| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Feature |
Department of Pediatrics, University of Montréal (R.P., G.V.V.), Montréal, Québec, Canada H3T 1C5; Department of Pediatrics (C.H., P.B.), and Interdisciplinary Research Institute and Department of Medical Genetics (G.V.), Free University of Brussels, 1070 Brussels, Belgium; Department of Pediatrics, University of Sherbrooke (K.K.), Sherbrooke, Québec, Canada J1H 5N4; and Departments of Medicine and Pediatrics, Laval University (F.S., J.H.D.), Québec City, Québec, Canada G1V 4G2
Address all correspondence and requests for reprints to: Dr. Guy Van Vliet, Endocrinology Service, Sainte Justine Hospital, 3175 Sainte Catherine Road, Montréal, Québec, Canada H3T 1C5. E-mail: . gvanvliet{at}justine.umontreal.ca
Abstract
Since the advent of biochemical screening for congenital hypothyroidism, the majority of monozygotic twins reported with thyroid dysgenesis have been discordant, and most were missed on neonatal screening, presumably due to fetal blood mixing. We hypothesized that there may be bias leading to preferential reporting of discordant twins and/or of false negative screening results. Therefore, we performed a systematic search for twins in two congenital hypothyroidism screening centers, Quebec and Brussels, that use a primary TSH approach.
In Quebec, 10 pairs of twins were identified, all discordant for congenital hypothyroidism due to thyroid dysgenesis (4 monozygotic and 4 dizygotic pairs) and dyshormonogenesis (2 dizygotic pairs). The 6 pairs identified in the Brussels database were also all discordant for congenital hypothyroidism due to thyroid dysgenesis (1 monozygotic and 3 dizygotic pairs) and dyshormonogenesis (2 dizygotic pairs). The median increase in TSH between screening and diagnosis was 7-fold in the monozygotic twins vs. 2-fold in matched singletons (P = 0.02), suggesting fetal blood mixing between the twins.
In summary, discordance for thyroid dysgenesis is the rule in monozygotic twins, and fetal blood mixing may result in delayed or missed diagnoses. We therefore conclude that 1) a second sample for congenital hypothyroidism screening at 14 d of age should be considered for all same-sex twins; and 2) thyroid dysgenesis generally results from epigenetic phenomena, early somatic mutations, or postzygotic stochastic events.
CONGENITAL HYPOTHYROIDISM is the most frequent preventable cause of mental deficiency. In iodine-sufficient countries, congenital hypothyroidism in approximately 85% of cases is due to defective differentiation or migration of the embryonic thyroid (thyroid dysgenesis), the cause of which is generally unknown (1). However, a recent survey in France showed that 2% of cases of thyroid dysgenesis are familial, 15-fold higher than what would be expected by chance alone. A model of dominant transmission with variable penetrance was proposed (2). The remaining cases of congenital hypothyroidism have a defect in one of the steps involved in thyroid hormone biosynthesis (dyshormonogenesis), an autosomal recessive condition (3).
Systematic biochemical screening of newborns has been remarkably successful in preventing mental retardation due to congenital hypothyroidism (4), but all possibilities of missing cases need to be considered. A potential cause of a false negative screening result is fetal blood mixing between twins, which can occur in the absence of readily recognizable features of the twin to twin transfusion syndrome.
We recently encountered a pair of monozygotic twin boys who are discordant for thyroid dysgenesis. The affected twin had only a very mild elevation of TSH at screening, possibly as a result of fetal blood mixing. A review of the literature published after systematic neonatal screening was introduced revealed mostly single case reports: seven of eight reported monozygotic twins were discordant for thyroid dysgenesis, and neonatal screening was falsely negative in the affected twin in six of these seven discordant pairs. We hypothesized that a bias may have led to preferential reporting of discordant twins and/or of false negative screening results. We therefore undertook a systematic search for twins in the computerized databases maintained by two large congenital hypothyroidism screening programs in Quebec and Brussels. In these databases monozygotic twins are always discordant for thyroid dysgenesis, and there is evidence of fetal blood mixing.
Subjects and Methods
The index twins were referred as newborns to the endocrinology clinic at the Sainte Justine Hospital in Montreal, Québec, in August 2000. Pathology reports were reviewed for the description of the placenta and its membranes. Monozygosity was confirmed by DNA analysis. The findings in these index cases (see below) prompted a retrospective review of our databases.
Screening procedures
All values refer to TSH and/or T4 values in whole blood. In Québec since 1987 the screening strategy has been based on measuring TSH first. Initially, the cut-off TSH value for a repeat filter paper request was 15 mIU/liter, and that for immediate referral was 50 mIU/liter. This latter value was decreased to 40 mIU/liter in 1988 and to 30 mIU/liter in 1997. In 1998, the screening procedure was altered so that if the initial TSH is more than 15 mIU/liter, a total T4 measurement is performed on the same filter paper, and the newborn is immediately referred if the total T4 value is less than 86 nmol/liter. Between 1987 and 2000, there were 1,164,600 neonates screened in Québec, of whom 392 had congenital hypothyroidism (incidence of 1 in 2971).
In Brussels screening was commenced in 1974 and is based on measuring TSH with a cut-off value of 10 mIU/liter. If TSH is more than 50 mIU/liter, the baby is immediately referred. If TSH is more than 10 mIU/liter, then a total T4 measurement is performed on the same filter paper. A second sample is requested for TSH values between 25–50 mIU/liter or for values between 10–25 mIU/liter if total T4 is less than 64 nmol/liter. To date, the Brussels program has screened 475,000 neonates, of whom 150 have been found to have congenital hypothyroidism (incidence of 1 in 3167).
Upon referral by the screening network, a venous blood sample is obtained for determination of TSH, free T4, T3, antithyroglobulin, and antithyroperoxidase antibodies in the infant and in the mother, and a thyroid radionuclide scan with 99m-technetium is performed in the infant (5).
Identification of twins and assessment of zygosity
The computerized databases were used in both Québec and Brussels to identify twins with congenital hypothyroidism, and the case notes were then reviewed. In addition, a survey of pediatric endocrinologists in the province of Québec and in the Brussels area was performed to check whether any twins with congenital hypothyroidism had been missed since the implementation of the screening procedures described above. The description of the placenta was obtained from the mothers’ charts, including histological examination when available.
We first considered same-sex twins as monozygotic if they had a strikingly similar physical appearance (beyond the neonatal period) (6). We next obtained a description of the placenta. Lastly, we sought consent from all families for analysis of leukocyte DNA. Molecular confirmation of monozygosity was based on 1) a panel of 9 microsatellite loci (Profiler Plus, AmpFISTR, Perkin-Elmer, Zaventem, Belgium), and 2) a multilocus DNA fingerprint by Southern blotting using a minisatellite probe (7). In all cases, genotyping at microsatellites and minisatellites were in agreement and confirmed assignment of monozygosity based on physical appearance (Tables 1
and 2).
|
|
Blood mixing between a hypothyroid and a euthyroid twin would result in a lowering of TSH at screening in the former and, consequently, a greater rise in TSH between screening and diagnosis than in hypothyroid singletons. Therefore, the change in the TSH level between screening and diagnosis (TSHdiag - TSHscr/TSHscr x 100) was compared (by the Mann-Whitney test) between monozygotic twins and singleton newborns with thyroid dysgenesis. This could only be done for the Québec patients because 1) TSH at screening and diagnosis was measured using the same reagents; and 2) TSH at screening was reported as an exact number. There were 111 singletons with ectopy and 36 with athyreosis who met these criteria. From this database, for each of the monozygotic twins in Québec, 4 singletons were selected with matching for the following criteria: 1) TSH at diagnosis (this is a better reflection of severity of hypothyroidism than TSH at screening, because it is further in time not only from the putative fetal blood mixing between twins but also from the well established transplacental transfer of T4 from mother to fetus) (8); 2) age at screening (because the effect of blood mixing between twins probably decreases quickly after separation of the twins’ circulations); and 3) age at diagnosis (because the increase in TSH between screening and diagnosis is also likely to be time dependent). In the monozygotic twins there were 2 with ectopy and 2 with athyreosis (Table 1
), whereas in the singletons these numbers were 9 and 7, respectively.
Results
Index cases
These twin boys (1A and 1B) were born in Quebec in July 2000 at 36 wk gestation after a normal pregnancy sustained by a monochorionic diamniotic placenta. Twin 1A had received phototherapy for neonatal jaundice. He had mildly elevated TSH at screening on d 3 and slightly low total T4 for gestational age. On d 14 plasma TSH was markedly elevated, and free T4 was very low (Table 1). Both twin 1B and the twins’ mother had normal thyroid function. On ultrasound of the neck, twin 1A had no detectable thyroid tissue in the pretracheal area, whereas twin 1B had a normal bilobed gland. Twin 1A had no uptake of 99m-technetium in the lingual, cervical, or mediastinal areas, and antiperoxidase antibodies were undetectable; however, plasma thyroglobulin was detectable (34.5 µg/liter), suggesting that there was some ectopic thyroid tissue too small for detection by nuclear medicine scanning (9). On x-ray of the knee, twin 1A had only a distal femoral ossification center, whereas twin 1B had both a distal femoral and a proximal tibial ossification center. The twins were shown to be monozygous by micro- and minisatellite analyses of leukocyte DNA. Twin 1B was admitted at 3 months of age with a urinary tract infection, and ultrasonography revealed an ectopic left kidney. Twin 1A had a normal renal ultrasound scan.
Review of the databases
In Québec, 10 pairs of twins (4 monozygotic and 6 dizygotic) were identified. They were all discordant for congenital hypothyroidism due to thyroid dysgenesis (in 4 monozygotic and 4 dizygotic) or dyshormonogenesis (2 dizygotic; Table 1). The Brussels database identified 6 pairs of twins (1 monozygotic and 5 dizygotic) also all discordant for congenital hypothyroidism due to thyroid dysgenesis (1 monozygotic and 3 dizygotic) and dyshormonogenesis (2 dizygotic; Table 2). Our survey of pediatric endocrinologists in Québec and Brussels did not reveal any twins with congenital hypothyroidism that had been missed on neonatal screening. No monozygotic twins were concordant for thyroid dysgenesis in either database (see below). None of the mothers had abnormal thyroid function tests, and in none of the families was there more than 1 case of congenital hypothyroidism.
Assessment of zygosity and of the euthyroid twin
Among the Québec patients, three pairs of same-sex twins with differing physical traits but reportedly monochorionic-diamniotic placentas were, in fact, dizygotic on DNA analysis. In one pair with identical physical traits but a dichorionic-diamniotic placenta (where the probability of monozygosity is only 33% (10), monozygosity was established by DNA fingerprinting (Table 1).
In view of a report of a pair of monozygotic twins, one of whom had ectopy and the other hemiagenesis (11), thyroid ultrasound was also performed in the euthyroid member of our monozygotic pairs whenever possible. This was performed and was found to be normal in the unaffected twin of pairs 1, 2, 4 (Québec), and 1 (Brussels).
Assessment of fetal blood mixing
In the four monozygotic pairs for whom screening TSH was reported as an exact number, the median increase in TSH between screening and diagnosis was 7-fold, as opposed to 2-fold in matched singletons (P = 0.02), suggesting fetal blood mixing (Table 3).
|
Our systematic survey of two large congenital hypothyroidism databases identified five pairs of monozygotic twins, all discordant for thyroid dysgenesis. These findings have important implications for our understanding of the molecular pathophysiology of thyroid dysgenesis as well as for the screening programs. In the present report the twins were all identified by the screening procedures (Tables 1 and 2), but there was evidence for fetal blood mixing in the monozygotic pairs (Table 3). This was most striking in our index case (twin 1A); although this neonate was correctly identified by biochemical screening, there would have been a delay in diagnosis if the total T4 had not been measured initially, as it was the combination of the mildly elevated TSH and decreased total T4 on the first filter paper sample that prompted immediate referral to a pediatric endocrinologist. A delay in diagnosis and treatment could have adversely affected his developmental outcome (12, 13). The initial TSH measurement of case 1A was most likely lowered due to transfusion of euthyroid blood from the unaffected twin by fetal blood mixing, but that pair did not demonstrate any of the cardinal features of the twin to twin transfusion syndrome (see below).
Of the seven pairs of monozygotic twins discordant for thyroid dysgenesis reported (14, 15, 16, 17, 18, 19) after the advent of neonatal screening, only one case (18) was correctly identified by routine screening; using the screening methods applicable at that time, the six other monozygotic twins affected with congenital hypothyroidism were missed, and in five of those it was presumably due to fetal blood mixing (Table 4). In our study, there is evidence for fetal blood mixing in monozygotic twins (Table 3), but none was missed with our screening procedures and cut-off values. This is supported by the fact that no missed case of congenital hypothyroidism in a monozygotic twin was identified by our survey of pediatric endocrine clinics in the areas studied.
|
Based on our findings we suggest that to avoid the potential pitfall of subclinical fetal blood mixing, a second sample at 14 d of life should be considered, but we would restrict this practice to same-sex twins because there is no report of a missed case in dizygotic twins. Blood mixing can occur in twins through in utero vascular connections, which are found in at least 70% of monozygotic placentas (22). These vascular connections allow the free transfer of T4 from the euthyroid twin and maintain TSH in the normal range in the hypothyroid twin until a few days after delivery. Our data suggest that fetal blood mixing can occur independently of the twin to twin transfusion syndrome (23) and hence remain unrecognized at the time of delivery. As T4 is estimated to have a biological half-life of 3.6 d in newborns (8), and the disappearance of T4 is reported to follow first order kinetics (24), it would seem reasonable to consider repeat screening at 14 d in same-sex twins to allow treatment to be started within the first month of life. The financial burden of additional neonatal screening tests is currently receiving intense scrutiny (25); however, based on our data and the literature (Table 4), rescreening same-sex twins for primary congenital hypothyroidism is more justified than rescreening all very low birth weight babies (26).
Assignments of zygosity by DNA fingerprinting and examination of placental membranes were in contradiction in 40% of the twin pairs from Québec. This suggests that placental examination is not a reliable basis for determining which twin pairs would need retesting at 14 d (15). As it is difficult to assess the identity of physical traits in the neonatal period, we advocate retesting of all same-sex twins. However, even retesting at 14 d would have missed one of the twins reported in the literature (16) (Table 4). The neonate is generally able to mount a brisk TSH response to primary thyroid insufficiency, as exemplified by acute iodine overload (27); on the other hand, plasma TSH may remain suppressed for a few weeks in infants with neonatal Graves’ disease, even after plasma T4 has become normal (28). The reason why plasma TSH was still not elevated on d 14 in the case reported by De Zegher and Vanderschueren-Lodeweyckx (16) is unclear, but it is noteworthy that this is the only twin pair with a monoamniotic placenta (Table 4). Therefore, even if a rescreening policy is implemented, a high clinical index of suspicion should be kept for same-sex twins.
The results from the screening databases indicate that discordance of monozygotic twins for thyroid dysgenesis is real and does not represent preferential reporting of discordant monozygotic twins because of their "curiosity value" (29). Estimates of the frequency of discordance between twins in the prescreening era are difficult, because of uncertainty regarding zygosity and etiology of hypothyroidism, but also suggest that monozygotic twins are generally discordant (30). Since the advent of neonatal screening and routine scintigraphy, there have been seven pairs of monozygotic twins reported to be discordant, and only one pair concordant (11), for thyroid dysgenesis. In this pair of male twins, the thyroid defects were not identical; twin A had permanent congenital hypothyroidism due to an ectopic thyroid gland in the suprahyoid position, and twin B had transient congenital hypothyroidism due to left thyroid hemiagenesis. In the four euthyroid monozygotic twins tested in our study, two thyroid lobes were identified by ultrasound.
As demonstrated by the cases identified within the Québec and Brussels databases and those reported in the screening literature, discordance for thyroid dysgenesis occurred in 12 of 13 monozygotic twin pairs (92%). This excludes classical Mendelian transmission of recessive alleles accompanied by severe reduction of reproductive fitness as well as a multigenic inheritance, two models that have been proposed to account for the sporadic occurrence of the disease (31). Interestingly, Léger et al. (32) recently reported that minor thyroid developmental abnormalities were identified by ultrasound in only 0.8% of controls, but in about 8% of first degree relatives of children with thyroid dysgenesis. This latter figure seems complementary to our estimate of discordance in monozygotic twins for thyroid dysgenesis. Our interpretation is that this malformation is most commonly due to noninheritable postzygotic events, which could include epigenetic modifications, early somatic mutations, or stochastic developmental events.
It is possible that the postzygotic events themselves, by creating two different cell populations in the inner cell mass, actually stimulate the twinning process itself (10). Moreover, if these two different cell populations in the inner cell mass are unequally distributed between the two embryos, the development of specific organs, such as the thyroid, could differ between monozygotic twins. The stochastic nature of these postzygotic events is highlighted by our index twins, who are also discordant for left-sided renal ectopy.
Consistent with these complex interrelations between monozygotic twinning and organogenesis, there is an increased incidence of congenital malformations in monozygotic twins (33). Establishing whether thyroid dysgenesis itself occurs significantly more often in monozygotic twins than in the general population would require larger numbers of patients.
In summary, our systematic study of two large congenital hypothyroidism databases shows that discordance of monozygotic twins for thyroid dysgenesis is the rule, and that because of fetal blood mixing, specific screening guidelines should be considered for all same-sex twins.
Acknowledgments
The invaluable cooperation of Ms. N. Belanger and Dr. A. Grenier (Québec screening laboratory), of Ms. C. Streydio (Laboratory of Medical Genetics, Brussels, Belgium; for genotyping of the twins), and of the nurse coordinators of the Québec congenital hypothyroidism clinics is gratefully acknowledged. We thank Drs. C. Rodd, C. Polychronakos, and J. De Schepper for providing clinical information about and blood samples from patients under their care; Drs. S. P. Taback and J. F. Connelly for sharing unpublished information; Drs. J. Michaud, L. Oligny, and J.-C. Fouron for helpful discussions; Dr. D. Jaquet for statistical advice; and Drs. D. Jaquet and M. Castanet for critical reading of the manuscript.
Footnotes
Research in thyroid diseases at Sainte Justine Hospital is supported by its Research Center and by Mr. John H. MacBain.
Received December 13, 2001.
Accepted May 8, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  A. Carre, G. Szinnai, M. Castanet, S. Sura-Trueba, E. Tron, I. Broutin-L'Hermite, P. Barat, C. Goizet, D. Lacombe, M.-L. Moutard, et al. Five new TTF1/NKX2.1 mutations in brain-lung-thyroid syndrome: rescue by PAX8 synergism in one case Hum. Mol. Genet., June 15, 2009; 18(12): 2266 - 2276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Deladoey, N. Pfarr, J.-M. Vuissoz, J. Parma, G. Vassart, S. Biesterfeld, J. Pohlenz, and G. Van Vliet Pseudodominant Inheritance of Goitrous Congenital Hypothyroidism Caused by TPO Mutations: Molecular and in Silico Studies J. Clin. Endocrinol. Metab., February 1, 2008; 93(2): 627 - 633. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Philibert, D. Zenaty, L. Lin, S. Soskin, F. Audran, J. Leger, J. C. Achermann, and C. Sultan Mutational analysis of steroidogenic factor 1 (NR5a1) in 24 boys with bilateral anorchia: a French collaborative study Hum. Reprod., December 1, 2007; 22(12): 3255 - 3261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Olivieri, E. Medda, S. De Angelis, H. Valensise, M. De Felice, C. Fazzini, I. Cascino, V. Cordeddu, M. Sorcini, M. A. Stazi, et al. High Risk of Congenital Hypothyroidism in Multiple Pregnancies J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3141 - 3147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Deladoey, N. Belanger, and G. Van Vliet Random Variability in Congenital Hypothyroidism from Thyroid Dysgenesis over 16 Years in Quebec J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3158 - 3161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Al Taji, H. Biebermann, Z. Limanova, O. Hnikova, J. Zikmund, C. Dame, A. Gruters, J. Lebl, and H. Krude Screening for mutations in transcription factors in a Czech cohort of 170 patients with congenital and early-onset hypothyroidism: identification of a novel PAX8 mutation in dominantly inherited early-onset non-autoimmune hypothyroidism Eur. J. Endocrinol., May 1, 2007; 156(5): 521 - 529. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Fagman, J. Liao, J. Westerlund, L. Andersson, B.E. Morrow, and M. Nilsson The 22q11 deletion syndrome candidate gene Tbx1 determines thyroid size and positioning Hum. Mol. Genet., February 1, 2007; 16(3): 276 - 285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 American Academy of Pediatrics, S. R. Rose, and the Section on Endocrinology and Committee on, American Thyroid Association, R. S. Brown, and the Public Health Committee, and Lawson Wilkins Pediatric Endocrine Society Update of newborn screening and therapy for congenital hypothyroidism. Pediatrics, June 1, 2006; 117(6): 2290 - 2303. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Vassart and J. E. Dumont Thyroid Dysgenesis: Multigenic or Epigenetic ... or Both? Endocrinology, December 1, 2005; 146(12): 5035 - 5037. [Full Text] [PDF]
|
|
|
|
|
|
E. Amendola, P. De Luca, P. E. Macchia, D. Terracciano, A. Rosica, G. Chiappetta, S. Kimura, A. Mansouri, A. Affuso, C. Arra, et al. A Mouse Model Demonstrates a Multigenic Origin of Congenital Hypothyroidism Endocrinology, December 1, 2005; 146(12): 5038 - 5047. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Eugene, A. Djemli, and G. Van Vliet Sexual Dimorphism of Thyroid Function in Newborns with Congenital Hypothyroidism J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2696 - 2700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. De Felice and R. Di Lauro Thyroid Development and Its Disorders: Genetics and Molecular Mechanisms Endocr. Rev., October 1, 2004; 25(5): 722 - 746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Meeus, B. Gilbert, C. Rydlewski, J. Parma, A. L. Roussie, M. Abramowicz, C. Vilain, D. Christophe, S. Costagliola, and G. Vassart Characterization of a Novel Loss of Function Mutation of PAX8 in a Familial Case of Congenital Hypothyroidism with In-Place, Normal-Sized Thyroid J. Clin. Endocrinol. Metab., September 1, 2004; 89(9): 4285 - 4291. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Fagman, M. Grande, A. Gritli-Linde, and M. Nilsson Genetic Deletion of Sonic Hedgehog Causes Hemiagenesis and Ectopic Development of the Thyroid in Mouse Am. J. Pathol., May 1, 2004; 164(5): 1865 - 1872. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Brown and L. A. Demmer The Etiology of Thyroid Dysgenesis--Still an Enigma after All These Years J. Clin. Endocrinol. Metab., September 1, 2002; 87(9): 4069 - 4071. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
