RightRep EN · PT Open the app
BLOG

Can You Train With Hypothyroidism? The Cold-Engine Barbell Audit: Where Thyroid Hormone Doesn't Build the Muscle But Signs the Building Permit ('Instead of Directly Stimulating Muscle Growth, They Create the Intracellular Environment Required for Effective Adaptation to Exercise'), the Prospective Electrodiagnostic Cohort Put Neuromuscular Complaints at 79% and Clinical Weakness at 38% While Creatine Kinase Correlated With Nothing, the 38-Study Tolerance Ledger Says Exercise Intolerance Is Multifactorial and 'Not Always Reversible' Even at Normalized TSH, Six Months of Replacement Made the Same Submaximal Stage Cost Less Oxygen, Ventilation and Heartbeat, 16 Weeks of Bike and Treadmill Repaid a 12-Point Functional-Capacity Deficit, the Heart-Rate Spool-Up Fell From 39.6 to 28.9 Seconds While the Control Arm Drifted, the 4-Arm On-Levothyroxine RCT Had Every Mode Paying and Combined Winning TSH and Mental Health, the Meta-Analysis Kept TSH Flat While the Symptom Columns Moved Anyway, Coffee and Calcium Steal the Pill Dose, and the Scale Reads Myxedema Water — Fix the Hormone First, Then Load the Bar

By RightRep AI Assistant · · 22 min read

Hypothyroidism is the condition this library's treatment twin — does exercise help hypothyroidism — answers from the medicine side: training barely moves the TSH column but reliably moves the fatigue, function and quality-of-life columns the pill leaves behind. This file answers the barbell side: what happens to a loaded training program when the hormone that sets the muscle's metabolic baseline is missing or marginal — the commonest thyroid dysfunction, its risk decided by iodine nutrition and ageing, belonging overwhelmingly to women, with a subclinical pool sitting under the overt one [1][2]. The default scripts fail in opposite directions. The wait-unil-it-passes script treats the fatigue as a reason to park the body — while the exercise files show the trained subclinical patient improves the columns the diagnosis drains [6][7][8]. The push-through script trains like the bloodwork doesn't exist — while the physiology says an untreated or under-treated thyroid runs the engine cold, and the weakness that comes with it is part of the disease, not a discipline failure [4][5]. The reframe that carries the whole program is one sentence from the newest file on the shelf: thyroid hormones do not directly stimulate muscle growth — they create the intracellular environment required for effective adaptation to exercise [3]. The hormone signs the building permit; the barbell still has to do the building. Which puts the fences in order. First: the diagnosis and the dose belong to a physician — levothyroxine is the standard treatment and it is prescribed, monitored and tuned by the prescriber [2]; training changes what the prescription needs, never edits it. Second: new or changed symptoms get priced by a doctor before they get programmed — chest pain, syncope, a fatigue that drops off a cliff, a pregnancy, a body weight that has changed materially — all of these refer before barbell, the same rule every chronic-condition file in this library carries (the type 2 diabetes audit and the fibromyalgia audit stage it the same way). Third: don't let the supplement shelf eat the dose — levothyroxine is a microgram-dosed, narrow-therapeutic-index drug whose absorption coffee, calcium and iron measurably intercept [11]; the timing rules are part of the program, not pharmacy trivia. With the fences standing, the file underneath is unusually complete: the gains-cap question has a mechanistic answer [3], the engine question has a 38-study ledger behind it [4], the weakness question has a prospective electrodiagnostic cohort [5], the replacement question has three trials [6][7][8], the mode split has a four-arm randomized answer [9] and a meta-analysis [10], and the scale question — the one that torments this diagnosis more than any other — has a one-line verdict: the weight that comes and goes with the thyroid is mostly water, not fat [12].

The short answer: yes — and the pill comes first, the barbell second

Strip the file to its spine. The permission is structural, not heroic: hypothyroidism is a common, readily diagnosed and managed condition, levothyroxine replacement is the standard treatment, and nothing in its symptom list — fatigue, lethargy, cold intolerance, weight gain, constipation [2] — contraindicates loading; the exercise literature in these patients reads "seemed to be safe" across ten pooled studies [10]. The engine runs cold until the hormone is fixed, and fixing it pays measurable performance dividends: in the Rio randomized trial, six months of TSH normalization on levothyroxine made the same submaximal treadmill stage cost less oxygen (24.1→17.1 ml/kg/min), less ventilation (28.0→23.5 l/min) and fewer heartbeats (128→121 bpm, all p=0.03) — while the untreated arm changed nothing [6]. Training then pays on the columns the pill leaves behind: sixteen weeks of bike-and-treadmill repaid a twelve-point functional-capacity deficit on the SF-36 in subclinical women while the sedentary arm stayed flat [7]; twelve weeks of endurance sped the heart-rate spool-up from 39.6 to 28.9 seconds against a control that drifted the wrong way (p=0.001) [8]. And in women already on levothyroxine, every training mode paid: aerobic, resistance and combined training all improved thyroid function markers, lipids, exercise capacity and quality of life against controls over 12 weeks — with the combined aerobic-plus-resistance arm winning TSH and the mental-health column outright [9]. That is a permission file with an order of operations under it: hormone first, load second, both ongoing — and the honest tail riding along: a substantial proportion of patients who reach biochemical treatment targets keep persistent complaints anyway [2], which the exercise files answer the only way they can — with the training log as the instrument panel that separates "the engine is still cold" from "the program isn't working."

The reframe: the hormone is the building permit, not the construction crew

Everything operational in this file hangs on one mechanistic sentence from the 2026 review that finally priced thyroid hormone for the weight room: thyroid hormones are "permissive regulators of skeletal muscle protein accretion and oxidative capacity" — "instead of directly stimulating muscle growth, they create the intracellular environment required for effective adaptation to exercise" [3]. Adequate thyroid-hormone availability supports mitochondrial integrity, ribosome biogenesis and coordinated protein turnover — the translation layer that converts anabolic signaling into structural adaptation [3]. Run the two directions. Low direction (this diagnosis): reduced translational efficiency during hypothyroidism may blunt hypertrophic responses [3] — the barbell still builds, but the construction site is short on permits and materials, so the same training returns a slower yield. High direction (not this file, but the mirror): hyperthyroidism accelerates proteolytic pathways and compromises net protein gain despite elevated turnover — more hormone does not mean more muscle either [3]. Two operational consequences follow. Consequence one: the hypertrophy question has an answer, and it is "slower, not never" — resistance training, "when sufficient biosynthetic capacity is present, can stimulate myofibrillar accretion" [3]: the capacity condition is exactly why the replacement file comes first [6] and why training through an untreated deficiency trains with the brake on. Consequence two: energy availability is a thyroid variable — reduced energy availability lowers T3 concentrations and may attenuate adaptive efficiency across both resistance and endurance modalities [3], which means the aggressive crash-cut that already costs a healthy lifter muscle costs this lifter hormone on top — the anti-starvation argument (does starvation mode exist owns the general audit) hits this diagnosis harder, and the protein floor is part of the thyroid program, not just the muscle program.

The engine file: why the tank feels empty

The complaint that brings this reader to the gym door — "I have no engine" — has a whole systematic review behind it, and it deserves to be read honestly. Thirty-eight studies covering 1,379 patients establish that exercise intolerance in untreated hypothyroidism is multifactorial: cardiovascular, cardiopulmonary, musculoskeletal, neuromuscular and cellular metabolic limitations acting in concert [4] — not one broken part, a whole system idling under its set-point. The same ledger carries the honest tail: "exercise intolerance in patients is not always reversible during adequate hormone replacement therapy" — despite a defined euthyroid status on paper, a significant group of treated patients keep exercise limitations that dent daily activities, sport and quality of life, and the review closes by calling this a "forgotten area" with only two training studies at the time [4]. That sentence pairs with the clinical spine's own admission — a substantial proportion of patients who reach biochemical treatment targets have persistent complaints [2] — and together they set the expectation this file programs around: the bloodwork normalizes faster than the body [2][4]. What the files also say is that this gap is exactly where training pays: the quality-of-life columns [7][9], the functional-capacity column [7], the spool-up kinetics [8] and the submaximal economy [6] all moved under training or replacement without the TSH pill needing to move further — and the meta-analysis across ten studies found aerobic and resistance training improved physical- and mental-health outcomes while TSH itself sat statistically flat [10]. Translation for the gym floor: the tank reading empty is a measured phenotype [4], not a character judgment, and it is trainable — but you train it like you warm up a cold engine, not like you floor a rental.

The weakness file: aches, strength and the CK red herring

The muscle question under this diagnosis — "is my muscle damaged? should it even be loaded?" — has a prospective cohort behind it, and its numbers price both the problem and the reassurance. In newly diagnosed hypothyroid patients evaluated clinically and electrodiagnostically: 79% carried neuromuscular complaints, 38% showed clinical weakness in at least one muscle group, 42% had signs of a sensorimotor axonal neuropathy, and 29% had carpal tunnel syndrome [5]. The muscle aches, the grip that fades, the tingling hands — all part of the diagnosis's presentation, all things a lifter will otherwise misread as overtraining or aging. Two sentences in that cohort do disproportionate work for this file. First: "serum creatine kinase did not correlate with weakness" [5] — the lab value that spikes in muscle-damage conversations is a red herring for the weakness question here; you cannot read your muscles off your CK in this diagnosis. Second: "weakness in hypothyroidism is more difficult to treat, suggesting myopathy" [5] — after a full year of treatment, 13% of the hypothyroid patients still had measurable weakness, in pointed contrast to the hyperthyroid weakness in the same cohort, which resolved completely within an average of 3.6 months [5]. The hypothyroid weakness lives deeper in the muscle. That is an argument for loading — the myopathy of under-use compounds the myopathy of the diagnosis, and progressive resistance is the specific tool for weakness [3][9] — and simultaneously an argument for patience with the clock: strength returns on the thyroid's schedule [5][6], the progressive overload staircase is climbed from a lower rung, and the program that wins is the one still running in month twelve, not the one that peaked in week three.

What the pill actually restores — and what it doesn't

The replacement file is the Brazilian chain, and it reads like a three-act structure. Act one — the economy run (Mainenti, Rio): twenty-three subclinical patients randomized to treated versus untreated; six months of TSH normalization on levothyroxine, and at the same submaximal exercise stage the treated group's oxygen uptake, minute ventilation and heart rate all fell significantly (all p=0.03) while the untreated arm changed nothing — the identical workload got cheaper to run [6]. That is what "fixed the hormone" means operationally: not a lightning bolt, a fuel-efficiency repair. Act two — the quality-of-life run (Werneck): women with subclinical hypothyroidism started twelve points below euthyroid controls on SF-36 functional capacity (77.0 vs 88.8, p=0.020); sixteen weeks of supervised bike-and-treadmill, three hours a week, moved functional capacity, general health, emotional aspects and both HRQoL components while the sedentary control arm moved nothing [7]. Act three — the spool-up run (Almas): eighteen women randomized; twelve weeks of endurance training and the heart-rate on-kinetics — the mean response time to 63% of steady-state HR, the literal warm-up lag this diagnosis is famous for — fell from 39.6 to 28.9 seconds, while the control group drifted from 53.6 to 55.0 (p=0.001) [8]. Read together: the pill restores the engine's economy [6], training restores the columns the pill leaves short [7][10], and the warm-up lag itself is trainable even with the thyroid still marginal [8] — which is why the warm-up section of this program is longer than a healthy lifter's, on purpose. And the restore-it-doesn't column stays honestly open: the 38-study ledger's persistent-intolerance tail [4] and the clinical review's persistent-complaints admission [2] both say some engines keep a cold streak no matter what the labs say — for that reader, the program's job becomes getting the most out of the engine that exists, and the instrument panel (the training log) is what keeps that effort honest instead of aspirational.

The mode split: what the iron pays when the thyroid is managed

The four-arm randomized trial is the mode-split file, and it was run on exactly the population this file serves: sixty women aged 35–45 with clinical hypothyroidism, all already on levothyroxine, randomized to aerobic, resistance, combined aerobic-plus-resistance, or control for twelve weeks at three days a week of low-to-moderate work [9]. Every exercise group improved every outcome — thyroid function markers, lipid profile, exercise capacity and quality of life — against baseline and against the controls (p<0.05) [9]. Then the arms separated in a way a program designer can use directly: the combined arm won TSH and the SF-12 mental component outright (p<0.05 vs the single-mode arms), the aerobic arm won estimated VO2max (aerobic > combined > resistance), and free T4, lipids and the SF-12 physical component tied across exercise arms [9]. The meta-analysis supplies the honesty under it: ten studies, low-to-moderate quality, mostly subclinical women — TSH under exercise training sat statistically flat (Hedges' g −0.96, 95% CI −2.71 to 0.79, p=0.160), FT3 and FT4 likewise, "exercise does not have a significant impact on thyroid function, although its practice can lead to secondary outcomes related to physical and mental health" [10]. So the mode split for this diagnosis: the barbell owns strength, the myopathy counter-file [5] and the hypertrophy question [3]; the treadmill owns VO2max [9] and the spool-up [8]; running both is the arm that won the most columns [9] — and neither mode is thyroid medicine [10]. The cardio-kills-gains fear has its own file and its own acquittal; here it dies faster, because the combined mode is literally the winning arm of the only four-arm trial on these shelves [9]. Two to three lifting days inside a sustainable set budget and training frequency, aerobic work stacked alongside [6][7][9] — the machines-vs-free-weights call stays a normal call, with machines earning their keep on the days the fatigue makes free-weight control expensive.

The supplement shelf that eats your dose

This is the section no general training guide carries, and for this diagnosis it is worth its own heading. Levothyroxine is dosed in micrograms with a narrow therapeutic index, which makes absorption-phase interactions clinically significant — the review's own framing [11]. The named interceptors read like the inventory of a gym bag: coffee and soy have the greatest impact in reducing absorption; calcium carbonate, calcium citrate, calcium acetate, iron sulfate, aluminum hydroxide and proton-pump inhibitors are established interceptors; vitamin C increases absorption — and the whole problem has a one-move solution: separating the administration of the pill from the interfering substance [11]. Translated to the lifter's morning: the levothyroxine goes in first, on an empty stomach, with water; the pre-workout coffee waits; the calcium or iron tablet from the creatine-and-friends shelf waits longer — hours, not minutes [11] — and the time-of-day question for the session itself is free to be answered by schedule, not by the pill clock. Why this belongs in a training file rather than a pharmacy pamphlet: the lifter's supplement stack is exactly the collection most likely to collide with a morning levothyroxine habit, the collision is silent (you feel nothing the day it happens — the TSH drift is what tells the story months later [11]), and the reader who fixes it protects the entire order of operations this file runs on — because every downstream claim, from the submaximal economy [6] to the hypertrophy permit [3], assumes the dose on the label is the dose in the blood. One more name from the same review, for completeness: gastrointestinal conditions — celiac disease, atrophic gastritis, lactose intolerance, Helicobacter pylori — can impede absorption too, which is the prescriber's lane to check when the dose stops behaving [11].

The scale file: myxedema water is not fat

No condition in this library tortures the scale like this one, and the file is short and strangely liberating. Small differences in thyroid function associate with up to about 5 kg of body-weight difference [12] — but the weight that hypothyroidism adds and treatment subtracts is, in the overt case, "caused by excretion of water bound in tissues (myxoedema)" [12]: the diagnosis moves water first and fat a distant second. Three consequences run the program. Consequence one: the scale is a lagging, misleading gauge in this diagnosis — the 3 kg that dropped in the first month of treatment is not a fat-loss victory, and the 3 kg that climbed before it was never a discipline failure; meter the program on loads, reps and sessions (the training log again — it is the only gauge on this car the water can't move). Consequence two: the mirror complaint often predates the thyroid — obese people with normal thyroid glands tend to run higher TSH through axis activation, not gland failure [12], so a high TSH on an obese bloodwork panel is a conversation with an endocrinologist, not a self-diagnosis; the twin file owns that audit in full (does exercise help hypothyroidism). Consequence three: the actual fat-loss program is the boring one, run gently — a modest deficit that does not crash energy availability (which lowers T3 and attenuates adaptation [3]), the protein floor held, the mode split doing both strength and aerobic work [9], and the cardio-or-weights-for-fat-loss question answered "both, at the doses the trials ran" — with sleep as the unpaid column it always is. The reader who dials a starvation cut because "the thyroid broke my metabolism" pays for it twice: muscle [3], and hormone [3] — the why-am-I-not-losing-weight audit owns the general file; this diagnosis adds the water caveat and the T3 warning on top.

The Brazil anchors: who the pt-BR twin serves

The pt-BR lane's reader is served by her own country's research base better than almost any condition on these shelves — one Brazilian chain (the UFRJ thyroid clinic with the Minas exercise-physiology groups) ran the replacement question three times. Mainenti's Rio trial randomized subclinical patients to treated versus untreated and showed the submaximal economy repair six months of TSH normalization buys [6]. Werneck's SF-36 trial priced the functional-capacity deficit at twelve points and repaid it with sixteen weeks of supervised aerobic work [7]. Almas's kinetics file took the warm-up lag this diagnosis is famous for and trained it from 39.6 to 28.9 seconds [8]. "Hipotireoidismo não pode fazer musculação" — the domestic default this file prices — dies against the four-arm on-levo trial where the resistance arm improved every column it touched [9] and the permissive-regulator file that says the loaded program builds as soon as the biosynthetic capacity is present [3]; the push-through half — training through an untreated deficiency because "it's just laziness" — dies against the 38-study ledger that measured exactly how multifactorial and real the cold engine is [4]. The pt-BR twin exists to replace both halves with the order of operations: o remédio primeiro, a barra depois, os dois para sempre — and the pill-clock rule that keeps the supplement shelf from silently eating the dose [11].

So what do you actually do? The practical file

Six rules, every one sourced from the files above:

  1. Fix the hormone before you chase the PR. Replacement is the standard treatment [2], and six months of normalized TSH measurably cheapened every submaximal stage [6] — an untreated deficiency trains with the brake on [3][4]; the dose and its tuning belong to the prescriber, always.
  2. Treat the engine as cold on every start. Longer warm-ups (the spool-up lag is real [8] and trainable [8]), first working sets conservative, and progression governed by the log — loads, reps, sessions — never by how the day feels [4].
  3. Keep the pill clock and the gym bag apart. Levothyroxine first thing, empty stomach, water; coffee, calcium, iron and antacids separated by hours [11]; the pre-workout question is answered around the pill, not over it.
  4. Meter the body on performance, not on the scale. The thyroid's weight is mostly water moving in and out of tissues [12]; fat loss runs on the gentle deficit that protects T3 [3], the protein floor, and both training modes [9] — the scale is the last gauge to consult, not the first.
  5. Run both modes; let each own its column. Combined aerobic-plus-resistance won the most columns including TSH and mental health [9]; aerobic owns VO2max [9] and the spool-up [8]; the iron owns strength and the myopathy counter-file [3][5][9] — and none of it is thyroid medicine [10], it is symptom medicine, which is exactly the column set this diagnosis needs moved [2][7].
  6. Program for slower, not zero — and keep the program running. Hypothyroidism may blunt hypertrophic responses [3]; the weakness can take a year and leave residue [5]; a normalized TSH does not always mean a normalized engine [2][4] — so deload on schedule, autoregulate the hard days (reps in reserve instead of forced failure), and judge the program on the twelve-month trend, because every positive file on these shelves ran twelve weeks minimum [7][8][9].

How RightRep fits into this

The cold-engine audit is an instrumentation question — the body's own gauges (felt energy, felt recovery, the scale) are the least reliable sensors in this diagnosis [4][12] — and the hard half is exactly what a counting camera does:

  • The instrument panel. Loads, reps, sessions — counted, not felt — are the meters the fatigue and the water can't corrupt [4][12]; every session logged is a dose-response data point for the next prescriber visit.
  • Form under a fatigued engine. When the tank reads empty mid-set, technique discipline has to come from outside the body — camera-scored form keeps the staircase safe without asking the cold engine to self-grade.
  • Progression on evidence. The overload decision fires on logged performance, not on felt readiness — the difference between adapting [3] and riding fatigue in circles [4].
  • The streak. Every quality-of-life file on these shelves ran twelve to sixteen weeks minimum [7][8][9]; in a diagnosis whose gains arrive on the thyroid's clock [5][6], a visible streak is the cheapest continuation device on the market.

The bottom line

  • The permission file: common condition, standard levothyroxine treatment, exercise safe across ten pooled studies — with the order of operations: hormone first, load second [1][2][10].
  • The reframe: thyroid hormones are permissive regulators — they create the intracellular environment required for adaptation; hypothyroidism may blunt hypertrophic responses; resistance training pays "when sufficient biosynthetic capacity is present" [3].
  • The engine file: 38 studies, 1,379 patients — exercise intolerance is multifactorial and "not always reversible" even at normalized TSH; a "forgotten area" [4].
  • The weakness file: 79% neuromuscular complaints, 38% clinical weakness, 42% axonal neuropathy, 29% carpal tunnel; CK correlated with nothing; 13% still weak at one year — myopathy, harder to treat [5].
  • The replacement file (Brazil): six months of normalized TSH cut the submaximal stage's oxygen, ventilation and heart-rate cost (all p=0.03) [6]; 16 weeks of aerobic work repaid a 12-point SF-36 functional-capacity deficit [7]; 12 weeks sped the HR spool-up 39.6→28.9 s vs a drifting control (p=0.001) [8].
  • The mode split: 60 women on levothyroxine, 4 arms, 12 weeks — every mode improved everything vs controls; combined won TSH and mental health; aerobic won VO2max [9].
  • The honesty column: TSH under training sat flat (g −0.96, CI −2.71 to 0.79, p=0.160) while physical- and mental-health outcomes moved — "exercise does not have a significant impact on thyroid function" [10]; and the pill itself doesn't fully close the complaint file either [2][4].
  • The dose trap: coffee and soy cut absorption most; calcium, iron and PPIs intercept; vitamin C helps; separate the doses — micrograms make it matter [11].
  • The scale file: the thyroid's weight is mostly myxedema water [12]; small thyroid-function differences ride up to ~5 kg [12]; the fat-loss program is the gentle one that protects T3 [3].

Fix the hormone, load the bar, meter the engine on instruments — and let the camera count the reps the fatigue is trying to blur at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Taylor PN, Albrecht D, Scholz A, et al. Global epidemiology of hyperthyroidism and hypothyroidism. Nature Reviews Endocrinology, 2018;14(5):301–316. pubmed.ncbi.nlm.nih.gov
  2. Chaker L, Bianco AC, Jonklaas J, Peeters RP. Hypothyroidism. The Lancet, 2017;390(10101):1550–1562. pubmed.ncbi.nlm.nih.gov
  3. Brenmoehl J, Walz C, Hoeflich A. Thyroid hormones as permissive regulators of skeletal muscle protein accretion and oxidative capacity: Implications for resistance and endurance training. Best Practice & Research Clinical Endocrinology & Metabolism, 2026;40(3):102110. pubmed.ncbi.nlm.nih.gov
  4. Lankhaar JA, de Vries WR, Jansen JA, Zelissen PM, Backx FJ. Impact of overt and subclinical hypothyroidism on exercise tolerance: a systematic review. Research Quarterly for Exercise and Sport, 2014;85(3):365–389. pubmed.ncbi.nlm.nih.gov
  5. Duyff RF, Van den Bosch J, Laman DM, van Loon BJ, Linssen WH. Neuromuscular findings in thyroid dysfunction: a prospective clinical and electrodiagnostic study. Journal of Neurology, Neurosurgery & Psychiatry, 2000;68(6):750–755. pubmed.ncbi.nlm.nih.gov
  6. Mainenti MR, Vigário PS, Teixeira PF, Maia MD, Oliveira FP, Vaisman M. Effect of levothyroxine replacement on exercise performance in subclinical hypothyroidism. Journal of Endocrinological Investigation, 2009;32(5):470–473. pubmed.ncbi.nlm.nih.gov
  7. Werneck FZ, Coelho EF, Almas SP, et al. Exercise training improves quality of life in women with subclinical hypothyroidism: a randomized clinical trial. Archives of Endocrinology and Metabolism, 2018;62(5):530–536. pubmed.ncbi.nlm.nih.gov
  8. Almas SP, Werneck FZ, Coelho EF, Teixeira PFS, Vaisman M. Endurance training improves heart rate on-kinetics in women with subclinical hypothyroidism: a preliminary study. Journal of Endocrinological Investigation, 2023;46(1):51–57. pubmed.ncbi.nlm.nih.gov
  9. Ahmad AM, Serry ZH, Abd Elghaffar HA, Ghazi HA, El Gayar SL. Effects of aerobic, resistance, and combined training on thyroid function and quality of life in hypothyroidism. A randomized controlled trial. Complementary Therapies in Clinical Practice, 2023;53:101795. pubmed.ncbi.nlm.nih.gov
  10. Duñabeitia I, González-Devesa D, Varela-Martínez S, Diz-Gómez JC, Ayán-Pérez C. Effect of physical exercise in people with hypothyroidism: systematic review and meta-analysis. Scandinavian Journal of Clinical and Laboratory Investigation, 2023;83(8):523–532. pubmed.ncbi.nlm.nih.gov
  11. Skelin M, Lucijanić T, Amidžić Klarić D, et al. Factors Affecting Gastrointestinal Absorption of Levothyroxine: A Review. Clinical Therapeutics, 2017;39(2):378–403. pubmed.ncbi.nlm.nih.gov
  12. Laurberg P, Knudsen N, Andersen S, Carlé A, Pedersen IB, Karmisholt J. Thyroid function and obesity. European Thyroid Journal, 2012;1(3):159–167. pubmed.ncbi.nlm.nih.gov