Fructose as a Driver of Diabetes: An Incomplete View of the Evidence
2015; Elsevier BV; Volume: 90; Issue: 7 Linguagem: Inglês
10.1016/j.mayocp.2015.04.017
ISSN1942-5546
AutoresJohn L. Sievenpiper, Luc Tappy, Fred Brouns,
Tópico(s)Liver Disease Diagnosis and Treatment
ResumoWe are concerned that the article by DiNicolantonio et al1DiNicolantonio J.J. O'Keefe J.H. Lucan S.C. Added fructose: a principal driver of type 2 diabetes mellitus and its consequences.Mayo Clin Proc. 2015; 90: 372-381Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar published in the March 2015 issue of Mayo Clinic Proceedings that implicated added fructose as a driver of type 2 diabetes misrepresented the data by placing undue emphasis on low-quality evidence from ecological observations, animal models of fructose overfeeding, and selected human studies assessed in isolation. It also ignored important biological mechanisms by which fructose may assist in the metabolic handling of glucose. If one considers the totality of the highest-quality evidence from controlled feeding trials and prospective cohorts, then different conclusions are reached. A series of carefully conducted systematic reviews and meta-analyses2Livesey G. Taylor R. Fructose consumption and consequences for glycation, plasma triacylglycerol, and body weight: meta-analyses and meta-regression models of intervention studies.Am J Clin Nutr. 2008; 88: 1419-1437PubMed Scopus (211) Google Scholar, 3Cozma A.I. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on glycemic control in diabetes: a systematic review and meta-analysis of controlled feeding trials.Diabetes Care. 2012; 35: 1611-1620Crossref PubMed Scopus (154) Google Scholar, 4Sievenpiper J.L. de Souza R.J. Mirrahimi A. et al.Effect of fructose on body weight in controlled feeding trials: a systematic review and meta-analysis.Ann Intern Med. 2012; 156: 291-304Crossref PubMed Scopus (222) Google Scholar, 5Sievenpiper J.L. Carleton A.J. Chatha S. et al.Heterogeneous effects of fructose on blood lipids in individuals with type 2 diabetes: systematic review and meta-analysis of experimental trials in humans.Diabetes Care. 2009; 32: 1930-1937Crossref PubMed Scopus (133) Google Scholar, 6Wang D.D. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on postprandial triglycerides: a systematic review and meta-analysis of controlled feeding trials.Atherosclerosis. 2014; 232: 125-133Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 7Ha V. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on blood pressure: a systematic review and meta-analysis of controlled feeding trials.Hypertension. 2012; 59: 787-795Crossref PubMed Scopus (143) Google Scholar, 8Wang D.D. Sievenpiper J.L. de Souza R.J. et al.The effects of fructose intake on serum uric acid vary among controlled dietary trials.J Nutr. 2012; 142: 916-923Crossref PubMed Scopus (160) Google Scholar, 9Chiu S. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on markers of non-alcoholic fatty liver disease (NAFLD): a systematic review and meta-analysis of controlled feeding trials.Eur J Clin Nutr. 2014; 68: 416-423Crossref PubMed Scopus (205) Google Scholar of more than 50 controlled trials that included over 1000 participants of the effect of fructose across a wide dose range have failed to document a signal for harm of fructose in isocaloric substitution for other carbohydrates likely to replace it (Figure 1). Contrary to the hypothesis put forward by DiNicolantonio et al,1DiNicolantonio J.J. O'Keefe J.H. Lucan S.C. Added fructose: a principal driver of type 2 diabetes mellitus and its consequences.Mayo Clin Proc. 2015; 90: 372-381Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar pooled analyses of the totality of the evidence from these trials show that fructose in isocaloric exchange for other sources of carbohydrate leads to clinically meaningful improvements in glycemic control as assessed by glycated blood proteins (equivalent to a 0.57% reduction in hemoglobin A1c, which exceeds the US Food and Drug Administration threshold of 0.3% for the development of new oral antihyperglycemic agents) in individuals with and without diabetes.2Livesey G. Taylor R. Fructose consumption and consequences for glycation, plasma triacylglycerol, and body weight: meta-analyses and meta-regression models of intervention studies.Am J Clin Nutr. 2008; 88: 1419-1437PubMed Scopus (211) Google Scholar, 3Cozma A.I. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on glycemic control in diabetes: a systematic review and meta-analysis of controlled feeding trials.Diabetes Care. 2012; 35: 1611-1620Crossref PubMed Scopus (154) Google Scholar Favorable results are also seen in blood pressure, without any adverse effects on other cardiometabolic risk factors including insulin sensitivity, body weight, fasting lipid levels, postprandial lipid levels, uric acid concentration, and markers of nonalcoholic fatty liver disease in individuals with varying metabolic phenotypes.2Livesey G. Taylor R. Fructose consumption and consequences for glycation, plasma triacylglycerol, and body weight: meta-analyses and meta-regression models of intervention studies.Am J Clin Nutr. 2008; 88: 1419-1437PubMed Scopus (211) Google Scholar, 3Cozma A.I. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on glycemic control in diabetes: a systematic review and meta-analysis of controlled feeding trials.Diabetes Care. 2012; 35: 1611-1620Crossref PubMed Scopus (154) Google Scholar, 4Sievenpiper J.L. de Souza R.J. Mirrahimi A. et al.Effect of fructose on body weight in controlled feeding trials: a systematic review and meta-analysis.Ann Intern Med. 2012; 156: 291-304Crossref PubMed Scopus (222) Google Scholar, 5Sievenpiper J.L. Carleton A.J. Chatha S. et al.Heterogeneous effects of fructose on blood lipids in individuals with type 2 diabetes: systematic review and meta-analysis of experimental trials in humans.Diabetes Care. 2009; 32: 1930-1937Crossref PubMed Scopus (133) Google Scholar, 6Wang D.D. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on postprandial triglycerides: a systematic review and meta-analysis of controlled feeding trials.Atherosclerosis. 2014; 232: 125-133Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 7Ha V. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on blood pressure: a systematic review and meta-analysis of controlled feeding trials.Hypertension. 2012; 59: 787-795Crossref PubMed Scopus (143) Google Scholar, 8Wang D.D. Sievenpiper J.L. de Souza R.J. et al.The effects of fructose intake on serum uric acid vary among controlled dietary trials.J Nutr. 2012; 142: 916-923Crossref PubMed Scopus (160) Google Scholar, 9Chiu S. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on markers of non-alcoholic fatty liver disease (NAFLD): a systematic review and meta-analysis of controlled feeding trials.Eur J Clin Nutr. 2014; 68: 416-423Crossref PubMed Scopus (205) Google Scholar DiNicolantonio et al implied that bias from industry funding might explain these favorable results among the available controlled trials, but very few of these trials were funded exclusively by industry. The majority were funded by a combination of agency and industry or agency alone, and there was no evidence of publication bias across the end points.2Livesey G. Taylor R. Fructose consumption and consequences for glycation, plasma triacylglycerol, and body weight: meta-analyses and meta-regression models of intervention studies.Am J Clin Nutr. 2008; 88: 1419-1437PubMed Scopus (211) Google Scholar, 3Cozma A.I. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on glycemic control in diabetes: a systematic review and meta-analysis of controlled feeding trials.Diabetes Care. 2012; 35: 1611-1620Crossref PubMed Scopus (154) Google Scholar, 4Sievenpiper J.L. de Souza R.J. Mirrahimi A. et al.Effect of fructose on body weight in controlled feeding trials: a systematic review and meta-analysis.Ann Intern Med. 2012; 156: 291-304Crossref PubMed Scopus (222) Google Scholar, 5Sievenpiper J.L. Carleton A.J. Chatha S. et al.Heterogeneous effects of fructose on blood lipids in individuals with type 2 diabetes: systematic review and meta-analysis of experimental trials in humans.Diabetes Care. 2009; 32: 1930-1937Crossref PubMed Scopus (133) Google Scholar, 6Wang D.D. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on postprandial triglycerides: a systematic review and meta-analysis of controlled feeding trials.Atherosclerosis. 2014; 232: 125-133Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 7Ha V. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on blood pressure: a systematic review and meta-analysis of controlled feeding trials.Hypertension. 2012; 59: 787-795Crossref PubMed Scopus (143) Google Scholar, 8Wang D.D. Sievenpiper J.L. de Souza R.J. et al.The effects of fructose intake on serum uric acid vary among controlled dietary trials.J Nutr. 2012; 142: 916-923Crossref PubMed Scopus (160) Google Scholar, 9Chiu S. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on markers of non-alcoholic fatty liver disease (NAFLD): a systematic review and meta-analysis of controlled feeding trials.Eur J Clin Nutr. 2014; 68: 416-423Crossref PubMed Scopus (205) Google Scholar Prospective cohort studies, which provide the greatest protection against bias among observational studies because of their long longitudinal follow-up and the ability to adjust for multiple confounding factors, have also failed to document a direct relationship between fructose and diabetes. Although pooled analyses of the available cohorts have revealed that sugar-sweetened beverages (SSBs) as a source of free fructose are associated with an increased risk of diabetes when the highest and lowest levels of exposure are compared,10Malik V.S. Popkin B.M. Bray G.A. Després J.P. Willett W.C. Hu F.B. Sugar-sweetened beverages and risk of metabolic syndrome and type 2 diabetes: a meta-analysis.Diabetes Care. 2010; 33: 2477-2483Crossref PubMed Scopus (1415) Google Scholar, 11Xi B. Li S. Liu Z. et al.Intake of fruit juice and incidence of type 2 diabetes: a systematic review and meta-analysis.PLoS One. 2014; 9: e93471Crossref PubMed Scopus (100) Google Scholar pooled analyses involving many of the same cohorts have not found the same relationship for total sugars, total sucrose, total fructose,12Tsilas CS, de Souza RJ, Tawfik R, et al. No relation between total sugars intake and incident diabetes: a systematic review and meta-analysis of cohorts. In: Proceedings of the 32nd International Symposium on Diabetes and Nutrition; June 25-27, 2014; Reykjavik, Iceland.Google Scholar or other sources of free fructose such as 100% fruit juice11Xi B. Li S. Liu Z. et al.Intake of fruit juice and incidence of type 2 diabetes: a systematic review and meta-analysis.PLoS One. 2014; 9: e93471Crossref PubMed Scopus (100) Google Scholar and cakes and cookies13Buijsse B. Boeing H. Drogan D. et al.Consumption of fatty foods and incident type 2 diabetes in populations from eight European countries.Eur J Clin Nutr. 2015; 69: 455-461Crossref PubMed Scopus (31) Google Scholar (Figure 2). The opposite relationship (benefit) has also been reported for fruit as a source of bound fructose.14Li M. Fan Y. Zhang X. Hou W. Tang Z. Fruit and vegetable intake and risk of type 2 diabetes mellitus: meta-analysis of prospective cohort studies.BMJ Open. 2014; 4: e005497Google Scholar Setting aside these discrepant findings, if one wants to invoke the evidence for SSBs as a proxy for all added fructose-containing sugars, then one has to ask how important a risk factor is the intake of SSBs. A recent comparative risk assessment revealed that the burden of disease and mortality attributable to SSBs (that is, population-attributable fraction) is still much less than that of other established risk factors measured among the cohort studies, ranking 32nd among 57 risk factors globally.15Lim S.S. Vos T. Flaxman A.D. et al.A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010.Lancet. 2012; 380 ([published correction appears in Lancet. 2013;381(9874):1276]): 2224-2260Abstract Full Text Full Text PDF PubMed Scopus (8623) Google Scholar Even among the dietary and physical inactivity risk factors, SSBs ranked 12th of 15 for both burden of disease and mortality, and no other sources of added fructose-containing sugars were identified as risk factors.15Lim S.S. Vos T. Flaxman A.D. et al.A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010.Lancet. 2012; 380 ([published correction appears in Lancet. 2013;381(9874):1276]): 2224-2260Abstract Full Text Full Text PDF PubMed Scopus (8623) Google Scholar The question becomes why the higher-level evidence disagrees with the current hypothesis from DiNicolantonio et al.1DiNicolantonio J.J. O'Keefe J.H. Lucan S.C. Added fructose: a principal driver of type 2 diabetes mellitus and its consequences.Mayo Clin Proc. 2015; 90: 372-381Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar One reason may be that the mechanisms being invoked are not as relevant in humans as in the animal models used to support them. For example, although de novo lipogenesis is extremely high (estimated at ≥50%) with excessive fructose feeding (typically 60% of total energy intake) in rodents, a careful review of stable isotope tracer studies reveals that de novo lipogenesis from fructose (as indeed from all carbohydrate) is a minor pathway for fructose disposal in humans (estimated to range from 0%-1% at moderate intake to up to 5% with overfeeding in humans).16van Buul V.J. Tappy L. Brouns F.J. Misconceptions about fructose-containing sugars and their role in the obesity epidemic.Nutr Res Rev. 2014; 27: 119-130Crossref PubMed Scopus (66) Google Scholar Another reason for the differences may lie in biologically plausible pathways that offset any harm and even explain some benefits. Fructose has a very low glycemic index (15), a factor that led to an early interest in fructose in diabetes management. Emerging evidence also shows that low-dose fructose (≤10 g per meal) may benefit glycemic control through its metabolite fructose-1-phosphate by inducing glucokinase activity. This catalytic effect of fructose on hepatic glucose metabolism has been reported to coincide with (1) a decrease in hepatic glucose production under hyperglycemic clamp conditions in patients with type 2 diabetes and (2) an increase in glycogen synthesis by carbon 13 nuclear magnetic resonance spectroscopy under euglycemic clamp conditions in participants without diabetes.17Petersen K.F. Laurent D. Yu C. Cline G.W. Shulman G.I. Stimulating effects of low-dose fructose on insulin-stimulated hepatic glycogen synthesis in humans.Diabetes. 2001; 50: 1263-1268Crossref PubMed Scopus (116) Google Scholar, 18Hawkins M. Gabriely I. Wozniak R. Vilcu C. Shamoon H. Rossetti L. Fructose improves the ability of hyperglycemia per se to regulate glucose production in type 2 diabetes.Diabetes. 2002; 51: 606-614Crossref PubMed Scopus (75) Google Scholar These mechanisms appear to be sustainable over the long term. A systematic review and meta-analysis of controlled trials of the effect of small "catalytic" fructose doses (≤36 g/d) in exchange for starch reproduced the favorable glycemic effects seen at higher doses, without any adverse effects on metabolic control over 1 to 52 weeks of follow-up.19Sievenpiper J.L. Chiavaroli L. de Souza R.J. et al.'Catalytic' doses of fructose may benefit glycaemic control without harming cardiometabolic risk factors: a small meta-analysis of randomised controlled feeding trials.Br J Nutr. 2012; 108: 418-423Crossref PubMed Scopus (80) Google Scholar Harm from fructose, however, is seen under certain conditions. Dose thresholds for harm have been previously identified for the effect of fructose on fasting and postprandial lipid levels,2Livesey G. Taylor R. Fructose consumption and consequences for glycation, plasma triacylglycerol, and body weight: meta-analyses and meta-regression models of intervention studies.Am J Clin Nutr. 2008; 88: 1419-1437PubMed Scopus (211) Google Scholar, 5Sievenpiper J.L. Carleton A.J. Chatha S. et al.Heterogeneous effects of fructose on blood lipids in individuals with type 2 diabetes: systematic review and meta-analysis of experimental trials in humans.Diabetes Care. 2009; 32: 1930-1937Crossref PubMed Scopus (133) Google Scholar Although these thresholds have not been reproduced in updated meta-analyses,6Wang D.D. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on postprandial triglycerides: a systematic review and meta-analysis of controlled feeding trials.Atherosclerosis. 2014; 232: 125-133Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar individual trials of fructose feeding at very high doses (>100 g/d) have found increases in fasting and postprandial triglyceride levels in energy-matched comparisons with glucose.20Egli L. Lecoultre V. Theytaz F. et al.Exercise prevents fructose-induced hypertriglyceridemia in healthy young subjects.Diabetes. 2013; 62: 2259-2265Crossref PubMed Scopus (83) Google Scholar The most consistent signal for harm remains restricted to hypercaloric addition trials, in which excess calories from pure fructose are added to a diet, compared with the same diet without these excess calories. Systematic reviews and meta-analyses of the available trials have found that fructose under these conditions leads to weight gain and increases in fasting and postprandial triglyceride levels, fasting glucose levels, whole-body and hepatic insulin resistance, uric acid concentrations, and markers of nonalcoholic fatty liver disease3Cozma A.I. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on glycemic control in diabetes: a systematic review and meta-analysis of controlled feeding trials.Diabetes Care. 2012; 35: 1611-1620Crossref PubMed Scopus (154) Google Scholar, 4Sievenpiper J.L. de Souza R.J. Mirrahimi A. et al.Effect of fructose on body weight in controlled feeding trials: a systematic review and meta-analysis.Ann Intern Med. 2012; 156: 291-304Crossref PubMed Scopus (222) Google Scholar, 5Sievenpiper J.L. Carleton A.J. Chatha S. et al.Heterogeneous effects of fructose on blood lipids in individuals with type 2 diabetes: systematic review and meta-analysis of experimental trials in humans.Diabetes Care. 2009; 32: 1930-1937Crossref PubMed Scopus (133) Google Scholar, 6Wang D.D. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on postprandial triglycerides: a systematic review and meta-analysis of controlled feeding trials.Atherosclerosis. 2014; 232: 125-133Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 7Ha V. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on blood pressure: a systematic review and meta-analysis of controlled feeding trials.Hypertension. 2012; 59: 787-795Crossref PubMed Scopus (143) Google Scholar, 8Wang D.D. Sievenpiper J.L. de Souza R.J. et al.The effects of fructose intake on serum uric acid vary among controlled dietary trials.J Nutr. 2012; 142: 916-923Crossref PubMed Scopus (160) Google Scholar, 9Chiu S. Sievenpiper J.L. de Souza R.J. et al.Effect of fructose on markers of non-alcoholic fatty liver disease (NAFLD): a systematic review and meta-analysis of controlled feeding trials.Eur J Clin Nutr. 2014; 68: 416-423Crossref PubMed Scopus (205) Google Scholar (Figure 1). The inability of fructose to induce the same adverse effects in isocaloric substitution for other carbohydrates suggests that the main determinant of this observed harm is excess calories rather than any special metabolic or endocrine responses to the fructose. Any adverse effects observed under conditions of overfeeding appear to be no more mediated by fructose than by other sources of carbohydrate used to replace it. That these adverse effects appear to be reversible by exercise also suggests that they may not be generalizable to all individuals.20Egli L. Lecoultre V. Theytaz F. et al.Exercise prevents fructose-induced hypertriglyceridemia in healthy young subjects.Diabetes. 2013; 62: 2259-2265Crossref PubMed Scopus (83) Google Scholar In conclusion, efforts to identify the factors that contribute to the development of type 2 diabetes remain an important goal. These efforts, however, must be conducted using systematic evidence-based approaches that protect against important sources of bias that include confounding from energy. In the case of fructose, the totality of the highest-level evidence from the systematic reviews and meta-analyses of controlled trials and prospective cohort studies fails to implicate fructose as an independent driver of type 2 diabetes. Added Fructose: A Principal Driver of Type 2 Diabetes Mellitus and Its ConsequencesMayo Clinic ProceedingsVol. 90Issue 3PreviewData from animal experiments and human studies implicate added sugars (eg, sucrose and high-fructose corn syrup) in the development of diabetes mellitus and related metabolic derangements that raise cardiovascular (CV) risk. Added fructose in particular (eg, as a constituent of added sucrose or as the main component of high-fructose sweeteners) may pose the greatest problem for incident diabetes, diabetes-related metabolic abnormalities, and CV risk. Conversely, whole foods that contain fructose (eg, fruits and vegetables) pose no problem for health and are likely protective against diabetes and adverse CV outcomes. Full-Text PDF In reply—Fructose as a Driver of Diabetes: An Incomplete View of the EvidenceMayo Clinic ProceedingsVol. 90Issue 7PreviewWe appreciate the response to our article from Dr Sievenpiper and colleagues, who argue that the highest-level evidence fails to implicate fructose as an independent driver of type 2 diabetes. We respectfully counter that the meta-analyses and systematic reviews the authors present may be misleading and that the totality of data supports our original contention. Full-Text PDF
Referência(s)