate
obtained by the random effects meta-analysis
(+13.02, SE 5.83 cm H2O, P=0.026; and +13.12 SE
5.78 cm H2O, P=0.024, respectively). The associated
weight change in the same four studies also signifi-
cantly favoured the nutritional support group (+3.10
SE 0.67 kg; P<0.001). Meta-regression found no sig-
nificant relationship between PE max (cm H20) and
each of the following variables: duration of interven-
tion (weeks) (slope =-0.071, SE 0.430, P=0.869), %
IBW (slope =-0.321, SE 0.809, P=0.691) and age
(years) (slope =1.494, SE 1.224, P=0.266).
Of the two studies that could not be included in the
meta-analysis, one reported no significant difference
in measurements between groups48 and the other no
significant change within groups.50
To assess respiratory accessory muscle strength, a
further study measured sternomastoid strength and
fatigability, and found that ONS resulted in signifi-
cantly increased strength (P<0.05) and reduced fati-
gability after 3 months of supplementation, while
non-significant changes in the opposite direction
occurred in the control group. The differences
between groups returned towards baseline after
cessation of treatment.13
Other functional measures
Maximum voluntary peripheral
(non-respiratory) muscle strength
Five studies assessed peripheral muscle strength
using HGS,13,15,24,25,50 with four of the five providing
nutritional support using ONS. Four studies13,15,24,25
were amenable to meta-analysis and all favoured
intervention, two significant in their own right13,15
(Fig. 4). The mean changes were +1.41 SE 0.66 kg,
P=0.032 (range 0.3–5.2 kg (1.3–18.5%) above baseline
in favour of the intervention group). In undertaking
Figure 3 Random effects meta-analysis
of four studies measuring changes in
maximal expiratory mouth pressure (PE
max) (cm H2O) in nutritional intervention
(n=58) and control groups (n=48). The
forest plot shows the difference in the
mean changes between groups. CI, con-
fidence interval.
PF Collins et al.622
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
this meta-analysis, two assumptions were made: the
SD of the change for the control group in the study of
Efthimiou et al.,13 was the same as that for the ONS
group, and the unrealistically high grip strength
values reported for both intervention and control
groups in the study of Rogers et al.15 were in pounds
rather than in kilograms. To address the latter uncer-
tainty about the units of measurement, the meta-
analysis was repeated using standardized differences
(overall point estimate 0.56, SE 0.22; P=0.009,
Figure 4). To address the former uncertainty, a sensi-
tivity analysis was undertaken by altering the SD of
the change by ⫾25% (75–125%), which produced little
change in the overall point estimate and the associ-
ated statistical significance of the differences between
groups (using a value of 75% for the SD of the change,
the overall effect size in the meta-analysis was 0.59, SE
0.23 (P=0.010); and with 125% for the SD of the
change, the effect size was 0.52, SE 0.20 (P=0.008)).
Even when the sensitivity analysis involved an altera-
tion of the SD of the change by as much as ⫾50%
(50–150%), there was little overall impact on the effect
sizes and P-values (effects sizes ranging from 0.48 to
0.58 and P-values from 0.021 to 0.018, respectively).
The associated change in body weight in the same
four studies significantly favoured the intervention
group (+2.06 SE 0.65 kg; P=0.001). Meta-regression
found no significant relationship between HGS (kg)
and each of the following variables: duration of inter-
vention (weeks) (slope =-0.033, SE 0.041, P=0.428)
and % IBW (slope =-0.322, SE 0.035, P=0.364).
Steiner et al.,24 reported that quadriceps muscle
strength increased more in the supplemented than
control group (+17.4 kg or ~5% vs +3.6 kg or ~1%
increase, P=0.068) after adjustment for baseline
values. When these results replaced those of HGS in
the random effects meta-analysis on muscle strength
and the amalgamated results analysed using stand-
ardized differences, the point estimate remained
significant (effect size 0.56, SE 0.22, P=0.010).
A small study of ETF reported no significant differ-
ences in the changes between intervention and
control groups in electrical stimulation tests involving
the adductor pollicis muscle.21
Walking distance and endurance
during walking
Seven studies examined the influence of nutritional
support on improving exercise tolerance.13,15,16,22,24,52,66
Four studies favoured the intervention group,13,15,24,66
one favoured the control group,52 and the remaining
two studies16,22 did not provide the necessary informa-
tion to assess which group was favoured. Meta-
analysis was not performed due to the use of different
methodologies, types of tests, and ways of reporting
results (e.g. some reporting median values24 and
others mean values) and lack of measures of variation
and/or P-values for some of the within-group
changes.13
Using the 6-min walk test, Efthimiou et al.13
reported significant improvements in the ONS group
(53 m (~+12.8%); P<0.05) but not in the control
group (1 m (~+1.4%) non-significant), and DeLetter
did the same66 [+35.4 m (~+11.6%) vs -1.2 m
(~-0.4%)]. Using the 12-min walk test, Rogers et al.15
found that the distance walked increased significantly
more in the ONS group (34 m (~7%) at 4 weeks and
143 m (~28%) at 4 months) compared with the control
group (a deterioration of 42 m (-8%) at 4 weeks and
0.3 m (-0.1%) at 4 months) (P=0.03 for the difference
in the mean change between the two groups). Otte
et al.52 reported no significant changes in the 12-min
walking distance in either the ONS or control groups
(-81 m (~-9.0%) vs +50 m (~6.3%), respectively).
Schols et al.16,22 reported an improvement in the
12-min walking distance in subgroups of depleted
(173 m, 29%) and non-depleted (147 m, 24%) patients
undergoing pulmonary rehabilitation, with no signifi-
cant differences between the intervention and control
groups. Using the shuttle walk tests in subjects under-
going pulmonary rehabilitation, Steiner et al.24 found
that performance improved to a greater extent in the
intervention (ONS) than control group with respect to
Figure 4 Random effects meta-analysis
of four studies measuring changes in
peripheral muscle strength (standardized
differences) in nutritional intervention
(n=77) and control groups (n=79). The
forest plot shows the difference in the
standardized mean changes between
groups. CI, confidence interval.
Nutrition support in COPD 623
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
endurance walk tests (mean increase in distance
walked, 60.0 m vs 42.6 m; ~29% vs 19%; P=0.182) and
the incremental walk tests (median increase in dura-
tion, 328 s (~¥2 baseline value vs 191 s, ~¥0.9 base-
line value; P=0.172), but the differences were not
significant.
QoL and subjective measures of breathlessness
Five studies examined the effect of nutritional
support on QoL, but because the results were
obtained using different tools and reported in differ-
ent ways, they were not subjected to meta-
analysis.13,15,24,25,52 One study reported a significant
improvement in total score in favour of the interven-
tion group: the St George’s Respiratory questionnaire
(with the components also favouring the intervention
group: Activity (P=0.06), Impacts (P=0.004), symp-
toms (P=0.73) using the intention-to-treat analysis),
as well as the short-form 36, which measured health
change. The overall scores represented an improve-
ment of 18% and 55% improvement, respectively,
according to the intention-to-treat analysis).25 These
changes were mirrored by significant differences in
breathlessness. Another study reported a significant
improvement in general well-being of the nutrition
intervention (ONS) group (~27%) and not in the
control group (~6%),13 which was paralleled by a sig-
nificant improvement in breathlessness in the inter-
vention and not in the control group. A third study of
ONS reported that a greater proportion of subjects felt
that their well-being had improved as a result of the
nutrition intervention compared with control (23% vs
13%) and a smaller proportion felt that it had deterio-
rated (15% vs 33%), but given the small sample size
(13 vs 15), the differences were not statistically signifi-
cant.52 This same study reported a tendency for
breathlessness to improve in favour of the interven-
tion group (P=0.20). One of the two remaining
studies briefly reported no significant differences in
health-related QoL assessed at enrolment or at 4
months using the Sickness Impact Scores15 and no
significant differences in breathlessness scores, which
contributed to the overall QoL. The final study involv-
ing subjects undergoing exercise rehabilitation found
significant improvements in health-related QoL and
in breathlessness (both assessed using the self-
reported Chronic Respiratory Questionnaire) in both
the control and intervention groups,24 with no signifi-
cant differences between them.
Other outcomes
The only study that examined activities of daily living
in malnourished patients with COPD reported that
the nutrition intervention group found it significantly
easier to perform everyday activities compared with
the control group (P=0.009 in the per-protocol analy-
sis and P=0.06 in the intention-to-treat analysis, the
difference in the changes being about 18% and 11% of
the baseline values, respectively25).
Four studies reported changes in immunological
tests.14,21,48,52 One of these14 found significant improve-
ments in delayed cutaneous hypersensitivity and
total circulating lymphocyte count, without associ-
ated changes in circulating immunoglobulin concen-
trations, during nutritional repletion of malnourished
patients with COPD. Two trials, one involving ETF21
and one ONS48 briefly reported that lymphocyte
counts remained unchanged over the study period.
Finally, a study52 involving ONS reported no signifi-
cant changes in T-helper/T-suppressor ratio and
mitogen reaction of T lymphocytes to phytohemag-
glutinin. None of these three studies reported sepa-
rately the changes that occurred in each group.
Furthermore, none of these three studies or any of the
other studies included in this systematic review
measured cytokines or acute phase proteins.
DISCUSSION
This review found that nutritional support in COPD
produces significant improvements in several func-
tional outcomes including respiratory and limb
muscle strength. These findings demonstrate the
positive effects of nutritional support on respiratory
muscle tests and other functional outcomes, building
on the conclusions of a recent review that nutritional
support significantly improves energy and protein
intakes with resulting increase in body weight.17 This
earlier review did not consider functional outcomes
other than change in HGS, which was reported as a
percentage change rather than in kilogram as in this
review. However, it did consider methodologi-
cal issues, including those involved in an earlier
Cochrane review. The findings of the present review
also strengthen the argument that a causal pathway
exists linking increased nutritional intake provided
primarily by standard ONS to increased body weight
and function. The current findings are in contrast
with the previous Cochrane review including the
same trials that reported no effect of nutritional
support in COPD12 probably because it considered
only cross-sectional differences between groups at
the end of the intervention period and not within-
group changes induced by the interventions (see
Collins et al.17 for a discussion on methodological
issues). However, an updated Cochrane review, which
appeared very recently following submission of the
present review for peer review, changed its conclu-
sions. A comparison of this review with the updated
Cochrane review is provided after the main findings of
the present review are discussed.
The current review with a series of meta-analyses
found that for each of the tests used to document
improved respiratory muscle function (PI max and PE
max) and non-respiratory (handgrip/quadriceps)
muscle strength, there was a highly significant
increase in body weight of more than 2 kg (2.1–3.1 kg)
in favour of the nutrition intervention group. Schols
et al.68 found that both an improved inspiratory
mouth pressure (PI max) and a weight gain of >2kg
were associated with significantly improved survival
in keeping with a previous review reporting that sig-
nificant functional improvements were seen in mal-
nourished patients receiving ONS when weight gain
was >2 kg.7It appears that this level of weight gain
should be a therapeutic target in malnourished COPD
PF Collins et al.624
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
patients, especially as recently reviewed evidence
highlights that this level of weight gain is achievable
in malnourished COPD patients.17 The current review
confirms that weight gain of this magnitude is associ-
ated with functional improvements in this patient
group.
In the clinical setting, increasing importance is
being placed on the assessment of functional out-
comes. HGS is not only a reliable marker of peripheral
muscle strength but it also predicts clinical out-
comes69 such as mortality, morbidity, postoperative
complications and increased length of hospital stay.
In the elderly, a loss of grip strength often means a loss
of independence. Although muscle strength is closely
related to mid-arm muscle area,70 whole body protein
content,71 and even body weight and BMI,72 a variety
of studies suggest that changes in muscle function
can occur independently of muscle mass.7It has
recently been suggested that muscle strength
responds faster to nutritional depletion and repletion
than anthropometric measures such as BMI and
fat-free mass69 probably as a result of increased avail-
ability of energy, electrolytes and micronutrients in
muscle. Therefore, the improvement in muscle
strength induced by nutritional intervention in mal-
nourished COPD patients is likely to be due to a com-
bination of increased force generated by the available
muscle and increased muscle mass, which is consist-
ent with the increase in mid-arm muscle circumfer-
ence (or area) reported in RCT of COPD17 and other
conditions.73 Mid-arm muscle area has been found to
be a better predictor of mortality than BMI in patients
with COPD.74 Therefore, nutritional support leading
to weight gain (>2 kg) and increased mid-arm muscle
circumference could confer survival benefits as sug-
gested by previous studies.68,75
Exercise tests in COPD have also been found to
predict outcomes,76 such as mortality and postopera-
tive complications.77,78 This review examined the
effect of nutritional support in COPD patients under-
taking different types of walking and shuttle tests per-
formed on a flat surface, but the reviewed studies
were not amenable to meta-analysis. However, four of
the five studies favoured the nutritional support
group, and the only studies reporting significant
improvements in performance also favoured those
receiving nutritional support. While these tests have
limitations (e.g. some patients still have difficulties
walking faster on flat surfaces as their condition
improves, but they can walk up a steeper slope), they
at least assess important aspects of the patient’s
ability to function in ways that are relevant to every-
day life.79
The evidence based on the effect of nutritional
support on immunological function is very limited
not least because none of the three studies14,21,48 that
assessed restricted aspects of immune function
reported the results separately for the intervention
and control groups. In addition, the total absence
from these studies of cytokine measurements and
acute phase proteins as markers of the inflammatory
response highlights the need to examine immune/
inflammatory-nutrition interactions. This is because
the immune system not only helps prevent and aid
recovery from respiratory infections but also because
it is linked to the processes involved in nutritional
depletion and repletion of body tissues and their
responsiveness to nutritional support.34 Whether
exercise has a pro- or anti-inflammatory role in COPD
is unclear;80,81 however, a recent trial82 involving a
combination of low-intensity exercise education ses-
sions and an ONS with immunomodulatory proper-
ties (immunonutrition) in a cohort of malnourished
(mean BMI 18.0 kg/m2) patients with moderate COPD
produced some very promising results that included
improvements in weight, peripheral and respiratory
muscle strength, exercise capacity, QoL, and a reduc-
tion in inflammation (measured by interleukin-6,
interleukin-8, tumour necrosis factor-a, high sensi-
tive C-reactive protein levels). Further work is
required to examine whether these improvements are
due to the exercise intervention, the immunonutri-
tion (or other components of the ONS) or a combina-
tion of these. Further work is also needed to examine
the extent to which the changes in outcome could be
reproduced by using a standard ONS.
An outcome that was found to be unresponsive to
nutritional support in the current review was lung
function (assessed by tests such as FEV1, forced vital
capacity and blood gases), but this is likely to reflect
the irreversible nature of lung pathology in COPD. It
may seem surprising that the lack of an effect of nutri-
tional support on objective tests of lung function were
sometimes associated with significant improvements
in subjective measures of breathlessness. However,
because malnutrition has effects on the central
nervous system, including modulation of the sensitiv-
ity of the respiratory centre to hypoxic stimulation,81
it is plausible that nutritional support influences
the sensation of breathlessness through centrally
mediated mechanisms. Interestingly, cross-sectional
studies of men with COPD have reported that breath-
lessness is inversely related to BMI independently of
respiratory function tests (diffusing capacity to
carbon monoxide, partial arterial oxygen concentra-
tion, P0.1/PI max).83 Breathlessness influences QoL,
which probably explains the striking concordance
between them, both within and between groups of
the reviewed RCT.
The extent to which changes in functional outcome
measures reflect clinically relevant improvements in
patient well-being can be difficult to establish. For
example, a small change in muscle strength (which
may be as little as a few percent, as in some of the
reviewed studies) may go totally unnoticed in strong
well-nourished subjects, but in malnourished
patients who are close to the threshold of disability,84
they may be easily noticed and make the difference
between being able to get up and not get up from a
bed or a chair, and between being independent and
dependent on others. Nevertheless, attempts have
been made to establish the minimum clinically rel-
evant changes associated with some of these tests. For
example, it has been conservatively estimated that
the minimum clinically important difference in 6-min
walking distance is 54–80 m85, which exceeds that
found in the only two nutrition intervention studies
that employed the 6-min walk test (an improvement
Nutrition support in COPD 625
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
in favour of ONS by a mean of 4713 and 37 m66).
However, much larger changes have been found with
the 12-min walk test, for example, an improvement of
143 m has been attributed to ONS in the study of
Rogers et al..15 The minimum clinically important
improvement in incremental shuttle walk test has
been estimated to be 47.5 m. The minimum benefit
distinguishable by patients relates to 78.7 m.86 Of the
studies considered in the present systematic review,
the only one that used the incremental shuttle walk
test to examine the effects of ONS during pulmonary
rehabilitation24 found improvements in walking dis-
tance in favour of the nutrition intervention group
that were less than the suggested thresholds.
However, the patients were studied during pulmonary
rehabilitation and the effect of ONS in combination
with the other treatments showed a statistically sig-
nificant overall improvement of 60 m. Because both
non-malnourished (87%) and malnourished patients
(13%) received ONS if they were randomized to the
nutritional support arm of the study, it is possible that
those with malnutrition responded differently from
those without malnutrition, but such information
was not reported. The benefits of exercise rehabilita-
tion are well established;87 however, as alluded to by
Steiner and colleagues,24 it can produce a negative
energy balance that might require reversal by supple-
mentation before an improvement in training out-
comes can be demonstrated. A recent RCT of patients
with chronic respiratory failure, the majority of whom
had COPD, participating in an exercise rehabilitation
programme and classified as malnourished (BMI
21.5, SD 3.8 kg/m2and fat-free mass deplete) found
that nutritional support (ONS 3¥per day), education
and oral testosterone undecanoate led to significant
improvements in body weight, fat-free mass, strength
and function above control.62 At present, it is unclear
whether all malnourished COPD patients undertak-
ing exercise training should receive additional nutri-
tional support or indeed whether training should
commence in those who are malnourished without
nutritional support.61 It would appear pertinent to
recommend that all COPD patients at risk of malnu-
trition should receive some form of nutritional
support during rehabilitation and recommendations
are required.
The present analysis also examined the effect of
potential explanatory variables, such as duration of
intervention with nutritional support, % IBW and age
of the participants, on functional outcomes (PI max,
PE max and HGS), but generally, they were not found
to be significantly related to the outcomes. This is not
too surprising given that the meta-regressions
involved a small numbers of studies that differed in
design and prescribed amounts of nutritional support
(see Table 2), and also involved examination of each
variable individually. In addition, there was only small
variation between the mean age of the populations
involved with different studies (62–69 years) and in
some cases involved significant effects on outcomes
followed short periods of supplementation (e.g. sig-
nificant improvement in PE max reported in one
study after 16 days of supplementation).21 Further
insights might emerge if individual patient data
(instead of mean study data) were analysed together,
but unfortunately, such data are not available.
To understand the significance, strengths and limi-
tations of this review and the way it differs from the
updated Cochrane review,88 it is necessary to consider
certain methodological issues. Although the conclu-
sions of both reviews appear to be similar (and both at
variance with those of the earlier Cochrane review,12
the two should not be confused because apart
from not addressing the same issues, they have used
different methodology to meta-analyse different
studies, which were selected according to different
criteria.
The present review excluded three studies,61,63,82
which were also absent from the previous Cochrane
review. However, in the updated Cochrane review,
these three studies contributed to the assessment of
almost all of the functional outcomes, dominating
some of the analyses such as the overall health-
related QoL (accounting for three out of the four
studies) and their domains (two out of three studies in
each domain61,82) and the 6-min walk test (three out of
five studies). The studies totally dominated the meta-
analyses of quadriceps strength (only two studies61,63),
yet in the present review, data on quadriceps strength
from another paper were able to be used.24 Two of the
three papers were excluded from the present review
because they both incorporated an exercise pro-
gramme in the nutritional support arm of the study
and not in the control arm,61,63 and one of them63 also
included additional interventions in the nutritional
support arm and not the control arm. (The depleted
patients who received nutritional support in this
study accounted for minority of the population,
which was predominantly overweight with mild
COPD.)These study designs make it difficult to isolate
the effects of nutritional support. The multiple
reported functional outcomes in favour of nutritional
support arm may have been due to the non-
nutritional interventions or a combination of nutri-
tional and non-nutritional interventions. The third
study82 was not included in the present review
because it became available after the literature search
was carried out. It involved a nutritional supplement
with immunomodulatory properties, which can be
attributed to an anti-inflammatory whey peptide,
pharmacological doses of the anti-oxidant vitamins
C, E and A, and fish oils. Arguably, this study should be
treated separately from other studies, none of which
involved an immunonutrition feed.
In the present review, the meta-analyses examined
whether the changes induced by the interventions
differed between the two arms of studies, whereas in
both the previous and updated Cochrane reviews,
most of the meta-analyses, including those involving
PI max and PE max, involved only the values at the
end of the intervention period. It is considered pref-
erable to generally use the ‘change’ method than the
‘end value’ method to assess the impact of interven-
tions, especially when there is a relatively large base-
line imbalance (which could fortuitously affect the
two arms of individual studies in opposite directions
(e.g. for variables such as PI max and PE max)).
Another difference between the two reviews concerns
PF Collins et al.626
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
the uncertainty associated with imputation of
missing data (missing SE or SD). In the present review,
imputation involved only 1 of the 12 studies,13
affected only the control group (for PI max, PE max
and HGS), and was always accompanied by sensitivity
(uncertainty) analysis, which assessed the potential
errors associated with imputation. The uncertainty
associated with the some of the analyses in the
updated Cochrane review appears is less clear
because imputation involved 5 out of 14 studies (and
more than half of the studies in some meta-analyses,
such as those involving changes in anthropometry
including weight, and 6-min walk test), often both
arms of some studies or the difference between them
and in the absence of sensitivity analyses.
Although the type of functional outcomes exam-
ined by the two reviews was similar, the present
review systematically considered HGS and immuno-
logical function, which was not the case with either of
the Cochrane reviews.12,88 In addition, the present
review provided new data using meta-regression
(involving duration of intervention, % IBW, age and as
moderators) and considered the minimally important
clinical differences and related the findings associ-
ated with one type of outcome variable to that of
another another (e.g. the extent of weight gain asso-
ciated with improvements in functional outcomes).
In contrast, the Cochrane review, following the formal
format of Cochrane Collaboration to produce a docu-
ment of almost 100 pages long, included more
detailed information about individual studies, listed
the excluded studies and undertook some subgroup
meta-analyses, such as those involving the compo-
nents of QoL (made possible by inclusion of the three
new studies that did not feature in this review).
However, the present review included a semiquanti-
tative and narrative description of QoL data and well-
being13,15,24,52 that were not synopsized by the
Cochrane review.
Because a cure is impossible for COPD patients, a
major goal in the management of the disease is the
improvement or maintenance of body function and
QoL. This systematic review describes the types and
magnitude of functional benefits that are likely to
arise through nutritional support. It suggests that at
least some of the adverse functional consequences of
severe COPD are reversible by nutritional support.
The review also suggests that while several of the
studies were judged to be of high quality, many were
of lower quality, and therefore, the evidence base for
the role of nutritional support in COPD needs to be
strengthened with sufficiently powered RCT.
REFERENCES
1 Gosselink AT, Smit AJ, Crijns HJ et al. Alteration of peripheral
vasodilatory reserve capacity after cardioversion of atrial fibril-
lation. Eur. Heart J. 1996; 17: 926–34.
2 Bernard S, LeBlanc P, Whittom F et al. Peripheral muscle weak-
ness in patients with chronic obstructive pulmonary disease.
Am. J. Respir. Crit. Care Med. 1998; 158: 629–34.
3 Seymour JM, Spruit MA, Hopkinson NS et al. The prevalence of
quadriceps weakness in COPD and the relationship with disease
severity. Eur. Respir. J. 2010; 36: 81–8.
4 Schols AM, Soeters PB, Dingemans AM et al. Prevalence and
characteristics of nutritional depletion in patients with stable
COPD eligible for pulmonary rehabilitation. Am. Rev. Respir. Dis.
1993; 147: 1151–6.
5 Cano NJ, Roth H, Court-Ortune I et al. Nutritional depletion in
patients on long-term oxygen therapy and/or home mechanical
ventilation. Eur. Respir. J. 2002; 20: 30–7.
6 Eisner MD, Iribarren C, Blanc PD et al. Development of disability
in chronic obstructive pulmonary disease: beyond lung func-
tion. Thorax 2011; 66: 108–14.
7 Stratton RJ, Green CJ, Elia M. Disease-Related Malnutrition: An
Evidence Based Approach to Treatment. CABI Publishing (CABI
International), Oxford, 2003.
8 Ezzell L, Jensen GL. Malnutrition in chronic obstructive pulmo-
nary disease. Am. J. Clin. Nutr. 2000; 72: 1415–6.
9 Coxson HO, Chan IH, Mayo JR et al. Early emphysema in
patients with anorexia nervosa. Am. J. Respir. Crit. Care Med.
2004; 170: 748–52.
10 Ferreira I, Brooks D, Lacasse Y et al. Nutritional intervention in
COPD: a systematic overview. Chest 2001; 119: 353–63.
11 Ferreira I. Chronic obstructive pulmonary disease and malnutri-
tion: why are we not winning this battle? J. Pneumol. 2003; 29:
107–15.
12 Ferreira IM, Brooks D, Lacasse Y et al. Nutritional supplementa-
tion for stable chronic obstructive pulmonary disease. Cochrane
Database Syst. Rev. 2005; (2): CD000998.
13 Efthimiou J, Fleming J, Gomes C et al. The effect of supplemen-
tary oral nutrition in poorly nourished patients with chronic
obstructive pulmonary disease. Am. Rev. Respir. Dis. 1988; 137:
1075–82.
14 Fuenzalida CE, Petty TL, Jones ML et al. The immune response to
short-term nutritional intervention in advanced chronic obstruc-
tive pulmonary disease. Am. Rev. Respir. D is. 1990; 142: 49–56.
15 Rogers RM, Donahoe M, Costantino J. Physiologic effects of oral
supplemental feeding in malnourished patients with chronic
obstructive pulmonary disease. A randomized control study. Am.
Rev. Respir. Dis. 1992; 146: 1511–7.
16 Schols AM, Soeters PB, Mostert R et al. Physiologic effects of
nutritional support and anabolic steroids in patients with
chronic obstructive pulmonary disease. A placebo-controlled
randomized trial. Am. J. Respir. Crit. Care Med. 1995; 152(4 Pt 1):
1268–74. [nourished patients].
17 Collins PF, Stratton RJ, Elia M. Nutritional support in chronic
obstructive pulmonary disease: a systematic review and meta-
analysis. Am. J. Clin. Nutr. 2012; 95: 1385–95.
18 Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews
of Interventions. Version 5.0.2. The Cochrane Collaboration,
2009. [Accessed Jun 2012] Available from URL: www.cochrane-
handbook.org
19 Centre for Reviews and Dissemination. Systematic Reviews:
CRD’s Guidance for Undertaking Reviews in Health Care. CRD,
University of York, York, UK, 2009.
20 Moher D, Cook DJ, Eastwood S et al. Improving the quality of
reports of meta-analyses of randomised controlled trials: the
QUOROM statement. Quality of reporting of meta-analyses.
Lancet 1999; 354: 1896–900.
21 Whittaker JS, Ryan CF, Buckley PA et al. The effects of refeeding
on peripheral and respiratory muscle function in malnourished
chronic obstructive pulmonary disease patients. Am. Rev. Respir.
Dis. 1990; 142: 283–8.
22 Schols AM, Soeters PB, Mostert R et al. Physiologic effects of
nutritional support and anabolic steroids in patients with
chronic obstructive pulmonary disease. A placebo-controlled
randomized trial. Am. J. Respir. Crit. Care Med. 1995; 152(4 Pt 1):
1268–74. [malnourished patients].
23 Jadad AR, Moore RA, Carroll D et al. Assessing the quality of
reports of randomized clinical trials: is blinding necessary?
Control. Clin. Trials 1996; 17: 1–12.
24 Steiner MC, Barton RL, Singh SJ et al. Nutritional enhancement
of exercise performance in chronic obstructive pulmonary
Nutrition support in COPD 627
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
disease: a randomised controlled trial. Thorax 2003; 58: 745–
51.
25 Weekes CE, Emery PW, Elia M. Dietary counselling and food
fortification in stable COPD: a randomised trial. Thorax 2009; 64:
326–31.
26 Adams G. Effect of nasogastric feedings on arterial oxygen
tension in patients with symptomatic chronic obstructive pul-
monary disease. Heart Lung 1988; 17: 594.
27 Aguilaniu B, Goldstein-Shapses S, Pajon A et al. Muscle protein
degradation in severely malnourished patients with chronic
obstructive pulmonary disease subject to short-term total
parenteral nutrition. J. Parenter. Enteral. Nutr. 1992; 16: 248–54.
28 Akrabawi SS, Mobarhan S, Stoltz RR et al. Gastric emptying, pul-
monary function, gas exchange, and respiratory quotient after
feeding a moderate versus high fat enteral formula meal in
chronic obstructive pulmonary disease patients. Nutrition 1996;
12: 260–5.
29 Angelillo VA, Bedi S, Durfee D et al. Effects of low and high car-
bohydrate feedings in ambulatory patients with chronic obstruc-
tive pulmonary disease and chronic hypercapnia. Ann. Intern.
Med. 1985; 103(6 Pt 1): 883–5.
30 Borum ML, Lynn J, Zhong Z et al. The effect of nutritional sup-
plementation on survival in seriously ill hospitalized adults: an
evaluation of the SUPPORT data. Study to Understand Prognoses
and Preferences for Outcomes and Risks of Treatments. J. Am.
Geriatr. Soc. 2000; 48(5 Suppl): S33–8.
31 Broekhuizen R, Creutzberg EC, Weling-Scheepers CA et al. Opti-
mizing oral nutritional drink supplementation in patients with
chronic obstructive pulmonary disease. Br.J.Nutr.2005; 93: 965–
71.
32 Broekhuizen R, Wouters EF, Creutzberg EC et al. Polyunsatu-
rated fatty acids improve exercise capacity in chronic obstructive
pulmonary disease. Thorax 2005; 60: 376–82.
33 Cai B, Zhu Y, Ma Y et al. Effect of supplementing a high-fat,
low-carbohydrate enteral formula in COPD patients. Nutrition
2003; 19: 229–32.
34 Creutzberg EC, Schols AM, Weling-Scheepers CA et al. Charac-
terization of nonresponse to high caloric oral nutritional therapy
in depleted patients with chronic obstructive pulmonary
disease. Am. J. Respir. Crit. Care Med. 2000; 161(3 Pt 1): 745–52.
35 Creutzberg EC, Wouters EF, Mostert R et al. Efficacy of nutri-
tional supplementation therapy in depleted patients with
chronic obstructive pulmonary disease. Nutrition 2003; 19:
120–7.
36 Deacon SJ, Vincent EE, Greenhaff PL et al. Randomized control-
led trial of dietary creatine as an adjunct therapy to physical
training in chronic obstructive pulmonary disease. Am. J. Respir.
Crit. Care Med. 2008; 178: 233–9.
37 Dore MF, Laaban JP, Orvoen-Frija E et al. Role of the thermic
effect of food in malnutrition of patients with chronic obstruc-
tive pulmonary disease. Am. J. Respir. Crit. Care Med. 1997; 155:
1535–40.
38 Efthimiou J, Mounsey PJ, Benson DN et al. Effect of carbohydrate
rich versus fat rich loads on gas exchange and walking perform-
ance in patients with chronic obstructive lung disease. Thorax
1992; 47: 451–6.
39 Faager G, Soderlund K, Skold CM et al. Creatine supplementa-
tion and physical training in patients with COPD: a double blind,
placebo-controlled study. Int. J. Chron. Obstruct. Pulmon. Dis.
2006; 1: 445–53.
40 Felbinger TW, Suchner U. Nutrition for the malnourished
patient with chronic obstructive pulmonary disease: more is
better! Nutrition 2003; 19: 471–2.
41 Ferreira IM, Verreschi IT, Nery LE et al. The influence of 6
months of oral anabolic steroids on body mass and respiratory
muscles in undernourished COPD patients. Chest 1998; 114:
19–28.
42 Forli L, Pedersen JI, Bjortuft O et al. Dietary support to under-
weight patients with end-stage pulmonary disease assessed for
lung transplantation. Respiration 2001; 68: 51–7.
43 Fuld JP, Kilduff LP, Neder JA et al. Creatine supplementation
during pulmonary rehabilitation in chronic obstructive pulmo-
nary disease. Thorax 2005; 60: 531–7.
44 Goris AH, Vermeeren MA, Wouters EF et al. Energy balance in
depleted ambulatory patients with chronic obstructive pulmo-
nary disease: the effect of physical activity and oral nutritional
supplementation. Br.J.Nutr.2003; 89: 725–31.
45 Grigorakos L, Sotiriou E, Markou N et al. Combined nutritional
support in patients with chronic obstructive pulmonary disease
(COPD), under mechanical ventilation (MV). Hepatogastroenter-
ology 2009; 56: 1612–4.
46 Hoogendoorn M, van Wetering CR, Schols AM et al. Is INTERdis-
ciplinary COMmunity-based COPD management (INTERCOM)
cost-effective? Eur. Respir. J. 2010; 35: 79–87.
47 Iovinelli G, Marinangeli F, Ciccone A et al. Parenteral nutrition in
ventilated patients with chronic obstructive pulmonary disease:
long chain vs medium chain triglycerides. Minerva Anestesiol.
2007; 73: 65–76.
48 Knowles JB, Fairbarn MS, Wiggs BJ et al. Dietary supplementa-
tion and respiratory muscle performance in patients with COPD.
Chest 1988; 93: 977–83.
49 Kubo H, Honda N, Tsuji F et al. Effects of dietary supplements on
the Fischer ratio before and after pulmonary rehabilitation. Asia
Pac. J. Clin. Nutr. 2006; 15: 551–5.
50 Lewis MI, Belman MJ, Dorr-Uyemura L. Nutritional supplemen-
tation in ambulatory patients with chronic obstructive pulmo-
nary disease. Am. Rev. Respir. Dis. 1987; 135: 1062–8.
51 Marchesani F, Valerio G, Dardes N et al. Effect of intravenous
fructose 1,6-diphosphate administration in malnourished
chronic obstructive pulmonary disease patients with chronic
respiratory failure. Respiration 2000; 67: 177–82.
52 Otte KE, Ahlburg P, D’Amore F et al. Nutritional repletion in mal-
nourished patients with emphysema. J. Parenter. Enteral. Nutr.
1989; 13: 152–6.
53 Planas M, Alvarez J, Garcia-Peris PA et al. Nutritional support
and quality of life in stable chronic obstructive pulmonary
disease (COPD) patients. Clin. Nutr. 2005; 24: 433–41.
54 Saudny-Unterberger H, Martin JG, Gray-Donald K. Impact of
nutritional support on functional status during an acute exacer-
bation of chronic obstructive pulmonary disease. Am. J. Respir.
Crit. Care Med. 1997; 156(3 Pt 1): 794–9.
55 Slinde F, Gronberg AM, Engstrom CR et al. Individual dietary
intervention in patients with COPD during multidisciplinary
rehabilitation. Respir. Med. 2002; 96: 330–6.
56 Vargas M, Puig A, de la Maza M et al. [Patients with chronic
airflow limitation: effects of the inspiratory muscle training with
threshold load valve, built with appropriate technology,
associated to nutritional support]. Rev. Med. Chil. 1995; 123:
1225–34.
57 Vermeeren MA, Wouters EF, Geraerts-Keeris AJ et al. Nutritional
support in patients with chronic obstructive pulmonary disease
during hospitalization for an acute exacerbation; a randomized
controlled feasibility trial. Clin. Nutr. 2004; 23: 1184–92.
58 Vermeeren MA, Wouters EF, Nelissen LH et al. Acute effects of
different nutritional supplements on symptoms and functional
capacity in patients with chronic obstructive pulmonary disease.
Am. J. Clin. Nutr. 2001; 73: 295–301.
59 Wilson DO, Rogers RM, Sanders MH et al. Nutritional interven-
tion in malnourished patients with emphysema. Am. Rev. Respir.
Dis. 1986; 134: 672–7.
60 Hoekstra J, Uil S, van den Berg J. Nutritional intervention versus
standard nutritional advice in malnourished COPD patients. Eur.
Respir. J. 2005; 26: 721s.
61 Sugawara K, Takahashi H, Kasai C et al. Effects of nutritional
supplementation combined with low-intensity exercise in mal-
nourished patients with COPD. Respir. Med. 2010; 104: 1883–9.
62 Pison CM, Cano NJ, Cherion C et al. Multimodal nutritional
rehabilitation improves clinical outcomes of malnourished
patients with chronic respiratory failure: a randomised control-
led trial. Thorax 2011; 66: 953–60.
PF Collins et al.628
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
63 van Wetering CR, Hoogendoorn M, Broekhuizen R et al. Efficacy
and costs of nutritional rehabilitation in muscle-wasted patients
with chronic obstructive pulmonary disease in a community-
based setting: a prespecified subgroup analysis of the INTER-
COM trial. J. Am. Dir. Assoc. 2010; 11: 179–87.
64 Baldi S, Aquilani R, Pinna GD et al. Fat-free mass change after
nutritional rehabilitation in weight losing COPD: role of insulin,
C-reactive protein and tissue hypoxia. Int. J. Chron. Obstruct.
Pulmon. Dis. 2010; 5: 29–39.
65 Laviolette L, Lands LC, Dauletbaev N et al. Combined effect of
dietary supplementation with pressurized whey and exercise
training in chronic obstructive pulmonary disease: a rand-
omized, controlled, double-blind pilot study. J. Med. Food 2010;
13: 589–98.
66 DeLetter M. A Nutritional Intervention for Persons with Chronic
Airflow Limitation. University of Kentucky, Lexington, KY, 1991.
67 The Global Initiative for Chronic Obstructive Lung Disease
(GOLD). Pocket guide to COPD diagnosis, management, and
prevention. 2011. [Accessed Jun 2012] Available from URL: www.
goldcopd.org
68 Schols AM, Slangen J, Volovics L et al. Weight loss is a reversible
factor in the prognosis of chronic obstructive pulmonary
disease. Am. J. Respir. Crit. Care Med. 1998; 157(6 Pt 1): 1791–7.
69 Norman K, Stobaus N, Gonzalez MC et al. Hand grip strength:
outcome predictor and marker of nutritional status. Clin. Nutr.
2011; 30: 135–42.
70 Hornby ST, Nunes QM, Hillman TE et al. Relationships between
structural and functional measures of nutritional status in a nor-
mally nourished population. Clin. Nutr. 2005; 24: 421–6.
71 Peng S, Plank LD, McCall JL et al. Body composition, muscle
function, and energy expenditure in patients with liver cirrhosis:
a comprehensive study. Am. J. Clin. Nutr. 2007; 85: 1257–66.
72 Budziareck MB, Pureza Duarte RR, Barbosa-Silva MC. Reference
values and determinants for handgrip strength in healthy sub-
jects. Clin. Nutr. 2008; 27: 357–62.
73 Cawood AL, Elia M, Stratton RJ. Systematic review and meta-
analysis of the effects of high protein oral nutritional supple-
ments. Ageing Res. Rev. 2012; 11: 278–96.
74 Soler-Cataluna JJ, Sanchez-Sanchez L, Martinez-Garcia MA et al.
Mid-arm muscle area is a better predictor of mortality than body
mass index in COPD. Chest 2005; 128: 2108–15.
75 Prescott E, Almdal T, Mikkelsen KL et al. Prognostic value of
weight change in chronic obstructive pulmonary disease: results
from the Copenhagen City Heart Study. Eur. Respir. J. 2002; 20:
539–44.
76 Oga T, Nishimura K, Tsukino M et al. Analysis of the factors
related to mortality in chronic obstructive pulmonary disease:
role of exercise capacity and health status. Am. J. Respir. Crit. Care
Med. 2003; 167: 544–9.
77 Arozullah AM, Khuri SF, Henderson WG et al. Development and
validation of a multifactorial risk index for predicting postopera-
tive pneumonia after major noncardiac surgery. Ann. Intern.
Med. 2001; 135: 847–57.
78 Datta D, Lahiri B. Preoperative evaluation of patients
undergoing lung resection surgery. Chest 2003; 123: 2096–
103.
79 Porszasz J, Emtner M, Goto S et al. Exercise training decreases
ventilatory requirements and exercise-induced hyperinflation at
submaximal intensities in patients with COPD. Chest 2005; 128:
2025–34.
80 Koechlin C, Couillard A, Cristol JP et al. Does systemic inflam-
mation trigger local exercise-induced oxidative stress in COPD?
Eur. Respir. J. 2004; 23: 538–44.
81 Garrod R, Ansley P, Canavan J et al. Exercise and the inflamma-
tory response in chronic obstructive pulmonary disease
(COPD)—does training confer anti-inflammatory properties in
COPD? Med. Hypotheses 2007; 68: 291–8.
82 Sugawara K, Takahashi H, Kashiwagura T et al. Effect of anti-
inflammatory supplementation with whey peptide and exercise
therapy in patients with COPD. Respir. Med. 2012; 106: 1526–
34.
83 de Torres JP, Casanova C, Montejo de Garcini A et al. Gender and
respiratory factors associated with dyspnea in chronic obstruc-
tive pulmonary disease. Respir. Res. 2007; 8: 1–17.
84 World Health Organization. Active Aging: A Policy Framework.
WHO, Madrid, Spain, 2002.
85 Wise RA, Brown CD. Minimal clinically important differences in
the six-minute walk test and the incremental shuttle walking
test. COPD 2005; 2: 125–9.
86 Singh SJ, Jones PW, Evans R et al. Minimum clinically important
improvement for the incremental shuttle walking test. Thorax
2008; 63: 775–7.
87 Lacasse Y, Goldstein R, Lasserson T et al. Pulmonary rehabilita-
tion for chronic obstructive pulmonary disease. Cochrane Data-
base Syst. Rev. 2006; (4): CD003793.
88 Ferreira IM, Brooks D, White J et al. Nutritional supplementation
for stable chronic obstructive pulmonary disease. Cochrane
Database Syst. Rev. 2012; (12): CD000998.
Nutrition support in COPD 629
© 2013 The Authors
Respirology © 2013 Asian Pacific Society of Respirology
Respirology (2013) 18, 616–629
