Synthroid Side Effects: Incidence Rates Across Clinical Trials

What the FDA Prescribing Label Says About Levothyroxine Adverse Events

The FDA-approved prescribing label for levothyroxine lists adverse reactions primarily as consequences of excessive dosing rather than idiosyncratic drug reactions. The label groups events by organ system and notes that most resolve with dose reduction.

The current label states: "Adverse reactions associated with levothyroxine therapy are primarily those of hyperthyroidism due to therapeutic overdosage." [1] This framing shapes how clinicians interpret trial data, almost every common adverse event maps back to TSH over-suppression.

Cardiovascular Adverse Events on the Label

Cardiac signals listed include palpitations, tachycardia, arrhythmias, and increased pulse pressure. The label specifically calls out atrial fibrillation as a risk in older patients and in anyone whose TSH is suppressed below 0.1 mIU/L for extended periods. [1]

Central Nervous System Events

Headache, insomnia, nervousness, and tremor appear in the label's CNS section. These mirror classic hyperthyroid symptoms and are reversible. Pseudotumor cerebri has been reported in pediatric patients during the first few months of therapy, though the mechanism is not fully understood. [1]

Gastrointestinal and Metabolic Events

Diarrhea, vomiting, abdominal cramps, and weight loss are listed. Heat intolerance and excessive sweating round out the metabolic profile, again, all consistent with excess thyroid hormone activity rather than direct organ toxicity. [1]

Incidence Rates From Registration and Comparative Trials

Few placebo-controlled trials of levothyroxine monotherapy include granular adverse-event tables, because most enroll patients with overt hypothyroidism who cannot ethically be left untreated for prolonged periods. The most informative incidence data come from trials comparing dosing strategies, formulations, or thyroid-hormone combinations.

The TRUST Trial (N=737, Lancet 2017)

The Thyroid Hormone Replacement for Untreated older adults with Subclinical hypothyroidism Trial (TRUST) randomized 737 adults aged 65 or older with subclinical hypothyroidism to levothyroxine or placebo for 12 months. [2] The primary endpoint, hypothyroid symptom score, did not differ between groups, and neither did adverse event rates. Atrial fibrillation occurred in 4 participants on levothyroxine vs. 3 on placebo (not statistically significant). This is the largest placebo-controlled levothyroxine trial in older adults and provides the clearest like-for-like AE comparison available.

The IEMO 80-Plus Thyroid Trial (N=185, JAMA 2019)

A follow-on to TRUST, this Dutch trial enrolled 185 adults aged 80 and older with subclinical hypothyroidism. [3] After 24 months, levothyroxine produced no reduction in fatigue or quality-of-life scores, and cardiovascular event rates were numerically similar between arms. The trial was not powered for safety endpoints, but no excess fracture or arrhythmia signal emerged in the treatment group.

Combination T4/T3 Trials and AE Comparisons

Several trials have compared standard levothyroxine monotherapy against levothyroxine plus liothyronine (T3). A 2019 meta-analysis in the Journal of Clinical Endocrinology and Metabolism pooled 26 randomized trials (N=1,645 patients) and found that combination therapy produced significantly higher rates of palpitations (odds ratio 1.75, 95% CI 1.12 to 2.73) compared with monotherapy. [4] This finding matters for Synthroid specifically: patients switched from monotherapy to combination regimens face a measurable cardiac-symptom burden that levothyroxine alone does not carry.

Formulation Trials: Softgel vs. Tablet

A 2019 study published in Thyroid compared the softgel formulation of levothyroxine (Tirosint) against standard tablet (Synthroid) in 85 patients with malabsorption conditions. [5] TSH over-correction (TSH <0.1 mIU/L) occurred in 9% of the softgel arm vs. 5% of the tablet arm over 6 months, relevant because over-correction is the proximate driver of cardiac and bone adverse events.

Bone Mineral Density: Quantifying the Fracture Signal

TSH suppression below 0.1 mIU/L in post-menopausal women reduces lumbar spine bone mineral density (BMD) at a rate of approximately 2% per year. [6] Pre-menopausal women and men show a smaller but still measurable effect.

Meta-Analytic Evidence on BMD Loss

A Cochrane-adjacent systematic review published in the Annals of Internal Medicine examined 41 studies of thyroid hormone therapy and bone outcomes. [6] The pooled analysis found that suppressive levothyroxine doses (used in thyroid cancer management) produced a mean BMD reduction of 1.1% per year at the femoral neck in post-menopausal women (P<0.001). Replacement-dose therapy targeting a normal TSH did not produce a statistically significant BMD change compared with controls.

Clinical Implication

Patients on suppressive therapy, thyroid cancer survivors, primarily, face a fracture risk roughly 1.2 times that of age-matched controls after 10 or more years of TSH suppression, according to a Danish registry study of 4,946 thyroid cancer patients. [7] The American Thyroid Association 2015 guidelines now stratify TSH targets by cancer risk category precisely to minimize this exposure. [8]

Cardiovascular Event Rates: Atrial Fibrillation and Beyond

Atrial fibrillation is the most clinically consequential cardiovascular adverse event associated with levothyroxine over-treatment.

Subclinical Hyperthyroidism and AF Risk

A 2015 individual-participant meta-analysis in JAMA Internal Medicine pooled data from 10 prospective cohort studies (N=52,674) and found that a TSH below 0.1 mIU/L was associated with a hazard ratio of 1.68 (95% CI 1.16 to 2.43) for incident atrial fibrillation. [9] The risk was not elevated for TSH in the 0.1 to 0.44 mIU/L range. This dose-response pattern supports the conclusion that keeping TSH within the low-normal range (0.5 to 2.0 mIU/L) for replacement therapy essentially eliminates the excess AF signal.

Heart Failure and Coronary Events

The same 2015 meta-analysis found a hazard ratio of 1.94 (95% CI 1.01 to 3.72) for coronary heart disease events with TSH below 0.1 mIU/L. [9] Heart failure risk followed a similar pattern. For patients whose TSH sits in the normal range on Synthroid, the cardiac event rate is not measurably different from the general population.

FAERS Data: Post-Market Adverse Event Signals

The FDA's Adverse Event Reporting System (FAERS) provides a real-world counterpoint to controlled trial data, capturing rare events and outcomes in populations underrepresented in trials.

Volume and Top Signal Categories

As of the 2023 FAERS quarterly data release, levothyroxine-containing products account for over 50,000 total adverse event reports in the database. [10] The five most frequently reported preferred terms are: palpitations, weight changes, fatigue, alopecia, and insomnia, consistent with the label. Alopecia appears more prominently in FAERS than in trials, likely reflecting reporting bias (patients notice hair loss and attribute it to a new medication), but also a genuine pharmacologic signal: thyroid hormone influences the hair cycle, and any flux in hormone levels during dose initiation or adjustment can precipitate telogen effluvium.

Disproportionality Analysis

A pharmacovigilance disproportionality analysis published in 2021 in Drug Safety identified a reporting odds ratio (ROR) of 4.2 for "decreased TSH" as a concomitant finding with levothyroxine-associated cardiac events, reinforcing the mechanistic link between over-dosing and cardiac signals rather than a direct drug toxicity. [11]

Pediatric FAERS Signals

In pediatric reports, pseudotumor cerebri (idiopathic intracranial hypertension) carries a disproportionate signal (ROR approximately 6.1 in under-18 reports), corroborating the label warning. [10] This event is rare in absolute terms, estimated at fewer than 1 in 10,000 treated children, but warrants monitoring during the first 12 weeks of therapy.

Hypersensitivity and Excipient Reactions

True allergic reactions to levothyroxine sodium are rare. Most reactions attributed to "Synthroid allergy" trace to inactive excipients: acacia, lactose, and dyes used to color-code tablet strengths.

Lactose Intolerance and GI Symptoms

A 2020 study in Nutrients found that hypothyroid patients with lactose intolerance required meaningfully higher levothyroxine doses to achieve the same TSH as lactose-tolerant patients, evidence that lactose in the tablet impairs absorption rather than causes overt GI side effects in most users. [12] Switching to a lactose-free formulation (such as the softgel or liquid levothyroxine) corrected TSH in a majority of affected patients.

Dye Reactions

FD&C Yellow No. 5 (tartrazine), used in the 100 mcg Synthroid tablet, carries a small risk of urticaria in aspirin-sensitive patients. The FDA requires a label disclosure for tartrazine when it appears in oral drug products. [1]

Drug Interactions That Amplify Adverse Event Risk

Levothyroxine interacts with a wide range of drugs in ways that either increase or decrease its effective concentration, and both directions carry AE implications. [13]

Absorption-Reducing Interactions

Calcium carbonate, ferrous sulfate, proton pump inhibitors, and cholestyramine each reduce levothyroxine absorption by 20 to 40% when co-administered. [13] Under-absorption produces under-treatment, not direct adverse events, but drives compensatory dose increases that then risk over-treatment if the interacting drug is later stopped.

Metabolism-Accelerating Interactions

Rifampin, phenytoin, carbamazepine, and sertraline all accelerate levothyroxine metabolism via CYP induction or increased thyroid-hormone binding. A case series published in Thyroid documented TSH rises from within-range to above 10 mIU/L within 4 to 6 weeks of starting rifampin, requiring dose increases of 25 to 50 mcg in most patients. [14]

Interactions That Increase Hyperthyroid AE Risk

Amiodarone contains approximately 37% iodine by weight and alters both T4 and T3 metabolism in complex ways. [15] Patients on amiodarone who also take Synthroid require more frequent TSH monitoring (every 3 to 6 months) because both under-treatment and over-treatment become more likely. The cardiac AE risk from TSH suppression is particularly significant in this group given their underlying arrhythmia history.

Thyroid Cancer Patients: A Higher-Risk Subpopulation

Patients treated for differentiated thyroid cancer frequently receive suppressive levothyroxine doses (target TSH <0.1 mIU/L for high-risk disease, 0.1 to 0.5 mIU/L for intermediate-risk). This intentional over-replacement creates the most concentrated adverse event exposure in the levothyroxine-treated population.

The American Thyroid Association's 2015 guidelines stratify TSH targets explicitly by recurrence risk to balance oncologic benefit against the bone and cardiac harms of chronic TSH suppression. [8] For low-risk, disease-free patients after 1 to 2 years, the ATA states: "We recommend that TSH be maintained in the low-normal range (0.5 to 2.0 mIU/L) indefinitely." This single guideline shift, applied to the large population of low-risk survivors, has substantially reduced long-term bone and cardiac AE exposure.

Fracture Rates in Thyroid Cancer Survivors

A nationwide Danish cohort study following 4,946 thyroid cancer patients for a median of 8.7 years found an age-adjusted fracture incidence rate ratio of 1.23 (95% CI 1.09 to 1.38) compared with matched population controls. [7] The excess risk was concentrated in patients whose TSH remained below 0.1 mIU/L for more than 5 years.

Atrial Fibrillation in Cancer Survivors

In the same Danish cohort, the incidence rate of atrial fibrillation was 6.7 per 1,000 person-years in the thyroid cancer group vs. 4.3 per 1,000 person-years in controls, a rate ratio of 1.56. [7] The excess was again concentrated in the suppressed-TSH subgroup.

Monitoring Protocols to Minimize Adverse Event Incidence

Matching the monitoring schedule to the patient's risk profile reduces AE incidence in a measurable and practical way.

Standard Replacement Monitoring

For patients on replacement-dose levothyroxine with a stable TSH target of 0.45 to 4.5 mIU/L, TSH measurement every 6 to 12 months is appropriate once stable, per the American Thyroid Association. [8] Any dose change prompts a repeat TSH at 6 to 8 weeks.

Suppressive Therapy Monitoring

Patients on intentional TSH suppression below 0.5 mIU/L should have TSH checked every 3 to 6 months, a baseline DEXA scan, and annual electrocardiographic review if TSH remains below 0.1 mIU/L for more than 12 months. [8]

Special Populations

Pregnancy raises levothyroxine requirements by 25 to 50% starting in the first trimester. The Endocrine Society recommends TSH monitoring every 4 weeks through 20 weeks gestation and at least once in the third trimester. [16] Failure to increase dose in pregnancy does not produce maternal adverse events in the classic sense, but fetal neurodevelopmental risk is the consequence, making monitoring the primary safety tool in this population.

Alopecia: Mechanism and Incidence Estimate

Hair loss is the side effect patients ask about most, and it deserves a separate accounting. Levothyroxine-related alopecia is almost always telogen effluvium, a temporary shedding phase triggered by the metabolic shift as thyroid function changes.

A prospective observational study of 314 newly treated hypothyroid patients found that 14.6% reported noticeable hair shedding in the first 3 to 4 months of levothyroxine therapy. [17] By 6 months, hair density had returned to baseline in 91% of those affected. Persistent alopecia beyond 6 months warrants evaluation for other causes, iron deficiency, autoimmune alopecia, or ongoing thyroid instability, rather than attributing it to the drug.

Psychiatric and Cognitive Adverse Events

Anxiety, emotional lability, and sleep disturbance appear in the label and in FAERS data, but controlled trial evidence for psychiatric adverse events at replacement doses is limited. The TRUST trial found no significant difference in anxiety scores between levothyroxine and placebo at 12 months (mean Generalized Anxiety Disorder-7 score: 3.1 vs. 3.0, P=0.84). [2]

At suppressive doses, cognitive symptoms including difficulty concentrating and irritability are reported more commonly. A small crossover trial (N=56) published in the Journal of Clinical Endocrinology and Metabolism found that working memory scores were lower when TSH was maintained below 0.1 mIU/L compared with TSH of 0.5 to 2.0 mIU/L (P=0.03), though effect sizes were modest. [18]