Serum Thyroglobulin Measured With a Second-Generation Assay in Patients Undergoing Total Thyroidectomy Without Radioiodine Remnant Ablation: A Prospective Study

Introduction

Indications for Iodine-131 (¹³¹I) ablation have decreased in the case of patients with papillary thyroid carcinoma (PTC) submitted to total thyroidectomy who have complete tumor resection and whose histological data suggest a low risk of persistent or recurrent disease. Follow-up consisting of the measurement of nonstimulated serum thyroglobulin (Tg) combined with neck ultrasonography (US) is recommended for these patients without indication of ¹³¹I. Consequently, if the risk of persistent or recurrent disease is sufficiently low to exclude the need of ¹³¹I ablation, suppression of thyrotropin (TSH) is not recommended. The latter should be maintained in the lower half of the normal range (1 –6).

Second-generation Tg assays (2,6 –9) have been increasingly used. These assays have been extensively evaluated in patients submitted to ablation with ¹³¹I, but few studies are available for patients treated only by total thyroidectomy (10 –12), and they have limitations. First, these studies were retrospective. Second, the time after surgery when Tg levels were determined was not uniform and varied widely; for example, the first Tg was obtained 3 to 12 months (10), 1 to 34 months (11), and <1 to 6 months (12) after thyroidectomy. Third, serum TSH was also not uniform at the time when Tg was measured, with TSH being suppressed or >2 mIU/L in many patients. Under these conditions, Tg can be lower or higher (11,12), respectively, than the concentration obtained with a TSH in the lower half of the normal range as recommended (1 –6). Fourth, neck US for the exclusion of small lymph node metastases was not performed in all patients (11,12). Finally, even in patients submitted to US, the measurement of remnant thyroid tissue was not reported (10 –12). This measurement is necessary to know to which volume of residual tissue the Tg values found would apply.

This prospective study was conducted in view of (i) the growing number of patients with PTC submitted only to total thyroidectomy (without radioiodine); (ii) the importance of knowing whether Tg levels in these individuals can be attributed to residual thyroid tissue or, in contrast, suggest residual disease (13); (iii) the increasing use of second-generation Tg assays; and (iv) the scarcity of data using this type of assay in these patients.

The objective of this study was to define the concentrations of nonstimulated Tg levels measured with a second-generation assay and a serum TSH between 0.5 and 2 mIU/L, 3, 6, 12, and 24 months after total thyroidectomy.

Patients and Methods

Results

Sixty-nine patients (54 women and 15 men), ranging in age from 21 to 72 years (median 43 years), all without DTC on histology, but eight with MTC, were studied. Basal serum calcitonin was undetectable in the latter during the postoperative period.

Residual thyroid tissue was <2 g in all patients [ranging from undetected to 1.3 g (median 0.4 g)], in agreement with the definition of total thyroidectomy (13,20,21).

The Tg concentrations obtained at the different time points after total thyroidectomy are shown in Table 2 (for the individual results of each patient, see Table 3). An important reduction in serum Tg was observed for the evaluation at 180 versus 90 days. A smaller variation was found for the evaluation at 360 versus 180 days. Finally, no change in serum Tg was observed between 360 and 720 days. There was no case of Tg conversion from undetectable to detectable among these patients in whom all Tg measurements were obtained with TSH in the same range (0.5–2 mIU/L), and none of the patients presented an increase in Tg.

In the last assessment performed 2 years after surgery, none of the patients exhibited suspicious lesions on neck US, as expected.

TgAb were undetectable in all patients before thyroidectomy (inclusion criterion) and continued to be undetectable in the subsequent four assessments in 67 patients. Two patients had positive TgAb 3 months after surgery (these patients were excluded from this evaluation of Tg), but TgAb became undetectable in the subsequent assessment. There was no case of conversion from undetectable TgAb 6 months after surgery to positive in the subsequent assessments.

Of note, the exclusion of 8 patients with MTC did not alter the results shown in Table 2.

Discussion

We first highlight some characteristics of the present study. In contrast to previous series that also measured Tg in patients not submitted to ablation with ¹³¹I using a second-generation assay (10 –12), this was a prospective study whose specific objective was to define the expected concentrations of Tg after total thyroidectomy. The sampling times were uniform. Since in many patients the nadir of Tg is not reached early, the inclusion of Tg obtained a few weeks after thyroidectomy (11,12) may overestimate, for example, the values expected for 3 months postsurgery; in contrast, the inclusion of Tg obtained for the first time 2 years after thyroidectomy (11) may underestimate, for example, the values expected for 6 or 12 months postsurgery. Furthermore, the TSH range at the time of Tg measurement (and also between measurements) was the same for all subjects and was defined as recommended for low-risk patients with PTC without indication for ¹³¹I (1 –6). Considering that the patients had normal thyroid remnants, which are known to respond well to serum TSH (21), it is reasonable to imagine that Tg obtained in the presence of suppressed TSH is lower and a Tg obtained in the presence of TSH levels that are >2 mIU/L or elevated is higher than a Tg obtained in the presence of a TSH between 0.5 and 2 mIU/L, especially when Tg is measured with a sensitive assay, as suggested by the results of previous studies (11,12). In the retrospective studies, TSH was not uniform at the time of Tg measurement. Since the objective was to define Tg concentrations that could be attributed to thyroid remnants, we chose to include patients without DTC but undergoing the same surgical procedure as patients with PTC. Additionally, all patients were submitted to neck US in the last assessment. Although unlikely, the presence of small lymph node metastases cannot be ruled out in some patients of the previous studies in which neck US was not performed in all patients (11,12). Finally, we estimated thyroid remnants in order to define to which patients the observed results would apply. This measurement was not performed in any of the previous studies (10 –12).

An important decrease was observed in the comparison of Tg obtained at 6 months versus 3 months after thyroidectomy, but a less marked reduction was also seen for Tg measured at 12 months versus 6 months after surgery. Measurement of Tg at 3 months after surgery may therefore be early, and measurement of the first Tg after 6 months would be more adequate, expecting stability after 1 year. Since this recommendation applies to patients with complete tumor resection who are at low risk, it is unlikely that this delay in evaluation of a few months would have negative repercussions. We did not measure preoperative Tg, but the patients included in this study did not have voluminous goiter or hyperthyroidism and were from an iodine-sufficient region (unpublished data); it is therefore unlikely that the patients had very high Tg concentrations before surgery. Nevertheless, if the half-life suggested in two previous studies is considered [approximately 28 h (22) and 65 h (23)], even starting from extremely high preoperative concentrations, the Tg nadir would be achieved early (within a maximum of 4 to 6 weeks). This fact suggests that the Tg reduction observed in many patients even 3 and 6 months after thyroidectomy may be due to thyroid cell apoptosis secondary to inflammation or postoperative ischemia, and/or to some degree to atrophy of these cells secondary to the maintenance of TSH at constant levels <2 mIU/L, which is still within the reference range.

In the present study, in which all Tg measurements were obtained with TSH between 0.5 and 2 mIU/L (predefined criterion), none of the patients presented an increase in Tg, confirming that a reduction or stability of this marker can be expected in the absence of the thyroid. However, small, nonprogressive, and transient elevations in Tg have been reported in tumor-free patients (12,24 –27). One explanation for this finding is that small increases in serum TSH stimulate normal residual tissue to secrete more Tg and that this increase, although mild, is detectable when new-generation assays are employed (27). A difference in the secretion of Tg by remnant thyroid tissue has been demonstrated for TSH levels <0.5 mIU/L versus >0.5 mIU/L, and a direct relationship between serum TSH and Tg is observed if the levels are >0.5 mIU/L (12). In practice, this means that Tg can become detectable or increase when TSH<0.5 mIU/L returns to a normal value, or when it oscillates from lower to upper limit but still within the reference range, or even when it increases (although discretely). Inversely, a small adjustment in L-T4 therapy aiming at TSH close to the lower limit of the reference range might be sufficient for discretely elevated detectable Tg to become undetectable, eliminating further investigation. This possibility (i.e., an increase in nonstimulated Tg without indicating recurrence) should be taken into consideration by clinicians when new-generation assays are used.

None of the patients presenting undetectable TgAb 6 months after thyroidectomy converted to positive, as observed in 290 patients of a previous series (10), confirming that this is expected in the absence of the thyroid and that conversion of TgAb (measured by the same method and especially if increasing) would be considered suspicious of recurrence. This does not apply to transiently positive TgAb observed immediately after thyroid tissue injury caused by surgery or radiation, as observed in two patients of the present study. Although not evaluated here, as seen in patients undergoing ablation with ¹³¹I (28), the presence of discrete thyroid remnants does not seem to prevent the reduction or even the conversion of initially TgAb-positive patients without residual disease to negative TgAb (11,29).

Despite differences in the time of Tg measurement after surgery and in serum TSH, the present results and those of the previous series (10 –12) generally show that more than half of the patients achieve an undetectable Tg one year after total thyroidectomy, even in the absence of radioiodine ablation and when measured with a second-generation assay. On that occasion, Tg is >2 ng/mL in exceptional cases [98% of 275 patients compiled from the present study and from other series (10 –12,30) had Tg≤2 ng/mL]. However, patients with a serum Tg>2 ng/mL do not necessarily have metastases. It should be noted that these Tg concentrations may reflect discrete thyroid remnants (<1.5 g estimated by US in the present study) and that elevated Tg levels can be seen after less extensive surgery. The estimation of thyroid remnants by US in these patient with Tg>2 ng/mL would therefore be interesting. For example, the finding of >2 g of residual thyroid tissue makes it more likely to be the source of Tg, while the lack of detection of residual tissue by US would alert to a higher probability of metastases. Furthermore, the first hypothesis is supported by a reduction or stability of Tg in subsequent measurements, while the latter becomes more likely in the presence of Tg progression.

In conclusion, an important decline in serum Tg occurs between 3 and 6 months after total thyroidectomy and, to a lesser extent, between 6 and 12 months after this procedure. One year after surgery, Tg was undetectable in approximately 60% of the patients and was ≤2 ng/mL in all of them, even when obtained with TSH between 0.5 and 2 mIU/L and measured with a second-generation assay. No increase in Tg or conversion of TgAb from undetectable to positive was observed in these patients without residual thyroid.