Anda di halaman 1dari 7

Graves disease in pregnancy: Prospective evaluation of a selective invasive treatment protocol

Zohar Nachum, MD,a Yardena Rakover, MD,b Ehud Weiner, MD,a and Eliezer Shalev, MDa Afula and Haifa, Israel
OBJECTIVE: Graves disease in pregnancy carries a risk of fetal thyrotoxicosis from the transplacental transfer of thyroid-stimulating antibodies or fetal hypothyroidism from transplacental transfer of antithyroid drugs and thyroid-blocking antibodies. STUDY DESIGN: From 1991 through 2002, all pregnant women with Graves disease underwent follow-up evaluations that included serial thyroid-stimulating antibody level, thyroid function, and ultrasound examinations. Umbilical blood sampling was recommended if the thyroid-stimulating antibody level was abnormally high or if fetal tachycardia, goiter, intrauterine growth retardation, or hydrops were present. For fetal hyperthyroidism, the mother received antithyroid drugs; for fetal hypothyroidism, maternal antithyroid treatment was reduced, and thyroxine was injected into the amniotic sac. RESULTS: Of 40,000 deliveries, 24 pregnancies (26 fetuses) occurred in 18 women with Graves disease. Nine of 14 mothers with positive findings elected umbilical blood sampling. In 4 of the mothers, the results were normal. Hyperthyroidism and hypothyroidism were diagnosed in 2 and 3 fetuses, respectively. All the fetuses were treated successfully by the protocol with up to four repeated umbilical blood samplings. No complications were recorded in any of the 20 umbilical blood sampling. In the 5 patients who had only elevated thyroid-stimulating antibody levels and who did not elect umbilical blood sampling, sonographic findings remained normal up to term, and the newborn infants were normal. One of 12 children (in whose case we did not recommend umbilical blood sampling) was born with transient hypothyroidism caused by maternal propylthiouracil treatment. All children, whose cases were followed for up to 9 years, are normal. CONCLUSION: In women with Graves disease, umbilical blood sampling in selected cases may improve the control of fetal thyroid function. (Am J Obstet Gynecol 2003;189:159-65.)

Key words: Graves disease, pregnancy, treatment protocol, umbilical blood sampling, ultrasound scan

In the United States, Graves disease affects 1 in 500 pregnancies and accounts for 95% of the cases with hyperthyroidism during pregnancy.1 Neonatal Graves disease occurs in 1.5% to 12% of infants who are born to affected mothers.2 In affected infants, a mortality rate of 12% to 16% has been reported.3,4 Furthermore, the fetus is at risk of thyrotoxicosis from the transplacental transfer of thyroid-stimulating antibodies (TSAb) and at risk for hypothyroidism that was due to the transplacental transfer of antithyroid drugs and thyroid-blocking antibodies

From the Department of Obstetrics and Gynecologya and the Endocrine Pediatric Unit,b HaEmek Medical Center, and the Rappaport Faculty of Medicine, TechnionIsrael Institute of Technology. Received for publication August 9, 2002; revised October 15, 2002; accepted January 31, 2003. Reprint requests: E. Shalev, MD, Department of Obstetrics and Gynecology, HaEmek Medical Center, Afula, 18101, Israel. E-mail: shalev_e@clalit.org.il 2003, Mosby, Inc. All rights reserved. 0002-9378/2003 $30.00 + 0 doi:10.1067/mob.2003.321

(TBAb).5,6 When TSAb were present, neonatal hyperthyroidism and hypothyroidism occurred in 38% and 29% of cases, respectively.7 Fetal hyperthyroidism may cause tachycardia, cardiac failure, hydrops, intrauterine growth restriction (IUGR), goiter,2 and serious neurologic sequelae that may appear in up to two thirds of cases.8 Fetal hypothyroidism may lead to IUGR, goiter, and impaired psychomotor development, even if immediate neonatal treatment is initiated.9 The goiter in itself may cause obstetric and neonatal difficulties.2 Umbilical blood sampling (UBS) has become a simple procedure.10 Data on fetal thyroid development and function has been elaborated.11,12 However, the use of UBS for the treatment of Graves disease has been reported only sporadically.5,6,13-19 In all these case reports, the procedure was performed when the fetuses were affected severely already, with varying degrees of short-term success. We conducted a prospective study to evaluate the impact on neonatal outcome of a protocol for the treatment of Graves disease in pregnancy with the use of UBS for predetermined indications. 159

160 Nachum et al

July 2003 Am J Obstet Gynecol

Table. Data for maternal, fetal, and neonatal status


Patient/case 1/1 Maternal TSAb level* 101-374, 804%-1060% Fetal tachycardia + + + + Fetal growth Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Fetal goiter + + + + Timing of UBS (wk) 20 24 28 32 22 27 24 27 30 36 21 24 34 37 31 35 26 33 24 22

2/2 3/3

259% 500%

3/4

470%

4/5 3/6 3/7 5/8 6/9 2/10 7/11 7/12 8/13 9/14 10/15 11/16 12/17 13/18 14/19 15/20 10/21 5/22 16/23 16/24 17/25 18/26

51-61 400% Normal 850% 35 240% 43-60 43-60 60 41 Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal

*Elevated are levels of bioassay >160% or immunoassay above 10 U/L. Fetal heart rate >160 beats/min.

Methods Clinical protocol. All pregnancies of mothers with Graves disease have been treated prospectively over the last 10 years by a protocol that was designed to control the metabolic state of both mother and fetus. Women were consulted, treated, and followed up by a team of maternal-fetal medicine specialists, adult and pediatric endocrinologists, obstetric sonographists, and neonatologists. The evaluation included serial maternal plasma level of TSAb, thyroid-stimulating hormone (TSH) thyroid receptor-blocking antibodies (TBAb) and thyroid function tests, and monthly ultrasound examinations for fetal heart rate, anthropometric measurements, signs of cardiac failure, and the appearance of goiter. UBS was recommended whenever TSAb were abnormally high (>10 U/L) or fetal tachycardia (>160 beats/min), goiter, IUGR (<10th percentile), or hydrops (accumulation of fluid in third-space cavities) was present. The patients signed a specific informed consent in every case of UBS. UBS was performed to measure TSH, free thyroxine (FT4), and TSAb. Concomitant amniotic fluid was taken for TSH and FT4 levels. UBS was not recommended when both TSAb and sonographic findings were normal. Fetal thyroid status served to modify treatment. When fetal TSH and FT4 were indicative of hyperthyroidism, we either began treatment or increased the dose of existing maternal antithyroid drugs (preferably, propylthiouracil). When observation suggested hypothyroidism, we responded by first reducing maternal antithyroid medication. When this was insufficient to alleviate the fetal problem, we administered intra-amniotic thyroxine (500 g, biweekly). UBS was repeated 2 to 4

Volume 189, Number 1 Am J Obstet Gynecol

Nachum et al 161

Fetal thyroid status Hyperthyroidism Hyperthyroidism Hyperthyroidism Hyperthyroidism Hyperthyroidism Normal Hypothyroidism Hypothyroidism Normal Normal Hypothyroidism Normal Hypothyroidism Normal Hypothyroidism Normal Normal Normal Normal Normal

Treatment LT4, after 131I ablation Propylthiouracil 100 mg/d Propylthiouracil 150 mg/d Propylthiouracil 200 mg/d Propylthiouracil 200 mg/d LT4, after 131I ablation Propylthiouracil 100 mg/d LT4 + Propylthiouracil 100 mg/d Propylthiouracil 150 100 mg/d Propylthiouracil 100 50 mg/d Propylthiouracil 50 mg/d Propylthiouracil 50 mg/d Propylthiouracil 100 50 mg/d Propylthiouracil 50 mg/d Propylthiouracil stopped, intra-amniotic LT4 500 g at 34 and 36 wk None Propylthiouracil 100 50 mg/d until stopped None Propylthiouracil stopped, maternal hypothyroidism LT4 LT4, after 131I ablation LT4, after 131I ablation Propylthiouracil 100 50 mg/d until stopped LT4, after 131I ablation Propylthiouracil 150 mg/d Propylthiouracil 150 mg/d LT4, after 131I ablation LT4, after 131I ablation Propylthiouracil 100 mg/d Propylthiouracil 50 mg/d Propylthiouracil 100 mg/d Propylthiouracil 100 mg/d Propylthiouracil 50 mg/d Propylthiouracil 50 mg/d None LT4, after 131I ablation LT4, after 131I ablation LT4, after 131I ablation None Propylthiouracil 100 50 mg/d until stopped

Neonatal thyroid status

Follow-up period 5y

At 6 d, thyrotoxicosis treated, propylthiouracil + propanolol for 3 wk From age 1 mo - normal

Normal Normal

6y 9y 9y

Normal

Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Normal Mild transient hypothyroidism, from age 2 d to normal Normal Normal Normal Normal Normal Normal Normal Normal Normal

14 mo 6y 2y 6y 1y 5y 4y 4y 4y 3y 8y 5y 5y 4y 4y 18 mo 6y 3y 3y 3y 15 mo 1y

weeks later to evaluate the effect of therapy in the fetus and to modify treatment accordingly. If the fetus was euthyroid and sonographic findings were normal, UBS was not repeated. During delivery, cord blood was sampled for TSH and FT4. Newborn infants were evaluated by a neonatologist and a specialist in pediatric endocrinology for clinical symptoms and signs of hyperthyroidism or hypothyroidism. Thyroid function tests were repeated at the age of 3 days, and again later, if the results were abnormal. Treatment was given accordingly. Follow-up examinations were preformed by a pediatric endocrinologist. Laboratory methods. Umbilical blood and amniotic fluid TSH were measured by time-resolved fluoroimmunoassay with the use of a commercial kit (DELFIA-Wallac, Turku, Finland), which is a highly sensitive method

(detection limit, 0.005 mU/L). Maternal and newborn TSH and FT4 levels were measured by enzyme-linked immunosorbent assay with a commercial kit (Enzymun; Boehringer Mannheim, Mannheim, Germany). TSAb were measured by biologic assay and immunoassay. The biologic assay measured the capacity of a serum immunoglobulin preparation to stimulate cyclic adenosine monophosphate in cultured human thyroid cells, as previously described.20 The immunoassay measured TSAb plasma levels with a commercial kit (RIA, Cis; Bio International, Gif-sur-Yvette, France). TBAb were assayed by measurement of the degree of inhibition of TSH-stimulated cyclic adenosine monophosphate production in cultured human thyroid cells by a serum immunoglobulin preparation, as described in detail previously.21 A normal reference value for amniotic fluid TSH and FT4 was established

162 Nachum et al

July 2003 Am J Obstet Gynecol

Fig 1. Cord blood values in cases 1 through 9. Values for TSH (A) and FT4 (B) are given with 5% to 95% CI for mean of normal values (linear regression for TSH concentrations: r = 0.32, n = 78, P =.005; linear regression for FT4 concentrations: r = 0.58; n = 78; P < .0001). Closed diamonds, Case 1; closed squares, case 2; closed circles, case 3; closed triangles, case 4; open circles, case 5; open diamonds, case 6; open triangles, case 7; open squares, case 8; gray squares, case 9.

Fig 2. Maternal plasma and fetal cord blood values (with 95% CI [fetal, solid lines; maternal, broken lines]) for TSH (A) and FT4 (B) in case 1. The table below the figures summarizes the concomitant sonographic data and maternal treatment. Closed diamonds, Fetal values; open diamonds, maternal values.

in our laboratory with the use of amniotic fluid samples that were taken for various clinical indications from pregnancies with normal neonates. The normal reference values for fetal TSH and FT4 are based on 78 fetal samples that were collected by sonographic-guided UBS from nonstressed fetuses for various clinical indications that were not known to interfere with thyroid function. Results In our center, 24 pregnancies of 40,000 deliveries were complicated by Graves disease over a 10-year period (1:1700 pregnancies). Eighteen women were treated during 24 pregnancies with 26 fetuses. The Table shows the maternal, fetal, and neonatal parameters. Two cases were twin pregnancies (cases 11 and 12 and cases 23 and 24).

In 12 fetuses, no criteria for UBS were met (cases 1526); in 14 fetuses, criteria were met (cases 1-14). Of these 14 cases, 9 patients (cases 1-9) elected to undergo the procedure. Fetal TSH and FT4 results are shown in Fig 1. Four patients (cases 6-9) underwent one UBS procedure, which resulted in normal findings and, subsequently, normal babies. In five patients, the findings were abnormal (cases 1-5), and repeat UBS was performed (2-4 times). Fetal hyperthyroidism was diagnosed in two cases, and hypothyroidism was diagnosed in three cases. In the hyperthyroid fetuses, both mothers were hypothyroid because of iodine 131 ablation of the thyroid and were treated with thyroxine (cases 1 and 2). In case 1, UBS was also performed to exclude fetal hypothyroidism that was due to 131I ablation treatment 2 months before conception. In cases 1 and 2, the first UBS was performed at 20 and 22 weeks of gestation, respectively. No abnormal

Volume 189, Number 1 Am J Obstet Gynecol

Nachum et al 163

findings were detected by the noninvasive follow-up examination. Maternal treatment with propylthiouracil was started, and the dose was increased, which resulted in fetal euthyroidism. Case 1, which represents the most severe case of fetal hyperthyroidism in our series, is presented in detail in Fig 2 and has been described in a recent publication.22 It was characterized by exceptionally high levels of TSAb. Fetal tachycardia and goiter, and later bradycardia and goiter, were followed by a good response to propylthiouracil, with the reversal of the goiter, the resumption of normal heart rate, and the delivery of a euthyroid newborn. Transient Graves disease developed in the baby on the third day of life because of the elevated TSAb level and the elimination of propylthiouracil from the neonatal circulation. At age 2 months, the elevated TSAb level normalized. The neonate was treated with propylthiouracil and propanolol for 3 weeks, which resulted in the reversal of symptoms. Follow-up examinations for 5 years revealed a normally developing child. In the hypothyroidic fetuses, one fetus improved with the reduction of the maternal propylthiouracil dose (case 3), one fetus improved with the cessation of propylthiouracil (case 5), and the third fetus (case 4) required two additional treatments with intra-amniotic thyroxine, which resulted in the reversal of the goiter. The last case represents the most severe case of fetal hypothyroidism in our study and is presented in detail in Fig 3. Figs 2 and 3 demonstrate the consideration of both maternal and fetal thyroid status in the tailoring of appropriate treatment. No complications were recorded in any of the 20 UBS procedures. In the five patients who did not elect to undergo proposed UBS (because of elevated TSAb levels only), ultrasound findings remained normal up to term, and the newborn infants were born normal. According to our criteria, there was no indication for UBS in 12 pregnancies. Six of the women did not have elevated TSAb levels and were treated with propylthiouracil. One of the newborn infants of this group was born with mild and transient hypothyroidism that was due to the maternal propylthiouracil treatment (case 17). He also had delayed epiphyseal maturation of the knees. In the 6 patients who did not have elevated TSAb levels and who were not treated with antithyroid drugs, all the babies were born euthyroid. All of the children, who have been followed for up to 9 years, are developing normally. Logarithmic correlation of concomitant fetal TSH and FT4 values was high (r = 0.86). As can also be clearly seen in Figs 1 through 3, this indicates the maturity of the fetal thyroid axis in the second half of pregnancy. All TBAb measurements were negative. Concomitant measurements of amniotic fluid TSH and FT4 levels were not correlated with fetal or maternal serum values. Correlation of concomitant maternal and fetal TSAb levels was very high (r = 0.94)

Fig 3. Maternal plasma and fetal cord blood values (with 95% CI [fetal, solid lines; maternal, broken lines]) for TSH (A) and FT4 (B) in case 4. The table below the figure summarizes concomitant sonographic data and maternal treatment. Closed diamonds, Fetal values; open diamonds, maternal values.

Comment Graves disease in pregnancy is not an uncommon disorder, accounting for most of the cases of hyperthyroidism during pregnancy.1 It has been estimated to affect 1 of 500 pregnancies.1 In our center, it was present in 24 pregnancies of 40,000 deliveries over a 10-year period, which was approximately 1 of 1700 pregnancies. Neonatal Graves disease occurs in 1.5% to 12% of infants who are born to affected mothers2 and in 38% when TSAb levels are elevated.7 Thyroid dysfunction was reported in approximately 17% of cases and overt thyrotoxicosis in 2.6% cases.8 In our series of 26 fetuses and neonates, there were two cases of hyperthyroidism (8%) and three cases of hypothyroidism (12%). Altogether, 19% of cases were affected. Of the 13 cases with elevated TSAb levels, 2 patients (15%) were hyperthyroidic, and 3 patients (23%) hypothyroidic. Both of the mothers with the hyperthyroidic fetuses had been rendered hypothy-

164 Nachum et al

July 2003 Am J Obstet Gynecol

roid previously because of 131I ablation of the thyroid and were treated with thyroxine. This seemed to place the fetuses of these woman in the at-risk group because of the lack of treatment with antithyroid drugs. The lack of proper attention to that risk because the patient is hypothyroidic may lead to fetal hyperthyroidism. Two of six women with elevated TSAb levels and no antithyroid treatment had hyperthyroidic fetuses. Treatment with antithyroid drugs may cause fetal hypothyroidism. Four of 12 women who received propylthiouracil therapy had hypothyroidic fetuses, even with low doses of 50 to 100 mg per day. Fetal hypothyroidism that is due to TBAb level is rare,23 with no case documented in our study. High mortality rate and severe fetal complications were reported in pregnancies that were complicated by maternal Graves disease.2,4,6,8,13-19,22-24 Our protocol enabled us to diagnose five of the six affected fetuses and to control thyroid function antenatally, which included the reversal of goiter in 2 cases. All five fetuses were born euthyroidic. The one case with neonatal Graves disease (case 1) had mild disease that started on day 3 of life and responded well to treatment. This later onset represents the faster clearance of maternally transferred propylthiouracil in the newborn infant compared with TSAb level. The treating team and the parents were well prepared. The only hypothyroidic newborn infant had mild laboratory values of hypothyroidism, which returned to normal on the third day of life. Mild delayed bone maturation resolved spontaneously during the follow-up period. This was 1 of the 12 cases for which we did not recommend UBS, and the mother was one of the six patients who were treated with propylthiouracil. The parents must be informed of this risk. The amount of antithyroid treatment should be reduced to the minimum possible. Long-term follow-up evaluation of all the children is normal. Despite the risk that is involved, the use of UBS improves the outcome of these pregnancies.5,6,13-19,25 UBS carries approximately a 0.5% to 1% risk of fetal complications.10 We estimated the risk of fetal hyperthyroidism and its consequences in cases of elevated TSAb levels (13 cases) or abnormal sonographic findings (1 case) to be high enough to suggest UBS. After consultation, 9 of the 14 women chose to undergo UBS. Children with mild congenital hypothyroidism, if treated promptly, are not affected by long-term mental disability.9 We assume that there is a negligible risk of fetal hyperthyroidism in women with low TSAb levels, and if hypothyroidism develops without sonographic signs, it is mild and transient. When hypothyroidism is severe enough to cause goiter or IUGR, we would recommend UBS. Monitoring fetal heart rate and thyroid size by ultrasonic imaging was suggested for the diagnosis and followup evaluation of fetal thyrotoxicosis.2 However, other

investigators have suggested that these data may be misleading.14,17 We found that normal ultrasound scanning and heart rate findings cannot exclude abnormal fetal thyroid function, as was the case in three fetuses (cases 1, 2, and 5). Furthermore, an abnormal fetal heart rate is not pathognomotic for fetal abnormality. For example, the fetus in case 1 had goiter and bradycardia and was hyperthyroidic. Conversely, cases 3 and 4 were tachycardic and hypothyroidic. Finally, case 7 was tachycardic and euthyroidic. The presence of goiter correlates well with severe thyroid abnormality (cases 1 and 4). This correlation, already described,16,25,26 justifies invasive evaluation. Our results support the notion that IUGR is a late manifestation of sustained fetal thyrotoxicosis.7,14 In our study, as in others, amniotic fluid levels of thyroid hormones were not correlated with fetal hormones.3 As expected, the evaluation of maternal, fetal, and neonatal TSAb revealed that it readily crosses the placenta and can be detected in neonatal plasma as long as 2 months after birth. Because fetal UBS has become a simple procedure, normal reference values for fetal thyroid function have been established.11,12 The coexistence of fetal thyrotoxicosis (cases 1 and 2), elevated fetal TSAb levels and an undetectable level of TSH, and the coexistence of fetal hypothyroidism (cases 3-5) with high TSH and low FT4 confirms that the fetal thyroid at 20 weeks of gestation is capable of responding to TSAb levels. Furthermore, fetal TSH was suppressed by the elevated thyroid hormones and was elevated when FT4 was low. One can assume that the pituitary-thyroid feedback mechanism is functional as early as the week 20 of gestation.23 This conclusion is supported by the high correlation between fetal TSH and FT4 and by the favorable fetal response to treatment. The capability of the fetal pituitary to respond to hypothyroxinemia by increasing TSH production was demonstrated in several published case reports.25,26 In summary, this is the first prospective study to evaluate the impact of a protocol for the treatment of Graves disease in pregnancy on neonatal outcome, with the use of UBS in selected cases. Although caution must be taken because of its limited size, it may serve as a database for the counseling of patients with Graves disease. We suggest adaptation of our treatment protocol, with close serial monitoring of maternal and fetal parameters. We suggest UBS when maternal TSAb levels are high or when sonographic findings of IUGR, cardiomegaly, or goiter are seen, with or without fetal tachycardia. We would start UBS at 20 to 24 weeks of gestation, when the fetal axis is reactive and there is still time to correct the disease. We would repeat the treatment after 2 to 4 weeks when the results are abnormal, to evaluate the effect of the treatment. Neonatal follow-up is necessary, because some infants will need antithyroid therapy. Future studies should include the search for other noninvasive parameters.

Volume 189, Number 1 Am J Obstet Gynecol

Nachum et al 165

We thank Dr Izhar Ben-Shlomo for constructive comments.


REFERENCES

1. Burrow GN. The management of thyrotoxicosis in pregnancy. N Engl J Med 1985;313:562-5. 2. Fisher DA. The thyroid. In: Kaplan SA, editor. Clinical pediatric endocrinology. 2nd ed. Philadelphia: WB Saunders; 1990. p. 87-126. 3. Hollingsworth DR, Mabry CC. Congenital Graves disease: four familial cases with long-term follow-up and perspective. Arch Pediatr Adolesc Med 1976;130:148-55. 4. Zimmerman D, Lteif AN. Thyrotoxicosis in children. Endocrinol Metab Clin North Am 1998;27:109-26. 5. McKenzie JM, Zakarija M. Fetal and neonatal hyperthyroidism and hypothyroidism due to maternal TSH receptor antibodies. Thyroid 1992;2:155-9. 6. Perelman AH, Clemons RD. The fetus in maternal hyperthyroidism. Thyroid 1992;2:225-8. 7. Mitsuda N, Tamaki H, Amino N, Hosono T, Miyai K, Tanizawa O. Risk factors for developmental disorders in infants born to women with Graves disease. Obstet Gynecol 1992;80:359-64. 8. Daneman D, Howard NJ. Neonatal thyrotoxicosis: intellectual impairment and craniosynostosis in later years. J Pediatr 1980; 97:257-9. 9. Van Vliet G. Treatment of congenital hypothyroidism. Lancet 2001;358:86-7. 10. Shalev E, Dan U, Weiner E, Romano S, Giselevitz J, Mashiach S. Prenatal diagnosis using sonographic guided cordocentesis. J Perinat Med 1989;17:393-8. 11. Ballabio M, Nicolini U, Jowett T, Ruiz de Elvira MC, Ekins RP, Rodeck CH. Maturation of thyroid function in normal human foetuses. Clin Endocrinol (Oxf) 1989;31:565-71. 12. Thorpe-Beeston JG, Nicolaides KH, Felton CV, Butler J, McGregor AM. Maturation of the secretion of thyroid hormone and thyroid-stimulating hormone in the fetus. N Engl J Med 1991; 324:532-6. 13. Porreco RP, Bloch CA. Fetal blood sampling in the management of intrauterine thyrotoxicosis. Obstet Gynecol 1990;76:509-12.

14. Wenstrom KD, Weiner CP, Williamson RA, Grant SS. Prenatal diagnosis of fetal hyperthyroidism using funipuncture. Obstet Gynecol 1990;76:513-7. 15. Polk DH. Diagnosis and management of altered fetal thyroid status. Clin Perinatol 1994;21:647-62. 16. Hadi HA, Strickland D. Prenatal diagnosis and management of fetal goiter caused by maternal Graves disease. Am J Perinatol 1995;12:240-2. 17. Wallace C, Couch R, Ginsberg J. Fetal thyrotoxicosis: a case report and recommendations for prediction, diagnosis, and treatment. Thyroid 1995;5:125-8. 18. Treadwell MC, Sherer DM, Sacks AJ, Ghezzi F, Romero R. Successful treatment of recurrent non-immune hydrops secondary to fetal hyperthyroidism. Obstet Gynecol 1996;87:838-40. 19. Abuhamad AZ, Fisher DA, Warsof SL, Slotnick RN, Pyle PG, Wu SY, et al. Antenatal diagnosis and treatment of fetal goitrous hypothyroidism: case report and review of the literature. Ultrasound Obstet Gynecol 1995;6:368-71. 20. Kraiem Z, Glaser B, Pauker J, Sadeh O, Sheinfeld M. Bioassay of thyroid-stimulating immunoglobulin in cryopreserved human thyroid cells: optimization and clinical evaluation. Clin Chem 1988;34:244-9. 21. Kraiem Z, Lahat N, Glaser B, Baron E, Sadeh O, Sheinfeld M. Thyrotrophin receptor blocking antibodies: incidence, characterization and in-vitro synthesis. Clin Endocrinol (Oxf) 1987; 27:409-21. 22. Rakover Y, Weiner E, Mosh N, Shalev E. Fetal pituitary negative feedback at early gestational age. Clin Endocrinol (Oxf) 1999; 50:809-14. 23. Fisher DA. Fetal thyroid function: diagnosis and management of fetal thyroid disorders. Clin Obstet Gynecol 1997;40:16-31. 24. Watson WJ, Fiegen MM. Fetal thyrotoxicosis associated with nonimmune hydrops. Am J Obstet Gynecol 1995;172:1039-40. 25. Van Loon AJ, Derksen JTH, Bos AF, Rouwe CW. In utero diagnosis and treatment of fetal goitrous hypothyroidism, caused by maternal use of propylthiouracil. Prenat Diagn 1995;15:599-604. 26. Davidson KM, Richards DS, Schatz DA, Fisher DA. Successful in utero treatment of fetal goiter and hypothyroidism. N Engl J Med 1991;324:543-6.