User login
Higher levels of LDL cholesterol, HDL cholesterol, and triglycerides over a lifetime are protective against type 2 diabetes, a Mendelian randomization study has shown.
The study also bolstered established evidence that LDL cholesterol and triglycerides boost the risk of coronary artery disease (CAD) but showed no contribution of HDL cholesterol to that risk.
Investigators sought to shed light on the role of the most commonly measured lipid fractions – LDL cholesterol, HDL cholesterol, and triglycerides – in the development of CAD and diabetes, particularly the observed link between statin therapy and an increased risk of diabetes.
Because genotype is not modifiable by disease, a genetic instrument can be used as an model for an exposure, and “Mendelian randomization generates unbiased, unconfounded effect estimates that are sometimes taken as evidence of a causal role,” Jon White, PhD, of University College London and his coinvestigators explained.
They used data from three genome-wide association studies involving 188,577 persons with blood lipid measures, 63,158 CAD cases, and 34,840 diabetes cases. All involved only people of European ancestry. Summary-level data for lipids were from the Global Lipids Genetics Consortium (Nat Genet. 2013;45[11]:1274-83), diabetes data came from the Diabetes Genetics Replication and Meta-analysis (Nat Genet. 2012;44[9]:981-90), and CAD data were from the Coronary Artery Disease Genome-wide Replication and Meta-analysis plus Coronary Artery Disease Genetics (Nat Genet. 2013;45[1]:25-33). From these, the investigators constructed genetic instruments comprised of single-nucleotide polymorphisms (SNPs) and conducted Mendelian randomizations designed to adjust for the SNPs’ possible associations with other traits, or pleiotropy.
The results showed that two lipid fractions were associated with reduced risk for type 2 diabetes and one had no discernible effect. LDL cholesterol showed the strongest association: An increase of 1 standard deviation, equivalent to 38 mg/dL, was tied to a 21% reduction in risk (odds ratio, 0.79) of diabetes. For HDL, a 1-SD rise of 16 mg/DL in HDL was associated with a 17% lower risk (OR, 0.83). A 1-SD rise of triglycerides, 89 mg/dL, also reduced risk by 17% (OR, 0.83), but there were statistical inconsistencies between analyses.
The associations between 1-SD increases and CAD were consistent with conventional wisdom: For LDL cholesterol, CAD risk rose by 68%; for triglycerides, the increase was 28%; and for HDL cholesterol, the risk was slightly reduced by 5% but was not statistically significant (JAMA Cardiol. 2016 Aug 3. doi: 10.1001/jamacardio.2016.1884).
These results can help to identify the potential effects of lipid-modifying drugs, yet “although all three lipids were associated with reduced risk of diabetes, it does not necessarily follow that lowering of LDL cholesterol or triglyceride levels through use of drugs that target specific proteins (eg, PCSK9) will alter the risk of diabetes,” Dr. White and his colleagues wrote. Large-scale genetic and clinical trials are needed to determine such dysglycemic associations.
This study was conducted by the Clinical Trial Service Unit of the University of Oxford through a grant by Merck Sharp & Dohme, with additional funding from numerous academic and research institutions. The funding sources had no role in the design or conduct of the study. Two of the investigators had ties to pharmaceutical companies.
The findings from Dr. White and his associates will no doubt fuel the controversy on the causal association of major plasma lipids with type 2 diabetes.
Because prior studies have shown that lowering LDL cholesterol with statins is associated with a 21% increased risk of diabetes, the finding that increasing LDL lowers the risk of diabetes is not inconsistent. The magnitude of risk, however, was much lower (absolute increase of 9%) in the statin trials. Nonetheless, if LDL is indeed shown to be protective against diabetes, this will have major implications for all lipid-lowering drugs, not just statins.
Regarding HDL cholesterol, the inconclusive findings of Dr. White and his coinvestigators, combined with prior research showing both direct and inverse associations between elevated HDL and type 2 diabetes, shed scant light on the role of HDL in diabetes.
The findings on the association of genetically mediated triglyceride levels and type 2 diabetes from this study are the most counterintuitive. They are opposite to previous epidemiological reports showing that elevated triglyceride levels are associated with higher risk for diabetes. If true, the implications of this are substantial, including the potential that intervention to reduce triglyceride levels could paradoxically increase the risk for type 2 diabetes.
This study, using Mendelian randomization and sophisticated analyses to adjust for pleiotropic effects, advances our knowledge; however, it seems that other approaches are required to further evaluate the causal relevance of each of these lipid fractions in association with type 2 diabetes.
Danish Saleheen, MBBS, PhD; Daniel J. Rader, MD; and Benjamin F. Voight, PhD, of the University of Pennsylvania, Philadelphia, made these comments in an accompanying editorial (JAMA Cardiol. 2016 Aug 3. doi: 10.1001/jamacardio.2016.2298). They had no disclosures to report.
The findings from Dr. White and his associates will no doubt fuel the controversy on the causal association of major plasma lipids with type 2 diabetes.
Because prior studies have shown that lowering LDL cholesterol with statins is associated with a 21% increased risk of diabetes, the finding that increasing LDL lowers the risk of diabetes is not inconsistent. The magnitude of risk, however, was much lower (absolute increase of 9%) in the statin trials. Nonetheless, if LDL is indeed shown to be protective against diabetes, this will have major implications for all lipid-lowering drugs, not just statins.
Regarding HDL cholesterol, the inconclusive findings of Dr. White and his coinvestigators, combined with prior research showing both direct and inverse associations between elevated HDL and type 2 diabetes, shed scant light on the role of HDL in diabetes.
The findings on the association of genetically mediated triglyceride levels and type 2 diabetes from this study are the most counterintuitive. They are opposite to previous epidemiological reports showing that elevated triglyceride levels are associated with higher risk for diabetes. If true, the implications of this are substantial, including the potential that intervention to reduce triglyceride levels could paradoxically increase the risk for type 2 diabetes.
This study, using Mendelian randomization and sophisticated analyses to adjust for pleiotropic effects, advances our knowledge; however, it seems that other approaches are required to further evaluate the causal relevance of each of these lipid fractions in association with type 2 diabetes.
Danish Saleheen, MBBS, PhD; Daniel J. Rader, MD; and Benjamin F. Voight, PhD, of the University of Pennsylvania, Philadelphia, made these comments in an accompanying editorial (JAMA Cardiol. 2016 Aug 3. doi: 10.1001/jamacardio.2016.2298). They had no disclosures to report.
The findings from Dr. White and his associates will no doubt fuel the controversy on the causal association of major plasma lipids with type 2 diabetes.
Because prior studies have shown that lowering LDL cholesterol with statins is associated with a 21% increased risk of diabetes, the finding that increasing LDL lowers the risk of diabetes is not inconsistent. The magnitude of risk, however, was much lower (absolute increase of 9%) in the statin trials. Nonetheless, if LDL is indeed shown to be protective against diabetes, this will have major implications for all lipid-lowering drugs, not just statins.
Regarding HDL cholesterol, the inconclusive findings of Dr. White and his coinvestigators, combined with prior research showing both direct and inverse associations between elevated HDL and type 2 diabetes, shed scant light on the role of HDL in diabetes.
The findings on the association of genetically mediated triglyceride levels and type 2 diabetes from this study are the most counterintuitive. They are opposite to previous epidemiological reports showing that elevated triglyceride levels are associated with higher risk for diabetes. If true, the implications of this are substantial, including the potential that intervention to reduce triglyceride levels could paradoxically increase the risk for type 2 diabetes.
This study, using Mendelian randomization and sophisticated analyses to adjust for pleiotropic effects, advances our knowledge; however, it seems that other approaches are required to further evaluate the causal relevance of each of these lipid fractions in association with type 2 diabetes.
Danish Saleheen, MBBS, PhD; Daniel J. Rader, MD; and Benjamin F. Voight, PhD, of the University of Pennsylvania, Philadelphia, made these comments in an accompanying editorial (JAMA Cardiol. 2016 Aug 3. doi: 10.1001/jamacardio.2016.2298). They had no disclosures to report.
Higher levels of LDL cholesterol, HDL cholesterol, and triglycerides over a lifetime are protective against type 2 diabetes, a Mendelian randomization study has shown.
The study also bolstered established evidence that LDL cholesterol and triglycerides boost the risk of coronary artery disease (CAD) but showed no contribution of HDL cholesterol to that risk.
Investigators sought to shed light on the role of the most commonly measured lipid fractions – LDL cholesterol, HDL cholesterol, and triglycerides – in the development of CAD and diabetes, particularly the observed link between statin therapy and an increased risk of diabetes.
Because genotype is not modifiable by disease, a genetic instrument can be used as an model for an exposure, and “Mendelian randomization generates unbiased, unconfounded effect estimates that are sometimes taken as evidence of a causal role,” Jon White, PhD, of University College London and his coinvestigators explained.
They used data from three genome-wide association studies involving 188,577 persons with blood lipid measures, 63,158 CAD cases, and 34,840 diabetes cases. All involved only people of European ancestry. Summary-level data for lipids were from the Global Lipids Genetics Consortium (Nat Genet. 2013;45[11]:1274-83), diabetes data came from the Diabetes Genetics Replication and Meta-analysis (Nat Genet. 2012;44[9]:981-90), and CAD data were from the Coronary Artery Disease Genome-wide Replication and Meta-analysis plus Coronary Artery Disease Genetics (Nat Genet. 2013;45[1]:25-33). From these, the investigators constructed genetic instruments comprised of single-nucleotide polymorphisms (SNPs) and conducted Mendelian randomizations designed to adjust for the SNPs’ possible associations with other traits, or pleiotropy.
The results showed that two lipid fractions were associated with reduced risk for type 2 diabetes and one had no discernible effect. LDL cholesterol showed the strongest association: An increase of 1 standard deviation, equivalent to 38 mg/dL, was tied to a 21% reduction in risk (odds ratio, 0.79) of diabetes. For HDL, a 1-SD rise of 16 mg/DL in HDL was associated with a 17% lower risk (OR, 0.83). A 1-SD rise of triglycerides, 89 mg/dL, also reduced risk by 17% (OR, 0.83), but there were statistical inconsistencies between analyses.
The associations between 1-SD increases and CAD were consistent with conventional wisdom: For LDL cholesterol, CAD risk rose by 68%; for triglycerides, the increase was 28%; and for HDL cholesterol, the risk was slightly reduced by 5% but was not statistically significant (JAMA Cardiol. 2016 Aug 3. doi: 10.1001/jamacardio.2016.1884).
These results can help to identify the potential effects of lipid-modifying drugs, yet “although all three lipids were associated with reduced risk of diabetes, it does not necessarily follow that lowering of LDL cholesterol or triglyceride levels through use of drugs that target specific proteins (eg, PCSK9) will alter the risk of diabetes,” Dr. White and his colleagues wrote. Large-scale genetic and clinical trials are needed to determine such dysglycemic associations.
This study was conducted by the Clinical Trial Service Unit of the University of Oxford through a grant by Merck Sharp & Dohme, with additional funding from numerous academic and research institutions. The funding sources had no role in the design or conduct of the study. Two of the investigators had ties to pharmaceutical companies.
Higher levels of LDL cholesterol, HDL cholesterol, and triglycerides over a lifetime are protective against type 2 diabetes, a Mendelian randomization study has shown.
The study also bolstered established evidence that LDL cholesterol and triglycerides boost the risk of coronary artery disease (CAD) but showed no contribution of HDL cholesterol to that risk.
Investigators sought to shed light on the role of the most commonly measured lipid fractions – LDL cholesterol, HDL cholesterol, and triglycerides – in the development of CAD and diabetes, particularly the observed link between statin therapy and an increased risk of diabetes.
Because genotype is not modifiable by disease, a genetic instrument can be used as an model for an exposure, and “Mendelian randomization generates unbiased, unconfounded effect estimates that are sometimes taken as evidence of a causal role,” Jon White, PhD, of University College London and his coinvestigators explained.
They used data from three genome-wide association studies involving 188,577 persons with blood lipid measures, 63,158 CAD cases, and 34,840 diabetes cases. All involved only people of European ancestry. Summary-level data for lipids were from the Global Lipids Genetics Consortium (Nat Genet. 2013;45[11]:1274-83), diabetes data came from the Diabetes Genetics Replication and Meta-analysis (Nat Genet. 2012;44[9]:981-90), and CAD data were from the Coronary Artery Disease Genome-wide Replication and Meta-analysis plus Coronary Artery Disease Genetics (Nat Genet. 2013;45[1]:25-33). From these, the investigators constructed genetic instruments comprised of single-nucleotide polymorphisms (SNPs) and conducted Mendelian randomizations designed to adjust for the SNPs’ possible associations with other traits, or pleiotropy.
The results showed that two lipid fractions were associated with reduced risk for type 2 diabetes and one had no discernible effect. LDL cholesterol showed the strongest association: An increase of 1 standard deviation, equivalent to 38 mg/dL, was tied to a 21% reduction in risk (odds ratio, 0.79) of diabetes. For HDL, a 1-SD rise of 16 mg/DL in HDL was associated with a 17% lower risk (OR, 0.83). A 1-SD rise of triglycerides, 89 mg/dL, also reduced risk by 17% (OR, 0.83), but there were statistical inconsistencies between analyses.
The associations between 1-SD increases and CAD were consistent with conventional wisdom: For LDL cholesterol, CAD risk rose by 68%; for triglycerides, the increase was 28%; and for HDL cholesterol, the risk was slightly reduced by 5% but was not statistically significant (JAMA Cardiol. 2016 Aug 3. doi: 10.1001/jamacardio.2016.1884).
These results can help to identify the potential effects of lipid-modifying drugs, yet “although all three lipids were associated with reduced risk of diabetes, it does not necessarily follow that lowering of LDL cholesterol or triglyceride levels through use of drugs that target specific proteins (eg, PCSK9) will alter the risk of diabetes,” Dr. White and his colleagues wrote. Large-scale genetic and clinical trials are needed to determine such dysglycemic associations.
This study was conducted by the Clinical Trial Service Unit of the University of Oxford through a grant by Merck Sharp & Dohme, with additional funding from numerous academic and research institutions. The funding sources had no role in the design or conduct of the study. Two of the investigators had ties to pharmaceutical companies.
FROM JAMA CARDIOLOGY
Key clinical point: Elevated LDL cholesterol and triglyceride levels increase the risk of coronary artery disease but are linked with a lower risk of diabetes.
Major finding: Rises in LDL cholesterol and triglycerides were associated with decreases in diabetes risk of 21% and 17%, respectively.
Data source: A Mendelian randomization analysis using three genome-wide association studies.
Disclosures: This study was conducted by the Clinical Trial Service Unit of the University of Oxford through a grant by Merck Sharp & Dohme, with additional funding from numerous academic and research institutions. The funding sources had no role in the design or conduct of the study. Two of the investigators had ties to pharmaceutical companies.
