What a Ketone Does Inside a Knee

Synovium, cartilage, bone, tendon — what the evidence actually shows, tissue by tissue

Before this experiment started I scored ninety-nine of my own symptoms on a scale of zero to ten. I have published those scores. My knees were a 10. Not a 9. A 10. I stopped running seven years ago because of them. Stairs were a negotiation. A block to the train was a decision. Up the stairs, I grabbed the bannister and pulled with my arms. Down the stairs, I wedged both arms against the rails on either side and lowered myself, stair by stair, so my knees would not have to take the weight all at once. In the past two years I used a cane for more than a month. This June, my wife asked if I wanted to use the cane to walk to the grocery store. I said it was too far.

On Saturday, September 5 — Day 81 — I ran for an hour. On the same track where a sprint routine broke me before COVID. Three point seven miles of running, five miles in all.

My knees: nothing. Not a twinge on the stairs the next morning. Not the morning after that. Five days out, as I write this, nothing.

My quads were wrecked. That is delayed onset muscle soreness, and it is on schedule. But the knees, the joints that were a 10 out of 10 on my own scorecard, the joints that ended my running career seven years ago, did not send a bill.

I do not know why. But I have candidates, and I am going to walk you through them tissue by tissue, because the knee is not one thing. It is a system. And BHB — the molecule I have been running on for months — touches every part of that system.

I want to be honest up front about the evidence. Some of what follows is established in animals. Some is established in cells in a dish. Some is plausible. Some is my hypothesis. I will label each one. I will not dress up a mouse study as a human trial. And where the science disagrees with what happened to me, I will tell you that too.

One more honesty note. My knees have never been imaged. The diagnosis is clinical, and it is mine. I treat people for knees; mine were worse than most of theirs.

The words I'll use

You should not have to stop and look anything up. So here is everything, in plain words, before I use it.

Ketones. A fuel your liver makes from fat when carbohydrate is scarce. Your brain and heart can run on it. So can your joints, as we will see. The one I measure is called beta-hydroxybutyrate — say "BHB" — a small molecule your liver assembles out of fat and ships into the blood. I will just say the ketone number, or the meter.

The meter. A finger-prick device. It reads blood BHB in millimoles per liter. You do not need the unit. Above 0.5 you are in ketosis. My floor is 4.0. My cruising range is above 5.

Cartilage. The smooth, slippery white cap on the ends of your bones where they meet in a joint. It is the cushion. It has no blood supply and no nerves — which is why cartilage can be badly damaged and not hurt, and why it heals so slowly when it does get damaged.

Subchondral bone. The bone directly underneath the cartilage. "Subchondral" means "under the cartilage." When osteoarthritis starts, this bone often changes before the cartilage does — it thickens, it develops little fluid-filled pockets, it gets inflamed. It has nerves. It hurts.

Synovium. The thin membrane that lines the inside of the joint capsule — the bag that holds the joint. It makes synovial fluid, the slippery liquid that lubricates the joint and feeds the cartilage. When the synovium gets inflamed — a condition called synovitis — it pours inflammatory chemicals into the fluid, and the whole joint suffers. This is where most of the pain in an arthritic knee actually comes from. Not the cartilage. The synovium.

Ligaments. The tough bands that hold the bones of a joint together. The ACL — anterior cruciate ligament — is the famous one. Ligaments have a poor blood supply too, which is why they heal slowly.

Tendons. The cords that attach muscle to bone. The quadriceps tendon runs over the front of the knee. Same poor blood supply, same slow healing.

NLRP3 inflammasome. Say "the inflammasome." A protein complex inside immune cells that acts like a fire alarm for cellular damage. When it goes off, it releases inflammatory chemicals called cytokines — IL-1β and IL-18 are the two big ones — that drive swelling, pain, and tissue breakdown. It is one of the body's master alarm switches. BHB turns it off.

HDAC. Histone deacetylase. Say "aitch-dee-ay-cee." A family of enzymes that act like librarians in your cell's nucleus, deciding which genes are open for reading and which are shelved. HDACs shelf genes. BHB blocks HDACs, which unshelves genes — including the ones that protect against oxidative stress.

HCAR2. Hydroxycarboxylic acid receptor 2. Say "aitch-car-two." A receptor — a docking station — on the surface of cells that BHB plugs into. When BHB docks, it sends an anti-inflammatory signal into the cell. Think of it as a keyhole that BHB fits, and when the key turns, the cell calms down.

Osteoarthritis. Say "OA." The most common joint disease on Earth. Not "wear and tear" — that is the old story, and it is wrong. OA is an inflammatory disease of the whole joint — cartilage, bone, synovium, ligaments, the fluid — driven by low-grade chronic inflammation. The whole joint is the patient.

Mitochondria. The power plants inside your cells. They burn fuel to make ATP — adenosine triphosphate, the cell's energy currency. Every step you take, every beat of your heart, every repair job your body runs is paid for in ATP minted inside mitochondria. The number and quality of your mitochondria determines how much you can do and how well you recover.

What is actually wrong inside a sixty-year-old arthritic knee

Before I can tell you what BHB might do, I have to tell you what is broken. Because "arthritis" is not one thing. It is a system failure, and it happens in layers.

The old story is wrong. For most of the twentieth century, osteoarthritis was called "wear and tear." The cartilage wears down like brake pads, the story went, and eventually bone rubs on bone. Simple. Mechanical. Inevitable.

That story is wrong. Or at least it is incomplete enough to be misleading. We know this because the pain of osteoarthritis does not correlate with the amount of cartilage loss on an X-ray. Some people with severe cartilage loss have no pain. Some people with mild cartilage loss are disabled by pain. If it were just brake pads, the picture and the pain would match. They do not.

The modern understanding — and this is established, not speculative — is that osteoarthritis is an inflammatory disease of the whole joint. Not just the cartilage. The cartilage, the bone underneath it, the synovium that lines the joint, the fluid inside it, the ligaments and tendons around it. All of it. The whole joint is the patient.

Here is what that looks like, layer by layer.

The cartilage. Cartilage is the smooth, slippery white cap on the ends of your bones. It is made of a mesh of collagen fibers — type II collagen, the tough structural kind — filled with a gel of water and large molecules called proteoglycans. The most important proteoglycan is called aggrecan. Think of cartilage as a wet sponge made of collagen netting. The sponge is what absorbs the shock when your heel hits the ground.

The cells that maintain this sponge are called chondrocytes — say "kon-dro-sites." They are the only cells in cartilage. There are not many of them, and they do not divide much. They sit inside the sponge like lighthouse keepers, maintaining the structure, replacing worn-out proteoglycans, patching the collagen mesh.

In osteoarthritis, the chondrocytes start to fail. They become senescent — a word that means "old and stopped working properly." Senescent chondrocytes do not just sit there quietly. They become zombie cells. They stop maintaining the sponge and start secreting inflammatory chemicals — the same cytokines, IL-1β and IL-18, that the inflammasome produces. They also secrete enzymes called matrix metalloproteinases — say "MMPs" — that chew up the collagen mesh. The cartilage starts to break down from the inside, digested by its own caretakers.

Why do chondrocytes go senescent? Many reasons. Age. Mechanical overload. Oxidative stress — the sparks that fly off damaged mitochondria. And inflammation from the synovium, which brings us to the next layer.

The synovium. The synovium is the thin membrane that lines the inside of the joint capsule. It is the bag that holds the joint. It makes synovial fluid — the slippery liquid that lubricates the joint and, critically, feeds the cartilage. Remember: cartilage has no blood supply. It gets all its nutrients from the synovial fluid, by diffusion. The synovium is the kitchen. The cartilage is the dining room. If the kitchen catches fire, the dining room starves.

In osteoarthritis, the synovium gets inflamed. This is called synovitis. Immune cells — macrophages, the same cells Youm's 2015 paper studied — infiltrate the synovium and start pouring inflammatory cytokines into the synovial fluid. IL-1β. IL-18. TNF-α. These chemicals diffuse into the cartilage and drive the chondrocytes toward senescence. They also make the synovial fluid itself thinner and less protective. The joint's lubricant becomes an irritant.

This is where most of the pain comes from. The synovium has nerves. The cartilage does not. When your knee hurts, it is usually the synovium screaming, not the cartilage.

The subchondral bone. Underneath the cartilage is bone. Subchondral bone — the bone directly below the cartilage. In a healthy joint, this bone is a shock absorber. It flexes slightly under load, distributing the force.

In osteoarthritis, the subchondral bone changes early — sometimes before the cartilage does. It thickens. It develops fluid-filled pockets called bone marrow lesions, which show up on MRI as bright spots and are strongly associated with pain. It becomes stiffer, which means it absorbs less shock, which means more force goes to the cartilage, which means the cartilage breaks down faster. The bone and the cartilage are in a vicious cycle. Hu's 2021 review in Bone Research is the standard reference: subchondral bone lesions, including bone marrow edema and abnormal blood vessel growth, develop earlier than cartilage damage in many cases of OA.

The subchondral bone has nerves. It hurts. When your knee aches at rest, deep inside, that is often the bone, not the cartilage.

The ligaments and tendons. The ligaments hold the bones together. The tendons attach the muscles to the bones. Both are made of dense collagen — type I collagen, the rope kind, not the sponge kind. Both have poor blood supply, which is why a torn ACL or a damaged patellar tendon heals so slowly.

In osteoarthritis, the ligaments and tendons around the knee are not innocent bystanders. They are exposed to the same inflammatory soup in the synovial fluid. They become stiffer, less elastic, more prone to micro-tears. The quadriceps tendon — the big one over the front of the knee — can become chronically inflamed, a condition called tendinopathy, which is a common source of knee pain in older adults.

So the arthritic knee is not a worn brake pad. It is a house on fire. The cartilage is being digested by its own caretakers. The synovium is pouring gasoline. The bone underneath is thickening and hurting. The ligaments and tendons are stiffening in the heat. And the whole system is running on inflammatory chemicals that keep the fire going.

Now — what happens when you flood that system with BHB?

What BHB does at each tissue — the evidence, tissue by tissue

I am going to walk through each tissue in the knee and tell you what the science actually shows about BHB. I will label every claim: established (in animals or humans), established in cells, plausible, or speculative. I will cite the papers. And where the science is thin or disagrees, I will say so.

1. The synovium — where BHB has the strongest case

The synovium is where the pain lives. It is also where BHB has its best evidence.

In 2015, Youm and colleagues at Yale published a paper in Nature Medicine that changed how we think about ketones. They showed that BHB — specifically BHB, not the other ketone bodies — blocks the NLRP3 inflammasome. The inflammasome is the fire alarm inside immune cells. When it goes off, it releases IL-1β and IL-18, the two cytokines that drive most of the inflammatory damage in osteoarthritis.

BHB blocks the alarm. It does this by preventing potassium from leaking out of the cell — the inflammasome needs that potassium leak to assemble itself — and by stopping a protein called ASC from clumping into the spike that activates the whole complex. No potassium leak, no ASC clump, no alarm, no cytokines.

The paper showed this in human monocytes — immune cells — and in mouse models of inflammatory disease. Mice given BHB had less inflammation. Mice genetically engineered to lack the inflammasome had the same protection. BHB was doing the same job as the genetic knockout.

That is established in cells and in mice. It is not a knee study. But the synovium in an arthritic knee is full of macrophages — the same immune cells Youm studied — and those macrophages are running the NLRP3 inflammasome, pouring IL-1β and IL-18 into the synovial fluid. If BHB is present at sufficient concentration, it should quiet those macrophages. That is the mechanism.

How much BHB does it take? Youm's cell studies used concentrations in the range of 1 to 10 millimolar. My cruising range is 4 to 6 on the meter, which is 4 to 6 millimolar. That is inside the range where the inflammasome blockade was demonstrated. I am not saying that proves anything about my knees. I am saying the concentration is not fantasy.

There is a second mechanism at the synovium, and it is the receptor. HCAR2 — the docking station I mentioned in the glossary. BHB plugs into HCAR2 on the surface of immune cells and sends an anti-inflammatory signal into the cell. This is separate from the inflammasome blockade. It is a second channel. Graff's 2016 paper in Metabolism showed that activating HCAR2 produces broad anti-inflammatory effects across multiple tissues. The synovium is full of cells that carry this receptor.

So at the synovium, BHB has two established mechanisms — inflammasome blockade and HCAR2 activation — both anti-inflammatory, both demonstrated in the immune cells that drive synovitis. The leap is from macrophages in a dish to a human knee. That leap is plausible. It is not proven.

2. The cartilage — where BHB has two animal studies and a cellular mechanism

The cartilage is the cushion. The chondrocytes are the caretakers. In osteoarthritis, the caretakers go senescent and start eating the cushion. Can BHB stop them?

Two papers say yes, in mice.

The first is from 2024, published in Aging Cell. A Chinese research group gave BHB to mice with surgically induced osteoarthritis — they destabilized the knee joint, which is the standard way to create OA in a mouse. The BHB-treated mice had less cartilage degeneration than controls. The mechanism they identified was mitophagy — the cellular recycling program that removes damaged mitochondria. BHB enhanced mitophagy in the chondrocytes through a pathway involving HCAR2, AMPK (an energy-sensing enzyme), and two proteins called PINK1 and Parkin that tag damaged mitochondria for disposal. Cleaner mitochondria meant less oxidative stress, which meant fewer senescent chondrocytes, which meant less cartilage breakdown.

That is a real paper, in a real journal, with a real mechanism. It is in mice. It is not in humans. I will say established in animals.

The second paper is from 2024, published in the Journal of Bone and Mineral Research. This group tested both a standard ketogenic diet and an intermittent ketogenic diet in mice with OA. Both reduced inflammatory cytokines, reduced cartilage loss, and improved joint function. They identified a signaling pathway — ERBB3 — as the mechanism. This one is notable because they tested the diet itself, not just injected BHB, which is closer to what I am doing.

Also established in animals.

At the cellular level, there is a third mechanism, and it is the HDAC inhibition I mentioned in the glossary. Shimazu's 2013 paper in Science — one of the most cited papers in the ketone literature — showed that BHB blocks HDACs, which unshelves genes that protect against oxidative stress. The two genes they focused on were FOXO3A and MT2. FOXO3A is a master regulator of stress resistance. When it is unshelved, cells produce more antioxidant enzymes — superoxide dismutase, catalase — that mop up the sparks from damaged mitochondria. This was demonstrated in mouse kidney cells, not chondrocytes. But the mechanism is general. Any cell with HDACs and FOXO3A should respond. Chondrocytes have both.

So at the cartilage, BHB has: enhanced mitophagy (established in mice), reduced inflammatory cytokines and cartilage loss (established in mice), and HDAC inhibition leading to oxidative stress protection (established in cells, mechanism general). The leap to a human knee is plausible.

6. Dyson G, et al. (Jeffries M, senior author). Ketogenic diet-associated worsening of osteoarthritis histologic severity, increased pain sensitivity and gut microbiome dysbiosis in mice. ACR Open Rheumatology 2025. — Oklahoma Medical Research Foundation. Eight weeks of ketogenic diet in a mouse OA model: weight loss, but worse cartilage histology, more pain sensitivity, higher inflammatory markers and gut dysbiosis. Counterevidence. Included for honesty. ✅ verified at PMC11760994 (Sep 10, 2026).

3. The subchondral bone — where BHB has early evidence

The bone under the cartilage is where the deep ache lives. It thickens, it develops fluid pockets, it gets inflamed. Can BHB help?

The evidence here is newer and thinner, but it exists.

A 2025 review by Luo and colleagues in the journal Nutrients covered the intersection of ketogenic diets, BHB, and osteoporosis. The key findings relevant to the knee: BHB promotes osteoblast function — osteoblasts are the cells that build bone — and suppresses osteoclast activity — osteoclasts are the cells that break bone down. In osteoarthritis, the subchondral bone undergoes abnormal remodeling: too much breakdown in some spots, too much buildup in others. BHB appears to normalize this balance, at least in animal models.

A 2016 study by Saito and colleagues looked directly at BHB and acetoacetate in mouse osteoblasts. They found that BHB enhanced alkaline phosphatase activity — a marker of bone formation — and mineralization, the process by which osteoblasts lay down new bone. Acetoacetate, the other ketone body, had mixed effects. BHB was the cleaner signal.

A 2026 study by He and colleagues — very new, in Cellular Signalling — gave BHB to mice in their drinking water and found it prevented bone loss in two models of osteoporosis: estrogen deficiency (the menopause model) and type 2 diabetes. Bone density went up, bone structure improved, and the bone-dissolving cells were held back. The mechanism ran through BHB attaching itself to histones — the spools DNA winds around — which is the same kind of gene-level switch as the HDAC story above.

So at the subchondral bone, BHB has: enhanced osteoblast function (established in cells), balanced bone remodeling (established in animals), prevention of bone loss in disease models (established in animals). The leap to subchondral bone in a human arthritic knee is plausible but early.

4. The ligaments and tendons — where I have to be honest about the gap

Here is where the science gets thin, and I am not going to pretend otherwise.

There is no published study that I can find showing that BHB directly improves ligament or tendon healing in the knee. Not in humans. Not in animals. Not in cells.

What exists is adjacent. A 2024 study in Frontiers in Immunology found that treadmill training upregulated serum BHB levels and promoted functional recovery after tendon-bone insertion injury in a mouse model. But the active ingredient was the exercise, not the BHB — BHB was a marker, not the treatment. And a 2022 review in Bioactive Materials described poly-BHB — a polymer made of BHB monomers — being used as a scaffold material for ACL repair in rabbits. But that is BHB as a building material, not BHB as a signaling molecule. Different thing entirely.

So for ligaments and tendons, I can tell you what is plausible based on mechanism — BHB's anti-inflammatory effects should reduce the inflammatory tendinopathy that makes the quadriceps tendon ache — but I cannot cite a direct study. I will say speculative. If a reader finds a paper I missed, I want to know about it.

The liver is a pharmacy

One thing ties all four tissues together, and it is worth saying plainly.

While I ran, my liver was doing two jobs at once. It was making glucose from scratch so my blood sugar would not crash — that is what emptied muscles demand. And it was making BHB from fat to feed my brain and heart. Both factories running, from different raw materials, without missing a beat.

The part that matters for the knee: the BHB the liver makes does not stop at the brain. It circulates to every tissue in the body — the synovium, the cartilage, the bone underneath. And in those tissues it is not just fuel. It is a signal. It is telling the macrophages in the synovium to stand down. It is telling the chondrocytes to clean up their mitochondria. It is telling the osteoblasts to keep building.

The liver is not just a fuel depot. On a ketogenic diet, it is dispensing anti-inflammatory medication with every shift. Whether the dose reaching a sixty-year-old knee is enough — that is the question I cannot answer from the literature. I can only answer it from the stairs.

Thirty years of knees

I should tell you that mine is not the first knee I have watched do this.

For the first fifteen years of my practice I ran medically supervised very-low-calorie programs. Patients with bad knees, hips and backs got better, and we credited the weight loss. We never measured ketones. We did not understand them as the agent.

Since I started measuring, I have watched knees, ankles, backs, multi-joint pain, muscle and tendon inflammation follow the same rule: reach the number and hold the diet, and the pain is less within a month and usually gone within two. I have not seen it fail in a patient who did what was asked. That is not a study. It is survivorship, with no imaging, from one physician's memory. I am telling you so that you can weigh it for what it is.

What I'd do with this, if I were you

If your knees hurt — if stairs are a negotiation, if a block to the train is a decision — I cannot tell you that ketosis will fix them. I can tell you that the mechanism is real, that the animal evidence is favorable, that the human evidence is me but growing, and that the risk of trying is a finger prick and a dietary change. The alternative is the bannister.

I will keep filming the stairs. If the knees hold up through weeks of daily exercise, that is not proof. But it is another data point on the line.

I ran for an hour. My knees did not send a bill. What the meter did afterward is the other half of the story: /blog/the-deconditioning-tax

— Barry

I post my blood ketone number every day — including the bad ones — in a 90-second video: youtube.com/@neuraliftmd. Day one is there. So is every failure since.

If you want to do something like this with a physician watching, I'm taking on a small founding cohort. It starts with a conversation: calendly.com/hello-neuralift/30min

Educational only, not medical advice. Talk to your doctor before changing your diet, medications, or exercise.

Where all this comes from

1. Youm YH, Nguyen KY, Grant RW, et al. The ketone metabolite β-hydroxybutyrate blocks NLRP3 inflammasome-mediated inflammatory disease. Nature Medicine 2015;21(3):263–269. — BHB, not acetoacetate, blocks the NLRP3 inflammasome in human monocytes and mouse models by preventing potassium efflux and ASC oligomerization. The foundational paper for BHB as an anti-inflammatory signal. Cited over 2,600 times. Verified.
2. Graff EC, Fang H, Wanders D, Judd RL. Anti-inflammatory effects of the hydroxycarboxylic acid receptor 2. Metabolism 2016;65(2):102–113. — HCA2 (HCAR2) activation produces broad anti-inflammatory effects across multiple tissues. Verified.
3. Shimazu T, Hirschey MD, Newman J, et al. Suppression of oxidative stress by β-hydroxybutyrate, an endogenous histone deacetylase inhibitor. Science 2013;339(6116):211–214. — BHB inhibits HDACs 1, 3, and 4, upregulating FOXO3A and MT2, which increases oxidative stress resistance. Demonstrated in mouse kidney; mechanism is general. Cited over 2,200 times. Verified.
4. β-Hydroxybutyrate enhances chondrocyte mitophagy and reduces cartilage degeneration in osteoarthritis via the HCAR2/AMPK/PINK1/Parkin pathway. Aging Cell 2024. — BHB-treated mice with surgically induced OA had less cartilage degeneration. Mechanism: enhanced mitophagy in chondrocytes through HCAR2/AMPK/PINK1/Parkin. Established in mice. Verified via PubMed 39126207.
5. β-Hydroxybutyrate ameliorates osteoarthritis through activation of the ERBB3 signaling pathway in mice. Journal of Bone and Mineral Research 2024;40(1):140. — Both standard and intermittent ketogenic diets reduced inflammatory cytokines, cartilage loss, and improved joint function in OA mice. Established in mice. Verified via PubMed 39498503.
6. Oklahoma Medical Research Foundation (OMRF) study. Keto diet may worsen osteoarthritis. 2025. — Ketogenic diet worsened arthritis and pain in their mouse model despite weight loss. Counterevidence. Included for honesty. Verified via OMRF press release.
7. Luo C, et al. Ketogenic diet and β-hydroxybutyrate in osteoporosis. Nutrients 2025. — Review: BHB promotes osteoblast function, suppresses osteoclast activity, normalizes bone remodeling. Verified via PMC11798809.
8. Saito A, et al. Enhanced and suppressed mineralization by acetoacetate and β-hydroxybutyrate in cultured osteoblasts. Biochemical and Biophysical Research Communications 2016. — BHB enhanced alkaline phosphatase activity and mineralization in mouse osteoblasts. Established in cells. Verified.
9. He Y, Mou C, Tang M, et al. β-Hydroxybutyrate prevents bone loss partially via the histone β-hydroxybutyrylation pathway. Cellular Signalling 2026;145:112562. — BHB given in drinking water raised bone mineral density, improved bone microstructure and inhibited osteoclast formation in two mouse models of osteoporosis: estrogen deficiency (ovariectomy) and type 2 diabetes (db/db). Mechanism: histone β-hydroxybutyrylation via ACSS2/P300. Established in animals. Verified at PubMed 42069096 (Sep 10, 2026).
10. Hu Y, Chen X, Wang S, Jing Y, Su J. Subchondral bone microenvironment in osteoarthritis and pain. Bone Research 2021;9(1):20. — Subchondral bone lesions including bone marrow edema and angiogenesis develop earlier than cartilage damage in OA; strongly associated with pain. Verified.
11. Sanchez-Lopez E, et al. Synovial inflammation in osteoarthritis progression. Nature Reviews Rheumatology 2022. — Synovitis is a key driver of OA progression and pain; HMGB1 and inflammatory cytokines elevated in synovium and synovial fluid of OA patients. Verified via PMC9050956.
12. My own record. Aug 27 workout: 63 min, average HR 126, next three mornings 2.6 / 2.6 / 3.7, glucose 89 / 106 / 105. Sep 5 run: approximately 1 h 15 min total session, 3.7 miles running continuous, 5 miles total including walking; readings 3.7 before, 3.3 after, 3.1 / 2.6 / 3.6 over the next 36 hours. Sleep tracker nervous-system score at the floor of its scale the night after. Baseline symptom scorecard, June 2026: knees 10 of 10. Orthostatic blood pressure changes noted this summer; four dehydration episodes.