- Email: [email protected]
Thyroid blood flow as a useful predictor of relapse of Graves’ disease after normal delivery in patients with Graves’ disease
Biomedicine & Pharmacotherapy 64 (2010) 113–117
Original article
Thyroid blood flow as a useful predictor of relapse of Graves’ disease after normal delivery in patients with Graves’ disease Toshiki Nagasaki a, Masaaki Inaba a,*, Misako Fujiwara-Ueda a, Junko Nishio b, Yasuro Kumeda a, Yoshikazu Hiura a, Hideki Tahara a, Eiji Ishimura a, Osamu Ishiko b, Yoshiki Nishizawa a a b
Department of Metabolism, Endocrinology and Molecular Medicine, Osaka City University Graduate School of Medicine, 1-4-3, Asahi-machi, Abeno-ku, Osaka 545-8585, Japan Department of Obstetrics and Gynecology, Osaka City University Graduate School of Medicine, 1-4-3, Asahi-machi, Abeno-ku, Osaka 545-8585, Japan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 April 2009 Accepted 22 April 2009 Available online 20 October 2009
Objective: Measurement of the peak systolic velocity (PSV) in the inferior thyroid artery (ITA) before withdrawal of an anti-thyroid drug (ATD) is useful for predicting relapse of Graves’ disease (GD). We further investigated whether the ITA–PSV can be used for prediction of GD relapse after delivery in euthyroid women with GD who stopped ATD administration during mid- to late pregnancy. Patients and methods: ITA–PSV was monitored monthly for three months after delivery in 42 women with GD aged from 24 to 45 years old (mean SE: 34.7 0.92 years old) who met the above criteria. To confirm the stability of the measurement, ITA–PSV was also measured monthly in 32 age-matched nonpregnant normal women and for three months after delivery in 10 age-matched women. Results: ITA–PSV and thyroid volume were higher in women with GD immediately after delivery compared to normal women, but the levels of TSH receptor antibody (TRAb) and thyroid-stimulating antibody (TSAb) did not differ significantly between the two groups. Of the 42 patients, 23 had relapse of GD and the smoker/non-smoker ratio and thyroid volume in these patients immediately after delivery were significantly higher than those in the 19 patients who did not undergo relapse (10/23 vs. 0/19, p < 0.0001; 24280.3 2280.9 vs. 19670.0 2103.7 mm3, p = 0.046), while ITA–PSV, TRAb and TSAb did not differ between the two groups of patients. The ITA–PSV ratio was calculated by dividing each value in the follow-up period by that obtained immediately after delivery. A significant increase in the mean ITA–PSV ratio occurred at least one month before the time of relapse (1.00 0.00 at –3 months before relapse vs. 1.46 0.12 at –1 month, p = 0.010; 1.00 0.00 at –3 months vs. 1.77 0.13 at the time of relapse, p = 0.0048). In contrast, there were no significant changes in this ratio during the follow-up period in nonrelapse patients. Conclusion: Monthly measurement of ITA–PSV after delivery in remitted euthyroid women with GD may assist in early prediction of GD relapse. ß 2009 Elsevier Masson SAS. All rights reserved.
Keywords: Inferior thyroid artery Color Doppler measurement Delivery Hyperthyroidism Graves’ disease Relapse
1. Introduction The overall relapse rate of Graves’ disease (GD) is as high as 30– 50% [1], and women in the postpartum period have a particularly higher frequency of GD relapse due to the profound autoimmune modification that occurs after delivery [2]. However, there is no established method for prediction of postpartum relapse of GD, except for detection of thyroid-stimulating antibodies (TSAb) early in pregnancy [3]. The increased sophistication of ultrasonographic instruments has made it possible to measure blood flow in the inferior thyroid artery (ITA), a major feeding artery in the thyroid (Fig. 1), and this is a useful method for differentiation of various autoimmune thyroid diseases [4]. Moreover, we recently reported
* Corresponding author. E-mail address: [email protected] (M. Inaba). 0753-3322/$ – see front matter ß 2009 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.biopha.2009.04.038
that the peak systolic velocity (PSV) in the ITA might assist in the prediction of early GD relapse after anti-thyroid drug (ATD) withdrawal [5] and that the ITA–PSV in untreated hyperthyroid patients with GD may reflect GD activity and methimazole sensitivity [6]. The aim of the current study was to assess the value of monthly measurement of ITA–PSV by a pulsed Doppler method for predicting relapse of stimulatory hyperthyroidism after normal delivery in pregnant women with GD who maintained an euthyroid state without ATD administration in mid- to late pregnancy.
2. Patients and methods 2.1. Patients Informed consent was obtained from all subjects. Pregnant women with GD aged from 24 to 45 years old (mean SE:
114
T. Nagasaki et al. / Biomedicine & Pharmacotherapy 64 (2010) 113–117
Fig. 1. Duplex Doppler recording of the right inferior thyroid artery in patients with Graves’ disease. A Doppler image was used to determine the appropriate sampling point for pulsed Doppler recordings. The sample volume was positioned in the right inferior thyroid artery at the inflow point to the thyroid (arrow, left). The velocity waveform is displayed above the baseline to indicate the arterial blood flow (right).
34.7 0.92 years old, n = 42) who had maintained an euthyroid state without ATD administration in mid- to late pregnancy were consecutively recruited into the study. ITA–PSV was measured at the time of enrolment and then every month until three months after delivery. Diagnosis of GD in these patients was based on a history of ATD administration during the first trimester of pregnancy. As controls, ITA–PSV was measured monthly for three months in 32 healthy women aged from 25 to 44 (35.7 1.2) years old who were not pregnant and in 10 healthy women aged from 29 to 37 (33.0 1.3) years old for three months after normal delivery. GD relapse was defined as an increase in the serum level of free thyroxine (FT4) above the normal upper range with suppression of serum TSH below the normal lower limit. A diagnosis of hyperthyroidism due to GD relapse rather than postpartum thyroiditis was made on the basis of a positive reaction for TSH receptor antibody (TRAb) or TSAb, or a requirement for ATD therapy to normalize serum levels of FT4 for several months. 2.2. Biochemical parameters for thyroid function and autoantibodies Blood samples drawn just before the ultrasound study after an overnight fast were kept on ice for 1 h and then centrifuged at 1000 g for 10 min. The serum and plasma samples obtained were stored in aliquots at –20 8C until assayed. Measurements were made immediately after thawing. Commercially available highcapacity, random-access immunoassay kits were used to measure FT4, FT3 and TSH levels (Chiron Diagnostics, East Walpole, MA, USA) [7]. TRAb and TSAb were measured by radioreceptor assay using a commercial kit (Cosmic Corp., Tokyo, Japan) [8] and a radioimmunoassay kit (Yamasa, Chiba, Japan) [9], respectively. 2.3. Blood flow in the inferior thyroid artery and thyroid volume Thyroid blood flow was measured in the ITA as previously reported [5,6] (Fig. 1). Ultrasound examinations were performed
using a duplex Doppler apparatus (SSD 2000, Aloka, Tokyo, Japan) with a 5 MHz convex array probe in both the color Doppler and pulsed Doppler modes. The angle correction cursor was parallel to the direction of flow. The peak systolic flow velocity (PSV) of the right ITA, which is automatically calculated by the ultrasound apparatus, was used as an index of ITA thyroid blood flow. All measurements were performed by one examiner (M. F.-U.), who was blinded to the subject characteristics. Thyroid blood flow was determined in the ITA since this artery is a major contributor to thyroid blood flow, identification and measurement are straightforward, and the procedure has a low coefficient of variation (CV) (< 5.2%). The change of blood flow in the ITA was expressed as the change in the ITA– PSV ratio, which was obtained by dividing ITA–PSV at each measurement time point by the value obtained immediately after delivery. In normal non-pregnant women, the ITA–PSV ratio was calculated by dividing ITA–PSV at each time point by the value from the first examination. Thyroid volume was measured by ultrasound and calculated using an ellipsoid model (width length thickness 0.7 for each lobe) [10]. 2.4. Statistical analysis Data are shown as means SE unless otherwise indicated. Statistical analysis was performed with StatView v.5.0 (SAS Institute, Cary, NC). Differences in clinical factors among GD patients, normal non-pregnant women, and normal women after delivery were examined using a two-tailed multiple t-test with a Bonferroni correction. Differences in clinical values between GD patients with and without relapse were examined using a Mann-Whitney U test for assessment of medians. The difference in the smoker/non-smoker ratio was analyzed by Chi2 test. Differences in mean monthly ITA–PSV values in each group were assessed by a two-tailed Student t-test for paired data. Values of p < 0.05 were considered to be statistically significant.
T. Nagasaki et al. / Biomedicine & Pharmacotherapy 64 (2010) 113–117
3. Results 3.1. ITA–PSV and clinical variables in women with GD and in normal subjects The clinical characteristics of women with GD determined immediately after delivery, in normal non-pregnant women, and in normal women after delivery are shown in Table 1. All the women with GD had a normal delivery in an euthyroid state. There were no significant differences in age and smoker/non-smoker ratio among the three groups, and serum TSH, FT4 and FT3 levels and serum markers of GD activity such as TRAb and TSAb did not differ significantly between the GD patients and the normal subjects. ITA–PSV and thyroid volume were significantly higher in the patients compared to normal subjects (p < 0.0001): ITA–PSV ranged from 8.8 to 69.3 cm/s in the patients, from 8.1 to 30.0 cm/s in normal non-pregnant women, and from 10.2 to 26.7 cm/s in normal women after delivery. 3.2. Clinical variables in patients with or without relapse of GD Twenty-three (54.8%) of the 42 patients exhibited GD relapse and 19 (45.2%) remained in an euthyroid state. Relapse occurred in seven patients within one month after delivery, in five within two months, and in 11 within three months. A comparison of variables immediately after delivery between patients with and without relapse is shown in Table 2. Age and duration of GD did not differ significantly between the two groups and there was no significant difference in ITA–PSV, FT3, FT4, TSH, TRAb or TSAb. However, the smoker/non-smoker ratio and thyroid volume were significantly higher in patients with relapse of GD compared to those without relapse (10/13 vs. 0/19, p < 0.0001; 24280.3 2280.9 vs. 19670.0 2103.7, p = 0.046). 3.3. Time course of ITA–PSV in patients with and without postpartum relapse of GD The time courses of changes of ITA–PSV in patients with GD during the postpartum period are shown in Fig. 2. The time of relapse, detected by an increase of serum FT4 to above the normal upper limit or the suppression of TSH to below the normal lower limit, was defined as the zero time point. As shown in Fig. 2(a), the seven patients (30.4%) who relapsed within one month after delivery showed simultaneous ITA–PSV elevation. In the five patients (21.7%) who relapsed within two months, an increase in ITA–PSV was evident one month before relapse, and of the 11 patients (47.8%) who relapsed within three months, seven and three showed an apparent increase in ITA–PSV at one and two
115
months prior to relapse, respectively. Only one patient had no increase of ITA–PSV. A statistical analysis of the changes in the ITA– PSV ratio is shown in Fig. 2(b). The mean ITA–PSV ratio showed a significant increase at least one month before GD relapse (1.00 0.00 at –3 months before relapse vs. 1.46 0.12 at –1 month, p = 0.010; 1.00 0.00 at –3 months vs. 1.77 0.13 at relapse, p = 0.0048). Monthly measurement of ITA–PSV did not allow prediction of GD relapse that occurred within one month after delivery, but did enable prediction of relapse at more than one month after delivery in 15 out of 16 patients. In patients without relapse, the ITA–PSV ratio did not change during the three months after delivery (Fig. 1(d)). 3.4. Time-course of changes in ITA–PSV in normal women Time courses of changes in ITA–PSV in 32 non-pregnant normal women and 10 normal women after delivery are shown in Fig. 3. In the non-pregnant women, there were no significant changes in individual ITA–PSV ratios (Fig. 3(a)) or in the mean ratio (Fig. 3(b)) over a three-month period. Similarly, in normal women in the three months after delivery there were no individual changes in ITA–PSV ratios (Fig. 3(c)) or in the mean ITA–PSV ratio (Fig. 3(d)). 4. Discussion We demonstrated that the ITA–PSV value is a useful predictor of GD relapse in pregnant patients during the postpartum period, since an increase in ITA–PSV was detected at least one month before relapse in 15 of 23 (65.2%) patients. Furthermore, seven of the 23 patients (30.4%) had a GD relapse within one month after delivery, which cannot be detected by monthly measurement of ITA–PSV after delivery. Therefore, weekly measurement of ITA– PSV may be appropriate in high-risk cases. Of the 23 patients who relapsed, 22 (93.8%) showed apparent increases in ITA–PSV compared to the value just after delivery, whereas there was no significant change in ITA–PSV in patients without relapse (Fig. 2). These findings suggest that ITA–PSV is a very useful marker for prediction of postpartum GD relapse. ITA–PSV and serum levels of thyroid hormones, TRAb and TSAb just after delivery did not differ significantly between patients with and without GD relapse. This suggests that the profound suppression of immune function during pregnancy (2) may mask the difference in GD activity in these patients. Therefore, measurement of these parameters just after delivery cannot distinguish between patients who are likely to relapse and those who will maintain a euthyroid state. Ten of the 23 relapsed patients (43.4%) were smokers, whereas all 19 patients who did not relapse were non-smokers. A tendency for smokers with GD to
Table 1 Clinical characteristics of pregnant patients with Graves’ disease and control subjects.
Number of subjects Duration of GD (years) Age (years) Smoker/ non-smoker FT4 (pmol/l) [9.01–24.45] FT3 (pmol/l) [4.00–7.70] TSH (mIU/l) [0.4–4.7] TRAb (%) [< 15] TSAb (%) [< 179] ITA–PSV (cm/s) Thyroid volume (mm3)
GD patients
Normal non-pregnant women
Normal women after delivery
42 3.0 0.68 34.7 0.92 10/32 16.0 0.74 4.7 0.16 2.1 0.52 1.2 1.1 112 16.8 32.4 3.4a 22274.3 1902.2a
32
10
35.7 1.2 8/24 15.1 0.55 4.5 0.11 1.8 0.40 –0.12 1.3 108.2 6.9 24.2 2.0 1234.1 1002.6
33.0 1.3 2/8 15.5 0.82 4.4 0.20 1.9 0.50 -0.14 1.6 106.5 12.2 25.0 2.4 1308.2 988.7
P ns ns ns ns ns ns ns ns < 0.0001 < 0.0001
Data are expressed as means SE. Differences in clinical factors among patients with GD, normal non-pregnant women, and normal women after delivery were examined using a two-tailed multiple t-test with a Bonferroni correction. The difference in the smoker/non-smoker ratio was analyzed by Chi2 test. ns: not significant among the three groups; GD: Graves’ disease; ITA–PSV: peak systolic velocity in the inferior thyroid artery. a Significant compared with the other two groups
T. Nagasaki et al. / Biomedicine & Pharmacotherapy 64 (2010) 113–117
116
Table 2 Differences in clinical variables at delivery between patients with Graves’ disease (GD) who did and did not relapse.
Number of subjects Age (years) Duration of GD (years) smoker/non-smoker ratio FT4 (pmol/l) [9.01–24.45] FT3 (pmol/l) [4.00–7.70] TSH (mIU/l) [0.4–4.7] TRAb (%) [< 15] TSAb (%) [< 179] ITA–PSV (cm/s) Thyroid volume (mm3)
Relapse
Non-relapse
P
23 34.1 1.2 3.73 0.91 10/13 16.4 1.0 4.80 0.24 1.65 0.64 0.89 1.9 116 20.2 30.2 3.3 24,280.3 2280.9
19 33.9 1.5 2.09 0.54 0/19 15.5 1.12 4.56 0.20 2.61 0.86 1.76 0.9 110 17.9 35.5 4.5 19,670.0 2103.7
ns ns < 0.0001 ns ns ns ns ns ns 0.046
Data are expressed as means SE. Differences between groups were examined using a Mann-Whitney U test for assessment of medians. The difference in the smoker/nonsmoker ratio was analyzed by Chi2 test; ns: not significant.
relapse after delivery has been reported previously [1,10]. The thyroid volume was also higher in patients who relapsed, but this parameter is not clinically useful for prediction of patients who are likely to relapse. The seven patients (30.4%) who relapsed within one month after delivery showed simultaneous ITA–PSV elevation. These patients had a mean age of 35.4 1.9 years old, and therefore were neither young nor old, and a smoker/non-smoker ratio of 3/4. The small number of patients makes a statistical comparison difficult, but the seven patients had a high mean ITA–PSV (41.2 9.0 cm/s) and a large thyroid volume (30,735.5 4498.5 mm3) compared to the 16 patients who relapsed at a later time, without high TRAb and TSAb titers (–1.0 4.3% and 110 42.6%, respectively) immediately after delivery. Early relapse cannot be predicted by monthly measurement of ITA–PSV and there is a need for more frequent ITA–PSV measurement, especially in high-risk patients such as those with a smoking habit, a high baseline ITA–PSV, or severe goiter.
Fig. 2. Time course of ITA–PSV ratios in the postpartum period in patients who did (left) and did not (right) relapse. (a) Individual changes in the ITA–PSV ratio in patients who relapsed. The time of relapse was defined as the zero time point. Triangles with a dotted line, squares with a dotted line, and open circles with a solid line indicate patients who relapsed within one, two and three months after delivery, respectively. (b) Means SE for patients who relapsed, showing a significant increase in the ITA–PSV ratio at least one month before relapse. (c) Individual changes in the ITA–PSV ratio in patients who did not relapse. Three months after delivery was defined as the zero time point. (d) The mean ITA– PSV ratio in non-relapse patients showed no significant change.
Fig. 3. Time courses of ITA–PSV ratios in 32 normal non-pregnant women (left) and 10 normal women after delivery (right). (a) Individual changes in the ITA–PSV ratio in nonpregnant women. Three months after the first measurement was defined as the zero time point. (b) The mean ITA–PSV ratio in non-pregnant women showed no significant change. (c) Individual changes in the ITA–PSV ratio in normal women after delivery. Three months after delivery was defined as the zero time point. (d) The mean ITA–PSV ratio in normal women after delivery showed no significant change.
T. Nagasaki et al. / Biomedicine & Pharmacotherapy 64 (2010) 113–117
A significant increase in ITA–PSV was seen in almost all cases one month prior to relapse. The mechanism underlying this increase of ITA–PSV is unclear, but these data suggest that an increase of thyroid blood flow is the first event leading to development of hyperthyroidism. The increased thyroid blood flow might result in an increase of autoimmune-antibody exposure to the TSH receptor, despite the absence of changes in the autoimmune-antibody titer and thyroid volume. Even subclinical hyperthyroidism is known to cause an increase in cardiac output [11], which might lead to an increase in blood flow in the common carotid artery. However, ITA–PSV elevation preceded suppression of TSH and we found no increase of blood flow in the common carotid artery with ITA–PSV elevation in an euthyroid state (data not shown). These findings further indicate that an increase of thyroid blood flow is specific for patients with GD. The reason why one of the patients that relapsed did not show an increase in ITA–PSV is unclear. This patient was a 29-year-old woman with no smoking habit. She did not have a high level of ITA–PSV (18.2 cm/s) immediately after delivery and had a smaller thyroid volume (18,469.4 mm3) than the mean, although this was within –1 SD of the mean volume in the relapsed patients (24531.3 10133 mm3). However, she had a high TSAb titer (158.2%) that was > +1 SD of the mean in the relapsed patients (116 40.2%), but within the normal limit in the assay kit (< 179%). Her TRAb titer (4.9%) was within the mean 1 SD of the relapsed patients (0.89 5.8%). This suggests that TSAb might cause GD relapse independently of an elevation of thyroid blood flow. Elevation of IgE in an untreated hyperthyroid GD state has been associated with the severity of GD [12,13] and a recent study of a Japanese population showed that serum levels of IgE are increased significantly in GD patients and are correlated with the recurrence rate of hyperthyroidism after ATD treatment [14]. These reports suggest that serum IgE should be measured in parallel to ITA–PSV in the postpartum period. There is currently no method to predict postpartum GD relapse. Hidaka et al. reported that women at high risk of postpartum onset of Graves’ thyrotoxicosis can be identified early in pregnancy by detection of TSAb [3], and TSAb is also useful for prediction of short-term relapse at the end of ATD treatment [15]. However, measurement of TSAb during the postpartum period cannot predict GD relapse, since elevation of the autoimmune antibody titer and recurrence occur almost simultaneously. Therefore, ITA– PSV measurement during the postpartum period may be more useful, especially for GD patients with a high TSAb titer early in pregnancy. All standard errors for the ITA–PSV ratio in normal nonpregnant women or in normal women after delivery were within 0.02 and were much lower than in women with GD. This indicates that the stability of thyroid blood flow is independent of monthly changes in women without thyroid disease, and further supports the clinical significance of ITA–PSV measurements over at least a few months in GD patients. We have previously reported the clinical usefulness of ITA–PSV measurement in GD patients in an untreated hyperthyroid state [5] or a euthyroid state before withdrawal of ATD [5]. Unlike 123I thyroidal uptake, measurement of ITA–PSV is easy to perform without exposure to radiation, and the CV of the measurement is
117
small enough to allow monitoring of the time course during highrisk periods for GD relapse, such as the postpartum period. Furthermore, the ITA–PSV value can be obtained instantly during measurement. We conclude that the results of the current study support the hypothesis that monthly measurement of ITA–PSV after delivery in remitted euthyroid women with GD may assist in prediction of GD relapse during the postpartum period.
5. Conflicts of interest We declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the reported research. This research was not funded by a specific grant from any funding agency in the public, commercial or not-for-profit sector. References [1] Glinoer D, de Nayer P, Bex M. Effects of l-thyroxine administration, TSHreceptor antibodies and smoking on the risk of recurrence in Graves’ hyperthyroidism treated with antithyroid drugs: a double-blind prospective randomized study. Eur J Endocrinol 2001;144:475–83. [2] Amino N, Tada H, Hidaka Y, Izumi Y. Postpartum autoimmune thyroid syndrome. Endocr J 2000;47:645–55. [3] Hidaka Y, Tamaki H, Iwatani Y, Tada H, Mitsuda N, Amino N. Prediction of postpartum Graves’ thyrotoxicosis by measurement of thyroid stimulating antibody in early pregnancy. Clin Endocrinol 1994;41:15–20. [4] Bogazzi F, Bartalena L, Brogioni S, Burelli A, Manetti L, Tanda ML, et al. Thyroid vascularity and blood flow are not dependent on serum thyroid hormone levels: studies in vivo by color flow Doppler sonography. Eur J Endocrinol 1999;140:452–6. [5] Ueda M, Inaba M, Kumeda Y, Nagasaki T, Hiura Y, Tahara H, et al. The significance of thyroid blood flow at the inferior thyroid artery as a predictor for early Graves’ diseases relapse. Clin Endocrinol 2005;63:657–62. [6] Nagasaki T, Inaba M, Kumeda Y, Fujiwara-Ueda M, Hiura Y, Ishikawa T, et al. Significance of thyroid blood flow as a predictor of methimazole sensitivity in untreated hyperthyroid patients with Graves’ disease. Biomed Pharmacother 2007;61:472–6. [7] Girgensohn S, Liedtke R, Balzer G, Castor S, Hauser M. Performance of the ACS: CentaurTM high-capacity, random-access immunoassay system. Clin Lab 1997;43:975–83. [8] Smith BR, Hall R. Thyroid-stimulating immunoglobulins in Graves’ disease. Lancet 1974;2:427–31. [9] Yoshikawa N, Nishikawa M, Horimoto M, Uno C, Taniguchi N, Inada M. Activity of thyroid stimulating antibody and thyroid stimulation blocking antibody. Endocrinol Jpn 1989;36:55–63. [10] Murakami Y, Takamatsu J, Sakane S, Kuma K, Ohsawa N. Changes in thyroid volume in response to radioactive iodine for Graves’ hyperthyroidism correlated with activity of thyroid-stimulating antibody and treatment outcome. J Clin Endocrinol Metab 1996;81:3257–60. [11] Danzi S, Klein I. Thyroid hormone-regulated cardiac gene expression and cardiovascular disease. Thyroid 2002;12:467–72. [12] Sato A, Takemura Y, Yamada T, Ohtsuka H, Sakai H, Miyahara Y, et al. A possible role of immunoglobulin E in patients with hyperthyroid Graves’ disease. J Clin Endocrinol Metab 1999;84:3602–5. [13] Yamada T, Sato A, Komiya I, Nishimori T, Ito Y, Terao A, et al. An elevation of serum immunoglobulin E provides a new aspect of hyperthyroid Graves’ disease. J Clin Endocrinol Metab 2000;85:2775–8. [14] Komiya I, Yamada T, Sato A, Kouki T, Nishimori T, Takasu N. Remission and recurrence of hyperthyroid Graves’ disease during and after methimazole treatment when assessed by IgE and interleukin 13. J Clin Endocrinol Metab 2001;86:3540–4. [15] Massart C, Orgiazzi J, Maugendre D. Clinical validity of a new commercial method for detection of TSH-receptor binding antibodies in sera from patients with Graves’ disease treated with antithyroid drugs. Clin Chim Acta 2001;304: 39–47.
Original article
Thyroid blood flow as a useful predictor of relapse of Graves’ disease after normal delivery in patients with Graves’ disease Toshiki Nagasaki a, Masaaki Inaba a,*, Misako Fujiwara-Ueda a, Junko Nishio b, Yasuro Kumeda a, Yoshikazu Hiura a, Hideki Tahara a, Eiji Ishimura a, Osamu Ishiko b, Yoshiki Nishizawa a a b
Department of Metabolism, Endocrinology and Molecular Medicine, Osaka City University Graduate School of Medicine, 1-4-3, Asahi-machi, Abeno-ku, Osaka 545-8585, Japan Department of Obstetrics and Gynecology, Osaka City University Graduate School of Medicine, 1-4-3, Asahi-machi, Abeno-ku, Osaka 545-8585, Japan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 April 2009 Accepted 22 April 2009 Available online 20 October 2009
Objective: Measurement of the peak systolic velocity (PSV) in the inferior thyroid artery (ITA) before withdrawal of an anti-thyroid drug (ATD) is useful for predicting relapse of Graves’ disease (GD). We further investigated whether the ITA–PSV can be used for prediction of GD relapse after delivery in euthyroid women with GD who stopped ATD administration during mid- to late pregnancy. Patients and methods: ITA–PSV was monitored monthly for three months after delivery in 42 women with GD aged from 24 to 45 years old (mean SE: 34.7 0.92 years old) who met the above criteria. To confirm the stability of the measurement, ITA–PSV was also measured monthly in 32 age-matched nonpregnant normal women and for three months after delivery in 10 age-matched women. Results: ITA–PSV and thyroid volume were higher in women with GD immediately after delivery compared to normal women, but the levels of TSH receptor antibody (TRAb) and thyroid-stimulating antibody (TSAb) did not differ significantly between the two groups. Of the 42 patients, 23 had relapse of GD and the smoker/non-smoker ratio and thyroid volume in these patients immediately after delivery were significantly higher than those in the 19 patients who did not undergo relapse (10/23 vs. 0/19, p < 0.0001; 24280.3 2280.9 vs. 19670.0 2103.7 mm3, p = 0.046), while ITA–PSV, TRAb and TSAb did not differ between the two groups of patients. The ITA–PSV ratio was calculated by dividing each value in the follow-up period by that obtained immediately after delivery. A significant increase in the mean ITA–PSV ratio occurred at least one month before the time of relapse (1.00 0.00 at –3 months before relapse vs. 1.46 0.12 at –1 month, p = 0.010; 1.00 0.00 at –3 months vs. 1.77 0.13 at the time of relapse, p = 0.0048). In contrast, there were no significant changes in this ratio during the follow-up period in nonrelapse patients. Conclusion: Monthly measurement of ITA–PSV after delivery in remitted euthyroid women with GD may assist in early prediction of GD relapse. ß 2009 Elsevier Masson SAS. All rights reserved.
Keywords: Inferior thyroid artery Color Doppler measurement Delivery Hyperthyroidism Graves’ disease Relapse
1. Introduction The overall relapse rate of Graves’ disease (GD) is as high as 30– 50% [1], and women in the postpartum period have a particularly higher frequency of GD relapse due to the profound autoimmune modification that occurs after delivery [2]. However, there is no established method for prediction of postpartum relapse of GD, except for detection of thyroid-stimulating antibodies (TSAb) early in pregnancy [3]. The increased sophistication of ultrasonographic instruments has made it possible to measure blood flow in the inferior thyroid artery (ITA), a major feeding artery in the thyroid (Fig. 1), and this is a useful method for differentiation of various autoimmune thyroid diseases [4]. Moreover, we recently reported
* Corresponding author. E-mail address: [email protected] (M. Inaba). 0753-3322/$ – see front matter ß 2009 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.biopha.2009.04.038
that the peak systolic velocity (PSV) in the ITA might assist in the prediction of early GD relapse after anti-thyroid drug (ATD) withdrawal [5] and that the ITA–PSV in untreated hyperthyroid patients with GD may reflect GD activity and methimazole sensitivity [6]. The aim of the current study was to assess the value of monthly measurement of ITA–PSV by a pulsed Doppler method for predicting relapse of stimulatory hyperthyroidism after normal delivery in pregnant women with GD who maintained an euthyroid state without ATD administration in mid- to late pregnancy.
2. Patients and methods 2.1. Patients Informed consent was obtained from all subjects. Pregnant women with GD aged from 24 to 45 years old (mean SE:
114
T. Nagasaki et al. / Biomedicine & Pharmacotherapy 64 (2010) 113–117
Fig. 1. Duplex Doppler recording of the right inferior thyroid artery in patients with Graves’ disease. A Doppler image was used to determine the appropriate sampling point for pulsed Doppler recordings. The sample volume was positioned in the right inferior thyroid artery at the inflow point to the thyroid (arrow, left). The velocity waveform is displayed above the baseline to indicate the arterial blood flow (right).
34.7 0.92 years old, n = 42) who had maintained an euthyroid state without ATD administration in mid- to late pregnancy were consecutively recruited into the study. ITA–PSV was measured at the time of enrolment and then every month until three months after delivery. Diagnosis of GD in these patients was based on a history of ATD administration during the first trimester of pregnancy. As controls, ITA–PSV was measured monthly for three months in 32 healthy women aged from 25 to 44 (35.7 1.2) years old who were not pregnant and in 10 healthy women aged from 29 to 37 (33.0 1.3) years old for three months after normal delivery. GD relapse was defined as an increase in the serum level of free thyroxine (FT4) above the normal upper range with suppression of serum TSH below the normal lower limit. A diagnosis of hyperthyroidism due to GD relapse rather than postpartum thyroiditis was made on the basis of a positive reaction for TSH receptor antibody (TRAb) or TSAb, or a requirement for ATD therapy to normalize serum levels of FT4 for several months. 2.2. Biochemical parameters for thyroid function and autoantibodies Blood samples drawn just before the ultrasound study after an overnight fast were kept on ice for 1 h and then centrifuged at 1000 g for 10 min. The serum and plasma samples obtained were stored in aliquots at –20 8C until assayed. Measurements were made immediately after thawing. Commercially available highcapacity, random-access immunoassay kits were used to measure FT4, FT3 and TSH levels (Chiron Diagnostics, East Walpole, MA, USA) [7]. TRAb and TSAb were measured by radioreceptor assay using a commercial kit (Cosmic Corp., Tokyo, Japan) [8] and a radioimmunoassay kit (Yamasa, Chiba, Japan) [9], respectively. 2.3. Blood flow in the inferior thyroid artery and thyroid volume Thyroid blood flow was measured in the ITA as previously reported [5,6] (Fig. 1). Ultrasound examinations were performed
using a duplex Doppler apparatus (SSD 2000, Aloka, Tokyo, Japan) with a 5 MHz convex array probe in both the color Doppler and pulsed Doppler modes. The angle correction cursor was parallel to the direction of flow. The peak systolic flow velocity (PSV) of the right ITA, which is automatically calculated by the ultrasound apparatus, was used as an index of ITA thyroid blood flow. All measurements were performed by one examiner (M. F.-U.), who was blinded to the subject characteristics. Thyroid blood flow was determined in the ITA since this artery is a major contributor to thyroid blood flow, identification and measurement are straightforward, and the procedure has a low coefficient of variation (CV) (< 5.2%). The change of blood flow in the ITA was expressed as the change in the ITA– PSV ratio, which was obtained by dividing ITA–PSV at each measurement time point by the value obtained immediately after delivery. In normal non-pregnant women, the ITA–PSV ratio was calculated by dividing ITA–PSV at each time point by the value from the first examination. Thyroid volume was measured by ultrasound and calculated using an ellipsoid model (width length thickness 0.7 for each lobe) [10]. 2.4. Statistical analysis Data are shown as means SE unless otherwise indicated. Statistical analysis was performed with StatView v.5.0 (SAS Institute, Cary, NC). Differences in clinical factors among GD patients, normal non-pregnant women, and normal women after delivery were examined using a two-tailed multiple t-test with a Bonferroni correction. Differences in clinical values between GD patients with and without relapse were examined using a Mann-Whitney U test for assessment of medians. The difference in the smoker/non-smoker ratio was analyzed by Chi2 test. Differences in mean monthly ITA–PSV values in each group were assessed by a two-tailed Student t-test for paired data. Values of p < 0.05 were considered to be statistically significant.
T. Nagasaki et al. / Biomedicine & Pharmacotherapy 64 (2010) 113–117
3. Results 3.1. ITA–PSV and clinical variables in women with GD and in normal subjects The clinical characteristics of women with GD determined immediately after delivery, in normal non-pregnant women, and in normal women after delivery are shown in Table 1. All the women with GD had a normal delivery in an euthyroid state. There were no significant differences in age and smoker/non-smoker ratio among the three groups, and serum TSH, FT4 and FT3 levels and serum markers of GD activity such as TRAb and TSAb did not differ significantly between the GD patients and the normal subjects. ITA–PSV and thyroid volume were significantly higher in the patients compared to normal subjects (p < 0.0001): ITA–PSV ranged from 8.8 to 69.3 cm/s in the patients, from 8.1 to 30.0 cm/s in normal non-pregnant women, and from 10.2 to 26.7 cm/s in normal women after delivery. 3.2. Clinical variables in patients with or without relapse of GD Twenty-three (54.8%) of the 42 patients exhibited GD relapse and 19 (45.2%) remained in an euthyroid state. Relapse occurred in seven patients within one month after delivery, in five within two months, and in 11 within three months. A comparison of variables immediately after delivery between patients with and without relapse is shown in Table 2. Age and duration of GD did not differ significantly between the two groups and there was no significant difference in ITA–PSV, FT3, FT4, TSH, TRAb or TSAb. However, the smoker/non-smoker ratio and thyroid volume were significantly higher in patients with relapse of GD compared to those without relapse (10/13 vs. 0/19, p < 0.0001; 24280.3 2280.9 vs. 19670.0 2103.7, p = 0.046). 3.3. Time course of ITA–PSV in patients with and without postpartum relapse of GD The time courses of changes of ITA–PSV in patients with GD during the postpartum period are shown in Fig. 2. The time of relapse, detected by an increase of serum FT4 to above the normal upper limit or the suppression of TSH to below the normal lower limit, was defined as the zero time point. As shown in Fig. 2(a), the seven patients (30.4%) who relapsed within one month after delivery showed simultaneous ITA–PSV elevation. In the five patients (21.7%) who relapsed within two months, an increase in ITA–PSV was evident one month before relapse, and of the 11 patients (47.8%) who relapsed within three months, seven and three showed an apparent increase in ITA–PSV at one and two
115
months prior to relapse, respectively. Only one patient had no increase of ITA–PSV. A statistical analysis of the changes in the ITA– PSV ratio is shown in Fig. 2(b). The mean ITA–PSV ratio showed a significant increase at least one month before GD relapse (1.00 0.00 at –3 months before relapse vs. 1.46 0.12 at –1 month, p = 0.010; 1.00 0.00 at –3 months vs. 1.77 0.13 at relapse, p = 0.0048). Monthly measurement of ITA–PSV did not allow prediction of GD relapse that occurred within one month after delivery, but did enable prediction of relapse at more than one month after delivery in 15 out of 16 patients. In patients without relapse, the ITA–PSV ratio did not change during the three months after delivery (Fig. 1(d)). 3.4. Time-course of changes in ITA–PSV in normal women Time courses of changes in ITA–PSV in 32 non-pregnant normal women and 10 normal women after delivery are shown in Fig. 3. In the non-pregnant women, there were no significant changes in individual ITA–PSV ratios (Fig. 3(a)) or in the mean ratio (Fig. 3(b)) over a three-month period. Similarly, in normal women in the three months after delivery there were no individual changes in ITA–PSV ratios (Fig. 3(c)) or in the mean ITA–PSV ratio (Fig. 3(d)). 4. Discussion We demonstrated that the ITA–PSV value is a useful predictor of GD relapse in pregnant patients during the postpartum period, since an increase in ITA–PSV was detected at least one month before relapse in 15 of 23 (65.2%) patients. Furthermore, seven of the 23 patients (30.4%) had a GD relapse within one month after delivery, which cannot be detected by monthly measurement of ITA–PSV after delivery. Therefore, weekly measurement of ITA– PSV may be appropriate in high-risk cases. Of the 23 patients who relapsed, 22 (93.8%) showed apparent increases in ITA–PSV compared to the value just after delivery, whereas there was no significant change in ITA–PSV in patients without relapse (Fig. 2). These findings suggest that ITA–PSV is a very useful marker for prediction of postpartum GD relapse. ITA–PSV and serum levels of thyroid hormones, TRAb and TSAb just after delivery did not differ significantly between patients with and without GD relapse. This suggests that the profound suppression of immune function during pregnancy (2) may mask the difference in GD activity in these patients. Therefore, measurement of these parameters just after delivery cannot distinguish between patients who are likely to relapse and those who will maintain a euthyroid state. Ten of the 23 relapsed patients (43.4%) were smokers, whereas all 19 patients who did not relapse were non-smokers. A tendency for smokers with GD to
Table 1 Clinical characteristics of pregnant patients with Graves’ disease and control subjects.
Number of subjects Duration of GD (years) Age (years) Smoker/ non-smoker FT4 (pmol/l) [9.01–24.45] FT3 (pmol/l) [4.00–7.70] TSH (mIU/l) [0.4–4.7] TRAb (%) [< 15] TSAb (%) [< 179] ITA–PSV (cm/s) Thyroid volume (mm3)
GD patients
Normal non-pregnant women
Normal women after delivery
42 3.0 0.68 34.7 0.92 10/32 16.0 0.74 4.7 0.16 2.1 0.52 1.2 1.1 112 16.8 32.4 3.4a 22274.3 1902.2a
32
10
35.7 1.2 8/24 15.1 0.55 4.5 0.11 1.8 0.40 –0.12 1.3 108.2 6.9 24.2 2.0 1234.1 1002.6
33.0 1.3 2/8 15.5 0.82 4.4 0.20 1.9 0.50 -0.14 1.6 106.5 12.2 25.0 2.4 1308.2 988.7
P ns ns ns ns ns ns ns ns < 0.0001 < 0.0001
Data are expressed as means SE. Differences in clinical factors among patients with GD, normal non-pregnant women, and normal women after delivery were examined using a two-tailed multiple t-test with a Bonferroni correction. The difference in the smoker/non-smoker ratio was analyzed by Chi2 test. ns: not significant among the three groups; GD: Graves’ disease; ITA–PSV: peak systolic velocity in the inferior thyroid artery. a Significant compared with the other two groups
T. Nagasaki et al. / Biomedicine & Pharmacotherapy 64 (2010) 113–117
116
Table 2 Differences in clinical variables at delivery between patients with Graves’ disease (GD) who did and did not relapse.
Number of subjects Age (years) Duration of GD (years) smoker/non-smoker ratio FT4 (pmol/l) [9.01–24.45] FT3 (pmol/l) [4.00–7.70] TSH (mIU/l) [0.4–4.7] TRAb (%) [< 15] TSAb (%) [< 179] ITA–PSV (cm/s) Thyroid volume (mm3)
Relapse
Non-relapse
P
23 34.1 1.2 3.73 0.91 10/13 16.4 1.0 4.80 0.24 1.65 0.64 0.89 1.9 116 20.2 30.2 3.3 24,280.3 2280.9
19 33.9 1.5 2.09 0.54 0/19 15.5 1.12 4.56 0.20 2.61 0.86 1.76 0.9 110 17.9 35.5 4.5 19,670.0 2103.7
ns ns < 0.0001 ns ns ns ns ns ns 0.046
Data are expressed as means SE. Differences between groups were examined using a Mann-Whitney U test for assessment of medians. The difference in the smoker/nonsmoker ratio was analyzed by Chi2 test; ns: not significant.
relapse after delivery has been reported previously [1,10]. The thyroid volume was also higher in patients who relapsed, but this parameter is not clinically useful for prediction of patients who are likely to relapse. The seven patients (30.4%) who relapsed within one month after delivery showed simultaneous ITA–PSV elevation. These patients had a mean age of 35.4 1.9 years old, and therefore were neither young nor old, and a smoker/non-smoker ratio of 3/4. The small number of patients makes a statistical comparison difficult, but the seven patients had a high mean ITA–PSV (41.2 9.0 cm/s) and a large thyroid volume (30,735.5 4498.5 mm3) compared to the 16 patients who relapsed at a later time, without high TRAb and TSAb titers (–1.0 4.3% and 110 42.6%, respectively) immediately after delivery. Early relapse cannot be predicted by monthly measurement of ITA–PSV and there is a need for more frequent ITA–PSV measurement, especially in high-risk patients such as those with a smoking habit, a high baseline ITA–PSV, or severe goiter.
Fig. 2. Time course of ITA–PSV ratios in the postpartum period in patients who did (left) and did not (right) relapse. (a) Individual changes in the ITA–PSV ratio in patients who relapsed. The time of relapse was defined as the zero time point. Triangles with a dotted line, squares with a dotted line, and open circles with a solid line indicate patients who relapsed within one, two and three months after delivery, respectively. (b) Means SE for patients who relapsed, showing a significant increase in the ITA–PSV ratio at least one month before relapse. (c) Individual changes in the ITA–PSV ratio in patients who did not relapse. Three months after delivery was defined as the zero time point. (d) The mean ITA– PSV ratio in non-relapse patients showed no significant change.
Fig. 3. Time courses of ITA–PSV ratios in 32 normal non-pregnant women (left) and 10 normal women after delivery (right). (a) Individual changes in the ITA–PSV ratio in nonpregnant women. Three months after the first measurement was defined as the zero time point. (b) The mean ITA–PSV ratio in non-pregnant women showed no significant change. (c) Individual changes in the ITA–PSV ratio in normal women after delivery. Three months after delivery was defined as the zero time point. (d) The mean ITA–PSV ratio in normal women after delivery showed no significant change.
T. Nagasaki et al. / Biomedicine & Pharmacotherapy 64 (2010) 113–117
A significant increase in ITA–PSV was seen in almost all cases one month prior to relapse. The mechanism underlying this increase of ITA–PSV is unclear, but these data suggest that an increase of thyroid blood flow is the first event leading to development of hyperthyroidism. The increased thyroid blood flow might result in an increase of autoimmune-antibody exposure to the TSH receptor, despite the absence of changes in the autoimmune-antibody titer and thyroid volume. Even subclinical hyperthyroidism is known to cause an increase in cardiac output [11], which might lead to an increase in blood flow in the common carotid artery. However, ITA–PSV elevation preceded suppression of TSH and we found no increase of blood flow in the common carotid artery with ITA–PSV elevation in an euthyroid state (data not shown). These findings further indicate that an increase of thyroid blood flow is specific for patients with GD. The reason why one of the patients that relapsed did not show an increase in ITA–PSV is unclear. This patient was a 29-year-old woman with no smoking habit. She did not have a high level of ITA–PSV (18.2 cm/s) immediately after delivery and had a smaller thyroid volume (18,469.4 mm3) than the mean, although this was within –1 SD of the mean volume in the relapsed patients (24531.3 10133 mm3). However, she had a high TSAb titer (158.2%) that was > +1 SD of the mean in the relapsed patients (116 40.2%), but within the normal limit in the assay kit (< 179%). Her TRAb titer (4.9%) was within the mean 1 SD of the relapsed patients (0.89 5.8%). This suggests that TSAb might cause GD relapse independently of an elevation of thyroid blood flow. Elevation of IgE in an untreated hyperthyroid GD state has been associated with the severity of GD [12,13] and a recent study of a Japanese population showed that serum levels of IgE are increased significantly in GD patients and are correlated with the recurrence rate of hyperthyroidism after ATD treatment [14]. These reports suggest that serum IgE should be measured in parallel to ITA–PSV in the postpartum period. There is currently no method to predict postpartum GD relapse. Hidaka et al. reported that women at high risk of postpartum onset of Graves’ thyrotoxicosis can be identified early in pregnancy by detection of TSAb [3], and TSAb is also useful for prediction of short-term relapse at the end of ATD treatment [15]. However, measurement of TSAb during the postpartum period cannot predict GD relapse, since elevation of the autoimmune antibody titer and recurrence occur almost simultaneously. Therefore, ITA– PSV measurement during the postpartum period may be more useful, especially for GD patients with a high TSAb titer early in pregnancy. All standard errors for the ITA–PSV ratio in normal nonpregnant women or in normal women after delivery were within 0.02 and were much lower than in women with GD. This indicates that the stability of thyroid blood flow is independent of monthly changes in women without thyroid disease, and further supports the clinical significance of ITA–PSV measurements over at least a few months in GD patients. We have previously reported the clinical usefulness of ITA–PSV measurement in GD patients in an untreated hyperthyroid state [5] or a euthyroid state before withdrawal of ATD [5]. Unlike 123I thyroidal uptake, measurement of ITA–PSV is easy to perform without exposure to radiation, and the CV of the measurement is
117
small enough to allow monitoring of the time course during highrisk periods for GD relapse, such as the postpartum period. Furthermore, the ITA–PSV value can be obtained instantly during measurement. We conclude that the results of the current study support the hypothesis that monthly measurement of ITA–PSV after delivery in remitted euthyroid women with GD may assist in prediction of GD relapse during the postpartum period.
5. Conflicts of interest We declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the reported research. This research was not funded by a specific grant from any funding agency in the public, commercial or not-for-profit sector. References [1] Glinoer D, de Nayer P, Bex M. Effects of l-thyroxine administration, TSHreceptor antibodies and smoking on the risk of recurrence in Graves’ hyperthyroidism treated with antithyroid drugs: a double-blind prospective randomized study. Eur J Endocrinol 2001;144:475–83. [2] Amino N, Tada H, Hidaka Y, Izumi Y. Postpartum autoimmune thyroid syndrome. Endocr J 2000;47:645–55. [3] Hidaka Y, Tamaki H, Iwatani Y, Tada H, Mitsuda N, Amino N. Prediction of postpartum Graves’ thyrotoxicosis by measurement of thyroid stimulating antibody in early pregnancy. Clin Endocrinol 1994;41:15–20. [4] Bogazzi F, Bartalena L, Brogioni S, Burelli A, Manetti L, Tanda ML, et al. Thyroid vascularity and blood flow are not dependent on serum thyroid hormone levels: studies in vivo by color flow Doppler sonography. Eur J Endocrinol 1999;140:452–6. [5] Ueda M, Inaba M, Kumeda Y, Nagasaki T, Hiura Y, Tahara H, et al. The significance of thyroid blood flow at the inferior thyroid artery as a predictor for early Graves’ diseases relapse. Clin Endocrinol 2005;63:657–62. [6] Nagasaki T, Inaba M, Kumeda Y, Fujiwara-Ueda M, Hiura Y, Ishikawa T, et al. Significance of thyroid blood flow as a predictor of methimazole sensitivity in untreated hyperthyroid patients with Graves’ disease. Biomed Pharmacother 2007;61:472–6. [7] Girgensohn S, Liedtke R, Balzer G, Castor S, Hauser M. Performance of the ACS: CentaurTM high-capacity, random-access immunoassay system. Clin Lab 1997;43:975–83. [8] Smith BR, Hall R. Thyroid-stimulating immunoglobulins in Graves’ disease. Lancet 1974;2:427–31. [9] Yoshikawa N, Nishikawa M, Horimoto M, Uno C, Taniguchi N, Inada M. Activity of thyroid stimulating antibody and thyroid stimulation blocking antibody. Endocrinol Jpn 1989;36:55–63. [10] Murakami Y, Takamatsu J, Sakane S, Kuma K, Ohsawa N. Changes in thyroid volume in response to radioactive iodine for Graves’ hyperthyroidism correlated with activity of thyroid-stimulating antibody and treatment outcome. J Clin Endocrinol Metab 1996;81:3257–60. [11] Danzi S, Klein I. Thyroid hormone-regulated cardiac gene expression and cardiovascular disease. Thyroid 2002;12:467–72. [12] Sato A, Takemura Y, Yamada T, Ohtsuka H, Sakai H, Miyahara Y, et al. A possible role of immunoglobulin E in patients with hyperthyroid Graves’ disease. J Clin Endocrinol Metab 1999;84:3602–5. [13] Yamada T, Sato A, Komiya I, Nishimori T, Ito Y, Terao A, et al. An elevation of serum immunoglobulin E provides a new aspect of hyperthyroid Graves’ disease. J Clin Endocrinol Metab 2000;85:2775–8. [14] Komiya I, Yamada T, Sato A, Kouki T, Nishimori T, Takasu N. Remission and recurrence of hyperthyroid Graves’ disease during and after methimazole treatment when assessed by IgE and interleukin 13. J Clin Endocrinol Metab 2001;86:3540–4. [15] Massart C, Orgiazzi J, Maugendre D. Clinical validity of a new commercial method for detection of TSH-receptor binding antibodies in sera from patients with Graves’ disease treated with antithyroid drugs. Clin Chim Acta 2001;304: 39–47.