nu ien s
A icle
Ul a-P ocessed Food Consump ion is Associa ed wi h Renal
Func ion Decline in Olde Adul s: A P ospec i e Coho S udy
Jimena Rey-Ga cía1,2 , Ca olina Dona -Va gas 1,3,4,*, Helena Sando al-Insaus i 1,5, Ana Bayan-B a o 6,
Belén Mo eno-F anco 7,8, JoséRamón Banegas 1, Fe nando Rod íguez-A alejo 1,3 and
Pila Gualla -Cas illón1,3


Ci a ion: Rey-Ga cía, J.;
Dona -Va gas, C.; Sando al-Insaus i,
H.; Bayan-B a o, A.; Mo eno-F anco,
B.; Banegas, J.R.; Rod íguez-A alejo,
F.; Gualla -Cas illón, P.
Ul a-P ocessed Food Consump ion
is Associa ed wi h Renal Func ion
Decline in Olde Adul s: A
P ospec i e Coho S udy. Nu ien s
2021,13, 428. h ps://doi.o g/
10.3390/nu13020428
Academic Edi o : Alessand o Leone
Recei ed: 5 Janua y 2021
Accep ed: 25 Janua y 2021
Published: 28 Janua y 2021
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2021 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
1Depa men o P e en i e Medicine and Public Heal h, School o Medicine, Uni e sidad Au ónoma de
Mad id-IdiPaz, CIBERESP (CIBER o Epidemiology and Public Heal h), 28029 Mad id, Spain;
jimena. eyga [email p o ec ed] (J.R.-G.); [email p o ec ed] (H.S.-I.); [email p o ec ed] (J.R.B.);
[email p o ec ed] (F.R.-A.); mpila [email p o ec ed] (P.G.-C.)
2In e nal Medicine Depa men , Ramón y Cajal Uni e si y Hospi al, 28034 Mad id, Spain
3IMDEA-Food Ins i u e, CEI UAM+CSIC, 28049 Mad id, Spain
4Uni o Nu i ional Epidemiology, Ins i u e o En i onmen al Medicine, Ka olinska Ins i u e ,
171 77 S ockholm, Sweden
5Depa men o Nu i ion, Ha a d T.H. Chan School o Public Heal h, Bos on, MA 02115, USA
6Depa men o Nu i ion, 12 de Oc ub e Hospi al, 28041 Mad id, Spain; [email p o ec ed]
7Depa men o Mic obiology, Radiology, Pedia ics and Public Heal h, Uni e sidad de Za agoza,
50009 Za agoza, Spain; [email p o ec ed].es
8Ins i u o de In es igación Sani a ia A agón, Hospi al Uni e si a io Miguel Se e , 50009 Za agoza, Spain
*Co espondence: ca [email p o ec ed]g
Abs ac :
Ul a-p ocessed ood (UPF) consump ion has been associa ed wi h inc eased isk o
ca dio ascula isk ac o s and mo ali y. Howe e , li le is known on he UPF e ec on enal unc ion.
The aim o his s udy is o assess p ospec i ely he associa ion be ween consump ion o UPF and enal
unc ion decline. This is a p ospec i e coho s udy o 1312 communi y-dwelling indi iduals aged 60
and olde ec ui ed du ing 2008–2010 and ollowed up o Decembe 2015. A baseline, a alida ed
die a y his o y was ob ained. UPF was iden i ied acco ding o NOVA classi ica ion. A baseline
and a ollow-up, se um c ea inine (SC ) and es ima ed glome ula il a ion a e (eGFR) le els we e
asce ained and changes we e calcula ed. A combined end-poin o enal decline was conside ed: SC
inc ease o eGFR dec eased beyond ha expec ed o age. Logis ic eg ession wi h adjus men o
po en ial con ounde s was pe o med. Du ing ollow-up, 183 cases o enal unc ion decline occu ed.
The ully adjus ed odds a ios (95% CI) o enal unc ion decline ac oss e ciles o pe cen age o
o al ene gy in ake om UPF we e 1.56 (1.02–2.38) o he second e cile, and 1.74 (1.14–2.66) o
he highes e cile; p- end was 0.026. High UPF consump ion is independen ly associa ed wi h an
inc ease highe han 50% in he isk o enal unc ion decline in Spanish olde adul s.
Keywo ds:
ul a-p ocessed ood; c ea inine se um le els; glome ula il a ion a e; enal unc ion decline
1. In oduc ion
Renal unc ion shows a s eady decline when ageing [
1
]. This decline migh be in-
c eased unde di e en ci cums ances (such as he p esence o ca dio ascula isk ac o s),
e en leading o he de elopmen o a Ch onic Kidney Disease (CKD). [
2
] CKD a ec s
10% o he wo ld’s popula ion [
3
] and anks in he op en non-communicable diseases
con ibu ing o disabili y and p ema u e dea h [
4
]. CKD is linked o high heal h ca e
cos s, a poo quali y o li e, se ious ad e se heal h ou comes [
5
,
6
] such as ca dio ascula
disease, enal ailu e equi ing eplacemen he apy, in ec ion, o dep ession, as well as
mo ali y [
3
]. O e he las decade, a 41.5% inc ease in CKD mo ali y has been obse ed
wo ldwide [
7
]. The e o e, he decline in he enal unc ion has subs an ial clinical and
he apeu ic consequences among he elde ly, as well as public heal h ele ance.
Nu ien s 2021,13, 428. h ps://doi.o g/10.3390/nu13020428 h ps://www.mdpi.com/jou nal/nu ien s
Nu ien s 2021,13, 428 2 o 12
The ex en o which age- ela ed enal decline is p e en able emains con o e sial [
8
].
While die a y changes a e one o he key modi iable isk ac o s o p og ession o CKD [
9
],
hey a e also use ul in he p e en ion o obesi y, hype ension, and diabe es [
10
]. In ac , a
ecen me a-analysis o coho s udies has shown ha a high adhe ence o a heal hy die a y
pa e n ( ich in whole g ains, ege ables, ui , legumes, nu s, and ish, while low in ed
and p ocessed mea , sodium, and suga swee ened be e ages) migh p e en CKD [11].
Despi e he e idence o he bene i s o a heal hy die , he e has been a apid wes e n-
iza ion o ou die , including a global inc ease in he consump ion o ul a-p ocessed ood
(UPF) [
12
,
13
]. In some high-income coun ies, he cu en consump ion o UPF accoun s
o mo e han 50% o he o al ene gy in ake [
14
,
15
]. In Spain, se e al s udies ha e shown
ha UPF consump ion ep esen s 24% o o al ene gy in ake [
16
], which could be om 1.5
o 5 se ings pe day [17].
UPF is o mula ed mos ly o en i ely wi h subs ances de i ed om ood wi h li le, i
any, o he o iginal ood emaining. Indus ial p ocessingalso in ol es he addi ion o chem-
icals o imp o e shel -li e and he o ganolep ic cha ac e is ics o UPF [
18
]. Compa ed wi h
o he ood g oups, UPF is ypically low-cos , eady o consume, and hype -pala able [
19
].
Mo eo e , UPF is low in ibe and mic onu ien s, while i is high in e ined ca bohyd a es,
added suga s, sa u a ed, and ans a y acids, sodium, and addi i es [20–22].
Sys ema ic e iews o he li e a u e ha e shown a posi i e associa ion be ween UPF
consump ion and ad e se heal h ou comes [
19
,
23
], including obesi y [
24
,
25
], diabe es [
26
],
hype ension [
27
], and mo ali y [
16
,
17
,
28
,
29
]. Howe e , o ou knowledge, no p e ious
s udy has e alua ed he ole o UPF consump ion enal unc ion decline. The e o e, his
s udy aimed o p ospec i ely assess he associa ion be ween UPF consump ion and he
isk o enal unc ion decline in he Senio s-ENRICA-1 s udy, which is a popula ion-based
coho o olde adul s om Spain. We ocused speci ically on olde adul s, whe e UPF
consump ion is inc easing and enal unc ion wo sens wi h age.
2. Ma e ials and Me hods
2.1. S udy Design and Pa icipan s
Da a we e collec ed om he Senio s-ENRICA-1 coho , which was es ablished du ing
2008–2010. This is a ep esen a i e coho o he non-ins i u ionalized popula ion aged 60
and olde in Spain, whose me hods ha e been p e iously epo ed [
30
,
31
]. A baseline, a
compu e -assis ed phone in e iew was pe o med o ob ain in o ma ion on sociodemo-
g aphic ac o s, li es yle, and mo bidi y. In addi ion, wo subsequen home isi s we e
pe o med o collec blood and u ine samples, o ob ain a die a y his o y, and o conduc a
physical examina ion. Pa icipan s we e ollowed-up un il 2015, when ano he wa e o
da a collec ion was ca ied ou . Pa icipan s who we e epo ed dead o declined ano he
in e iew o we e o ha e a blood sample collec ed we e excluded. S udy pa icipan s ga e
w i en in o med consen . The Clinical Resea ch E hics Commi ee o La Paz Uni e si y
Hospi al in Mad id (Spain) app o ed he s udy.
2.2. S udy Va iables
2.2.1. Die and Co a iables
A baseline, in o ma ion on die was collec ed h ough a alida ed compu e ized
ace- o- ace die a y his o y (DH-ENRICA), de eloped om ha used in he EPIC (Eu o-
pean P ospec i e In es iga ion in o Cance and Nu i ion) coho s udy in Spain [
32
,
33
].
Pa icipan s we e asked abou he ood consumed in a ypical week o he p eceding yea .
The HD-ENRICA egis e s 860 oods and 24 di e en cooking me hods, and uses 120 se s
o pho og aphs o help in es ima ing he po ion sizes. T ained and ce i ied in e iewe s
pe o med he da a collec ion. The in ake o mac o and mic onu ien s was es ima ed using
s anda d ood composi ion ables o Spain.
A baseline, sel - epo ed in o ma ion was ob ained on sex, age, educa ional le el (no
o mal educa ion o p ima y, seconda y, and uni e si y), smoking s a us (ne e , o me ,
and cu en smoke s), and o me -d inke s a us. Physical ac i i y was asce ained a
Nu ien s 2021,13, 428 3 o 12
baseline wi h he ques ionnai e de eloped by he EPIC g oup and was exp essed in
me abolic equi alen s (MET)–hou /week [
34
], and he numbe o hou s wa ching TV
pe week was sel - epo ed. Pa icipan s epo ed he ollowing physician-diagnosed
ch onic condi ions: co ona y hea disease, hea ailu e, s oke, ch onic espi a o y disease,
cance , os eoa h i is, and dep ession equi ing ea men . A nu se checked he numbe
o he epo ed medica ions used agains d ug packages. Hype ension was de ined as
ha ing a sys olic blood p essu e
≥
140 mmHg o a dias olic blood p essu e
≥
90 mmHg o
an ihype ensi e ea men . Diabe es was de ined as ha ing a as ing glucose
≥
126 mg/dl
o an idiabe ic ea men . Hype choles e olemia was de ined as ha ing a o al choles e ol
≥
200 mg/dl o d ug ea men . Finally, weigh and heigh we e measu ed a home unde
s anda dized condi ions, and body mass index (BMI) was calcula ed as weigh in kg
di ided by heigh in m squa ed.
Exposu e Assessmen and NOVA Classi ica ion
Foods consumed we e ca ego ized in o ou g oups acco ding o he NOVA classi i-
ca ion based on he ex en and pu pose o indus ial ood p ocessing [
35
,
36
]. In b ie , he
i s g oup includes unp ocessed o minimally p ocessed ood such as ui and ege ables,
g ains, nu s and seeds, esh and pas eu ized milk, and na u al yogu wi h no added
suga s o a i icial swee ene s. The second g oup comp ises p ocessed culina y ing edien s
(sal , suga , honey, ege able oils, bu e , la d, and inega ). The hi d g oup includes
p ocessed oods, o example canned o bo led ege ables and legumes, ui in sy up,
canned ish, unpackaged cheeses, eshly made b ead, and sal ed o suga ed nu s and
seeds. Finally, he ou h g oup consis s o UPF, which a e hose o mula ed mos ly o
en i ely om ood-de i ed subs ances con aining li le o none o he o iginal ood o m,
o example snacks, cookies, swee s, ice-c eam, pizza, ins an soup, p ocessed mea , o
so -d inks. The eco ded oods as well as hei g oup acco ding o he NOVA classi ica ion,
has been p e iously epo ed in de ail [16].
2.2.2. Renal Func ion Decline
A baseline and a he end o ollow-up, 12 h as ing blood was ex ac ed and a spo
u ine sample was p o ided du ing he home isi . Labo a o y de e mina ions we e pe -
o med cen ally a he Cen e o Biological Diagnosis o he Hospi al Clinic in Ba celona,
using s anda d p ocedu es and app op ia e quali y con ols [
30
]. Se um c ea inine (SC )
was de e mined by he Ja é, alkaline pic a e by kine ic eac ion. The es ima ed glome ula
il a ion a e (eGFR) was es ima ed om SC wi h he Ch onic Kidney Disease Epidemiol-
ogy Collabo a ion (CKD-EPI) Equa ion [
37
]. Changes in SC and eGFR le els om baseline
o he end o ollow-up we e calcula ed. Renal unc ion decline was de ined as a SC
inc eased o an eGFR dec eased beyond ha expec ed o age. Change in eGFR beyond
ha expec ed o age was calcula ed in 3 s eps: (i) eGFR based on baseline c ea inine and
age in 2015; (ii) eGFR in 2015 based on bo h SC and eGFR in 2015; and (iii) sub ac ing ii
om i.
2.3. S a is ical Analysis
The Senio s-ENRICA coho comp ises 2519 pa icipan s om he ENRICA (S udy on
Nu i ion and Ca dio ascula isk ac o s in Spain) s udy, who we e 60 o o e a baseline,
and who p o ided da a in 2015. O hem, 118 we e excluded i hei eGFR a baseline was
<60 mL/min/1.73 m
2
[
38
]. These pa icipan s we e excluded because hey could ha e
ecei ed ad ice o a oid UPF consump ion. We also excluded 19 pa icipan s wi h ex eme
da a in ene gy consump ion ( o al ene gy in ake ou o ange: 600–4200 kcal/day in men o
400–3500 in women). We excluded 1062 pa icipan s wi h missing in o ma ion on eGFR a
ollow-up, and 8 pa icipan s wi h missing co a ia e in o ma ion. Finally, he s udy sample
consis ed o 1312 pa icipan s wi hou e idence o enal unc ion impai men (Figu e 1).
Nu ien s 2021,13, 428 4 o 12
Figu e 1. Pa icipan s’ lowcha .
UPF consump ion was exp essed as he pe cen age o ene gy om UPF o o al ene gy
in ake, as well as in g/day o UPF pe kg o body weigh . Logis ic eg ession models
we e used o assess he associa ion be ween UPF consump ion, in sex-speci ic e iles,
and enal unc ion decline; esul s we e exp essed as odds a ios (OR) and hei 95%
con idence in e al (CI), using he i s (lowes ) e ile as e e ence. The p o linea end
was calcula ed modelling he UPF e iles as a con inuous a iable.
We buil ou logis ic eg ession models wi h consecu i e adjus men le els: model 1
was adjus ed o sex, age and o al ene gy in ake; model 2 was u he adjus ed o ed-
uca ional le el (no o mal educa ion o p ima y, seconda y, and uni e si y), smoking
s a us (ne e , o me , cu en smoke ), o me -d inke s a us (yes, no), physical ac i i y
(METs-hou /week), ime spen wa ching TV (hou /week), and ibe consump ion (g/day);
model 3 was u he adjus ed o he numbe o ch onic condi ions, aking in o accoun
he ollowing: co ona y hea disease, hea ailu e, s oke, ch onic espi a o y disease,
cance , os eoa h i is, and dep ession equi ing ea men (con inuous), numbe o medica-
ions used pe day (con inuous), and wo well-es ablished enal isk ac o s: hype ension
(yes/no) and diabe es (yes/no); hype choles e olemia (yes/no), and body mass index
(BMI) (con inuous).
S a is ical signi icance was se a wo-sided p< 0.05. The analyses we e pe o med
wi h S a a/SE, e sion 13.1 (S a aCo p, College S a ion, TX). This a icle ollows he ecom-
menda ions o he STROBE—Nu i ional Epidemiology ini ia i e [39–41].
3. Resul s
Among he 1312 pa icipan s (51% women; mean age 67
±
5.5), 183 cases o enal
unc ion decline occu ed by he end o he 6-yea ollow-up. The a e age pe cen ages
o ene gy consump ion om UPF o each consecu i e e ile o UPF (% o ene gy) we e:
8.6%, 18.7%, and 33.0% in men; and 6.8%, 16.2%, and 29.8% in women (Table 1). Compa ed
wi h pa icipan s in he lowes e ile o UPF consump ion hose in he highes e ile o
consump ion had a highe ene gy in ake and BMI, we e mo e equen ly ex-d inke s, and
pe o med less physical ac i i y (Table 1).
Nu ien s 2021,13, 428 5 o 12
Table 1.
Baseline cha ac e is ics o he coho pa icipan s acco ding o sex-speci ic e ciles o Ul a-p ocessed Food
Consump ion as a pe cen age o o al ene gy (% ene gy), in he Senio s-ENRICA Coho S udy (n= 1312).
Ul a-P ocessed Food Consump ion (% ene gy)
T1 (Lowes ) (n= 438) T2 (n= 438) T3 (Highes ) (n= 436) pT end
To al ene gy (kcal/day),
mean ±SD 1948 ±549 2053 ±565 2161 ±569 <0.001
Ul a-p ocessed ood consump ion
(% ene gy), mean ±SD 7.7 ±3.5 17.5 ±3.0 31.5 ±7.7 <0.001
Ul a-p ocessed ood consump ion
(g ams pe day), mean ±SD 128 ±99 251 ±141 379 ±177 <0.001
Weigh (kg), mean ±SD 73.7 ±13 74.8 ±13 76.0 ±13.0 0.044
Ul a-p ocessed ood consump ion
(g/kg), mean ±SD 1.8 ±1.3 3.4 ±1.9 5.1 ±2.5 <0.001
Age, yea s, mean ±SD 67.4 ±5.5 67 ±5.2 67 ±5.8 0.823
Educa ional le el, % 0.735 †
No o mal educa ion o p ima y 23.7 23.9 23.9
Seconda y 25.6 25.8 29.1
Uni e si y 50.7 50.2 47.0
Smoking s a us, % 0.356 †
Ne e smoke 57.8 58.9 54.6
Fo me smoke 32.2 28.3 32.1
Cu en Smoke 10.1 12.8 13.3
Fo me -d inke s a us, % 8.7 4.6 10.8 0.003 †
Physical ac i i y, MET-hou /week,
mean ±SD 63 ±34 60 ±32 58 ±34 0.035
Time spen wa ching TV,
hou /week, mean ±SD 2.4 ±1.5 2.5 ±1.5 2.4 ±1.6 0.500
Fibe (g ams/day), mean ±SD 24.4 ±8.0 25.1 ±8.0 24 ±7.6 0.477
Numbe o ch onic condi ions,
mean ±SD 0.7 ±0.7 0.7 ±0.7 0.7 ±0.8 0.400
Numbe o medica ions pe day,
mean ±SD 1.7 ±1.7 1.8 ±1.8 1.7 ±1.9 0.389
Hype ension, % 63.7 64.3 56.4 0.118 †
Diabe es melli us, % 13 12.8 15.3 0.471 †
Hype choles e olemia, % 70.8 72.1 73.6 0.643
BMI baseline, mean ±SD 28.1 ±4 28.5 ±4.4 28.6 ±4.2 0.124
SD = S anda d de ia ion.
†
Chi-Squa ed. The cu o poin s o e iles o he pe cen age o ene gy om ul a-p ocessed ood we e: Te ile 1
(0–11.8), Te ile 2 (11.8–20.9), Te ile 3 (21–57.5) in men; Te ile 1 (0–13.9), Te ile 2 (14–23.8), Te ile 3 (23.9–66.7) in women.
Pa icipan s wi h a highe baseline UPF consump ion we e mo e likely o ha e a
decline in hei enal unc ion o e he ollow-up. Compa ed wi h he lowes e ile o he
pe cen age o ene gy in ake om UPF, he ully adjus ed OR (95% CI) o enal unc ion
decline was 1.56 (1.02–2.38) o he second e ile and 1.74(1.14–2.66) o he highes e ile;
p- end 0.023 (Table 2, Model 3). The co esponding alues when UPF in ake was exp essed
in g/kg/day we e 1.28 (0.85–1.85) and 1.62 (1.06–2.49); p- end 0.043 (Table 2, Model 3).
When s a i ied analyses we e pe o med, esul s we e simila acco ding o he p e a-
lence o se e al ca dio ascula isk ac o s (ha ing a leas one ch onic condi ion, hype en-
sion, diabe es, hype choles e olemia, and obesi y). Howe e , in indi iduals wi h diabe es
and wi hou obesi y, he associa ion was s onge (Table 3).
When compa ing ex eme e iles o speci ic UPF ood g oups, al hough none o hem
eached s a is ical signi icance, b eak as ce eals, non-alcoholic be e ages (e.g., indus ial
ui juices), cakes and pas ies, and mea p oduc s, we e he ones ha con ibu ed he
mos o his associa ion (Figu e 2).

Nu ien s 2021,13, 428 6 o 12
Table 2.
Associa ion be ween he consump ion o ul a-p ocessed ood exp essed as a pe cen age
o o al ene gy (% ene gy) o as g ams pe day/weigh and he isk o enal unc ion decline a e
6-yea o ollow-up (2008/10–2015) (n= 1312).
T1 (Lowes )
OR (95% CI)
T2
OR (95% CI)
T3 (Highes )
OR (95% CI) pT end
Ul a-P ocessed Food Consump ion (% Ene gy)
n438 438 436
Cases 47 67 69
Model 1 Re . 1.63 (1.08–2.44) 1.75 (1.16–2.64) 0.008
Model 2 Re . 1.56 (1.04–2.35) 1.69 (1.11–2.55) 0.014
Model 3 Re . 1.56 (1.02–2.38) 1.74 (1.14–2.66) 0.026
Ul a-P ocessed Food Consump ion (g/kg/Day)
n438 437 437
Cases 55 61 67
Model 1 Re . 1.26 (0.84–1.89) 1.56 (1.03–2.35) 0.034
Model 2 Re . 1.25 (0.84–1.88) 1.57 (1.04–2.38) 0.033
Model 3 Re . 1.28 (0.85–1.95) 1.62 (1.06–2.49) 0.043
OR: Odds Ra io. CI: Con idence in e al. Model 1: Logis ic eg ession model adjus ed o sex, age, and o al
ene gy in ake. Model 2: As in Model 1 and addi ionally adjus ed o educa ion le el (p ima y, seconda y,
uni e si y), smoking s a us (ne e , o me , cu en smoke ), o me -d inke s a us (yes, no), physical ac i i y
(MET-hou /week), ime spen wa ching TV (hou /week), and o al ibe consump ion (g ams/day). Model 3:
As in Model 2 and addi ionally adjus ed o numbe o ch onic condi ions (con inuous), numbe o medica ions
used (con inuous), hype ension (yes/no), and diabe es (yes/no), hype choles e olemia (yes/no) and body mass
index (con inuous).
Table 3.
Associa ion be ween he consump ion o ul a-p ocessed ood exp essed as a pe cen age o o al ene gy (% ene gy)
and he isk o enal unc ion decline a e 6-yea o ollow-up (2008/10–2015) (n= 1312) acco ding o mo bidi y and se e al
ca dio ascula isk ac o s.
Ul a-P ocessed Food Consump ion (% Ene gy)
T1 (Lowes ) T2 T3 (Highes ) pT end
Wi h a leas one ch onic condi ion
n/cases 229/25 241/38 232/38
OR (95% CI) 1 (Re .) 1.49 (0.85–2.62) 1.5(0.84–2.68) 0.174
Wi hou any ch onic condi ion
n/cases 209/22 197/29 204/31
OR (95% CI) 1 (Re .) 1.47 (0.78–2.76) 1.61 (0.86–3.03) 0.137
Wi h hype ension
n/cases 279/33 282/48 246/45
OR (95% CI) 1 (Re .) 1.54 (0.94–2.53) 1.65 (0.99–2.75) 0.055
Wi hou hype ension
n/cases 159/14 156/19 190/24
OR (95% CI) 1 (Re .) 1.49 (0.68–3.24) 1.52 (0.70–3.27) 0.305
Wi h diabe es
n/cases 57/8 56/14 67/22
OR (95% CI) 1 (Re .) 1.86 (0.62–5.6) 3.08 (1.08–8.75) 0.034
Wi hou diabe es
n/cases 381/39 382/53 369/47
OR (95% CI) 1 (Re .) 1.43 (0.91–2.25) 1.36 (0.85–2.19) 0.200
Wi h hype choles e olemia
n/cases 310/34 316/47 321/53
OR (95% CI) 1 (Re .) 1.50 (0.92–2.46) 1.67 (1.03–2.73) 0.042
Wi hou hype choles e olemia
n/cases 128/13 122/20 115/16
OR (95% CI) 1 (Re .) 1.63 (0.72–3.70) 1.38 (0.59–3.27) 0.474
Wi h obesi y (BMI ≥30 kg/m2)
n/cases 125/17 141/27 128/18
OR (95% CI) 1 (Re .) 1.55 (0.76–3.14) 1.08 (0.50–2.32) 0.833
Wi hou obesi y (BMI < 30 kg/m2)
n/cases 313/30 297/40 308/51
OR (95% CI) 1 (Re .) 1.49 (0.88–2.53) 1.90 (1.13–3.19) 0.015
All adjus men s as in model 3. OR: Odds Ra io. CI: Con idence In e al. The conside ed ch onic condi ions we e: ch onic espi a o y
disease, co ona y hea disease, s oke, hea ailu e, os eoa h i is, cance , and dep ession equi ing ea men . Hype ension: sys olic
blood p essu e
≥
140 mmHg o dias olic blood p essu e
≥
90 mmHg o an ihype ensi e ea men . Diabe es: as ing glucose
≥
126 mg/dl
o an idiabe ic ea men . Hype choles e olemia: o al choles e ol ≥200 mg/dl o d ug ea men . Obesi y: BMI ≥30 kg/m2.
Nu ien s 2021,13, 428 7 o 12
Figu e 2.
Odds a io o enal decline isk and 95% con idence in e als (95% CIs) o e ile 3 (highes
in ake) o g oups o ul a-p ocessed ood consump ion as a pe cen age o o al ene gy (% o ene gy)
e sus e ile 1 (lowes in ake), in he Senio s-ENRICA 1 coho s udy, n= 1312. When he in ake o
ul a-p ocessed ood om a speci ic ood g oup occu ed in less han 25% o he pa icipan s, he odds
a io (95% CI) was calcula ed be ween subjec s who consumed he ood compa ed o hose who did
no (as in b eak as ce eals, dai y desse s, so d inks and spi i s). p o linea end was calcula ed
using e iles as a con inuous a iable. Adjus men s as in model 3. O he non-alcoholic be e age
g oups include ins an co ee d inks and cocoa d inks, packaged juices and o he non-alcohol d inks,
excluding so d inks.
4. Discussion
A e 6 yea s o ollow-up, UPF consump ion was independen ly associa ed wi h enal
unc ion decline in a coho o communi y-dwelling indi iduals aged 60 o olde om
Spain. This s udy ound ha pa icipan s wi h he highes UPF consump ion a baseline
ha e a ound a 50% highe isk o enal unc ion decline compa ed o pa icipan s wi h
he lowes consump ion a e adjus ing o a se ies o demog aphic, li es yle as well as
clinical and biological co a ia es. This associa ion could be s onge among diabe ics. Ou
s udy ex ends indings om p e ious s udies o die a y pa e ns and enal impai men
isk, and ein o ces he impo ance o die in he p ima y p e en ion o enal decline a a
popula ion le el.
Some heal hy die s ha e been linked o a lowe isk o CKD. A highe adhe ence o he
Medi e anean die ( ich in ui , ege ables, ce eals, legumes, and ish) has been associa ed
wi h a lowe CKD incidence in a mul i-e hnic coho [
42
]. Simila ly, s udies o young and
middle-aged adul s in he Uni ed S a es ha e ound ha hose wi h a highe adhe ence
o he Die a y App oaches o S op Hype ension (DASH) die ( ich in ui , ege ables,
legumes, nu s, and low- a dai y) also had a lowe isk o inciden CKD [
43
]. Ano he
s udy examining die a y sou ces o p o eins and CKD ound ha , when one se ing o ed
and p ocessed mea was eplaced wi h plan p o eins, he isk o CKD was signi ican ly
lowe [
12
]. Likewise, he p opo ion o CKD a ibu able isk o a lowe adhe ence o a
heal hy plan -based die is es ima ed a a ound 4% [44].
By con as , UPF consump ion has p e iously been associa ed wi h hype ension,
diabe es, obesi y and me abolic synd ome [
45
], which in u n, a e majo isk ac o s o
enal unc ion impai men . In addi ion, in ou s udy UPF consump ion inc eased he isk
o enal unc ion decline independen ly o hype ension, diabe es, BMI, as well as o he
ch onic condi ions. Ou indings sugges ha UPF consump ion may be di ec ly associa ed
wi h an impai men o he enal unc ion.
Se e al mechanisms may be in ol ed in he esul s ob ained, such as he consump ion
o a low amoun o ibe , as well as high amoun s o sodium, suga s, and phospha es, when
ollowing a die high in UPF [
14
]. In a ecen s udy, ibe in ake has shown an in e se
associa ion wi h inciden CKD, wi h a 11% dec ease in he isk o CKD o e e y 5-g
Nu ien s 2021,13, 428 8 o 12
inc ease in ib e in ake [
46
]. Fibe in ake also imp o es glycemic con ol as well as insulin
sec e ion, which is associa ed wi h a lowe isk o mic oalbuminu ia and p o einu ia [
47
].
In addi ion, ibe can educe he isk o CKD by mi iga ing he e ec o some o hei
well-es ablished isk ac o s such as hype ension and diabe es [48].
A highe in ake o sodium is also consis en ly associa ed wi h an inc eased isk o CKD
and eGFR decline [
11
]. The me a-analysis om Bach e al. shows wi h a mode a e quali y o
e idence ha sodium in ake was associa ed wi h a highe isk o CKD in 6 s udies in ol ing
43,772 pa icipan s [
11
]. Likewise, in coho s udies, a die low in sodium is associa ed
wi h a lowe CKD isk among high- isk indi iduals. This is he case o he Teh an Lipid
and Glucose S udy (TLGS) whose pa icipan s had dysglycemia, dyslipidemia, and high
blood p essu e and whe e a nega i e associa ion was ound be ween a low-sodium die
(DASH-s yle die ) and CKD a e 3 yea s o ollow-up [
49
]. In addi ion, a U-shaped dose-
esponse has been obse ed be ween sodium in ake and he incidence o CKD, al hough i
could be due o e e se causa ion when he e is low sodium in ake. Howe e , e en hough
mo e esea ch on he op imal le el o sodium in ake is needed, i is clea ha high sodium
consump ion is de imen al, especially o hype ensi e pa ien s [50,51]. Mo eo e , when
CKD is es ablished, he e is clinical ial e idence o he de imen al e ec o sodium in ake
on blood p essu e and p o einu ia [52].
In ake o simple suga s may also play a ole. In an analysis o 1630 I anians om he
TLGS, a die high in a s and suga was ela ed o a 46% inc ease in he isk o CKD. Like-
wise, consump ion o >4 se ings/week o so d inks doubled he isk o de eloping CKD
(Yuzbashian e al., 2016) [
53
]. Simila esul s we e ob ained in a communi y-based coho o
A ican Ame icans ( he Jackson Hea S udy) a e 8 yea s o ollow-up (Rebholz, 2020) [
54
].
Some ood addi i es om UPF, pa icula ly phospha es, dese e a men ion [
55
].
Unlike o ganic phospho us, which is p esen in plan -based ood (wi h low phospho us
a ailabili y due o phy a e con en ) [
56
], ino ganic phospha e is p esen in many UPF as
an addi i e, and has a e y high bioa ailabili y [
57
,
58
]. This is why ino ganic phospha e
is disp opo iona ely high in wes e nized die s compa ed wi h o ganic phospho us om
na u al ood sou ces [
57
]. UPFs ha con ain high amoun s o ino ganic phospha es include
p ocessed mea , ham, sausages, canned ish, baked goods, cola d inks, and o he so
d inks. Indi iduals wi h a high consump ion o UPF also had a high consump ion o added
phospha e ha could be 250–1000 mg highe han in indi iduals wi h a low consump ion o
UPF [
59
]. Howe e , die a y phospha e assessmen is complex and mos ly unde es ima ed,
as he amoun o ood addi i es con aining phospha e is no epo ed on ood labels [
57
,
60
].
The apid consump ion o la ge amoun s o phospha e leads o acu e kidney inju y
and ul ima ely o CKD [
55
]. Fu he mo e, la ge doses o phospha e, as in o al he apy
(>2250 mg/day on op o die a y phospha e), o e 1–7 yea s inc eased he occu ence o
calci ica ions in so issues and impai ed enal unc ion [
61
]. High se um phospha e in
olde adul s is also associa ed wi h enal dys unc ion, ca dio ascula isk and p ema u e
dea h [55,62,63].
Phospha e could also ope a e h ough o he enal isk ac o s such as hype ension o
diabe es [
10
]. In an 11 week in e en ion s udy in young adul s wi h no mal enal unc ion,
high-phospha e in ake was linked o inc eased pulse a e and sys olic and dias olic blood
p essu e [
64
]. Likewise, in he F ench E3N coho , a highe phospha e in ake was associa ed
wi h a highe incidence o diabe es [65].
The social and heal h ele ance o enal damage is subs an ial, and i s p e en ion is
a public heal h p io i y because: (1) CKD imposes a signi ican economic bu den. Many
de eloped coun ies spend be ween 2–3% o hei annual heal h ca e budge o ea he
mos ad anced o ms o he disease, o example kidney eplacemen he apy. (2) Mo ali y
due o enal ailu e also ose be ween 2005 and 2017 om 0.9 million o 1.2 million dea hs
annually [
7
]. (3) The e is a signi ican po en ial o p e en ion (wi h o e 497 million adul s
wo ldwide wi h CKD s ages 1–5 [
11
]). (4) The consump ion o UPF con inuous o s eadily
inc ease globally.
Nu ien s 2021,13, 428 9 o 12
Ou s udy has some limi a ions. Fi s , as in mos nu i ional epidemiology s udies,
die was sel - epo ed so a plausible ecall bias canno be excluded. Second, ce ain
misclassi ica ion o UPF canno be uled ou ei he . Howe e , he NOVA classi ica ion
is easy o apply, and i is he mos equen ly used in epidemiological s udies. Thi d,
he numbe o people who de eloped a decline in enal unc ion was small, al hough
he s a is ical powe was enough o ind signi ican associa ions. Finally, enal decline is
based on one c ea inine measu emen in each ime pe iod ha may lead o non-di e en ial
misclassi ica ion and conse a i e esul s. This s udy also had some s eng hs. These
include he p ospec i e design wi h a ela i ely long ollow-up pe iod, which educed he
possibili y o e e se causa ion. Mo eo e , die was collec ed using a comp ehensi e and
alida ed die a y his o y, and analyses we e adjus ed o he main po en ial con ounde s.
Finally, his is he i s s udy showing his associa ion.
5. Conclusions
In conclusion, in his p ospec i e s udy o communi y-dwelling olde adul s om
Spain, a highe in ake o UPF was associa ed wi h a decline in enal unc ion. These
obse a ional indings add e idence o he es ic ion o he consump ion o UPF as a
p ima y p e en ion s a egy o CKD, and he need o p omo e he consump ion o esh
o minimally p ocessed oods o e UPF o educe he bu den o disease in he gene al
olde popula ion.
Au ho Con ibu ions:
Concep ualiza ion, J.R.-G., C.D.-V. and P.G.-C.; me hodology, J.R.-G., C.D.-V.
and P.G.-C.; o mal analysis, J.R.-G., C.D.-V., H.S.-I. and P.G.-C.; w i ing—o iginal d a p epa a ion,
J.R.-G., C.D.-V. and P.G.-C.; w i ing— e iew and J.R.-G., C.D.-V., H.S.-I., A.B.-B., B.M.-F., J.R.B., F.R.-A.
and P.G.-C.; edi ing, supe ision, J.R.-G., C.D.-V. and P.G.-C.; All au ho s ha e ead and ag eed o he
published e sion o he manusc ip .
Funding:
Da a collec ion was unded by he ollowing g an s: FIS PI09/1626, PI16/00609, PI16/01460,
PI17/1709, PI20/144 (S a e Sec e a y o R + D and FEDER/FSE), and he CIBERESP, Ins i u o de
Salud Ca los III. Mad id, Spain. Ca olina Dona -Va gas has an “A acción de alen o” con ac om
he Regional Go e nmen o he Mad id Communi y/IMDEA Food.
Ins i u ional Re iew Boa d S a emen :
The s udy was conduc ed acco ding o he guidelines o he
Decla a ion o Helsinki and app o ed by he E hics Commi ee o La Paz Uni e si y Hospi al in
Mad id (Spain). The e hical app o al numbe is: HULP-PI-1793 om The Regional E hics Commi ee
o Clinical Resea ch o he “Hospi al Uni e si a io La Paz” (CEIC-R IdiPAZ).
In o med Consen S a emen :
In o med consen was ob ained om all subjec s in ol ed in he s udy.
Con lic s o In e es :
The au ho s decla e no con lic o in e es . The unde s had no ole in he design
o he s udy; in he collec ion, analyses, o in e p e a ion o da a; in he w i ing o he manusc ip , o
in he decision o publish he esul s.
Re e ences
1.
Hommos, M.S.; Glassock, R.J.; Rule, A.D. S uc u al and unc ional changes in human kidneys wi h heal hy aging. J. Am. Soc.
Neph ol. 2017,28, 2838–2844. [C ossRe ] [PubMed]
2.
Kidney Disease: Imp o ing Global Ou comes (KDIGO) CKD Wo k G oup. KDIGO 2012 Clinical P ac ice Guideline o he
E alua ion and Managemen o Ch onic Kidney Disease. Kidney In . Suppl. 2013,3, 1–150. [C ossRe ]
3.
Bello, A.K.; Le in, A.; Tonelli, M.; Okpechi, I.G.; Feehally, J.; Ha is, D.; Jindal, K.; Salako, B.L.; Ra eb, A.; Osman, M.A.; e al.
Assessmen o global kidney heal h ca e s a us. JAMA J. Am. Med. Assoc. 2017,317, 1864–1881. [C ossRe ] [PubMed]
4.
Nagha i, M.; Abajobi , A.A.; Abba a i, C.; Abbas, K.M.; Abd-Allah, F.; Abe a, S.F.; Aboyans, V.; Ade okunboh, O.; Ä nlö , J.;
A shin, A.; e al. Global, egional, and na ional age-sex speci c mo ali y o 264 causes o dea h, 1980–2016: A sys ema ic analysis
o he Global Bu den o Disease S udy 2016. Lance 2017,390, 1151–1210. [C ossRe ]
5.
Couse , W.G.; Remuzzi, G.; Mendis, S.; Tonelli, M. The con ibu ion o ch onic kidney disease o he global bu den o majo
noncommunicable diseases. Kidney In . 2011,80, 1258–1270. [C ossRe ]
6.
Palme , S.; Vecchio, M.; C aig, J.C.; Tonelli, M.; Johnson, D.W.; Nicolucci, A.; Pelleg ini, F.; Saglimbene, V.; Log oscino, G.;
Fishbane, S.; e al. P e alence o dep ession in ch onic kidney disease: Sys ema ic e iew and me a-analysis o obse a ional
s udies. Kidney In . 2013,84, 179–191. [C ossRe ]