| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Pediatric Endocrinology (M.J.E.K., A.S.P.v.T., D.A.v.T., E.B., B.M.W., J.J.M.d.V., T.V.), Department of Clinical Chemistry, Laboratory of Endocrinology (E.E.), Emma Children’s Hospital, Academic Medical Center, University of Amsterdam, 1100 DE Amsterdam, The Netherlands
Address all correspondence and requests for reprints to: M. J. E. Kempers, M.D., Emma Children’s Hospital Academic Medical Center, Department of Pediatric Endocrinology, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands. E-mail: m.j.kempers{at}amc.uva.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Methods: Retrospectively, plasma FT4, TSH, and T3 concentrations were analyzed in T4-supplemented children aged 0.5–20.0 yr with thyroidal CH, central (secondary or tertiary) CH, or autoimmune thyroid disease and in control children with type 1 diabetes mellitus.
Results: When TSH was within the reference range (0.4–4.0 mU/liter), mean FT4 in thyroidal CH [1.65 ng/dl; 95% confidence interval (CI), 1.62–1.67] was significantly higher than in autoimmune thyroid disease (1.15 ng/dl; 95% CI, 1.11–1.19) and diabetes (1.08 ng/dl; 95% CI, 1.06–1.10). In central CH, when TSH was less than or equal to 0.02 mU/liter, mean FT4 was 1.27 ng/dl (95% CI, 1.24–1.29). When FT4 was within the reference range (0.78–1.79 ng/dl), 43% of the TSH measurements in thyroidal CH were more than 4.0 mU/liter, compared with 18% in autoimmune thyroid disease and 0% in type 1 diabetes mellitus; in central CH, 95% of TSH measurements were less than 0.4 mU/liter.
Conclusions: In T4-supplemented patients with thyroidal CH, when TSH concentrations are established within the reference range, FT4 concentrations tend to be elevated, and vice versa. Because this phenomenon could not be observed in acquired thyroidal hypothyroidism, we hypothesize that a pre- and/or perinatal hypothyroid state shifts the setpoint of the thyroid’s regulatory system. In central CH, when FT4 concentrations are established within the reference range, the pituitary secretes only minute amounts of TSH. For monitoring T4 supplementation, reference ranges for FT4 and TSH should be adapted to the etiology of hypothyroidism.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Specific clinical determinants that accurately reflect the patient’s thyroid hormone state are lacking. Therefore, diagnosis and treatment monitoring of thyroid disease depend on the measurement of thyroid function determinants in blood. It appears a matter of course that especially the plasma free T4 (FT4) concentration is brought to and maintained at an adequate level after the diagnosis of (congenital) hypothyroidism, because the brain preferentially uses T4 for its intracellular T3 provision (10). However, generally, the plasma concentration of TSH is considered the most sensitive and objective indicator of the thyroid hormone state in thyroidal (primary) hypothyroidism (11, 12, 13).
Previously we have shown that during the first weeks of T4 supplementation in newborns with thyroidal CH (CH-T), their plasma FT4 concentrations could be established within the age-specific reference range within 3–4 d, whereas for plasma TSH concentrations it took 3–4 wk. We concluded that during this initial phase, plasma FT4 and not TSH is useful for monitoring T4 supplementation (14). Also, after this initial phase, TSH concentrations above the reference range are frequently encountered despite FT4 concentrations within the reference range (15, 16, 17, 18). In patients with central (secondary or tertiary) hypothyroidism, the opposite has been observed during T4 supplementation. Despite FT4 concentrations within the reference range, low or undetectable TSH concentrations are encountered. So in central hypothyroidism, FT4 is the determinant of choice for treatment monitoring (13, 19, 20).
To investigate the variability of the relationship between thyroid function determinants, we analyzed and compared plasma FT4, TSH, and T3 concentrations during long-term T4 supplementation in children with a variety of thyroid disorders and in a control population.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Patients, followed by the Pediatric Endocrinology Department of the Emma Children’s Hospital Academic Medical Center, were included when data on the etiology of hypothyroidism were available and when they were considered compliant with their T4 supplementation. Three groups were defined: CH-T, central CH (CH-C), and autoimmune thyroid disease (AITD). The first group was subdivided, according to the absence, CH-T(a), or presence, CH-T(b), of functional thyroid tissue.
Children with CH were detected by the Dutch neonatal T4-based screening program (21). Etiological classification was based upon initial presentation, thyroid function determinants, thyroid imaging, and/or TRH test results (22).
Patients with CH-T(a) (n = 40) had thyroid agenesis (n = 6), cryptopic thyroid remnants (n = 15), or total iodide organification defects (n = 19) (22). Patients with CH-T(b) had dystopic thyroid rudiments, usually sublingually located (n = 18). In patients with CH-C (n = 28), other hormonal deficiencies (e.g. caused by ACTH, somatotropin, or gonadotropin deficiency) were, if present, supplemented.
T4 supplementation in CH-T children was monitored in agreement with the guidelines of the European Society of Pediatric Endocrinology (23). In short, T4 supplementation aimed at keeping plasma TSH concentration well within the reference range, thereby maintaining a satisfactory clinical condition. In children with CH-C, T4 supplementation aimed at keeping plasma FT4 concentration well within the reference range, thereby maintaining a satisfactory clinical condition.
Children with AITD (n = 17) suffered either from Hashimoto’s disease (n = 8) or Graves’ disease (n = 9). Children with Hashimoto’s disease were supplemented with T4. Children with Graves’ disease were treated according to the block-and-replace method with antithyroid drugs and T4. Because of the antithyroid drug inhibition, children with Graves’ disease were considered as patients with (acquired) functional athyroidism. In both groups, treatment aimed at normalizing TSH concentrations and establishing a satisfactory clinical condition. Because TSH in children with Graves’ disease often remains suppressed for several months after initiation of the antithyroid drug therapy, even after normalization of FT4, the first 6 months of treatment were excluded from analysis.
Children with type 1 diabetes mellitus (T1DM) (n = 39) treated with sc insulin and without any thyroid disease were used as controls. In our hospital, children with diabetes are checked yearly for AITD (when their metabolic status is stable). If anti-thyroid peroxidase concentration was above 30 kU/liter at any time, patients were excluded for this study.
Data retrieval and analysis
When FT4, TSH, or T3 are used in the text, we imply their plasma concentrations unless otherwise stated. In the five groups of patients, FT4, TSH, and T3 obtained during routine tests between 0.5 and 20.0 yr of age were analyzed retrospectively. In total, 1969 blood samples (in which FT4, TSH, and sometimes T3 were measured) of 142 patients were analyzed retrospectively. Daily T4 dose (µg/kg body mass) was calculated, and FT4 was analyzed when TSH was between 0.4 and 4.0 mU/liter (the reference range) in CH-T(a), CH-T(b), AITD, and T1DM at different age intervals. In patients with CH-C, daily T4 dose was calculated and FT4 was analyzed at different age intervals, when TSH was no more than 0.02 mU/liter. The cutoff at 0.02 for TSH was empirically chosen; with a TSH assay detection limit of 0.01 mU/liter, 0.02 is just measurable.
FT4 and T3 were analyzed in relation to distinctive, arbitrarily chosen, TSH ranges (0.4–1.3, 1.4–2.2, 2.3–3.1, and 3.2–4.0 mU/liter) in CH-T(a), CH-T(b), AITD, and T1DM. T3 was also analyzed when TSH was slightly below (0.10–0.39 mU/liter) or above (4.1–10.0 mU/liter) the reference range.
We analyzed the log TSH in relation to distinctive, arbitrarily chosen, FT4 ranges (0.93–1.24, 1.25–1.55, 1.56–1.86, and 1.87–2.18 ng/dl). Before log transformation was performed, TSH concentrations of less than 0.02 mU/liter and less than 0.01 mU/liter were set at 0.010 and 0.005 mU/liter, respectively. The mean log TSH and its 95% confidence interval (CI) were then back transformed and presented as the geometric mean and its 95% CI.
The age groups 1, 2, and 3 represent children aged 0.5 to less than 5.0, 5.0 to less than 10.0, and 10.0–20.0 yr, respectively.
Statistical analysis
Mean, SD scores, and 95% CI were calculated using SPSS version 11 (SPSS, Inc., Chicago, IL). The one-way ANOVA was used to compare, for each etiology group, FT4 and T4 dose between the different age groups; to compare, for each age group, FT4 and T4 dose between the etiology groups; to compare, for each TSH category (0.4–1.3, 1.4–2.2, 2.3–3.1, and 3.2–4.0 mU/liter), FT4 between the etiology groups; and to compare, for each FT4 category (0.93–1.24, 1.25–1.55, 1.56–1.86, and 1.87–2.18 ng/dl), log TSH between the etiology groups. Correlation coefficients were obtained and linear regression models were applied to investigate the relationship between FT4 and TSH in CH-T, AITD, and T1DM.
Laboratory methods
FT4 and TSH were measured by time-resolved fluoroimmunoassays (Delfia Free T4 and Delfia hTSH Ultra; PerkinElmer, Wallac Oy, Turku, Finland). The reference range, as established in our laboratory, was 0.78–1.79 ng/dl for FT4 (multiply by 12.87 for pmol/liter) and 0.4–4.0 mU/liter for TSH. T3 was measured by an in-house RIA method, with a reference range of 85–176 ng/dl (multiply by 0.0154 for nmol/liter).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The dose needed to obtain TSH within the reference range decreased significantly in children with CH-T(a), CH-T(b), and CH-C comparing age groups 1 and 2 and in children with CH-T(a) and CH-C comparing age groups 2 and 3 (Fig. 1A
and Table 1). The mean T4 dose in children with CH-T(a) was significantly higher than in CH-T(b) (P = 0.014, < 0.001, and 0.003 for the different age groups, respectively) and CH-C (P < 0.001 for all three age groups). After the age of 10 yr, children with CH-T needed significantly higher T4 doses than children with AITD (P < 0.001).
|
|
In all groups, FT4 tended to be higher in infancy compared with adolescence when TSH was within the reference range (Fig. 1B
and Table 2). At all ages, children with CH-T(a) and CH-T(b) showed significantly higher FT4 than children with CH-C, AITD or T1DM (P < 0.001). After the age of 10 yr, children with CH-T(a) had higher FT4 than children with CH-T(b) (P < 0.001), and children with CH-C had higher FT4 than children with AITD (P = 0.050) and T1DM (P < 0.001).
|
). In children with CH-C, when TSH was no more than 0.02 mU/liter during T4 supplementation, mean FT4 was 1.27 ng/dl (95% CI, 1.24–1.29), significantly lower than in children with CH-T (P < 0.001) and higher than in children with AITD or T1DM (P < 0.001).
For each TSH category within the reference range, children with CH-T have significantly higher FT4 than the other groups (P < 0.001) (Fig. 2A). Comparison of FT4 of children with Graves’ disease and age-matched children with CH-T(a) revealed that when TSH was in the same range, differences in FT4 were still present (Fig. 2B).
|
When FT4 was within the reference range (0.78–1.79 ng/dl), 43% of the TSH measurements in CH-T were more than 4.0mU/liter, compared with 18% in AITD and 0% in T1DM; in CH-C, 95% of TSH measurements were less than 0.4 mU/liter.
Figure 3, A and B, shows that with comparable FT4, TSH is higher in children with CH-T compared with AITD and T1DM. When FT4 was 0.93–1.24 ng/dl, TSH (geometric mean) was significantly higher in children with CH-T(a) compared with CH-T(b) (P = 0.003), AITD (P < 0.001), and T1DM (P < 0.001) and in CH-T(b) compared with AITD (P = 0.037) and T1DM (P < 0.001). When FT4 was 1.25–1.55 or 1.56–1.86 ng/dl, TSH (geometric mean) was significantly higher in children with CH-T compared with AITD (P < 0.01). Although the regression lines of children with CH-T(a) and AITD had a similar slope, both lines are about 0.6 ng/dl apart on the horizontal axis (FT4).
|
). Plasma T3 concentration
T3 remained rather constant within the etiology groups (Fig. 4), even when TSH was somewhat beyond the reference range (0.10–10.0 mU/liter) for thyroidal hypothyroidism. Children with CH-C had a mean T3 of 163 ng/dl (95% CI, 156–169) when TSH was no more than 0.02 mU/liter.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
In this study, we show that both T4 doses and FT4 concentrations are consistently highest in children with CH-T compared with children with CH-C, AITD, or T1DM. In CH-C we have calculated the T4 dose and analyzed FT4 and T3 when TSH was less than or equal to 0.02 mU/liter. Our choice to use a TSH cutoff to analyze FT4 is debatable because TSH is generally considered useless for monitoring T4 supplementation in central hypothyroidism. At TSH less than or equal to 0.02 mU/liter, the FT4 distribution curve in the CH-C group turned out to be similar to the reference range (Fig. 1C), which indirectly justifies the TSH cutoff for the analysis of CH-C patients. The FT4 distribution curve of children with CH-T shows an upward shift of 0.5–0.6 ng/dl when TSH is within the reference range, compared with AITD and T1DM (Fig. 1C). Their TSH concentrations are shifted upwards when FT4 is within the reference range (Fig. 3, A and B). In patients with CH-C, TSH is extremely low when FT4 is within the reference range (Fig. 3C). In all groups, T3 remains stable over a wide range of TSH concentrations.
TSH elevations in children with CH-T during T4 supplementation (15, 16) have been attributed to resistance of the pituitary to thyroid hormone (15). This resistance was thought to diminish with age because TSH elevations were found more frequently among younger patients (15). Indeed, we did observe that FT4 decreases slightly with age, whereas TSH remains within the reference range (Fig. 1B). This might be a result of a gradual increase of the pituitary’s responsiveness to FT4 with age. Yet, this phenomenon is not exclusively seen in CH-T children but is ubiquitous among all treated as well as control children. We suppose that the high frequency of elevated TSH reported in CH-T infants is the consequence of a higher need of T4 per kilogram (Fig. 1A) and a faster increase of body mass in younger compared with older children, which increase the risk to outgrow the T4 dose. This phenomenon can probably be prevented by more frequent testing and earlier dose adjustments especially in young children.
Apparently in T4-supplemented children with CH-T, the setpoint of the thyroid’s regulatory system, defined as the individual’s specific combination of mutually dependent FT4 and TSH, differs substantially from untreated control children with T1DM and from T4-supplemented children with AITD, at least up to adulthood. In contrast to controls, patients treated for hypothyroidism will have nonphysiological circadian variations in FT4 and TSH, as a consequence of a once-a-day T4 dose scheme (24, 25). However, children with CH-T(a) were shown to have substantially higher FT4 than children with Graves’ disease (also having functional athyroidism because of antithyroid drugs), despite a similar T4 supplementation strategy. Therefore, the treatment strategy cannot explain the difference in setpoint.
In patients with severe thyroidal hypothyroidism (either congenital or acquired), the secretion of both T4 and T3 is negligible. Also in patients with moderate thyroidal hypothyroidism, the endogenous production of T4 and T3 becomes very low once TSH is normalized by treatment. The T4 supplementation therapy has to compensate for the impaired thyroidal secretion of T4 as well as that of T3. Because this is similar in congenital and acquired thyroidal hypothyroidism, it does not explain the difference in dose of T4 to maintain euthyroidism.
A fundamental difference between patients with congenital and acquired hypothyroidism is the timing of onset of thyroid hormone deficiency. Despite the maternal-fetal transfer of T4, patients with CH have decreased thyroid hormone concentrations during the period that the hypothalamic-pituitary-thyroid system matures toward an integrated system for control of the thyroid hormone state (26, 27, 28, 29). In patients with CH-T, this condition is accompanied by strongly increased TSH concentrations (29, 30). The normalization of TSH usually takes several weeks after initiating T4 supplementation, whereas FT4 is within the reference range within a few days (14). It is largely unknown which processes are involved in establishing and maintaining the setpoint. Previously we have demonstrated that after exposure of the fetus to a hyperthyroxinemic environment caused by inadequately treated maternal Graves’ disease, the child’s thyroid regulatory system may become disturbed up to at least several months after birth (21). Apparently, a pathological condition like CH-T, with decreased (F)T4 and increased TSH concentrations during the early phase of life, can induce a setpoint shift as well.
If a fetal hypothyroid state causes isolated pituitary hyporesponsiveness to thyroid hormone, it might be less obvious to rely on TSH as an adequate indicator of the overall thyroid hormone state. Although TSH of children with CH-T is still sensitive to changes in FT4 (Fig. 3, A and B), increased FT4 concentrations are needed to establish TSH within the reference range. Such increased FT4 concentrations are potentially harmful for the brain and other target tissues, unless fetal hypothyroidism induces generalized hyporesponsiveness to thyroid hormone. In that case, the increased FT4 concentrations must be considered the proper adaptation to realize normal metabolism. Indeed, we generally experience that patients with CH-T maintain a satisfying clinical and mental condition when TSH is kept well within its reference range. This, however, could also be explained by an alternative hypothesis that normalization of TSH itself optimizes the clinical condition (despite elevated FT4 concentrations). However, although the TSH receptor is known to be expressed in, e.g. brain tissue (31), until now any evidence is lacking that a normal TSH receptor occupation is more important than a normal thyroid hormone receptor occupation.
In patients with central hypothyroidism, the lack of integrity of the regulatory system inhibits the pituitary to produce sufficient amounts of bioactive TSH. During treatment, when FT4 is kept within the reference range, CH-C patients have significantly decreased TSH concentrations. So, in contrast to CH-T, patients with CH-C are never subject to increased TSH concentrations, neither during the fetal period nor during T4 supplementation. However, if exposure to pre- and perinatal decreased T4 concentrations induces a generalized hyporesponsiveness to thyroid hormone, CH-C patients might also need an upward-shifted FT4 reference range to normalize metabolism. Certainly, when thyroid hormone deficiency is part of multiple pituitary hormone deficiency, the treatment goal to achieve and maintain a fulfilling clinical and mental condition is even more challenging because of the simultaneous supplementation of multiple hormones. On the condition that the other hormones are adequately supplemented, we experience that the well-being of patients with CH-C is maintained by FT4 concentrations well within the reference range, which seems to be more in line with the alternative hypothesis.
In conclusion, in patients with CH-T, establishing FT4 well within the reference range will result in elevated TSH concentrations, whereas TSH concentrations within the reference range can be accomplished only by elevated FT4 concentrations. Because this phenomenon is not seen in acquired thyroidal hypothyroidism, we hypothesize that hypothyroidism in prenatal life is responsible for the apparent shift in the setpoint of the thyroid’s regulatory system. In patients with CH-C, TSH concentrations are strongly decreased when FT4 is kept well within the reference range. Apparently, while treating any CH patient, it is inevitable to use an adapted reference range for FT4 and/or for TSH. Future prospective research should reveal which strategy is most capable to ascertain optimal tissue metabolism, especially in the developing brain.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AITD, Autoimmune thyroid disease; CH, congenital hypothyroidism; CH-C, central CH; CH-T, thyroidal CH; CH-T(a), CH-T with no functional thyroid tissue; CH-T(b), CH-T with functional thyroid tissue; CI, confidence interval; FT4, free T4; T1DM, type 1 diabetes mellitus.
Received January 31, 2005.
Accepted April 4, 2005.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. F. La Gamma, A. G. van Wassenaer, S. Ares, S. G. Golombek, J. H. Kok, J. Quero, T. Hong, M. H. Rahbar, G. M. de Escobar, D. A. Fisher, et al. Phase 1 Trial of 4 Thyroid Hormone Regimens for Transient Hypothyroxinemia in Neonates of <28 Weeks' Gestation Pediatrics, August 1, 2009; 124(2): e258 - e268. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. St. Germain, V. A. Galton, and A. Hernandez Defining the Roles of the Iodothyronine Deiodinases: Current Concepts and Challenges Endocrinology, March 1, 2009; 150(3): 1097 - 1107. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. E. Kempers, A. S. P. van Trotsenburg, R. R. van Rijn, A. M. J. B. Smets, B. J. Smit, J. J. M. de Vijlder, and T. Vulsma Loss of Integrity of Thyroid Morphology and Function in Children Born to Mothers with Inadequately Treated Graves' Disease J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 2984 - 2991. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Quignodon, C. Grijota-Martinez, E. Compe, R. Guyot, N. Allioli, D. Laperriere, R. Walker, P. Meltzer, S. Mader, J. Samarut, et al. A combined approach identifies a limited number of new thyroid hormone target genes in post-natal mouse cerebellum J. Mol. Endocrinol., July 1, 2007; 39(1): 17 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. E. Kempers, L. van der Sluijs Veer, M. W. G. Nijhuis-van der Sanden, L. Kooistra, B. M. Wiedijk, I. Faber, B. F. Last, J. J. M. de Vijlder, M. A. Grootenhuis, and T. Vulsma Intellectual and Motor Development of Young Adults with Congenital Hypothyroidism Diagnosed by Neonatal Screening J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 418 - 424. [Abstract] [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 |
0.02 mU/liter for CH-C]. Concentrations are expressed as mean surrounded by CI. Within all groups, data of children 16.5–20.0 yr of age are grouped together and labeled 
16.5. Red filled square, CH-T(a); green filled square, CH-T(b); blue 
, CH-C; pink 
, AITD; brown X, T1DM. C, Distribution curves of FT4 of the different groups, when the simultaneously measured TSH is within the reference range (same as for A and B). The table in the right upper corner represents FT4 mean ± 2 SD for each group. The bar that incorporates the mean FT4 concentration of the etiology group is striped.
