Right to Adequate Coenzyme Q10

“The demonstration of the ubiquitous distribution of coenzyme Q10 in tissues, and its central role in the mitochondrial respiratory chain, should have made its adequate supply a matter of routine clinical concern decades ago.” — Karl Folkers, Proceedings of the National Academy of Sciences, 1985

The case-study of a depletion no one denies

Every drug in the statin class inhibits the same enzyme, 3-hydroxy-3-methylglutaryl-coenzyme A reductase, at the same step of the same pathway. That pathway does not lead only to cholesterol. It leads, through a branch that separates below the point where statins act, to ubiquinone, the mobile electron carrier the mitochondrion needs at every second of its life. Blocking the enzyme lowers cholesterol. It also lowers coenzyme Q10. This is not a hypothesis, and it is not disputed. It is the textbook consequence of the drug’s mechanism of action.

Forty million adults in the United States take a statin. The prescribing information does not mention Q10 depletion. The clinical follow-up does not measure it. The insurance formularies do not cover the substrate that would replace it. When patients report muscle pain, fatigue, exercise intolerance, or cognitive haze, the response is a discussion of whether the symptoms are “real” or “nocebo”, not a discussion of the enzyme that was inhibited.

The right to adequate coenzyme Q10 is the case-study of a depletion no one denies, taking place under a prescription no one questions, monitored by an instrument no one runs, corrected by a substrate no one provides.

Folkers and the antecedence

Coenzyme Q10 was isolated in 1957 by two teams working independently: Frederick Crane at the University of Wisconsin, working on beef heart mitochondria, and R. A. Morton in Liverpool, working on rat liver. It was Karl Folkers, then research director at Merck, who elucidated its structure the following year and gave it the name that stuck.

Folkers spent the next four decades arguing that Q10 was not a curiosity of biochemistry but a nutrient whose adequacy was clinically decisive. He treated cardiac patients from the early 1970s and published dozens of papers on Q10 in congestive heart failure, hypertension, periodontal disease, and mitochondrial encephalomyopathies. His last paper, in 1997, was on Q10 and Parkinson’s disease. He died in 1997.

Merck itself understood the co-supplementation logic. In May 1990 the company was issued two US patents (4,929,437 and 4,933,165) covering combination formulations of an HMG-CoA reductase inhibitor with coenzyme Q10, explicitly designed to prevent the depletion the statin would cause. The patents were never developed into commercial products. Lovastatin was marketed alone. Simvastatin was marketed alone. Every statin since has been marketed alone. The depletion was known at the time of approval and was left uncorrected, by choice, at the point of prescription.

The mevalonate pathway

Cholesterol biosynthesis begins with acetyl-CoA, which is condensed to acetoacetyl-CoA, then to HMG-CoA, then reduced by HMG-CoA reductase to mevalonate. Mevalonate is phosphorylated, decarboxylated, and rearranged to isopentenyl pyrophosphate and dimethylallyl pyrophosphate, which polymerize to farnesyl pyrophosphate. From farnesyl pyrophosphate the pathway branches. One branch leads to squalene and, from squalene, to cholesterol. The other branch elongates farnesyl pyrophosphate to decaprenyl pyrophosphate, which is joined to the benzoquinone ring to form ubiquinone. A third branch leads to the isoprenylation of proteins, including small GTPases like Ras and Rho, and to the isoprenyl side chain of heme A and dolichol.

Statins block the pathway at HMG-CoA reductase. Every branch downstream of that step is affected. This is why the class produces effects beyond cholesterol reduction: reduced protein prenylation, reduced dolichol, reduced heme A, and reduced coenzyme Q10. The pleiotropic effects of statins, sometimes claimed as bonuses, are the same molecular fact seen from a different angle. There is no known way to inhibit HMG-CoA reductase without also inhibiting the synthesis of coenzyme Q10.

What the statins do to the Q10 pool

The plasma coenzyme Q10 depletion under statin therapy has been documented since Ghirlanda’s 1993 study, which showed reductions of roughly 40 percent within a month of initiation. Rundek and colleagues in 2004 showed atorvastatin 80 mg/day reducing plasma Q10 by 22 percent within thirty days. A dozen further studies in the following decade replicated the finding across the class: simvastatin, pravastatin, rosuvastatin, atorvastatin. The magnitude varies, the direction does not.

The muscle Q10 pool is more contested. Some biopsy studies show depletion, others do not. The measurement is technically demanding, the biopsies are painful, the assay is not standardized, and the compartments matter: total muscle Q10, mitochondrial Q10, and the ratio of ubiquinol to ubiquinone are three different quantities that respond differently. What the “muscle Q10 is not depleted” studies often show is that total Q10 in a homogenate can be maintained by compensatory mitochondrial biogenesis, even as functional Q10 in the respiratory chain per unit of ATP produced is dropping. The debate is a debate about the wrong compartment.

The buffered deficit

The coenzyme Q10 that matters is not in the plasma and not in the muscle homogenate. It sits in the inner mitochondrial membrane, embedded in a chain of five complexes that carries electrons from NADH and succinate to molecular oxygen, pumping protons across the membrane to generate the electrochemical gradient that drives ATP synthesis. It receives electrons from Complex I (NADH dehydrogenase) and Complex II (succinate dehydrogenase), it delivers them to Complex III (the cytochrome bc1 complex), and it does so by diffusing laterally within the lipid bilayer, oxidized and reduced dozens of times per second at each site. It is not an enzyme. It is a shuttle.

A shuttle whose concentration falls does not produce an acute failure. It produces a buffered deficit, absorbed by a machinery of compensation that is one of the oldest and most redundant in the eukaryotic cell.

• When electron transfer through the reduced Q10 pool slows, electrons leak. The leaked electrons produce superoxide. The mitochondrion responds by upregulating the antioxidant system that neutralizes superoxide: manganese superoxide dismutase, glutathione peroxidase, thioredoxin. The glutathione pool is drawn down to handle the leak. The compensation works, and the visible endpoint (cell death, apoptosis) does not occur. The invisible endpoint (chronic drawdown of the reducing capacity of the cell) accumulates.
• When ATP output per unit of substrate falls, the cell shifts fuel selection. Glycolysis increases relative to oxidative phosphorylation. The Warburg-like shift is subtle in a healthy tissue and only unmasked at high demand, but it changes the metabolic signature of the tissue over years.
• Uncoupling proteins UCP2 and UCP3 are induced. They dissipate part of the proton gradient as heat rather than as ATP. This reduces the electron leak but also reduces the efficiency of oxidative phosphorylation. The cell burns more substrate for less ATP.
• PGC-1α is activated. Mitochondrial biogenesis increases. The cell responds to per-mitochondrion insufficiency by making more mitochondria. Total mitochondrial mass rises. A muscle biopsy that measures total Q10 in a homogenate sees the pool preserved, because the compensation succeeded. The per-mitochondrion pool is lower, the demand per unit of ATP produced is higher, and the tissue is running on a wider but shallower machinery.
• Damaged mitochondria are cleared by mitophagy. Cristae are remodeled by OPA1. The architecture of the organelle is rebuilt around the shortage.

None of this fails visibly. All of it costs. The costs accumulate on the decade rather than the month, and they show up as the diffuse phenomena a modern medical vocabulary attributes to “aging”: reduced VO2max, exercise intolerance, muscle fatigue, cognitive slowing, cardiac reserve loss. These are not diagnoses. They are the accumulated interest on a buffered deficit.

The founding case of this site was a woman whose serum potassium was normal while her tissue potassium was collapsed, and whose symptoms tracked the tissue value that no test measured. The coenzyme Q10 case is that argument extended to the mitochondrial membrane. The compartment that matters is not the compartment that is sampled. The homeostatic buffering is not a nuisance to be corrected for. It is the reason the deficit is invisible. And when the drug that causes the deficit was designed to treat cardiovascular disease, the buffered organ is the heart.

Statin-associated muscle symptoms, or the measurement of the damage

The reported incidence of statin-associated muscle symptoms (SAMS) ranges from about 5 percent in blinded randomized trials to 10-30 percent in observational cohorts. The gap between the two is one of the most contested numbers in modern cardiology.

The RCT number is treated by the cardiological establishment as the true one. The observational number is treated as inflated by nocebo. The SAMSON trial, published in 2020, used N-of-1 crossover between statin, placebo, and no pill; it reported that 90 percent of the symptom burden was reproduced on placebo, and the conclusion widely drawn was that SAMS is largely psychological. What that conclusion elides is that patients recruited into SAMSON were patients already convinced they could not tolerate statins, that the crossover periods were short, that the placebo arm still produced measurable symptoms (which the design cannot separate from adaptation or from residual statin effect), and that a chronic buffered deficit is not the class of harm a two-month crossover is calibrated to detect.

The pattern to notice is the following. Symptoms that are attributed by patients to the drug are re-attributed by trialists to the patient’s expectation. The drug’s mechanism (which is undisputed) is separated from the drug’s effects (which are contested), and the burden of proof is placed on the patient who has to demonstrate their symptoms are “real”. The buffered deficit framework predicts exactly this pattern. What is measurable is intact. What has changed is a reserve that is not measured. The patient reports the change in reserve. The trial does not detect it. The trial concludes there is no change. The patient is told the symptoms are in their head.

This is not a claim that every reported statin symptom is a true Q10-related deficit. It is a claim that the mechanism predicts exactly the class of harm the current instruments cannot see, and the current instruments’ silence is being treated as evidence of the harm’s absence.

The absurdity at the core

The mainstream cardiological position on coenzyme Q10 co-supplementation with statins is, in essence: we do not have sufficient randomized controlled trial evidence to recommend it. Cochrane reviews find modest benefit for SAMS. Meta-analyses reach mixed conclusions depending on inclusion criteria. Guidelines do not recommend routine co-supplementation. Therefore, forty million people take a drug that depletes a nutrient, and neither the depletion nor the substrate is on their prescription.

The absurdity is structural. Consider what would be required for the mainstream to change its position: a large, multi-year, adequately powered RCT of coenzyme Q10 versus placebo in statin-treated patients, with clinically meaningful endpoints, funded by an entity that would benefit from its outcome. No such entity exists. The statin manufacturers have no interest in the trial: the drugs are off-patent, the profits are captured, and the trial could only complicate the label. The CoQ10 supplement industry has no capacity for a phase-III cardiology trial. Public funders do not prioritize nutrient trials in the shadow of an established pharmacological class. The absent trial is not an accident of research priorities. It is the predictable outcome of the funding architecture described in Behind Every Test, an Industry.

There is a deeper problem. Even if the trial were run, it would very probably measure the wrong endpoints on the wrong compartment on the wrong timescale. A six-month RCT with CPK elevation and self-reported myalgia as endpoints will not detect a buffered deficit whose consequences unfold over a decade in reserve capacity. The evidence base the mainstream demands is an evidence base whose methodological choices are incompatible with the class of harm being investigated. To wait for it is to make the wait permanent.

The parallel with vitamin D in COVID-19 is exact in its logic. In both cases: the mechanism is established, the intervention is essentially without downside, the observational and mechanistic evidence converge, the demand is for a class of RCT that the funding architecture will not produce, and while the discussion proceeds the population is denied the substrate whose absence the pharmacological choice has created. This is not the neutral application of an epistemic standard. It is a doctrine whose effect, if not its intent, is to protect the pharmacological status quo by requiring of any correction a burden of proof the correction cannot meet.

This is the argument developed in All Medicine Is Preventive: the essentiality of a nutrient is a relationship between the organism, its state, and the metabolic load imposed on it. Under a statin, coenzyme Q10 becomes essential in a way it was not before. The pharmacological choice creates the essentiality. The right to the substrate follows from the choice.

Coenzyme Q10 beyond the statin case

The statin case is the sharpest, but it is not the only one. Coenzyme Q10 is a mitochondrial cofactor whose adequacy conditions the function of every energy-demanding tissue.

Congestive heart failure. The Q-SYMBIO trial, published by Svend Aage Mortensen and colleagues in 2014 in the Journal of the American College of Cardiology, randomized 420 patients with NYHA class III-IV heart failure to CoQ10 300 mg/day or placebo on top of standard therapy. Over two years, the CoQ10 arm showed a 43 percent reduction in major adverse cardiovascular events, with reductions in cardiovascular mortality, hospitalization, and all-cause mortality. This is one of the rare nutrient RCTs in cardiology with an unambiguously positive outcome and clinically meaningful magnitude. It has not become standard of care. The reasons offered are that Q-SYMBIO was a single trial, that it was industry-adjacent, that the mechanism is unclear. The reasons offered do not track any principle that would be applied consistently to a pharmaceutical intervention with the same effect size.

Hypertension. Franklin Rosenfeldt and colleagues in 2007 published a meta-analysis of Q10 in hypertension showing systolic reductions of 11-17 mmHg and diastolic of 8-10 mmHg. The magnitude is comparable to first-line antihypertensives. It is not part of any hypertension guideline.

Migraine prophylaxis. Q10 100 mg three times daily reduced migraine frequency in several placebo-controlled trials, often studied alongside riboflavin and magnesium as a mitochondrial cofactor cocktail. It is listed by the American Academy of Neurology as level C evidence for prevention.

Parkinson’s disease. Shults and colleagues in 2002 reported a slowing of progression at high-dose CoQ10 (1200 mg/day), consistent with the complex I deficit documented in the substantia nigra. The larger 2014 QE3 trial did not confirm; the interpretation depends on whether the earlier signal was a Type I error or the later trial recruited too late in disease progression. The mitochondrial deficit in Parkinson’s is not in dispute.

Mitochondrial disorders. In primary Q10 deficiencies caused by mutations in the biosynthetic pathway (COQ2, COQ4, COQ6, ADCK4 and others), CoQ10 supplementation is life-changing and, in some cases, life-saving. The response of these patients to supplementation is a proof of principle that pool restoration matters, extrapolable at least in principle to the acquired deficits that constitute the ordinary case.

Aging. Tissue coenzyme Q10 declines with age in humans, most steeply in cardiac muscle. The decline is one of many strands of what is sometimes called mitochondrial aging. Whether restoring the pool alters trajectory in otherwise-healthy individuals is unresolved. That the pool falls is not.

Sources and forms

Coenzyme Q10 exists in two interconvertible forms: ubiquinone (the oxidized, quinone form) and ubiquinol (the reduced, quinol form). The two convert within the electron transport chain and within extra-mitochondrial redox systems. Nutritionally they are distinct in one respect only: ubiquinol appears to be better absorbed in older adults, whose intestinal reduction capacity may be limited. In younger adults the two forms produce comparable plasma levels at equivalent doses.

Absorption is lipophilic and improves substantially when Q10 is taken with a meal containing fat. Solubilized, emulsified, and lipid-carrier formulations show meaningfully higher bioavailability than crystalline powders. Dosing in the clinical literature ranges from 100 mg/day (general adequacy) to 300 mg/day (Q-SYMBIO heart failure protocol) to 1200 mg/day (Parkinson’s trials). There is no established toxicity ceiling; doses to 3000 mg/day have been used without serious adverse effects.

Dietary sources are modest. Organ meats (heart, liver), fatty fish (sardines, mackerel), and beef contain measurable amounts; plant sources are lower. The endogenous biosynthesis, when unimpaired, contributes more than the diet in most people. Under a statin, the diet contribution matters more, because the endogenous synthesis is what the drug is blocking.

Measurement, or its absence

Plasma coenzyme Q10 is measured by high-performance liquid chromatography with electrochemical or ultraviolet detection. The assay is well established, is offered by a handful of specialty and reference laboratories, and is essentially never ordered in ordinary practice.

The reference range in most laboratories is derived from a general population that includes statin users, that includes older adults with age-related decline, and that includes no reference to the mitochondrial pool. A plasma Q10 in the low-normal range in a person on a statin is not evidence of adequacy. It is evidence that the buffered deficit is doing what buffered deficits do: keeping the compartment sampled at the concentration that will not alarm the assay. This is the general problem described in Calibrated for Nothing: the reference range is calibrated to the population, not to the tissue.

More informative measurements exist and are almost never used:

• The ubiquinol-to-ubiquinone ratio in plasma, which reports on the systemic redox environment.
• Platelet or muscle Q10, which correlates better than plasma with mitochondrial content but requires invasive sampling.
• Functional indices of mitochondrial reserve: VO2 max, cardiac output at maximal effort, muscle phosphocreatine recovery after exercise (by 31P-MRS). These integrate over all components of the respiratory chain, but a chronic drop is not compatible with an adequate Q10 pool.

The right that follows is not to a single test. It is to a diagnostic culture that would take the mechanism seriously enough to check.

What the right to coenzyme Q10 requires

The right to adequate coenzyme Q10 is not the right to a supplement. It is the right to a clinical practice that recognizes what the pharmacological choices already made have done, and repairs them at the point of the choice.

In statin prescription:

• Automatic co-prescription of coenzyme Q10 100-200 mg/day with any statin initiation, treated as part of the drug’s indication rather than as an optional add-on.
• Plasma Q10 available on request, and offered to any patient reporting muscle, energy, or cognitive symptoms on a statin, before the symptoms are re-attributed to psychology.
• Reference thresholds that reflect the mitochondrial pool the population needs, not the population’s current distribution while half of it is being depleted by prescription.

In cardiology:

• Coenzyme Q10 recognized as an adjunct in heart failure at Q-SYMBIO-level indications, not as an alternative or curiosity.
• Coenzyme Q10 available to patients with cardiomyopathy of any origin.

In neurology and other:

• Coenzyme Q10 available for migraine prophylaxis, mitochondrial disorders, and any condition with a documented mitochondrial component.

In diagnosis:

• Plasma coenzyme Q10 and the ubiquinol/ubiquinone ratio available on request, without gatekeeping around a prior authorization that requires proof of the deficit the test would establish.

In population practice:

• Recognition, in the same language as The Life We Call Normal and The Tragedy of Diagnosis, that the buffered deficit is not the absence of harm. It is the accumulation of harm at an interest rate the current instruments do not read.

The right is the right to have the mechanism of one’s medication matched by the substrate that the mechanism removes. It is a right the pharmaceutical architecture has known since 1990 and has chosen, at every step of its distribution to forty million people, to leave unmet.

References

Crane, F. L., et al. (1957). “Isolation of a quinone from beef heart mitochondria.” Biochimica et Biophysica Acta 25: 220-221.

Folkers, K., Vadhanavikit, S., Mortensen, S. A. (1985). “Biochemical rationale and myocardial tissue data on the effective therapy of cardiomyopathy with coenzyme Q10.” PNAS 82(3): 901-904.

Ghirlanda, G., et al. (1993). “Evidence of plasma CoQ10-lowering effect by HMG-CoA reductase inhibitors: a double-blind, placebo-controlled study.” Journal of Clinical Pharmacology 33(3): 226-229.

Langsjoen, P. H., Langsjoen, A. M. (2003). “The clinical use of HMG CoA-reductase inhibitors and the associated depletion of coenzyme Q10. A review of animal and human publications.” Biofactors 18(1-4): 101-111.

Rundek, T., et al. (2004). “Atorvastatin decreases the coenzyme Q10 level in the blood of patients at risk for cardiovascular disease and stroke.” Archives of Neurology 61(6): 889-892.

Marcoff, L., Thompson, P. D. (2007). “The role of coenzyme Q10 in statin-associated myopathy: a systematic review.” Journal of the American College of Cardiology 49(23): 2231-2237.

Rosenfeldt, F. L., et al. (2007). “Coenzyme Q10 in the treatment of hypertension: a meta-analysis of the clinical trials.” Journal of Human Hypertension 21(4): 297-306.

Shults, C. W., et al. (2002). “Effects of coenzyme Q10 in early Parkinson disease: evidence of slowing of the functional decline.” Archives of Neurology 59(10): 1541-1550.

Mortensen, S. A., et al. (2014). “The effect of coenzyme Q10 on morbidity and mortality in chronic heart failure: results from Q-SYMBIO.” JACC: Heart Failure 2(6): 641-649.

Banach, M., et al. (2015). “Statin therapy and plasma coenzyme Q10 concentrations: a systematic review and meta-analysis of placebo-controlled trials.” Pharmacological Research 99: 329-336.

Wood, F. A., et al. (2020). “N-of-1 trial of a statin, placebo, or no treatment to assess side effects.” New England Journal of Medicine 383(22): 2182-2184. (SAMSON)

Golomb, B. A., Evans, M. A. (2008). “Statin adverse effects: a review of the literature and evidence for a mitochondrial mechanism.” American Journal of Cardiovascular Drugs 8(6): 373-418.

US Patent 4,929,437 (1990). Combined use of HMG-CoA reductase inhibitor and coenzyme Q10.

US Patent 4,933,165 (1990). Combination formulation.