Does Low Carb Raise Cortisol? Yes, and the One Test That Would Show It Is the One Nobody Orders

Cold hands in a warm room. A pulse that sits in the fifties and gets called "athletic." Sleep that breaks at 3 a.m. for no reason you can name. Hair thinning at the temples. A workout that used to feel good and now costs you two days.

You went low carb, and for a while it was the best you had felt in years. Then somewhere around month four or month eight, the energy quietly left and did not come back. So you did the responsible thing. You asked for labs. Your cortisol came back normal. Your TSH came back normal. And you were told, in the kindest possible way, that nothing is wrong with you.

Here is the part nobody explains: on a long-term low carbohydrate diet, a normal resting cortisol is not evidence that your stress physiology recovered. It is evidence that the single measurement your doctor ordered is the one measurement that recovers. Underneath a normal number, the way your body makes, holds, and disposes of cortisol has been rearranged, and the study that demonstrated this used a method almost no clinic runs.

This article walks through exactly what the research shows about keto and cortisol, which part of the standard low carb defense is actually true (some of it is), which part falls apart, and why your thyroid panel is quietly caught up in the same problem. Then we will get to what to do about it.

Does Low Carb Raise Cortisol? The Honest Answer Has Two Halves

Ask whether cutting carbs spikes cortisol and you will get a confident yes from one camp and a confident no from the other. Both are quoting real data. They are quoting different halves of it.

The most complete answer comes from a 2022 systematic review and meta-analysis in Nutrition and Health, which pooled 27 intervention studies covering 309 men on diets of 35% carbohydrate or less (Whittaker and Harris, 2022). It is the paper both camps cite, and it is worth reading carefully, because it does not say what either side claims.

What it found in the first three weeks: low versus high carbohydrate diets moderately raised resting cortisol (standardized mean difference 0.41, p < 0.01). This is the uncontroversial part. When you stop eating carbohydrate, liver glycogen empties within roughly a day, and your body has to manufacture glucose internally to supply the tissues that cannot run on fat. Red blood cells cannot run on fat. Neither can parts of the brain, the kidney, or the lens of the eye. Cortisol is one of the hormones that makes that internal manufacture happen, by breaking down tissue for raw material and by blocking muscle and fat from consuming the glucose so the nervous system can have it.

What it found after three weeks: resting cortisol showed no consistent effect. Ketone production ramps up, the brain takes some of its fuel from ketones instead of glucose, the emergency glucose-sparing job gets smaller, and the resting number drifts back toward baseline.

This is the finding the low carb world quotes, and it is a fair quote. Resting cortisol really does normalize. If the question is strictly "does a long-term low carb diet leave you with a high cortisol blood draw," the honest answer is usually no.

But the same meta-analysis found something else in the same subjects, and it does not get quoted at all. After exercise lasting 20 minutes or more, cortisol on low carbohydrate diets was much higher, and stayed that way: at 0 hours (SMD 0.78), at 1 hour (SMD 0.81), and at 2 hours (SMD 0.82). All three, p < 0.01. The authors put it plainly in their own conclusion: resting cortisol appears to return to baseline, "whilst post-exercise cortisol remains elevated."

So the picture is not "low carb is stressful at first and then it is fine." The picture is that low carbohydrate diets leave your stress response intact and exaggerated the moment anything demands energy from you, while the one condition under which you get measured, lying still in a lab chair after an overnight fast, is the condition under which you look fine.

That is one line of evidence. The second one is stranger, and it goes deeper than blood levels entirely.

The Keto Cortisol Study That Measured What Blood Tests Cannot See

In 2007, a group at the University of Edinburgh published a study in the Journal of Clinical Endocrinology and Metabolism that asked a question almost nobody asks: not how much cortisol is in the blood, but what the body is doing with it (Stimson et al., 2007).

The design matters, so here it is in plain terms. Seventeen obese men ate, for four weeks, either a high fat, low carbohydrate diet (66% fat, 4% carbohydrate) or a moderate fat, moderate carbohydrate diet (35% fat, 35% carbohydrate), and then crossed over to the other. A second group of six men did the same thing under isocaloric feeding, meaning the calories were matched on purpose so weight loss could not explain the result. While all this was happening, the researchers infused a labeled form of cortisol, a version tagged with deuterium, so they could trace individual cortisol molecules through the body and watch where they went.

Cortisol in the blood did not change. Plasma cortisol was unaffected. So was the salivary cortisol rhythm across the day. If you had drawn a standard cortisol on these men, you would have found nothing.

Underneath that unchanged number, three things had shifted on the low carbohydrate diet and only on the low carbohydrate diet:

1. Cortisol regeneration went up. An enzyme called 11beta-HSD1 takes cortisone, the inactive form, and converts it back into active cortisol inside tissues. On the high fat, low carb diet, that regeneration increased. The moderate carbohydrate diet did not do this.
2. Cortisol disposal went down. A separate set of enzymes, the 5alpha- and 5beta-reductases, inactivate cortisol and clear it. On low carbohydrate, urinary excretion of those inactivated cortisol metabolites dropped, and the clearance of the labeled cortisol slowed.
3. It was not the weight loss. Both diets caused weight loss. Only the low carbohydrate one changed cortisol handling. Under matched calories, the same thing happened. And the effect was already visible after one week, when weight loss was minimal.

The authors identified the driver directly: "Low-carbohydrate intake appears to be the key factor responsible for alterations in glucocorticoid metabolism." Protein was nearly identical between the two diets. Fat was only marginally higher than baseline. Carbohydrate was the variable that moved.

Two details deserve honesty, because they are usually left out when this study gets summarized online.

First, this is a liver effect, not a fat-tissue effect. The paper states explicitly that in subcutaneous abdominal adipose tissue, 11beta-HSD1 messenger RNA and activity were unaffected by diet. Diagrams circulating on social media that show this happening in "liver and fat cells" are overstating it. The change is hepatic: more cortisol regenerated in the liver, less inactivated there.

Second, four weeks is not decades. This study establishes that the shift begins fast and does not depend on weight loss. It does not establish what year five looks like. Anyone telling you it proves lifelong keto wrecks your adrenals is going further than the data. What it does prove is that the reassurance built on normal resting cortisol was never as solid as it sounded, because resting cortisol was normal in this study too, in the exact men whose cortisol metabolism had measurably changed.

There is a useful piece of vocabulary for what 11beta-HSD1 is doing. In a landmark review, Seckl and Walker described it as "a tissue-specific amplifier of glucocorticoid action" (Seckl and Walker, 2001). The enzyme does not change how much cortisol you make. It changes how loudly the cortisol you already have gets heard inside a given tissue. Turn the amplifier up and the signal at the tissue gets stronger while the reading on the meter never moves.

The One Test That Would Show It Is the One Nobody Orders

Now the title of this article has to pay for itself, so let me be precise rather than dramatic about it.

The measurement Stimson's group used was a stable-isotope tracer infusion combined with 24-hour urinary steroid metabolite profiling. The tracer half is a research tool. You cannot go get a deuterated cortisol infusion at a hospital outpatient lab, and I am not going to pretend otherwise.

The other half is real and available. A 24-hour urinary steroid metabolite profile, run by mass spectrometry, measures the downstream products of cortisol metabolism rather than the hormone itself. It shows the ratio of regenerated to inactivated cortisol, which is precisely the thing that moved in the study and precisely the thing a serum cortisol cannot see. Specialty and reference laboratories run it. It is used routinely in endocrine research and in workups for cortisol excess syndromes.

It is almost never ordered for a tired person on a low carbohydrate diet, for three understandable reasons. It costs more than a serum cortisol. It is rarely covered. And it answers a question most clinicians were never trained to ask, because the training frames cortisol as a concentration rather than as a flow.

Here is what I want you to take from this, and it is not "go buy an expensive urine test." It is this: a normal cortisol result tells you the concentration in one tube of blood at one moment. It does not tell you the rate at which your tissues are regenerating cortisol from cortisone. It does not tell you how fast you are clearing it. It does not tell you what happens to your cortisol 90 minutes after you exercise. Those are different questions with different answers, and on a low carbohydrate diet the research says at least two of them come back abnormal while the first one comes back fine.

A normal test is a real piece of information. It is just a much smaller piece than it feels like when you are handed it.

Why Your Thyroid Panel Looks Normal Too

The cortisol story does not stay in the cortisol lane, and this is where most people on long-term low carb are actually losing their quality of life.

Start with what carbohydrate restriction does to active thyroid hormone. The cleanest study on this is old and still unmatched in its design. In 1976, Spaulding and colleagues put obese subjects on 800-calorie diets in which only the carbohydrate content varied, from 0% of calories up to 100% (Spaulding et al., 1976). Same calories. Different carbohydrate.

The results:

• Total fasting: T3 fell 53%.
• The no-carbohydrate 800-calorie diet: T3 fell 47%. Nearly as much as total starvation.
• The same 800 calories containing at least 50 grams of carbohydrate: no significant change in T3 at all.

Read that last line again, because it is the most practically useful sentence in this entire article. Fifty grams of carbohydrate, inside an aggressive 800-calorie deficit, was enough to completely prevent the collapse in active thyroid hormone. The authors concluded that "dietary carbohydrate is an important regulatory factor in T3 production in man." It was never the calorie deficit alone doing the damage. It was the absence of carbohydrate.

And now a correction to something you have probably read on a bioenergetic blog, including in places I otherwise respect. The popular version of this story says low carb shunts T4 into reverse T3, the inactive mirror-image form, which then blocks your receptors. The Spaulding data do not support that. On the no-carbohydrate diet, rT3 showed no significant change. Reverse T3 rose only with total fasting, which the authors attributed to more severe caloric restriction than the 800-calorie diets produced. Their words: "rT3 concentration is not significantly affected by changes in dietary carbohydrate."

I looked for evidence that reverse T3 physically occupies and blocks the T3 receptor, because that claim is repeated constantly. I could not find a primary study establishing it. If you see that claim made confidently, ask for the citation.

The truth is less baroque and more damning: carbohydrate restriction lowers your active thyroid hormone directly, and it does not need a reverse T3 conspiracy to do it. Your liver converts most of your T4 into T3, that conversion depends on glucose and liver glycogen status, and you have removed the substrate.

So why does your TSH come back normal, when low T3 should push TSH up?

Because cortisol suppresses TSH. This was demonstrated in humans in 1976 and has held up since (Re et al., 1976). When investigators gave normal subjects dexamethasone, a potent glucocorticoid, basal TSH fell from 2.2 to 0.8 microunits per milliliter. TSH production fell from 79 to 30 milliunits per day per square meter. The pituitary's response to TRH stimulation was blunted from 16.4 down to 5.8. And T3 fell from 106 to 61 nanograms per deciliter, with free T3 falling from 174 to 76.

The most telling arm of that study is the one that ran in reverse. In 25 euthyroid patients, the investigators used metyrapone to lower cortisol, and serum TSH rose from 1.6 to 3.1. The authors concluded that "physiologic levels of circulating cortisol also have a suppressive effect on serum TSH."

Put the two findings together and you have the trap:

Cold hands, a slow pulse, afternoon exhaustion and thinning hair are not contradicted by a normal TSH. On a low carbohydrate diet, a normal TSH is exactly what the mechanism predicts.

A note on dexamethasone, since honesty matters more than a clean narrative: that was a high dose of a synthetic glucocorticoid, not a low carb diet. It establishes the direction and the mechanism, not the magnitude you would get from eating this way. The metyrapone arm is what tells us ordinary circulating cortisol is already exerting this effect at physiologic levels.

The Biospark Approach

Here is where we land, and I am not going to pretend this is a toss-up.

Low carbohydrate diets work for some people on real endpoints. Weight comes off. Appetite drops. Blood sugar improves. I will not tell you those results are imaginary, because they are not, and anyone who does is not being straight with you. In the Stimson study, the low carbohydrate group did lose weight.

But look at what else that study found, because it is the whole argument in one comparison: the moderate carbohydrate group lost a similar amount of fat mass without shifting cortisol metabolism at all. Same fat loss. No amplifier turned up in the liver. You did not have to pay the cortisol price to get the result.

That is the bioenergetic position in a sentence. Low carbohydrate works by lowering the demand on a struggling system. It does not rebuild the system's capacity to make energy. Those are different goals, and ours is the second one. Warmth, a decent pulse, hair, sleep and a workout you can recover from are outputs of a body making energy well, not rewards for restricting harder.

So the reflex and the alternative, side by side. The reflex when energy drops is to cut carbohydrate further, add a fast, and manage the stress. What that actually does is remove the substrate your liver needs to make T3, force your body to mobilize its own fat for fuel, and lean harder on the same cortisol machinery that got amplified in the first place. What we do instead is put the fuel back: carbohydrate restored deliberately, fat kept moderate rather than eliminated, and the results tracked with a thermometer instead of a feeling.

Practically, for someone coming off long-term low carb:

• Reintroduce carbohydrate gradually rather than all at once. Start low and build. Ripe fruit, orange juice, honey, well-cooked potato and root vegetables, and dairy if you tolerate it. Easy to digest beats high fiber here, because the goal is fuel arriving without gut irritation.
• Eat within an hour of waking, and then every three to four hours. The longest cortisol demand in your day is the overnight fast. Extending it is the opposite of the intervention.
• Keep protein adequate and fat moderate. Not zero fat. Moderate. High protein is its own problem on low carbohydrate: the same 2022 meta-analysis found that high-protein (35% or more) low carbohydrate diets dropped resting total testosterone substantially, around 5.23 nmol/L.
• Track temperature and pulse, not just how you feel. These cost nothing and they move before labs do. Waking oral temperature and a resting pulse, taken daily, tell you whether the intervention is working. A rising waking temperature is the readout you want.
• Expect a few pounds of water in the first two weeks. Refilling liver and muscle glycogen brings water with it. That is glycogen returning, not fat gained, and mistaking it for failure is the most common reason people quit the fix in week two.
• Be careful with 5alpha-reductase inhibitors while you do this. Finasteride, dutasteride and saw palmetto inhibit the enzyme family responsible for clearing cortisol, the same family already downregulated by carbohydrate restriction. Dual 5alpha-reductase inhibition in humans has been shown to promote hepatic lipid accumulation (Hazlehurst et al., 2016), in the same organ where the cortisol shift occurs. If you take one of these, that is a conversation to have with your prescriber, not a medication to stop on your own.

I want to be clear about which claims carry which weight. The cited findings are that carbohydrate restriction alters cortisol metabolism independent of weight loss, that 50 grams of carbohydrate prevented the T3 fall in a hypocaloric diet, that cortisol suppresses TSH, and that post-exercise cortisol stays elevated long term. The recommendation to restore carbohydrate to a specific level is our clinical position, reasoned from those mechanisms. Spaulding tested 50 grams inside an 800-calorie diet. He did not test our protocol. That distinction is yours to know.

---

Ready to stop guessing from a normal lab result?

The Bioenergetic Reset Program addresses the root cause, not the number on the page. Join 300+ members who have rebuilt their metabolic capacity instead of restricting their way around it.

• 9+ hours of comprehensive training
• Twice-monthly LIVE coaching calls
• Complete meal plans and carbohydrate reintroduction protocols
• Direct email support

$133/month | Cancel anytime

Start Your Reset Today

---

Cortisol and Thyroid Testing in Reading and Berks County, PA

If you have been handed a normal TSH and a normal cortisol in the Reading or Wyomissing area and told your fatigue is stress, you are in a very large group. Many residents across Berks County have spent years cycling between low carbohydrate diets, intermittent fasting and more aggressive restriction, each one working briefly before the energy leaves again.

At Biospark Health, we look at the readouts that move first. Waking body temperature. Resting pulse. Post-meal warmth. A full thyroid picture rather than TSH in isolation. These tell us what a single cortisol draw cannot, and they tell us early enough to act.

We serve clients throughout southeastern Pennsylvania, including Lancaster, Downingtown, Allentown, West Chester and the greater Philadelphia suburbs. Virtual and in-person options both work for this, because the most important measurements in metabolic recovery are the ones you take at home every morning.

If you are looking for a hormone specialist in Wyomissing or metabolic health support anywhere in Berks County, the first step is understanding what your labs did not measure.

Frequently Asked Questions

Does cutting carbs spike cortisol?

Yes, in the first three weeks. A 2022 meta-analysis of 27 studies found low carbohydrate diets moderately raised resting cortisol short term. After three weeks, resting cortisol returns toward baseline, which is where the "keto is not stressful" claim comes from. But post-exercise cortisol stays elevated in the same subjects, and separate research shows tissue-level cortisol handling stays altered even when the blood level looks normal.

What diet is best for lowering cortisol?

One that supplies enough carbohydrate to keep your liver glycogen topped up, eaten in regular meals rather than compressed into a window. Cortisol rises when fuel runs short, so the interventions that lower it are the ones that stop the shortage: eating within an hour of waking, eating every three to four hours, and including carbohydrate at each meal. Fasting and further restriction move it the wrong direction.

What happens when you stop keto?

Two things, and people confuse them. Liver and muscle glycogen refill, and glycogen holds water, so most people gain several pounds in the first one to two weeks. That is not fat. Separately, if your cells have spent months adapted to running on fat, glucose tolerance can be temporarily poor, and a large carbohydrate load may produce a bigger blood sugar rise than you expect. Both resolve. Reintroducing carbohydrate gradually rather than all at once makes both much easier.

Can keto cause anxiety?

It can contribute to it. The mental sharpness many people feel in early ketosis is partly driven by adrenaline and cortisol, which are the hormones your body uses to mobilize fuel when glucose is scarce. The same chemistry that produces focus at 10 a.m. can produce a racing pulse and 3 a.m. waking a few months later. Waking in the early hours is a classic pattern in someone whose overnight fuel supply is running out.

What are the 5 signs of high cortisol?

Waking between 2 and 4 a.m., central weight gain despite eating little, thinning skin or hair, wired-but-tired energy that crashes in the afternoon, and poor recovery from exercise. Worth knowing: on a low carbohydrate diet you can have several of these with a completely normal cortisol blood test, because a single resting measurement does not capture how much cortisol your tissues are regenerating.

Do carbs lower cortisol?

They lower the demand for it. Cortisol's job during carbohydrate shortage is to manufacture glucose out of your own tissue. Supply the glucose and the job gets smaller. This is why 50 grams of carbohydrate protected T3 in a starvation-level calorie deficit while zero carbohydrate did not: the body reads available carbohydrate as "the emergency is over."

Conclusion

The defense of long-term low carbohydrate diets rests on one true fact: resting cortisol normalizes after about three weeks. That fact is real, and the people quoting it are not lying to you.

It is also the narrowest possible reading of the evidence. In the same pooled analysis, post-exercise cortisol stayed elevated. In a controlled crossover study, cortisol regeneration in the liver went up and cortisol clearance went down, within one week, independent of weight loss, in men whose blood cortisol was perfectly normal. And carbohydrate restriction cuts active T3 by close to half while elevated cortisol keeps TSH from ever reporting it.

You were not imagining the fatigue. You were being measured with the one instrument that could not see it.

The way out is not a better test, though a better test exists. It is putting the fuel back. Carbohydrate restored on purpose, fat kept moderate, meals spaced so you are never running an internal deficit, and a thermometer on the nightstand to tell you it is working. Your body was never asking you to restrict harder. It was asking to be fed.