2025 Volume 72 Issue 7 Pages 813-818
No evidence-based standards exist regarding levothyroxine (LT4) replacement therapy initiation timing in patients with hyperthyroid Graves’ disease undergoing total thyroidectomy. Although LT4 replacement from the first postoperative day has been the standard of care at our hospital, its clinical validity has not been thoroughly examined. This study investigated the perioperative kinetics of thyroid hormones to assess the safety and efficacy of early LT4 initiation. Thirty patients with Graves’ disease (18 hyperthyroid and 12 euthyroid) and 12 with thyroid nodules who underwent total thyroidectomy were included. Blood samples were collected from each patient for thyroid hormone measurement on the day before surgery (D-1), 15 min after surgery (D0), at 8:00 am on days 1 (D1) and 3 (D3), and 3 weeks (W3) and 3 months (M3) after surgery. In 18 patients with hyperthyroid Graves’ disease, serum free triiodothyronine (FT3) levels significantly decreased immediately after surgery and were within the normal range by D1. Although LT4 was started on D1, FT3 levels continued to decline by D3 and remained low at W3 and M3. Serum FT4 levels followed a slower decline but remained within the normal range for M3. In patients with euthyroid Graves’ disease and those with thyroid nodules, hormone levels stayed within or around the reference range throughout the observation period. In conclusion, initiating LT4 on the day after surgery is safe and effective for maintaining thyroid function in patients with hyperthyroid Graves’ disease undergoing total thyroidectomy. These results could inform future guidelines, supporting earlier postoperative LT4 initiation.
Total thyroidectomy is a common treatment for thyroid diseases, including Graves’ disease and thyroid cancer [1]. Endocrine societies generally recommend that patients with Graves’ disease should be rendered euthyroid with antithyroid drugs (ATD) and/or iodine before surgery to minimize the risk of thyroid storm [2-4]. However, surgery is the only option in some cases, particularly when thyroid function is poorly controlled due to medication side effects. While surgery in the context of active thyrotoxicosis poses risks, several studies have demonstrated that thyroidectomy can be safely performed in these patients [5, 6].
The American Association of Endocrine Surgeons guidelines suggest that in patients with hyperthyroid Graves’ disease undergoing thyroidectomy, ATD should be discontinued postoperatively, and levothyroxine (LT4) replacement should begin approximately 1 week after surgery to allow thyroid hormone levels to normalize [4]. However, this recommendation is not supported by direct evidence. To our knowledge, only two studies have investigated postoperative thyroid hormone kinetics in patients with hyperthyroid Graves’ disease undergoing subtotal or total thyroidectomy [7, 8]. Hermann et al. have observed a rapid decline in free thyroid hormone levels, particularly free triiodothyronine (FT3), within the first postoperative day. FT3 levels normalized by day 2, while free thyroxine (FT4) levels took longer to reach normal ranges. Based on their findings, the authors suggested initiating LT4 replacement therapy within 2 weeks after surgery.
At our hospital, contrary to the guidelines mentioned above [4], LT4 therapy has traditionally been initiated on postoperative day 1, even in patients with hyperthyroid Graves’ disease, with favorable hormonal control outcomes. This practice is based on studies showing that elevated thyroid hormone levels accelerate hormone metabolism, leading to shorter half-lives [9, 10], and decrease the hormonal responses due to downregulation of thyroid hormone receptors and desensitization of the target genes [11, 12]. These findings suggest that despite normal thyroid hormone levels after surgery, patients experience a transient hypothyroid state, which could have clinical implications, such as impaired wound healing and the progression of thyroid eye disease. Given these observations, we aimed to investigate the perioperative kinetics of thyroid hormones in our patient population and provide new insights into the optimal timing for initiating LT4 replacement therapy.
This study was approved by the Ethics Committee of Yamashita Thyroid Hospital (approval no. 2022-13). Written informed consent was obtained from all participants. All procedures were performed in accordance with the principles of the Declaration of Helsinki.
Study population and proceduresThis study initially included 27 patients with Graves’ disease and 12 with thyroid nodules who underwent total thyroidectomy at our hospital between March and July 2022. Due to varying degrees of hyperthyroidism, three additional patients with Graves’ disease with severely elevated FT3 levels (>20 ng/mL) who underwent total thyroidectomy between March 2023 and August 2024 were also included. In total, the study comprised 18 patients with Graves’ disease who had elevated thyroid hormone levels (FT3 and/or FT4) the day before surgery (classified as the hyperthyroid group [HT group]), 12 with Graves’ disease who were in a euthyroid state at the time of surgery (euthyroid group [ET] group), and 12 with thyroid nodules (nodule group [N group]).
The most common indications for surgery in patients with Graves’ disease were poor control of thyroid function and adverse reactions to ATD. Blood samples were collected at four different time points as follows: the day before surgery at 2:00 pm (preoperative = D-1), 15 min after thyroid removal (intraoperative, D0), postoperative day 1 at 8:00 am (D1), and postoperative day 3 at 8:00 am (D3). Additional samples were taken in the outpatient department at 3 weeks (W3) and 3 months (M3) postoperatively. Serum thyroid stimulating hormone (TSH), thyroid hormones, and thyroglobulin (Tg) levels were measured. LT4 replacement therapy (1.6 μg/kg) was initiated on postoperative day 1 after blood sample collection (D1).
All surgeries were performed by five endocrine surgeons following standardized procedures. Each patient underwent total thyroidectomy using the Thunderbeat Open Fine Jaw (Olympus, Japan) or Harmonic Focus (Ethicon, Johnson & Johnson, Cincinnati, OH, USA) under general anesthesia with total intravenous anesthesia. Thyroid vessels were carefully dissected prior to gland mobilization to prevent a surge of thyroid hormones. The thyroid gland was removed via extracapsular dissection with intraoperative nerve monitoring [13] of the recurrent laryngeal nerves and external branches of the laryngeal nerve, ensuring the preservation of the parathyroid glands.
Laboratory testsSerum levels of FT3 (the normal reference range; 2.3–4.3 pg/mL), FT4 (0.9–1.7 ng/dL), TSH (0.5–5.0 mU/L), and Tg (<35 ng/mL) were measured using an electrochemiluminescence immunoassay kit (COBAS 8000 e801 analyzer, Roche Diagnostics, Indianapolis, IN, USA). The normal reference range of the FT3/FT4 ratios calculated from the data obtained using these kits was 1.92–2.95 [14].
Statistical analysesData are presented as mean ± standard deviation (SD). Statistical differences between the three study groups were assessed using the Kruskal–Wallis and Chi-square tests for continuous variables and categorical variables, respectively. Thyroid hormone levels across different time points were analyzed using Friedman’s test, and correlations were determined using Pearson’s correlation coefficient.
All statistical analyses were performed using EZR version 1.68 (Saitama Medical Center, Jichi Medical University, Saitama, Japan) [15], which provides a graphical user interface for R version 4.02 (The R Foundation for Statistical Computing, Vienna, Austria). EZR is a modified version of the R commander that adds commonly used biostatistical functions. Statistical significance was considered at p < 0.05.
The clinical and biochemical characteristics of the patients are summarized in Table 1. No significant differences were observed in age or sex ratios between the groups. TSH levels were below detectable limits in the HT group and within normal ranges in the ET and N groups. FT3 and FT4 levels were significantly elevated in the HT group compared with the ET and N groups (p < 0.01 for both). In the HT group, 14 of the 18 patients had elevated FT3 and FT4 levels, while 4 had elevated FT3 levels alone. All patients in the HT group and 7 of the 12 in the ET group were preoperatively treated with potassium iodine. Of these, six and three patients in the HT and ET groups, respectively, were treated with potassium iodide alone. Nine and four patients in the HT and ET groups, respectively, received potassium iodide combined with ATD such as thiamazole or propylthiouracil, and three in the HT group were treated with potassium iodide combined with dexamethasone. The remaining five patients in the ET group were treated with ATDs.
| Groups | HTc | ETc | Nc |
|---|---|---|---|
| Number of patients | 18 | 12 | 12 |
| Ages (means, ranges) | 40.2 (14–65) | 47.2 (18–62) | 50.9 (34–75) |
| M:Fa | 3:15 | 3:9 | 3:9 |
| TSHb (mIU/mL) (mean ± SD) | <0.05 in all | 2.0 ± 2.2 | 1.4 ± 1.0 |
| FT3b (pg/mL) (mean ± SD) | 11.5 ± 6.6 | 2.8 ± 0.4* | 2.9 ± 0.5* |
| FT4b (ng/dL) (mean ± SD) | 2.7 ± 1.9 | 0.9 ± 0.2* | 1.2 ± 0.1* |
aM and F indicate male and female, respectively.
bTSH, FT3, and FT4 indicate thyroid stimulating hormone, free triiodothyronine, and free thyroxine, respectively.
cHT, ET, and N groups indicate hyperthyroid, euthyroid, and nodule groups, respectively.
*p < 0.01 vs. the HT group.
Figs. 1 and 2 illustrate the perioperative changes in thyroid hormone (FT3, FT4 and FT3/FT4 ratios) and Tg concentrations in the three groups. The actual data for each case in the HT group are shown in Supplementary Fig. 1. In the HT group, FT3 levels significantly decreased immediately after surgery (D0 versus [vs.] D-1, p < 0.01), and the mean FT3 level was within the normal reference range on D1 (D1 vs. D0 or D-1, p < 0.01) (Fig. 1A). Although LT4 replacement was started on D1, the mean FT3 levels continued to decline to the lower limit of the reference range on D3 and remained around this level at W3 and M3 (Fig. 1A). In contrast, the kinetics of serum FT4 were slower than those of T3 (Fig. 1B). The mean FT4 level was still higher than the upper limit of the reference range on D1. Similar to FT3 levels, even after the initiation of LT4 replacement, the mean FT4 levels continued to decline on D3 and remained within the reference range for the next 3 months. As previously reported [16], the FT3/FT4 ratios were higher in the hyperthyroid state (D-1 and D0) than in the non-hyperthyroid state and then gradually decreased (Fig. 1C).
The closed circles and error bars represent mean ± SD. The gray areas indicate the normal reference ranges. *, p < 0.01 vs. D-1. FT3, free triiodothyronine; FT4, free thyroxine; Tg, thyroglobulin; HT, hyperthyroid; SD, standard deviation.
(A) Data on T3, r = 0.497, p = 0.035. (B) Data on T4, r = 0.68, p = 0.002. Open circles depict three patients with extremely high thyroid hormone levels. HT, hyperthyroid; T3, triiodothyronine; T4, thyroxine.
As shown in Fig. 2, significant positive correlations were found between thyroid hormone levels immediately after surgery (D0) and their diminution rates from D0 to D1 (r = 0.50, p = 0.04 for FT3 level and r = 0.68, p < 0.01 for FT4 level), demonstrating that the higher the thyroid hormone levels, the more accelerated the thyroid hormone metabolism. It should be noted that thyroid hormone levels continued to decline after the commencement of LT4 replacement therapy on D1, even in three patients with extremely high levels of thyroid hormones (FT3 >20 pg/mL) on D-1 in the HT group (shown as open circles in Fig. 2 and Supplementary Fig. 1).
In contrast, in the ET and N groups, serum FT3 and FT4 levels remained within or around the reference range throughout the observation period (Fig. 3A and B). The FT3/FT4 ratios that were slightly higher preoperatively in the ET group and within the reference range in the N group remained around the lower limit of the reference range on LT4 treatment after surgery (Fig. 3C).
The closed circles and error bars represent the mean ± SD. The gray areas indicate the normal reference ranges. *, p < 0.01 vs. D-1. FT3, free triiodothyronine; FT4, free thyroxine; Tg, thyroglobulin; ET, euthyroid; N, nodule; SD, standard deviation.
Preoperative Tg concentrations varied widely across all groups, with values ranging from 0.4 to 2,153.0 ng/mL in the HT group, 2.1 to 2,145 ng/mL in the ET group, and 13.0 to 1,104.0 ng/mL in the N group. Postoperative thyroid gland manipulation caused a significant increase in Tg levels in all groups: from 411.7 ± 967.7 to 5,251.0 ± 6,636.2 ng/mL in the HT group, 592.0 ± 686.0 to 3,501.0 ± 4,382.2 ng/mL in the ET group, and 309.1 ± 340.4 to 1,242.5 ± 1,345.5 ng/mL in the N group (p < 0.01 for all groups) (Figs. 1D and 3D).
Finally, we assessed for thyrotoxic symptoms such as tachycardia and tremor after initiating LT4 replacement therapy, but found no evidence of these symptoms.
This study demonstrates that, in patients with hyperthyroid Graves’ disease who underwent total thyroidectomy, postoperative declines in thyroid hormone levels, particularly FT3, were rapid, and the early initiation of weight-based LT4 replacement therapy maintained thyroid hormone levels within or near the normal reference ranges without adverse effects. Consequently, the FT3/FT4 ratio decreased, and Tg levels showed marked elevations postoperatively. While we did not assess thyroid hormone levels when LT4 replacement therapy was initiated 1 week later, as recommended by the guidelines [4], it is reasonable to expect a decrease in thyroid hormone levels below the lower reference limit based on the present data.
Rapid declines in serum FT3 levels in our patients likely reflect accelerated metabolism of FT3, as noted in previous studies [9, 10], despite enhanced T4 to T3 conversion in a hyperthyroid state [17]. Elevated serum reverse T3 levels, a marker of increased thyroid hormone metabolism, have been similarly reported [7]. As T3 is the biologically active thyroid hormone, early initiation of LT4 replacement therapy appears beneficial for maintaining its normal levels, even in patients with hyperthyroid Graves’ disease. Although serum FT4 levels were elevated on postoperative day 1 (D1 in Fig. 1B), they quickly decreased to the reference range after LT4 replacement therapy, suggesting that initiating therapy on D1 was appropriate and not excessive. For patients with significantly high preoperative thyroid hormone levels (FT3 >20 pg/mL), such as three patients in this study (Fig. 2 and Supplementary Fig. 1), delaying LT4 replacement therapy by a few days may be advisable. Therefore, thyroid hormone levels on the day after surgery could inform clinical decision-making in these cases.
Maintaining FT4 levels near the upper reference range is reported critical for ensuring sufficient amounts of serum FT3 in athyreotic patients, because T3 is no longer secreted from the thyroids in such cases [18], although conflicting data exist [19]. This is also a reason for decreased FT3/FT4 ratios after total thyroidectomy. Previous studies showing excess T3-mediated down-regulation of thyroid hormone receptors by accelerating ubiquitin-proteasome system-mediated degradation and desensitization of thyroid hormone-target genes [11, 12] also suggest the importance of maintaining FT4 levels near the upper limit of the reference range in patients with hyperthyroid Graves’ disease after total thyroidectomy.
The periods during which FT4 levels declined to the upper limit of the reference range were shorter in our study (1 day) than in Hermann’s reports (5 days) [7, 8]. This is at least partly due to the higher preoperative concentrations of thyroid hormones in their studies since they selected cases with thyroid hormone levels more than three times the upper limits of the maximal normal values and without previous administration of ATDs or iodine. In contrast, our study included patients with mildly hyperthyroid Graves’ disease with elevated FT3 and normal FT4 levels, and all patients were treated with ATDs, iodine, and/or dexamethasone. The different surgical modalities, namely total thyroidectomy in our study and subtotal or near-total thyroidectomy in their studies may also explain the variations in postoperative thyroid hormone kinetics.
Interestingly, while serum Tg (the precursor for thyroid hormones) levels increased markedly in all groups following thyroidectomy, a corresponding rise in serum thyroid hormone levels was observed during thyroid manipulation and incision. This finding aligns with the previous reports [7, 8, 20] and is due to the lack of enzymes necessary in the periphery to generate thyroid hormones from Tg. These data have been used to support the safety of total thyroidectomy in patients with hyperthyroid Graves’ disease [6, 21].
While our study supports the clinical validity of initiating LT4 replacement therapy on postoperative day 1 in patients with uncontrolled Graves’ disease, it has some limitations. First, the relatively small sample size and a single-center design at a high-volume institution may reduce the generalizability of our results. Second, we did not evaluate thyroid hormone kinetics in the first week after surgery without LT4 replacement, as recommended by the American Association of Endocrine Surgeons [4]. This is because this study was a prospective study to confirm the validity of our clinical practice at our hospital rather than to determine the optimal starting timing for postoperative T4 administration in hyperthyroidism. Third, the potential effects of medications that influence T4 to T3 conversion were not examined [22-24]. Finally, while early LT4 supplementation may promote wound healing and improve postoperative mental status, we could not assess these outcomes. Therefore, these issues need to be explored in future studies.
In conclusion, initiating weight-based LT4 replacement therapy on the day after surgery is safe and effective for managing thyroid function in patients with hyperthyroid Graves’ disease undergoing total thyroidectomy (Graphical Abstract). As the existing guidelines for LT4 administration timing are based on limited evidence [4], our findings may prompt reconsideration of these recommendations in future guidelines. However, as mentioned above, in cases with extremely high thyroid hormone levels (e.g., FT3 >20 pg/mL), it may be advisable to delay the start of LT4 for a few days. Further studies are needed to clarify this issue.
We thank Dr. Tomoyuki Ohara for his advice on statistical processing, and Editage (www.editage.jp) for English language editing.
There is no funding for this work for all the authors.
None of the authors have any potential conflicts of interest associated with this research.
Y.N. contributed to conceptualization (supporting), data interpretation, and writing (original draft); S.T. and T.F. to methodology, interpretation of data, statistical analysis, and writing (original draft); K.K., D.T., and Y.M. to data collection, article review, and the discussion; H.S. to conceptualization (supporting), visualization, and writing (review & editing); H.T., M.O., and A.Y. to conceptualization (supporting), data interpretation, and discussion; S.S. to conceptualization (supporting), data interpretation, and writing (review & editing); and H.Y. to conceptualization, design of the study, supervision, and writing (review & editing).
