Amino acids, lipid metabolites, and ferritin as potential mediators linking red meat consumption to type 2 diabetes

In the paper

“Our sensitivity analysis evaluating different subtypes of exposure revealed that the attenuation of the associated diabetes risk after mutual adjustment for all selected biomarkers was also significant if only unprocessed red meat (P = 0.007) or processed meat (P = 0.002) consumption was considered the exposure variable (reference: model 3).”

In the paper

“When using the BMI-adjusted model 3 as reference, the attenuation was significant after adjustment for diacyl phosphatidylcholine 36:4, hydroxy-sphingomyelin 14:1, and glycine and borderline significant after adjustment for diacyl phosphatidylcholine 38:4 (P = 0.05) and ferritin (P = 0.06).”

In the paper

“Diacyl phosphatidylcholines 36:4 and 38:4 were highly correlated, and adjustment for diacyl phosphatidylcholine 36:4 did not further attenuate the association between total red meat consumption and type 2 diabetes risk if diacyl phosphatidylcholine 38:4 was also included in the same model (data not shown).”

In the paper

“In the fully adjusted model 3, the proportion of associated diabetes risk explainable by simultaneous biomarker adjustment (median and IQR from 1000 bootstrap replicates) was 49% (IQR: 29%, 89%) for unprocessed red meat consumption, whereas it was 71% (IQR: 45%, 115%) for processed meat consumption (Supplemental Figure 8).”

In the paper

“Thus, results were robust to the choice of selection strategy and P value adjustment method for glycine, ferritin, and phosphatidylcholines 36:4 and 38:4 and were somewhat sensitive to that choice for lysophosphatidylcholine 17:0 and hydroxy-sphingomyelin 14:1.”

In the paper

“Our sensitivity analysis applying different P value adjustments and differently defined selection criteria showed that selection was very robust for most potential mediators but was somewhat sensitive to these choices for lysophosphatidylcholine 17:0 and hydroxy-sphingomyelin 14:1.”

In the paper

“In conclusion, high ferritin, low glycine, and altered hepaticderived lipids in the circulation were associated with both total red meat consumption and diabetes risk.”

In the paper

“Components of red meat that were proposed to affect diabetogenic pathways include heme iron, cholesterol, fatty acids (FAs), amino acids, and advanced glycation and lipoxidation end products (1), among others.”

What the sources say

• Not checked yetcites [1] Feskens, 2013, Meat consumption, diabetes, and its complications, Curr Diab Rep source

In the paper

“Habitual diet was assessed with validated semiquantitative food-frequency questionnaires.”

What the sources say

• Not checked yetcites [16] Kroke, 1999, Am J Clin Nutr source

In the paper

“To reduce the probability of spurious associations, we selected biomarkers only if they were associated with total red meat consumption in both sexes in stratified analyses.”

In the paper

“In EPIC-Potsdam, metabolomic data were used to identify lipid metabolites and amino acids as biomarkers of diabetes risk (12).”

What the sources say

• Not checked yetcites [12] Floegel, 2013, Identification of serum metabolites associated with risk of type 2 diabetes using a targeted metabolomic approach, Diabetes source

In the paper

“Reported associations of red meat consumption with biomarkers of inflammation, glucose metabolism, and hepatocellular and oxidative stress were mostly explainable by differences in BMI (7,8).”

What the sources say

• Not checked yetcites [7] Montonen, 2013, Consumption of red meat and whole-grain bread in relation to biomarkers of obesity, inflammation, glucose metabolism and oxidative stress, Eur J Nutr source
• Not checked yetcites [8] Ley, 2014, Associations between red meat intake and biomarkers of inflammation and glucose metabolism in women, Am J Clin Nutr source

In the paper

“Recently, change in red meat consumption was observed to be related to a change in diabetes risk among 3 large US cohorts (4), suggesting a dose-response relation.”

What the sources say

• Not checked yetcites [4] Pan, 2013, Changes in red meat consumption and subsequent risk of type 2 diabetes mellitus: three cohorts of US men and women, JAMA Intern Med source

In the paper

“On the basis of a previous evaluation of the reliability of measurements (23), we excluded metabolites below the limit of detection (n = 30) and those with very high analytic variance (n = 6).”

What the sources say

• Not checked yetcites [23] Floegel, 2011, Reliability of serum metabolite concentrations over a 4-month period using a targeted metabolomic approach, PLoS ONE source

In the paper

“We selected mediators of the red meat-diabetes association according to 4 predefined mediation criteria (25) (see Supplemental Figure 1A).”

What the sources say

• Not checked yetcites [25] Baron, 1986, The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations, J Pers Soc Psychol source

In the paper

“The case-cohort design was accounted for by Prentice weighting (26).”

What the sources say

• Not checked yetcites [26] Prentice, 1988, Aspects of the use of relative risk models in the design and analysis of cohort studies and prevention trials, Stat Med source

In the paper

“Furthermore, the red meat-associated diabetes risk was also observed in models in which red meat was substituted by foods with a similar macronutrient content (2), suggesting the relation to be independent of macronutrient composition of the diet.”

What the sources say

• Not checked yetcites [2] Pan, 2011, Red meat consumption and risk of type 2 diabetes: 3 cohorts of US adults and an updated meta-analysis, Am J Clin Nutr source

In the paper

“The P values were controlled for false discovery rate (FDR) within metabolite class by using the linear step-up method by Benjamini and Hochberg (30).”

What the sources say

• Not checked yetcites [30] Benjamini, 1995, Controlling the false discovery rate—a practical and powerful approach to multiple testing, J R Stat Soc B source

In the paper

“Potential confounders to be included as covariates were a priori selected based on the literature on the association between red meat intake and diabetes risk (2, 3) and prior investigations in EPIC-Potsdam regarding risk factors of type 2 diabetes (5), dietary patterns (27), and determinants of metabolic profiles (28,29).”

What the sources say

• Not checked yetcites [2] Pan, 2011, Red meat consumption and risk of type 2 diabetes: 3 cohorts of US adults and an updated meta-analysis, Am J Clin Nutr source
• Not checked yetcites [3] Benedinelli, 2013, Association between dietary meat consumption and incident type 2 diabetes: the EPIC-InterAct study, Diabetologia source
• Not checked yetcites [5] Mühlenbruch, 2014, Update of the German Diabetes Risk Score and external validation in the German MONICA/KORA study, Diabetes Res Clin Pract source
• Not checked yetcites [27] Schulze, 2001, Dietary patterns and their association with food and nutrient intake in the European Prospective Investigation into Cancer and Nutrition (EPIC)–Potsdam study, Br J Nutr source

In the paper

“The red meat-associated diabetes risk was largely attenuated after adjustment for selected biomarkers, which is consistent with the presumed mediation hypothesis.”

In the paper

“In the literature, several lines of experimental evidence relate lipid metabolism to the pathogenesis of type 2 diabetes, among them a challenged mitochondrial metabolism, altered cellular signaling processes at the membrane, and modulation of gene expression (34,35).”

What the sources say

• Not checked yetcites [34] Ye, 2013, Mechanisms of insulin resistance in obesity, Front Med source
• Not checked yetcites [35] Amati, 2012, Revisiting the diacylglycerol-induced insulin resistance hypothesis, Obes Rev source

In the paper

“In our study, high ferritin, low glycine, and altered hepatic-derived lipid concentrations in the circulation were associated with total red meat consumption and, independent of red meat, with diabetes risk.”

In the paper

“Given that an observational study is generally prone to confounding and other sources of bias, interpretation of our findings with regard to biological paths is rather hypothesis generating than able to proof their causal nature (45).”

What the sources say

• Not checked yetcites [45] Hafeman, 2009, “Proportion explained”: a causal interpretation for standard measures of indirect effect?, Am J Epidemiol source

In the paper

“Thus, our results generate and underline experimentally testable hypotheses that can inform future dietary interventions in terms of design and biomarker assessment (46).”

What the sources say

• Not checked yetcites [46] Pearl, 2010, Causal Inference, Int J Biostat source

In the paper

“described by us and others (12,41).”

What the sources say

• Not checked yetcites [12] Floegel, 2013, Identification of serum metabolites associated with risk of type 2 diabetes using a targeted metabolomic approach, Diabetes source
• Not checked yetcites [41] Wang-Sattler, 2012, Novel biomarkers for pre-diabetes identified by metabolomics, Mol Syst Biol source

In the paper

“Consistently, endogenous FA composition in several lipid compartments has been related to the risk of type 2 diabetes in several prospective cohorts, including EPIC-Potsdam (36).”

What the sources say

• Not checked yetcites [36] Kröger, 2011, Erythrocyte membrane phospholipid fatty acids, desaturase activity, and dietary fatty acids in relation to risk of type 2 diabetes in the European Prospective Investigation into Cancer and Nutrition (EPIC)–Potsdam Study, Am J Clin Nutr source

In the paper

“Underlying mechanisms seem to be manifold and likely include oxidative stress and modulation of intracellular signaling cascades (44).”

What the sources say

• Not checked yetcites [44] Simcox, 2013, Iron and diabetes risk, Cell Metab source

In the paper

“This should have reduced the likelihood of false discoveries.”

In the paper

“Furthermore, estrogen is a modulator of mitochondrial function, and associations of red meat consumption with acylcarnitines might be sex specific.”

In the paper

“For example, advanced glycation end products, trans FA, nitrites and nitrates, and methylamines are discussed as meat type-specific mediators (1).”

What the sources say

• Not checked yetcites [1] Feskens, 2013, Meat consumption, diabetes, and its complications, Curr Diab Rep source

In the paper

“Furthermore, we evaluated the robustness of our results in a bootstrapping procedure.”

In the paper

“Still, random measurement error may have masked some associations.”

In the paper

“In addition, there is the possibility that cases remained undiagnosed during follow-up.”

In the paper

“However, this misclassification should not bias the associations between red meat consumption and diabetes given that false-positive case definitions should have been rare due to the strict verification procedures in EPIC-Potsdam (49).”

What the sources say

• Not checked yetcites [49] Rothman, 2008

In the paper

“This effect, however, may reflect mitochondrial metabolism under acute dietary challenge (48) rather than effects of long-term habitual diet as assessed in our study.”

What the sources say

• Not checked yetcites [48] Krug, 2012, The dynamic range of the human metabolome revealed by challenges, FASEB J source

In the paper

“However, FFQs showed good reliability and relative validity to assess meat intake, and investigated metabolites showed good reliability in validation studies (23).”

What the sources say

• Not checked yetcites [23] Floegel, 2011, Reliability of serum metabolite concentrations over a 4-month period using a targeted metabolomic approach, PLoS ONE source

In the paper

“Still, the a priori design of our search strategy included arbitrary decisions, and P values lose their straightforward interpretability with regard to the probability of type I errors in a step-by-step selection procedure such as ours (47).”

What the sources say

• Not checked yetcites [47] Gelman, 2014, The statistical crisis in science, Am Sci source

In the paper

“Albeit large, the set of targeted metabolites was not exhaustive with regard to potential mediating mechanisms.”

In the paper

“Furthermore, acylcarnitines have previously been linked to red meat intake (38).”

What the sources say

• Not checked yetcites [38] O’Sullivan, 2011, Dietary intake patterns are reflected in metabolomic profiles: potential role in dietary assessment studies, Am J Clin Nutr source

In the paper

“Our results, however, cannot prove causality of the observed associations and, where possible, the suggested single relations should inform the design and the biomarker assessment of interventional studies.”