a provocative question anchors the conversation: how much of cardiovascular risk is about cholesterol itself versus the context it operates in? Guest Josh Wageman, DPT, PA-C, PhD—a clinical lipid specialist and researcher—makes the case for using metaphor, metabolic health, and imaging to think differently about risk. The discussion spans coronary calcium scoring, lipid particles, insulin and inflammation, endurance training extremes, neurodegeneration, and why communicating science clearly actually matters.

generated summary

Illness, recovery, and why I study lipids

  • A CAC score of zero does not eliminate atherosclerotic events: Western Denmark patients with zero CAC who still experienced events often had diabetes, smoked, or had low HDL, while existing plaque predicts more plaque. [1]
  • Wageman trained in physical therapy, physician-assistant medicine, and cholesterol disturbances in Alzheimer's disease; he coaches cross-country, works in youth ministry, and wrote The Home Security System and the Lipid Neighborhood to make lipid science memorable. [2]
  • In 2009, while running roughly 90–100 miles weekly at Idaho State, Wageman contracted H1N1 influenza with a fever near 104°F, competed while unwell, repeatedly collapsed, and developed severe recurrent rhabdomyolysis.
  • His endocrine and neurological problems included thyroiditis and a TSH near 35; he became unable to climb stairs, withdrew from school, and remained ill despite many consultations and confusing laboratory findings.
  • Working with special-needs children during recovery shifted his attention from athletic performance and grades to learning, compassion, and using difficult experiences to help others.
  • Wageman returned to finish physical therapy school, became absorbed in endocrinology and biochemistry while investigating his illness, and later pursued further medical and lipid training without ever fully explaining his own illness.
  • Wageman uses humor and analogies to make lipid mechanisms memorable and encourage learning beyond isolated cholesterol numbers.

Home security, the lipid neighborhood, and coronary calcium

  • The home-security analogy begins with four defenses: control insulin and glucose, control blood pressure, minimize chronic metabolic inflammation, and avoid smoking, recreational drugs, and alcohol; the number of ApoB particles reflects the neighborhood's potential intruders.
  • Peter Attia's Straight Dope on Cholesterol series and Wageman's book both use lively metaphors to help readers understand complex lipid biology. [3]
  • Lp(a) adds a particularly dangerous intruder, and established plaque is evidence of a previous break-in: a vessel already affected by atherosclerosis is more vulnerable to subsequent events.
  • A fit friend with low inflammatory markers and otherwise favorable measurements had Lp(a) near 128 mg/dL, a CAC score around 2,814 at age 49, and five-vessel coronary bypass surgery despite a calculated 10-year risk of 2.3%.
  • Risk calculators flatten individual differences; Wageman prefers interpreting each biomarker and using coronary artery calcium (CAC) imaging to establish whether coronary disease is already detectable.
  • Feldman cites roughly 23,000 symptomatic Western Denmark patients: a zero CAC score has meaningful diagnostic value but cannot rule out all noncalcified plaque or every obstructive lesion on CT angiography. [4]
  • A young-adult CAC analysis puts the onset of detectable calcium in White men around 36 years of age; risk burden and demographics influence when imaging may become useful. [5]
  • The "power of zero" means low near-term risk after a CAC score of zero, with roughly three to five years as a possible interval before rescanning depending on risk; MESA supports risk-dependent warranty periods. [6]
  • In CARDIA, adults scanned at ages 32–46 who had any detectable CAC had approximately five times the later coronary-event risk; a low positive score at a young age warrants attention. [7]
  • The other Western Denmark analysis followed roughly 23,000 symptomatic patients: LDL-C was associated with events mainly when CAC was already positive; among CAC-zero patients, diabetes, smoking, and low HDL identified residual risk. [1]
  • Low HDL can be a marker of insulin resistance and high triglyceride turnover, while very high HDL can accompany heavy alcohol use; neither HDL value alone explains the whole metabolic environment.

The lipid energy model and direct LDL secretion

  • Lean mass hyper-responders (LMHRs) are lean people on carbohydrate-restricted diets who develop high LDL-C, high HDL-C, and low triglycerides; the cause of the dramatic LDL change is the central physiological question.
  • The lipid energy model links carbohydrate restriction and lower hepatic glycogen to increased hepatic VLDL-triglyceride secretion and faster lipoprotein-lipase-mediated triglyceride delivery to peripheral tissues, leaving more LDL particles in circulation. [8]
  • Circulating nonesterified fatty acids supply much of the immediate fuel; VLDL helps shuttle and replenish triglyceride stores, and increased throughput can coexist with low measured triglyceride concentrations.
  • Wageman's proposed lean-mass hyperabsorber mechanism begins with low insulin releasing adipose fatty acids; the liver receives abundant fatty-acid substrate without the high-insulin SREBP-1c and de novo lipogenesis response typical of insulin-resistant diabetes.
  • The mechanism diverts much of that substrate to ketones and ACAT2-mediated cholesterol esterification, with a substantial fraction of ApoB particles secreted directly as IDL or LDL, not necessarily as triglyceride-rich VLDL.
  • A hepatocyte experiment found that LDL uptake increased cholesterol esterification by about 280% and ACAT2 expression by about 870% without strongly suppressing cholesterol synthesis, an observation relevant to the hepatic export mechanism. [9]
  • Kinetic work indicates that a sizeable portion of circulating LDL can enter plasma without first passing through VLDL catabolism; Wageman cites about 38% as a physiological possibility. [10]
  • The central disagreement is the amount of triglyceride exported by the liver, not whether low insulin mobilizes fat or whether ApoB particles can travel through several lipoprotein classes.
  • Dietary chylomicrons may also replenish adipose stores in a high-fat diet; hepatic VLDL flux cannot be determined from a fasting triglyceride result or a single ApoB concentration.
  • Distinguishing the mechanisms requires isotope-tracer and lipoprotein kinetic studies of secretion, clearance, and tissue flux; model-based estimates depend on the chosen compartments and assumptions.

Cholesterol absorption, HDL, and different responders

  • Wageman focuses on chylomicron-remnant cholesterol entering a hepatic regulatory pool, altering SREBP-2, cholesterol synthesis, LDL-receptor recycling, and PCSK9; absorption and clearance can change the LDL level without a single universal pathway.
  • In the intestine, NPC1L1 is the cholesterol "ticket taker" and ABCG5/G8 the "bouncer"; Wageman hypothesizes that low insulin favors absorption or recycling of sterols in susceptible lean responders.
  • ApoE4 and ABCG5/G8 variation may contribute to strong LDL responses, but a lean carnivore runner had ApoB around 75–79 mg/dL after three years, and Wageman’s own ketogenic trials did not cause a large LDL rise.
  • Individual lean responders have shown unusually large LDL-C reductions with ezetimibe, including approximately 200 mg/dL in an anecdotal case; this motivates an intestinal-absorption intervention study, not a universal expected response.
  • A meta-analysis of 41 randomized trials found LDL-C increases after carbohydrate restriction were concentrated in participants with lower starting BMI, supporting the importance of body composition. [11]
  • Wageman's HDL mechanism relies on CETP exchanging HDL cholesteryl esters for ApoB-particle triglycerides, followed by hepatic and endothelial lipase remodeling; intestinal ABCA1 and ApoA1 may raise nascent HDL in strong absorbers.
  • Feldman gives greater weight to HDL formation during lipoprotein-triglyceride turnover and transfer of lipoprotein surface material; the same high-HDL phenotype need not prove a single mechanism.
  • Plant-sterol blood markers may fail to identify absorption accurately when a carnivore diet supplies almost no plant sterols; the marker depends partly on what enters the gut.
  • Feldman's own repeated-measurement experiment compared short and long daily eating windows and found substantial changes within a day, including total cholesterol around 329 to 277 mg/dL and calculated LDL-C about 257 to 189 mg/dL.
  • Post-meal lipemia, time of sampling, and meal patterns can confound comparisons between people and between tests, even before deciding which LDL mechanism is correct.

Falsifying the lipid hypothesis and testing clinical decisions

  • Feldman proposes a hypothetical TRIAD trial: roughly 200 LMHR participants randomized to ezetimibe or placebo, assessed over two to five years with coronary CT angiography and cardiovascular outcomes, to test whether a large ApoB reduction changes plaque.
  • A negative intervention result could test the causal LDL hypothesis for metabolically healthy LMHRs; whether a null result would change Wageman's recommendation when CAC is already positive remains unanswered.
  • For Wageman, detectable coronary plaque is the deciding clinical distinction: "plaque begets plaque," so he favors lowering ApoB or LDL in someone with established disease even if glucose and insulin measurements look excellent.
  • Wageman sees no demonstrated benefit from keeping LDL-C extraordinarily high, but recognizes that ketogenic diets can yield substantial quality-of-life improvements, including in neurological or psychiatric conditions.
  • Feldman separates the evidence in insulin-resistant or familial-hypercholesterolemia cohorts from the unanswered question of whether equally high LDL carries the same risk in lean, insulin-sensitive ketogenic individuals.
  • Metabolic health and lipoprotein concentrations are often correlated; resolving causation requires interventions and outcomes in the population whose risk is being disputed, not merely extrapolation.
  • Failed LDL-lowering drug trials and successful interventions pose different challenges for broad causal inference; abetalipoproteinemia and hepatic failure demonstrate the difference between producing and clearing ApoB particles.
  • Imaging-defined plaque and clinical events are more decisive than isolated blood tests, while the duration, scale, medication choice, and cost of a definitive trial remain difficult.

KETO-CTA, progression data, and the boundary of CAC zero

  • The initial KETO analysis compared roughly 80 ketogenic individuals with very high LDL-C against 80 matched comparators and found no greater coronary plaque burden at baseline, despite years of carbohydrate restriction. [12]
  • The subsequent KETO-CTA publication tracked 100 lean ketogenic participants for one year; baseline plaque, rather than ApoB or LDL-C, correlated with plaque progression in the analysis. [13]
  • Another set of coronary-image measurements showed unusually widespread and rapid plaque growth; Feldman identifies inconsistencies, including zero-CAC participants progressing faster than participants with existing CAC and an absence of any regression.
  • A formal inquiry into those imaging analyses is under way and Feldman has a private list of 24 problems; detailed findings and the ongoing review are not fully available in the conversation.
  • Wageman remains concerned by any existing atherosclerosis and would manage a young patient with CAC above zero more aggressively; a CAC score of zero gives more room for individualized decisions, not permanent immunity.
  • Both speakers retain a common requirement: revise beliefs when good evidence conflicts with them and avoid turning lipid research into competing identities.

Lp(a), oxidized phospholipids, and individual risk

  • Lp(a) combines an ApoB lipoprotein with apo(a), structurally related to plasminogen; its kringle domains and carried oxidized phospholipids connect it with vascular inflammation, thrombosis-related biology, and calcific aortic-valve disease. [14][15]
  • The protein may have physiological roles in injury and wound healing, and its levels can change in some inflammatory contexts; a possible function does not erase the risk of lifelong genetically high Lp(a).
  • Statins can sometimes increase Lp(a), so LDL-C lowering does not necessarily correct every lipoprotein-linked risk mechanism.
  • Wageman's patients illustrate wide variation: one person can have Lp(a) over 500 nmol/L without coronary calcification in later life, while another develops detectable calcium young; a high Lp(a) value is consequential but not a complete prognosis.
  • Lp(a) measurement can identify inherited vascular risk that low triglycerides, a good fasting insulin, or high exercise capacity cannot rule out.

ANGPTL proteins, insulin, lipase activity, and particle quality

  • ANGPTL3, ANGPTL4, and ANGPTL8 regulate lipoprotein lipase in response to feeding and tissue needs; insulin helps shift triglyceride-derived fuel between adipose storage and working muscle.
  • In the fed state, ANGPTL3/8 affects adipose triglyceride partitioning; in fasting, ANGPTL4 influences where circulating triglycerides are hydrolyzed, so the system is more than a simple on-off inhibition of lipase.
  • ANGPTL4/8 complexes in adipose tissue can interact with tissue plasminogen activator and plasminogen, producing plasmin that cleaves the complex and restores lipase activity; this connects lipid storage with local fibrinolytic biology. [16]
  • Wageman regards fasting insulin as an underused test: some people with glucose values that look acceptable have markedly elevated insulin, including cases around 45 or even 128, signaling a very different metabolic environment.
  • Historical oral-glucose-tolerance insulin curves, including Joseph Kraft's work, illustrate why glucose alone may conceal hyperinsulinemia; the response to an oral load can reveal more than a fasting glucose value.
  • After Wageman's wife began regular resistance and aerobic exercise, her LDL-C remained around 80 mg/dL while LDL particle count fell roughly from 1,300 to 994 and smaller LDL particles from about 630 to the 350s.
  • Ron Krauss's small-dense LDL "pattern B" threshold comes up near fasting triglycerides around 95 mg/dL; Wageman questions whether the common triglyceride cutoff of 150 mg/dL is metabolically reassuring.
  • Fasting triglyceride tests differ from postprandial excursions, which can be altered by ezetimibe, PCSK9 inhibition, and incretin-based drugs even when standard fasting tests miss the change.

Alzheimer's disease, brain cholesterol, and metabolic risk

  • Wageman's grandmother's death from Alzheimer's led him to study lipids in dementia; Alzheimer's, Lewy-body disease, vascular dementia, and Parkinson's share aspects of disrupted energy use and lipid handling but retain distinct pathological features.
  • Roughly 20–25% of body cholesterol resides in the brain; astrocytes normally supply cholesterol to neurons through ApoE-containing HDL-like particles for membranes, synapses, and signaling.
  • Wageman identifies three linked disturbances in Alzheimer's pathology: impaired cholesterol transport, inappropriate cholesterol or lipid-droplet accumulation with oxidized sterols such as 7-ketocholesterol, and reduced cholesterol synthesis.
  • Wageman's working mechanism places impaired neuronal substrate utilization and lipid trafficking upstream of amyloid-beta buildup and tau hyperphosphorylation, while abnormal amyloid production and clearance amplify the downstream problem.
  • ApoE4 particles can be less efficiently lipidated, creating misplaced lipid deposits, glial inflammation, and additional demands on neurons; two ApoE4 copies carry an approximately 8–12-fold Alzheimer's risk but are not a deterministic diagnosis.
  • Midlife coronary disease is linked to 29% higher Alzheimer's risk, hyperinsulinemia to 2.2-fold risk, midlife obesity to fourfold risk, and diabetes to 56% higher risk.
  • Exercise correlates with about 45% lower Alzheimer's risk, and the highest cardiorespiratory fitness accompanies the lowest dementia incidence.
  • MESA participants with greater coronary artery calcium also experienced greater dementia incidence, reinforcing the overlap between cardiovascular and brain metabolic health. [17]
  • UK Biobank data associate greater circulating ApoB with increased dementia risk, while a higher LDL-C-to-ApoB ratio correlates inversely with dementia; blood markers still do not directly quantify cholesterol traffic within the brain. [18]
  • Desmosterol tracks the Bloch cholesterol-synthesis pathway prominent in brain biology, while lathosterol reflects another pathway; the Helsinki Businessmen Study linked reduced cholesterol-synthesis markers with Alzheimer's disease. [19]
  • A Japanese cohort linked lower desmosterol-to-cholesterol ratios to Alzheimer's disease and cognitive decline, leading Wageman to question excessive cholesterol-synthesis suppression in some patients without using a blood ratio as a diagnosis of brain cholesterol deficiency. [20]
  • More accessible blood biomarkers, including phosphorylated tau, may enable earlier monitoring; preliminary individual lifestyle cases do not establish the cause of biomarker changes.

Sleep, fructose, exercise, and metabolic flexibility

  • Feldman identifies sleep and stress as continuing personal concerns, and Wageman prioritizes adequate sleep alongside diet because disturbed sleep can rapidly undermine glucose regulation.
  • One population analysis associated habitual daytime napping with greater brain volume; Wageman uses it humorously to defend occasional naps, not to prove that every nap prevents dementia. [21]
  • In a sleep-restriction experiment in healthy young adults, five nights of less than four hours' sleep produced prediabetic oral-glucose-tolerance values in half of the sleep-deprived group, indicating rapid metabolic disruption.
  • Fructose elicits little direct insulin response but can transiently deplete cellular ATP, elevate uric acid, stimulate hepatic lipogenesis, and change the redox environment when repeatedly consumed in large amounts.
  • Wageman connects increased uric acid to interference with acyl-CoA hydratase and aconitase, potentially hampering fatty-acid oxidation and the TCA cycle, and mentions observational links between uric acid and coronary-calcification progression.
  • SGLT2 inhibitors can lower uric acid, and alcohol can share aspects of fructose's hepatic metabolic burden; the physiological context matters more than a simplistic rule that everything raising insulin is harmful.
  • Carbohydrates can serve intense glycolytic training and replenish glycogen; Wageman favors matching carbohydrate intake to activity without maintaining the food habits of high-volume training after becoming sedentary.
  • Combining resistance training, endurance exercise, daily walking, and maintained muscle mass improves the capacity to clear glucose and handle fuel, particularly as activity declines with age.

Meaning, faith, care for others, and open inquiry

  • Wageman connects severe illness, a later medical emergency involving hypoglycemia, and his Christian faith with a change from achievement-centered identity toward service; Romans 8:28 anchors his interpretation of adversity.
  • Feldman does not share the same religious certainty but finds meaning in the idea of a hero's journey, voluntary service, and doing consequential work without agreement on metaphysical questions.
  • Serving children, visiting nursing homes, helping patients, and receiving heartfelt letters matter more to Wageman than credentials, social prestige, or winning an intellectual contest.
  • Wageman's purpose in lipidology is to make complex biology useful while remaining humble about what remains unknown; patients benefit when experts can disagree without tribal hostility.

References

  1. [00:00] Low-Density Lipoprotein Cholesterol Is Predominantly Associated With Atherosclerotic Cardiovascular Disease Events in Patients With Evidence of Coronary Atherosclerosis: The Western Denmark Heart Registry — https://doi.org/10.1161/CIRCULATIONAHA.122.061010
  2. [00:01] The Home Security System and the Lipid Neighborhood — Joshua Wageman (official Yale School of Medicine book feature) — https://medicine.yale.edu/news-article/alumni-bookshelf-joshua-wageman-pa-c-20/
  3. [00:25] The Straight Dope on Cholesterol — Part I — https://peterattiamd.com/the-straight-dope-on-cholesterol-part-i/
  4. [00:35] Association of Age With the Diagnostic Value of Coronary Artery Calcium Score for Ruling Out Coronary Stenosis in Symptomatic Patients — https://doi.org/10.1001/jamacardio.2021.4406
  5. [00:36] Race- and Sex-Specific Age at Which Coronary Artery Calcium Becomes Detectable Among Young Adults — https://doi.org/10.1161/CIRCIMAGING.124.016599
  6. [00:37] Warranty Period of a Calcium Score of Zero: Comprehensive Analysis From MESA — https://doi.org/10.1016/j.jcmg.2020.06.048
  7. [00:38] Association of Coronary Artery Calcium in Adults Aged 32 to 46 Years With Incident Coronary Heart Disease and Death — https://doi.org/10.1001/jamacardio.2016.5493
  8. [00:43] The Lipid Energy Model: Reimagining Lipoprotein Function in the Context of Carbohydrate-Restricted Diets — https://doi.org/10.3390/metabo12050460
  9. [00:52] Hepatic Cholesterol Homeostasis: Is the Low-Density Lipoprotein Pathway a Regulatory or a Shunt Pathway? — https://doi.org/10.1161/ATVBAHA.113.301517
  10. [00:57] The Crucial Roles of Apolipoproteins E and C-III in ApoB Lipoprotein Metabolism in Normolipidemia and Hypertriglyceridemia — https://doi.org/10.1097/MOL.0000000000000146
  11. [01:17] Increased Low-Density Lipoprotein Cholesterol on a Low-Carbohydrate Diet in Adults With Normal but Not High Body Weight: A Meta-analysis — https://doi.org/10.1016/j.ajcnut.2024.01.009
  12. [01:36] Carbohydrate Restriction-Induced Elevations in LDL-Cholesterol and Atherosclerosis: The KETO Trial — https://doi.org/10.1016/j.jacadv.2024.101109
  13. [01:36] RETRACTED: Longitudinal Data From the KETO-CTA Study: Plaque Predicts Plaque, ApoB Does Not — https://doi.org/10.1016/j.jacadv.2025.101686
  14. [02:06] Oxidized Phospholipid Modification of Lipoprotein(a): Epidemiology, Biochemistry and Pathophysiology — https://doi.org/10.1016/j.atherosclerosis.2022.04.001
  15. [02:07] Lipoprotein(a) and Oxidized Phospholipids Promote Valve Calcification in Patients With Aortic Stenosis — https://doi.org/10.1016/j.jacc.2019.01.070
  16. [02:20] Angiopoietin-Like Protein 4/8 Complex-Mediated Plasmin Generation Leads to Cleavage of the Complex and Restoration of LPL Activity — https://doi.org/10.1073/pnas.2214081120
  17. [02:37] Coronary Artery Calcium and Risk of Dementia in the Multi-Ethnic Study of Atherosclerosis — https://doi.org/10.1161/CIRCIMAGING.116.005349
  18. [02:37] Serum Lipid Traits and the Risk of Dementia: A Cohort Study of 254,575 Women and 214,891 Men in the UK Biobank — https://doi.org/10.1016/j.eclinm.2022.101695
  19. [02:39] Serum Noncholesterol Sterols in Alzheimer's Disease: The Helsinki Businessmen Study — https://doi.org/10.1016/j.trsl.2018.07.002
  20. [02:39] Reduced Plasma Desmosterol-to-Cholesterol Ratio and Longitudinal Cognitive Decline in Alzheimer's Disease — https://doi.org/10.1016/j.dadm.2014.11.009
  21. [02:44] Is There an Association Between Daytime Napping, Cognitive Function, and Brain Volume? A Mendelian Randomization Study in the UK Biobank — https://doi.org/10.1016/j.sleh.2023.05.002
you are viewing a single comment's thread
view the rest of the comments
[–] [S] 1 point 2 hours ago

this is a excellent debate, demonstrating how two curious people who don't agree can have a constructive debate and both push themselves to be better without name calling.

  • source