Levothyroxine Dosing and the Risk of Suboptimal Thyroid Control: Insights from a Large Real-World Longitudinal Study

Highlight

This longitudinal, real-world study involving 4,449 adult patients treated with levothyroxine (LT4) reveals a clear dose-dependent increase in suboptimal thyroid hormone control, as evidenced by thyroid stimulating hormone (TSH) values outside normal ranges. Patients receiving LT4 doses greater than 150 µg exhibited over a 45% incidence of out-of-range TSH measurements within 6 months. These findings suggest a critical need for more frequent TSH surveillance, particularly among medically complex patients on higher LT4 doses.

Study Background

Hypothyroidism affects millions worldwide, typically requiring lifelong thyroid hormone replacement with levothyroxine to restore normal metabolism and prevent complications. Despite its widespread use, achieving optimal thyroid hormone balance remains challenging. Suboptimal treatment—whether under- or overtreatment—is common and has been linked to adverse cardiovascular outcomes, cognitive impairment, and decreased quality of life. Patients on higher LT4 doses may be at greater risk of such fluctuations, potentially due to underlying comorbidities, altered pharmacokinetics, or malabsorption issues. Understanding the relationship between LT4 dose and TSH control in routine clinical practice is essential to inform monitoring strategies and improve treatment outcomes.

Study Design

This longitudinal observational study included 4,449 adults treated with levothyroxine, each undergoing at least two serum TSH measurements at least 30 days apart during a 6-month follow-up period. Patients were categorized by LT4 daily dose: ≤50 µg, 51–100 µg, 101–150 µg, and >150 µg. The primary outcome was the presence of at least one TSH value outside an age-adjusted normal reference range during follow-up; a secondary outcome was TSH outside a broader range (0.1–10.0 mIU/L), representing clinically significant deviation.

Covariates adjusted for in the analyses included sociodemographic variables, initial TSH level, prevalent comorbid diagnoses, and medications known to affect LT4 metabolism or absorption. Statistical approaches comprised repeated measures analysis using generalized estimating equations (GEE) to account for correlated TSH measurements, and survival analysis with Cox proportional hazards to evaluate time to the first out-of-range TSH event. Kaplan-Meier curves illustrated the cumulative incidence of suboptimal TSH control across LT4 dose groups.

Key Findings

Across all patients, approximately 25% experienced at least one TSH measurement outside the age-adjusted normal range during the 6-month period. This frequency rose substantially among patients on the highest LT4 doses (>150 µg daily), where the incidence reached over 45%.

The GEE repeated measures analysis revealed a strongly dose-dependent association with suboptimal TSH control. Compared to patients on ≤50 µg daily, odds ratios (ORs) for TSH outside the normal range were 1.98 (95% CI not specified) for 51–100 µg, 3.39 for 101–150 µg, and 5.65 for >150 µg groups (all p < 0.001). A similar pattern was observed when using the wider TSH range of 0.1–10.0 mIU/L, supporting the robustness of the findings.

In the time-to-event (hazard) analysis, the risk of experiencing a suboptimal TSH measurement within 6 months correlated with LT4 dose, with hazard ratios (HRs) of 1.58, 2.73, and 3.08 for the 51–100, 101–150, and >150 µg groups, respectively. Kaplan-Meier curves visually demonstrated an accelerating incidence of out-of-range TSH values with increasing LT4 dose category.

These findings persisted after controlling for relevant covariates such as baseline TSH, comorbidities, and medications affecting thyroid hormone metabolism.

Expert Commentary

This study provides compelling real-world evidence that higher daily doses of levothyroxine are significantly associated with a greater likelihood of suboptimal thyroid hormone control, as reflected by TSH instability. This observation aligns with clinical experience showing the challenges in titrating LT4 doses in medically complex patients who often require higher doses due to factors such as impaired absorption, resistance states, or concurrent illnesses.

From a pathophysiological perspective, larger LT4 doses may increase the risk of overshooting or undershooting optimal thyroid hormone levels due to narrow therapeutic windows and patient-specific variability in metabolism. The findings underscore the importance of individualized monitoring strategies that go beyond the standard practice of annual or biannual TSH assessments.

Current guidelines typically recommend periodic TSH surveillance but do not specify enhanced monitoring for patients on high LT4 doses. This study supports revisiting these recommendations, especially for patients with prior TSH excursions or complex medical conditions.

Limitations include the observational design, which cannot establish causality, and the reliance on electronic medical record data that may not capture all confounders or reasons for dose adjustments. Nevertheless, the large sample size and robust statistical methods enhance the validity of the conclusions.

Conclusions

The present real-world cohort analysis reveals that levothyroxine doses above 100 µg daily correspond with a markedly increased risk of suboptimal hypothyroidism treatment, evidenced by out-of-range TSH values within six months. This dose-dependent relationship highlights the critical need for intensified TSH monitoring schedules in patients receiving higher LT4 doses, particularly those with a history of unstable thyroid function or complex comorbidities. Enhanced surveillance may allow timely dose adjustments to mitigate risks of both overtreatment and undertreatment, ultimately improving clinical outcomes and reducing cardiovascular and metabolic risks associated with thyroid dysfunction.

Further prospective interventional studies are warranted to determine optimal monitoring intervals and to assess whether more frequent TSH testing in this population improves long-term health outcomes.

Funding and Study Registration

The study by Ettleson et al. was supported by institutional and research grants, details of which are documented in the original publication. There is no mention of clinical trial registration, consistent with the retrospective observational nature of the study.

References

1. Ettleson MD, Wan W, Laiteerapong N, Kim BW, Bianco AC. A Longitudinal, Real-World Study of the Relationship between Levothyroxine Dose and Suboptimal Hypothyroidism Treatment. Thyroid. 2026 Jun 17:10507256261462468. PMID: 42308153.

2. Jonklaas J, Bianco AC, Bauer AJ, et al. Guidelines for the Treatment of Hypothyroidism: Prepared by the American Thyroid Association Task Force on Thyroid Hormone Replacement. Thyroid. 2014 Dec;24(12):1670-751.

3. Wiersinga WM. Subclinical hypothyroidism and thyroid hormone therapy—when to treat and when not to treat. Nat Rev Endocrinol. 2012 Oct;8(10):594-9.