Sage — A Metabolic Hypothesis of Ocular Surface Disease
We are born with perfect biology. That is the first and most radical claim of the metabolic philosophy of medicine. Perfect for intermittent fuel availability, natural circadian entrainment, and the obligatory movement-strength hormesis that defined hominin life. What we label disease is, in the overwhelming majority of chronic cases, not a random error or a software bug in the genome. It is the predictable downstream consequence of a biology operating in chronic compensatory mode because its upstream ATP-generating capacity has been chronically compromised by mismatch.
The ocular surface cluster—meibomian gland dysfunction (MGD), keratoconjunctivitis sicca (KCS), blepharitis, and ocular rosacea—belongs to this pattern. This essay proposes that these are not primarily disorders of "tear deficiency" or "local immune over-activation" that require perpetual replacement and suppression. On this hypothesis, they are high-energy tissues whose secretory, barrier, and lipid-synthetic functions degrade when mitochondrial ATP flux is chronically insufficient, a shortfall driven in large part by the modern dietary substrate mismatch. Conventional management—artificial tears, punctal plugs, warm compresses, and immune-modulating drops (cyclosporine, lifitegrast)—addresses the compensatory downstream signals and the symptoms they produce. On this view, it leaves the proposed upstream energy deficit unaddressed.
Energy Demand and the Compensatory Cascade
The ocular surface is metabolically expensive. Meibomian acini must synthesize and secrete complex lipids—cholesteryl esters, wax esters, and O-acyl-ω-hydroxy fatty acids—under tight temperature and compositional control. The process is energy-intensive, redox-sensitive, and dependent on continuous oxidative phosphorylation plus adequate NAD+ for local steroidogenic activity. Lacrimal acinar cells must produce aqueous tears with precise osmolarity, proteins, and growth factors. Corneal and conjunctival epithelium must maintain barrier integrity, mucin production, and rapid repair against hyperosmolar and evaporative stress. All of these functions require sustained mitochondrial ATP generation.
This hypothesis proposes that when ATP-generating capacity falls, early changes include altered lipid composition (higher melting point, greater viscosity, elevated free fatty acids and cholesterol relative to wax and cholesteryl esters), reduced expressibility, epithelial barrier leak, and local inflammatory signaling. What clinicians later name "obstructive MGD," "evaporative dry eye," or "blepharitis" may be, on this hypothesis, the visible result of that bioenergetic shortfall. Hyperosmolar stress and evaporative loss further damage mitochondria, release mtDNA, and activate cGAS-STING and NLRP3 pathways, amplifying ROS and cytokines in a self-reinforcing loop. AMPK, the cellular energy sensor, is downregulated in meibomian glands under hyperglycemic stress; its restoration (in one diabetic rat study, by metformin) preserves gland morphology and reduces local oxidative stress and inflammation. Metformin is a prescription drug; here it appears only as a finding from one high-dose rat study, untested for this use in people, and any question about it belongs with your own physician. In human corneal epithelial cells under hyperosmolar challenge, energy failure activates AMPK-mediated mitochondrial fission and mitophagy that can escalate ROS and inflammatory signaling when quality control is overwhelmed.
Dietary Drivers and Systemic Associations
This hypothesis holds that the dominant modern driver of the shortfall is dietary. Persistent availability of refined carbohydrates and excess linoleic acid may push mitochondria toward chronic compensatory modes—elevated reverse electron transport, NAD+/NADH imbalance, excess reactive oxygen species, and progressive loss of respiratory-chain efficiency. Insulin resistance and low-grade systemic inflammation follow. The same tissues that must produce meibum and tears now operate under substrate and redox stress.
Epidemiologic associations are consistent with this framing. Multiple meta-analyses and cross-sectional studies link components of metabolic syndrome—hyperglycemia, dyslipidemia, hypertension—to higher rates of dry eye disease, elevated tear osmolarity, reduced Schirmer scores, and shorter tear breakup time. A recent meta-analysis reported an odds ratio of approximately 1.66 for the association between diabetes mellitus and MGD. In a diabetic rat model, phospho-AMPK in the meibomian glands fell within two months of hyperglycemia, and glands showed dropout by four months; metformin given to these rats restored AMPK signaling, preserved gland structure, and reduced local inflammation and oxidative stress. This is a single animal study; it has not been tested in people. High-fat (Western-style) feeding in mice induces meibomian inflammation via PPAR-γ downregulation, MAPK/NF-κB activation, mitochondrial damage, and acinar apoptosis; these changes improve with dietary correction or PPAR-γ agonism. Mitochondrial inner-membrane area declines in aging lacrimal acinar cells, and mitophagy pathways are engaged. These fit the hypothesis that they are local expressions of the same ATP-generating deficit visible in muscle, liver, and brain under chronic dietary mismatch.
Ocular Rosacea as the Same Process
This essay proposes that ocular rosacea is less a skin disease that spreads to the lids than an expression of the same upstream deficit. On this hypothesis, that deficit is concentrated in tissues that share developmental and metabolic programs—the facial sebaceous units and the meibomian glands. Dysfunctional meibum (altered lipid composition, higher cholesterol content, increased viscosity) produces ductal obstruction, evaporative instability, hyperosmolar stress, and local inflammation—the clinical picture of posterior blepharitis and evaporative dry eye that defines most ocular rosacea. Facial erythema, telangiectasia, and papulopustular changes are the cutaneous counterpart. Ocular signs can precede cutaneous disease, consistent with the meibomian glands being especially vulnerable high-energy targets.
Supporting associations include higher rates of dyslipidemia and elevated CRP in rosacea cohorts, and elevated oxidative-stress markers (total oxidant status, oxidative stress index) in both tears and serum of ocular rosacea patients that correlate with meibomian gland dysfunction severity. Small intestinal bacterial overgrowth was far more common in rosacea patients than controls in one study (46% vs 5%), and treating it cleared or greatly improved skin lesions in most of those treated; the finding is from a single, partly unblinded study and has not been shown for ocular rosacea. Dietary mismatch simultaneously impairs clean mitochondrial oxidation, promotes gut dysbiosis, and raises systemic inflammatory tone; the meibomian and facial sebaceous units register the consequence as dysfunctional lipid output and persistent low-grade inflammation. On this view, inflammation, Demodex, and microbial shifts act mainly as amplifiers; Demodex blepharitis still has its own physician-directed treatment.
Why Conventional Approaches Remain Downstream
On this hypothesis, inflammation here is largely compensatory. Reduced meibum quality raises evaporative stress; damaged mitochondria release mtDNA and ROS that activate innate pathways; barrier disruption invites further cytokine signaling. Prescription anti-inflammatory drops can improve symptoms and signs, but on this hypothesis they do not restore mitochondrial capacity. Punctal plugs conserve remaining aqueous volume without correcting the lipid layer or the energy cost of producing it. On this view, the result is often a managed rather than resolved condition.
Layered Support After the Pillars
The EnSurf pillars—ancestral Fuel (prioritizing substrates that support clean oxidative phosphorylation while reducing refined carbohydrate and excess linoleic-acid load), Circadian alignment, and Strength hormesis—are proposed to address the root by restoring substrate preference, NAD+/NADH balance, and adaptive mitochondrial capacity. Once those are established, secondary compounds may amplify residual throughput.
Astaxanthin acts primarily on the redox/ROS limb. In a 30-day open-label study without a placebo group (60 middle-aged and older adults with mild-to-moderate dry eye), astaxanthin at a study amount of 6 mg twice daily was followed by improvements in OSDI, tear breakup time, corneal staining, eyelid margin signs, meibomian expressibility, and meibum quality; tear volume and gland dropout did not change. That is a study amount, shown as information only, not a recommendation. Proposed mechanisms, mostly from laboratory studies, include membrane stabilization, Nrf2/HO-1 activation, reduction of oxidative stress and ferroptosis (via GPX4), and dampening of hyperosmolarity-driven inflammation. This is consistent with the ATP-generating deficit that produces excess mitochondrial ROS.
NAD precursors are of interest because of a specific cofactor problem found in aging mouse meibomian glands. In aged mice, falling NAD+ limits 3β-HSD, the enzyme the glands use to make their own local steroid hormones; loss of this activity shrinks the glands. Whether it affects meibum quality has not been shown. In aged mice, NMN or NR eye drops raised meibomian NAD+, reactivated the enzyme, partly restored gland size (about 30% of the way back toward young levels), and reduced evaporative dry-eye signs. Whether oral precursors would help the meibomian glands in people has not been tested; the idea that they could support the same pathway is a hypothesis. Information, not a recommendation; see Sage — NAD precursors for how this hypothesis discusses the choice of precursor.
CoQ10 is a core electron-transport-chain cofactor supporting ATP generation and mitochondrial membrane antioxidant capacity. Eye drops combining CoQ10 with cross-linked hyaluronic acid outperformed plain hyaluronic acid in one small trial, though the different base makes it hard to credit CoQ10 alone. Oral CoQ10 has not been tested for dry eye; that it might help once the pillars are in place is a hypothesis, not an established use. Information, not a recommendation.
Low-dose methylene blue is proposed to act as an alternative electron carrier at hormetic doses, accepting electrons from NADH and donating them toward cytochrome c/Complex IV. In laboratory studies, this may bypass impaired Complex I or III, support ATP, and reduce ROS at primary leak sites. It aligns with mito-stack logic but remains secondary; higher doses become pro-oxidant. Methylene blue carries an FDA boxed warning for serotonin syndrome with SSRIs, SNRIs, MAOIs and other serotonergic drugs, and it is avoided in G6PD deficiency and pregnancy; any use belongs with your own physician. On EnSurf, methylene blue is information only and is discussed within the three windows on Sage — Methylene Blue; the low range here corresponds to Window A, and nothing here is a recommendation.
A GLA-containing formulation such as HydroEye (gamma-linolenic acid from black currant seed oil plus EPA/DHA, with supportive cofactors) is proposed to supply lipid building blocks and to support resolution of residual inflammation. A small randomized, placebo-controlled trial (38 postmenopausal women with moderate-to-severe dry eye, 6 months) found better OSDI symptom scores than placebo at 24 weeks, kept the conjunctival inflammation markers HLA-DR and CD11c from rising as they did on placebo, and kept the corneal surface smoother; tear production, tear breakup time, and staining did not change. The manufacturer’s label notes that it may add to the effect of blood thinners; people on blood thinners should check with their physician. Plain omega-3 (EPA/DHA) has not held up: the largest trial, DREAM (535 people, 12 months), found fish oil no better than an olive-oil placebo for symptoms or signs. Smaller studies report some benefits, and TFOS still considers omega-3 a reasonable option, but the best-controlled evidence is negative. GLA is thought to feed anti-inflammatory prostaglandin (PGE1) pathways that fish oil alone does not.
These compounds are additive rather than interchangeable. Astaxanthin and NAD precursors may more directly support the upstream ROS and cofactor deficits generated by chronic ATP shortfall. GLA-plus-omega-3 formulations may improve the lipid products the glands can assemble once energy throughput is better. Plain omega-3 contributes to broader lipid balance but is the least supported of the three by controlled trials.
Boundaries of the Hypothesis
The hypothesis is strongest for the common chronic secretory and barrier cluster. Other conditions merit brief acknowledgment as related but distinct.
Limbal stem cell deficiency. Severe, prolonged ocular surface inflammation, hyperosmolar stress, and oxidative damage can eventually exhaust or damage the limbal niche. LSCD may, in that limited sense, be a late downstream consequence of the same energy and redox failure. Most clinical LSCD, however, is driven by chemical injury, thermal burns, contact-lens trauma, or primary inflammatory attack. Metabolic support may help protect residual niche function, but it is not the primary etiology in the majority of cases.
Primary autoimmune disease. Classic Sjögren's and broader systemic autoimmunity (rheumatoid arthritis, lupus, mucous-membrane pemphigoid) that secondarily damages the ocular surface involve antigen-specific immune dysregulation and loss of tolerance. Mitochondrial dysfunction, redox imbalance, and dietary factors can amplify inflammatory tone and impair resolution, so the metabolic lens remains useful as a modifier. It is not a sufficient explanation for the disease itself.
Eyelid and ocular surface tumors. Benign lesions such as chalazia sit comfortably inside the hypothesis as local expressions of disordered meibum composition and ductal obstruction. Malignant tumors (ocular surface squamous neoplasia, sebaceous carcinoma, melanoma, basal-cell carcinoma of the lid margin) are separate diseases; at most, chronic inflammation may contribute to a permissive setting. Chronic inflammation, oxidative stress, and compensatory metabolic shifts can contribute to a permissive microenvironment—sebaceous carcinoma in particular is a sebaceous (lipid-producing) tumor. These remain neoplastic processes with additional classical oncogenic drivers and are not core targets of the present hypothesis.
Conclusion
The large majority of chronic MGD, evaporative dry eye, blepharitis, and rosacea-associated surface disease may share the same upstream bioenergetic logic suggested by associations in metabolic syndrome and diabetes and by mitochondrial studies, mostly in animals and cells. Conventional therapy has optimized the management of downstream compensation. On this hypothesis, the higher-leverage opportunity is to restore the ATP-generating capacity the tissues were built to use. On this view, the order would be pillars first, so mitochondria are no longer operating under chronic substrate and redox stress. Then, if anything, would come redox support (astaxanthin), NAD repletion, electron-transport cofactors, and targeted lipid substrates, discussed with your own physician. Local measures remain useful for breaking secondary cycles, but durable improvement may follow restoration of upstream energy capacity rather than perpetual suppression of the inflammatory response.
Autoimmune Sjögren's, severe meibomian atrophy after long-standing obstruction, limbal stem cell failure from primary trauma or inflammation, and medication toxicity remain distinct and may still require conventional measures. Individual responses vary by severity, residual glandular architecture, and concurrent metabolic status. The hypothesis attempts to unify the common secretory and barrier cluster under the most conserved imperative in biology: the metabolic persistence of ordered, energy-transforming systems against the mismatch of modern substrate availability.
A working hypothesis, open to correction and revised as new evidence arrives. Last revised October 17, 2026.
Related: Sage — Eye · Sage — Astaxanthin · Sage — NAD precursors · Sage — CoQ10 · Sage — Methylene Blue · Sage — Fewer Double Bonds · Sage — Fuel · Sage — HydroEye and membrane fats · Sage — Fuchs’ Dystrophy. Shorter version: Coach — Dry eye and the ocular surface.
See the notice at the top of this page. Working hypothesis, not medical advice.