Effectsoffruitandvegetableconsumptiononinflammatory ...€¦ · Design: Electronic databases...
Transcript of Effectsoffruitandvegetableconsumptiononinflammatory ...€¦ · Design: Electronic databases...
Effects of fruit and vegetable consumption on inflammatorybiomarkers and immune cell populations: a systematic literaturereview and meta-analysis
Banafshe Hosseini,1 Bronwyn S Berthon,1 Ahmad Saedisomeolia,2,3 Malcolm R Starkey,1 Adam Collison,1
Peter AB Wark,1 and Lisa G Wood1
1Grow Up Well Priority Research Centre and Priority Research Centre for Healthy Lungs and, Hunter Medical Research Institute, University of Newcastle,Newcastle, Australia; 2School ofMedicine,Western Sydney University, Sydney, Australia; and 3School of Nutritional Sciences and Dietetics, Tehran Universityof Medical Sciences, Tehran, Iran
ABSTRACTBackground: Inflammation is associated with an increased risk ofa range of chronic diseases. A diet high in fruit and vegetables mayhelp to reduce inflammation, as fruit and vegetables are rich sourcesof antioxidants and other biologically active substances, which mayimprove immune function.Objective: To summarize the evidence, we executed a systematicreview and meta-analysis examining the effects of fruit and/or veg-etable intake on inflammatory biomarkers and immune cells in hu-mans with different diseases and conditions.Design: Electronic databases including PubMed, Cochrane,CINAHL, and EMBASE were systematically searched up toMarch 2018.Results: Eighty-three studies were included. Of these, 71 (86%)were clinical trials, and 12 were observational studies (n= 10 cross-sectional and n = 2 cohort). Amongst the observational research,n = 10 studies found an inverse association between intakes of fruitor vegetables and inflammatory biomarkers. Similarly, the majorityof the intervention studies (68%, n = 48) reported beneficial effectsof fruit or vegetable intake on ≥1 biomarker of systemic or airwayinflammation. A meta-analysis of included studies showed that fruitor vegetable intake decreased circulating levels of C-reactive proteinand tumor necrosis factor-α (P < 0.05) and increased the γ δ-T cellpopulation (P < 0.05).Conclusions: In conclusion, this review suggests that higher intakesof fruit and vegetables lead to both a reduction in proinflammatorymediators and an enhanced immune cell profile. Am J Clin Nutr2018;108:136–155.
Keywords: fruits, vegetables, antioxidants, inflammation,immunity
INTRODUCTION
The relation between oxidative stress, inflammation, andthe risk of a wide range of chronic health conditions has beenfrequently described in the literature (1–3). Inflammation is
essential for protecting the body against insult and injury.However, when inflammation becomes persistent, the mediatorsproduced by activated immune cells can lead to tissue damageand development of disease (3). A chronic inflammatory stateis characterized by increased levels of circulating inflammatorybiomarkers such as C-reactive protein (CRP), TNF-α, and IL-6(4). Therapeutic strategies that target reducing inflammationhave the potential to dramatically reduce the burden of manychronic diseases.
Sufficient fruit and vegetable (F&V) intake is one of thecornerstones of a healthy diet, and may provide protectionagainst cardiovascular disease (CVD), several cancers, and otherchronic diseases (5). F&Vs are rich dietary sources of variousimmune-protective substances such as fiber, folate, vitamins, aswell as non-nutrient phytochemicals, including carotenoids andflavonoids, such as β-carotene, anthocyanins, flavanols, and fla-vanones (6). These substances can have profound effects on cel-lular growth and differentiation, and are needed for the optimalfunctioning of the immune system (7).
Various studies have reported that a high F&V consumptioncan decrease systemic inflammation (3). Some studies have ex-amined the effect of F&V interventions on immune cell popula-tions. Increased natural killer cell (NK cell) cytotoxicity and lym-phocyte proliferation have been reported following consumptionof different F&V juices (8–10). In addition, some epidemiologicstudies have also reported that F&V intake is inversely associ-
Address correspondence to LGW (e-mail: [email protected]).Abbreviations used: CRP, C-reactive protein; CVD, cardiovascular dis-
ease; F&V, fruit and vegetable; GPR, G protein-coupled receptor; hsCRP,high-sensitivity C-reactive protein; ICAM-1, intercellular adhesionmolecule-1; iNOS, inducible nitric oxide synthase; MCP-1, monocyte chemotacticprotein-1; MD, mean difference; NK cell, natural killer cell; PBMC, pe-ripheral blood mononuclear cell; sICAM-1, soluble intercellular adhesionmolecule-1; sVCAM-1, soluble vascular adhesion molecule-1; VCAM-1,vascular adhesion molecule-1; WBC, white blood cell count.Received December 15, 2017. Accepted for publication March 28, 2018.First published online 0, 2018; doi: https://doi.org/10.1093/ajcn/nqy082.
136 Am J Clin Nutr 2018;108:136–155. Printed in USA. © 2018 American Society for Nutrition. All rights reserved.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 137
FIGURE 1 Example of search strategy with the use of PubMed for studies investigating the effect of fruit and/or vegetable on inflammatory biomarkers andimmune cell population in humans. CRP, C-reactive protein; eNO, exhaled nitrico oxide; ICAM-1, intercellular adhesion molecule; IFN, interferon; VCAM-1,vascular cell adhesion protein 1.
ated with proinflammatory biomarkers, such as CRP (1, 11, 12)and TNF-α (13). However, there is heterogeneity in the literatureand not all studies have reported beneficial effects (14, 15).
Although previous meta-analyses have assessed the relationbetween F&V intake and risk of CVD (16) or other chronic dis-eases such as type 2 diabetes (17), to our knowledge, there is nocomprehensive study that reviews the effects of F&V intake oninflammation and immune cell populations. Hence, the aim ofthis systematic literature review and meta-analysis is to evalu-ate the association between F&V consumption and inflammationand immunity. We specifically aimed to identify and examine theavailable evidence for the effects of F&V intake on inflammatorybiomarkers and immune cell populations in humans.
METHODS
Search strategy
A systematic search of relevant papers published beforeMarch 2017 was performed with the use of PubMed, Cochrane,CINAHL, and EMBASE with the keyword search term only(inclusion and exclusion criteria were not considered in the
search strategy). Studies were limited to humans with no timerestriction. In addition, the reference lists of retrieved articles andrelevant systematic reviews were searched to identify other rele-vant studies. See Figure 1 for an example of the search strategyused. The search was conducted again in March 2018 to ensurethat any relevant articles published after the initial search wereidentified. The Medical Subject Headings search terms included:fruit, fruit extract, vegetable, vegetable products, fruit and veg-etable juices, inflammatory markers, IL, CRP, TNF-α, IL-6, im-mune cells, T cells, NK cells, B cells, dendritic cells, lympho-cytes, and antigen-presenting cells.
Study selection
This systematic review considered only original studies withthe following designs: randomized controlled trials, cohortstudies, case-control studies, before and after studies, and cross-sectional studies. Animal models, in-vitro studies, systematicreviews, narrative reviews, opinion papers, case studies, casereports, and conference abstracts were excluded. Review articleswere collected for the purposes of reviewing the reference listsand did not contribute to the final number of included studies.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
138 HOSSEINI ET AL.
TABLE 1Methodologic quality rating of each study as determined by the judgments of the authors1
First author (year) Q12 Q23 Q34 Q45 Q56 Q67 Q78 Q89 Q910 Q1011 QA
Amagase (2009) (25) Y Y Y N Y Y Y Y Y Y PosVidal (2012) (52) Y Y Y Y Y Y Y Y Y UC PosGuo (2014) (37) Y Y Y Y Y Y Y Y Y UC PosKarlsen (2010) (39) Y Y Y Y N Y Y Y Y UC PosKolehmainen (2012) (41) Y Y Y Y N Y Y Y Y Y PosLarmo (2008) (43) Y Y Y Y Y Y Y Y Y UC PosMcAnulty (2011) (45) Y Y N N N Y Y Y Y UC NeuRiso (2013) (47) Y N Y Y NA Y Y Y Y Y NeuBasu (2014) (28) Y Y Y Y N Y Y Y Y Y PosXie (2017) (55) Y Y Y Y Y Y Y Y Y Y PosNilsson (2017) (58) Y Y Y Y Y Y Y Y Y UC PosDuffey (2015) (59) Y Y Y Y N Y Y Y Y N PosHosseini (2016) (2) Y Y Y Y Y Y Y Y Y Y PosAsgary (2013) (26) Y Y N Y N Y Y Y Y Y NeuDavidson (2009) (33) Y Y Y Y Y Y Y Y Y UC PosShema-Didi (2012) (49) Y Y Y Y Y Y Y Y Y Y PosSohrab (2014) (50) Y Y Y Y Y Y Y Y Y Y PosHan (2016) (38) Y Y Y N Y Y Y Y Y Y PosTomé-Carneiro (2012) (51) Y Y Y Y Y Y Y Y Y UC PosCastilla (2006) (31) Y Y Y Y N Y Y Y Y Y PosZunino (2014) (54) Y Y Y Y Y Y Y Y Y Y PosBarona (2012) (27) Y Y Y Y Y Y Y Y Y UC PosBell (2014) (29) Y Y Y Y N Y Y Y Y Y PosKelley (2013) (40) Y N N Y NA Y Y Y Y UC NeuKent (2017) (56) Y Y Y Y Y Y Y Y Y UC PosBuscemi (2012) (30) Y Y Y Y N Y Y Y Y Y PosDalgard (2009) (32) Y Y Y N Y Y Y Y Y Y PosSanchez-Moreno (2003) (48) Y Y N N NA Y Y Y Y UC NeuDeopurkar (2010) (34) Y UC Y NA NA N Y Y Y Y NeuGammon (2014) (36) Y N Y Y N Y Y Y Y Y NeuHunter (2012) (53) Y Y Y Y Y Y Y Y Y UC PosDow (2013) (35) Y Y Y N N Y Y Y Y UC NeuLeelarungrayub (2016) (44) Y Y N Y N Y Y Y Y UC NeuKanellos (2017) (57) Y Y Y Y Y Y Y Y Y Y PosNishizawa (2011) (46) Y N Y N Y Y Y Y Y N NeuKuntz (2014) (42) Y N Y N Y Y Y Y Y Y NeuBub (2003) (8) Y N Y N N Y Y Y Y UC NeuGhavipour (2013) (4) Y N Y Y N Y Y Y Y Y NeuWatzl (2003) (9) Y N Y Y Y Y Y Y Y Y NeuWatzl (1999) (10) Y N N Y N Y Y Y Y UC NeuBriviba (2004) (14) Y N Y Y Y Y Y Y Y UC NeuWatzl (2000) (15) Y Y Y Y Y Y Y Y Y UC PosAalami-Harandi (2015) (60) Y Y Y N Y Y Y Y Y Y PosRiso (2014) (61) Y N Y Y NA Y Y Y Y UC NeuUpritchard (2000) (62) Y Y Y Y N Y Y Y Y UC PosInserra (1999) (65) Y Y N Y NA Y Y Y Y Y NeuJin (2010) (3) Y Y Y Y Y Y Y Y Y Y PosMacready (2014) (6) Y Y Y Y N Y Y Y Y UC PosBaldrick (2012) (1) Y Y Y Y N Y Y Y Y Y PosFreese (2004) (11) Y Y Y NA N Y Y Y Y N PosMcCall (2011) (12) Y Y Y Y NA Y Y Y Y Y PosLamprecht (2007) (13) Y N Y Y Y Y Y Y Y UC NeuBobe (2010) (63) Y Y Y Y N Y Y Y Y Y PosHunter (2012) (64) Y Y Y Y N Y Y Y Y Y PosKnab (2013) (66) Y N Y Y N Y Y Y Y Y NeuKnab (2014) (67) Y Y Y N N Y Y Y Y Y PosLamprecht (2013) (68) Y N Y Y Y Y Y Y Y UC NeuNadeem (2014) (69) Y Y Y Y NA Y Y Y Y Y PosNantz (2006) (70) Y Y Y Y Y Y Y Y Y UC PosWatzl (2005) (71) Y N Y NA Y Y Y Y Y UC NeuWilliams (2017) (72) Y Y Y Y Y Y Y Y Y Y PosWood (2012) (73) Y Y Y Y NA Y Y Y Y Y PosRomieu (2009) (23) Y Y Y Y N Y Y Y Y Y Pos
(Continued)
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 139
TABLE 1(Continued)
First author (year) Q12 Q23 Q34 Q45 Q56 Q67 Q78 Q89 Q910 Q1011 QA
Holt (2009) (20) Y Y Y Y N Y Y Y Y Y PosHermsdorff (2010) (74) Y Y Y Y N Y Y Y Y Y PosLopez-Garcia (2004) (75) Y N Y Y N Y Y Y Y Y NeuRoot (2012) (76) Y Y Y Y N Y Y Y Y Y PosWannamethee (2006) (77) Y N Y Y N Y Y Y Y Y NeuAlmeida-de-souza (2017) (22) Y Y Y Y NA Y Y Y Y Y PosRowe (2011) (80) Y Y Y Y Y Y Y Y Y Y PosKontiokari (2005) (24) Y Y Y Y Y Y Y Y Y UC PosZunino (2013) (81) Y Y Y Y Y Y Y Y Y Y PosMcAnulty (2014) (78) Y Y Y N N Y Y Y Y UC PosNantz (2013) (79) Y Y Y Y Y Y Y Y Y UC PosBohn (2010) (82) Y N Y Y N Y Y Y Y Y NeuNantz (2012) (84) Y Y Y Y Y Y Y Y Y UC PosMurashima (2004) (83) Y N N Y N Y Y Y Y Y NeuLiu (2012) (85) Y N Y Y N Y Y Y Y UC NeuGibson (2012) (5) Y Y Y Y Y Y Y Y Y Y PosWinkler (2004) (86) Y Y Y Y NA Y Y Y Y UC PosHendricks (2008) (87) Y N Y Y N Y Y Y Y Y NeuNettleton (2010) (88) Y Y Y Y N Y Y Y Y Y PosKruzich (2004) (21) Y Y Y Y N Y Y Y Y Y Pos
1N, no; NA, not applicable; Neu, neutral; Pos, positive; Q, question; QA, quality assessment; UC, unclear; Y, yes.2Q1: Was the research question clearly stated?3Q2: Was the selection of study subjects/patients free from bias?4Q3: Were study groups comparable?5Q4: Was the method of handling withdrawals described?6Q5: Was blinding used to prevent introduction of bias?7Q6: Were intervention/therapeutic regimens/exposure factor or procedure and any comparisons described in detail? Were intervening factors described?8Q7: Were outcomes clearly defined and the measurements valid and reliable?9Q8: Was the statistical analysis appropriate for the study design and type of outcome indicators?10Q9: Are conclusions supported by results with biases and limitations taken into consideration?11Q10: Is bias due to study’s funding or sponsorship unlikely?
The target study population was humans of all ages, genders, orethnicities, with any health status. The exposure of interest wasconsumption of whole or extracted fruit or vegetables. The studyoutcome measures were inflammatory biomarkers such as ILs,CRP, and TNF-α, as well as markers of immune function suchas immune cell populations.
Citations from electronic databases were imported into thereferencing software Endnote X8. Two reviewers (BH andAS) independently reviewed and evaluated all articles by ti-tle and abstract, and thereafter performed independent full-text assessments. A third independent reviewer (BSB) wasconsulted in cases where reviewers’ evaluations were notin agreement.
Study quality
Eligible studies were assessed independently in terms of theirmethodological quality by 2 reviewers (BH and AS) based on astandardized critical appraisal checklist designed by the Ameri-can Dietetic Association (18). The tool considered the reliability,validity, and generalizability of the included studies. This toolcomprises 4 relevance questions that address the applicabilityof the study findings to practice and 10 validity questions thataddress scientific rigor, including risk of bias. Based on the re-sponses to these questions as determined by the reviewers (BH
and AS), each study was rated as having negative, positive, orneutral quality. The methodologic quality rating of the includedstudies is detailed in Table 1.
Data extraction and study synthesis
Study details were extracted and recorded into a custom-designed database. Data extracted included title, authors,country, study design, participant characteristics, study factor(e.g., dosage/dietary intake of fruits and vegetables), studyduration, main outcome measures, findings including statisticalsignificance, analysis with adjustment for confounding factors,and limitations. The Preferred Reporting Items for SystematicReviews and Meta-Analyses (PRISMA) protocol was followedduring data extraction and study synthesis (19).
Statistical methods
A meta-analysis was performed to investigate the effects ofF&V intake on inflammatory biomarkers (CRP, TNF-α, andIL-6) and immune cell population percentage (αβ-T cells, γ δ-T cells, and NK cells) with Review Manager (RevMan, version5.3, Nordic Cochrane Centre). Only studies thatmet the followinginclusion criteria were included in the meta-analysis: 1) F&V in-take reported; 2) the mean and SD, SE, or IQRs were reported; 3)
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
140 HOSSEINI ET AL.
FIGURE2 Search and inclusion process flowchart of studies for inclusion in a systematic review of the effects of fruit and vegetable intakes on inflammatorybiomarkers and immune cell population.
the OR or the relative risks and the corresponding 95% CIs werereported. All reported SEs, 95% CIs, and IQRs were converted toSDs. However, due to the differences in exposure and outcome as-sessments, meta-analysis of all included studies was not possible.Appreciable heterogeneity was assumed if I2 > 50 and P < 0.1.Meta-analysis was performed with the use of fixed-effect mod-elling if I2 < 50, and random-effectmodellingwas used if I2 > 50.Included studies used different interventions; thus meta-analysiswas performed in studies with similar outcome measurements.The inverse-variance statistical method was used, and the meandifference (MD) and corresponding 95% CIs were calculated.
RESULTS
Search results
A flow chart showing the article retrieval and inclusion processis presented in Figure 2. A total of 8853 articles were identified
with n= 530 duplicates identified and excluded. The titles of theremaining 8323 articles were reviewed with n= 295 retrieved forabstract appraisal. Abstracts from 131 articles met the inclusioncriteria and full texts were retrieved for the further review. Eighty-three articles met the criteria and were included in this systematicliterature review.
Study characteristics
The majority of studies (94%, n = 78) were performed inadults (≥18 y of age) with only 3 in adolescents (20–22) and 2 inchildren (23, 24). In terms of study design, experimental was themost common (86%, n= 71), with 10 cross-sectional studies and2 cohort studies. A total of n = 5420 participants were includedin clinical trials with a mean intervention period of 85 d, rangingfrom 6 h to 4 y. Overall, 17,710 participants were included fromcross-sectional studies, and 556 individuals with a mean follow-up of 5.22 y from cohort studies. The methodological quality of
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 141
TABLE2
Summaryof
includ
edstud
iesexam
iningtheeffectsof
fruitand
vegetableintake
oninflammatorybiom
arkers1
Firstautho
r(year)
Cou
ntry
Design
Popu
latio
n,n
Age,y
Interventio
nEffectson
inflammatorybiom
arkers
Amagase(2009)
(25)
USA
RCT
Health
y,60
55–7
212
0mLstandardized
barberry
fruitjuice
(equ
ivalentto
≥150
gfreshfruit)or
placebodaily
for30
d↑S
erum
levelsof
IL-2
Vidal(201
2)(52)
China
RCT
Health
y,15
065
–70
Lacto-w
olfberry
orplacebo(13.7g/d)
for3mo
↔Any
biom
arkers
Guo
(201
4)(37)
China
RCT
Obesity,44
18–2
525
0mLof
either
barberry
juiceor
placebotw
icedaily
for4wk
↓Plasm
alevelsof
TNF-
α,and
IL-8
Karlsen
(201
0)(39)
USA
RCT
Atriskof
CVD,
6230
–70y(m
en)
and45
–70y
(wom
en)
330mLbilberry
juice(n
=31
)or
water
(n=
31)for4wk
↓Plasm
aconcentrations
ofCRP,IL-6,IL-15,and
MIG
,↑TNF-
α,↔
otherbiom
arkers
Kolehmainen(201
2)(41)
Finland
RCT
Metabolic
synd
rome,27
BLB:5
3±
6CO:5
0±
7Adietrich
inBLB(400
gfreshBLB)or
acontrold
ietfor
8wk
↓Serum
hsCRP,IL-6,IL-12
Larmo(200
8)(43)
Finland
RCT
Health
y,25
419
–50
28gfrozen
seabuckthornpu
reeor
placebofor90
d↓S
erum
CRPconcentrations
McA
nulty
(201
1)(45)
USA
RCT
Welltrained,25
BB (31.1
±12
.6)
CO
(33.4
±16
.0)
250gblueberries/dfor6wkand37
5ggiven1hpriorto
2.5h
ofrunn
ingor
placebo
↔Any
biom
arkers
Riso(201
3)(47)
Italy
Crossover
RCT
Atriskof
CVD,
1847
.8±
9.7
AWB(25gfreeze-dried
powder,providing37
5mgof
ACNs)
oraplacebodrinkfor6wk,
spaced
bya6-wkwash-ou
t↔
Serum
levelsof
IL-6,T
NF-
α,C
RP,sV
CAM-1
Basu(201
4)(28)
UK
RCT
Abd
ominal
adiposity,60
49±
10One
ofthefollo
wing4beveragesfor12
wks:1
)LD-FDS(25
g/d),2)LDcontrol,3)
HD-FDS(50g/d),and
4)HDcontrol
↔Serum
levelsof
hsCRP,sICAM,and
sVCAM
Xie(201
7)(55)
USA
RCT
Health
y,49
18–6
550
0mgberryextracto
rplacebofor12
wk
↔Any
biom
arkers
Nilsson(201
7)(58)
Sweden
Crossover
RCT
Health
y,40
50–7
0Berry
beverage
basedon
amixture
ofberries(150
gblueberries,50
gblackcurrant,50gelderberry,50g
lingo
nberries,50gstrawberry,and10
0gtomatoes)or
acontrolb
everage,daily
for5wk
↔PlasmaIL-6
andIL-18levels
Duffey(201
5)(59)
USA
Cross-
sectional
Health
y,10,334
≥19
Average
404mLof
CJC
vs.<
404ml
↓Serum
levelsof
CRP
Hosseini(20
16)(2)
Iran
RCT
Overw
eight/
obese,42
30–6
010
00mgpo
megranateextract,or
aplacebo,daily
for30
d↓P
lasm
alevelsof
IL-6
andhsCRP
Asgary(2013)
(26)
Iran
Exp
erim
ental
13hypertensive
men
39–6
8NaturalPJ
(150
ml/d
)follo
winga12
-hfast
↔Blood
levelsof
hsCRP,ICAM-1,V
CAM-1,
E-selectin
,and
IL-6
Davidson(200
9)(33)
USA
RCT
Atriskof
CHD,
289
45–7
424
0mlP
J/dor
controlb
everagefor18
mo
↔Blood
levelsof
CRP
Shem
a-Didi(20
12)
(49)
Israel
RCT
Chron
ichemod
ialysis
patie
nts,10
1
>18
100mLPJ,ormatchingplacebodu
ring
each
dialysis(3/w
k)for1y
↓Plasm
alevelsof
IL-6,T
NF-
α
Sohrab
(201
4)(50)
Iran
RCT
T2D
M,50
40–6
525
0mLPJ/d
oracontrolb
everagefor12
wk
↓Plasm
aCRPandIL-6
levels
Han
(201
6)(38)
Korea
RCT
Overw
eigh
t/ob
ese,76
30–7
0Con
trol
(starch,4g/d,n
=24),low-G
O[low
-doseGO,g
rape
pomaceextract(34
2.5mg/d)
+om
ijafruitextract(57.5
mg/d),n
=26
],andhigh
-GO[high-do
seGO,g
rape
pomace
extract(68
5mg/d)
+om
ijafruitextract(115
mg/d),n
=26
]grou
psfor10
wk
↓Plasm
aIL-1b,andTNF-
αlevels
Tomé-Carneiro(201
2)(51 )
Spain
RCT
Und
ergo
ing
prim
ary
preventio
nof
CVD,7
5
18–8
0A:resveratrol-richgrapesupp
lement(resveratrol8
mg);B
:conventio
nalg
rape
supp
lementlacking
resveratrol;C:
placebo(m
altodextrin)
forthefirst6moandado
ubledo
seforthenext
6mo
A:↓
PlasmahsCRP;
B:↔
anybiom
arker
(Continued)
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
142 HOSSEINI ET AL.
TABLE2
(Continued)
Firstautho
r(year)
Cou
ntry
Design
Popu
latio
n,n
Age,y
Interventio
nEffectson
inflammatorybiom
arkers
Castilla(200
6)(31)
Spain
RCT
Receiving
hemod
ialysis,
38;h
ealth
y,15
30–7
0Hem
odialytic
patie
nts:RGJ(100
mL/d)or
controlfor
14d.
Health
ycontrols:1
00mLRGJ/dfor14
d↓P
lasm
alevelsof
MCP-1,
↔VCAM1,IC
AM1,andCRP
Zun
ino(201
4)(54)
USA
Crossover
RCT
Obesity,24
20–6
092
ggrapepowder(∼
4servings
ofgrape)
orplacebofor9wk
↔Plasmalevelsof
CRP
Baron
a(201
2)(27)
USA
RCT
Metabolic
synd
rome,24
30–7
046
glyop
hiliz
edgrapepowder(equ
ivalentto25
2gfreshgrapes),
orplacebofor4wk
↑mRNAexpression
ofIL-10,
andtheindu
cibleiN
OSin
thoseindividu
alswith
outd
yslip
idem
ia,↔
plasmaIL-6,
IL-8,and
TNF-
αlevels
Bell(20
14)(29)
UK
RCT
Trained
cyclists,
1630
±8
30mLMCor
placebo,2tim
es/d
for7consecutivedays
↓Plasm
alevelsof
IL-6
andhsCRP,
↔IL-1-β
orTNF-
α
Kelley(201
3)(40)
USA
Beforeand
after
Health
y,18
45–6
1Su
pplementedthedietswith
Bingsw
eetcherries(280
g/d)
for28
d↓P
lasm
alevelsof
CRP,plasminog
enactiv
ator
inhibitor-1,endo
thelin-1,epiderm
algrow
thfactor,and
IL-18,
↑IL-1
receptor
antago
nist
Kent(20
17)(56)
Australia
RCT
Mild
-to-mod
erate
dementia
,49
>70
200mLeither
acherry
juiceor
controlb
everage/dfor12
wk
↔PlasmaCRPandIL-6
levels
Buscemi(20
12)(30)
Italy
Sing
le-blin
dcrossover
stud
y
Atriskof
CVD,
18;h
ealth
y,12
27–5
62period
sof
7deach
with
a3-dinterval,eachparticipant
alternativelyrand
omized
toreceive50
0mLorange
juiceor
placebo
↓Plasm
aconcentrations
ofhsCRP,IL-6,and
TNF-
α
Dalgard
(200
9)(32)
Denmark
2×
2factorial,
crossover
RCT
Periph
eralarterial
disease,48
60.57
Juice
+vitamin
E,juice
+placebo,referencebeverage
(sugar
drink)
+vitamin
E,reference
beverage
+placebofor28
d,separatedby
a4-wkwash-ou
tperiod
↓Plasm
alevelsof
CRPandfib
rino
genby
11%
and3%
,respectiv
ely,
↔IL-6
Sanchez-Moreno(200
3)(48)
Spain
Exp
erim
ental
Health
y,12
20–3
250
0mLhigh
-pressureorange
juice/dfor14
d↓P
lasm
alevelsof
PGE2,
↔CRP
Deopu
rkar
(201
0)(34)
USA
Trial
Health
y,48
25–4
730
0-kcaldrinks
ofeither
glucose,saturatedfatascream,o
rang
ejuice,or
only
water
toingestin
afastingstate
↔Exp
ressionof
TLR-4,N
F-kB
bind
ing,TNF-
α,IL-1
β
inMNCsandin
plasmaLPS
.Gam
mon
(201
4)(36)
New
Zealand
Crossover
RCT
Hyp
er-
cholesterol,85
27–7
34-wkhealthydietrun-in,beforetwo4-wkinterventio
nsof
2green
kiwifruit/dplus
healthydiet(intervention)
orhealthydietalon
e(con
trol)
↓Serum
CRPandIL-6
concentrations
Hun
ter(201
2)(53)
New
Zealand
Crossover
RCT
Elderly,32
≥65
The
equivalent
of4kiwifruitor
2bananas/dfor4wk,with
treatm
entsseparatedby
a4-wkwasho
utperiod
↔Plasmalevelsof
hsCRP,ho
mocysteine
Dow
(201
3)(35)
USA
RCT
Overw
eigh
t/ob
ese,69
;metabolic
synd
rome,29
>18
Alowbioactivedietplus
1.5grapefruit/dfor6wk(n
=37
,n=
14with
MetS)
oracontrolcon
ditio
nin
which
alowbioactivediet
devoid
ofcitrus
was
consum
ed(n
=32
,n=
15with
MetS)
↓Plasm
alevelsof
hsCRP,
↔sV
CAM-1
Leelarung
rayu
b(201
6)(44)
Thaila
ndBeforeand
after
Elderly,29
54–7
8A2-wkcontrolp
eriodwas
follo
wed
by4wkof
100gfreshstar
fruitjuice
consum
ption2tim
es/d
aftermeals
↓Blood
levelsof
NO,T
NF-
αandIL-23
,↔IL-2
Kanellos(201
7)(57)
Greece
RCT
Health
y,36
20–4
0Raisins
equalto5fruitserving
s(90g/d)
orplacebofor4wk
↔PlasmaCRPandleptin
levels
Nishizawa(201
1)(46)
Japan
RCT
Run
ners,20
20.6
±1.25
Twice-daily
FRLFE
(100
mg/d,or
50mg/do
se)or
placebofor2
mo
↓Serum
levelo
fIL-6,↔
IL-10
Kun
tz(201
4)(42)
Germany
Crossover
RCT
Health
y,30
23–2
733
0mLbeverages(placebo
,fruitjuice,andfruitsmoo
thiewith
8.9
±0.3,98
3.7
±37
,and
840.9
±10
mgACN/L,respectively)
over
14d
↔Any
biom
arkers
Bub
(200
3)(8)
Germany
Crossover
RCT
Health
y,27
35±
42po
lyph
enol-richjuices
[236
mg(A
)and22
6mg(B
)po
lyph
enols
with
cyanidin
glycosides
(A)andepigallocatechin
galla
te(B
)](330
mL/d)supp
lementedfor2wk
↑IL-2
secretionby
activ
ated
lymph
ocytes
(Continued)
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 143
TABLE2
(Continued)
Firstautho
r(year)
Cou
ntry
Design
Popu
latio
n,n
Age,y
Interventio
nEffectson
inflammatorybiom
arkers
Ghavipo
ur(201
3)(4)
Iran
RCT
Overw
eigh
t/obese,
106
22–2
433
0mLof
tomatojuiceor
water/d
for20
d↓S
erum
levelsof
IL-8
andTNF-
α
Watzl(200
3)(9)
Germany
Crossover
RCT
Health
y,22
28.7
±5.9
330mLof
either
tomatojuice(37.0mglycopene)or
carrot
juice
(27.1mg
β-caroteneand13
.1mg
α-carotene/d)
for2wkon
alow-carotenoiddietwith
a2-wkdepletionperiod
afterjuice
interventio
n
Low
-carotenoiddiet:↓
IL-2;juice
interventio
n:↔
any
biom
arkers
Watzl(199
9)(10)
Germany
Exp
erim
ental
Health
y,23
27–4
0Weeks
1–2:
low-carotenoidperiod
,throu
ghou
tweeks
3–8:
daily
consum
ptionof
330mLtomatojuice(40mg)
(weeks
3–4),330
mLcarrot
juice(w
eeks
5–6),10gdriedspinachpowder(w
eeks
7–8)
Low
-carotenoiddiet:↓
IL-2
andIL-4
secretion;
tomato
juiceconsum
ption:
↑IL-2
andIL-4
secretion;
carrot
juiceandspinachpowderconsum
ption:
↔any
biom
arker
Briviba
(200
4)(14)
Germany
RCT
Non
smoker,30;
smoker,25
30–4
5Three
tomatooleoresinextractcapsulesdaily
(eachcontaining
4.88
mglycopene,0.48mgphytoene,0.44mgphytofl
uene,and
1.18
1mg
α-tocop
herol)or
placebofor2wk
↔IL-2
andTNF-
αsecretion
Watzl(200
0)(15)
Germany
RCT
Health
y,50
63–8
633
0mLtomatojuice/d(47.1mglycopene/d)or
mineralwater
for
8wk
↔Se
cretionof
TNF-
α,IL-2,and
IL-4
Aalam
i-Harandi
(201
5)(60)
Iran
RCT
Pregnant
wom
enat
27wkof
gestation,44
18–4
0One
garlictablet(equ
alto
400mggarlicand1mgallic
in)
(n=
22)or
placebo(n
=22
)on
cedaily
for9wk
↓Serum
levelsof
hsCRP
Riso(201
4)(61)
Italy
Crossover
trial
Smokers(>
10cigarettes/d),17
21.8
±2.7
Broccolid
iet(25
0g/d)/wash-ou
t/con
trol
dietvs.con
trol
diet/wash-ou
t/broccolid
iet.Eachanalysiswas
separatedby
15d
ofwash-ou
tperiod
↓Plasm
aCRP,
↔TNF-
α,IL-6,IL-6sR
oradipon
ectin
,
Upritc
hard
(200
0)(62)
New
Zealand
RCT
T2D
M,57
<75
Tomatojuice(500
mL/d),or
vitamin
E(800
U/d)or
vitamin
C(500
mg/d)
orplacebofor4wk
↑Plasm
alycopene
levels,↔
CRP,VCAM
Inserra(199
9)(65)
USA
Beforeand
after
Elderly,53
60–8
6Fruitjuice
supp
lementscontained85
0mgfruitp
owder/capsule
madefrom
extractsof
apples,orang
es,pineapp
les,papaya,
cranberries,andpeaches.Vegetablesupp
lementscontained75
0mgvegetablepowder/capsuleandcontainedextractsof
carrots,
parsley,beets,broccoli,
kale,cabbage,spinach,and
tomatoes.
Participantsconsum
edextractfor
80d
↑IL-2
secretion,
↔TNF-
α
Jin(201
0)(3)
USA
RCT
Health
y,11
722
–55
Placebo,FV
,orFV
Bcapsulefora60
-dperiod
Bothinterventio
ns:↓
MCP-1,MIP-1b,andRANTES
Macready(201
4)(6)
UK
RCT
Atriskof
CVD,
174
26–7
0High-flavono
idFV
,low
-flavon
oidFV
,orhabitualdiet,w
ithhigh
-andlow-flavon
oidFV
amou
ntssequ
entia
llyincreasing
by2,
4,and6(+
2,+4
,and
+6)po
rtions/d
every6wkover
habitual
intakes
Low
-flavon
oidFV
diet:↓
CRP,Eselectin,and
VCAM
with
14po
rtions/d
Baldrick(201
2)(1)
UK
RCT
COPD
with
ahabitually
low
FVintake
(≤2
portions
FV/d),
81
Low
FV:
61.2
±8.3;
HighFV
:63
.2±
9.1
≥5po
rtions
ofFV
/dor
≤2po
rtions
FV/d
for12
wk
↔Any
biom
arkers
Freese
(200
4)(11)
Finland
RCT
Health
y,77
19–5
2A6-wkdietcontaining
either
810or
196gof
vegetables,berries,
andapples/d,and
rich
either
inlin
oleicacid
(11%
energy
)or
oleicacid
(12%
energy
)
↔Plasmalevelsof
ICAM,and
hsCRP
(Continued)
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
144 HOSSEINI ET AL.
TABLE2
(Continued)
Firstautho
r(year)
Cou
ntry
Design
Popu
latio
n,n
Age,y
Interventio
nEffectson
inflammatorybiom
arkers
McC
all(20
11)(12)
UK
RCT
Hyp
ertensive,11
740
–65
1,3or
6po
rtions
ofFV
/dfor8wk
↔Se
rum
levelsof
hsCRP,sICAM,V
CAM
Lam
prech(200
7)(13)
Austria
RCT
Health
y,41
34±
5Placeboor
JPCcapsules
for28
wk
↓Serum
concentrations
ofCRPandTNF-
α
Bob
e(201
0)(63)
USA
RCT
Colorectal
adenom
a,87
252
–71
Extensive
dietaryandbehavioralcoun
selling
toachievethePP
Tdietarygo
alsof
20%
oftotalenergyfrom
fat,18
g/10
00kcalof
dietaryfib
er,and
5–8daily
servings
(depending
ontotalcaloric
intake)of
FV;o
rcontrol(didno
treceive
such
coun
selling
and
wereexpected
tocontinue
theirusualintake)
grou
psfor4y
↓Serum
IL-6
concentrations
Hun
ter(201
2)(64)
New
Zealand
Crossover
RCT
Smokers,42
30–6
3Non
supp
lementedmilk
,prototype
milk
Acontainedacombinatio
nof
grapeseed
extract(0.21
3g/serving),app
leextract(0.21
3g/serving),and
tomatoconcentrate(0.106
g/serving),prototype
milk
Bcontainedboysenberryjuiceconcentrate(0.285
g/serving)
andappleextract(0.28
5g/serving)
for6wk
↔Any
inflammatorybiom
arkers
Knab(201
3)(66)
USA
Cross
over
RCT
Swim
mers,9
24.6
±0.7
Com
pleted
10-d
training
with
orwith
out1
6floz
offreshFV
juice
(230
mgflavono
ids)ingested
before
andafer
workout
↔Any
inflammatorybiom
arkers
Knab(201
4)(67)
USA
RCT
Cyclists,34
In:3
5.1
±8.0;
Pl:
35.5
±8.2
Afreeze-dried
FVjuicepowder(230
mgflavono
ids/d)
orplacebo
for17
d↔
Exercise-indu
cedalteratio
nsin
inflammatory
biom
arkers(IL-6,IL-8,T
NF-
α,M
CP-1,CRP)
Lam
precht
(201
3)(68)
Austria
RCT
Obese
prem
enop
ausal,
42
41±
5FV
Bcapsules
thatprovided
7.5mg
β-carotene,20
0mgvitamin
C,
60mgRRR-α-tocop
herol,60
0mgfolateand63
kJ/d
orplacebo
for8wk
↓Serum
levelsof
CRPandTNF-
α
Nadeem
(201
4)(69)
UK
RCT
Hyp
ertensive,11
240
–65
Interventio
nof
1,3,or
6po
rtions
ofFV
/dfor8wkaftera4-wk
washo
utperiod
(1servingFV
/d)
↓Serum
amyloidArelatedinflammation,
↔IL-6,
hsCRP,E-selectin
Nadeem
(201
4)(69)
UK
RCT
Low
FVintake
(≤2serving/d),
82
65–8
5Interventio
nof
2or
5po
rtions
ofFV
/dfor16
wk
FV:↓
serum
amyloidArelatedinflammation,
↔IL-6,
hsCRP,E-selectin
Nantz(200
6)(70)
US
RCT
Health
y,59
21–5
3FV
CJ4capsules/d
orplacebofor77
d↓S
erum
IFN-γ
Watzl(200
5)(71)
Germany
RCT
Low
FVintake
(≤2serving/d),
64
31±
92,5,or
8servings/d
ofcaroteno
id-richvegetables
andfruitfor
4-wkperiod
8servings
FV/d:↓
serum
CRP
Williams(201
7)(72)
Australia
RCT
Overw
eigh
t/ob
esity,56
≥40
6capsules
ofFV
juiceconcentrateor
placebofor8wk
↓Plasm
aTNF-
αlevel
Woo
d(201
2)(73)
Australia
RCT
Asthm
a,13
7HAD:5
4±
14;
LAD:5
8±
15Ahigh
-antioxidant
diet(5
servings
ofvegetables
and2servings
offruit/d
,n=
46)or
alow-antioxidant
diet(<
2servings
ofvegetables
and1servingof
fruit/d
,n=
91)for14
d,andthen
thosewho
consum
edthehigh
-antioxidant
dietreceived
placebo.
Subjectswho
consum
edthelow-antioxidant
dietreceived
placeboor
tomatoextract(45
mglycopene/d).The
interventio
ncontinuedun
tilweek14
orun
tilan
exacerbatio
noccurred
Low
-antioxidant
diet:↑
plasmaCRPatweek14
;tom
ato
extract:
↔airw
ayor
system
icinflammation
Rom
ieu(200
9)(23)
Mexico
Coh
ort
Asthm
atic
child
ren,15
8vs.
50healthy
controlfollowed
foran
average
of22
wk
6–14
>1/dvs.≥
4tim
esFV
/mo
FVI:
↓IL-8
levelsin
nasallavage
Holt(20
09)(20)
US
Cross-
sectional
Health
y,28
513
–17
Highestvs.low
estserving
/d↓P
lasm
alevelsof
CRP,IL-6,and
TNF-
α
(Continued)
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 145
TABLE2
(Continued)
Firstautho
r(year)
Cou
ntry
Design
Popu
latio
n,n
Age,y
Interventio
nEffectson
inflammatorybiom
arkers
Hermsdorff(201
0)(74)
Spain
Translatio
nal
Health
y,12
020
.8±
2.6
Highestvs.low
esttertile
↓Plasm
alevelsof
CRP,ho
mocysteine,IC
AM1,IL1-R1,
IL-6,T
NF-
α;↓
NF-
κB1gene
expression
inPB
MC
Lop
ez-G
arcia(200
4)(75)
USA
Cross-
sectional
Health
y,73
243
–69
Highestvs.low
estq
uartile
↓plasm
aCRPandE-selectin
Roo
t(20
12)(76)
USA
Cross-
sectional
Health
y,10
0018
–85
Fruit:
≤2vs.>
2servings/d;v
egetables:
≤3vs.>
3servings/d
↓Serum
levelsof
CRP,IL-6,T
NF-
α
Wannamethee(200
6)(77)
UK
Cross-
sectional
Health
y,32
5860
–79
<1/wkvs.>
7/wk
Fruitintake:
↓serum
CRPlevels;v
egetableintake:↔
Alm
eida-de-souza
(201
7)(22)
Portugal
Cross-
sectional
Health
y,41
212
–18
Highestvs.low
esttertile
Greater
varietyof
vegetableconsum
ption:
↓serum
levels
ofCRP,IL-6
andoverallinfl
ammatoryscore;fruit
variety:
↔1ACN,antho
cyanin;B
B,b
lueberries;BLB,b
ilberries;CHD,coron
aryheartd
isease;CJC
,cranb
erry
juicecocktail;
CO,con
trol;C
OPD
,chron
icob
structivepu
lmon
arydisease;
CRP,C-R
eactiveProtein;
CVD,cardiovasculardisease;FD
S,freeze-dried
strawberries;floz,fl
uidounce;FR
LFE
,flavonol-richlychee
fruitextract;F
V,fruitandvegetable;FV
B,fruitandvegetablejuicepowderconcentratewith
added
berrypowder;FV
CJ,fruitand
vegetableconcentratejuice;FV
I,fruitand
vegetableindex;GO,grape
pomaceandom
ijiafruitethanolextracts;H
AD,highantio
xidantdiet;H
D,highdo
se;hsC
RP,high
-sensitiv
ityC-reactiveprotein;
ICAM,intracellu
laradhesion
molecule;IFN,interferon;
In.,interventio
n;iN
OS,
isoform
ofnitricox
idesynthase;JPC
,juice
powderconcentrate;LAD,low
antio
xidant
diet;L
D,low
dose;
LPS
,lipop
olysaccharide;MC,M
ontm
orency
cherries;M
CP-1,
mon
ocytechem
otactic
protein-1;
MetS,
metabolicsynd
rome;MIG
,mon
okineindu
cedby
interferon
-gam
maMIP-1b,
macroph
ageinflammatory
protein1-b;
MNC,m
onocytes;P
BMC,p
eripheralb
lood
mon
ocytecell;
PGE2,
prostaglandinE2;
PJ,p
omegranatejuice;Pl.,Placebo;
PPT,
polyppreventio
ntrial;RANTES,
regu
latedup
onactiv
ationno
rmal
Tcellexpressedandsecreted;RCT,
random
ized
controlledtrial;RGJ,redgrapejuice;
sICAM,soluble
intracellularadhesion
molecule;
sVCAM,solub
levascular
adhesion
molecule;
TLR,toll-lik
ereceptor;
T2D
M,type2diabetes
mellitus;V
CAM,vascularadhesionomolecule;WB,w
ildblueberries;
↑,increase;↓
,decrease;
↔,noeffect.
n = 55 studies was positive. These studies were methodologi-cally strengthened by their use of random allocation to the in-tervention and control group or treatment arm (cross over trials),double blinding, and comparability of study group; n = 28 stud-ies were neutral. Factors that limited the methodologic quality ofthe studies rated as neutral included, insufficient detail providedregarding the study inclusion/exclusion criteria, group compara-bility, intervening factors, and data collection and analysis. Nostudy was excluded due to negative quality.
Effects of fruit and vegetable intake on systemic/airwayinflammatory biomarkers
Fruit only
Thirty-six experimental trials (2, 8, 25–58), and one cross-sectional study (59) assessed the association of fruit intake and in-flammatory biomarkers and circulating levels of high-sensitivityCRP (hsCRP), TNF-α, IL-6, IL-8, soluble vascular adhesionmolecule-1 (sVCAM), soluble intercellular adhesion molecule-1 (sICAM), or E-selectin (Table 2). Fruit intake was reported tohave beneficial effects on at least 1 marker of systemic inflam-mation such as hsCRP or TNF-α in most of the studies (2, 8,25, 27, 29–32, 35–41, 43–46, 48–51, 59), although some stud-ies (26, 28, 33, 34, 42, 47, 52–58) found no significant effects.The type of fruits investigated varied, and included berries, suchas strawberries, blueberries, or barberry (n = 12) (25, 28, 37,39, 41, 43, 45, 47, 52, 55, 58, 59), pomegranate juice (26, 33,49, 50) or extract (2) (n = 5), grape products such as grape ex-tract (38, 51), powder (27, 54) or juice (31), and cherries (29,40, 56). Other studies examined whether consumption of mixed-fruit juice (8, 42), orange juice (30, 32, 34, 48), kiwifruit (36, 53),grapefruit (35), star fruit (44), raisins (57), or lychee extract (46)beneficially altered circulating levels of hsCRP, TNF-α, IL-6,IL-8, sVCAM, sICAM, or E-selectin.
Amagase et al. (25) examined the effects of daily barberry juiceon immune function, and reported that the intervention group hadsignificantly decreased serum levels of IL-2 compared to baselineand the control group, whereas serum concentration of IL-4 wasnot altered by the treatment. In contrast, a 3-mo trial by Vidalet al. (52) found that there was no difference in CRP, IL-6, ororosomucoid following barberry supplementation compared toplacebo. Guo et al. (37) described a 4-wk trial comparing bay-berry juice with placebo and reported the intervention group hadsignificantly lower plasma levels of TNF-α and IL-8 compared toplacebo. Similarly, 2 studies (39, 41) showed that supplementa-tion with bilberries resulted in decreased CRP and IL-6; however,plasma concentration of TNF-α unexpectedly increased follow-ing bilberry intervention in 1 study (39). In another trial by Larmoet al. (43) consumption of frozen sea buckthorn berries for 90d significantly decreased serum levels of CRP compared to thebaseline and placebo. Two studies (45, 47) reported that blueberryintervention did not reduce plasma levels of CRP, IL-6, TNF- α,and sVCAM. One study assessed the effects of strawberries onbiomarkers of inflammation, and showed no significant changeon blood levels of hsCRP, sICAM, and sVCAM (28). Two stud-ies reported no significant changes in IL-6, CRP, IL-1β, or TNF-αfollowing consumption of berry extract (55) or mixed-berry bev-erage (58). A cross-sectional study by Duffey et al. (59) reported
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
146 HOSSEINI ET AL.
TABLE3
Summaryof
includ
edstud
iesexam
iningtheeffectsof
fruitand
vegetableintake
onim
mun
efunctio
n1
Firstautho
r(year)
Cou
ntry
Stud
ydesign
Popu
latio
n,n
Age,y
Interventio
nEffectson
immun
ecells
Row
e(201
1)(80)
USA
RCT
Health
y,85
50–7
536
0mL(12oz)of
beverage
(100
%grapejuiceor
placebo)
daily
for9wk
↑Num
bersof
circulating
γδ-T,↔
proliferationof
γδ-T
cells
Bub
(200
3)(8)
Germany
Crossover
RCT
Non
smok
ing,27
35±
42po
lyph
enol-richjuices
[236
mg(A
)and22
6mg(B
)po
lyph
enolswith
cyanidin
glycosides
(A)and
epigallocatechin
galla
te(B
)](330
mL/d)supp
lemented
for2wk
↑Lym
phocyteproliferativerespon
siveness,the
lytic
activ
ityof
NKcells
Amagase(200
9)(25)
USA
RCT
Health
y,60
55–7
212
0mLstandardized
barberry
fruitjuice
(equ
ivalentto
≥150
gfreshfruit)or
placebodaily
for30
d↑N
umberof
lymph
ocytes,↔
numberof
CD4,CD8,and
NKcells
McA
nulty
(201
1)(45)
USA
RCT
Well-trained
subjects,25
BB: 31.1
±12
.6;
placebo:
33.4
±16
.0
250gblueberries/dfor6wkand37
5ggiven1hpriorto
2.5hof
runn
ingor
placebo
↑NKcellcoun
ts
Kon
tiokari(200
5)(24)
Finland
RCT
341child
ren
4.3
Cranb
erry
juice(5
mL/kg)
oraplacebofor3mo
↔Carriageof
respiratorybacteria
Shem
a-Didi(20
12)(49)
Israel
RCT
Chron
ichemod
ialysis
patie
nts,10
1
>18
100mLPJ,ormatchingplacebodu
ring
each
dialysis
(3/w
k)for1y
↓Infectio
nincidencerate
Nishizawa(201
1)(46)
Japan
RCT
Lon
g-distance
runn
ers,20
20.6
±1.25
Twice-daily
FRLFE
(100
mg/d,or
50mg/do
se)or
placebo
for2mo
↔Neutrop
hilo
rlymph
ocytecoun
ts,totalWBCmod
ified
Vidal(201
2)(52)
China
RCT
Elderly,150
65–7
0Lacto-w
olfberry
orplacebo(13.7g/d)
for3mo
↔Im
mun
ecells
compo
sitio
n,autoantib
odieslevels
Zun
ino(201
4)(54)
USA
Crossover
RCT
Obesity,24
20–6
092
ggrapepowder(∼
4servings
ofgrape)
orplacebofor9
wk
↔Proliferativerespon
sesof
CD4+
orCD8+
Tcells,
cytokine
prod
uctio
nby
activ
ated
Tcells
Zun
ino(201
3)(81)
USA
Crossover
RCT
Obesity,20
20–5
0Strawberrypowder(∼
4servings
offrozen
strawberries/d)
orplacebofor3wk,
andthen
crossedover
toanother
treatm
entfor
thelasttreatm
ent
↔Cellcou
nts,theoverallp
ercentageof
CD4+
and
CD8+
cells
at24
,48,and72
h,cytokine
prod
uctio
nby
T-lymph
ocyte
Hun
ter(201
2)(53)
New
Zealand
Crossover
RCT
Elderly,32
≥65
The
equivalent
of4kiwifruitor
2bananas/dfor4wk,
with
treatm
entsseparatedby
a4-wkwasho
utperiod
↔Innateim
mun
efunctio
n(N
Kcellactiv
ity,
phagocytosis)
McA
nulty
(201
4)(78)
USA
RCT
Health
y,25
18–5
0Blueberry
(equ
ivalentto25
0gberries)or
placebodaily
for6wk
↑AbsoluteNKcells
Nantz(201
3)(79)
USA
RCT
Health
y,45
21–5
0Low
-caloriecranberrybeverage
(450
mL)madewith
ajuice-derived,powderedcranberryfractio
n(n
=22
)or
aplacebobeverage
(n=
23),daily
for10
wk
↑Proliferationindexof
γδ-T
cells,↔
Bcells,N
Kcells,
αβ-T
cells,and
mon
ocytes
Boh
n(201
0)(82)
Norway
Trial
Health
ysm
okers,29
45–7
5Antioxidant-richdiet,a
kiwifruitdiet(3
kiwifruits/d
addedto
theregu
lardiet)or
acontrolg
roup
for8wk
Upregulationgrou
psof
genesinvolved
inregu
latio
nof
immun
ecells
Nantz(201
2)(84)
USA
RCT
Health
y,12
021
–50
2.56
ggarlicextract/d
orplacebofor90
d↑P
roliferationof
NKcells,activationstateof
theNK
popu
latio
n,γδ-T
cells
proliferation
Watzl(200
3)(9)
Germany
Crossover
RCT
Health
y,22
28.7
±5.9
330mL/d
ofeither
tomatojuice(37.0mglycopene/d)or
carrot
juice(27.1mg
β-carotene/dand13
.1mg
α-carotene/d)
for2wkon
alow-carotenoiddietwith
a2-wkdepletionperiod
afterjuiceinterventio
n
Low
-carotenoiddiet:↓
NKcellcytotoxicity,and
lymph
ocyteproliferation;
juiceinterventio
n:↔
Watzl(199
9)(10)
Germany
Exp
erim
ental
Health
y,23
27–4
0Weeks
1–2:
low-carotenoidperiod
;throu
ghou
tweeks
3–8:
daily
consum
ptionof
330mLtomatojuice(40mg)
(weeks
3–4),330
mLcarrot
juice(w
eeks
5–6),10g
driedspinachpowder(w
eeks
7–8)
Low
-carotenoiddiet:↓
proliferationof
PBMCs;tomato
juiceconsum
ption:
↔lymph
ocyteproliferation;
carrot
juiceandspinachpowderconsum
ption:
↔
(Continued)
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 147
TABLE3
(Continued)
Firstautho
r(year)
Cou
ntry
Stud
ydesign
Popu
latio
n,n
Age,y
Interventio
nEffectson
immun
ecells
Briviba
(200
4)(14)
Germany
RCT
Smoker,25,
nonsmoker,
30
30–4
53tomatooleoresinextractcapsules/d(eachcontaining
4.88
mglycopene,0.48mgphytoene,0.44mg
phytofl
uene
and1.18
1mg
α-tocop
herol)or
placebofor
2wk
↔Ly
mph
ocyteproliferation,NKcellactiv
ity
Watzl(200
0)(15)
Germany
RCT
Elderly,50
63–8
633
0mLtomatojuice/d(47.1mglycopene/d)or
mineral
water
for8wk
↔Ly
mph
ocyteproliferation
Murashima(200
4)(83)
Japan
Exp
erim
ental
Health
y,12
20–3
6Freshbroccolisprouts(100
g/d)
for1wk
↔NKcellactiv
ityLiu
(201
2)(85)
USA
Self-con
trolled,
long
itudinal
stud
y
HIV-infected
wom
en,77
≥18
“Every
oralmostevery
day”
vs.“on
lyas
needed”
↔Any
immun
olog
icmarkers
Inserra(199
9)(65)
USA
Beforeandafter
Elderly,53
60–8
6y
Fruitjuice
supp
lementscontained85
0mgfruit
powder/capsulemadefrom
extractsof
apples,orang
es,
pineapples,papaya,cranberries,andpeaches.Vegetable
supp
lementscontained75
0mgvegetable
powder/capsuleandcontainedextractsof
carrots,
parsley,beets,broccoli,
kale,cabbage,spinach,and
tomatoes.Participantsconsum
edextractfor
80d
↑Spo
ntaneous
proliferationof
PBMCs,NKcell
cytotoxicity
Gibson(201
2)(5)
UK
RCT
Health
y,83
65–8
5≤2
FVpo
rtions/d
or≥5
FVpo
rtions/d
for16
wk
↑Antibod
ybind
ingto
pneumococcalcapsularLPS
(total
IgG),
↔antib
odybind
ingto
tetanu
stoxo
idJin(201
0)(3)
USA
RCT
Health
y,11
722
–55
Placebo,FV
,orFV
Bcapsulefora60
-dperiod
↔WBCcoun
tsKnab(201
4)(67)
USA
RCT
Cyclists,43
In:3
5.1
±8.0;
Pl:3
5.5
±8.2
Afreeze-dried
FVjuicepowder(230
mgflavono
ids/d)
orplacebofor17
d↔
Immun
efunctio
n(G
-PHAGandM-PHAG)
Nantz(200
6)(70)
USA
RCT
Health
y,59
21–5
3FV
CJ4capsules/d
orplacebofor77
d↑N
umberof
γδ-T
cells
Watzl(200
5)(71)
Germany
RCT
Intake
of<2
servings
FV/d
31±
92,5,or
8servings/d
ofcaroteno
id-richvegetables
and
fruitfor
4-wkperiod
8servings
FV/d
↔im
mun
olog
icmarkers
Winkler
(200
4)(86)
Germany
RCT
HIV,2
3vs.
healthy
controls,18
20–5
31Lfruitjuice
(group
J)or
30mLFV
concentrate(group
C)daily
for16
wk
Fruitjuice:↑
PHA-ind
uced
lymph
ocyteproliferationin
HIV
+andin
HIV-subject,
↔thecoun
tofCD4+
and
CD8+
cells,the
CD4+
:CD8+
ratio
andHIV
viralload
Hendricks
(200
8)(87)
USA
Coh
ort
HIV,3
48≥1
8Highestvs.low
estm
eanintake
↑CD4coun
tRoo
t(20
12)(76)
USA
Cross-sectio
nal
Health
y,10
0018
–85
Fruit:
≤2vs.>
2servings/d;v
egetables:
≤3vs.>
3servings/d
↓WBCcoun
ts
Nettle
ton(201
0)(88)
USA
Cross-sectio
nal
Health
y,11
01F:
70.7
±5.4;
M:71.8
±5.5
Highestvs.low
estq
uartile
↔Im
mun
olog
icmarkers
Kruzich
(200
4)(21)
USA
Cross-sectio
nal
HIV-infected,
264andHIV
uninfected,
127
13–2
3Maxim
umvs.m
inim
umscore
↔Im
mun
olog
icmarkers
1BB,blueberries;C
D,cluster
ofdifferentia
tion;
FRLFE
,flavon
ol-richlychee
fruitextract;F
V,fruitandvegetables;F
VB,fruitandvegetablejuicepowderc
oncentratewith
addedberrypowder;FV
CJ,fruit
andvegetableconcentratejuice;G-PHAG,g
ranulocytephagocytosisandoxidativeburstactivity
;HD,H
eartDisease;In.,intervention;
M-PHAG,m
onocyteph
agocytosisandox
idativeburstactivity
;NKcells,
naturalk
iller
cells;P
BMC,p
eripheralb
lood
mon
ocytecell;
PHA,p
hytohemagglutinin;P
J,po
megranatejuice;Pl.,placebo;
RCT,
rand
omized
controlle
dtrial;WBC,w
hitebloo
dcell;
↑,increase;↓
,decrease;
↔,n
oeffect.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
148 HOSSEINI ET AL.
FIGURE 3 Forest plot of randomized controlled trials investigating the effects of fruit and vegetable intake on circulating C-reactive protein levels (mil-ligrams per liter). Values are mean differences with 95% CIs determined with the use of generic IV random-effects models. Heterogeneity was quantified byI2 at a significance of P < 0.10. IV, inverse variance.
that consumption of cranberry juice was significantly associatedwith lower CRP levels.
We have previously investigated the effects of pomegranateextract consumption on inflammation (2), and reported thatpomegranate extract intervention for 30 d reduced plasma CRPand IL-6. Similarly, consumption of pomegranate juice resultedin reduction of plasma IL-6, TNF-α, and CRP levels in 2 stud-ies (49, 50). However, some studies (26, 33) did not observe anychanges in circulating inflammatory biomarkers, such as hsCRP,intercellular adhesion molecule-1 (ICAM-1), vascular adhesionmolecule-1 (VCAM-1), and IL-6, following pomegranate juiceconsumption. Three studies reported significantly decreased lev-els of TNF-α (38), CRP (51), or monocyte chemotactic protein-1(MCP-1) (31) following grape supplement intervention. In con-trast, intake of grape powder for 9 wk had no significant effecton blood CRP level (54). In another trial by Barona et al. (27)consumption of grape powder for 4 wk increased expression ofinducible nitric oxide synthase (iNOS) and IL-10. However, nosignificant changes were observed in any inflammatory biomark-ers, such as IL-6 and TNF-α.
Interventions to increase cherry consumption reduced circulat-ing levels of IL-6 and hsCRP, compared with control groups in 2studies (29, 40); however, 1 study reported no significant changein plasma CRP and IL-6 concentrations (56). Four trials exam-ined inflammatory responses following orange juice intake, with3 showing a protective effect against systemic inflammation (pri-marily hsCRP, IL-6, or prostaglandin E2) (30, 32, 48). However,Deopurkar et al. (34) found no effect in a similar orange juiceintervention trial.
Gammon et al. (36) described a 4-wk trial of 2 greenkiwifruit/d plus healthy diet compared with healthy diet, and
reported that kiwifruit resulted in a significant reduction in serumlevels of CRP and IL-6. In another trial by Hunter et al. (53), nosignificant difference was observed in the serum level of hsCRPbetween those who received kiwifruit compared with those whoreceived bananas daily for 4 wk.
Three studies investigated the effects of consumption of grape-fruit (35), star fruit (44), or lychee extract (46), and all showed asignificant reduction in either hsCRP or TNF-α compared withplacebo. No changes in plasma levels of CRP and leptin werereported in a raisin intervention study (57). Two trials assessedthe effects of mixed-fruit juice intake on IL-2, IL-6, IL-8, CRP,or TNF-α. One study (42) did not observe any changes, whereasthe other (8) reported a significant increase in IL-2 secretion byactivated lymphocytes compared with baseline.
Vegetables only
Eight trials (4, 9, 10, 14, 15, 60–62) examined anti-inflammatory responses to increased vegetable intake, with3 showing a protective effect on inflammation including reducedhsCRP, TNF-α, and IL-6 (4, 60, 61) (Table 2). In a 9-wk trial(60), intake of garlic supplements decreased serum levels of CRPcompared to placebo. In another trial (14), no change in TNF-αwas observed following a low-carotenoid diet supplemented witha tomato extract for 2 wk. Daily consumption of tomato juicealso had no significant effect on CRP, IL-2, IL-4, and TNF-α (9,10, 15, 62). In contrast with these findings, we previously showedthat supplementation with tomato juice for 20 d resulted in asignificant reduction in serum concentrations of IL-8 and TNF-αcompared to placebo and baseline (4). In another study (61), 10-dbroccoli consumption significantly decreased plasma CRP levels.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 149
FIGURE 4 Forest plot of randomized controlled trials investigating the effects of fruit and vegetable intake on circulating TNF-α (pg/mL). Values aremean differences with 95% CIs determined with the use of generic IV random-effects models. Heterogeneity was quantified by I2 at a significance of P< 0.10.IV, inverse variance.
Fruit and vegetables
Twenty-five studies, including 17 experimental trials (1, 3, 6,11–13, 63–73), 1 cohort (23) and 6 cross-sectional studies (20,22, 74–77), assessed the effects of F&V intake on systemic orairway inflammation (Table 2). One cohort (23) and 5 cross-sectional studies (20, 22, 74–76) reported that F&V intake wasinversely associated with circulating levels of inflammatory cy-tokines, such as CRP, IL-6, and TNF-α. Wannamethee et al. (77)reported that fruit intake was significantly associated with lowerplasma CRP levels, though vegetable intake was not. The major-ity of experimental trials (3, 6, 13, 63, 65, 68–73) (66%, n = 11)
reported a significant reduction in systemic inflammation as indi-cated by inflammatory biomarkers including hsCRP, TNF-α, IL-6, and E-selectin following F&V consumption, whereas 6 trials(1, 11, 12, 64, 66, 67) found no effect of F&V intake on systemicor airway inflammation.
Studies on the effects of F&V intake on immune cellpopulations
Fruit only
Twelve experimental trials (8, 25, 45, 46, 52–54, 78–82) ex-amined the effects of fruit intake on immune cell function, and
FIGURE 5 Forest plot of randomized controlled trials investigating the effects of fruit and vegetable intake on circulating IL-6 (pg/ml). Values are meandifferences with 95% CIs determined with the use of generic IV random-effects models. Heterogeneity was quantified by I2 at a significance of P < 0.10. IV,inverse variance.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
150 HOSSEINI ET AL.
FIGURE 6 Forest plot of randomized controlled trials investigating the effects of fruit and vegetable intake on γ δ-T cell population percentage. Values aremean differences with 95% CIs determined with the use of generic IV random-effects models. Heterogeneity was quantified by I2 at a significance of P< 0.10.IV, inverse variance.
1 study (24) assessed whether fruit intake can beneficially affectrespiratory bacterial flora (Table 3). Nine studies (8, 45, 46, 54,78–82) reported that fruit consumption induced beneficial alter-ations in immune cell function, whereas some studies (24, 25, 52,53) found no effect (on immune cell populations).More than 50%of studies (n= 7) (24, 25, 45, 52, 78, 79, 81) examined the effectsof different kind of berries, mostly cranberries and blueberries,on immune cell function. Other studies tested the hypothesis thatconsumption of grapes (54, 80), kiwifruit (53, 82), lychee extract(46), or mixed-fruit juice (8) can modify immune function.
Two studies (45, 78) found that intake of blueberries increasedthe NK cell population. In another trial by Nantz et al. (79) in-take of a low-calorie cranberry beverage increased the prolifera-tion index of γ δ-T cells. In agreement with these findings, Bub etal. (8) found that fruit juice consumption significantly increasedlymphocyte proliferative responses as well as the lytic activity ofNK cells. In another study (81), consumption of strawberries for 3wk was found to increase the production of TNF-α by the mono-cyte population after stimulation with LPS, which may improvethe effectiveness of an immune response to invading pathogens.In contrast, 3 studies examining consumption of either barberries(25), goji berries (52), or kiwifruit (53) found no significant effecton immune cell proliferation. Similarly, consumption of lycheeextract (46) for 2 mo did not change proliferation of neutrophilsor lymphocytes; however, total white blood cell counts weremod-ified. Increased numbers of circulating γ δ-T cells were reportedfollowing grape juice consumption for 9 wk (80). In contrast, inanother trial (54) consumption of grape powder for 9 wk did notaffect the immune cell population; however, a significant increasein the production of IL-6 and IL-1β was reported in LPS-activatedmonocytes.
In terms of other effects on immunity, adding kiwifruits dailyto a regular diet resulted in upregulation of 11 genes involved inimmune-related processes compared to baseline and control (82).
Also, composition of respiratory bacteria was not altered aftercranberry juice intake for 3 mo (24).
Vegetables only
Seven studies (9, 10, 14, 15, 83–85) examined immune cellpopulations following vegetable intake, with n= 3 showingmod-ulation of immune cell function (9, 10, 84) (Table 3). In a 3-motrial (84), intake of garlic supplements increased proliferation ofNK cells and γ δ-T cells compared to placebo, whereas garlic hadno effect on immune cells in another supplementation study (85).Watzl et al. (10) found that carotenoid depletion decreased prolif-eration of peripheral blood mononuclear cells (PBMCs). In onestudy (9), supplementation with tomato juice increased lytic ac-tivity of NK cells, although no effects were seen in other studiesinvestigating tomato extract (14) and tomato juice (15). In addi-tion, no beneficial effect was observed on immune cell functionfollowing consumption of broccoli (83).
Fruit and vegetables
Twelve studies, including 8 experimental trials (3, 5, 65, 67,70, 71, 82, 86), 1 cohort (87) and 3 cross-sectional studies (21,76, 88), assessed the effects of F&V intake on immune cells(Table 3). More than 60% of studies (n = 7) showed a beneficialassociation of F&V consumption and function of immune cells(5, 65, 70, 76, 82, 86, 87). Bøhn et al. (82) reported upregulationof genes involved in the immune-related processes followingconsumption of a high antioxidant diet (>5 servings of F&V/d)compared to baseline and a control group. In 3 studies (65,70, 86), increased proliferation of lymphocytes were observedin the group supplemented with F&V extract compared toplacebo. Similarly, a prospective study (87) of 348 HIV-positiveadults reported that F&V intake was correlated with number
FIGURE 7 Forest plot of randomized controlled trials investigating the effects of fruit and vegetable intake on αβ-T cell population percentage. Values aremean differences with 95% CIs determined with the use of generic IV random-effects models. Heterogeneity was quantified by I2 at a significance of P< 0.10.IV, inverse variance.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 151
FIGURE 8 Forest plot of randomized controlled trials investigating the effects of fruit and vegetable intake on NK cell population percentage. Values aremean differences with 95% CIs determined with the use of generic IV random-effects models. Heterogeneity was quantified by I2 at a significance of P< 0.10.IV, inverse variance.
of CD4+ T lymphocyte cells. A cross-sectional study by Rootet al. (76) showed that fruit intake was inversely associated withwhite blood cell count (WBC), although, they did not find anyassociation between vegetable intake and WBC. In contrast, 5cross-sectional studies (3, 21, 67, 71, 88) found no significantassociation between F&V intake and immune function.
Findings from meta-analysis
Through the use of meta-analysis, we examined changes incirculating levels of the most commonly evaluated inflammatorybiomarkers, including CRP, TNF-α, and IL-6, following F&Vintervention. Higher F&V intake was correlated with lower bloodlevels of CRP (MD: –0.34; 95% CI: –0.58, –0.11; P < 0.01;I2 = 71%) (Figure 3) and TNF-α (MD: –0.87; 95% CI: –1.59,–0.15; P= 0.02; I2 = 86) (Figure 4). However, the meta-analysisshowed F&V intervention did not have a significant effect onIL-6 levels (MD: –0.12; 95% CI: –0.31, 0.08; P = 0.2; I2 = 0)(Figure 5). Changes in immune cell populations following F&Vconsumption were also analyzed. F&V intake was associatedwith a higher percentage of γ δ-T cells (MD: 1.68; 95% CI:0.29, 3.07; P = 0.02; I2 = 72) (Figure 6), although F&V intakehad no significant effect on other immune cell populations(Figures 7 and 8). No meta-analysis was possible on theobservational research, as the studies used heterogeneousmethods in reporting outcomes of interest.
DISCUSSION
This systematic review and meta-analysis supports the epi-demiologic evidence that higher consumption of F&V is asso-ciated with reduced inflammation. We found that the majority of
studies (n = 56; 71%) reported protective effects of F&V intakeon inflammation or immune function. The meta-analysis showedthat higher intake of F&V is associated with lower levels of CRPand TNF-α (P < 0.05) (Figure 9). The effect of F&V intake onimmune cell function was less clear due to the limited numberof available studies; however, meta-analysis showed that higherconsumption of F&V was associated with an increased γ δ-T cellpopulation (P < 0.05) (Figure 10).
Chronic systemic inflammation leads to reduced human lifespan, because it increases the prevalence of various diseasespredominantly found in developed countries, including obesity(4), diabetes (50), and some cancers (63). The ability of F&Vto protect against the development of these chronic conditionsis well documented (3, 16, 17, 89, 90). The mechanisms of pro-tection are not yet clear; however, there are numerous beneficialnutrients in F&V that have antioxidant and anti-inflammatoryproperties, such as carotenoids, vitamin C, vitamin E, flavonoids,and soluble fiber. Individually these nutrients have been shownto be protective against systemic inflammation. For example, across-sectional study (77) of 3254 elderly participants found aninverse association between vitamin C intake and plasma CRPlevels. Similarly, in another study, high plasma vitamin C levelswere correlated with a reduced circulatory CRP concentration(91). Another study reported that supplementation with vitaminE (800 IU/d) significantly decreased CRP levels in diabeticpatients (62). Hence, it is likely that the positive effects ofF&V can be attributed to the combination of these nutrients andsubsequent displacement of proinflammatory nutrients from thediet, such as dietary fat.
Epidemiologic data across the general population has demon-strated that plasma lycopene and β-carotene were inverselyassociated with inflammatory biomarkers, including CRP and
FIGURE 9 Overall effects of fruit and vegetable intake on inflammatory biomarkers. CRP, C-reactive protein; N, number of participants included in themeta-analysis; n, number of studies included in the meta-analysis.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
152 HOSSEINI ET AL.
FIGURE 10 Overall effects of fruit and vegetable intake on different immune cell population percentage. N, number of participants included in the meta-analysis; n, number of studies included in the meta-analysis; NK cells, natural killer cells.
sICAM (92). In addition, consumption of carotenoid-rich foodhas been shown to reduce systemic inflammation in overweightand obese people (4). In another study, lycopene inhibited TNF-α–induced endothelial ICAM expression (93). Flavonoids area group of polyphenols found in F&V with antioxidant andanti-inflammatory effects (94). Short-term supplementation ofpure flavonoids, including quercetin and epicatechin, decreasedIL-1β and E-selection (95). The exact mechanism by whichflavonoids inhibit inflammation is not clear. It has been suggestedthat quercetin, in particular, inhibits both cyclo-oxygenase andlipoxygenase activities, thus diminishing the formation of inflam-matory metabolites (96). Another immune-protective compoundof F&V is fiber. Recently, we reported that dietary fiber was in-versely associated with airway inflammation (97). In line withthese findings, several studies in the general population reportedthat dietary fiber intake was inversely related to systemic inflam-mation (98, 99). It has been suggested that dietary soluble fiber in-take can modulate inflammatory responses, due to production ofshort-chain fatty acids, which activate the free fatty acid receptorsG protein–coupled receptor 41 and 43 (GPR41 and GPR43), andinhibit histone deactylases, with both of thesemechanisms havinganti-inflammatory actions (100, 101). Since various phytochem-icals have mutually beneficial effects, it has been suggested thatcombinations of such nutrients, found in F&V, can enhance theirroles in immune responses to inflammation (102).
To our knowledge, the present systematic literature reviewand meta-analysis is the first to assess the effects of F&V con-sumption on inflammation and immune function. Previous meta-analyses have investigated the effects of F&V intake on the riskof chronic diseases such as type 2 diabetes (89, 90), and coronaryheart disease (16). A meta-analysis of 5 clinical trials (103) as-sessed the effects of pomegranate juice consumption on plasmaCRP levels, and showed no significant effect, which may be aresult of the small number of studies.
The strengths of this review include the broad systematic lit-erature search, defined inclusion criteria and clear approach tocollecting data, thorough assessment of the evidence, inclusionof studies with various designs, and meta-analysis. Furthermore,all studies were assessed for quality and validity.
However, our study has some limitations worth noting that arecommon to analysis of dietary interventions. First, the includedstudies were heterogeneous with respect to the intervention or thediet exposure, design, study size, and duration of follow-up. Somestudies assessed the effects of specific types of F&V, whereas oth-ers investigated the effects of total F&V intake. However, this
limitation has been addressed by presenting the results accordingto study exposure. The meta-analysis was also performed basedon the intervention employed. Second, dissimilar populations interms of age and health condition were observed among the stud-ies. Some studies were performed in healthy populations (11, 14,25, 34), whereas other studies were conducted in patients withtype 2 diabetes (50), metabolic syndrome (27), high blood pres-sure (26), high risk of CVD (30), or in other conditions such aspregnancy (60). Due to the limited number of studies availablefor inclusion in the meta-analysis, separating the analysis accord-ing to health status was not possible. Moreover, different studiesused different methods of dietary assessments, such as 24-h re-call, food record, or food-frequency questionnaire. Finally, thepossibility of publication bias exists, as studies with positive find-ings are more likely to be published.
In conclusion, this systematic review of the literature suggeststhat a higher intake of F&V decreases inflammation, and also en-hances immune cell populations. These findings support recom-mendations to increase F&V intake for the primary prevention ofmany chronic diseases. However, well-designed randomized con-trolled trials are needed to confirm our results. Future research isalso required to explore potential mechanisms underlying the ob-served associations.
The authors’ responsibilities were as follows—BH, BSB, and LGW: de-signed the study; BH and AS: conducted the search; BH: analyzed the dataand wrote the paper; BSB, AS, and LGW: revised the paper; MRS, AC, andPABW: reviewed the paper; LGW: had primary responsibility for the finalcontent; and all authors: read and approved the final manuscript. None of theauthors reported a conflict of interest.
REFERENCES1. Baldrick F, Elborn J, Woodside J, Treacy K, Bradley J, Patterson
C, Schock B, Ennis M, Young I, McKinley M. Effect of fruit andvegetable intake on oxidative stress and inflammation in COPD: arandomised controlled trial. Eur Respir J 2012;39:1377–84.
2. Hosseini B, Saedisomeolia A, Wood LG, Yaseri M, Tavasoli S.Effects of pomegranate extract supplementation on inflammation inoverweight and obese individuals: a randomized controlled clinicaltrial. Complement Ther Clin Pract 2016;22:44–50.
3. Jin Y, Cui X, Singh U, Chumanevich A, Harmon B, CavicchiaP, Hofseth A, Kotakadi V, Stroud B, Volate S et al. Systemicinflammatory load in humans is suppressed by consumption of twoformulations of dried, encapsulated juice concentrate. Mol Nutr FoodRes 2010;54:1506–14.
4. Ghavipour M, Saedisomeolia A, Djalali M, Sotoudeh G, EshraghyanM, Moghadam A, Wood L. Tomato juice consumption reducessystemic inflammation in overweight and obese females. Br J Nutr2013;109:2031–5.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 153
5. Gibson A, Edgar J, Neville C, Gilchrist S, McKinley M, Patterson C,Young I, Woodside J. Effect of fruit and vegetable consumption onimmune function in older people: a randomized controlled trial. Am JClin Nutr 2012;96:1429–36.
6. Macready A, George T, Chong M, Alimbetov D, Jin Y, Vidal A,Spencer J, KennedyO, TuohyK,Minihane A et al. Flavonoid-rich fruitand vegetables improve microvascular reactivity and inflammatorystatus in men at risk of cardiovascular disease—FLAVURS: arandomized controlled trial. Am J Clin Nutr 2014;99:479–89.
7. Wintergerst ES, Maggini S, Hornig DH. Immune-enhancing role ofvitamin C and zinc and effect on clinical conditions. Ann Nutr Metab2006;50:85–94.
8. Bub A, Watzl B, Blockhaus M, Briviba K, Liegibel U, Müller H,Pool-Zobel B, Rechkemmer G. Fruit juice consumption modulatesantioxidative status, immune status and DNA damage. J Nutr Biochem2003;14:90–8.
9. Watzl B, Bub A, Briviba K, Rechkemmer G. Supplementation of alow-carotenoid diet with tomato or carrot juice modulates immunefunctions in healthy men. Ann Nutr Metab 2003;47:255–61.).
10. Watzl B, Bub A, Brandstetter BR, Rechkemmer G. Modulation ofhuman T-lymphocyte functions by the consumption of carotenoid-richvegetables. Br J Nutr 1999;82:383–9.
11. Freese R, Vaarala O, Turpeinen A, Mutanen M. No difference inplatelet activation or inflammation markers after diets rich or poorin vegetables, berries and apple in healthy subjects. Eur J Nutr2004;43:175–82.
12. McCall D, McGartland C, McKinley M, Sharpe P, McCance D,Young I, Woodside J. The effect of increased dietary fruit andvegetable consumption on endothelial activation, inflammation andoxidative stress in hypertensive volunteers. NutrMetab Cardiovasc Dis2011;21:658–64.
13. Lamprecht M, Ottl K, Schwaberger G, Hofmann P, Greilberger J.[Supplementation with a fruit and vegetable concentrate reduces theoxidative protein damage and the concentration of the cytokine TNF-alpha]. Dtsch Z Sportmed 2007;58:203.
14. Briviba K, Kulling SE, Moseneder J, Watzl B, Rechkemmer G, BubA. Effects of supplementing a low-carotenoid diet with a tomatoextract for 2 weeks on endogenous levels of DNA single strandbreaks and immune functions in healthy non-smokers and smokers.Carcinogenesis 2004;25:2373–8.
15. Watzl B, Bub A, Blockhaus M, Herbert BM, Luhrmann PM,Neuhauser-Berthold M, Rechkemmer G. Prolonged tomato juiceconsumption has no effect on cell-mediated immunity of well-nourished elderly men and women. J Nutr 2000;130(7):1719–23.
16. GanY, TongX, Li L, Cao S, YinX, GaoC, Herath C, LiW, Jin Z, ChenY et al. Consumption of fruit and vegetable and risk of coronary heartdisease: a meta-analysis of prospective cohort studies. Int J Cardiol2015;183:129–37.
17. Wang PY, Fang JC, Gao ZH, Zhang C, Xie SY. Higher intake of fruits,vegetables or their fiber reduces the risk of type 2 diabetes: a meta-analysis. J Diabetes Investig 2016;7:56–69.
18. American Dietetic Association [Internet]. Availabe from:http://andevidencelibrary.com/files/Docs/2012_Jan_EA_Manual.pdf[cited August 2016].
19. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting itemsfor systematic reviews and meta-analyses: the PRISMA statement. JClin Epidemiol 2009;62:1006–12.
20. Holt EM, Steffen LM, Moran A, Basu S, Steinberger J, Ross JA, HongC-P, Sinaiko AR. Fruit and vegetable consumption and its relation tomarkers of inflammation and oxidative stress in adolescents. J AmDietAssoc 2009;109:414–21.
21. Kruzich LA, Marquis GS, Wilson CM, Stephensen CB. HIV-infectedUS youth are at high risk of obesity and poor diet quality: a challengefor improving short- and long-term health outcomes. J Am Diet Assoc2004;104:1554–60.
22. Almeida-de-Souza J, Santos R, Lopes L, Abreu S, Moreira C, PadraoP, Mota J, Moreira P. Associations between fruit and vegetable varietyand low-grade inflammation in Portuguese adolescents from LabMedPhysical Activity Study. Eur J Nutr 2017;110:1650–55.
23. Romieu I, Barraza-Villarreal A, Escamilla-Núñez C, Texcalac-Sangrador J, Hernandez-Cadena L, Díaz-Sánchez D, Batlle J, Rio-Navarro B. Dietary intake, lung function and airway inflammationin Mexico City school children exposed to air pollutants. Respir Res2009;10:122.
24. Kontiokari T, Salo J, Eerola E, Uhari M. Cranberry juice and bacterialcolonization in children—a placebo-controlled randomized trial. ClinNutr 2005;24:1065–72.
25. Amagase H, Sun B, Nance D. Immunomodulatory effects of astandardized Lycium barbarum fruit juice in Chinese older healthyhuman subjects. J Med Food 2009;12:1159–65.
26. Asgary S, Keshvari M, Sahebkar A, Hashemi M, Rafieian-Kopaei M.Clinical investigation of the acute effects of pomegranate juice onblood pressure and endothelial function in hypertensive individuals.ARYA Atheroscler 2013;9:326–31.
27. Barona J, Blesso C, Andersen C, Park Y, Lee J, Fernandez M. Grapeconsumption increases anti-inflammatory markers and upregulatesperipheral nitric oxide synthase in the absence of dyslipidemias in menwith metabolic syndrome. Nutrients 2012;4:1945–57.
28. Basu A, Betts NM, Nguyen A, Newman ED, Fu D, Lyons TJ. Freeze-dried strawberries lower serum cholesterol and lipid peroxidation inadults with abdominal adiposity and elevated serum lipids. J Nutr2014;144:830–7.
29. Bell P, Walshe I, Davison G, Stevenson E, Howatson G.Montmorencycherries reduce the oxidative stress and inflammatory responses torepeated days high-intensity stochastic cycling. Nutrients 2014;6:829–43.
30. Buscemi S, Rosafio G, Arcoleo G, Mattina A, Canino B, MontanaM, Verga S, Rini G. Effects of red orange juice intake on endothelialfunction and inflammatory markers in adult subjects with increasedcardiovascular risk. Am J Clin Nutr 2012;95:1089–95.
31. Castilla P, Echarri R, Dávalos A, Cerrato F, Ortega H, Teruel JL, LucasMF, Gómez-Coronado D, Ortuño J, Lasunción MA. Concentrated redgrape juice exerts antioxidant, hypolipidemic, and antiinflammatoryeffects in both hemodialysis patients and healthy subjects. Am J ClinNutr 2006;84:252–62.
32. Dalgård C, Nielsen F, Morrow J, Enghusen-Poulsen H, Jonung T,Hørder M, Maat M. Supplementation with orange and blackcurrantjuice, but not vitamin E, improves inflammatory markers in patientswith peripheral arterial disease. Br J Nutr 2009;101:263–9.
33. Davidson MH, Maki KC, Dicklin MR, Feinstein SB, Witchger M,Bell M, McGuire DK, Provost JC, Liker H, Aviram M. Effects ofconsumption of pomegranate juice on carotid intima-media thicknessin men and women at moderate risk for coronary heart disease. Am JCardiol 2009;104:936–42.
34. Deopurkar R, Ghanim H, Friedman J, Abuaysheh S, Sia CL, MohantyP, Viswanathan P, Chaudhuri A, Dandona P. Differential effects ofcream, glucose, and orange juice on inflammation, endotoxin, andthe expression of Toll-like receptor-4 and suppressor of cytokinesignaling-3. Diabetes Care 2010;33:991–7.
35. Dow C, Wertheim B, Patil B, Thomson C. Daily consumption ofgrapefruit for 6 weeks reduces urine F2-isoprostanes in overweightadults with high baseline values but has no effect on plasma high-sensitivity C-reactive protein or soluble vascular cellular adhesionmolecule 1. J Nutr 2013;143:1586–92.
36. Gammon CS, Kruger R, Conlon CA, von Hurst PR, JonesB, Stonehouse W. Inflammatory status modulates plasma lipidand inflammatory marker responses to kiwifruit consumption inhypercholesterolaemic men. Nutr Metab Cardiovasc Dis 2014;24:91–9.
37. GuoH, Zhong R, Liu Y, Jiang X, Tang X, Li Z, XiaM, LingW. Effectsof bayberry juice on inflammatory and apoptotic markers in youngadults with features of non-alcoholic fatty liver disease. Nutrition2014;30:198–203.
38. Han H, Jung U, Kim H-J, Cho S-J, Kim A, Han Y, Choi M-S.Combined supplementation with grape pomace and omija fruit ethanolextracts dose-dependently improves body composition, plasma lipidprofiles, inflammatory status, and antioxidant capacity in overweightand obese subjects. J Med Food 2016;19:170–80.
39. Karlsen A, Paur I, Bøhn S, Sakhi A, Borge G, Serafini M, Erlund I,Laake P, Tonstad S, Blomhoff R. Bilberry juice modulates plasmaconcentration of NF-kappaB related inflammatory markers in subjectsat increased risk of CVD. Eur J Nutr 2010;49:345–55.
40. Kelley DS, Adkins Y, Reddy A, Woodhouse LR, Mackey BE,Erickson KL. Sweet bing cherries lower circulating concentrations ofmarkers for chronic inflammatory diseases in healthy humans. J Nutr2013;143(3):340–4.
41. Kolehmainen M, Mykkänen O, Kirjavainen P, Leppänen T, MoilanenE, Adriaens M, Laaksonen D, Hallikainen M, Puupponen-Pimiä
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
154 HOSSEINI ET AL.
R, Pulkkinen L et al. Bilberries reduce low-grade inflammation inindividuals with features of metabolic syndrome. Mol Nutr Food Res2012;56:1501–10.
42. Kuntz S, Kunz C, Herrmann J, Borsch C, Abel G, Fröhling B, DietrichH, Rudloff S. Anthocyanins from fruit juices improve the antioxidantstatus of healthy young female volunteers without affectinganti-inflammatory parameters: results from the randomised, double-blind, placebo-controlled, cross-over ANTHONIA (ANTHOcyaninsin Nutrition Investigation Alliance) study. Br J Nutr 2014;112:925–36.
43. Larmo P, Alin J, Salminen E, Kallio H, Tahvonen R. Effects of seabuckthorn berries on infections and inflammation: a double-blind,randomized, placebo-controlled trial. Eur J Clin Nutr 2008;62:1123–30.
44. Leelarungrayub J, Laskin JJ, Bloomer RJ, Pinkaew D. Consumptionof star fruit juice on pro-inflammatory markers and walking distance inthe community dwelling elderly. Archs Gerontol Geriatr 2016;64:6–12.
45. McAnulty L, Nieman D, Dumke C, Shooter L, Henson D, Utter A,Milne G, McAnulty S. Effect of blueberry ingestion on natural killercell counts, oxidative stress, and inflammation prior to and after 2.5 hof running. Appl Physiol Nutr Metab 2011;36:976–84.
46. Nishizawa M, Hara T, Miura T, Fujita S, Yoshigai E, Ue H, HayashiY, Kwon A, Okumura T, Isaka T. Supplementation with a flavanol-rich lychee fruit extract influences the inflammatory status of youngathletes. Phytother Res 2011;25:1486–93.
47. Riso P, Klimis-Zacas D, Bo C, Martini D, Campolo J, VendrameS, Møller P, Loft S, Maria R, Porrini M. Effect of a wildblueberry (Vaccinium angustifolium) drink intervention on markers ofoxidative stress, inflammation and endothelial function in humans withcardiovascular risk factors. Eur J Nutr 2013;52:949–61.
48. Sanchez-Moreno C, Cano MP, de Ancos B, Plaza L, Olmedilla B,Granado F, Martin A. High-pressurized orange juice consumptionaffects plasma vitamin C, antioxidative status and inflammatorymarkers in healthy humans. J Nutr 2003;133:2204–9.
49. Shema-Didi L, Sela S, Ore L, Shapiro G, Geron R, Moshe G,Kristal B. One year of pomegranate juice intake decreases oxidativestress, inflammation, and incidence of infections in hemodialysispatients: a randomized placebo-controlled trial. Free Radic Biol Med2012;53:297–304.
50. Sohrab G, Nasrollahzadeh J, Zand H, Amiri Z, Tohidi M, Kimiagar M.Effects of pomegranate juice consumption on inflammatory markers inpatients with type 2 diabetes: a randomized, placebo-controlled trial.J Res Med Sci 2014;19:215–20.
51. Tomé-Carneiro J, Gonzálvez M, Larrosa M, Yáñez-Gascón M,García-Almagro F, Ruiz-Ros J, García-Conesa M, Tomás-BarberánF, Espín J. One-year consumption of a grape nutraceutical containingresveratrol improves the inflammatory and fibrinolytic status ofpatients in primary prevention of cardiovascular disease. Am J Cardiol2012;110:356–63.
52. Vidal K, Bucheli P, Gao Q, Moulin J, Shen L, Wang J, Blum S,Benyacoub J. Immunomodulatory effects of dietary supplementationwith a milk-based wolfberry formulation in healthy elderly: arandomized, double-blind, placebo-controlled trial. Rejuvenation Res2012;15:89–97.
53. Hunter D, Skinner M, Wolber F, Booth C, Loh J, Wohlers M,Stevenson L, Kruger M. Consumption of gold kiwifruit reducesseverity and duration of selected upper respiratory tract infectionsymptoms and increases plasma vitamin C concentration in healthyolder adults. Br J Nutr 2012;108:1235–45.
54. Zunino S, Peerson J, Freytag T, Breksa A, Bonnel E, WoodhouseL, Storms D. Dietary grape powder increases IL-1? and IL-6production by lipopolysaccharide-activated monocytes and reducesplasma concentrations of large LDL and large LDL-cholesterolparticles in obese humans. Br J Nutr 2014;112:369–80.
55. Xie L, Vance T, Kim B, Lee SG, Caceres C, Wang Y, Hubert PA,Lee JY, ChunOK, Bolling BW.Aronia berry polyphenol consumptionreduces plasma total and low-density lipoprotein cholesterol in formersmokers without lowering biomarkers of inflammation and oxidativestress: a randomized controlled trial. Nutr Res 2017;37:67–77.
56. Kent K, Charlton K, Roodenrys S, Batterham M, Potter J, Traynor V,Gilbert H, Morgan O, Richards R. Consumption of anthocyanin-richcherry juice for 12 weeks improves memory and cognition in olderadults with mild-to-moderate dementia. Eur J Nutr 2017;56:333–41.
57. Kanellos PT, Kaliora AC, Protogerou AD, Tentolouris N, Perrea DN,Karathanos VT. The effect of raisins on biomarkers of endothelialfunction and oxidant damage; an open-label and randomizedcontrolled intervention. Food Res Int 2017;102:674–80.
58. Nilsson A, Salo I, Plaza M, Bjorck I. Effects of a mixed berrybeverage on cognitive functions and cardiometabolic risk markers; arandomized cross-over study in healthy older adults. PLoS One 2017;12:e0188173.
59. Duffey KJ, Sutherland LA. Adult consumers of cranberry juicecocktail have lower C-reactive protein levels compared withnonconsumers. Nutr Res 2015;35:118–26.
60. Aalami-Harandi R, Karamali M, Asemi Z. The favorable effects ofgarlic intake on metabolic profiles, hs-CRP, biomarkers of oxidativestress and pregnancy outcomes in pregnant women at risk forpre-eclampsia: randomized, double-blind, placebo-controlled trial. JMatern Fetal Neonatal Med 2015;28:2020–7.
61. Riso P, Vendrame S, Del Bo C, Martini D, Martinetti A, SeregniE, Visioli F, Parolini M, Porrini M. Effect of 10-day broccoliconsumption on inflammatory status of young healthy smokers. Int JFood Sci Nutr 2014;65:106–11.
62. Upritchard JE, Sutherland WH, Mann JI. Effect of supplementationwith tomato juice, vitamin E, and vitamin C on LDL oxidation andproducts of inflammatory activity in type 2 diabetes. Diabetes Care2000;23:733–8.
63. BobeG,Albert P, Sansbury L, Lanza E, SchatzkinA, ColburnN, CrossA. Interleukin-6 as a potential indicator for prevention of high-riskadenoma recurrence by dietary flavonols in the polyp prevention trial.Cancer Prev Res (Phila) 2010;3:764–75.
64. Hunter D, Brown R, Green T, Thomson C, Skeaff M, Williams S,Todd J, Lister C, McGhie T, Zhang J et al. Changes in markers ofinflammation, antioxidant capacity and oxidative stress in smokersfollowing consumption of milk, and milk supplemented with fruit andvegetable extracts and vitamin C. Int J Food Sci Nutr 2012;63:90–102.
65. Inserra PF, Jiang S, Solkoff D, Lee J, Zhang Z, Xu M, HesslinkR Jr, Wise J, Watson RR. Immune function in elderly smokers andnonsmokers improves during supplementation with fruit and vegetableextracts. Integr Med 1999;2:3–10.
66. Knab A, Nieman D, Gillitt N, Shanely R, Cialdella-Kam L, HensonD, Sha W. Effects of a flavonoid-rich juice on inflammation, oxidativestress, and immunity in elite swimmers: a metabolomics-basedapproach. Int J Sport Nutr Exerc Metab 2013;23:150–60.
67. Knab A, Nieman D, Gillitt N, Shanely R, Cialdella-Kam L, HensonD, Sha W, Meaney M. Effects of a freeze-dried juice blend powder onexercise-induced inflammation, oxidative stress, and immune functionin cyclists. Appl Physiol Nutr Metab 2014;39:381–5.
68. Lamprecht M, Obermayer G, Steinbauer K, Cvirn G, HofmannL, Ledinski G, Greilberger J, Hallstroem S. Supplementation witha juice powder concentrate and exercise decrease oxidation andinflammation, and improve the microcirculation in obese women:randomised controlled trial data. Br J Nutr 2013;110:1685–95.
69. Nadeem N, Woodside J, Neville C, McCall D, McCance D, EdgarD, Young I, McEneny J. Serum amyloid A-related inflammation islowered by increased fruit and vegetable intake, while high-sensitiveC-reactive protein, IL-6 and E-selectin remain unresponsive. Br J Nutr2014;112:1129–36.
70. Nantz M, Rowe C, Nieves C, Percival S. Immunity and antioxidantcapacity in humans is enhanced by consumption of a dried,encapsulated fruit and vegetable juice concentrate. J Nutr2006;136:2606–10.
71. Watzl B, Kulling SE, Moseneder J, Barth SW, Bub A. A 4-wkintervention with high intake of carotenoid-rich vegetables and fruitreduces plasma C-reactive protein in healthy, nonsmoking men. Am JClin Nutr 2005;82:1052–8.
72. Williams EJ, Baines KJ, Berthon BS, Wood LG. Effects of anencapsulated fruit and vegetable juice concentrate on obesity-inducedsystemic inflammation: a randomised controlled trial. Nutrients2017;9. doi:10.3390/nu9020116.
73. Wood LG, Garg ML, Smart JM, Scott HA, Barker D, Gibson PG.Manipulating antioxidant intake in asthma: a randomized controlledtrial. Am Journal Clin Nutr 2012;96:534–43.
74. Hermsdorff HHM, Zulet MA, Puchau B, Martínez JA. Fruit andvegetable consumption and proinflammatory gene expression fromperipheral blood mononuclear cells in young adults: a translationalstudy. Nutr Metab 2010;7:11p–p.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018
FRUIT AND VEGETABLE INTAKE AND IMMUNE RESPONSES 155
75. Lopez-Garcia E, Schulze MB, Fung TT, Meigs JB, Rifai N, MansonJE, Hu FB. Major dietary patterns are related to plasma concentrationsof markers of inflammation and endothelial dysfunction. Am J ClinNutr 2004;80:1029–35.
76. Root MM,McGinnMC, Nieman DC, Henson DA, Heinz SA, ShanelyRA,KnabAM, Jin F. Combined fruit and vegetable intake is correlatedwith improved inflammatory and oxidant status from a cross-sectionalstudy in a community setting. Nutrients 2012;4:29–41.
77. Wannamethee SG, Lowe GDO, Rumley A, Bruckdorfer KR,WhincupPH. Associations of vitamin C status, fruit and vegetable intakes,and markers of inflammation and hemostasis. Am J Clin Nutr2006;83:567–730.
78. McAnulty L, Collier S, Landram M, Whittaker D, Isaacs S, KlemkaJ, Cheek S, Arms J, McAnulty S. Six weeks daily ingestion ofwhole blueberry powder increases natural killer cell counts andreduces arterial stiffness in sedentary males and females. Nutr Res2014;34:577–84.
79. Nantz M, Rowe C, Muller C, Creasy R, Colee J, Khoo C, Percival S.Consumption of cranberry polyphenols enhances human γ δ-T cellsproliferation and reduces the number of symptoms associated withcolds and influenza: a randomized, placebo-controlled interventionstudy. Nutr J 2013;12:161.
80. Rowe C, NantzM, Nieves C,West R, Percival S. Regular consumptionof concord grape juice benefits human immunity. J Med Food2011;14:69–78.
81. Zunino S, Storms D, Freytag T, Mackey B, Zhao L, Gouffon J,Hwang D. Dietary strawberries increase the proliferative response ofCD3/CD28-activated CD8+ T cells and the production of TNF-α inlipopolysaccharide-stimulated monocytes from obese human subjects.Br J Nutr 2013;110:2011–9.
82. Bøhn S, Myhrstad M, Thoresen M, Holden M, Karlsen A, TunheimS, Erlund I, Svendsen M, Seljeflot I, Moskaug J et al. Blood cellgene expression associated with cellular stress defense is modulatedby antioxidant-rich food in a randomised controlled clinical trial ofmale smokers. BMC Med 2010;8:54.
83. Murashima M, Watanabe S, Zhuo XG, Uehara M, Kurashige A. Phase1 study of multiple biomarkers for metabolism and oxidative stressafter one-week intake of broccoli sprouts. Biofactors 2004;22:271–5.
84. Nantz MP, Rowe CA, Muller CE, Creasy RA, Stanilka JM, PercivalSS. Supplementation with aged garlic extract improves both NK andγ δ-T cell function and reduces the severity of cold and flu symptoms:a randomized, double-blind, placebo-controlled nutrition intervention.Clin Nutr 2012;31:337–44.
85. Liu C, Wang C, Robison E, Levine AM, Gandhi M, Schwartz R,Weber KM, Merenstein D. Short-term garlic supplementation andhighly active antiretroviral treatment adherence, CD4+ cell counts,and human immunodeficiency virus viral load. Altern Ther HealthMed 2012;18:18–22.
86. Winkler P, Ellinger S, Boetzer A, Arendt B, Berthold H, RockstrohJ, Spengler U, Goerlich R. Lymphocyte proliferation and apoptosis inHIV-seropositive and healthy subjects during long-term ingestion offruit juices or a fruit-vegetable-concentrate rich in polyphenols andantioxidant vitamins. Eur J Clin Nutr 2004;58:317–25.
87. Hendricks KM, Mwamburi DM, Newby P, Wanke CA. Dietarypatterns and health and nutrition outcomes in men living with HIVinfection. Am J Clin Nutr 2008;88:1584–92.
88. Nettleton JA, Matijevic N, Follis JL, Folsom AR, BoerwinkleE. Associations between dietary patterns and flow cytometry-
measured biomarkers of inflammation and cellular activation in theAtherosclerosis Risk in Communities (ARIC) Carotid Artery MRIStudy. Atherosclerosis 2010;212:260–7.
89. Carter P, Gray LJ, Troughton J, Khunti K, Davies MJ. Fruitand vegetable intake and incidence of type 2 diabetes mellitus:systematic review and meta-analysis. Br Med J (Clin Res Ed) 2010;341:c4229.
90. Li M, Fan Y, Zhang X, HouW, Tang Z. Fruit and vegetable intake andrisk of type 2 diabetes mellitus: meta-analysis of prospective cohortstudies. BMJ Open 2014;4:e005497.
91. Block G, Jensen C, Dietrich M, Norkus EP, Hudes M, PackerL. Plasma C-reactive protein concentrations in active and passivesmokers: influence of antioxidant supplementation. J Am Coll Nutr2004;23:141–7.
92. van Herpen-Broekmans WM, Klopping-Ketelaars IA, Bots ML, KluftC, Princen H, Hendriks HF, Tijburg LB, van Poppel G, KardinaalAF. Serum carotenoids and vitamins in relation to markers ofendothelial function and inflammation. Eur J Epidemiol 2004;19:915–21.
93. Hung CF, Huang TF, Chen BH, Shieh JM, Wu PH, Wu WB.Lycopene inhibits TNF-alpha-induced endothelial ICAM-1 expressionand monocyte-endothelial adhesion. Eur J Pharmacol 2008;586:275–82.
94. Gonzalez-Gallego J, Garcia-Mediavilla MV, Sanchez-Campos S,Tunon MJ. Fruit polyphenols, immunity and inflammation. Br J Nutr2010;104 Suppl 3:S15–27.
95. Dower JI, Geleijnse JM, Gijsbers L, Schalkwijk C, Kromhout D,Hollman PC. Supplementation of the pure flavonoids epicatechin andquercetin affects some biomarkers of endothelial dysfunction andinflammation in (pre)hypertensive adults: a randomized double-blind,placebo-controlled, crossover trial. J Nutr 2015;145:1459–63.
96. Nijveldt RJ, van Nood E, van Hoorn DE, Boelens PG, van Norren K,van Leeuwen PA. Flavonoids: a review of probable mechanisms ofaction and potential applications. Am J Clin Nutr 2001;74:418–25.
97. Berthon BS, Macdonald-Wicks LK, Gibson PG, Wood LG.Investigation of the association between dietary intake, diseaseseverity and airway inflammation in asthma. Respirology2013;18:447–54.
98. Ma Y, Griffith JA, Chasan-Taber L, Olendzki BC, Jackson E, StanekEJ 3rd, Li W, Pagoto SL, Hafner AR, Ockene IS. Associationbetween dietary fiber and serum C-reactive protein. Am J Clin Nutr2006;83:760–6.
99. Ajani UA, Ford ES, Mokdad AH. Dietary fiber and C-reactive protein:findings from national health and nutrition examination survey data. JNutr 2004;134:1181–5.
100. Halnes I, Baines KJ, Berthon BS, MacDonald-Wicks LK, GibsonPG, Wood LG. Soluble fibre meal challenge reduces airwayinflammation and expression of GPR43 and GPR41 in asthma.Nutrients 2017;9(1):50–2.
101. McLoughlin RF, BerthonBS, JensenME, BainesKJ,WoodLG. Short-chain fatty acids, prebiotics, synbiotics, and systemic inflammation: asystematic review and meta-analysis. Am J Clin Nutr 2017.
102. Wood LG, Gibson PG. Dietary factors lead to innate immuneactivation in asthma. Pharmacol Ther 2009;123:37–53.
103. Sahebkar A, Gurban C, Serban A, Andrica F, Serban M-C. Effects ofsupplementation with pomegranate juice on plasma C-reactive proteinconcentrations: a systematic review and meta-analysis of randomizedcontrolled trials. Phytomedicine 2016;23:1095–102.
Downloaded from https://academic.oup.com/ajcn/advance-article-abstract/doi/10.1093/ajcn/nqy082/5042153by MIT Libraries useron 22 June 2018