Muscle

The enzyme, and what it is actually for

Creatine kinase exists to solve a timing problem. A muscle fibre holds only enough ATP, the molecule cells spend to do work, for a second or two of hard effort. Rebuilding it from scratch takes longer than that.

The solution is a rechargeable buffer. Muscle stores creatine with a phosphate group attached, and creatine kinase transfers that phosphate onto spent ATP to regenerate it almost instantly. That system is what carries you through the first several seconds of a maximal effort, before slower energy pathways catch up. The enzyme sits inside muscle fibres in large quantities because that is where the demand is.

It leaks into the blood when those fibres are disrupted. So the test measures muscle breakdown rather than muscle function. It tells you fibres have been damaged, not how strong you are.

And this is where creatinine comes from. Creatine in muscle converts to creatinine spontaneously, at a slow and fairly constant daily rate, and the kidneys clear it. That is the entire reason creatinine reflects muscle mass, and therefore the reason eGFR does. The kidney marker and the muscle marker are two ends of the same chemistry.

Exercise moves it further than anything moves any other marker

An unaccustomed heavy session can raise creatine kinase many times over baseline, and eccentric work does it most, meaning the lowering half of a lift. It also lasts longer than most people assume. In some cases the peak comes at around day 4, and the level is still roughly 4 times baseline at day 10. This is not injury. It is the ordinary consequence of training, and the rise is part of how muscle adapts.

The adaptation is measurable, and it changes how two results should be compared. Repeat the same eccentric session a few weeks later and the damage is markedly smaller, and the creatine kinase rise with it, sometimes a fraction of the first. The protection develops after a single bout and lasts weeks to months. It is why the session that wrecked you in January does nothing by March.

The consequence is that a creatine kinase taken after a novel session and one taken after a familiar session are not comparable, even in the same person doing the same exercise. What to do about that before a test is in the Biomarker Guide entry.

One number, three different enzymes

Creatine kinase comes in 3 forms, built from 2 subunits in different combinations. Skeletal muscle is about 98 parts the muscle form to 2 parts the heart-associated form. Heart muscle carries far more of the second, somewhere between 15% and 30% of its total. A third form sits in brain.

A standard creatine kinase result is the total and does not separate them, which matters for one specific reason. Trained skeletal muscle carries a larger share of the heart-associated form than untrained muscle does. Measuring that fraction was once how heart attacks were detected, and the overlap with hard training was a recognised source of confusion. It is why troponin replaced it for that purpose. Skeletal muscle has troponins of its own, but they are different proteins from the cardiac ones and the test is built to read only the cardiac form. That separation is very good rather than perfect. The troponin T version can read high in muscle disease, where damaged muscle starts making the cardiac form again, while the troponin I version stays normal. A raised troponin is still treated as a heart result until the heart has been looked at.

Two results that are muscle wearing other clothes

This is why creatine kinase earns a place on a panel that is not otherwise about muscle at all.

The liver one. ALT and AST are described as liver enzymes, and AST in particular is also abundant in muscle. A heavy session 2 days before a test can raise both, on a report where they are printed under the heading Liver Profile. The pattern that separates them is simple once you know it. Raised ALT and AST, with a normal GGT and a raised creatine kinase, is muscle rather than liver. Without creatine kinase on the panel there is no way to tell, and this is the most common false alarm in people who train.

The kidney one. Because creatinine comes from muscle, more muscle means more creatinine produced at exactly the same kidney function, and therefore a lower calculated filtration rate. A heavily built person can return an eGFR that looks reduced while their kidneys are entirely normal. Creatine supplementation does the same thing by enlarging the pool that converts, and so does a large meal of cooked meat, because cooking converts creatine into creatinine directly, before you have eaten it.

This is the situation cystatin C exists for. It is produced at a constant rate by nearly every cell in the body rather than by muscle specifically, so it estimates filtration without inheriting the muscle problem. The Kidneys section covers what current guidance says about using it alongside creatinine.

What changes when muscle is under strain

At the extreme end is rhabdomyolysis, which is muscle breakdown severe enough to injure the kidneys. The mechanism connects this section to the first one on the page. Damaged muscle releases myoglobin, the oxygen-holding protein inside muscle fibres. Myoglobin is small enough to pass freely through the kidney's filter, and in quantity it damages the tubules it passes through. So the kidney injury is not caused by the creatine kinase at all. Creatine kinase is simply the easiest thing to measure.

There is no single agreed definition, and the numbers move depending on who is defining it. Most published work uses a creatine kinase above roughly 5 times the upper limit of normal, or above about 1,000 units per litre, together with muscle symptoms. Definitions written specifically for statin-related cases sit much higher. European consensus guidance uses above 40 times the upper limit of normal with kidney involvement or myoglobin in the urine, and older American guidance used well above 10 times with a raised creatinine. The definitions vary because the number was never the diagnosis. The kidney injury is.

Exercise is a category of its own, and it cuts both ways. Very high creatine kinase after prolonged hard exercise is common and frequently harmless, with figures above 100,000 units per litre recorded in ultramarathon runners and no apparent consequence, and there is no agreed level after exercise at which concern begins. But exercise-related rhabdomyolysis is a real event that does injure kidneys. Fewer than 1 in 10 of the people admitted to hospital with it develop acute kidney injury, which is lower than for rhabdomyolysis from other causes. The creatine kinase level is not what separates the two groups. Anti-inflammatory painkillers taken beforehand and being dehydrated are. Running and weightlifting are the 2 activities behind most reported cases. A very high number after a hard session usually means nothing. It is not a guarantee, and severe pain, weakness, swelling or dark urine after training is a reason to be seen rather than to look up a threshold.

Statins, and what the evidence actually shows

Muscle symptoms are the most common reason people stop taking statins, so this needs doing properly.

Statin myopathy is defined as muscle pain or weakness together with a creatine kinase above 10 times the upper limit of normal. It affects somewhere between 1 in 1,000 and 1 in 10,000 people a year on standard doses. Statin-related rhabdomyolysis is rarer again, at about 1 in 100,000 a year.

Because marked rises are that rare, routine creatine kinase monitoring on a statin is not recommended. The test is used when someone has symptoms, not as surveillance. That is a case where more testing is not better, and it is worth saying on a page published by a company that sells tests.

The harder part is that most statin muscle symptoms come with a normal or barely raised creatine kinase. Muscle complaints are reported by somewhere between 7% and 29% of people taking statins outside of trial conditions, and at a small fraction of that where neither the person nor the doctor knows who is taking the drug. That gap is real, and it means expectation contributes to the symptom for a meaningful number of people. It does not mean the symptoms are imagined, and it does not tell any individual which group they are in. What it does mean is that a creatine kinase can help separate muscle injury from muscle symptoms, which are not the same thing.

And one interaction connects this to another section entirely. An underactive thyroid raises the risk of statin-related muscle injury, and it is frequently missed because both are common in the same age group. It also raises creatine kinase on its own. European guidance names an underactive thyroid as one of the causes to consider where a raised creatine kinase persists, and lists it among the conditions that raise the risk of statin muscle injury in the first place. So a raised creatine kinase with muscle symptoms on a statin is a reason to check the thyroid rather than to assume the statin is the whole story.

At the other end, a low creatine kinase usually reflects carrying little muscle rather than anything being wrong, though an overactive thyroid can produce that too.

The mitochondria, which is the part nobody mentions

Muscle pain is the statin side effect everyone has heard of. The mechanism underneath it is not, and it is more interesting than the symptom.

Statins work by blocking an enzyme at the top of the mevalonate pathway. Cholesterol is made down that pathway, which is the point of the drug. But so are several other things, and one of them is coenzyme Q10, which sits in the electron transport chain inside your mitochondria and is part of how every cell in your body makes ATP. Block the top of a pathway and you reduce the output of everything downstream, not only the bit you were aiming at.

That is not speculation. Pooled trial data shows statins lower circulating coenzyme Q10 by a measurable amount. And muscle biopsies taken from people with statin muscle symptoms and a completely normal creatine kinase have shown the classic appearances of mitochondrial trouble: ragged red fibres, fat accumulating inside the muscle cells, and reduced activity of one of the respiratory chain enzymes. Their blood test was normal. Their muscle was not.

What has not been established is the direction of the arrow. Lower coenzyme Q10 in muscle does not prove it is what causes the symptoms, because the causation could run the other way. Sore muscles mean less training, less training means fewer mitochondria, and fewer mitochondria means less coenzyme Q10. Nobody has separated those cleanly.

And the obvious fix does not work. A double-blind trial gave statin users 400 mg of coenzyme Q10 a day for 8 weeks and then took muscle biopsies. Muscle coenzyme Q10 did not rise. Mitochondrial function did not change. Symptoms did not improve. So the mechanism is real and under-discussed, and the supplement sold on the back of it is running ahead of what anyone has shown.

What the benefit actually looks like in absolute terms

None of this is an argument for or against taking one. It is an argument for knowing what you are weighing, and the benefit side is routinely presented in a way that makes it look bigger than it is.

Trials report relative risk reduction. A statin cutting your risk by 30% sounds decisive. What it means is that if your ten-year risk was 10%, it is now 7%. Three people in a hundred avoid an event. Ninety-seven get nothing from it either way, and nobody knows which group they are in at the point of swallowing the tablet.

For primary prevention, meaning people who have not already had a heart attack or stroke, the published numbers sit in exactly that territory. Across the major trials the absolute risk reduction ran between roughly 1% and 2% over the few years each trial lasted. Scaled to 10 years of treatment, the number of people needing to be treated to prevent one event ranged from about 15 to about 51, depending on the trial and on how high the baseline risk was. Pooled analyses of primary prevention now do show a reduction in death from any cause, but a small one, at roughly 3 fewer deaths per 1,000 people treated over the few years the trials ran. Older pooled analyses found none at all.

Secondary prevention is a different proposition, and the 2 get conflated constantly. In people who have already had an event, the absolute risk reduction is around 5%, so roughly 1 in 20 treated for 5 years avoids one. Same drug, 3 or 4 times the benefit.

There is a second trap hiding in the relative numbers. In one analysis the relative risk reduction looked better in the lowest-risk group, 38% against 21% in the highest-risk group, while the absolute benefit ran the opposite way, at 0.18% against 1.18%. The lower your risk to start with, the better the percentage looks and the less the drug actually does for you.

So the size of the benefit depends almost entirely on how high your risk was to begin with, and that is a number you can work out yourself. The Heart and blood vessels section covers QRISK, which is the tool for working it out.

The way to actually find out, and it is built on exactly this idea

If muscle symptoms on a statin are hard to pin down across a population, there is a design that settles it for one person. Run the trial on yourself.

That is what StatinWISE did. Two hundred people in ordinary UK general practice, all of them either about to stop their statin over muscle symptoms or having already stopped in the previous 3 years. Each one did 6 blocks of 2 months across a year, 3 on atorvastatin and 3 on placebo, in a random order that neither they nor their doctor knew. They rated their muscle symptoms every day for the last week of every block.

Across the group there was no difference in muscle symptom scores between the statin blocks and the placebo blocks. And at the end, every participant was shown their own chart.

The group result is the headline. The individual charts are the point. A population average cannot tell you whether the statin is doing it to you. Six blinded blocks of your own data can. That is an n of 1 trial, and it is the same argument this entire site is built on, which is that the only comparison that means anything is you against you.

You cannot buy it, and it needs a doctor willing to arrange the blinding. But it has been done at scale in ordinary general practice, and it is the thing to ask about before you conclude anything about your own tablets.