Is high LDL in lean keto "hyper-responders" benign and harmless?
Claim attributed to The low-carb and ketogenic community and Lipid Energy Model researchers (Dave Feldman, Nick Norwitz and colleagues), spread on social media as reassurance for keto dieters with high LDL , The phenotype and its Lipid Energy Model are described in peer-reviewed papers and the metabolic explanation is plausible; the contested leap is the community reassurance that this LDL is therefore safe. The lead investigators' own KETO-CTA paper did not claim the LDL is harmless, and that paper was retracted in 2026.
The lipid pattern is real, but the reassurance built on top is not. The one prospective imaging study used to support it was retracted in 2026 and, on its own numbers, shows plaque progressing, not staying benign.
A real lipid pattern with a plausible cause, sold as proof of safety it never had: the one prospective study was retracted and its own scans show plaque getting worse.
What it’s supposed to target
- Lipid Energy Model (VLDL export)
- Lipoprotein lipase turnover
- ApoB particle number
- Arterial-wall retention
The phenotype has a real proposed mechanism, the Lipid Energy Model. On a low-carbohydrate diet a lean, insulin-sensitive person runs mostly on fat, so the liver exports triglyceride-rich VLDL to ship that fat to the tissues. Lipoprotein lipase strips the triglyceride out for fuel, and the particle left behind is an LDL particle. Where this fat traffic runs fast, LDL-C and HDL-C climb while triglycerides stay low, the LMHR 'lipid triad'. As an account of WHY the numbers move, the model is coherent and probably right.
Explaining the rise is not the same as showing it is harmless. Atherosclerosis tracks the number of apoB particles retained in the artery wall over time (the same response-to-retention biology that makes LDL causal), and lean hyper-responders carry a high apoB count however healthy they are otherwise. The decisive counterexample is familial hypercholesterolemia: lean, fit people with lifelong high LDL still get early heart disease, so 'metabolically healthy' is not a known exemption. A genuine mechanism for the high LDL, then wrongly stretched into a guarantee that this particular LDL cannot harm.
Mechanism is theory, not proof. A plausible pathway explains why something might work, not whether it does. The verdict rests on the evidence below, not the elegance of the theory.
What would have to be true
LDL would have to be atherogenic mainly through metabolic ill-health, so that healthy lean people are exempt. This link FAILS: apoB/LDL causes atherosclerosis via cumulative magnitude and duration of exposure, and lean people with familial hypercholesterolemia still get premature disease.
The LMHR phenotype would need a metabolic off-switch sparing the artery wall. No such switch is demonstrated.
Long-term hard-outcome data in LMHRs would have to show no excess events. Such data do not exist; the one prospective imaging study is short, uncontrolled, retracted, and shows plaque progressing.
What the evidence actually shows
The pattern is real; the safety claim was never tested
The phenotype-defining paper (Norwitz et al., 2022) was a web survey of 548 self-selected carb-restricted dieters identifying a subgroup with LDL-C >=200, HDL-C >=80 and triglycerides <=70 mg/dL at lower BMI. It measured no atherosclerosis and no cardiovascular outcomes, so it cannot show the LDL is benign. The companion Lipid Energy Model paper is explicitly a mechanism/hypothesis paper that explains why LDL rises, not whether it is harmless. A peer-reviewed critique (Moore et al., 2022) found clinically significant LDL rises across all BMI categories, that the original models were weak (adjusted R-squared about 0.07), and that even a 35 to 44 mg/dL rise implies roughly 20% higher 5-year coronary risk.
The flagship study was retracted and its own data show plaque worsening
The KETO-CTA longitudinal paper (promoted as "Plaque Begets Plaque," published as "Plaque Predicts Plaque, ApoB Does Not") was the headline evidence for reassurance. It was retracted in 2026 (an expression of concern, then a full retraction, confirmed by Retraction Watch and the journal JACC: Advances) over reliability concerns: unblinded image analysis by the imaging vendor Cleerly and an equity conflict the published paper did not disclose (coauthor James Earls, Cleerly's chief medical officer, held equity in the firm, though he had told the journal on acceptance). On the participants' own scans plaque progressed over the year, and independent reviewers flagged selective reporting, the paper foregrounding a secondary measure (percent atheroma volume) over its preregistered primary endpoint (noncalcified plaque volume). Against all of this, the EAS consensus concludes LDL unequivocally causes atherosclerosis, with no demonstrated exemption for any lean phenotype.
Studies, graded, and who paid
Defined in a 548-person web survey; a real lipid phenotype, but self-selected and not outcome data.
A coherent mechanistic hypothesis (VLDL secretion plus LPL turnover); explains the rise, not its safety.
Unestablished and against strong priors; the key study was retracted and its own imaging shows plaque worsening.
| # | Study | Type | Size | Funding / COI | Key limitations |
|---|---|---|---|---|---|
| 1 | Norwitz et al. 2022, phenotype-defining survey | Cross-sectional web survey (observational) | 548 respondents | Mixed Citizen Science Foundation model; authors disclose low-carb book ties; self-selected sample. | No atherosclerosis or CV outcomes measured; self-selected; defines lipid pattern only. |
| 2 | Norwitz et al. 2022, Lipid Energy Model | Hypothesis/mechanistic review | No primary participants | Mixed Disclosed: low-carb book royalties; HDL-assay patent; consulting/membership income. | No clinical outcome data; does not show LDL is non-atherogenic. |
| 4 | Soto-Mota, Norwitz et al. 2025, KETO-CTA longitudinal (RETRACTED) | Prospective single-arm imaging study, retracted | 100 LMHRs, ~1 year | Industry-funded Crowdfunded; vendor Cleerly; coauthor held undisclosed Cleerly equity. | Retracted 2026; no control arm; ~1 year; no hard endpoints; plaque progressed. |
| 3 | Moore et al. 2022, critique/reanalysis | Peer-reviewed critique letter | Reanalysis of the 548-person dataset | Independent No industry funding; independent academic critics. | Letter format; reanalysis of the same self-selected data. |
| 6 | Ference et al. 2017, EAS Consensus Statement | Expert consensus synthesizing genetics, epidemiology, RCTs | >2 million in cohorts; >200,000 in RCTs | Independent European Atherosclerosis Society panel; typical individual pharma disclosures. | Consensus synthesis, not a single trial; not specific to LMHR. |
The strongest counterexample is familial hypercholesterolemia: lean, otherwise healthy people with lifelong high LDL get premature atherosclerosis, showing high LDL is not benign just because someone is metabolically healthy.
Unproven ≠ disproven
No long-term hard-outcome (heart attack, stroke, death) data exist in LMHRs; the entire safety claim rests on short-term surrogate imaging and mechanism.
Where claim and evidence diverge
A finding of no LDL-to-plaque correlation WITHIN a group where everyone already has very high LDL (a restricted range) is not the same as 'high LDL is safe,' yet that is how it was popularized.
The money trail
The reassuring evidence is led by low-carb advocates and funded via crowdfunding/citizen-science with commercial imaging vendor Cleerly; a coauthor's undisclosed Cleerly equity featured in the retraction.
The contrary anchors (EAS consensus, the critique letter, Retraction Watch) are independent of the proponents.
The honest read
The lipid pattern is real and the reason LDL rises is plausibly explained, but 'this high LDL is harmless' is not established; the key study was retracted and its own data show plaque getting worse, against strong evidence that LDL causes atherosclerosis regardless of metabolic health.
What would change this verdict
A long (5 to 10 year), controlled, prospective study of LMHRs showing no excess atherosclerosis progression or cardiovascular events versus matched controls.
Replicated, independently funded, blinded imaging showing plaque does not progress in LMHRs despite sustained high LDL.
Sources
- Norwitz NG, Feldman D, Soto-Mota A, Kalayjian T, Ludwig DS. Elevated LDL Cholesterol with a Carbohydrate-Restricted Diet: Evidence for a "Lean Mass Hyper-Responder" Phenotype. Curr Dev Nutr. 2022;6(1):nzab144. PMID 35106434.
- Norwitz NG, Soto-Mota A, Kaplan B, et al. The Lipid Energy Model: Reimagining Lipoprotein Function in the Context of Carbohydrate-Restricted Diets. Metabolites. 2022;12(5):460. PMID 35629964.
- Moore JM, Diefenbach D, Nadendla M, Hiebert N. Evidence for a Lean Mass Hyperresponder Phenotype Is Lacking with Increases in LDL Cholesterol of Clinical Significance in All Categories of Response to a Carbohydrate-Restricted Diet. Curr Dev Nutr. 2022;6(5):nzac043. PMID 35669041.
- RETRACTED: Soto-Mota A, Norwitz NG, et al. Longitudinal Data From the KETO-CTA Study: Plaque Predicts Plaque, ApoB Does Not. JACC Adv. 2025;4(7):101686 (retracted 2026). PMID 40192608.
- Widely criticized keto diet study retracted. Retraction Watch. 2026.
- Ference BA, Ginsberg HN, Graham I, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. A consensus statement from the European Atherosclerosis Society Consensus Panel. Eur Heart J. 2017;38(32):2459-2472. PMID 28444290.
People also ask
- What is a lean mass hyper-responder (LMHR)?
- A lean, metabolically healthy person whose LDL rises sharply on a low-carb or ketogenic diet, showing the 'lipid triad': LDL-C of 200 mg/dL or higher, HDL-C of 80 or higher, and triglycerides of 70 mg/dL or lower. The defining 2022 survey of 548 dieters (Norwitz et al.) described this lipid pattern only, not heart-disease outcomes.
- Is the high LDL in lean mass hyper-responders harmless?
- Not established. The main reassurance came from the KETO-CTA imaging study, which was retracted in 2026; on the participants' own scans, plaque progressed over the year. Decades of genetic and trial evidence show apoB particles (LDL) cause atherosclerosis through cumulative exposure, regardless of how metabolically healthy someone is.
- What did the KETO-CTA study actually find?
- Over about one year in 100 LMHRs, baseline plaque rather than LDL or ApoB best predicted plaque change, but plaque still progressed. The paper (published as 'Plaque Predicts Plaque, ApoB Does Not') was retracted in 2026 over unblinded analysis by the imaging vendor Cleerly, an undisclosed equity conflict, and selective reporting of a secondary endpoint.
- Does being lean and metabolically healthy protect you from high LDL?
- There is no known exemption. Familial hypercholesterolemia shows that lean, fit people with lifelong high LDL still develop premature heart disease. No long-term outcome study (heart attacks, strokes, death) has yet been done in lean mass hyper-responders, so safety remains unproven.
Part of our guide: Longevity diets, fact-checked
Caveat is journalism, not medical advice. We check public claims against published evidence; we don’t diagnose, treat, or tell you what to take.