Abstract P407: Epicatechin intake and markers of glucose and insulin metabolism: the Zutphen elderly study

M. Obura, J.I. Dower, D. Kromhout, P.C.H. Hollman, J.M. Geleijnse, S.S. Soedamah-Muthu, J. de Goede

Research output: Contribution to journalAbstractAcademic


Background: Consumption of flavonoid-rich foods like cocoa and tea
is associated with a reduced risk of type 2 diabetes and improved
glycemic control. Cocoa and tea are both important dietary sources of
the flavonoid epicatechin. Our objective therefore was to examine
whether dietary epicatechin intake is associated with markers of
glucose and insulin metabolism.
Methods: We conducted a prospective study in 437 men from the
Zutphen Elderly Study aged 65–84 years at baseline (1985), free of
diabetes and cancer. Dietary epicatechin and other nutrients were
assessed at baseline using a validated cross-check dietary history
method. Glucose and insulin areas under the curve (AUC) were
determined in response to an oral glucose tolerance test (OGTT)
performed in 1990. Homeostasis model assessment of insulin resistance
(HOMA-IR) as well as Matsuda index were calculated from
fasting glucose (mmol/l) and insulin (mU/l) levels. Glycated hemoglobin
(HbA1c) was measured in 1993. Adjusted means and 95 %
confidence intervals (95 % CI) across quintiles of epicatechin intake
and trend analysis were calculated using multivariable linear regression.
Models were adjusted for age, BMI, socioeconomic status,
family history of diabetes, lifestyle (smoking, physical activity and
alcohol consumption), dietary factors (total energy and dietary fiber)
and cardiovascular disease.
Results: The median intake of epicatechin in the highest quintile (Q5)
was 26.6 mg/day compared to 5.5 mg/day in the lowest quintile (Q1).
Tea contributed 50 % of epicatechin intake, followed by apples
(29 %), cocoa (8 %) and red wine (2 %). Epicatechin intake was not
associated with any of the markers of glucose and insulin metabolism
(p trends[0.05 for all). The adjusted mean (95 % CI) for HbA1c (%)
was 4.9 (4.7–5.1) in Q1 and 4.7 (4.5–4.9) in Q5 (p trend = 0.32). The
adjusted means (95 % CI) for Q1 and Q5 were 90.5 (81.4–100.6)and
93.4 (83.6–104.4) (p trend = 0.52) for the Matsuda index, 891
(842–942) and 882 (832–935) (p trend = 0.94) for glucoseAUC
(mmol/L min), 4777 (4323–5279) and 4995 (4498.8–5547.4)
(p trend = 0.86) for insulinAUC (mU/L min) and 2.3 (2.0–2.5) and
2.1 (1.9–2.4) (p trend = 0.29) for HOMA-IR. There was no interaction
with age, family history of diabetes, smoking status and BMI.
Excluding men with cardiovascular disease (n = 57) did not alter the
Conclusions: In Dutch elderly non-diabetic men, epicatechin intake
was not associated with markers of glucose and insulin metabolism.
Original languageEnglish
Pages (from-to)915
Number of pages1
JournalEuropean Journal of Epidemiology
Issue number6
Publication statusPublished - 2015
EventHealthy Living: The European Congress of Epidemiology - MECC Maastricht, Maastricht, Netherlands
Duration: 25 Jun 201527 Jun 2015


  • epidemiology
  • diabetes

Fingerprint Dive into the research topics of 'Abstract P407: Epicatechin intake and markers of glucose and insulin metabolism: the Zutphen elderly study'. Together they form a unique fingerprint.

Cite this