Concept
Lipids & Cardiometabolic Risk
4 studiesEvidence last moved Sep 20, 2026
Blood lipids are the most-measured modifiable cardiovascular risk markers, and the literature on them spans short randomised trials of what people eat, large observational cohorts, genetic causal inference and animal mechanism work. Each answers a different question, and their answers are not interchangeable.
Dietary fat advice is contested in public in a way it is not in the trial literature, largely because a four-week change in a lipid marker and a lifetime change in event risk get quoted as if they were the same evidence. Seeing the designs side by side is what makes the difference legible.
Studies
4
Findings
4
4 supporting · 0 challenging · 0 qualifying citations
Open tensions
1
Latest change
Concept page published
Lipids & Cardiometabolic Risk
Currently
What we know
- A randomised marker effect over four weeks — with no clinical events measured.
- Mendelian randomisation raises this above the observational association it accompanies.
- No single index dominated; the better predictor changed with the endpoint.
- The knockdown reproducing the drug effect is what places SGK1 on the path.
Largest unresolved question
The randomised trial and the genetic study support different claims about the same molecules. The trial randomises a real dietary exposure but measures markers over four weeks in healthy volunteers; the Mendelian randomisation estimates lifelong effects on hard endpoints but from a single baseline lipid measurement in a healthier-than-average cohort, with instrument pleiotropy unresolved.
Common misconceptions
A trial showing coconut oil does not raise LDL more than olive oil shows it is heart-healthy.
The trial ran four weeks in healthy volunteers and measured lipids, not cardiovascular events. It establishes a marker comparison over a month; it cannot establish event risk, and the authors say it does not displace guidance on saturated fat.
A risk index that predicts better is measuring the underlying biology better.
TyG-WHtR beat TyG for mortality and heart failure while TyG beat TyG-WHtR for coronary disease and angina, in the same dataset. An index that wins on one endpoint and loses on another is capturing endpoint-specific prediction, not a single better measure of risk.
Related
Claim ledger
What the evidence shows
Drawn from 4 studies in this library. Mix labels say which citation roles are present; they are not a strength score. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope.
A randomised marker effect over four weeks — with no clinical events measured.
In a randomised comparison, butter raised LDL cholesterol relative to both coconut oil (+0.42 mmol/L) and olive oil (+0.38 mmol/L), while coconut and olive oil did not differ (-0.04 mmol/L, P = 0.74). Coconut oil raised HDL relative to butter (+0.18) and olive oil (+0.16). Weight, adiposity, glucose and blood pressure did not differ.
Mendelian randomisation raises this above the observational association it accompanies.
Genetic instruments support a causal role for triglyceride-rich lipoproteins in cardiometabolic multimorbidity. Remnant cholesterol at or above 1.0 against below 0.4 mmol/L carried HR 2.46 for type 2 diabetes and 1.63 for ischaemic heart disease; causal odds per 1.0 mmol/L triglycerides were 1.21 for any multimorbidity, with associations for ischaemic heart disease and diabetes but not stroke.
No single index dominated; the better predictor changed with the endpoint.
Composite indices can outperform their components, and which one wins depends on the outcome. TyG-WHtR predicted all-cause mortality (HR 1.31), cardiovascular mortality (HR 2.22) and heart failure (OR 3.99) more strongly than TyG alone, while TyG was stronger for total cardiovascular disease (OR 2.00), coronary heart disease (1.85) and angina (2.93).
The knockdown reproducing the drug effect is what places SGK1 on the path.
Mechanistic work supplies a pathway the human studies cannot. In diabetic animals and H9C2 cells, dapagliflozin attenuated diastolic dysfunction and reduced fibrosis via SGK1 and ENaC inhibition, with SGK1 knockdown reproducing the drug's effect on ENaC and NHE1.
Debates
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes.
The randomised trial and the genetic study support different claims about the same molecules. The trial randomises a real dietary exposure but measures markers over four weeks in healthy volunteers; the Mendelian randomisation estimates lifelong effects on hard endpoints but from a single baseline lipid measurement in a healthier-than-average cohort, with instrument pleiotropy unresolved.
The randomised trial and the genetic study support different claims about the same molecules. The trial randomises a real dietary exposure but measures markers over four weeks in healthy volunteers; the Mendelian randomisation estimates lifelong effects on hard endpoints but from a single baseline lipid measurement in a healthier-than-average cohort, with instrument pleiotropy unresolved.
- Does coconut oil raise LDL like butter?
- Do remnant cholesterol and triglycerides cause cardiometabolic multimorbidity?
Study Role Design N Population Outcome Does coconut oil raise LDL like butter? Supports RCTCommunity RCT in Cambridgeshire (BBC-recruited) assigning extra virgin coconut oil, extra virgin olive oil, or unsalted butter 50 g/day for 4 weeks N=96 · 96 randomised; 94 attended baseline; 91 completed 4-week follow-up Healthy volunteers aged 50–75 years without cancer, CVD, diabetes, or lipid-lowering drugs Change in serum LDL-C (primary); other lipids, weight, BP, glucose, CRP Do remnant cholesterol and triglycerides cause cardiometabolic multimorbidity? Supports Mendelian randomisationUK Biobank multistate cohort plus one-sample Mendelian randomisation using 13 triglyceride-rich-lipoprotein SNPs. N=334030 · Observational: 334,030 (remnant cholesterol) and 365,577 (triglycerides); MR: 376,712 and 411,930. UK Biobank adults free of diabetes, ischemic heart disease, and stroke at baseline (lipid-lowering users excluded from observational analyses). Progression from first cardiometabolic disease to multimorbidity (especially IHD plus type 2 diabetes), plus causal ORs from MR.
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
Timeline
How understanding moved
Study years are when the paper was published. Evidence edits are dated changes to this page's claims. Explanations are when PaperFren added a Discovery — not a claim that the science happened that day.
2026
Concept page published
Lipids & Cardiometabolic Risk
Change log
What changed
Dated edits to this page's evidence: studies added or removed from a claim, claims added or withdrawn, and new explanations tagged here. Rewordings are not listed.
- Concept page published
Papers
4 studies in this library bear on Lipids & Cardiometabolic Risk, ordered by citations.
- Does coconut oil raise LDL like butter?
In 96 healthy adults, 50 g/day butter for 4 weeks raised LDL-C by ~0.4 mmol/L versus coconut or olive oil; coconut oil did not differ from olive oil on LDL and raised HDL-C.
- Do TyG and TyG-WHtR predict death and CVD in NAFLD?
In 6,627 NHANES adults with NAFLD, TyG-WHtR better predicted all-cause death, CVD death, and heart failure, while the TyG index tracked total CVD, CHD, and angina.
- Do remnant cholesterol and triglycerides cause cardiometabolic multimorbidity?
In >300,000 UK Biobank participants, higher remnant cholesterol and triglycerides tracked—and MR supported a causal role in—progression to cardiometabolic multimorbidity, especially IHD plus type 2 diabetes.
- How might dapagliflozin ease diabetic stiff hearts?
In 30 rabbits, dapagliflozin lessened diabetes-related diastolic dysfunction and fibrosis by downregulating SGK1 and ENaC, with matching H9C2 cell results.
Compare studies
Select 2–10 studies. Design and N are labels, not a ranking.
Nothing selected yet.
Questions
What is still open
The randomised trial and the genetic study support different claims about the same molecules. The trial randomises a real dietary exposure but measures markers over four weeks in healthy volunteers; the Mendelian randomisation estimates lifelong effects on hard endpoints but from a single baseline lipid measurement in a healthier-than-average cohort, with instrument pleiotropy unresolved.
Ask PaperFren about Lipids & Cardiometabolic Risk
Study this conceptflashcards and short-answer questions
What does Mendelian randomisation add over a lipid cohort association, and what does it still assume?
It uses genetic variants fixed at conception as instruments, so reverse causation and most lifestyle confounding are excluded by design — yielding causal odds of 1.21 per 1.0 mmol/L triglycerides for cardiometabolic multimorbidity. It assumes those variants affect the outcome only through the lipid, which pleiotropy can violate, and the UK Biobank sample is healthier than the general population.
Why is a four-week lipid trial weak evidence for a dietary recommendation even though it is randomised?
Because randomisation secures internal validity for the outcome measured, and the outcome here was a surrogate. Butter raised LDL by about 0.4 mmol/L relative to both oils over four weeks in healthy volunteers, with no events recorded. Translating that into a lifetime dietary recommendation requires assuming the marker change maps onto event risk at that dose and duration — which this design cannot test.
How should an animal mechanism study be weighted against a human outcome trial for the same drug?
They answer different questions and neither substitutes. The dapagliflozin work identifies SGK1-ENaC signalling in diabetic animals and H9C2 cells, with SGK1 knockdown reproducing the drug effect — which explains how a benefit might arise. Whether the benefit occurs in patients, and how large it is, comes only from outcome trials, and the authors are explicit that this does not replace them.