Clinical Course of Congenital Hypothyroidism With Gland In Situ and Necessity of L-Thyroxine Therapy After Re-evaluation

Abstract

Context

The clinical course and need for long-term L-thyroxine (LT4) therapy of congenital hypothyroidism (CH) with gland in situ (GIS) remain unclear.

Objective

To describe the clinical history of CH with GIS and evaluate the proportion of patients who can suspend therapy during follow-up.

Design and Setting

Retrospective evaluation of patients followed at referral regional center for CH of Pisa.

Patients

Seventy-seven patients with confirmed primary CH and GIS after positive neonatal screening were included. All children started LT4 upon CH confirmation on serum assay.

Interventions

At 3 years of age, 55 children underwent a clinical reevaluation after withdrawal of therapy with hormonal examinations, imaging of the thyroid gland with ultrasonography, and 123-iodine with perchlorate discharge test. Subsequent periodic controls of thyroid function were executed, and, when possible, a new attempt to stop LT4 was performed. Adequate follow-up data (at least 6 months after treatment suspension trial) were available for 49 patients.

Results

Among the 55 patients who were reassessed, 18 (32.7%) were euthyroid. Considering subsequent follow-up, 49% of patients were no longer treated and 51% were taking therapy. No differences in neonatal parameters were observed between the 2 groups; LT4 dose before the last trial off medication was higher in permanent CH (P .016).

Conclusion

Monitoring thyroid function in children with CH and GIS is necessary to evaluate the need for substitution and avoid overtreatment. Even if therapy can be suspended, patients need to be monitored because apparently normal thyroid function may decline several months after withdrawal of LT4.

Congenital hypothyroidism (CH) is the most common endocrine disease of childhood, and it is defined as a condition of insufficient thyroid hormone supply to the conceptus beginning in utero (1). Owing to the essential role played by thyroid hormone in growth and brain development, CH can lead to growth retardation and neurological impairment if not immediately treated. The introduction of the neonatal screening program has allowed an early diagnosis of the disease, preventing these consequences through early hormone replacement therapy (2).

In recent years, an increase in CH incidence has been reported, mostly due to the lowering of TSH threshold used in neonatal screening, which allowed detection of mild defects in hormonogenesis, particularly with a normally located gland of normal size. In Italy, an increased incidence of CH has been documented (from 1:3000-1:4000 in 1980 to 1:2000 in 2022) (3). Other demographic factors are implicated, such as the more frequent use of assisted reproductive technologies, with an increase in the number of multiple pregnancies and preterm deliveries, which are risk factors for neonatal thyroid dysfunctions (4). With neonatal screening, executed through analysis of a blood drop from the heel in the first 3 days of life, if elevated TSH levels are found, the neonate is recalled to assess thyroid function with a serum assay (excluding false positive) and, if indicated, starts L-thyroxine (LT4) therapy. In some cases, such as preterm delivery, birth in an intensive neonatal care unit, low birth weight, Down's syndrome, or multiples birth, collection of a second specimen at 10 to 14 days of life is necessary to avoid false negatives in the TSH-based neonatal screening programs (5).

The pathogenetic cause of CH with gland in situ (GIS) remains largely unknown, even if molecular analyses of genes implicated in the synthesis of thyroid hormones and iodine metabolism, such as NIS, DUOX2, TG, DHEAL-1, TPO, and TSH receptor, has led to the identification in some cases of mutations that can be responsible for this dysfunction (6).

At present, long-term evolution of these forms of CH is unclear; in fact, clinical phenotypes have a great variability; duration of hypothyroidism is also variable. Different strategies of reevaluation have been adopted to determine the necessity of treatment during and after childhood. In fact, in different cohorts, a significative prevalence of a transient form of CH has been demonstrated, but the right timing of suspension and appropriate monitoring strategies thereafter remain unclear. The aim of this study is to describe the clinical evolution of CH in patients with thyroid gland in place of normal dimension and to evaluate the proportion of patients who can discontinue substitutive treatment during follow-up.

Patients and Methods

We selected 77 patients born between 1987 and 2018 in Italy, from regions characterized by a borderline-mild iodine deficiency, and subsequently referred to our center after positive neonatal screening. All patients were screened for CH utilizing a blood sample obtained from the baby's heel between 48 and 72 hours of life, subsequently confirmed by a serum analysis, and they had a normally located thyroid gland of normal size detected through ultrasonography (US).

Initial Data Recording

The neonatal screening was performed in elected screening centers, adopting a primary TSH method; the TSH cut-off used in our center has been progressively lowered: it was 20 mIU/L in the period between 1982 and 2002, 15 mIU/L between 2003 and 2015, 10 mIU/L between 2015 and 2016, and 9 mIU/L since 2017. When TSH exceeded the threshold, an assay on neonatal serum to measure TSH, free T4 and free T3 was done. The following information was retrospectively recorded: gestational age at birth (preterm birth defined as delivery <37 gestational weeks) and birth weight, maternal or neonatal history of iodine exposure, maternal history of thyroid disease (and, if possible, a maternal thyroid function profile and thyroid autoimmunity panel), pregnancy and perinatal history, family history of thyroid disease, blood spot TSH level. We collected data regarding first recall after screening, in which serum TSH, FT4, FT3, thyroglobulin (Tg), and urinary iodine excretion have been measured and size and position of thyroid gland have been assessed through a neck US or scintigraphy. All patients with Down's syndrome or other chromosomal abnormalities and thyroid dysgenesis, including agenesis and ectopic and hypoplastic thyroid gland, were excluded from the study. Subjects with central hypothyroidism were also excluded. Children born prematurely or with low birth weight were included in the follow-up study if hypothyroidism was confirmed at subsequent visits (TSH persistently higher than 10 mIU/L with low or normal FT4 values), while transient forms with spontaneous resolution within the first year of life were secondarily excluded.

Clinical Reassessment and Follow-up

After 3 years of age, in a group of patients (minimum follow-up from the start of therapy 6 months), LT4 therapy was suspended for 4 weeks, to perform a clinical and hormonal retesting after withdrawal. We retrospectively collected biochemical data regarding thyroid function, at least FT4, FT3, TSH, serum Tg, antithyroglobulin thyroid autoantibodies (Tg-Ab), antithyroperoxidase thyroid autoantibodies (TPO-Ab), and urinary iodine excretion at clinical reassessment. In all patients, a neck US and, in some cases, a scintigraphy with 123-iodine (123-I) and perchlorate discharge test were performed.

Based on thyroid function after LT4 withdrawal, patients were divided into 3 groups:

1. Normal thyroid function or euthyroidism, defined by TSH level <4 mIU/L and normal FT4 level;

2. Hyperthyrotropinemia, defined by TSH level between 4 and 10 mIU/L with normal FT4 level;

3. Hypothyroidism, defined by TSH level >10 mIU/L with normal or below normal FT4 level.

The resumption of LT4 treatment was managed by clinicians on an individual basis. If treatment was restarted after clinical reassessment, age, thyroid function, and LT4 dosage at restart of treatment were recorded.

For all patients, the follow-up period was defined as the time interval between the first and the last access to our center. We recorded the following information from the last visit: thyroid function profile with TSH, FT3, FT4, and LT4 dosage; when available, we recorded Tg-Ab and TPO-Ab positivity. For patients who suspended therapy during the follow-up period, we recorded age and LT4 dosage at cessation of treatment and duration of follow-up period after suspension.

Based on a longer follow-up data, we divided patients into 2 groups: 1 group was composed of patients who at the end of follow-up were off LT4 (that was stopped at clinical reassessment or at any time during the following years) and who were euthyroid for an adequate period after withdrawal (transient hypothyroidism); the other group included patients who were still treated with LT4 at the end of follow-up (permanent hypothyroidism). We excluded from the outcome comparison patients with an insufficient time of follow-up after suspension of LT4 therapy (ie, <6 months) (Fig. 1).

Thyroid Function Tests

The values of the free fractions of thyroid hormones were assayed at our department by immunoradiometric assay (Vitro System, Ortho-Clinical Diagnostics, Rochester, NY, USA). TSH was measured with a chemoluminescent method (Immulite 2000, DPC, Los Angeles, CA, USA). Tg-Ab and TPO-Ab were measured by enzyme immunoassay two-step (AIA-Pack Tg-Ab and TPO-Ab; Tosoh, Tokyo, Japan; RRID:AB_290885; RRID:AB_2920886).

Serum Tg was measured by chemoluminescent immunometric assay (Beckman UniCeL Dx1800, Access Thyroglobulin, Beckman Coulter Inc, Brea, CA, USA, RRID:AB_2756379).

Thyroid US

Thyroid US examination was performed using an US apparatus to real time (TechnosEsaote Biomedical) connected to a linear transducer (7.5 MHz). Thyroid volume was calculated according to the formula of the ellipsoid model.

Thyroid Scintigraphy and Perchlorate Discharge Test

The test was performed according to the method described by Meller et al in 1997 (7). Three hours after the oral administration of 123-I, activity 2.96 MBq, a frontal image of the thyroid was acquired, for 5 minutes, with a gamma-camera with a low energy, high resolution collimator (AtomLab 960 Thyroid Uptake System). 123-I uptake was calculated as the ratio of the activity detected on the thyroid and the activity of a syringe containing a known amount of 123-I. Then, potassium perchlorate (KClO₄, 400 mg) was administered orally immediately after scintigraphy. An hour later, a thyroid scan with the same method used for the baseline image was acquired. The disposal of iodine was calculated as the ratio between the uptake value after perchlorate administration and the uptake at the third hour. A reduction >10% of the uptake was indicative of a partial iodine organification defect (PIOD) and a reduction of more than 90% was indicative of a total iodine organification defect (TIOD).

Statistics

Continuous variables were expressed as median (minimum-maximum). Comparison between groups was performed using IBM SPSS Statistics ve. 26 (IBM Co. Armonk, NY, USA) with nonparametric Mann–Whitney U-test for continuous variables and chi-square test for categorical variables. A P-value of <.05 was considered statistically significant.

Results

Description of Cases at First Evaluation

77 patients (females = 34, males = 43) were positively screened for CH, confirmed with subsequent serum assays. Median birth weight was 3150 g (range 900-4170) and median length was 50 cm (40-53) with a median gestational age of 39 weeks (24-42). Cesarean delivery occurred in 33% of patients. Twelve children were born preterm (15.6%). Two children were born from a twin pregnancy and 1 child was born from pregnancy after intracytoplasmic sperm injection. Associated malformations were present in 6 children (7.8%), all were minor; most were heart defects (patent foramen ovale with mild shunt in 4 cases, 1 of them associated with mild pulmonary valve stenosis); inguinal hernia was found in 1 case and congenital stapes fixation in another case. Maternal thyroid disease was reported in 14 patients (18%): 2 mothers had Graves disease, and 1 of them was treated with methimazole during the third trimester of pregnancy; 1 had multinodular goiter for which she received LT4 treatment; 8 had autoimmune thyroiditis treated with LT4; 1 had untreated isolated hyperthyrotropinemia. Only 1 patient had a known family history of CH (maternal cousin with CH and GIS).

In 1 case both Tg-Ab and TPO-Ab were detected. Neonatal iodine overload was excluded in all but 3 cases with iodine exposure during prenatal or perinatal period. A neck US showed in all cases the presence of a thyroid localized in its anatomical position with normal size.

Children With Transient CH Identified in the First Year of Life

All patients were treated with LT4 within the first month of life, at an initial dosage of 8 to 12 µg/kg per day. Two patients were excluded because they showed a diminishing need for LT4 during the initial control interval and until the cessation of therapy within the first year, remaining euthyroid after withdrawal (follow-up after suspension respectively of 1.91 and 3.35 years). In 1 case, iodine exposure during the neonatal period might explain a transient form of CH.

Clinical reevaluation

After 3 years of age, 55 patients were clinical reassessed after LT4 discontinuation for 4 weeks, while 22 had not yet been reevaluated at the end of our follow-up and were still receiving treatment with LT4. Five patients did not perform scintigraphy with perchlorate discharge test at clinical reevaluation because they were unable to maintain position or the parents did not give consent.

Fifteen patients (27.3%) were frankly hypothyroid at clinical reassessment and resumed LT4 therapy. At the end of follow-up, 13 of them had maintained LT4 therapy; 1 patient did not continue clinical care at our center. In 1 case, the LT4 dose was gradually reduced until withdrawal at the age of 9, but treatment was reintroduced after 2 years because of increasing levels of TSH (7.9 mIU/L). One patient developed papillary thyroid carcinoma at the age of 23 years, requiring therapy with TSH-suppressing doses of LT4, so the follow-up period was considered until the carcinoma diagnosis.

Twenty-two children (40%) presented hyperthyrotropinemia at the time of clinical reassessment. In 1 case, Tg-Ab was positive with low titer, while TPO-Ab was negative. Seventeen of these patients resumed LT4 therapy while 5 patients discontinued it. At last evaluation at our center, 8 patients of this group were being treated with LT4 (resumed at clinical reassessment or later) while 12 showed a normal thyroid function without treatment. Two patients had a too short a follow-up period and were excluded from follow-up analysis. The child who had showed slight Tg-Ab positivity at clinical reassessment developed autoimmune thyroiditis during the follow-up period.

Eighteen patients (32.7%) had normal thyroid function at clinical reassessment. Fifteen patients from this group did not resume therapy at clinical reassessment, while in 3 cases therapy was reintroduced. In 2 cases, therapy was resumed during follow-up, respectively at 2 and 9 years after clinical reassessment, because of elevated TSH values (respectively 17 mIU/L and 9.2 mIU/L). Three patients were lost to follow-up. The clinical course and necessity of therapy in time of patients who performed reevaluation are depicted in Fig. 2.

Perchlorate discharge test showed a TIOD in 2 cases and a PIOD in 7 cases (4% and 14% of the total patients who performed the test, respectively). Features of these patients at clinical reassessment and their clinical course are depicted in Table 1.

Comparison Between “Transient” and “Permanent” Hypothyroidism During Follow-up

Subsequent periodic controls of thyroid function were executed, and, when possible (ie, patients who were adequately compensated with low LT4 dose or who had previously showed a negative perchlorate discharge test/PIOD), a new attempt to stop LT4 was performed. In order to assess any differences between patients in therapy at the last control and patients without therapy, we considered only patients with an adequate period of follow-up after suspension of therapy (ie, 6 months). Using these criteria, follow-up data were available in 49 patients (Table 2). Twenty-four were untreated, with a median duration of period after suspension of 6.04 years (0.5-12) and a median age at last suspension trial of 6.6 years (2-17.7); 25 patients were taking LT4 therapy at the end of follow-up. In the group of “permanent” hypothyroidism, we also included 2 patients who had not performed clinical reassessment because they had required high and increasing LT4 doses.

We observed no statistical differences between the 2 groups in first serum TSH levels. Among other parameters at birth, we found no statistical difference between sex, gestational age at birth, or birth weight. TSH at clinical reassessment showed a significant difference between the 2 groups (P < .001), being higher in the group of patients who showed a permanent form of hypothyroidism (10.6 vs 3.9 mIU/L). Median LT4 dose before clinical reassessment, expressed as µg/kg per day, was not significantly different between the 2 groups (1.75 in the group of permanent CH and 1.35 in transient CH, with P .082). LT4 dose before the last trial off therapy showed a significant difference between the 2 groups (P .016) and was higher in the group of patients with permanent hypothyroidism (1.58 vs 1.2). It must be specified that the age of patients at last trial off therapy was variable, ranging from 3 to 16 years, with a median age of 6 years. Comparing patients with neonatal blood-TSH at recall less than 20 mIU/L and patients with neonatal blood-TSH at recall ≥20 mIU/L, we found no statistically significant differences in terms of clinical outcome (transient vs permanent CH at the end of follow-up) (Fig. 3).

Discussion

In this retrospective study, we evaluated the clinical and biochemical outcome of a cohort of patients with CH and GIS, followed at a single center after diagnosis through periodic controls on average every 6 months. At clinical reassessment, patients were divided into 3 groups (euthyroidism, hypothyroidism, and hyperthyrotropinemia) according to thyroid function after LT4 suspension. In our cohort, over a third of patients were euthyroid at clinical reassessment. When we collected data at subsequent follow-up, for simplicity, we divided our cohort into 2 groups: patients with transient CH who had suspended therapy and were euthyroid at least for 6 months after suspension and patients with permanent CH who were still taking LT4 at the end of follow-up. Half of our patients showed a normalization of thyroid function without LT4, suspended at a variable age, with a median period of euthyroidism after suspension of 6 years (0.5-12). This result has an important prognostic value because it confirms that not all patients with CH with thyroid in situ require lifelong treatment. These data confirm what has been reported in previous studies on this specific form of CH. However, the follow-up period necessary to classify a form of CH as permanent or transient is still unclear and very variable among different studies. According to the Michigan Newborn Screening program, CH is defined as transient if the patient was trialed off therapy after 3 years of age and had a normal thyroid function test a month or more after discontinuation of LT4. Using these criteria, Korzeniewski et al (8) reported that 25% of children had transient hypothyroidism; however, 83% of them had stopped treatment without medical supervision and 45% of eligible cases were lost to follow-up. Rabbiosi et al (9) showed that 38% of patients with CH and a GIS were euthyroid after LT4 suspension for a period of 4 weeks; in this study the authors included children with both hypoplastic and hyperplastic glands. Castanet et al (10) reported that 37% of children with CH and GIS thyroid had normal thyroid function at clinical reassessment at a mean age of 54 months; the authors excluded premature children and included 5 neonates that were exposed to iodine overload. In Gaudino et al (11), transient CH was demonstrated in 31.38% of cases during a relatively short follow-up of 0.7 to 4.5 years, with a high proportion of premature births and iodine overload. Kemper et al showed similar results (12), demonstrating that 38% of CH had transient hypothyroidism at 3 years follow-up. In our cohort we chose to define permanent and transient CH considering also the data of the period after clinical reevaluation; in fact, in some cases that showed moderately elevated TSH at clinical reassessment, thyroid function completely normalized even several years later. On the other hand, some children, after a variably long period of mild hyperthyrotropinemia showed a worsening of thyroid function requiring treatment, and we classified them as permanent CH. This result adds important information to our knowledge about the clinical history of CH, showing that with a longer follow-up the ratio of permanent and transient CH could be different and in particular the proportion of patients with transient CH is higher than at clinical reassessment.

We compared clinical and biochemical parameters between the 2 groups of permanent and transient CH, but we did not find any differences in birth characteristics that could predict different outcomes in these patients. In preterm children, TSH can be slightly elevated immediately after birth, but it frequently normalizes during the first weeks of life. In our cohort, there was a prevalence of 15.6% of preterm births. We excluded from the comparisons preterm children with CH not confirmed at controls within the first year of life and included preterm children with persistent TSH >10 mIU/L over time; using these criteria we did not find any differences in the outcome of patients regarding gestational age.

Patients with screening blood TSH values ≥20 mIU/L had no significant differences in clinical outcomes from patients with screening blood TSH values <20 mIU/L. It must be noted that the second group would have not been identified through screening before 1982 in our region, missing the chance to identify those apparently mild forms of CH with subsequent worsening, which require prolonged treatment. This observation supports the more recent screening strategies that use a lower screening TSH cut-off, leading to the diagnosis of milder forms of CH, persistent over time (13).

The family history of thyroid disease, which has been associated with permanent CH (9), was not different between groups.

Thyroid US, including only patients with thyroid of normal size and excluding frankly hypoplastic glands, did not help to predict the course of disease. In cases of iodide organification defect, associated with permanent hypothyroidism, the thyroid gland was slightly hyperplastic when first evaluated, but thyroid volume normalized during subsequent years due to correct replacement therapy.

Regarding the value of perchlorate discharge test at reevaluation, in our cohort it helped the detection of organification defect (both total and partial) in 18% of patients undergoing the test. As also demonstrated by other authors (9), these cases showed a wide clinical variability, probably due to the different genetic backgrounds. In fact, if on the one hand all patients with TIOD had a permanent CH, among patients with PIOD, 2 of them showed a permanent CH while 3 showed a transient CH. We think that it is difficult to select a candidate category for this test. We suggest that the perchlorate test should be performed in specialized centers together with serum Tg measurement with the main aim to target the genetic analysis.

LT4 dose at time of clinical reassessment was higher in patients who had hypothyroidism after therapy withdrawal compared to patients with hyperthyrotropinemia and those with euthyroidism; however, the influence of these differences in subsequent outcomes (permanent vs transient CH during follow-up) was attenuated and not significant. The cut-off of LT4 daily requirement predictive of permanent CH varied among the studies between above 2.25 µg/kg during first 2 years and 4.1 µg/kg (14-16). We found a significant difference between the 2 groups of transient and permanent CH considering the LT4 dose before the last trial off medication whose dose was significantly higher in permanent CH. However, we must consider that the age at the last trial of suspension of therapy was variable in our cohort, and the LT4 dose is age dependent during growth. Despite this limit, these results encourage a trial of suspension of LT4 even in children with “apparent” permanent CH.

Despite the presence of guidelines (17), there is still no agreement in the literature about the necessity of hormone replacement therapy in hyperthyrotropinemia during the pediatric age range; due to the lack of randomized controlled trials specifically addressing this question, the clinical management of these forms varies significantly among different centers. During the neonatal period, concerns exist about neurocognitive outcome; on the other hand, during childhood and adolescence, current data do not suggest a clear effect of hyperthyrotropinemia on growth or neurodevelopment (18-21). We think that in the absence of strong evidence, considering that substitutive therapy with LT4 is safe and well tolerated, treatment of mild cases should be undertaken, especially in crucial stages of growth, in the presence of goiter or an ascertained organification defect or symptoms suggestive of hypothyroidism.

Our data show that monitoring of thyroid function through the years is important in children with CH and thyroid in situ. In well-known cases of iodine overload or maternal TRAb positivity, a transient form of CH is probably present, and the therapy should be suspended during the first months of life; in the other cases, a trial off therapy at the age of 3 years is necessary to evaluate the real need for substitution. Even after suspension of therapy, patients need to be followed closely and thyroid function monitored. We could not exclude that relapses of moderate hyperthyrotropinemia might happen later in life, especially at puberty or during pregnancy. Long-term studies are required to determine the outcome of this specific form of CH and to identify evidence-based criteria to assess hypothyroidism permanence.

Funding

This research did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.

Disclosures

The authors have nothing to disclose.

Data Availability

Some or all data generated or analyzed during this study are included in this published article or in the data repositories listed in References.

Ethics Approval

Ethical approval was waived by the local Ethics Committee of University of Pisa in view of the retrospective nature of the study and all the procedures being performed were part of the routine care.

References

Abbreviations

 
• 123-I

  123-iodine

 
• CH

  congenital hypothyroidism

 
• FT3

  free T3

 
• FT4

  free T4

 
• GIS

  gland in situ

 
• LT4

  levothyroxine

 
• PIOD

  partial iodine organification defect

 
• Tg

  thyroglobulin

 
• Tg-Ab

  antithyroglobulin thyroid antibodies

 
• TIOD

  total iodine organification defect

 
• TPO-Ab

  antithyroperoxidase thyroid autoantibodies

 
• TRAb

  anti-thyroid-stimulating hormone receptor antibodies

 
• US

  ultrasonography

Author notes