TY - JOUR
T1 - A dietary biomarker approach captures compliance and cardiometabolic effects of a healthy nordic diet in individuals with metabolic syndrome
AU - Marklund, Matti
AU - Magnusdottir, Ola K.
AU - Rosqvist, Fredrik
AU - Cloetens, Lieselotte
AU - Landberg, Rikard
AU - Kolehmainen, Marjukka
AU - Brader, Lea
AU - Hermansen, Kjeld
AU - Poutanen, Kaisa
AU - Herzig, Karl-Heinz
AU - Hukkanen, Janne
AU - Savolainen, Markku J.
AU - Dragsted, Lars O.
AU - Schwab, Ursula
AU - Paananen, Jussi
AU - Uusitupa, Matti
AU - Åkesson, Björn
AU - Thorsdottir, Inga
AU - Risérus, Ulf
PY - 2014
Y1 - 2014
N2 - Assessment of compliance with dietary interventions is
necessary to understand the observed magnitude of the
health effects of the diet per se. To avoid reporting
bias, different dietary biomarkers (DBs) could be used
instead of self-reported data. However, few studies
investigated a combination of DBs to assess compliance
and its influence on cardiometabolic risk factors. The
objectives of this study were to use acombination of DBs
toassesscompliance andto investigatehowahealthyNordic
diet(ND) influencescardiometabolic risk factors in
participantswith high apparent compliance comparedwith
the whole study population. From a recently conducted
isocaloric randomized trial, SYSDIET (Systems Biology in
Controlled Dietary Interventions and Cohort Studies), in
166 individuals withmetabolic syndrome, several DBs were
assessed to reflect different key components of the ND:
canola oil (serumphospholipid a-linolenic acid), fatty
fish [eicosapentaenoic acid (EPA) and docosahexaenoic
acid (DHA)], vegetables (plasma b-carotene), and whole
grains (plasma alkylresorcinols). High-fat dairy intake
(expectedly low in the ND) was reflected by serum
pentadecanoic acid. All participants with biomarker data
(n = 154) were included in the analyses. Biomarkers were
combined by using a biomarker rank score (DB score) and
principal component analysis (PCA). The DB score was then
used to assess compliance. During the intervention,
median concentrations of alkylresorcinols, a-linolenic
acid,EPA, and DHA were>25%higher in the ND individuals
than in the controls (P>0.05),where as median
concentrations of pentadecanoic acid were14%higher in
controls (P>0.05). MedianDBscorewas57%higher in the ND
than in controls (P > 0.001) during the intervention, and
participants were ranked similarly by DB score and PCA
score. Overall, estimates of group difference in
cardiometabolic effects generally appeared to be greater
among compliant participants than in the whole study
population (e.g., estimates of treatment effects on blood
pressure and lipoproteinswere;1.5- to 2-fold greater in
the most compliant participants), suggesting that poor
compliance attenuated the dietary effects. With adequate
consideration of their limitations, DB combinations
(e.g., DB score) could be useful for assessing compliance
in intervention studies investigating cardiometabolic
effects of healthy dietary patterns. The study was
registered at clinicaltrials.gov as NCT00992641
AB - Assessment of compliance with dietary interventions is
necessary to understand the observed magnitude of the
health effects of the diet per se. To avoid reporting
bias, different dietary biomarkers (DBs) could be used
instead of self-reported data. However, few studies
investigated a combination of DBs to assess compliance
and its influence on cardiometabolic risk factors. The
objectives of this study were to use acombination of DBs
toassesscompliance andto investigatehowahealthyNordic
diet(ND) influencescardiometabolic risk factors in
participantswith high apparent compliance comparedwith
the whole study population. From a recently conducted
isocaloric randomized trial, SYSDIET (Systems Biology in
Controlled Dietary Interventions and Cohort Studies), in
166 individuals withmetabolic syndrome, several DBs were
assessed to reflect different key components of the ND:
canola oil (serumphospholipid a-linolenic acid), fatty
fish [eicosapentaenoic acid (EPA) and docosahexaenoic
acid (DHA)], vegetables (plasma b-carotene), and whole
grains (plasma alkylresorcinols). High-fat dairy intake
(expectedly low in the ND) was reflected by serum
pentadecanoic acid. All participants with biomarker data
(n = 154) were included in the analyses. Biomarkers were
combined by using a biomarker rank score (DB score) and
principal component analysis (PCA). The DB score was then
used to assess compliance. During the intervention,
median concentrations of alkylresorcinols, a-linolenic
acid,EPA, and DHA were>25%higher in the ND individuals
than in the controls (P>0.05),where as median
concentrations of pentadecanoic acid were14%higher in
controls (P>0.05). MedianDBscorewas57%higher in the ND
than in controls (P > 0.001) during the intervention, and
participants were ranked similarly by DB score and PCA
score. Overall, estimates of group difference in
cardiometabolic effects generally appeared to be greater
among compliant participants than in the whole study
population (e.g., estimates of treatment effects on blood
pressure and lipoproteinswere;1.5- to 2-fold greater in
the most compliant participants), suggesting that poor
compliance attenuated the dietary effects. With adequate
consideration of their limitations, DB combinations
(e.g., DB score) could be useful for assessing compliance
in intervention studies investigating cardiometabolic
effects of healthy dietary patterns. The study was
registered at clinicaltrials.gov as NCT00992641
U2 - 10.3945/jn.114.193771
DO - 10.3945/jn.114.193771
M3 - Article
SN - 0022-3166
VL - 144
SP - 1642
EP - 1649
JO - Journal of Nutrition
JF - Journal of Nutrition
IS - 10
ER -