nu ien s
Re iew
E ec s o LC-PUFA Supplemen a ion in Pa ien s wi h
Phenylke onu ia: A Sys ema ic Re iew o
Con olled T ials
Ma ía Luz Couce 1,2,3,4,*, Ma ía Joséde Cas o 1,2,3, Ca mela de Lamas 3,4,5 and Rosau a Leis 1,2,3,4
1Depa men o Pedia ics, Uni e si y Clinical Hospi al o San iago de Compos ela,
15706 San iago de Compos ela, Spain
2IDIS-Heal h Resea ch Ins i u e o San iago de Compos ela, 15706 San iago de Compos ela, Spain
3CIBERER, Pabellón 11, 28029 Mad id, Spain
4Uni e sidade de San iago de Compos ela, 15704 San iago de Compos ela, Spain
5
Depa men o Pedia ics, Pedia ic Me abolism and Resea ch Uni , Reina So ia Uni e si y Hospi al, IMIBIC,
14004 Co doba, Spain
*Co espondence: [email p o ec ed]; Tel.: +34-98-195-0151
Recei ed: 4 June 2019; Accep ed: 2 July 2019; Published: 6 July 2019
Abs ac :
E idence sugges s a ole o long chain polyunsa u a ed a y acids (LC-PUFA), in which
animal oods a e especially ich, in op imal neu al de elopmen . The LC-PUFAs docosahexaenoic
acid (DHA) and a achidonic acid, ound in high concen a ions in he b ain and e ina, ha e po en ial
bene icial e ec s on cogni ion, and mo o and isual unc ions. Phenylke onu ia (PKU) is he mos
common inbo n e o o amino acid me abolism. The ea men o PKU consis s o a phenylalanine- ee
die , which limi s he in ake o na u al p o eins o high biological alue. In his sys ema ic e iew, we
summa ize he a ailable e idence suppo ing a ole o LC-PUFA supplemen a ion as an e ec i e
means o inc easing LC-PUFA le els and imp o ing isual and neu ocogni i e unc ions in PKU
pa ien s. Da a om con olled ials o child en and adul s (up o 47 yea s o age) we e ob ained by
sea ching he MEDLINE and SCOPUS da abases ollowing P e e ed Repo ing I ems o Sys ema ic
Re iews and Me a-Analyses (PRISMA) guidelines. Fo each selec ed s udy, he isk o bias was
assessed applying he me hodology o he Coch ane Collabo a ion. The indings indica e ha DHA
supplemen a ion in PKU pa ien s om 2 weeks o 47 yea s o age imp o es DHA s a us and dec eases
isual e oked po en ial P100 wa e la ency in PKU child en om 1 o 11 yea s old. Neu ocogni i e
da a a e inconclusi e.
Keywo ds:
a achidonic acid; cogni i e unc ion; docosahexaenoic acid; long-chain polyunsa u a ed
a y acids; phenylke onu ia; isual unc ion
1. In oduc ion
Phenylke onu ia (PKU; OMIM 261600) is an inbo n e o o phenylalanine (Phe) me abolism caused
by an inhe i ed de iciency in L-phenylalanine-4-hyd oxylase (PAH; EC 1.14.16.1) ac i i y, leading o
ele a ed le els o Phe in body luids [
1
]. O pa ien s wi h high phenylalanine concen a ions, 98% ha e a
de ec in PAH and 1–2% in e ahyd obiop e in me abolism. Child en wi h PKU diagnosed by newbo n
sc eening who begin die a y ea men du ing he neona al pe iod usually show no mal neu ological
de elopmen [
2
,
3
]. Howe e , hese pa ien s may ha e lowe in elligence quo ien s [
4
] and exhibi mild
neu opsychological dis u bances including impai ed mo o skills, isual unc ion, a en ion, inhibi ion,
and memo y [
5
,
6
], especially when compa ed wi h non-phenylke onu ic siblings [
7
] and heal hy
indi iduals [
8
,
9
]. PKU ea men consis s o li elong es ic ion o Phe in ake by limi ing he amoun o
na u al p o ein in he die , combined wi h adminis a ion o a Phe- ee amino-acid mix u e [
10
]. Mo e
Nu ien s 2019,11, 1537; doi:10.3390/nu11071537 www.mdpi.com/jou nal/nu ien s
Nu ien s 2019,11, 1537 2 o 14
ecen ly, a syn he ic o m o e ahyd obiop e in (6R-BH4) has been used o ea selec ed pa ien s
who ha e mode a e o ms o PKU and espond o he BH4 loading es [
11
,
12
]. Owing o a endency
o exclude p o ein- ich animal ood om hei die , mic onu ien de iciencies a e common in PKU
pa ien s [13–15].
Mea and ish a e he main sou ces o long-chain polyunsa u a ed a y acids (LC-PUFA) in
humans, and al hough hey a e p oduced endogenously, die a y in ake is he key de e minan o
LC-PUFA le els [
16
]. Docosahexaenoic acid (DHA) and a achidonic acid (AA) a e he mos impo an
LC-PUFAs o he n-3 and n-6 se ies, espec i ely [
17
]. Bo h a e s uc u al componen s o cell memb anes
and in luence hei biological unc ions, including enzyma ic ac i i y, anspo h ough ion channels,
and signal ansduc ion [
18
], especially in he ne ous sys em and he e ina [
19
]. Inco po a ion o
DHA and AA in hese issues du ing he p e- and pos na al pe iods has been co ela ed wi h isual,
cogni i e, and mo o unc ions in humans [20–22].
The low-Phe die o PKU pa ien s has been linked o insu icien blood le els o LC-PUFAs,
which may con ibu e o he mild neu ological, cogni i e, and isual al e a ions desc ibed in hese
pa ien s [
23
]. Howe e , o da e no conclusi e e idence suppo s a link be ween he PKU die , he
LC-PUFA p o ile and he clinical s a us o PKU pa ien s [
24
]. In his sys ema ic e iew, we p esen a
comp ehensi e o e iew o e idence om clinical ials assessing a co ela ion be ween he PKU die ,
LC-PUFA s a us, and neu ocogni i e and isual unc ions.
2. Me hods
This e iew was conduc ed ollowing he guidelines o P e e ed Repo ing I ems o Sys ema ic
Re iews and Me a-Analyses (PRISMA) [
25
] and was egis e ed in he In e na ional P ospec i e
Regis e o Sys ema ic Re iews (PROSPERO) wi h he numbe CRD42019133315. The e iew ques ion,
which was o mula ed ollowing he PICOS (Popula ion, In e en ion, Compa ison, Ou comes and
Se ings) c i e ia [26] (Table 1), was as ollows: Does LC-PUFA supplemen a ion in luence isual and
neu ocogni i e unc ions in pa ien s wi h phenylke onu ia?
Table 1.
PICOS (Popula ion, In e en ion, Compa ison, Ou comes and Se ings) c i e ia [
26
] o
he inclusion o s udies e alua ing he e ec s o long-chain polyunsa u a ed a y acids (LC-PUFA)
supplemen a ion in phenylke onu ia pa ien s.
Pa ame e 1Inclusion C i e ia
Popula ion Pa ien s wi h phenylke onu ia
In e en ion Con olled LC-PUFA in ake
Compa ison Non-exposed con ol g oup
Ou come Visual and neu ocogni i e unc ions and a y acid le els
Se ing Con olled ials
1PICOS c i e ia [26].
2.1. Inclusion and Exclusion C i e ia
S udies we e selec ed applying he ollowing inclusion c i e ia: all con olled s udies, andomized
o no , o pa ien s wi h PKU o any age and e hnici y ha we e published be ween 1 Janua y 1995 and
1 Ap il 2019. S udies in which LC-PUFA supplemen a ion was adminis e ed pa en e ally and hose
lacking a con ol g oup ha did no ecei e LC-PUFA supplemen a ion we e excluded.
2.2. In e en ion Types
S udies we e no es ic ed acco ding o he du a ion o supplemen a ion o he ype o dose o
LC-PUFAs adminis e ed. All s udies in which pa ien s ecei ed o al LC-PUFA supplemen a ion and
he e ec s we e compa ed wi h a non-supplemen ed g oup we e conside ed o inclusion.
Nu ien s 2019,11, 1537 3 o 14
2.3. P ima y Ou come Measu es
Visual e oked po en ial (VEP) a iables, speci ically al e a ions in P100 wa e and P1 peak la encies
(in ms), we e he p ima y ou come measu es used o assess he e ec s o LC-PUFA supplemen a ion
on isual unc ion. Fo he assessmen o neu ocogni i e unc ion, da a om any s udy ha included
some o m o e alua ion o psychomo o de elopmen we e conside ed. Ci cula ing and e y h ocy e
lipid le els and changes in lipid le els (mg/L, mmol/L, o % change) a e supplemen a ion we e
conside ed alid measu es o he assessmen o e ec s on lipid s a us.
2.4. Li e a u e Sea ch
The PUBMED and SCOPUS da abases we e sea ched using he MeSH e ms “Fa y
Acids, Unsa u a ed” and “Phenylke onu ias”. “Fa y Acids, Unsa u a ed” (Mesh Te ms) AND
“Phenylke onu ias” (Mesh Te ms) was he sea ch s a egy used in PUBMED. SCOPUS was sea ched
using he ollowing o mula: “Fa y acids” AND “Phenylke onu ias”, excluding esul s om
animal s udies.
2.5. S udy Selec ion
Two au ho s (MJDC and CDL) independen ly selec ed s udies om he 33 a icles e iewed in
ull. In cases in which he e was a lack o consensus ega ding selec ion, disc epancies we e a bi a ed
by MLC and RL. Nine a icles [27–35] we e ul ima ely selec ed o inclusion in he e iew.
2.6. Da a Ex ac ion
The ollowing da a we e ex ac ed om each s udy: publica ion yea ; numbe o pa icipan s
by sex; age; s udy ype; in e en ion cha ac e is ics. Da a on s udy du a ion, ou comes, esul s, and
conclusions we e sepa a ely ex ac ed by wo in es iga o s. Any disc epancies in opinions we e
a bi a ed by MLC and RL.
2.7. Assessmen o Risk o Bias
Following he me hodology o he Coch ane Collabo a ion, London, UK [
36
], wo e alua o s
independen ly s udied he isks o bias. The a icles we e analyzed indi idually, and he co esponding
isk o bias was classi ied as high, unce ain, o low depending on he isk o selec ion bias ( andom
sequence gene a ion, alloca ion concealmen ), pe o mance bias (blinding o pa icipan s and pe sonnel),
de ec ion bias (blinding o ou come assessmen ), a i ion bias (incomple e ou come da a), epo ing
bias (selec i e epo ing), and o he o ms o bias. A hi d and a ou h e iewe a bi a ed in cases o
disc epancies o opinion. We also e alua ed he gene al isk o bias in he g oup o a icles included in
he sys ema ic e iew, exp essed as a pe cen age o a icles ha p esen a isk o speci ic bias in ela ion
o he o al numbe o s udies included.
3. Resul s
The esul s o each s ep o he bibliog aphic sea ch a e shown in Figu e 1. O he 84 esul s om
he i s sea ch (PUBMED, 53; SCOPUS, 30; o he sou ces, 1), 27 duplica e a icles we e excluded.
Twen y- ou a icles we e excluded a e e ision o he abs ac . O he 33 a icles conside ed o
e alua ion o he comple e ex , nine a icles we e ul ima ely included in he sys ema ic e iew [
27
–
35
].
Nu ien s 2019,11, 1537 4 o 14
Nu ien s 2019, 11, x FOR PEER REVIEW 4 o 14
Figu e 1. Flow cha depic ing he li e a u e sea ch p ocess.
Tables 2–4 show he main cha ac e is ics o he selec ed clinical ials. Publica ion da es ange
om 1995 o 2017. Only one [35] o he nine a icles included was published be o e he yea 2000. The
combined s udy popula ion o he nine clinical ials included in his e iew was 419 indi iduals, 365
(87.1%) o whom had phenylke onu ia. Only wo s udies included a heal hy con ol g oup [29,33]
and he emaining se en, a PKU con ol g oup [27,28,30–32,34,35]. Ages anged om 2.1 weeks o 47
yea s. The mean sample size was 46 ± 27 pa icipan s ( ange, 20–109). The median in e en ion pe iod
was 6 mon hs ( ange, 3–12 mon hs). Supplemen a ion consis ed o a phenylalanine- ee LC-PUFA-
supplemen ed in an o mula [30–32] o DHA [27,28] o ish-oil [29,33–35] capsules. In one s udy [32],
no in o ma ion on he DHA supplemen a ion dose was p o ided. In all o he s udies, he DHA dose
anged om 0.1 o 15 mg/kg/day. In ou s udies, supplemen a ion consis ed o a combina ion o
omega 6 and omega 3 [30–32,34], wi h omega 6 o omega 3 a ios anging om 1:1 o 3:1.
3.1. LC-PUFA Supplemen a ion and Ci cula ing and E y h ocy e Lipids
Changes in ci cula ing and e y h ocy e lipids we e e alua ed in se en o he nine s udies
[27,28,30–32,34,35] (Table 2), all o which epo ed signi ican di e ences in DHA le els in LC-PUFA-
supplemen ed g oups e sus con ols. O hose se en ials, i e [27,28,32,34,35] epo ed highe
le els o DHA in he LC-PUFA-supplemen ed g oup, while wo ials conce ning LC-PUFA
supplemen a ion in in an s [30,31] epo ed a signi ican ly lowe dec ease in DHA le els compa ed
wi h con ols. Fou o he a icles [27,28,30,35] e alua ed DHA le els in plasma and ou [28,31,32,34]
in e y h ocy e lipids. Only one s udy [28] e alua ed DHA le els in bo h plasma and e y h ocy e
lipids. O he wo s udies in which choles e ol and iglyce ide le els we e measu ed [27,35], he
s udy by Demmelmai e al., in which pa ien s ecei ed he lowe DHA dose [27], epo ed no
di e ences in hese a iables be ween g oups, while he o he [35] epo ed signi ican ly lowe
iglyce ide le els in he LC-PUFA-supplemen ed g oup.
3.2. LC-PUFA Supplemen a ion and Visual Func ion
Fou o he s udies included in ou e iew p o ided VEP da a [27,31,33,34] (Table 3). All
measu ed P100 wa e la ency (Pa e n VEP), and one [31] also measu ed he P1 peak (Flash VEP).
Two o he ou s udies in which he highes dose (15 mg/kg) o DHA [33,34] was adminis e ed
epo ed signi ican dec eases in P100 wa e la ency in he supplemen a ion g oup, which ecei ed
DHA capsules in bo h cases. The wo emaining s udies [27,31], one [31] in which newbo ns we e
Figu e 1. Flow cha depic ing he li e a u e sea ch p ocess.
Tables 2–4show he main cha ac e is ics o he selec ed clinical ials. Publica ion da es ange
om 1995 o 2017. Only one [
35
] o he nine a icles included was published be o e he yea 2000. The
combined s udy popula ion o he nine clinical ials included in his e iew was 419 indi iduals, 365
(87.1%) o whom had phenylke onu ia. Only wo s udies included a heal hy con ol g oup [
29
,
33
]
and he emaining se en, a PKU con ol g oup [
27
,
28
,
30
–
32
,
34
,
35
]. Ages anged om 2.1 weeks o
47 yea s. The mean sample size was 46
±
27 pa icipan s ( ange, 20–109). The median in e en ion
pe iod was 6 mon hs ( ange, 3–12 mon hs). Supplemen a ion consis ed o a phenylalanine- ee
LC-PUFA-supplemen ed in an o mula [
30
–
32
] o DHA [
27
,
28
] o ish-oil [
29
,
33
–
35
] capsules. In one
s udy [
32
], no in o ma ion on he DHA supplemen a ion dose was p o ided. In all o he s udies,
he DHA dose anged om 0.1 o 15 mg/kg/day. In ou s udies, supplemen a ion consis ed o a
combina ion o omega 6 and omega 3 [
30
–
32
,
34
], wi h omega 6 o omega 3 a ios anging om 1:1 o
3:1.
3.1. LC-PUFA Supplemen a ion and Ci cula ing and E y h ocy e Lipids
Changes in ci cula ing and e y h ocy e lipids we e e alua ed in se en o he nine
s udies
[27,28,30–32,34,35] (Table 2
), all o which epo ed signi ican di e ences in DHA le els in
LC-PUFA-supplemen ed g oups e sus con ols. O hose se en ials, i e [
27
,
28
,
32
,
34
,
35
] epo ed
highe le els o DHA in he LC-PUFA-supplemen ed g oup, while wo ials conce ning LC-PUFA
supplemen a ion in in an s [
30
,
31
] epo ed a signi ican ly lowe dec ease in DHA le els compa ed
wi h con ols. Fou o he a icles [
27
,
28
,
30
,
35
] e alua ed DHA le els in plasma and ou [
28
,
31
,
32
,
34
]
in e y h ocy e lipids. Only one s udy [
28
] e alua ed DHA le els in bo h plasma and e y h ocy e lipids.
O he wo s udies in which choles e ol and iglyce ide le els we e measu ed [
27
,
35
], he s udy by
Demmelmai e al., in which pa ien s ecei ed he lowe DHA dose [
27
], epo ed no di e ences in
hese a iables be ween g oups, while he o he [
35
] epo ed signi ican ly lowe iglyce ide le els in
he LC-PUFA-supplemen ed g oup.
3.2. LC-PUFA Supplemen a ion and Visual Func ion
Fou o he s udies included in ou e iew p o ided VEP da a [
27
,
31
,
33
,
34
] (Table 3). All measu ed
P100 wa e la ency (Pa e n VEP), and one [
31
] also measu ed he P1 peak (Flash VEP). Two o he ou
s udies in which he highes dose (15 mg/kg) o DHA [
33
,
34
] was adminis e ed epo ed signi ican
dec eases in P100 wa e la ency in he supplemen a ion g oup, which ecei ed DHA capsules in bo h
cases. The wo emaining s udies [
27
,
31
], one [
31
] in which newbo ns we e supplemen ed wi h o mula
Nu ien s 2019,11, 1537 5 o 14
and he o he one [
27
] wi h 0.1–7 mg/kg o DHA supplemen a ion, epo ed no signi ican di e ences
in isual unc ion be ween g oups.
3.3. LC-PUFA Supplemen a ion and Neu ocogni i e Func ion
Neu ocogni i e unc ion was e alua ed in ou s udies [
27
–
29
,
31
] (Table 4): h ee s udies [
27
,
28
,
31
]
assessed he cogni i e a ea and wo s udies [
27
,
29
] he mo o unc ion. Only one [
29
] epo ed
signi ican di e ences in psychomo o de elopmen be ween supplemen ed and non-supplemen ed
g oups. The e was conside able a iabili y among hese ou s udies in e ms o age and ou comes.
Ages anged om 20
±
6.9 weeks in he s udy by Agos oni e al. [
31
] o 12–47 yea s in he s udy by Yin
e al. [28].
The e was also conside able he e ogenei y ega ding he scales used o e alua e cogni ion: one
s udy [
28
] assessed e bal abili y using he Peabody pic u e ocabula y es , execu i e unc ion using
he Delis Kaplan execu i e unc ion sys em and p ocessing speed using he Woodcock–Johnson III es s
o cogni i e abili y and achie emen ; ano he [
27
] calcula ed he in ellec ual quo ien using Ra en’s
p og essi e ma ices; and a hi d s udy [
31
] assessed he coe icien o de elopmen using he Bayley
es . The wo s udies ha e alua ed mo o unc ion [27,29] bo h used he Ros ock–Ose e zky scale.
3.4. Risk-o -Bias Assessmen
Fo all s udies included in ou e iew, we concluded ha he e was a low isk o selec ion bias
(alloca ion concealmen ) and an unclea isk o epo ing bias (selec i e epo ing). The pe cen age
o s udies o which he isk o di e en o ms o bias was conside ed low was as ollows: a i ion
bias (incomple e ou come da a), 89%; selec ion bias ( andom sequence gene a ion), 66%; pe o mance
bias (blinding o pa icipan s and pe sonnel), 66%; de ec ion bias (blinding o ou come assessmen ),
55%. We concluded ha he e was a isk o o he o ms o bias in 55% o s udies, due o a lack o
s anda dized p o ocols in h ee mul icen e s udies [
27
,
31
,
32
] and he lack o a con ol g oup composed
o non-LC-PUFA-supplemen ed PKU pa ien s in wo s udies [29,33].
The s udy o which he isk o biased esul s was g ea es was ha o Clea y e al. [
32
]; a
mul icen e s udy o which no s anda dized p o ocol was desc ibed. The isk o a i ion bias was also
high o his s udy, gi en he omission o an in en ion- o- ea analysis. The isk o bias was lowes o
he s udies by Kole zko e al. [29] and Yin e al. [28].
Addi ional in o ma ion on he isk-o -bias analysis ( isk-o -bias g aph and summa y) is p o ided
in he Supplemen a y Ma e ials Figu es S1 and S2.
Nu ien s 2019,11, 1537 6 o 14
Table 2. E ec s o LC-PUFA supplemen a ion on ci cula ing and e y h ocy e lipids in 299 phenylke onu ia pa ien s in con olled ials.
Re e ence nAge 1In e en ion
T ial Type
(Du a ion o
In e en ion)
Ou come Measu e Resul s 2Conclusion
Demmelmai e al.
(2018) [27]109 5–13 yea s
DHA capsules (IG1,
0.1–1.8 mg/kg/day; IG2,
1.9–7 mg/kg/day)
RCT (6 mon hs) Change in plasma lipid
concen a ion
DHA (mg/L): IG1, 5.1 ±10.3; IG2, 3.19.5 ±13.6; CG, 0.0 ±9.1
Signi ican inc ease in
DHA le els
TC (mmol/L): IG1, 0.0 ±0.5; IG2, 0.1 ±0.5; CG, −0.1 ±0.6
HDL (mmol/L): IG1, 0.0 ±0.3; IG2, 0.0 ±0.2; CG, −0.0 ±0.3
LDL (mmol/L): IG1, 0.2 ±0.5; IG2, 0.1 ±0.5; CG, −0.3 ±1.2
TG (mmol/L): IG1, 0.0 ±0.5; IG2, −0.1 ±0.5; CG, −0.1 ±0.6
Yi e al. (2011) [28] 33 (33 F) 12–47 yea s DHA capsules (10 mg/kg/day) RCT (4.5 mon hs) Plasma and e y h ocy e FA DHA (weigh % FA): IG, 3.14 ±0.57; CG, 0.97 ±0.34 Signi ican inc ease in
plasma DHA and
e y h ocy e FA le els
E y h ocy e DHA (weigh % e y h ocy e FA): IG, 5.82 ±1.26; CG, 2.35 ±
0.78
Kole zko e al. (2007)
[30]21 (8 F) 2.1 ±0.9 weeks
Tes - ea men o mula (DHA,
0.23 g/100 g FA. RCT (12 mon hs) Plasma phospholipid FA DHA (weigh % FA): IG, 3.08 ±0.1; CG, 1.52 ±0.19 Signi ican less decline
o DHA le els
Omega 6: omega 3 a io, 2:1)
Agos oni e al. (2006)
[31]42 (22 F) 20 ±6.9 weeks
Tes - ea men o mula (DHA,
0.3 g/100 g FA. RCT (12 mon hs)
Median change in LC-PUFA
concen a ion in e y h ocy e
MB phospholipids
% change DHA:
Signi ican less decline
o DHA le els
IG, −22%; CG, −64%
% change AA:
Omega 6: omega 3 a io, 2.5:1 IG, –5%; CG, –19%
Clea y e al. (2006)
[32]53 1–10 yea s
Tes - ea men o mula (PUFA,
2.8 g/100 g. RCT (20 weeks) Median change in LC-PUFA
concen a ion in e y h ocy e
MB phospholipids
% change DHA: IG, +19%; CG, +0.5% Signi ican inc ease in
DHA le els
Omega 6: omega 3 a io, 3:1) % change AA: IG, +0.5%; CG, +7.6%
Agos oni e al. (2000)
[34]20 (9 F)
10.7 ±2.4 yea s (IG)
Fish oil capsules (DHA,
15 mg/kg/day.
RCT (12 mon hs) LC-PUFA concen a ion in
e y h ocy e lipids
E y h ocy e PC (weigh % FA): EPA: IG, 0.1 ±0.07; CG, 0.1 ±0.04
Signi ican inc ease o
DHA le els
DHA: IG, 0.9 ±0.3; CG, 0.4 ±0.2. AA: IG, 5.39 ±1.16; CG, 5.83 ±0.98
AA:DHA a io, 1:1
10.5 ±2.8 yea s (CG)
E y h ocy e PEA (weigh % FA): EPA: IG, 0.3
±
0.1; CG, 0.2
±
0.1 DHA: IG,
3.7 ±1.7; CG, 1.3 ±0.9. AA: IG, 16.1 ±5.2; CG, 14.5 ±7.3
Agos oni e al. (1995)
[35]21 5–10 yea s
Fish oil capsules (DHA,
15 mg/kg/day; EPA,
22.5 mg/kg/day)
RCT (6 mon hs) Plasma lipid concen a ion
TC (mmol/L): IG, 3.12 ±0.67; CG, 3.41 ±0.28
Signi ican dec ease in
iglyce ides and
inc ease in n-3 LC-PUFA
le els
HDL (mmol/L): IG, 1.06 ±0.18; CG, 1.18 ±0.23
LDL (mmol/L): IG, 1.75 ±0.72; CG, 1.73 ±0.49
TG (mmol/L): IG, 0.68 ±0.16; CG, 1.09 ±0.47
LC-PUFA (weigh % FA). EPA: IG, 1.96 ±0.79; CG, 0.27 ±0.06
DHA: IG, 2.94 ±0.88; CG, 0.73 ±0.08; AA: IG, 5.39 ±1.16; CG, 5.83 ±0.98
AA, a achidonic acid; CG, con ol g oup; CT, con olled ial; DHA, docosahexaenoic acid (22:6, n-3); EPA, eicosapen aenoic acid (20:5, n-3); F, emale; FA, a y acid; HDL, high densi y
lipop o ein choles e ol; IG, in e en ion g oup; LC-PUFA, long chain polyunsa u a ed a y acid; LDL, low-densi y lipop o ein choles e ol; MB, memb ane; PC, phospha idylcholine;
PEA, phospha idyle hanolamine; RCT, andomized con olled ial; TC, o al choles e ol;. TG, iglyce ide.
1
Values (a en y) ep esen he ange o he mean
±
SD, as epo ed in he
co esponding a icle. 2Values ep esen he mean o mean ±SD, as epo ed in he co esponding a icle.
Nu ien s 2019,11, 1537 7 o 14
Table 3. E ec s o LC-PUFA supplemen a ion on isual unc ion in 237 subjec s in con olled ials.
Re e ence nAge 1In e en ion Type and Du a ion o
In e en ion Ou come Measu e Resul s 2Conclusion
Demmelmai e al.
(2018) [27]109 5–13 yea s DHA capsules (IG1, 0.1–1.8 mg/kg/day;
IG2, 1.9–7 mg/kg/day) RCT—6 mon hs Change in P100 wa e
la ency (ms)
Pa e n- e e sal. 15: IG1,
0.5 ±8.7; IG2, −0.6 ±4.7;
CG, 1.3 ±3.5
No signi ican di e ences
Agos oni e al.
(2006) [31]42 (22 F) 20 ±6.9 weeks
Tes - ea men o mula (DHA, 0.3 g/100 g
FA. RCT—12 mon hs P100 wa e (pa e n
VEP) and P1 peak ( lash
VEP) la encies (ms)
Pa e n- e e sal: IG, 120 ±
24; CG, 107 ±8No signi ican di e ences
Omega 6: omega 3 a io, 2.5:1) Flash: IG, 108 ±15; CG, 115
±24
Beblo e al. (2001)
[33]66 (34 F) 6.6 ±1.5 yea s (CG) Fish oil capsules (DHA, 15 mg/kg/day;
EPA, 22.5 mg/kg/day)
CT—3 mon hs Change in P100 wa e
la ency No da a Signi ican dec ease in P100
wa e la ency (5’, 15’)
CG heal hy child en
Agos oni e al.
(2000) [34]20 (9 F)
10.7 ±2.4 yea s (IG) Fish oil capsules (DHA, 15 mg/kg/day)
RCT—12 mon hs P100 wa e la ency (ms)
Pa e n- e e sal. 60’: IG,
104
±
4; CG, 109
±
9. 15’: IG,
107 ±6; CG, 118 ±11. Signi ican dec ease in P100
wa e la ency (15’, 2 Hz-1 J)
Flash. 1 Hz-2 J: IG, 113
±
10;
CG, 114 ±8.
10.5 ±2.8 yea s (CG) AA:DHA a io, 1:1 2 Hz-1 J: IG, 111 ±12; CG,
121 ±8
AA, a achidonic acid; CG, con ol g oup; CT, con olled ial; DHA, docosahexaenoic acid (22:6, n-3); F, emale; EPA, eicosapen aenoic acid (20:5, n-3); FA, a y acids; IG, in e en ion
g oup; RCT, andomized con olled ial; VEP, isual e oked po en ials.
1
Values (a en y) ep esen he ange o mean
±
SD, as epo ed in he co esponding a icle.
2
Values ep esen
he mean ±SD, as epo ed in he co esponding a icle.
Nu ien s 2019,11, 1537 8 o 14
Table 4. E ec s o LC-PUFA supplemen a ion on neu ocogni i e unc ion in 238 subjec s in con olled ials.
Re e ence nAge 1In e en ion Type and Time o
In e en ion Ou come Measu e Resul s 2Conclusion
Demmelmai e al.
(2018) [27]109 5–13 yea s DHA capsules (IG1, 0.1–1.8 mg/kg/day;
IG2, 1.9–7 mg/kg/day) RCT—6 mon hs Changes in mo ome ic Ros ock–Ose e zky
scale and Ra en´s p og essi e ma ices
Ros ock–Ose e zky scale:
IG1, 4.2 ±6.3; IG2, 0.8 ±9.1;
CG, 2.9 ±7.0 No signi ican di e ences
Ra en’s p og essi e
ma ices: IG1, 2.2 ±15.8;
IG2, 1.6 ±13.8; CG, 9.5 ±
13.5
Yi e al. (2011) [28] 33 (33 F) 12–47 yea s DHA capsules (10 mg/kg/day) RCT—4.5 mon hs
Ve bal abili y (Peabody pic u e ocabula y es ,
hi d edi ion), execu i e unc ion
(Delis-Kaplan execu i e unc ion sys em), and
cogni i e p ocessing speed
(Woodcock–Johnson III es s o cogni i e
abili y and achie emen )
Cogni i e p ocessing speed,
ac o sco e: IG, 98.8 ±5.3;
CG, 101 ±5.4 No signi ican di e ences
Cogni i e inhibi ion: IG,
11.3 ±1.5; CG, 11.4 ±1.5
Cogni i e lexibili y: IG,
11.1 ±1.4; CG, 10.8 ±1.4
Kole zko e al.
(2009) [29]54 6.3 ±0.6 yea s Fish oil capsules (DHA, 15 mg/kg/day;
EPA, 22.5 mg/kg/day)
CT—3 mon hs Changes in mo ome ic Ros ock–Ose e zky
scale No da a Signi ican imp o emen in ine mo o skills (especially coin
so ing), dynamic balance, and o al sco e in in e en ion g oup
CG, heal hy child en
Agos oni e al.
(2006) [31]42 (22 F) 20 ±6.9 weeks
Tes - ea men o mula (DHA, 0.3 g/100 g
FA. RCT—12 mon hs
Men al and psychomo o de elopmen (Bailey
es , second edi ion)
Men al de elopmen : IG,
92.67 ±16.02; CG, 93.19 ±
16.60 No signi ican di e ences
Physical de elopmen : IG,
92
±
13.32; CG, 97.69
±
15.57
Omega 6: omega 3 a io, 2.5:1)
AA, a achidonic acid; CG, con ol g oup; CT, con olled ial; DHA, docosahexaenoic acid (22:6, n-3); EPA, eicosapen aenoic acid (20:5, n-3); F, emale; FA, a y acids; IG, in e en ion
g oup; LC-PUFA, long chain polyunsa u a ed a y acid; RCT, andomized con olled ial; VEP, isual e oked po en ials.
1
Values (a en y) ep esen he ange, mean, o mean
±
SD, as
epo ed in he co esponding a icle. 2Values ep esen he mean ±SD, as epo ed in he co esponding a icle.
Nu ien s 2019,11, 1537 9 o 14
4. Discussion
This sys ema ic e iew o con olled ials ega ding LC-PUFA supplemen a ion in child en
and adul s wi h PKU e eals ha he addi ion o DHA a doses
≥
10 mg/kg/day o he pa ien ’s
Phe- es ic ed die dec eases VEP la encies. Howe e , no conclusi e e idence suppo s a ela ionship
be ween LC-PUFA supplemen a ion and neu ocogni i e ou comes in hese pa ien s.
DHA and AA a e he mos impo an LC-PUFAs o he n-3 and n-6 se ies, espec i ely. These key
s uc u al componen s o neu onal cell memb anes a e o c ucial impo ance o b ain de elopmen
and e inal unc ion [
37
]. In andomized clinical ials, LC-PUFA supplemen a ion is associa ed wi h
imp o ed isual and cogni i e ma u a ion in ull- e m and, in pa icula , p e e m in an s [
38
–
40
].
These ou comes in p e e m in an s may be linked o he g ea e p edisposi ion o hese child en o
LC-PUFA de iciency due o e al acc e ion o DHA (which usually occu s du ing he hi d imes e ),
an inabili y o con e p ecu so a y acids o DHA, and low pos na al DHA in ake [41].
PKU pa ien s a e ano he popula ion a isk o LC-PUFA de iciency; he ypical Phe- es ic ed
die s o hese pa ien s p o ide low amoun s o animal p oduc s, which a e he main sou ce o
LC-PUFAs [
42
,
43
]. Mo eo e , excess Phe is ca abolized o phenylpy u a e and phenyllac a e, which
a e epo ed o inhibi endogenous syn hesis o DHA and AA [
44
]. A 2013 sys ema ic e iew and
me a-analysis o nine case con ol s udies and six andomized con olled ials concluded ha PKU
pa ien s ha e signi ican ly lowe le els o bo h DHA and AA in all bioma ke s s udied han heal hy
con ols [
24
]. In line wi h his subop imal LC-PUFA s a us in PKU pa ien s, s udies o child en wi h
amino acid me abolism diso de s ha e desc ibed educed LC-PUFA in ake (a consequence o die a y
p o ein es ic ion) and lowe plasma and e y h ocy e memb ane concen a ions o DHA han heal hy
con ols [
45
–
47
]. The esul s o s udies o AA s a us in hese pa ien s a e inconclusi e ( anging om
no mal o educed), sugges ing ha endogenous syn hesis may be su icien o ensu e adequa e AA
s a us in some cases [48].
The indings o his sys ema ic e iew indica e ha DHA supplemen a ion in PKU pa ien s
signi ican ly inc eases DHA le els in plasma and/o e y h ocy e memb anes [
27
–
35
]. I should be
no ed ha e y h ocy e a y acid composi ion yields mo e in o ma ion ega ding long- e m LC-PUFA
s a us and is less in luenced by as ing, appea ing o be a mo e aluable bioma ke [
49
,
50
]. While he
mos commonly adminis e ed dose o DHA was 10–15 mg/kg/day, signi ican inc eases in DHA le els
we e obse ed e en wi h lowe doses (0.1–7 mg/kg/day) [
27
]. The e a e insu icien da a o de ine an
op imal LC-PUFA supplemen a ion dose o PKU pa ien s o di e en age g oups, and op imal DHA
in ake in in an s and child en emains a opic o deba e acco ding o bo h he ESPGHAN Commi ee on
Nu i ion and The Eu opean Food Sa e y Au ho i y Panel on Die e ic P oduc s, Nu i ion and Alle gies
(NDA). Despi e he lack o app op ia e da a on which o base die a y e e ence alues o pedia ic
pa ien s, he a o emen ioned o ganiza ions ha e p oposed an in ake o 100 mg/day o pa ien s aged
6–24 mon hs and 250 mg/day o hose aged 2 yea s and olde [51].
VEP es ing was conduc ed o assess cen al ne ous sys em (CNS) unc ion in ou o he s udies
included in his e iew [
27
,
31
,
33
,
34
]. VEP is widely used in s udies o neu al ma u a ion as i p o ides
a sensi i e means o assessing he unc ion o a majo CNS pa hway. P100 wa e la ency is conside ed
he mos eliable clinical indica o , as is he a iable leas a ec ed by echnical ac o s and he deg ee
o pa ien coope a ion [
52
]. Longe VEP la encies, which a e obse ed in PKU pa ien s no ecei ing
LC-PUFA supplemen a ion [
52
], indica e a lowe speed o in o ma ion p ocessing om he e ina o
he isual co ex. I should be no ed ha he con olled ials (CTs) and andomized con olled ials
(RCTs) in which sho e P100 wa e la encies we e obse ed a e in e en ion we e hose in which
he pa ien s ecei ed highe doses o DHA (10–15 mg/kg/day) [
33
,
34
]. In he wo RCTs [
27
,
31
] ha
epo ed no di e ences in VEP la encies a e LC-PUFA supplemen a ion, pa ien s ecei ed lowe
(0.1–7 mg/kg/day) o uncon olled doses (i.e., a LC-PUFA-supplemen ed, Phe- ee o mula) o DHA. In
he la e s udy [
31
], highe le els o DHA in e y h ocy e memb anes we e signi ican ly co ela ed wi h
a sho e P100 wa e la ency a e adjus men o age. This obse a ion sugges s ha he LC-PUFA
in ake o hese pa ien s was i egula and, in many cases, insu icien o al e he clinical ou come.
