- Email: [email protected]
Follow-up of newborns with elevated screening T4 concentrations
ORIGINAL ARTICLES
FOLLOW-UP OF NEWBORNS WITH ELEVATED SCREENING T4 CONCENTRATIONS STEPHEN H. LAFRANCHI, MD, DAVID B. SNYDER, MD, DAVID E. SESSER, MICHAEL R. SKEELS, PHD, NALINI SINGH, MD, GREGORY A. BRENT, MD, AND JERALD C. NELSON, MD
Objective To determine the type and incidence of hyperthyroxinemic disorders detected by follow-up of infants with elevated screening total T4 (TT4) values. Study design Infants born in Oregon with a screening TT4 measurement >3 SD above the mean were offered enrollment. Serum TT4, free T4, total T3, free T3, and thyroid-stimulating hormone concentrations were measured in study infants and their mothers. Results Over a 20-month period, 101 infants (51 boys) and their mothers enrolled in the study (of 241 eligible infants), from a total screening population of 80,884; 17 infants were identified with persistent hyperthyroxinemia (TT4 >16 mg/dL). Ten had thyroxine-binding globulin excess (1:8088), 5 had evidence for increased T4 binding but not thyroxine-binding globulin excess (1:16,177), and 2 had findings compatible with thyroid hormone resistance (1:40,442); the other 84 infants had transient hyperthyroxinemia. Sequence analysis revealed a point mutation in the thyroid hormone receptor-b gene in one infant with thyroid hormone resistance; no mutation was identified in the other infant. Conclusions Although neonatal Graves’ disease occurs in approximately 1 in 25,000 newborn infants, we did not detect any case among 80,884 infants, most likely because their mothers were receiving antithyroid drugs. Although the other hyperthyroxinemic disorders in the aggregate occur frequently (1:4758) and may benefit from detection, in general they do not require treatment. (J Pediatr 2003;143:296-301)
he goal of newborn screening is to identify babies with hypothyroidism. However, infants with high screening T4 concentrations are at risk for several hyperthyroxinemic disorders. An infant born in Oregon was diagnosed with neonatal Graves’ disease at 14 days of age. The mother was not known to have Graves’ disease, but after her infant’s diagnosis she was also diagnosed with Graves’ disease. On review, the infant’s filter paper screening test T4 at 2 days of age was 32.15 lg/dL, >4 SD above the assay mean. Mortality rates in neonatal Graves’ disease in the past have been reported as high as 20%,1 but it probably is much lower with earlier diagnosis and modern treatment. Furthermore, these infants are at risk for some degree of intellectual impairment; approximately half have IQs in the 75 to 85 range.2 Report of the screening results probably would lead to earlier diagnosis and treatment, perhaps improving the chance for survival without neurologic sequelae. Other potential causes of elevated screening total T4 (TT4) values include activating mutations of the thyroid-stimulating hormone (TSH) receptor, activating mutations of the a-subunit of the guanine nucleotide-binding protein (as is typically seen in McCuneAlbright syndrome), loss-of-function mutations in the thyroid hormone receptor causing thyroid hormone resistance, and increases in a T4-binding protein, most commonly
T
FT4 FT3 G-protein T3 T3RU T4
296
Free T4 measurement Free T3 measurement Guanine nucleotide-binding protein Triiodothyronine T3 resin uptake Thyroxine
TBG TR TT3 TT4 TSH
Thyroxine-binding globulin Thyroid hormone receptor Total T3 measurement Total T4 measurement Thyroid-stimulating hormone
See editorial, p 285. From the Department of Pediatrics, Oregon Health and Science University, Pediatric Endocrinology, Legacy Emanuel Children’s Hospital, and Oregon State Public Health Laboratory, Portland, Oregon; the Department of Medicine, VA Greater Los Angeles Health Care System, UCLA School of Medicine, Los Angeles, Quest Diagnostics Nichols Institute, San Juan Capistrano, and the Departments of nternal Medicine and Pathology, Loma Linda University Medical Center, Loma Linda, California. Submitted for publication Sept 20, 2002; revision received Feb 14, 2003; accepted Mar 12, 2003. Reprints not available from the authors. Please address correspondence to: Stephen H. LaFranchi, MD, Department of Pediatrics [CDRCP], Oregon Health and Science University, 707 SW Gaines Rd, Portland, OR 97239-3098. E-mail: [email protected]. Copyright ª 2003 Mosby, Inc. All rights reserved. 0022-3476/2003/$30.00 + 0 10.1067/S0022-3476(03)00184-7
thyroxine binding globulin (TBG), or serum protein mutations resulting in increased affinity for T4 (familial dysalbuminemic hyperthyroxinemia). Some of these conditions, such as activating mutations of the TSH receptor or the Gprotein a-subunit, would also result in hyperthyroidism. Infants with thyroid hormone resistance can manifest clinical features of hypothyroidism and hyperthyroidism, depending on which organ systems are most affected. Infants with increases in binding proteins, such as TBG, have normal free hormone concentrations with no clinical evidence of thyroid dysfunction. It is unclear whether these hereditary disorders manifest hyperthyroxinemia in the newborn period, and, if so, whether they can be detected by follow-up of elevated screening T4 concentrations. The incidence of each of these disorders is unclear. Neonatal Graves’ disease is estimated to occur in 1 in 25,000 newborn infants,1 based on the incidence of maternal Graves’ disease and the percent of babies affected. Hereditary TBG excess also is estimated to occur in 1 in 25,000 individuals.3 There is no estimate for the incidence of activating mutations of the TSH receptor, of the G- protein a-subunit, or of thyroid hormone resistance. The objective of our study was to determine the type and incidence of hyperthyroxinemic disorders that might be detected by follow-up of infants with elevated screening T4 values. Ultimately, detection of such disorders is likely to allow the best opportunity to follow their natural course and determine the benefit of treatment.
Capistrano, California. FT4 was measured by direct equilibrium dialysis, FT3 was measured by indirect equilibrium dialysis, TT4 and TT3 were measured by radioimmunoassay, and TSH was measured by chemiluminescent immunometric assay. Results were compared with age-related normal ranges provided by Quest Diagnostics Nichol’s Institute.4-6 To convert to SI units, multiply TT4 (lg/dL) and FT4 (ng/dL) by 12.87 to determine nmol/L and pmol/L, respectively; multiply TT3 (ng/dL) and FT3 (pg/dL) by 15.4 to determine pmol/L. Further investigations were carried out in infants with confirmed hyperthyroxinemia to determine the underlying cause. Serum TBG was measured in 10 of the 15 infants with results compatible with excess thyroid binding protein (also by Quest Diagnostics Nichols Institute). By history, none of the subjects were taking drugs known to elevate serum T4 concentrations. Sequence analysis of the carboxy-terminus (exons 7 through 10) of the thyroid hormone receptor (TR)-b gene was carried out in the two infants with thyroid function test results compatible with thyroid hormone resistance (Drs Singh and Brent). Genomic DNA was extracted from peripheral blood leukocytes, and exons 7 through 10 were amplified by polymerase chain reaction with intronic primers. The sequence was confirmed in at least two independent clones. The TR-b gene mutation identified in one patient was introduced into wild-type human TR-b cDNA, and in vitro studies of T3 binding were carried out.
RESULTS METHODS Infants born in Oregon and undergoing newborn screening were eligible for the study. T4 screening tests are carried out by the Oregon State Public Health Laboratory (PHL) on approximately 350 Oregon infants daily. Two routine screening specimens are collected in Oregon, the first at approximately 2 days of life and the second at approximately 2 weeks of life. As the serum TT4 rises and peaks 24 hours after birth and then falls over the next week of life, samples collected at approximately the same postnatal age are batched together for testing. The TT4 means and standard deviations are calculated by computer program for one month’s data and then applied to the following month’s data. Infants whose filter paper TT4 value was >3 SD above the mean (0.3%) were eligible for enrollment in the study. Infants were identified from test results in either the first or second routine specimen. Once identified, the study coordinator contacted the infant’s physician of record, who then presented the study to the infant’s parents. Once permission for contact was obtained, the coordinator contacted the family to explain full details of the study. The study protocol and consent form were approved by the Institutional Review Board of the Oregon Health and Science University. In those families who chose to participate, blood was obtained as soon as practical (mean age = 68 days) on infant and mother for measurement of TT4, free T4 (FT4), total T3 (TT3), free T3 (FT3), and TSH. Thyroid function tests were performed at Quest Diagnostics Nichols Institute, San Juan Follow-up of Newborns With Elevated Screening T4 Concentrations
Over a 20-month period, we enrolled 101 infants (51 boys) into the study of 241 eligible infants whose mothers were offered participation, from a total screening population of 80,884 infants born in Oregon over that time. Of the 101 infants who participated, 17 had persistent TT4 elevation confirmed on serum testing, whereas the serum TT4 measurement in the other 84 infants was normal. A comparison of the screening test results in the 17 infants with persistent hyperthyroxinemia and the 84 infants with transient hyperthyroxinemia (infants with false-positive results) is presented in Table I. There were no differences in the sex ratio, age of screening test collection, or screening test TT4 value between these 2 groups, nor were there any differences compared with the 140 infants whose mothers chose not to participate in the study. The screening TT4 level tended to fall between collection of the first specimen on day 2 of life and the second specimen at 2 weeks in the normal infants, whereas it remained elevated in the second test in the hyperthyroxinemic infants. A summary of the thyroid function test results and the estimated incidence of each disorder is presented in Table II. Infants judged to have TBG excess had elevated serum TT4 and TT3 values but normal FT4, FT3, and TSH concentrations. Infants judged to have some form of increased serum T4 binding but not TBG excess had similar findings but normal TBG concentrations. The two infants judged to have thyroid hormone resistance had elevated serum TT4 and TT3 but also elevations of serum FT4 and FT3, without suppressed TSH concentrations. The estimated incidence for these three 297
Table I. Comparison of screening test results in infants with confirmed persistent hyperthyroxinemia and infants with transient hyperthyroxinemia (age = mean; TT4 = mean 6 1 SD) Filter Paper #1 Group
n
M:F
Age
Confirmed persistent hyperthyroxinemia Transient hyperthyroxinemia Normal range for age
17 84
8:9 43:41
29 h 31 h
disorders probably is the minimal incidence, as a little more than half of eligible infants did not participate in the study. If the thyroid disorders occurred at a similar frequency in those latter infants, the incidence of each of these disorders could be up to approximately 2-fold higher. We did not detect any infants with unrecognized neonatal Graves’ disease, nor did we detect any with activating mutations of the TSH receptor or the G-protein a-subunit. Individual thyroid function tests for the 10 infants with results compatible with TBG excess are presented in Table III. Serum TBG was not determined in 5 of these infants. Infant 2 had a markedly low T3 resin uptake (T3RU), which confirms elevated binding protein. In infant 3, the sample was insufficient to measure TSH. However, this infant did not have clinical hyperthyroidism, and his mother had thyroid function results compatible with elevated binding protein, so we believe that he had the same disorder. Thyroid tests in the mother of infant 9 are compatible with postpartum thyrotoxicosis. Of the 5 infants with measured elevated serum TBG concentrations, 1 (infant 8) and possibly another (infant 7) of their mothers also appeared to have TBG excess. This pattern fits TBG excess, an X-linked disorder, in which the heterozygous female is less affected than the hemizygous male. Serum TBG appears less elevated in the 2 female infants (mean = 4.0 mg/dL) compared with the 3 male infants (mean = 5.5 mg/dL) in whom it was measured. Overall, 4 of the 10 infants are female, and their mean serum TT4 is less than that in the 6 male infants (19.0 lg/dL vs 26.3 lg/dL, respectively). A similar pattern is seen with serum TT3 (282 ng/dL in female infants vs 480 ng/dL in male infants). It is possible that some of the 5 infants in whom we were unable to determine serum TBG may have increased serum T4 binding but not TBG excess. Individual thyroid function tests for the 5 infants with increased serum T4 binding but not TBG excess are presented in Table IV. Serum TBG was normal in all 5 infants. Only one mother (subject 5) had clearly elevated T4 binding to proteins. The serum TT4 and TT3 concentrations were not different between the 3 female and 2 male infants; serum TT4 was 19.9 versus 20.0 lg/dL, respectively, and TT3 was 310 versus 304 ng/dL, respectively. Although the number of subjects was small, these results fit an autosomal disorder better, such as dysalbuminemic hyperthyroxinemia or a mutation of the 298
LaFranchi et al
mean TT4 (mg/dL)
22.76 22.67 8.2-19.9
Filter Paper #2 SD
Age
3.25 3.23
15 d 14 d
Mean TT4 (mg/dL)
23.71 16.90 8.1-15.7
SD
4.74 3.38
transthyretin gene leading to an elevation of this binding protein. Thyroid function tests in the two infants with thyroid hormone resistance are presented in Table V. Thyroid hormone resistance is inherited as an autosomal dominant disorder. Measurement of thyroid function was normal in both mother and father of both subjects. Serial thyroid function testing out to a year of age continued to show abnormal elevations of serum T4, FT4, T3, and FT3, with normal TSH concentrations in both subjects. We carried out sequence analysis of the TR-b gene carboxy-terminus (exons 7 through 10) in both infants. No coding sequence mutations were found in subject 1. In subject 2, we found a point mutation (G1663A) changing codon 460 from GAG to AAG, resulting in a glutamine-to-lysine substitution. There are no reported mutations of amino acid 460 associated with thyroid hormone resistance. The G460L mutation was introduced into a human TR-b cDNA, and T3 binding affinity of the in vitro expressed protein was compared with wild-type TR-b. T3 binding was not significantly diminished in the TR-b G460L compared with wild-type TR-b. Additional in vitro functional characterization, however, has shown that TR-b G460L antagonizes wild-type TR in T3-dependent gene activation and repression (Singh and Brent, manuscript in preparation). These findings are consistent with those from other resistance to thyroid hormone-associated TR-b mutations. Both of these infants are currently followed with no treatment.
DISCUSSION Overall, we detected 17 infants with confirmed hyperthyroxinemia of 80,884 newborn infants. The incidence of these combined disorders is approximately 1 in 4750, which, as described above, probably is an underestimate. Our main goal in follow-up of newborn infants with elevated screening T4 concentrations was to detect infants with some form of hyperthyroidism who would benefit from early detection and treatment. However, we did not detect any infant with unrecognized neonatal Graves’ disease or activating mutations of the TSH receptor or the G-protein a-subunit. At an estimated incidence of 1:25,000 for neonatal Graves’ disease, we would have expected 3 or 4 cases in our screened population. In fact, we were aware of cases, but they The Journal of Pediatrics September 2003
Table II. Serum thyroid test results (means) in 17 infants with confirmed persistent hyperthyroxinemia and estimated (minimal) incidence Disorder
n
TBG excess Increased T4 binding Thyroid hormone resistance Transient hyperthyroxinemia
10 5 2 84
Estimated incidence
TT4 (mg/dL)
FT4 (ng/dL)
TT3 (ng/dL)
FT3 (pg/dL)
TSH (mU/L)
1:8088 1:16,177 1:40,442
23.3 20.0 22.9 13.6
1.8 1.5 4.1 1.6
401 307 305 243
527 514 730 480
1.94 2.80 2.07 2.43
8.1-15.7
0.8-2.7
105-245
180-770
1.7-9.1
Normal range for age
Table III. Thyroid function test results in infants and mothers with TBG excess Infant Patient No. 1 2 3 4 5 6 7 8 9 10 Normal range for age
Mother
TT4 FT4 TT3 FT3 TSH TBG Sex (mg/dL) (ng/dL) (ng/dL) (pg/dL) (mU/L) (mg/dL) M M M F F M M M F F
28.2 17.1 29.0 17.9 18.3 20.9 33.5 28.8 22.6 17.0
1.5 2.3 2.0 2.1 2.1 1.8 1.5 1.1 1.8 1.7
8.1-15.7
0.9-2.2
718 342 419 323 280 414 518 471 292 232
775 545 509 552 458 638 612 500 409 370
105-245 180-770
1.91 1.53 — 3.96 1.69 1.72 1.71 1.12 2.59 1.26
1.7-9.1
— ***
— — — 4.1 7.1 5.3 3.9 4.1
1.7-3.7
TT4 FT4 TT3 FT3 TSH (mg/dL) (ng/dL) (ng/dL) (pg/dL) (mU/L) 11.3 10.9 16.7 7.2 7.7 7.5 13.3 14.3 14.7 8.1
0.9 1.0 1.5 1.2 1.3 1.4 1.0 — 8.1 1.1
316 179 268 119 95 167 262 282 390 146
319 268 364 287 252 338 366 324 1555 283
2.51 1.28 0.71 2.64 0.29 0.98 1.59 1.07 0.13 1.13
5.6-13.7
0.8-2.7
87-180
210-440
0.40-4.2
*T3RU 18% (normal range, 23%-38%).
all initially had hypothyroidism, as their mothers were known to have Graves’ disease and were receiving antithyroid drug therapy. Antithyroid drugs cross to the fetus; depending on the dose, newborn infants often have transient hypothyroidism until the drug is metabolized and excreted by the neonate. A combination of stimulating and blocking antibodies may also cloud the initial presentation.7 Neonatal Graves’ disease has been detected by using tests on the newborn screening specimen.8 However, in these cases, the mothers were already known to have Graves’ disease, and so the newborn screening specimen was singled out for testing. We are not aware of any case of unrecognized neonatal Graves’ disease detected by follow-up after T4 population screening. Congenital hyperthyroidism resulting from an activating mutation of the TSH receptor has been reported.9-11 In three cases cited, the infants had clinical features of hyperthyroidism and a goiter; in the other, a mother and her two sons had congenital hyperthyroidism.12 Thyroid function tests confirmed hyperthyroidism; in all cases, there was no evidence of autoimmune hyperthyroidism. Activating mutations of the TSH receptor genes were documented in each case; the first three were de novo mutations. All patients were treated initially with antithyroid drugs; one case underwent subtotal Follow-up of Newborns With Elevated Screening T4 Concentrations
thyroidectomy at 8.7 years followed by radioactive iodine treatment at 9.2 years of age.9 Neurodevelopmental testing was reported only in this latter case; his IQ was 75 to 85.9 The incidence of this disorder is unknown. Although newborn screening results were not reported, it is clear from these three cases that activating mutations of the TSH receptor are associated with elevated neonatal T4 concentrations that might be detected by screening. Hyperthyroidism occurs in 20% to 42% of children with McCune-Albright syndrome, resulting from somatic cell mutations of the G-protein a-subunit within the thyroid.13,14 Hyperthyroidism present at birth has been reported in at least two cases.15,16 It is clear from these two cases that at least some infants with activating mutations of the G-protein a-subunit have elevated neonatal T4 concentrations; we are unaware of any published reports of detection by newborn screening. The incidence of this disorder is unknown. In patients with thyroid hormone resistance, serum T4 (and T3) are elevated as a result of thyroid hormone receptor mutations, decreased pituitary T3 binding, and increased TSH secretion. Thyroid hormone resistance has been detected prenatally17 and in the newborn period. In one case, the infant’s mother was known to have thyroid hormone 299
Table IV. Thyroid function test results in infants and mothers with increased serum T4 binding but not TBG excess* Infant
Mother
Patient TT4 FT4 TT3 FT3 TSH TBG No. Sex (mg/dL) (ng/dL) (ng/dL) (pg/dL) (mU/L) (mg/dL) 1 2 3 4 5
M F F M F
19.3 17.2 21.5 20.8 21.0
1.6 1.6 1.2 1.7 1.6
272 284 362 336 283
526 517 473 504 550
1.77 3.44 1.65 4.17 2.95
3.4 3.1 3.1 3.3 2.3
TT4 FT4 TT3 FT3 (mg/dL) (ng/dL) (ng/dL) (pg/dL) 8.0 7.3 9.5 8.6 16.7
0.8 0.8 1.2 1.5 1.1
163 185 178 146 152
381 345 332 335 310
TSH (mU/L) 1.13 1.01 0.52 0.43 0.95
*See Table III for normal range for age.
Table V. Thyroid function test results in infants with thyroid hormone resistance and their mothers and fathers* Patient TT4 FT4 TT3 FT3 TSH No. Sex (mg/dL) (ng/dL) (ng/dL) (pg/dL) (mU/L)
1
2
F
F
22.4
23.3
3.6
Infant 246
4.6
Infant 364
491
968
2.34
1.8
TT4 FT4 TT3 FT3 (mg/dL) (ng/dL) (ng/dL) (pg/dL)
10.7
2.4
9.4
1.2
7.0
1.2
7.9
1.1
Mother 162 Father 190 Mother 116 Father 158
TSH (mU/L)
383
0.66
358
0.95
248
0.62
332
0.56
*See Table III for normal range for age.
resistance.18 At 2 days of life, testing using the newborn screening specimen showed a TT4 of 51 lg/dL and a TSH of 26 mU/L. The other case was detected by routine neonatal cord blood screening in Hong Kong.19 Other case reports show even higher TSH elevations when using the newborn screening specimen.20 Children with this disorder may have clinical features compatible with hypothyroidism, including growth retardation, delayed skeletal maturation, learning disabilities, attention deficit-hyperactivity disorder, sensorineural deafness, and strabismus. There are reports that some of these features may improve with T3 treatment.21 This has led to the suggestion by some investigators to consider screening newborn infants for thyroid hormone resistance.22 Our results suggest the minimum incidence of this disorder is approximately 1 in 40,000. One of the two infants with thyroid hormone resistance (patient 1) did not have a TR-b gene mutation identified. Among families with clinical features of thyroid hormone resistance, approximately 10% do not have TR gene mutations.23 Defects in a variety of nuclear receptor cofactors may be involved, although these associations have not yet been shown for patients with thyroid hormone resistance without TR-b gene mutations.24 Familial TBG excess, an X-linked disorder, is estimated to occur in approximately 1 in 25,000 individuals.25 Heterozygous female infants usually have TBG concentrations intermediate between normal and affected male infants, a finding similar to our results. Griffiths et al reported 5 cases of TBG 300
LaFranchi et al
excess detected during newborn screening for congenital hypothyroidism out of a population of 30,108 in Birmingham, England, an incidence of approximately 1 in 6000.26 This figure is closer to our finding, 1 in 8000. Thus, either hereditary TBG excess is more common than previously estimated, or some acquired forms are included in estimates. Familial dysalbuminemic hyperthyroxinemia may be as common as TBG excess27 or more common in some populations.28 It may be that some of our 5 subjects with evidence of increased T4 binding, not TBG excess, have this disorder. To our knowledge, there have been no reports of infants with dysalbuminemic hyperthyroxinemia detected by newborn screening. In summary, follow-up of infants with elevated screening T4 concentrations led to an incidence of 1 in 4750 for all hyperthyroxinemia disorders combined. However, we did not detect any infant with hyperthyroidism, the case most likely to benefit from early diagnosis and treatment. Treatment of infants with thyroid hormone resistance remains unclear. Although infants have features of hypothyroidism, and some physicians have prescribed large doses of T4 and/or T3 treatment,18 most patients with generalized thyroid hormone resistance are followed without treatment. Infants with elevations of TBG or other binding proteins have normal thyroid function and do not require treatment. Some would argue that detection would help to prevent inappropriate treatment, as some patients are mistakenly diagnosed as having The Journal of Pediatrics September 2003
a hyperthyroid state. Our screening program continues to follow-up infants with T4 concentrations >3.5 SD, so that we can gain experience with a larger population over a longer time period. We chose this higher cutoff to reduce the cost of follow-up and the number of false-positive cases. On balance, we do not recommend that follow-up of such infants become part of routine newborn screening programs at this time.
REFERENCES 1. LaFranchi S, Hanna CE. Graves’ disease in the neonatal period and childhood. In: Braverman L, Utiger RD, eds. Werner and Ingbar’s the thyroid. 8th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2000. p. 989-97. 2. Daneman D, Howard NJ. Neonatal thyrotoxicosis: intellectual impairment and craniosynostosis. J Pediatr 1980;97:257-9. 3. Refetoff S. Inherited thyroxine-binding globulin abnormalities in man. Endocrinol Rev 1989;10:275-93. 4. Nelson JC, Clark SJ, Borut DL, Tomei RT, Carlton EI. Age-related changes in serum free thyroxine during childhood and adolescence. J Pediatr 1993;123:899-905. 5. Adams LM, Emery JR, Clark SJ, Carlton EI, Nelson JC. Reference ranges for newer thyroid function tests in premature infants. J Pediatr 1995;126:122-7. 6. Elmlinger MW, Kuhnel W, Lambrecht H-G, Ranke MB. Reference intervals from birth to adulthood for serum thyroxine (T4), triiodothyronine (T3), free T3, free T4, thryoxine binding globulin (TBG) and thyrotropin (TSH). Clin Chem Lab Med 2001;39:973-9. 7. Zakarija M, McKenzie JM, Hoffman WH. Prediction and therapy of intrauterine and late-onset neonatal hyperthyroidism. J Clin Endocrinol Metab 1986;62:368-71. 8. Walker P, Dussault JH, Hart IR, Langelier P, Szots F. Thyrotoxicosis detected in a mass-screening program for neonatal hyperthyroidism: demonstration of placental transfer of an immunoglobulin with marked lipolytic activity. J Pediatr 1997;91:400-3. 9. Kopp P, Van Sande J, Parma J, Duprez L, Gerber H, Joss E, et al. Brief report: congenital hyperthyroidism caused by a mutation in the thyrotropinreceptor gene. N Engl J Med 1995;332:150-4. 10. De Roux N, Polak M, Couet J, Leger J, Czernichow P, Milgrom E, et al. A neomutation of the thyroid-stimulating hormone receptor in a severe neonatal hyperthyroidism. J Clin Endocrinol Metab 1996;81:2023-36. 11. Tonacchera M, Agretti P, Rosellini V, Ceccarini G, Perri A, Zampolli M, et al. Sporadic nonautoimmune congenital hyperthyroidism due to a strong activating mutation of the thyrotropin receptor gene. Thyroid 2000;10: 859-63. 12. Schwab KO, Gerlich M, Broecker M, Sohlemann P, Dervahl M, Lohse MJ. Constitutively active germline mutation of the thyrotropin receptor gene as a cause of congenital hyperthyroidism. J Pediatr 1997;131:899-904.
Follow-up of Newborns With Elevated Screening T4 Concentrations
13. Feuillan PP, Shawker T, Rose SR, Jones J, Jeevanram RK, Nissula BC. Thyroid abnormalities in the McCune-Albright syndrome: ultrasonography and hormonal studies. J Clin Endocrinol Metab 1990;71:1596-601. 14. Lee PA, Van Dop C, Migeon CJ. McCune-Albright syndrome: longterm follow-up. JAMA 1986;256:2980-4. 15. Mastorakos G, Mitsiades NS, Doufas AG, Koutras DA. Hyperthyroidism in MCCune-Albright syndrome with a review of thyroid abnormalities sixty years after the first report. Thyroid 1997;7:433-9. 16. Yoshimoto M, Nikayama M, Baba T, Uehara Y, Niikawa N, Ito M, et al. A case of neonatal McCune-Albright syndrome with Cushing’s syndrome and hyperthyroidism. Acta Paediatr Scand 1991;984-7. 17. Asteria C, Rajanayagam O, Collingwood TN, Persani L, Romoli R, Mannavola D, et al. Prenatal diagnosis of thyroid hormone resistance. J Clin Endocrinol Metab 1999;84:405-10. 18. Weiss RE, Balzano S, Scherberg NH, Refetoff S. Neonatal detection of generalized resistance to thyroid hormone. JAMA 1990;264:2245-50. 19. Wong GWK, Skek CC, Lam STS, Tsui MKM, Leung SSF. Detection of resistance to thyroid hormone by cord blood screening. Acta Paediatr 1995;84:335-6. 20. Pohlenz J, Schonberger W, Koffler T, Refetoff S. Resistance to thyroid hormone caused by a new mutation (V336M) in the thyroid hormone receptor b gene. Thyroid 1999;9:1001-4. 21. Hauser P, Zametkin AJ, Martinez P, Vitiello B, Matochik JA, Mixson AJ, et al. Attention deficit-hyperactivity disorder in people with generalized resistance to thyroid hormone. N Engl J Med 1993;328:997-1001. 22. Pass KA, Hedden JE, Morris JE, Hauser P, Weintraub BD, Mizejewski GJ. Newborn screening for thyroid resistance: implications for attention deficit-hyperactivity disorder. In: Farriaux J-P, Dhondt J-L, eds. New horizons in neonatal screening. Amsterdam (The Netherlands): Elsevier Science BV; 1994. p. 141-4. 23. Reutrakul S, Sadow PF, Pannain S, Pohlenz J, Carvalho GA, Macchia PE, et al. Search for abnormalities of nuclear corepressors, coactivators, and a coregulator in families with resistance to thyroid hormone without mutations in thyroid hormone receptor b or a genes. J Clin Endocrinol Metab 2000;85:3609-17. 24. Pohlenz J, Weiss RE, Macchia PE, Pannain S, Lau IT, Ho H, et al. Five new families with resistance to thyroid hormone not caused by mutations in the thyroid hormone receptor b gene. J Clin Endocrinol Metab 1999;84: 3919-28. 25. Refetoff S. Inherited thyroxine-binding globulin abnormalities in man. Endocrinol Rev 1989;10:275-93. 26. Griffiths KD, Virdi NK, Rayner PH, Green A. Neonatal screening for congenital hypothyroidism by measurement of plasma thyroxine and thyroid stimulating hormone concentrations. BMJ 1985;291:117-20. 27. Jensen IW, Fabaer J, Grunnet N. Familial occurrence of dysalbuminemic hyperthyroxinemia, lipomatosis and ankylosing spondylitis. Scand J Rheumatol 1990;19:303-5. 28. Sunthornthepvarakul T, Angkeow P, Weiss RE, Hayashi Y, Refetoff S. An identical missense mutation in the albumin gene results in familial dysalbuminemic hyperthyroxinemia in 8 unrelated families. Biochem Biophys Res Commun 1994;202:781-7.
301
FOLLOW-UP OF NEWBORNS WITH ELEVATED SCREENING T4 CONCENTRATIONS STEPHEN H. LAFRANCHI, MD, DAVID B. SNYDER, MD, DAVID E. SESSER, MICHAEL R. SKEELS, PHD, NALINI SINGH, MD, GREGORY A. BRENT, MD, AND JERALD C. NELSON, MD
Objective To determine the type and incidence of hyperthyroxinemic disorders detected by follow-up of infants with elevated screening total T4 (TT4) values. Study design Infants born in Oregon with a screening TT4 measurement >3 SD above the mean were offered enrollment. Serum TT4, free T4, total T3, free T3, and thyroid-stimulating hormone concentrations were measured in study infants and their mothers. Results Over a 20-month period, 101 infants (51 boys) and their mothers enrolled in the study (of 241 eligible infants), from a total screening population of 80,884; 17 infants were identified with persistent hyperthyroxinemia (TT4 >16 mg/dL). Ten had thyroxine-binding globulin excess (1:8088), 5 had evidence for increased T4 binding but not thyroxine-binding globulin excess (1:16,177), and 2 had findings compatible with thyroid hormone resistance (1:40,442); the other 84 infants had transient hyperthyroxinemia. Sequence analysis revealed a point mutation in the thyroid hormone receptor-b gene in one infant with thyroid hormone resistance; no mutation was identified in the other infant. Conclusions Although neonatal Graves’ disease occurs in approximately 1 in 25,000 newborn infants, we did not detect any case among 80,884 infants, most likely because their mothers were receiving antithyroid drugs. Although the other hyperthyroxinemic disorders in the aggregate occur frequently (1:4758) and may benefit from detection, in general they do not require treatment. (J Pediatr 2003;143:296-301)
he goal of newborn screening is to identify babies with hypothyroidism. However, infants with high screening T4 concentrations are at risk for several hyperthyroxinemic disorders. An infant born in Oregon was diagnosed with neonatal Graves’ disease at 14 days of age. The mother was not known to have Graves’ disease, but after her infant’s diagnosis she was also diagnosed with Graves’ disease. On review, the infant’s filter paper screening test T4 at 2 days of age was 32.15 lg/dL, >4 SD above the assay mean. Mortality rates in neonatal Graves’ disease in the past have been reported as high as 20%,1 but it probably is much lower with earlier diagnosis and modern treatment. Furthermore, these infants are at risk for some degree of intellectual impairment; approximately half have IQs in the 75 to 85 range.2 Report of the screening results probably would lead to earlier diagnosis and treatment, perhaps improving the chance for survival without neurologic sequelae. Other potential causes of elevated screening total T4 (TT4) values include activating mutations of the thyroid-stimulating hormone (TSH) receptor, activating mutations of the a-subunit of the guanine nucleotide-binding protein (as is typically seen in McCuneAlbright syndrome), loss-of-function mutations in the thyroid hormone receptor causing thyroid hormone resistance, and increases in a T4-binding protein, most commonly
T
FT4 FT3 G-protein T3 T3RU T4
296
Free T4 measurement Free T3 measurement Guanine nucleotide-binding protein Triiodothyronine T3 resin uptake Thyroxine
TBG TR TT3 TT4 TSH
Thyroxine-binding globulin Thyroid hormone receptor Total T3 measurement Total T4 measurement Thyroid-stimulating hormone
See editorial, p 285. From the Department of Pediatrics, Oregon Health and Science University, Pediatric Endocrinology, Legacy Emanuel Children’s Hospital, and Oregon State Public Health Laboratory, Portland, Oregon; the Department of Medicine, VA Greater Los Angeles Health Care System, UCLA School of Medicine, Los Angeles, Quest Diagnostics Nichols Institute, San Juan Capistrano, and the Departments of nternal Medicine and Pathology, Loma Linda University Medical Center, Loma Linda, California. Submitted for publication Sept 20, 2002; revision received Feb 14, 2003; accepted Mar 12, 2003. Reprints not available from the authors. Please address correspondence to: Stephen H. LaFranchi, MD, Department of Pediatrics [CDRCP], Oregon Health and Science University, 707 SW Gaines Rd, Portland, OR 97239-3098. E-mail: [email protected]. Copyright ª 2003 Mosby, Inc. All rights reserved. 0022-3476/2003/$30.00 + 0 10.1067/S0022-3476(03)00184-7
thyroxine binding globulin (TBG), or serum protein mutations resulting in increased affinity for T4 (familial dysalbuminemic hyperthyroxinemia). Some of these conditions, such as activating mutations of the TSH receptor or the Gprotein a-subunit, would also result in hyperthyroidism. Infants with thyroid hormone resistance can manifest clinical features of hypothyroidism and hyperthyroidism, depending on which organ systems are most affected. Infants with increases in binding proteins, such as TBG, have normal free hormone concentrations with no clinical evidence of thyroid dysfunction. It is unclear whether these hereditary disorders manifest hyperthyroxinemia in the newborn period, and, if so, whether they can be detected by follow-up of elevated screening T4 concentrations. The incidence of each of these disorders is unclear. Neonatal Graves’ disease is estimated to occur in 1 in 25,000 newborn infants,1 based on the incidence of maternal Graves’ disease and the percent of babies affected. Hereditary TBG excess also is estimated to occur in 1 in 25,000 individuals.3 There is no estimate for the incidence of activating mutations of the TSH receptor, of the G- protein a-subunit, or of thyroid hormone resistance. The objective of our study was to determine the type and incidence of hyperthyroxinemic disorders that might be detected by follow-up of infants with elevated screening T4 values. Ultimately, detection of such disorders is likely to allow the best opportunity to follow their natural course and determine the benefit of treatment.
Capistrano, California. FT4 was measured by direct equilibrium dialysis, FT3 was measured by indirect equilibrium dialysis, TT4 and TT3 were measured by radioimmunoassay, and TSH was measured by chemiluminescent immunometric assay. Results were compared with age-related normal ranges provided by Quest Diagnostics Nichol’s Institute.4-6 To convert to SI units, multiply TT4 (lg/dL) and FT4 (ng/dL) by 12.87 to determine nmol/L and pmol/L, respectively; multiply TT3 (ng/dL) and FT3 (pg/dL) by 15.4 to determine pmol/L. Further investigations were carried out in infants with confirmed hyperthyroxinemia to determine the underlying cause. Serum TBG was measured in 10 of the 15 infants with results compatible with excess thyroid binding protein (also by Quest Diagnostics Nichols Institute). By history, none of the subjects were taking drugs known to elevate serum T4 concentrations. Sequence analysis of the carboxy-terminus (exons 7 through 10) of the thyroid hormone receptor (TR)-b gene was carried out in the two infants with thyroid function test results compatible with thyroid hormone resistance (Drs Singh and Brent). Genomic DNA was extracted from peripheral blood leukocytes, and exons 7 through 10 were amplified by polymerase chain reaction with intronic primers. The sequence was confirmed in at least two independent clones. The TR-b gene mutation identified in one patient was introduced into wild-type human TR-b cDNA, and in vitro studies of T3 binding were carried out.
RESULTS METHODS Infants born in Oregon and undergoing newborn screening were eligible for the study. T4 screening tests are carried out by the Oregon State Public Health Laboratory (PHL) on approximately 350 Oregon infants daily. Two routine screening specimens are collected in Oregon, the first at approximately 2 days of life and the second at approximately 2 weeks of life. As the serum TT4 rises and peaks 24 hours after birth and then falls over the next week of life, samples collected at approximately the same postnatal age are batched together for testing. The TT4 means and standard deviations are calculated by computer program for one month’s data and then applied to the following month’s data. Infants whose filter paper TT4 value was >3 SD above the mean (0.3%) were eligible for enrollment in the study. Infants were identified from test results in either the first or second routine specimen. Once identified, the study coordinator contacted the infant’s physician of record, who then presented the study to the infant’s parents. Once permission for contact was obtained, the coordinator contacted the family to explain full details of the study. The study protocol and consent form were approved by the Institutional Review Board of the Oregon Health and Science University. In those families who chose to participate, blood was obtained as soon as practical (mean age = 68 days) on infant and mother for measurement of TT4, free T4 (FT4), total T3 (TT3), free T3 (FT3), and TSH. Thyroid function tests were performed at Quest Diagnostics Nichols Institute, San Juan Follow-up of Newborns With Elevated Screening T4 Concentrations
Over a 20-month period, we enrolled 101 infants (51 boys) into the study of 241 eligible infants whose mothers were offered participation, from a total screening population of 80,884 infants born in Oregon over that time. Of the 101 infants who participated, 17 had persistent TT4 elevation confirmed on serum testing, whereas the serum TT4 measurement in the other 84 infants was normal. A comparison of the screening test results in the 17 infants with persistent hyperthyroxinemia and the 84 infants with transient hyperthyroxinemia (infants with false-positive results) is presented in Table I. There were no differences in the sex ratio, age of screening test collection, or screening test TT4 value between these 2 groups, nor were there any differences compared with the 140 infants whose mothers chose not to participate in the study. The screening TT4 level tended to fall between collection of the first specimen on day 2 of life and the second specimen at 2 weeks in the normal infants, whereas it remained elevated in the second test in the hyperthyroxinemic infants. A summary of the thyroid function test results and the estimated incidence of each disorder is presented in Table II. Infants judged to have TBG excess had elevated serum TT4 and TT3 values but normal FT4, FT3, and TSH concentrations. Infants judged to have some form of increased serum T4 binding but not TBG excess had similar findings but normal TBG concentrations. The two infants judged to have thyroid hormone resistance had elevated serum TT4 and TT3 but also elevations of serum FT4 and FT3, without suppressed TSH concentrations. The estimated incidence for these three 297
Table I. Comparison of screening test results in infants with confirmed persistent hyperthyroxinemia and infants with transient hyperthyroxinemia (age = mean; TT4 = mean 6 1 SD) Filter Paper #1 Group
n
M:F
Age
Confirmed persistent hyperthyroxinemia Transient hyperthyroxinemia Normal range for age
17 84
8:9 43:41
29 h 31 h
disorders probably is the minimal incidence, as a little more than half of eligible infants did not participate in the study. If the thyroid disorders occurred at a similar frequency in those latter infants, the incidence of each of these disorders could be up to approximately 2-fold higher. We did not detect any infants with unrecognized neonatal Graves’ disease, nor did we detect any with activating mutations of the TSH receptor or the G-protein a-subunit. Individual thyroid function tests for the 10 infants with results compatible with TBG excess are presented in Table III. Serum TBG was not determined in 5 of these infants. Infant 2 had a markedly low T3 resin uptake (T3RU), which confirms elevated binding protein. In infant 3, the sample was insufficient to measure TSH. However, this infant did not have clinical hyperthyroidism, and his mother had thyroid function results compatible with elevated binding protein, so we believe that he had the same disorder. Thyroid tests in the mother of infant 9 are compatible with postpartum thyrotoxicosis. Of the 5 infants with measured elevated serum TBG concentrations, 1 (infant 8) and possibly another (infant 7) of their mothers also appeared to have TBG excess. This pattern fits TBG excess, an X-linked disorder, in which the heterozygous female is less affected than the hemizygous male. Serum TBG appears less elevated in the 2 female infants (mean = 4.0 mg/dL) compared with the 3 male infants (mean = 5.5 mg/dL) in whom it was measured. Overall, 4 of the 10 infants are female, and their mean serum TT4 is less than that in the 6 male infants (19.0 lg/dL vs 26.3 lg/dL, respectively). A similar pattern is seen with serum TT3 (282 ng/dL in female infants vs 480 ng/dL in male infants). It is possible that some of the 5 infants in whom we were unable to determine serum TBG may have increased serum T4 binding but not TBG excess. Individual thyroid function tests for the 5 infants with increased serum T4 binding but not TBG excess are presented in Table IV. Serum TBG was normal in all 5 infants. Only one mother (subject 5) had clearly elevated T4 binding to proteins. The serum TT4 and TT3 concentrations were not different between the 3 female and 2 male infants; serum TT4 was 19.9 versus 20.0 lg/dL, respectively, and TT3 was 310 versus 304 ng/dL, respectively. Although the number of subjects was small, these results fit an autosomal disorder better, such as dysalbuminemic hyperthyroxinemia or a mutation of the 298
LaFranchi et al
mean TT4 (mg/dL)
22.76 22.67 8.2-19.9
Filter Paper #2 SD
Age
3.25 3.23
15 d 14 d
Mean TT4 (mg/dL)
23.71 16.90 8.1-15.7
SD
4.74 3.38
transthyretin gene leading to an elevation of this binding protein. Thyroid function tests in the two infants with thyroid hormone resistance are presented in Table V. Thyroid hormone resistance is inherited as an autosomal dominant disorder. Measurement of thyroid function was normal in both mother and father of both subjects. Serial thyroid function testing out to a year of age continued to show abnormal elevations of serum T4, FT4, T3, and FT3, with normal TSH concentrations in both subjects. We carried out sequence analysis of the TR-b gene carboxy-terminus (exons 7 through 10) in both infants. No coding sequence mutations were found in subject 1. In subject 2, we found a point mutation (G1663A) changing codon 460 from GAG to AAG, resulting in a glutamine-to-lysine substitution. There are no reported mutations of amino acid 460 associated with thyroid hormone resistance. The G460L mutation was introduced into a human TR-b cDNA, and T3 binding affinity of the in vitro expressed protein was compared with wild-type TR-b. T3 binding was not significantly diminished in the TR-b G460L compared with wild-type TR-b. Additional in vitro functional characterization, however, has shown that TR-b G460L antagonizes wild-type TR in T3-dependent gene activation and repression (Singh and Brent, manuscript in preparation). These findings are consistent with those from other resistance to thyroid hormone-associated TR-b mutations. Both of these infants are currently followed with no treatment.
DISCUSSION Overall, we detected 17 infants with confirmed hyperthyroxinemia of 80,884 newborn infants. The incidence of these combined disorders is approximately 1 in 4750, which, as described above, probably is an underestimate. Our main goal in follow-up of newborn infants with elevated screening T4 concentrations was to detect infants with some form of hyperthyroidism who would benefit from early detection and treatment. However, we did not detect any infant with unrecognized neonatal Graves’ disease or activating mutations of the TSH receptor or the G-protein a-subunit. At an estimated incidence of 1:25,000 for neonatal Graves’ disease, we would have expected 3 or 4 cases in our screened population. In fact, we were aware of cases, but they The Journal of Pediatrics September 2003
Table II. Serum thyroid test results (means) in 17 infants with confirmed persistent hyperthyroxinemia and estimated (minimal) incidence Disorder
n
TBG excess Increased T4 binding Thyroid hormone resistance Transient hyperthyroxinemia
10 5 2 84
Estimated incidence
TT4 (mg/dL)
FT4 (ng/dL)
TT3 (ng/dL)
FT3 (pg/dL)
TSH (mU/L)
1:8088 1:16,177 1:40,442
23.3 20.0 22.9 13.6
1.8 1.5 4.1 1.6
401 307 305 243
527 514 730 480
1.94 2.80 2.07 2.43
8.1-15.7
0.8-2.7
105-245
180-770
1.7-9.1
Normal range for age
Table III. Thyroid function test results in infants and mothers with TBG excess Infant Patient No. 1 2 3 4 5 6 7 8 9 10 Normal range for age
Mother
TT4 FT4 TT3 FT3 TSH TBG Sex (mg/dL) (ng/dL) (ng/dL) (pg/dL) (mU/L) (mg/dL) M M M F F M M M F F
28.2 17.1 29.0 17.9 18.3 20.9 33.5 28.8 22.6 17.0
1.5 2.3 2.0 2.1 2.1 1.8 1.5 1.1 1.8 1.7
8.1-15.7
0.9-2.2
718 342 419 323 280 414 518 471 292 232
775 545 509 552 458 638 612 500 409 370
105-245 180-770
1.91 1.53 — 3.96 1.69 1.72 1.71 1.12 2.59 1.26
1.7-9.1
— ***
— — — 4.1 7.1 5.3 3.9 4.1
1.7-3.7
TT4 FT4 TT3 FT3 TSH (mg/dL) (ng/dL) (ng/dL) (pg/dL) (mU/L) 11.3 10.9 16.7 7.2 7.7 7.5 13.3 14.3 14.7 8.1
0.9 1.0 1.5 1.2 1.3 1.4 1.0 — 8.1 1.1
316 179 268 119 95 167 262 282 390 146
319 268 364 287 252 338 366 324 1555 283
2.51 1.28 0.71 2.64 0.29 0.98 1.59 1.07 0.13 1.13
5.6-13.7
0.8-2.7
87-180
210-440
0.40-4.2
*T3RU 18% (normal range, 23%-38%).
all initially had hypothyroidism, as their mothers were known to have Graves’ disease and were receiving antithyroid drug therapy. Antithyroid drugs cross to the fetus; depending on the dose, newborn infants often have transient hypothyroidism until the drug is metabolized and excreted by the neonate. A combination of stimulating and blocking antibodies may also cloud the initial presentation.7 Neonatal Graves’ disease has been detected by using tests on the newborn screening specimen.8 However, in these cases, the mothers were already known to have Graves’ disease, and so the newborn screening specimen was singled out for testing. We are not aware of any case of unrecognized neonatal Graves’ disease detected by follow-up after T4 population screening. Congenital hyperthyroidism resulting from an activating mutation of the TSH receptor has been reported.9-11 In three cases cited, the infants had clinical features of hyperthyroidism and a goiter; in the other, a mother and her two sons had congenital hyperthyroidism.12 Thyroid function tests confirmed hyperthyroidism; in all cases, there was no evidence of autoimmune hyperthyroidism. Activating mutations of the TSH receptor genes were documented in each case; the first three were de novo mutations. All patients were treated initially with antithyroid drugs; one case underwent subtotal Follow-up of Newborns With Elevated Screening T4 Concentrations
thyroidectomy at 8.7 years followed by radioactive iodine treatment at 9.2 years of age.9 Neurodevelopmental testing was reported only in this latter case; his IQ was 75 to 85.9 The incidence of this disorder is unknown. Although newborn screening results were not reported, it is clear from these three cases that activating mutations of the TSH receptor are associated with elevated neonatal T4 concentrations that might be detected by screening. Hyperthyroidism occurs in 20% to 42% of children with McCune-Albright syndrome, resulting from somatic cell mutations of the G-protein a-subunit within the thyroid.13,14 Hyperthyroidism present at birth has been reported in at least two cases.15,16 It is clear from these two cases that at least some infants with activating mutations of the G-protein a-subunit have elevated neonatal T4 concentrations; we are unaware of any published reports of detection by newborn screening. The incidence of this disorder is unknown. In patients with thyroid hormone resistance, serum T4 (and T3) are elevated as a result of thyroid hormone receptor mutations, decreased pituitary T3 binding, and increased TSH secretion. Thyroid hormone resistance has been detected prenatally17 and in the newborn period. In one case, the infant’s mother was known to have thyroid hormone 299
Table IV. Thyroid function test results in infants and mothers with increased serum T4 binding but not TBG excess* Infant
Mother
Patient TT4 FT4 TT3 FT3 TSH TBG No. Sex (mg/dL) (ng/dL) (ng/dL) (pg/dL) (mU/L) (mg/dL) 1 2 3 4 5
M F F M F
19.3 17.2 21.5 20.8 21.0
1.6 1.6 1.2 1.7 1.6
272 284 362 336 283
526 517 473 504 550
1.77 3.44 1.65 4.17 2.95
3.4 3.1 3.1 3.3 2.3
TT4 FT4 TT3 FT3 (mg/dL) (ng/dL) (ng/dL) (pg/dL) 8.0 7.3 9.5 8.6 16.7
0.8 0.8 1.2 1.5 1.1
163 185 178 146 152
381 345 332 335 310
TSH (mU/L) 1.13 1.01 0.52 0.43 0.95
*See Table III for normal range for age.
Table V. Thyroid function test results in infants with thyroid hormone resistance and their mothers and fathers* Patient TT4 FT4 TT3 FT3 TSH No. Sex (mg/dL) (ng/dL) (ng/dL) (pg/dL) (mU/L)
1
2
F
F
22.4
23.3
3.6
Infant 246
4.6
Infant 364
491
968
2.34
1.8
TT4 FT4 TT3 FT3 (mg/dL) (ng/dL) (ng/dL) (pg/dL)
10.7
2.4
9.4
1.2
7.0
1.2
7.9
1.1
Mother 162 Father 190 Mother 116 Father 158
TSH (mU/L)
383
0.66
358
0.95
248
0.62
332
0.56
*See Table III for normal range for age.
resistance.18 At 2 days of life, testing using the newborn screening specimen showed a TT4 of 51 lg/dL and a TSH of 26 mU/L. The other case was detected by routine neonatal cord blood screening in Hong Kong.19 Other case reports show even higher TSH elevations when using the newborn screening specimen.20 Children with this disorder may have clinical features compatible with hypothyroidism, including growth retardation, delayed skeletal maturation, learning disabilities, attention deficit-hyperactivity disorder, sensorineural deafness, and strabismus. There are reports that some of these features may improve with T3 treatment.21 This has led to the suggestion by some investigators to consider screening newborn infants for thyroid hormone resistance.22 Our results suggest the minimum incidence of this disorder is approximately 1 in 40,000. One of the two infants with thyroid hormone resistance (patient 1) did not have a TR-b gene mutation identified. Among families with clinical features of thyroid hormone resistance, approximately 10% do not have TR gene mutations.23 Defects in a variety of nuclear receptor cofactors may be involved, although these associations have not yet been shown for patients with thyroid hormone resistance without TR-b gene mutations.24 Familial TBG excess, an X-linked disorder, is estimated to occur in approximately 1 in 25,000 individuals.25 Heterozygous female infants usually have TBG concentrations intermediate between normal and affected male infants, a finding similar to our results. Griffiths et al reported 5 cases of TBG 300
LaFranchi et al
excess detected during newborn screening for congenital hypothyroidism out of a population of 30,108 in Birmingham, England, an incidence of approximately 1 in 6000.26 This figure is closer to our finding, 1 in 8000. Thus, either hereditary TBG excess is more common than previously estimated, or some acquired forms are included in estimates. Familial dysalbuminemic hyperthyroxinemia may be as common as TBG excess27 or more common in some populations.28 It may be that some of our 5 subjects with evidence of increased T4 binding, not TBG excess, have this disorder. To our knowledge, there have been no reports of infants with dysalbuminemic hyperthyroxinemia detected by newborn screening. In summary, follow-up of infants with elevated screening T4 concentrations led to an incidence of 1 in 4750 for all hyperthyroxinemia disorders combined. However, we did not detect any infant with hyperthyroidism, the case most likely to benefit from early diagnosis and treatment. Treatment of infants with thyroid hormone resistance remains unclear. Although infants have features of hypothyroidism, and some physicians have prescribed large doses of T4 and/or T3 treatment,18 most patients with generalized thyroid hormone resistance are followed without treatment. Infants with elevations of TBG or other binding proteins have normal thyroid function and do not require treatment. Some would argue that detection would help to prevent inappropriate treatment, as some patients are mistakenly diagnosed as having The Journal of Pediatrics September 2003
a hyperthyroid state. Our screening program continues to follow-up infants with T4 concentrations >3.5 SD, so that we can gain experience with a larger population over a longer time period. We chose this higher cutoff to reduce the cost of follow-up and the number of false-positive cases. On balance, we do not recommend that follow-up of such infants become part of routine newborn screening programs at this time.
REFERENCES 1. LaFranchi S, Hanna CE. Graves’ disease in the neonatal period and childhood. In: Braverman L, Utiger RD, eds. Werner and Ingbar’s the thyroid. 8th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2000. p. 989-97. 2. Daneman D, Howard NJ. Neonatal thyrotoxicosis: intellectual impairment and craniosynostosis. J Pediatr 1980;97:257-9. 3. Refetoff S. Inherited thyroxine-binding globulin abnormalities in man. Endocrinol Rev 1989;10:275-93. 4. Nelson JC, Clark SJ, Borut DL, Tomei RT, Carlton EI. Age-related changes in serum free thyroxine during childhood and adolescence. J Pediatr 1993;123:899-905. 5. Adams LM, Emery JR, Clark SJ, Carlton EI, Nelson JC. Reference ranges for newer thyroid function tests in premature infants. J Pediatr 1995;126:122-7. 6. Elmlinger MW, Kuhnel W, Lambrecht H-G, Ranke MB. Reference intervals from birth to adulthood for serum thyroxine (T4), triiodothyronine (T3), free T3, free T4, thryoxine binding globulin (TBG) and thyrotropin (TSH). Clin Chem Lab Med 2001;39:973-9. 7. Zakarija M, McKenzie JM, Hoffman WH. Prediction and therapy of intrauterine and late-onset neonatal hyperthyroidism. J Clin Endocrinol Metab 1986;62:368-71. 8. Walker P, Dussault JH, Hart IR, Langelier P, Szots F. Thyrotoxicosis detected in a mass-screening program for neonatal hyperthyroidism: demonstration of placental transfer of an immunoglobulin with marked lipolytic activity. J Pediatr 1997;91:400-3. 9. Kopp P, Van Sande J, Parma J, Duprez L, Gerber H, Joss E, et al. Brief report: congenital hyperthyroidism caused by a mutation in the thyrotropinreceptor gene. N Engl J Med 1995;332:150-4. 10. De Roux N, Polak M, Couet J, Leger J, Czernichow P, Milgrom E, et al. A neomutation of the thyroid-stimulating hormone receptor in a severe neonatal hyperthyroidism. J Clin Endocrinol Metab 1996;81:2023-36. 11. Tonacchera M, Agretti P, Rosellini V, Ceccarini G, Perri A, Zampolli M, et al. Sporadic nonautoimmune congenital hyperthyroidism due to a strong activating mutation of the thyrotropin receptor gene. Thyroid 2000;10: 859-63. 12. Schwab KO, Gerlich M, Broecker M, Sohlemann P, Dervahl M, Lohse MJ. Constitutively active germline mutation of the thyrotropin receptor gene as a cause of congenital hyperthyroidism. J Pediatr 1997;131:899-904.
Follow-up of Newborns With Elevated Screening T4 Concentrations
13. Feuillan PP, Shawker T, Rose SR, Jones J, Jeevanram RK, Nissula BC. Thyroid abnormalities in the McCune-Albright syndrome: ultrasonography and hormonal studies. J Clin Endocrinol Metab 1990;71:1596-601. 14. Lee PA, Van Dop C, Migeon CJ. McCune-Albright syndrome: longterm follow-up. JAMA 1986;256:2980-4. 15. Mastorakos G, Mitsiades NS, Doufas AG, Koutras DA. Hyperthyroidism in MCCune-Albright syndrome with a review of thyroid abnormalities sixty years after the first report. Thyroid 1997;7:433-9. 16. Yoshimoto M, Nikayama M, Baba T, Uehara Y, Niikawa N, Ito M, et al. A case of neonatal McCune-Albright syndrome with Cushing’s syndrome and hyperthyroidism. Acta Paediatr Scand 1991;984-7. 17. Asteria C, Rajanayagam O, Collingwood TN, Persani L, Romoli R, Mannavola D, et al. Prenatal diagnosis of thyroid hormone resistance. J Clin Endocrinol Metab 1999;84:405-10. 18. Weiss RE, Balzano S, Scherberg NH, Refetoff S. Neonatal detection of generalized resistance to thyroid hormone. JAMA 1990;264:2245-50. 19. Wong GWK, Skek CC, Lam STS, Tsui MKM, Leung SSF. Detection of resistance to thyroid hormone by cord blood screening. Acta Paediatr 1995;84:335-6. 20. Pohlenz J, Schonberger W, Koffler T, Refetoff S. Resistance to thyroid hormone caused by a new mutation (V336M) in the thyroid hormone receptor b gene. Thyroid 1999;9:1001-4. 21. Hauser P, Zametkin AJ, Martinez P, Vitiello B, Matochik JA, Mixson AJ, et al. Attention deficit-hyperactivity disorder in people with generalized resistance to thyroid hormone. N Engl J Med 1993;328:997-1001. 22. Pass KA, Hedden JE, Morris JE, Hauser P, Weintraub BD, Mizejewski GJ. Newborn screening for thyroid resistance: implications for attention deficit-hyperactivity disorder. In: Farriaux J-P, Dhondt J-L, eds. New horizons in neonatal screening. Amsterdam (The Netherlands): Elsevier Science BV; 1994. p. 141-4. 23. Reutrakul S, Sadow PF, Pannain S, Pohlenz J, Carvalho GA, Macchia PE, et al. Search for abnormalities of nuclear corepressors, coactivators, and a coregulator in families with resistance to thyroid hormone without mutations in thyroid hormone receptor b or a genes. J Clin Endocrinol Metab 2000;85:3609-17. 24. Pohlenz J, Weiss RE, Macchia PE, Pannain S, Lau IT, Ho H, et al. Five new families with resistance to thyroid hormone not caused by mutations in the thyroid hormone receptor b gene. J Clin Endocrinol Metab 1999;84: 3919-28. 25. Refetoff S. Inherited thyroxine-binding globulin abnormalities in man. Endocrinol Rev 1989;10:275-93. 26. Griffiths KD, Virdi NK, Rayner PH, Green A. Neonatal screening for congenital hypothyroidism by measurement of plasma thyroxine and thyroid stimulating hormone concentrations. BMJ 1985;291:117-20. 27. Jensen IW, Fabaer J, Grunnet N. Familial occurrence of dysalbuminemic hyperthyroxinemia, lipomatosis and ankylosing spondylitis. Scand J Rheumatol 1990;19:303-5. 28. Sunthornthepvarakul T, Angkeow P, Weiss RE, Hayashi Y, Refetoff S. An identical missense mutation in the albumin gene results in familial dysalbuminemic hyperthyroxinemia in 8 unrelated families. Biochem Biophys Res Commun 1994;202:781-7.
301