In Categories: Science & Research

Thyroid hormones have an important role in regulating metabolic rate, and that’s why they have emerged front and center as a common concern associated with nutritional ketosis among some internet bloggers. There’s no question that when some people adopt a low-carbohydrate lifestyle, their thyroid hormone levels may change.

Several factors may contribute to the thyroid response. For example, eating fewer calories than you expend, causing weight loss, prompts the body to reduce thyroid function to slow its metabolism. That’s because the body interprets any form of energy restriction from any cause as a sign of famine, leading it to reduce metabolism by 5-15% to conserve energy stores. Second, even when energy is not restricted, a ketogenic diet is associated with sharply reduced blood levels of active thyroid hormone. In this case, even when energy is plentiful and body weight stable, either the low intake of carbs or the presence of ketones reduces circulating concentrations of active thyroid hormone.

To the casual observer, this reduction in active thyroid hormone (called T3) has been taken as evidence that carbohydrate restriction impairs thyroid function. Some have opined that carb restriction should never be maintained below 100 grams per day in order to prevent this effect. Others advocate that people on a low-carbohydrate, high fat (LCHF) diet take intermittent “holidays” from carb restriction to boost thyroid function back up to “normal.”

An alternative explanation for these changes in thyroid hormones when one is weight stable on a LCHF diet is that the body becomes more responsive to these hormones due to beneficial changes in cell structure and function when in nutritional ketosis. As a result, it can function normally at lower T3 levels. Put another way, a ketogenic diet seems to result in improved thyroid hormone sensitivity (i.e., it takes less hormone to produce the same effect), which, if anything, puts less of a burden on thyroid hormone (T4) production in the thyroid gland and its conversion to T3 in the liver.

Is this hypothesis of improved thyroid hormone sensitivity on a ketogenic diet far-fetched? In fact, it has been repeatedly demonstrated that a well formulated ketogenic diet improves insulin sensitivity, often dramatically so in people with insulin resistance or type 2 diabetes. There is also strong evidence that the brain’s sensitivity to the satiety hormone leptin is improved during nutritional ketosis as well. Thus an improved thyroid hormone response during a LCHF diet would be quite consistent with these other documented improvements in hormone sensitivity.

How many published studies from well designed, prospective human trials have shown that impaired thyroid function (i.e., hypothyroidism) occurs on a ketogenic diet? The answer to this question is quick and simple – NONE!

This fact notwithstanding, it is common to find recommendations on the internet for daily carb intakes at or above 100 grams per day to maintain “normal thyroid function.” Since 100 grams per day of even low-glycemic carbohydrates will completely suppress nutritional ketosis in most adults, this “remedy” will definitely increase blood thyroid hormone levels. But here’s the key question: are these higher thyroid hormone levels really normal, or is this a relatively hyperthyroid state (compared to nutritional ketosis) driven by excess dietary carbohydrates? What if the lower T3 levels associated with a well formulated ketogenic diet are indicative of optimum T3 sensitivity and thus the true physiologic norm for humans?

The other proposed “cure” for this “problem” is to intermittently eat lots of carbs. Be this a few days per week or one week per month, the result is a metabolic roller coaster ride that is not without consequences. We have established that it takes many weeks to fully adapt to the state of nutritional ketosis, but most of this adaptation is reversed with just a few days of eating 100 grams or more per day of carbohydrates. What sense does it make to be constantly coaxing the body to keto-adapt and then jerking it back to deal with the disposal of high carbohydrate loads? Given that blood beta-hydroxybutyrate levels characteristic of nutritional ketosis reduce oxidative stress and inflammation, why would you want to shut down this beneficial effect even part of the time?

If carbohydrate-restricted diets were having a negative effect on thyroid function, it would be predicted that a disproportionate number of people would develop clinically obvious cases of thyroid failure (hypothyroidism) while following a ketogenic diet. Indeed, it could be that there is some percentage of particularly vulnerable people who respond to a low-carbohydrate diet in this way. How could we tell if this were happening? Well, there have been a number of large randomized studies published in the last decade lasting three months or longer using low-carbohydrate or ketogenic diets. Let’s look and see how many new cases of low-carb-induced hypothyroidism were reported in these studies

Reference # Subjects LCHF Duration # Cases Hypothyroid
Foster NEJM 2003 28 1 yr 0
Yancy Ann Int Med 2004 59 6 mos 0
Gardner JAMA 2007 68 1 yr 0
Shai NEJM 2008 85 2 yrs 0
Forsythe Lipids 2008 20 3 mos 0
Saslow PlosOne 2014 15 1 yr 0
Bazzano Ann Int Med 2014 75 1 yr 0
Totals 350 0

While these studies were not purposefully designed to look for thyroid dysfunction, overt hypothyroidism is hard to miss. These studies were all run by (or involved monitoring by) top-notch physicians, so a new case of hypothyroidism would definitely have been reported as a “serious adverse event” associated with the LCHF diet. And yet, out of 350 closely monitored patients, there were none!

Tens (if not many hundreds) of thousands of people in the United States – not to mention the developed world — make a serious effort to adapt to a low-carbohydrate diet each year. For many of them, they choose this “radical” diet only after conventional diets haven’t worked. In this context, it is likely that quite a few individuals with early thyroid issues (e.g., latent Hashimoto’s thyroiditis) try low-carbohydrate in response to their early and subtle symptoms, only to develop obvious symptoms of their slow-onset hypothyroidism months or years later. So if there actually is an increased incidence of diagnosed hypothyroidism after starting a LCHF diet, this could explain an association that has nothing to do with causality. But given this analysis of the published data to date, there is little if any likelihood that a well formulated ketogenic diet causes significant thyroid dysfunction.

Medical Geek Section

Along with his early mentors and collaborators, Dr. Phinney was involved in three studies of ketogenic diets in which they measured one or more parameters of thyroid hormone response. The first study involved six subjects on a very low-calorie ketogenic diet (VLCKD) for six weeks, the second evaluated nine men given a weight-maintenance ketogenic diet for 4 weeks, and the third studied 12 adults given a VLCKD with or without exercise training for 4-5 weeks.

In addition, in 2005 Yancy et al. published a study (Nutr & Metab) of 28 diabetics given a LCHF diet for four months, during which their mean TSH values did not change significantly (1.6 to 1.4 uU/L).

So let’s query these data from three perspectives: 1) failure of the thyroid to make enough T4, 2) failure of the liver to turn enough T4 into T3, and 3) markedly improved T3 sensitivity. Starting with the latest data from Yancy et al. first, if either the thyroid gland or the liver were failing to maintain adequate thyroid hormone effect, TSH should increase. It does not, but this was only measured in one study. Second, if the liver is failing to make enough T3, then TSH and T4 values should rise. In both our 1980 and 1988 reports, T4 went down slightly. And third, in all three studies blood T3 levels went down sharply (from a mean of 151 to 92) but clinical signs and symptoms were not indicative of overt hypothyroidism.

Case in point: although unpublished, Dr. Volek ran some thyroid tests on 14 overweight/obese men whose other results were reported in a 2004 study. Eight of these men consumed a reduced energy ketogenic diet for six weeks and then switched to a low-fat diet for another six weeks. Six other men consumed the low-fat diet first and then switched to a ketogenic diet. Average free T3 concentrations in the 14 individuals were significantly lower after the ketogenic plan than the low-fat diet (3.5 vs 4.2 pmol/L). Regardless of diet order, free T3 concentrations were lower during the ketogenic diet in 13 out of 14 men (see figure). However, despite the lower free T3 levels, the measured resting metabolic rates of these subjects were not different between diets.

Statistics? The chances of tossing a coin 14 times and getting 13 heads are less than 1 in 1000 (P<0.00085).

And finally, although it was a relatively short 11-day study, Bisschop et al. fed six men weight-maintaining diets containing 85%, 44%, and 2% of energy as carbohydrates. As shown in the figure below, although TSH and REE did not decline, serum T3 values plummeted (heavy black bars in the figure). These results again show a disconnect between circulating T3 and REE in the context of a ketogenic diet.

The only viable interpretation of these data is that ketogenic diets markedly increase tissue sensitivity to T3, and thus serum T3 levels decline while the physiological response to T3 remains normal. In this scenario, both the thyroid and the liver have to do far less “work” to maintain a normal thyroid physiologic response. Taking this one step further, why would anyone want to force their thyroid or liver to greater levels of thyroid hormone production by eating lots more carbohydrates? Forcing the pancreas to make more insulin by eating more carbs clearly doesn’t do a lot of good for type 2 diabetics, and we think the same logic applies here for thyroid function.

It’s understandable why some people may be concerned or misled by the changes in thyroid hormones that occur when one adopts a LCHF lifestyle. Clearly there is a need for one or more large prospective studies to shed additional light on this issue. But given the past attitude of the NIH towards low-carbohydrate research, don’t hold your breath. In the meantime, don’t be pressured into eating more carbs than you really need on the mistaken assumption that they are required to maintain normal function of your thyroid gland.

Have more questions about nutritional ketosis? Check out our FAQ by Dr. Steve Phinney and the Virta team.

The information we provide at and is not medical advice, nor is it intended to replace a consultation with a medical professional. Please inform your physician of any changes you make to your diet or lifestyle and discuss these changes with them. If you have questions or concerns about any medical conditions you may have, please contact your physician.


  1. Avatar
    Valerie Halling April 2, 2018 at 12:47 pm

    How would this affect a person with no thryroid and on a synthetic thyroid hormone?


    1. Virta Health

      Dr. Phinney and Dr. McKenzie have a response on this here:


  2. Avatar

    I am LCHF and lost over 100 lbs. I began having issues with my thyroid about a yr into my weight loss. Tests showed everything to be normal except for my reverse t3 to ft3 ratio. I have no inflammation and my fasting blood glucose is 73. I was having anxiety daily and if I didn’t maintain a strict way of eating I would put weight on. As soon as i started eating 50 to 60 grams of carbs daily, the brutal fatigue started to lift.
    I think I may be one of the highly sensitive when it comes to a ketogenic way of eating.


  3. Avatar

    I fail to understand the logic behind measuring RT3 and then using T3 rather than FT3. In using total T3 as your measurement you are including bound hormone. Measuring RT3, and then failing to measure FT3 seems pointless to me, as the only way RT3 provides true information is in ratio to FT3.


    1. Virta Health
      Virta Health May 3, 2018 at 5:26 pm

      Thanks for your question! We asked Dr. Phinney, and he said: “This is an excellent question in the current context of thyroid function analysis. The study we did involving the rT3 analysis was completed in 1977 and published in 1980, at which time these nuances of T4 and T3 distribution and binding were in their infancy. Hopefully we can find the funding to do a modern comprehensive study of thyroid function changes through the keto-adaptation process.”


      1. Avatar
        Catherine lindsay January 8, 2019 at 12:36 am

        Looking forward to that


  4. Avatar

    Any data on low tsh ? Total thyroidectomy here and 7 months Keto … showing .01 tsh with joint pain


    1. Virta Health

      Dr. Phinney and Dr. McKenzie have a video on this here:


  5. Avatar
    Anuradha leburu May 20, 2018 at 12:14 pm

    How about amiodarone induced hypothyroidism and ketogenic diet
    Does it help weight loss


    1. Virta Health

      Dr. Phinney said he doesn’t know of any research that addresses this fascinating question!


  6. Avatar

    I began a Ketogenic diet 10 months ago and have seen relief in many of my symptoms including water retention and increased weight loss. After about 10 months however I have noticed symptoms of hair loss and with research saw that carb ups might help but I have noticed weight gain and water retention almost instantly when increasing my carbs over 35. I have done tests recently showing low copper, magnesium, and b6, but am not sure whether it could have been induced by a low carb way of eating.


  7. Avatar

    Any data or mentions of testosterone plummeting in men who restrict carbs long term, and perhaps end up in a caloric deficit some days because of the reduced hunger in nutritional ketosis?


    1. Virta Health

      We asked Dr. Phinney and this was his answer:

      There are lots of self-reports of issues like this but little quality published data to provide a definitive answer. To be useful, a published study would need to involve a sizable group of men who remained in nutritional ketosis “long term”. There are two published studies by our co-founder Dr. Jeff Volek, one lasting 6 weeks (1) and the other lasting 10 weeks (2) in which the testosterone levels were significantly increased in the men following the ketogenic diets. The problem with most studies lasting a year or longer is that low carbohydrate diet adherence is questionable at best, and thus the results are not applicable to the effects of a well-formulated ketogenic diet.

      Jeff S. Volek, Matthew J. Sharman, Dawn M. Love, Neva G. Avery, Ana L. Gomez, Timothy P. Scheett, William J. Kraemer. Body composition and hormonal responses to a carbohydrate-restricted diet. Metabolism. 2002; 51:864–870.

      Wilson JM, Lowery RP, Roberts MD, Sharp MH, Joy JM, Shields KA, Partl J, Volek JS, D’Agostino D. The Effects of Ketogenic Dieting on Body Composition, Strength, Power, and Hormonal Profiles in Resistance Training Males. J Strength Cond Res. 2017 Apr 7. doi: 10.1519/JSC.0000000000001935.


  8. Avatar

    I am on LCHF since 14 weeks (under 30g carbs since 2 weeks) and IF 16:8 as a diabetic with a daily intake of 75mg thyroxine since a decade.
    Now, when I need less T3, should I ask my doctor to lower my thyroxine prescription? As I understand it, he will measure lower values in my blood. Therefor he might want to raise the prescription …


    1. Virta Health

      We recommend discussing your dietary changes with your doctor so that they can manage your medications and adjust them appropriately.


  9. Avatar

    What is your opinion on this study

    Hypothyroidism was diagnosed and L-thyroxine medication was initiated for eight, seven and five patients (20 patients in total, 16.7%) at 1, 3, and 6 months of KD therapy, respectively. Logistic regression analysis showed that baseline TSH elevation [odds ratio (OR): 26.91, 95% confidence interval (CI) 6.48-111.76, p<0.001] and female gender (OR: 3.69, 95% CI 1.05-12.97, p=0.042) were independent risk factors for development of hypothyroidism during KD treatment in epileptic children."


    1. Virta Health

      We asked Dr. Phinney and here’s his response:

      This is a very interesting study done in Turkey looking at standard thyroid hormone testing in 128 children who had failed treatment with multiple anti-seizure drugs. Perhaps the most important observation in this report has noting to do with the ketogenic diet itself, but with these kids’ prior pharmaceutical experience.

      Specifically: “baseline TSH elevation [odds ratio (OR): 26.91, 95% confidence interval (CI) 6.48-111.76, p<0.001] and female gender (OR: 3.69, 95% CI 1.05-12.97, p=0.042) were independent risk factors for development of hypothyroidism during KD treatment in epileptic children.”

      Translation – the predominant factor associated with diagnosed hypothyroidism “during KD treatment” was an elevated TSH at baseline! Kids with an elevated TSH at baseline were 27 times more likely to appear clinically hypothyroid after starting the KD than those with normal TSH values.

      So what’s going on here? If this paper’s authors had reviewed the literature, they would have discovered that multiple anti-seizure medications interfere with thyroid function. For example, this review paper from a decade ago:
      Verrotti A, Scardapane A, Manco R, Chiarelli F. Antiepileptic drugs and thyroid function. J Pediatr Endocrinol Metab. 2008 May;21(5):401-8.

      In essence, what the authors of this recent paper have done is blame the ketogenic diet for the side effects of the prior failed drug treatment. That’s like convicting the bullet for murder and ignoring the shooter.

      We screen all of our patients for thyroid function before starting a well-formulated ketogenic diet, correct any hypothyroidism that we detect before starting the Virta Treatment, and (as emphasized in our blog post) we do not ‘chase’ a reduced free T3 value as long as the TSH and T4 values are normal. Following this protocol, we rarely observe functional hypothyroidism occurring during a well-formulated ketogenic diet.


  10. Avatar

    I have autoimmune thyroiditis and have suffered a number of hypo-throid incidents as a result of reducing my carbs on a 25:25:50 ratio diet. I also referred to the above mentioned clinical studies in seeking to understand what was happening, and based on those studies and dietary manipulation, I have now narrowed this down to the reduction in sugars i.e. I can eat plenty of carby nuts and legumes and still have a clear, what I am calling, ‘hypothyroid episode’. My conclusion is that I am suffering these effects due to reduction in T3. Given that we know that the pituitary gland takes 6 weeks to fully respond to a change in weather temperatures or levothyroxine dosage in the form of a TSH adjustment, I question whether TSH would increase quickly enough to reflect these changes in T3 in the duration of these studies. I have found that a dose of any form of sugar: fructose, sucrose, other will resolve my symptoms in a matter of a couple of hours and I suddenly get quite warm and feel more energetic, I believe this is as the build up of T4 converts to T3.


  11. Avatar

    “We have established that it takes many weeks to fully adapt to the state of nutritional ketosis, but most of this adaptation is reversed with just a few days of eating 100 grams or more per day of carbohydrates.”

    What about someone who is fat-adapted for let’s say 6+ months or longer? It doesn’t make sense to me that the body that spent the last 6 months becoming adapted to fat burning simply loses it in a few days.


    1. Virta Health

      Due to the body’s preferential usage of carbohydrates, if you increase your carb intake over 100 grams per day, you will reverse fat adaptation. If you have been fat adapted for a long period of time, you may be able to more quickly return to fat adaptation than someone who has been in ketosis for a shorter amount of time, however you will still have to adapt and may have side effects during that period of time.


  12. Avatar

    This is an amazing article and fascinating discussion. I can give my contribution to confirm the theory that we become T3 efficient the longer we are keto adapted.
    I was diagnosed with hypothyroidism of hashimoto about 25 years ago. It’s been under control ever since. I’m now 43 years old and I’ve been keto for the past 12 years. I am also an athlete and I have high energy demands on my trainings. I feel great and have no clinical symptoms. I did a routine exam in Sept 2018 and it showed high FT4, low FT3, high RT3 and almost undetectable TSH. TSH 0,075- FT4 1,71- FT3 0,18 ng/dl- RT3 23 ng/dl. I’ve been taking syhthroid for the past 25 years.
    I guess this could be interpreted as a sign that my thyroid gland is working more and no longer needs as much medication?
    I lowered my medication and I’ve been following up with routine exams. I have never taken T3. I lowered Synthroid twice since September and I’m testing every 6 weeks. My last exam, last week, shows an improvement on TSH and FT4, but FT3 and RT3 remain pretty much the same.
    Last results:
    TSH 0,22- FT4 1,50- FT3 0,14 ng/dl- RT3 32 ng/dl. Anti TPO: 44 Ui/ml.

    However, like I said, despite the blood work, I feel excellent!! My performance is amazing. My body composition is amazing. Could I feel even better?? I don’t know. Maybe?

    I lowered my thyroid med once again (it went from 155 to 112). I’m still considering supplementing with T3 but I honestly feel that I am living proof of the adaptation that you mention on the article, that after being Keto adapted for so long my physiology just functions really well and it adapted to function with a lower T3 level than would be considered normal.
    My only concern is the RT3. Should that be a concern at all?


    1. Virta Health

      Hi Diva! We aren’t able to provide specific medical advice but congratulations on feeling great!


  13. Avatar

    Why not eliminate any concern and uncertainty over thyroid hormone suppression on LCHF/keto by supplementing with thyroid hormone replacement (i.e. natural desiccated thyroid) which contains both T3 and T4 (as well as T1, T2 and calcitonin)? Hypothyroid patients don’t have a choice and are already on thyroid meds while on LCHF type diet. The question is – does anyone know of any studies which show that these patients must adjust their medication to compensate for the reduced T3?



    1. Virta Health

      We initially addressed this topic a couple of years ago in our post here: Like the myth of adrenal fatigue, the myth that a low carbohydrate diet causes thyroid dysfunction refuses to die. Yes, when following a well-formulated ketogenic diet, serum values for T3 drop, but serum T4 and TSH values don’t change. More importantly, as we noted in our blog post, despite this reduction in T3, resting metabolic rate and other biomarkers of hypothyroidism are not consistent with impaired thyroid function. Rather, all of the evidence points to the body being much more responsive to active thyroid hormone (aka free T3). This is similar to the changes in insulin sensitivity and leptin sensitivity that we have reported with initiation of a well-formulated ketogenic diet. In other words, in ketoadapted humans, we need less of hormones like insulin, leptin, and T3 to achieve better regulatory results.

      As for giving active thyroid hormone (T3) during a low carbohydrate diet, it is not clear what the purpose would be since there is not credible human evidence of hypothyroid effects associated a well-formulated ketogenic diet. In fact, well-done studies from a few decades ago noted that giving humans T3 during total fasting actually accelerated the rate of lean tissue loss above the level caused by fasting alone (Vignati 1978, Gardner 1979). This is consistent with the observation that elevated thyroid activity (i.e., hyperthyroidism) is associated with lean tissue loss.

      Gardner DF, Kaplan MM, Stanley CA, Utiger RD. Effect of Tri-Iodothyronine Replacement on the Metabolic and Pituitary Responses to Starvation. N Engl J Med 1979; 300:579-584. DOI: 10.1056/NEJM197903153001102

      Vignati L, Finley RJ, Hagg S, Aoki TT. Protein conservation during prolonged fast:
      a function of triiodothyronine levels. Trans Assoc Am Physicians. 1978;91:169-79.


Leave a reply

Your email address will not be published.