Higher plasma selenium concentrations are associated with increased odds of prevalent type 2 diabetes

Lindsay N. Kohler, Ana Florea, Connor P. Kelley, Hsiao-Hui Chow, Chiu-Hsieh Hsu, Ken Batai, Kathylynn Saboda, Michael P Lance, Elizabeth T Jacobs

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Background: Selenium, an essential trace element, has been investigated as a potential cancer prevention agent. However, several studies have indicated that selenium supplementation may be associated with an increased risk of type 2 diabetes (T2D), although an equivocal relation of this nature requires confirmation. Objective: We examined the association between baseline plasma concentrations of selenium and the prevalence of T2D, as well as whether participant characteristics or intake of other antioxidant nutrients modified this relation. Methods: We conducted cross-sectional analyses of 1727 participants from the Selenium Trial, a randomized clinical trial of selenium supplementation for colorectal adenoma chemoprevention that had data for baseline selenium plasma concentrations, T2D status, and dietary intake. Logistic regression modeling was used to evaluate the associations between plasma selenium concentrations and prevalent T2D, adjusting for confounding factors. Heterogeneity of effect by participant characteristics was evaluated utilizing likelihood-ratio tests. Results: Mean ± SD plasma selenium concentrations for those with T2D compared with those without T2D were 143.6 ± 28.9 and 138.7 ± 27.2 ng/mL, respectively. After adjustment for confounding, higher plasma selenium concentrations were associated with a higher prevalence of T2D, with ORs (95% CIs) of 1.25 (0.80, 1.95) and 1.77 (1.16, 2.71) for the second and third tertiles of plasma selenium, respectively, compared with the lowest tertile (P-trend = 0.007). No significant effect modification was observed for age, sex, body mass index, smoking, or ethnicity. Increased odds of T2D were seen among those who were in the highest tertile of plasma selenium and the highest category of intake of β-cryptoxanthin (P-trend = 0.03) and lycopene (P-trend = 0.008); however, interaction terms were not significant. Conclusions: These findings show that higher plasma concentrations of selenium were significantly associated with prevalent T2D among participants in a selenium supplementation trial. Future work is needed to elucidate whether there are individual characteristics, such as blood concentrations of other antioxidants, which may influence this relation.

Original languageEnglish (US)
Pages (from-to)1333-1340
Number of pages8
JournalJournal of Nutrition
Volume148
Issue number8
DOIs
StatePublished - Jan 1 2018

Fingerprint

Selenium
Type 2 Diabetes Mellitus
Antioxidants
Chemoprevention
Trace Elements
Adenoma
Body Mass Index
Randomized Controlled Trials
Cross-Sectional Studies
Logistic Models
Smoking

Keywords

  • Antioxidants
  • Selenium
  • Supplementation
  • Trace elements
  • Type 2 diabetes

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Nutrition and Dietetics

Cite this

Higher plasma selenium concentrations are associated with increased odds of prevalent type 2 diabetes. / Kohler, Lindsay N.; Florea, Ana; Kelley, Connor P.; Chow, Hsiao-Hui; Hsu, Chiu-Hsieh; Batai, Ken; Saboda, Kathylynn; Lance, Michael P; Jacobs, Elizabeth T.

In: Journal of Nutrition, Vol. 148, No. 8, 01.01.2018, p. 1333-1340.

Research output: Contribution to journalArticle

Kohler, Lindsay N. ; Florea, Ana ; Kelley, Connor P. ; Chow, Hsiao-Hui ; Hsu, Chiu-Hsieh ; Batai, Ken ; Saboda, Kathylynn ; Lance, Michael P ; Jacobs, Elizabeth T. / Higher plasma selenium concentrations are associated with increased odds of prevalent type 2 diabetes. In: Journal of Nutrition. 2018 ; Vol. 148, No. 8. pp. 1333-1340.
@article{e8a2c16b20674aa68e9741b452bfd7eb,
title = "Higher plasma selenium concentrations are associated with increased odds of prevalent type 2 diabetes",
abstract = "Background: Selenium, an essential trace element, has been investigated as a potential cancer prevention agent. However, several studies have indicated that selenium supplementation may be associated with an increased risk of type 2 diabetes (T2D), although an equivocal relation of this nature requires confirmation. Objective: We examined the association between baseline plasma concentrations of selenium and the prevalence of T2D, as well as whether participant characteristics or intake of other antioxidant nutrients modified this relation. Methods: We conducted cross-sectional analyses of 1727 participants from the Selenium Trial, a randomized clinical trial of selenium supplementation for colorectal adenoma chemoprevention that had data for baseline selenium plasma concentrations, T2D status, and dietary intake. Logistic regression modeling was used to evaluate the associations between plasma selenium concentrations and prevalent T2D, adjusting for confounding factors. Heterogeneity of effect by participant characteristics was evaluated utilizing likelihood-ratio tests. Results: Mean ± SD plasma selenium concentrations for those with T2D compared with those without T2D were 143.6 ± 28.9 and 138.7 ± 27.2 ng/mL, respectively. After adjustment for confounding, higher plasma selenium concentrations were associated with a higher prevalence of T2D, with ORs (95{\%} CIs) of 1.25 (0.80, 1.95) and 1.77 (1.16, 2.71) for the second and third tertiles of plasma selenium, respectively, compared with the lowest tertile (P-trend = 0.007). No significant effect modification was observed for age, sex, body mass index, smoking, or ethnicity. Increased odds of T2D were seen among those who were in the highest tertile of plasma selenium and the highest category of intake of β-cryptoxanthin (P-trend = 0.03) and lycopene (P-trend = 0.008); however, interaction terms were not significant. Conclusions: These findings show that higher plasma concentrations of selenium were significantly associated with prevalent T2D among participants in a selenium supplementation trial. Future work is needed to elucidate whether there are individual characteristics, such as blood concentrations of other antioxidants, which may influence this relation.",
keywords = "Antioxidants, Selenium, Supplementation, Trace elements, Type 2 diabetes",
author = "Kohler, {Lindsay N.} and Ana Florea and Kelley, {Connor P.} and Hsiao-Hui Chow and Chiu-Hsieh Hsu and Ken Batai and Kathylynn Saboda and Lance, {Michael P} and Jacobs, {Elizabeth T}",
year = "2018",
month = "1",
day = "1",
doi = "10.1093/jn/nxy099",
language = "English (US)",
volume = "148",
pages = "1333--1340",
journal = "Journal of Nutrition",
issn = "0022-3166",
publisher = "American Society for Nutrition",
number = "8",

}

TY - JOUR

T1 - Higher plasma selenium concentrations are associated with increased odds of prevalent type 2 diabetes

AU - Kohler, Lindsay N.

AU - Florea, Ana

AU - Kelley, Connor P.

AU - Chow, Hsiao-Hui

AU - Hsu, Chiu-Hsieh

AU - Batai, Ken

AU - Saboda, Kathylynn

AU - Lance, Michael P

AU - Jacobs, Elizabeth T

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Background: Selenium, an essential trace element, has been investigated as a potential cancer prevention agent. However, several studies have indicated that selenium supplementation may be associated with an increased risk of type 2 diabetes (T2D), although an equivocal relation of this nature requires confirmation. Objective: We examined the association between baseline plasma concentrations of selenium and the prevalence of T2D, as well as whether participant characteristics or intake of other antioxidant nutrients modified this relation. Methods: We conducted cross-sectional analyses of 1727 participants from the Selenium Trial, a randomized clinical trial of selenium supplementation for colorectal adenoma chemoprevention that had data for baseline selenium plasma concentrations, T2D status, and dietary intake. Logistic regression modeling was used to evaluate the associations between plasma selenium concentrations and prevalent T2D, adjusting for confounding factors. Heterogeneity of effect by participant characteristics was evaluated utilizing likelihood-ratio tests. Results: Mean ± SD plasma selenium concentrations for those with T2D compared with those without T2D were 143.6 ± 28.9 and 138.7 ± 27.2 ng/mL, respectively. After adjustment for confounding, higher plasma selenium concentrations were associated with a higher prevalence of T2D, with ORs (95% CIs) of 1.25 (0.80, 1.95) and 1.77 (1.16, 2.71) for the second and third tertiles of plasma selenium, respectively, compared with the lowest tertile (P-trend = 0.007). No significant effect modification was observed for age, sex, body mass index, smoking, or ethnicity. Increased odds of T2D were seen among those who were in the highest tertile of plasma selenium and the highest category of intake of β-cryptoxanthin (P-trend = 0.03) and lycopene (P-trend = 0.008); however, interaction terms were not significant. Conclusions: These findings show that higher plasma concentrations of selenium were significantly associated with prevalent T2D among participants in a selenium supplementation trial. Future work is needed to elucidate whether there are individual characteristics, such as blood concentrations of other antioxidants, which may influence this relation.

AB - Background: Selenium, an essential trace element, has been investigated as a potential cancer prevention agent. However, several studies have indicated that selenium supplementation may be associated with an increased risk of type 2 diabetes (T2D), although an equivocal relation of this nature requires confirmation. Objective: We examined the association between baseline plasma concentrations of selenium and the prevalence of T2D, as well as whether participant characteristics or intake of other antioxidant nutrients modified this relation. Methods: We conducted cross-sectional analyses of 1727 participants from the Selenium Trial, a randomized clinical trial of selenium supplementation for colorectal adenoma chemoprevention that had data for baseline selenium plasma concentrations, T2D status, and dietary intake. Logistic regression modeling was used to evaluate the associations between plasma selenium concentrations and prevalent T2D, adjusting for confounding factors. Heterogeneity of effect by participant characteristics was evaluated utilizing likelihood-ratio tests. Results: Mean ± SD plasma selenium concentrations for those with T2D compared with those without T2D were 143.6 ± 28.9 and 138.7 ± 27.2 ng/mL, respectively. After adjustment for confounding, higher plasma selenium concentrations were associated with a higher prevalence of T2D, with ORs (95% CIs) of 1.25 (0.80, 1.95) and 1.77 (1.16, 2.71) for the second and third tertiles of plasma selenium, respectively, compared with the lowest tertile (P-trend = 0.007). No significant effect modification was observed for age, sex, body mass index, smoking, or ethnicity. Increased odds of T2D were seen among those who were in the highest tertile of plasma selenium and the highest category of intake of β-cryptoxanthin (P-trend = 0.03) and lycopene (P-trend = 0.008); however, interaction terms were not significant. Conclusions: These findings show that higher plasma concentrations of selenium were significantly associated with prevalent T2D among participants in a selenium supplementation trial. Future work is needed to elucidate whether there are individual characteristics, such as blood concentrations of other antioxidants, which may influence this relation.

KW - Antioxidants

KW - Selenium

KW - Supplementation

KW - Trace elements

KW - Type 2 diabetes

UR - http://www.scopus.com/inward/record.url?scp=85055573837&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055573837&partnerID=8YFLogxK

U2 - 10.1093/jn/nxy099

DO - 10.1093/jn/nxy099

M3 - Article

C2 - 29924331

AN - SCOPUS:85055573837

VL - 148

SP - 1333

EP - 1340

JO - Journal of Nutrition

JF - Journal of Nutrition

SN - 0022-3166

IS - 8

ER -