Free T3 peaks and valleys in T3 and NDT therapy

Free T3 peaks and valleys in T3 and NDT therapy

Lack of knowledge about Free T3 peaks and valleys contributes to failure and frustration in T3-based therapies, including desiccated thyroid and T3-T4 combination therapy.

On patient support groups, I often see conversations that go like this:

  • “My FT3 result was over reference range, FT4 low in range, and my TSH was low, so my doctor reduced my Cytomel T3 dose. Now I feel horribly hypothyroid.”
  • Another patient replies to ask, “How many hours post-dose did you do your lab test?”
  • She answers, “My doctor didn’t say. I think it was about 3 hours after my AM dose. Why, does it matter?”

Another common patient exchange usually goes like this:

  • “I am taking ## mg of NDT (Desiccated Thyroid) but I still feel tired for half the day.”
  • Another patient replies to ask, “Are you splitting your daily dose into two doses?”
  • She answers, “My doctor didn’t say I needed to split my dose. Why, should I?”

So many times, I’ve thought to myself, “I really wish all doctors who prescribe T3 or NDT medication understood the science behind the T3 dosing effect.”

In fact, even thyroid patients taking T3 or NDT should know the Free T3 dosing curve because it strongly influences their symptoms and lab results. Yes, you might be hypothyroid for half the day or more, and your lab tests might tell an uninformed doctor you’re thyrotoxic when you’re not.

This post displays one graph of normal TSH, FT4 and FT3 circadian rhythms, and then six graphs showing Free T3 fluctuations after doses of T3.

You can learn a lot just by simply scrolling through and eyeballing these graphs.

To help intelligent thyroid doctors interpret and apply them, I summarize their research context and give some commentary.

Once you understand how T3 behaves in isolation, you can apply your knowledge to various T3-T4 combinations and NDT dosing.

Therefore, most of the thyroid therapy images and notes (after the first two images) focus on T3 monotherapy because it isolates the dynamics of T3 dosing in the body.


  • NORMAL Circadian rhythm, no thyroid medication (“click to reveal” section)
    • Russell, 2008 – the natural TSH, FT3 and FT4 curves
  • When does a thyroid patient need T3 dosing? (“click to reveal” section)
  • The T3 Dosing curve
    • Saravanan et al, 2007 – Combination T3-T4 therapy, hypothyroid result
    • Hershman, 2009 – The different timing of peak T3 in NDT
    • Jonklaas et al, 2015 – A single overdose of 50 mcg T3 in euthyroid people
    • Saberi & Utiger, 1973 – A single dose of 50 mcg T3 in hypothyroid people
  • T3 Dose-splitting effects on FT3 / T3 levels
    • Ben-Shachar et al, 2012 – 1x a day versus 2x a day, 3x a day
    • Busnardo et al, 1980 – 3 times a day, euthyroid with suppressed TSH
    • Celi et al, 2010, 2011 – 3 times a day, hypothyroid with normal TSH
  • Blood test timing
  • Lessons learned

NORMAL Circadian rhythm, no thyroid medication

Click to expand and read this section

What kind of daily fluctuation in FT3 is natural and normal?

This is what the FT3, FT4 and TSH natural circadian rhythm looks like in 2 healthy people who are NOT taking any thyroid medication:

From Russell et al, 2008


Reference ranges:

  • TSH 0.35–3.5 mU/L | FT4 10.3–21.9 pmol/L | FT3 3.5– 6.5 pmol/L

SEE MORE: To view Russell et al’s average curves for their entire study cohort see the “circadian rhythm” PDF attachment at the bottom of Paul Robinson’s page on the Circadian T3 Method.

General insights

  • The degree of thyroid hormone rhythmicity is individualized. Averages and ranges do not describe the individual. Therefore, statistics must be used with great caution when determining what is “safe,” “normal” and “healthy” for an individual.
  • TSH has the largest peak and trough because it’s the hormone that initiates the circadian rhythm. The thyroid responds by secreting.
  • The TSH test result is highly dependent on the time of day. The person with high rhythmicity uses 2.5 mU/L (79.3%) of the population reference range every day. Their TSH is 1 and their TSH is 3.5 on the same day. However, their midpoint is 2.5.
  • Population stats depend on lab test results during office hours, 8am – 4pm. They ignore the important rise in TSH stimulation and higher FT3 that occurs at night.
  • FT4 fluctuation is far wider than FT3 fluctuation in pmol/L. The visuals are deceptive because the Y-axis scale shifts. The reference range for FT4 is about 3x as wide as the FT3 reference range.

Insights into Free T3, specifically

a ) Patient 27, high rhythmicity

  • FT3 variation is only 4.5 – 6.0, which is only 50% as wide as the reference range.
  • FT3 avoids lower 1/3 and upper 1/6 of reference while FT4 uses the full range.
  • The peak FT4 is followed 6 hours later by the lowest FT3. When FT4 is lower in evening, FT3 is simultaneously higher.
  • Lowest FT3:FT4 ratio is 0.34 pmol/L (around 18:00h, FT3 4.5 divided by FT4 13.0)

How would this person fare after gland loss, on standard thyroid therapy? Would they lose their FT3 and FT4 rhythm?

b ) Patient 34, low rhythmicity

  • FT3 variation is 4.7 – 5.5, which is only 26% as wide as the reference range.
  • Lowest FT3:FT4 ratio is 0.36 pmol/L (around 18:00h, FT3 4.7 divided by FT4 13.0)

Their TSH and FT4 are effectively defending T3 supply at a very healthy level. This is a person with a tightly controlled homeostatic set point in their HPT axis.

c ) FT3 In the population average graphs

(Not shown: see “circadian rhythm” PDF )

  • FT3 Has a clear rhythm, but it is even MORE tightly controlled around an average FT3 of 5.45 pmol/L (65% of reference).
  • Peak and trough fluctuation varies only between 5.2 – 5.6 pmol/L (only fluctuates within 13% of the reference range).
  • FT3 Does not fall below 5.2 pmol/L, which is 56% of range.
  • Lowest FT3:FT4 ratio is 0.33 pmol/L (around 15:00h, FT3 5.2 divided by FT4 15.8)

Lessons learned about Free T3 variation

  • FT3 has a rhythm that echoes TSH. The FT3 wave, they said, was 0.5 to 2.5 hours shifted to the right (later). This is because the thyroid gland shifts its T3:T4 secretion ratio, emphasizing T3 synthesis under higher TSH stimulation (Citterio et al, 2017).
  • Healthy people take an extra daily “T3 dose” after midnight. Population statistics do not account for this daily dose. How much higher would the FT3 laboratory reference range shift if it included tests taken in the evening, or at 3am?
  • Free T3 varies the least among all three indicators. The graphs above are visually deceptive because Y-axis and the reference range is 3x smaller for FT3 than for FT4. The body tightly controls around the mid-point of reference (5.0). Therefore, FT3 is actually a very reliable laboratory test.

Baseline Free T3 in health

Bianco et al, 2019 emphasizes the importance of maintaining a sufficient baseline “circulating T3” to maintain a healthy level of thyroid receptor occupancy in nuclei of cells.

There’s a higher likelihood of “tissue hypothyroidism” if FT3 falls lower into the lower half of reference and remains there on a chronic basis.  In severe nonthyroidal illness syndrome (NTIS) / “Low T3 syndrome,” people are at high risk of death or morbidity when unable to recover from illness-induced T3 loss. 

When does a thyroid patient need T3 dosing?

Click to expand and read this section

When a person cannot achieve optimal levels of Free T3 that maintain euthyroid T3 status in the bloodstream and all their tissues, they will experience chronic symptoms of hypothyroidism.

The telltale sign of a poor converter is a very high-normal or high FT4 and low-normal or low FT3 while on T4 monotherapy (a low FT3:FT4 ratio and chronically low FT3 level).

The term “poor converter” is from Midgley et al, 2015Go to their article and view the red, green and blue graphs that show poor, intermediate, and good converters on LT4 monotherapy.

  • To calculate an individual’s FT3:FT4 ratio, one must first be on no meds or T4 monotherapy because T3 dosing interferes with the calculation. Next, one must have results in pmol/L because that puts the relationship of T4 to T3 into molar units used in most articles about the ratio. Simply divide the FT3 result in pmol/L by the FT4 result in pmol/L (same blood draw, same lab, same assay manufacturer) and compare with Gullo’s figure 3, OR use the free SPINA-Thyr app for more precision.
  • A low FT3:FT4 ratio below 0.25 pmol/L is equivalent to Midgley’s threshold of “GD 23 nmol/s” according to SPINA. Being below 0.25 puts a huge deficit of FT3 per unit of FT4 that is common only in 2 conditions: 1) severe illness, and 2) dosing T4 alone, without a thyroid. In some people, severe loss of thyroid function exposes them to inefficient T4-T3 conversion beyond the thyroid (See Gullo et al, 2011, figure 3 based on 3,800+ healthy controls and 1800+ T4-treated people).

In recent research, low FT3:FT4 ratios are being associated with morbidity and mortality, especially with cardiovascular mortality:

  • Ataoğlu, H. E., Ahbab, S., Serez, M. K., Yamak, M., Kayaş, D., Canbaz, E. T., Çetin, F., Seçmeler, Ş., Şar, F., & Yenigün, M. (2018). Prognostic significance of high free T4 and low free T3 levels in non-thyroidal illness syndrome. European Journal of Internal Medicine.
  • Brozaitiene, J., Mickuviene, N., Podlipskyte, A., & Burkauskas, J. (2016). Relationship and prognostic importance of thyroid hormone and N-terminal pro-B-Type natriuretic peptide for patients after acute coronary syndromes: A longitudinal observational study. BMC Cardiovascular Disorders, 16, 45.
  • Gao, W., Guo, W., Guo, Y., Shi, M., Dong, G., Wang, G., Ge, Q., Zhu, J., & Zhou, X. (2021). Thyroid hormone concentrations in severely or critically ill patients with COVID-19. Journal of Endocrinological Investigation, 44(5), 1031–1040.
  • Hong, J., Liu, W.-Y., Hu, X., Chen, W.-W., Jiang, F.-F., Xu, Z.-R., Shen, F.-X., & Zhu, H. (2022). Free Triiodothyronine and Free Triiodothyronine to Free Thyroxine Ratio Predict All-Cause Mortality in Patients with Diabetic Foot Ulcers. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy, 15, 467–476.
  • Lang, X., Li, Y., Zhang, D., Zhang, Y., Wu, N., & Zhang, Y. (2022). FT3/FT4 ratio is correlated with all-cause mortality, cardiovascular mortality, and cardiovascular disease risk: NHANES 2007-2012. Frontiers in Endocrinology, 13.
  • Wang, C., Han, S., Li, Y., Tong, F., Li, Z., & Sun, Z. (2022). Value of FT3/FT4 Ratio in Prognosis of Patients With Heart Failure: A Propensity-Matched Study. Frontiers in Cardiovascular Medicine, 9.
  • Yuan, D., Jia, S., Zhu, P., Zhang, C., Liu, Y., Liu, R., Xu, J., Tang, X., Zhao, X., Gao, R., Yang, Y., Xu, B., Gao, Z., & Yuan, J. (2021). Usefulness of FT3 to FT4 Ratio to Predict Mortality in Euthyroid Patients With Prior Cardiovascular Events Undergoing PCI: Five-Year Findings From a Large Single-Center Cohort Study. Frontiers in Endocrinology, 12, 700349.

During illness, a person with a low FT3:FT4 ratio induced by T4 treatment is at risk of losing even more of their T3 at an intracellular level due to upregulation of deiodinase type 3 (D3 enzyme) by inflammatory cytokines and hypoxia.

If a person does not have the aid of a TSH-stimulted thyroid, will recovery from diseases take longer? Are thyroidless people at higher risk of dying if they cannot recover enough FT3 to aid recovery from their illness?

We don’t know. No research has ever focused on what happens to T4-treated patients with low FT3:FT4 ratios who subsequently become severely ill. T4-treated thyroidless patients are routinely excluded from such research studies.

The T3 dosing curve

Saravanan et al, 2007

CONTEXT: This is a study of 10 hypothyroid patients on synthetic T3/T4 combination therapy and 10 hypothyroid patients on T4 alone. They were treated autoimmune thyroid disease patients with Hashimoto’s thyroiditis or Graves’ patients after radioactive iodine therapy who were rendered hypothyroid. Judging by the Free T3 baseline around mid-reference, these were fairly good converters of T4 hormone who were already getting good value out of their T4 dosing. They likely had some functional thyroid gland tissue helping them to convert T4 into T3.

Laboratory reference ranges:

  • fT3: 2.8 – 7.1 pmol / L | fT4: 10.0 – 24.0 pmol / L, | TSH: 0.3 – 4.0 mU / L.

Why the higher TSH in T3/T4 combo therapy?

Saravanan’s T3/T4 combo used an incorrect pharmaceutical equivalence ratio that rendered some patients quite hypothyroid.

The combination therapy group had been on T3/T4 therapy for 3 months. They had been taking more than 100 mcg of T4 only. Then they had 50 mcg of their T4 dose taken away and replaced by only 10 mcg of T3 (a substitution ratio of 1:5). As a result, when the FT4 dropped by 6-8 pmol/L down to 12, close to bottom of range, the extra FT3’s temporary rise did not do enough to compensate for the FT4 loss.

  • The T3 should have replaced T4 at a ratio of at least 1:3, 1 mcg LT3 per 3 mcg LT4.
  • They needed at least 16.5 mcg of T3 (not 10 mcg) to replace their 50 mcg of T4.

Observations on circadian rhythm during thyroid therapy

For advanced readers: Click to reveal section

Natural Free T3 and Free T4 rhythmicity is lost. There’s a small bump in FT4 concentration 4 hours post T4 dose. In the graphs, we see a huge FT3 hill during the first 8 hours after T3 dose (peaking at 4 hours, when tested by Saravanan’s team every 2 hours).

The TSH rhythm is abnormal. TSH was shifted by 8 hours and was largely flattened in T4 monotherapy and influenced by T3 dosing. 

1. Lost thyroid gland function is a major variable.

Too many people think a pill can fully replace a thyroid gland. It cannot. Only living thyroid tissue can respond to TSH circadian rhythm by giving you more or less T3 and T4 over a 24 hour period.

If you want personalized T3 and T4 dosing 24/7, keep your thyroid gland as healthy as you can.

After the thyroid gland is mostly dead or gone, normal TSH will rarely be associated with truly normal FT3 results in any type of therapy.

2. Dosing of thyroid hormone now leads everything.

Even TSH is now powerfully manipulated by dosing.  Dosing will cause “turbulence” that can minimize and shift TSH rhythmicity and secretion. Thyroid hormone negative feedback to the hypothalamus & pituitary is not coming back at the time or levels these organs expect. 

Neither Saravanan’s T4 therapy nor T3/T4 therapy was doing enough to provide healthy mid-range FT3 levels most of the day. These average results are disappointing; they mean therapy failure for some patients whose results must have been below the reported average. These doctors clearly don’t know how to dose either therapy to achieve true T3-based euthyroid status.

The good news is that the timing of T3 hormone dosing can help revive a semblance of T3 circadian rhythm, if timed appropriately. Patients who lose their T3 rhythmicity often acquire dysfunctional cortisol rhythm. Natural cortisol rhythm and T3 rhythm depend on each other and reinforce each other in the early morning hours. Paul Robinson has researched the topic extensively, calling his system of T3 dosing the CT3M – Circadian T3 Method. T3 rhythm works hand in hand with cortisol rhythm. His second book The CT3M Handbook focuses on healing the adrenals through this circadian dosing method, and his blog includes a post on the Circadian Method.

Hershman, 2009

In an endocrinology textbook chapter, Jerome Hershman expressed his opinions of the various thyroid hormone preparations available on the market, steering readers away from all except synthetic T4 monotherapy.

One very useful piece of information on the T3 post-dose peak is highlighted by the pink bars in the margin of this quotation:

For those whose resolution does not permit reading the fine print, here are the relevant sections:

“2. Desiccated thyroid extract, USP … Approximately 4 to 8 hours after ingestion, T3 levels may rise to the supranormal range.”

“3. Synthetic T3 … Patients taking T3 therapy have elevated T3 concentrations for several hours after ingestion, which gradually fall to much lower levels 24 hours later.”

(Hershman, 2009)

I cite this rather biased piece of scholarship for the reason that I have yet to find a scientific source with a graph. Hershman gave no citations to research publications to back up his claims regarding the 4-8 hour post dose peak for NDT.

If the 4-8 hour time to peak is indeed the case, it confirms what some patients experience, a gentler and more delayed post-dose peak in desiccated thyroid (NDT/DTE) dosing. This may also be the reason why some thyroid patients (based on anecdotal evidence) who do not respond well to the fast FT3 peak in synthetic are able to tolerate the T3 obtained from NDT.

In theory, a T3 peak that is placed 4-8 hours post NDT dose may be partly due to the fact that its thyroid hormone molecules T4 and T3 are bound to thyroglobulin, which is a protein. The GI tract must separate the protein from the hormone molecules before they can pass into circulation. In contrast, synthetic T3 is bound to sodium.

Hershman’s attestation of the T3 peaks in these medications appears to be a very imprecise estimate, and must be taken with a grain of salt.

  • His explanation is misleading, making it seem as if, regardless of the dose strength, T3 will always surpass the top of the reference range in both medications. Of course, in light of Saravanan et al’s 24-hour study shown above, this characterization is worthy of robust ridicule. It’s brazen fearmongering. It also reveals undue obsession regarding the top of the T3 reference range considered in isolation from concurrent T4 levels.
  • In addition, there is no reason why LT3’s fall from its T3 peak should be described as “gradual” over 24 hours post dose. It has a steep curve for elimination, as shown below.

Jonklaas et al, 2015

CONTEXT: Jonklaas et al performed a pharmacokinetics study of one 50mcg OVERdose of T3 on 12 healthy people (4 females, 8 males). Most pharmaceutical studies use healthy volunteers.

Reference ranges: TSH: 0.4–4.5 mIU/L | FT3: 2.18–3.98 pg/mL, |TT3: 76–181 ng/dL


Significant results:

Triiodothyronine concentrations peaked at 2.5 hours following liothyronine administration.” This varied from patient to patient, as seen in the graph above with overlapping curves. Some people are faster absorbers or metabolizers than others.

“Heart rate increased by 5 hours after liothyronine [T3] administration, subsequently reaching a value higher than baseline.”  “No individual participant had a baseline HR of greater than 75 beats/min. Following T3 administration the highest individual HR recorded was 94 beats/min, and no significant ECG abnormalities developed. Mean HR increased by a maximum of 18 beats/min within the first 12 hours after T3 administration.” “The increased heart rate was transient and was followed by a reduction in thyroid stimulating hormone concentration.”

“Oral temperature increased from a mean of 36.2 C to a maximum of 36.6 C, and then dropped back to 35.9 C (see figure 1c). The increase in temperature was not significant.” Body temperature never reached 37.0°C (98.6°F).

Women had a slightly higher peak concentration than men (404 vs. 342 ng/dL Total T3), which may be because of their smaller total blood volume in relation to the 50mcg dose.

There were “no notable changes in blood pressure (SBP, DBP), respiratory rate, oral temperature, ECG recordings or body weight” with a single dose.


  • A single OVERDOSE of 50 mcg causes a very high peak above reference. However, the metabolic effects on the patients were very minimal (likely due to higher FT3’s ability to enhance Deiodinase Type 3 conversion of T3 into T2).
  • The peak shape is similar to Saravanan et al.
  • Peak FT3 arrives earlier, at 2.5 h post-dose.
  • Individual variation exists in the peak shape, and the time to peak.
  • Both Total T3 and Free T3 have a similar shaped peak.
  • There’s a long tail of T3 (FT3) over 72 h.
  • It takes a long time for TSH to recover after a single 50 mcg overdose.
  • Learn more about T3’s variable half-life in our article “LT3 monotherapy withdrawal: Clearance and effects

Therefore, a single dosing mistake is extremely unlikely to kill you. The body has ways of breaking down this hormone fast.  In the short term, you’ll be fine. If you’re worried, take some Milk Thistle, which blocks a dominant thyroid hormone transporter, MCT8. The short FT3 peak will pass soon.

Saberi & Utiger, 1974

Context: In Saberi & Utiger’s study, they tried to discover the equivalence of 200 mcg / day of T4 compared to [X] mcg / day of T3 dosing, solving for [X].

They estimated incorrectly. A T3:T4 equivalence ratio of 1:4 was too low.

  • 50 mcg 1x/day was not high enough a dose to remove hypothyroidism, in spite of average Total T3 levels at top of reference range and fluctuations far above. “The elevated basal serum TSH levels indicate that, in these 8 patients as a group, 50 mcg T3 was not adequate replacement.”
  • 75 mcg/ day, divided into 3 smaller doses throughout the day, was a more reasonable euthyroid dose for adults on T3 monotherapy. Saberi & Utiger referred to Chopra’s study, which proved this.
  • The test technology was older: TSH was not as sensitive, especially at the lower end of the range. There was no Free T3 test yet.


Are big T3 fluctuations like this harmful?

Click to reveal commentary

Today, thyroid scientists are fearmongering about the FT3 fluctuations during T3 therapy because they want to market a sustained release T3 pharmaceutical, or because they just want to keep people on T4 monotherapy and scare them away from trying T3.

For example, Saravanan’s abstract ends with this bogeyman:

“Our data suggests that despite chronic combined T3/T4 therapy, wide peak-to-trough variation in fT3 levels persists. Although no immediate cardiovascular effects were seen, the long-term consequences for patients on combined therapy are unknown.”

So, Saravanan, you say no cardio effects were seen! But “wide peak-to-trough variation”! Panic button! But wait a minute, we saw it’s no more variation than seen in Russell, 2008’s high-rhythmicity person! Sigh. They feel required to play mysterious organ music to generate fear of the unknown, fear of T3 dosing.

Researchers who were more experienced with T3-based therapies didn’t think FT3 far above and within reference were harmful unless the daily dose itself was too high. Saberi and Utiger reasoned about their graph above showing 50 mcg 1x a day,

  • “These fluctuations in serum T3 levels must be sufficiently attenuated at the tissue level that such a sensitive indicator of tissue thyroid hormone action as TSH secretion is constant despite the widely varying T3 concentrations which follow once daily T3 administration.”

They judged that the wide T3 fluctuations in blood could not influence tissues very much because the effect of the T3 wave on TSH was “attenuated” (reduced).

It is easy to learn if a patient can tolerate fast-release T3 dosing by giving them a very tiny dose, 2.5 to 5 mcg, whose effect would be very mild and transient. Get over this fear.

The other simple solution is … split the daily dose up. Bye-bye to big fluctuation fears.

But they have an aggravating way to dismiss this simple T3-dosing solution.

They say multi-dosing is just too inconvenient for us. Too impractical. We’re already noncompliant with 1x a day and confess to occasionally forgetting to take a dose or delaying it.  Therefore, we can’t handle multi-dosing.

What??  Are we more incompetent than diabetes patients who must test blood sugar and inject themselves with insulin? I just can’t believe it, but yes, some would withhold a vital hormone from all thyroid patients in policy, just because they imagine we are unable to manage more than 1 dose per day.

T3 Dose-splitting effects on FT3 / T3 levels

In the history of thyroid therapy, desiccated thyroid (NDT) has been routinely dosed at least 2 times a day, and T3 monotherapy about 3 times a day.

A person taking a very small dose, or a person who is a very slow metabolizer, could be fine with a 1x/day dose.

These graphs show why we split the dose in T3-based therapy: to avoid daily hypo / hyper rollercoasters that can cause daily discomfort to us, and to bring down the height of the Peak FT3, at which point T3 is triggering its own inactivation to T2 hormone secretly within our cells as we continue to resupply it by dosing.

Ben-Shachar et al, 2012

This simulation graph shows the potential peak-reducing effect of divided doses.


CONTEXT: Ben-Shachar et al’s simulations (not real dosing) were for the potential use of LT3 therapy in pediatric thyroid cancer patients (children aged 3 to 9 years old). Their aim was to benignly suppress TSH during the replacement period of 3 weeks.

NOTE: Withdrawal period? This is part of thyroid cancer therapy. After thyroidectomy for cancer, thyroid remnants are neutralized with radioactive iodine. First, TSH is suppressed to prevent cancer resurgence during recovery from surgery. Withdrawal of T3 increases TSH and induces iodine uptake into the thyroid tissue. Using T3 alone shortens time to TSH suppression and post-withdrawal time to TSH elevation. It reduces the patient’s time spent suffering in a hypothyroid state. Read more: “LT3 monotherapy withdrawal: Clearance and effects.”

Busnardo et al, 1980

After Ben-Shachar’s simulations, this graph shows REAL people with 3 doses per day, and even a table showing their doses.

CONTEXT: Busnardo’s study was also of patients receiving TSH-suppressive yet non-thyrotoxic LT3 monotherapy.


NOTES: Their goal was achieved: “The results of the present study confirm that T3, when given three times a day, at the minimum dosage necessary to inhibit TSH response to TRH, assures a constant suppression of TSH secretion without causing important side effects.” “TSH suppression is complete within 10-12 days” and withdrawal required the same period to achieve an elevated TSH.

Celi et al, 2010, 2011

CONTEXT: These researchers’ two articles are based on a single experiment that compared T4 monotherapy and T3 monotherapy to achieve a target TSH level in patients.

They discovered that a pharmaceutical substitution of 1mcg T3 for every 3 mcg T4 achieved similar TSH responses. Yes, a 1:3 substitution ratio confirms a lot of earlier science.

However, circulating T3 levels had to be significantly higher to achieve this TSH target. T3 had to fluctuate in some patients above the reference range (note the vertical error bars) and in some patients, it fell to the bottom of reference range before the morning dose.


Unfortunately, Celi’s target TSH of 1-1.5 was NOT truly euthyroid for such patients.

  1. Celi and team later published an article that admitted strong doubts (Yavuz et al, 2013): “the data suggest that pituitary euthyroidism, both assessed by basal [TSH] or TRH-stimulated TSH, does not necessarily equate to a state of generalized euthyroidism at the level of the different targets of the hormonal action.”
  2. The average T3 dose in Celi’s study was only 40 mcg / day, but most studies have proven that is insufficient to achieve body-wide euthyroid status in adults. The range for T3 monotherapy dosing is 50-100 mcg with 75 mcg being the average dose, as shown above, and as shown in a compilation of studies here.
  3. The T3 level during the L-T4 therapy experiment (not shown) was at the low borderline. However, the T3 and FT3 average is mid-range in all healthy people and rarely fluctuates so low, even during circadian rhythm, as shown in Russell, 2008 above.
  4. Three recent research studies on thyroidless patients have proven that euthyroidism is only achieved, on average, when TSH is below reference range in such patients treated on LT4 (Larisch et al, 2018; Ito et al, 2019a, 2019b).

In addition, the normalized TSH was extremely difficult to achieve on T3 monotherapy, requiring fine adjustments of compounded medication for each individual. Imagine a dose of 15.5 mcg, followed by a dose of 11.5, then 12.5.

Such a delicate balancing act on the edge of a knife would be necessary if you want to target a very narrow TSH using T3-dominant dosing. The emphasis on T3 hormone will induce an abnormally swift change of TSH from elevated to normal to suppressed as one inches closer to euthyroid status, as researchers have discovered:

  • TSH behaves abnormally at T3:T4 dosing ratios equivalent to desiccated thyroid (NDT) At a certain level of FT3, different for each patient, TSH suddenly takes a nosedive, but euthyroid status is achieved at the point of suppression (Snyder & Utiger, 1972).
  • TSH suppression is achieved by T3-dominant dosing without causing any thyrotoxicosis (as noted by Busnardo and Ben-Shachar, above).
  • Researchers have discovered the swift TSH suppression is a side effect at the hypothalamus, not the pituitary. The pituitary often still responds to TRH stimulation even after basal TSH is suppressed at 60 mcg/day. This means that TSH is often not “truly” suppressed, even when FT3 is elevated (Koutras et al, 1981).

This is why most clinicians in the history of thyroid therapy, before TSH-test obsession took over, calmly accepted the non-thyrotoxic fully suppressed TSH in T3-dominant therapy, rather than the challenging balancing act that would titrate doses down to the microgram to force the TSH to stay normal while the patient suffers.

What do you get from your heroic TSH balancing act if the T3-dosed patient isn’t necessarily euthyroid at a normalized TSH, as research confirms?

Blood test timing

Now you can see why the number of hours post dose matters.

Without consistency from test to test, you can’t compare progress between two lab tests months apart.

Why should we test after the Free T3 peak?

Measuring during peak FT3 is not going to yield consistent results.

Even 20 minutes sitting in the laboratory can make a big difference in FT3 levels during its volatile peak rise and fall. The time it takes for FT3 to rise and fall from that peak is highly individualized and can shift based on D3 conversion activity that will likely be triggered at this time.

As you can see in Saravanan’s graph above, FT3 levels can change 10% from 8 hours post-dose to 10 hours post-dose, and that’s why at least 12 hours is going to yield more consistency from lab test to lab test.

12 hours post T3 / NDT dose is a good rule of thumb

Why? because it is biased neither to the volatile peak, nor to the hypothyroid trough that would deepen after that point, especially by 24h.

Delaying or skipping doses on lab test day

The day of the lab test can be hard on the body if it involves skipping or delaying a dose more than 12 hours. Who wants to cause undue suffering by delaying their next T3-containing dose too long? If most of your therapy is in the form of T3, starving the body of T3 for too long can have side effects afterward for days or a week.

The best idea is to take your final dose in the evening before the lab test. Have a lab appointment already set up for early in the morning the next day so you can time the test appropriately.

After the blood draw, if you delayed your first dose of the day, take it after the lab test and just shift your doses forward in time to avoid taking two doses too close together.

How frequently can you do a test?

  • If all you want to check are Free T4 and/or Free T3, you can test at least 3 weeks after a stable T4 dose increase or decrease, and at least 3 days after a stable T3 dose increase or decrease, considering factors like half life and clearance rate.
  • If you want an accurate TSH test, then give TSH time to “equilibrate” to a new stable Free T4 level. It can take 6 weeks to 3 months for TSH response accuracy.

Lessons learned from the graphs

Our modern T3-T4 combination therapy trials have too often failed like Saravanan’s because they target TSH as the highest priority even though TSH is no longer in charge.

In addition, too many researchers and doctors like Saravanan also misunderstand T3 dosing dynamics and the complex FT3-FT4 relationship. These are not independent hormones. They interact in a dance in which FT3 dominates in health and must never fall below a baseline, and FT3 can compensate for less FT4, but never vice versa.

Using desiccated thyroid (NDT) or T4 monotherapy with the same classic misunderstandings will also lead to mixed results, and the most vulnerable patients will suffer.

Celi and team would have achieved euthyroid status if they had permitted the TSH to fall and allowed T3 to rise to compensate for no T4 in bloodstream. In all these graphs, the only one that used T3 alone and achieved true euthyroid status was Busnardo, 1980.

If you DO have T4 in bloodstream, the target is still at least a mid range T3 — as it is in health — or higher when needed. You will need to raise T3 to the degree you need to compensate for a reduction in T4 availability. Reducing T4 while adding T3 can be good for poor converters, because a higher FT4 can be an obstacle to achieving sufficient FT3.

Normalcy in TSH and T4 are never the targets of a fully healthy thyroid metabolism, according to the thyroid research leading the way in 2019. These two hormones are a support and a means to an end insofar as your body can still use them well enough as tools. TSH is for stimulating thyroid tissue, because a thyroid has the ability to both secrete variable ratios of T4:T3 and convert T4 to T3 at higher rates when stimulated. T4 is good for people who still convert it well despite thyroid disease. T3? well, supplying the right amount is good for everyone and every situation because it’s not a tool, it’s the end goal.

In other posts on this website, I’ve cited numerous articles to prove these basics:

  • T4 is nature’s sustained release T3, but it converts to T3 at a variable rate.
  • When T4 is more abundant, it converts poorly and can trigger both T4 and T3 depletion.
  • When T3 is more abundant, it inactivates to T2 at a higher rate.
  • When neither are sufficient, or T3 alone is insufficient for individual health, you’re hypothyroid.
  • When both are overabundant, or T3 alone is in excess of individual health requirements, you’re hyper.

Without a TSH-driven healthy thyroid, your body can’t compensate naturally for thyroid metabolic dysfunction and imbalance to sustain T3 supply to cells.

T3 hormone sufficiency in blood and cells is always the target and the foundation of health:

  • with or without a healthy thyroid
  • with or without TSH
  • and with or without T4.


Click to view references

Ben-Shachar, R., Eisenberg, M., Huang, S. A., & DiStefano, J. J. (2012). Simulation of Post-Thyroidectomy Treatment Alternatives for Triiodothyronine or Thyroxine Replacement in Pediatric Thyroid Cancer Patients. Thyroid, 22(6), 595–603.

Busnardo, B., Girelli, M. E., Bui, F., Zanatta, G. P., & Cimitan, M. (1980). Twenty-four hour variations of triiodothyronine (T3) levels in patients who had thyroid ablation for thyroid cancer, receiving T3 as suppressive treatment. Journal of Endocrinological Investigation, 3(4), 353–356.

Busnardo, B., Bui, F., & Girelli, M. E. (1983). Different rates of thyrotropin suppression after total body scan in patients with thyroid cancer: Effect of an optimal saturation regimen with thyroxine or triiodothyronine. Journal of Endocrinological Investigation, 6(6), 455–461.

Celi, F. S., Zemskova, M., Linderman, J. D., Babar, N. I., Skarulis, M. C., Csako, G., … Pucino, F. (2010). The pharmacodynamic equivalence of levothyroxine and liothyronine. A randomized, double blind, cross-over study in thyroidectomized patients. Clinical Endocrinology, 72(5), 709–715.

Celi, F. S., Zemskova, M., Linderman, J. D., Smith, S., Drinkard, B., Sachdev, V., … Pucino, F. (2011). Metabolic effects of liothyronine therapy in hypothyroidism: A randomized, double-blind, crossover trial of liothyronine versus levothyroxine. The Journal of Clinical Endocrinology and Metabolism, 96(11), 3466–3474.

Citterio, C. E., Veluswamy, B., Morgan, S. J., Galton, V. A., Banga, J. P., Atkins, S., … Arvan, P. (2017). De novo triiodothyronine formation from thyrocytes activated by thyroid-stimulating hormone. The Journal of Biological Chemistry, 292(37), 15434–15444.

Jonklaas, J., Burman, K. D., Wang, H., & Latham, K. R. (2015). Single Dose T3 Administration: Kinetics and Effects on Biochemical and Physiologic Parameters. Therapeutic Drug Monitoring, 37(1), 110–118.

Koutras, D. A., Malamitsi, J., Souvatzoglou, A., Piperingos, G. D., Livadas, D. P., Boukis, M., … Moulopoulos, S. D. (1981). Relative ineffectiveness of exogenous triiodothyronine as a thyroid suppressive agent. Journal of Endocrinological Investigation, 4(3), 343–347.

Russell, W., Harrison, R. F., Smith, N., Darzy, K., Shalet, S., Weetman, A. P., & Ross, R. J. (2008). Free Triiodothyronine Has a Distinct Circadian Rhythm That Is Delayed but Parallels Thyrotropin Levels. The Journal of Clinical Endocrinology & Metabolism, 93(6), 2300–2306.

Saberi, M., & Utiger, R. D. (1974). Serum Thyroid Hormone and Thyrotropin concentrations during thyroxine and triiodothyronine therapy. The Journal of Clinical Endocrinology & Metabolism, 39(5), 923–927.

Saravanan, P., Siddique, H., Simmons, D. J., Greenwood, R., & Dayan, C. M. (2007). Twenty-four hour hormone profiles of TSH, Free T3 and free T4 in hypothyroid patients on combined T3/T4 therapy. Experimental and Clinical Endocrinology & Diabetes: Official Journal, German Society of Endocrinology [and] German Diabetes Association, 115(4), 261–267.

Snyder, P. J., & Utiger, R. D. (1972). Inhibition of thyrotropin response to thyrotropin-releasing hormone by small quantities of thyroid hormones. Journal of Clinical Investigation, 51(8), 2077–2084. Retrieved from

Yavuz, S., Linderman, J. D., Smith, S., Zhao, X., Pucino, F., & Celi, F. S. (2013). The Dynamic Pituitary Response to Escalating-Dose TRH Stimulation Test in Hypothyroid Patients Treated With Liothyronine or Levothyroxine Replacement Therapy. The Journal of Clinical Endocrinology & Metabolism, 98(5), E862–E866.

19 thoughts on “Free T3 peaks and valleys in T3 and NDT therapy

  1. You mentioned articles by Larisch, et al. I thought you might be interested in a paper i co-authored with one of Larisch’s co-authors, Dr. Rudolf Hoermann, a retired endocrinologist. The paper is written to help hypothyroid patients understand why they are not being diagnosed or treated adequately, with supporting scientific evidence that encompasses the Total Thyroid Process. This is a link.

    1. Dear Mel Rowe, thanks so much for commenting here and leaving the link.

      Yes I came across that article a few years ago in June 2018 and I see it in my database and my folder. I read it through and did a lot of highlighting on it. I had many insights while reading it. It helped me to understand some of the things that were a little unclear in Hoermann and team’s scientific articles due to their extremely complex sentence grammar and compact reasoning.

      It sounds like you and I have a lot in common in terms of a passion for reading thyroid research — Let’s be in contact. Please contact us at our organization’s email address — though I warn you I’m very slow at responding via email.

      1. I am glad you found some points of interest in our previous paper. Afterward we decided to expand our efforts to include the total thyroid process, but in a condensed version that could be more readily used by patients with their doctors. We included all the steps in the total process, each with their numerous confounding variables. Statistically, all these variables clearly preclude TSH from being a valid indicator of thyroid status, except at extreme levels indicating overt primary hypothyroidism. We also make clear why TSH should not be used to determine med dosage levels for thyroid patients under treatment. I think you will find this new paper even more revealing.

  2. Thank you so much for this text! Best thing I have read on this topic. I finally understand my own issues now!!

    I am an ICU nurse, hypothyroid. Tried t4/levothyroxine for half a year, felt horrific. More hypothyroid, despite perfect labs. Then tried Armour Thyroid. Tolerated Armour sooo much better, no problems at all with pulse and BP (suffered through that with levothyroxine).

    But, I was dizzy and felt all hypo starting appr one hour after each dose.

    Now I have tried the doses mentioned here and take NDT at 10:30 am, 3 pm and 8 pm. Also a small dose at midnight.

    MY GOD what a difference. I feel so good! This knowledge has given me my life back!!!!!

    Still struggle to accept the fact it can make such a difference. I know I am not low on cortisol …. but yeah, it has been a life saver.

    I tried everything before this, checked ferritin, b12, nothing helped.

    If anyone is reading this and feeling bad, please try and adjust the timing on your NDT doses!!

Leave a public reply here, on our website.

This site uses Akismet to reduce spam. Learn how your comment data is processed.