Introduction
Congenital hypothyroidism (CH) is the most common congenital endocrine disorder in neonates. The thyroid hormone has a fundamental role in the growth and development of the brain. Hypothyroidism remains asymptomatic in many neonates. Therefore, a thyroid-screening test is carried out for all newborns during the first three to five days after birth, by measuring the thyroid-stimulating hormone (TSH) from the heel prick blood spot, to prevent the consequences of late diagnosis and treatment of CH. In the case of abnormal results of the primary screening test, a confirmatory test is requested (1, 2). Newborns who have an abnormal screening test and elevated TSH and normal T4 in the confirmatory test are considered hyperthyrotropinemia (HT).
HT has two types, namely transient and permanent. Transient HT is often resolved after a few weeks to a few months, but neonates with permanent HT will maintain the abnormal TSH level (3). A specific incidence rate for transient HT has not yet been determined, and various studies suggested an incidence range of 1 in every 8000 live births to 1 in every 17600 live births per year (4).
Different studies have suggested several causes for transient HT, such as prenatal iodine deficiency, prenatal iodine excess, maternal antithyroid medication, maternal hypothyroidism, and prematurity (5). Notably, 10% of neonatal hypothyroidism cases in the world are not detected despite the implementation of the thyroid screening protocol (6). Accordingly, a study recommended that hypothyroidism should be considered a potential cause in neonates with symptoms of developmental delay and prolonged jaundice (7).
Although a specific definition for transient HT has been proposed, there is no specific protocol in the confirmatory tests to discriminate between transient and permanent forms of HT in neonates (4). The management approach for these patients has been controversial. Some clinicians withhold treatment. Others treat the HT patients with thyroid hormone replacement for the first few years of life, and then gradually reduce the dose of levothyroxine (3, 8-10).
Although the latter approach can reduce the risk of side effects of neonatal hypothyroidism, it can increase the risk of overtreatment and complications for transient HT neonates (11). This study aimed to investigate the prevalence of transient HT, its associated factors, and the recovery rate without medical treatment to prevent the long-term potential side effects of this disorder, especially on brain development.

Methods
Study design
The present observational prospective study was conducted on neonates with HT referred to the pediatric clinic of Boali Hospital, affiliated to Islamic Azad University Tehran Faculty of Medicine, Tehran, Iran during September 2017 and January 2019. The recovery rate was investigated during a three-month follow-up, withholding medical treatment. The protocol of this study was approved by the ethics committee of Islamic Azad University Tehran Faculty of Medicine (Ethics code: IR.IAU.TMU.REC.1396.182). Written informed consent was obtained from the parents of all neonates before their participation in the study. All information was used privately and without mentioning the patients’ names.
Study population
During this study, 2,556 neonates were referred to the vaccination section or pediatric clinic of Boali Hospital for the thyroid screening test. To measure the qualitative TSH level in these neonates, the heel prick blood spot was conducted in the first three to five days after birth. Those who had an abnormal screening test were referred to the laboratory department of Boali Hospital to measure the serum level of TSH and free T4 (FT4). All neonates with an abnormal screening thyroid test who had a TSH level between 6 and 60 mIU/L and normal FT4 (1.3 to 2.8 ng/dL) entered the study with the diagnosis of hyperthyrotropinemic hypothyroidism. The secondary assessment was done by measuring TSH at three months of age. During these three months, the neonates underwent a routine physical examination, every two weeks. The exclusion criteria were as follows: The primary TSH level of lower than 6 mIU/L or higher than 60 mIU/L, or the primary FT4 level lower than 1.3 ng/dL. Known cases of metabolic diseases and those whose parents did not give consent were also excluded from the study. Sampling was done with the convenience method, and all neonates who fulfilled our criteria entered the study.

Data gathering
In this study, the neonate’s data including gender and maturity status (term, preterm), and history of jaundice (physiological, pathological, or none), congenital anomaly, and collagen vascular diseases were collected. Maternal data, including age and history of hypothyroidism, were also recorded. Weight measurement was done by a pediatrician using the same scaling instrument. The TSH and FT4 levels were assessed using the automated immune chemiluminescence assay kits (Abbott, IL, USA) in the laboratory of Boali Hospital. To investigate the recovery rate, the TSH level was measured again after three months of follow-up, and recovery was defined as a TSH level of 0.5 to 5.5 mIU/L.

Statistical analysis:
All data were analyzed with SPSS Version 22.0 (SPSS Inc., Chicago, IL, USA). Quantitative variables are described using mean ± standard deviation, and qualitative variables are presented using frequency and percentages. Chi-square test, one-way ANOVA, and independent sample T-test were applied to evaluate the difference between groups in terms of the target variables. A P-value < 0.05 was considered statistically significant.

Results
This study included 80 neonates with HT (65% girls). None of the neonates had a congenital anomThis study included 80 neonates with HT (65% girls). None of the neonates had a congenital anomaly or collagen vascular disease. The descriptive information of the participants is shown in Table 1. Overall, 62 neonates (77.5%) recovered without any treatment during three months of follow-up.
The mean level of TSH in all neonates was 12.2 ± 6.6 mIU/L, and the TSH level was significantly higher in boys and preterm newborns (p < 0.05). The recovery rate was also significantly higher in term newborns (p < 0.01). Neonates with a history of maternal hypothyroidism had a lower level of TSH, but the difference (versus others) was not statistically significant (p > 0.05). Further, 60% of neonates had a history of physiological jaundice, and the level of TSH was higher among them, but the relation was not statistically significant (p > 0.05). The recovery rate did not have any significant relation with gender, maternal hypothyroidism, and history of jaundice (p>0.05).
The mean weight of the participants was 2626.8 ± 584.4 grams. The mean weight of the neonates who recovered (2767.7 ± 440.2 g) was significantly higher than that of the neonates who did not recover (2141.7 ± 755.2 g) (p < 0.01). There was also a significant relationship between the TSH level and birth weight (p < 0.01). Moreover, the mean age of mothers in participants was 26.9 ± 3.7 years, with a minimum and maximum of 12 and 35 years, and did not have any significant relation with the level of TSH and with the recovery rate (p > 0.05).
The mean level of TSH in neonates who recovered within three-month was 9.4 ± 3 mIU/L, and in neonates who did not recover was 22 ± 6.5 mIU/L. The relation between the recovery and TSH level

Discussion
The present study showed that the prevalence of febrile seizures was associated with gender, living place, temperature, family history of seizure, and the serum level of zinc. In this regard, the frequency of zinc deficiency was higher in patients with febrile seizures compared to febrile patients without seizure, before and after adjusting for gender.
Zinc plays a vital role in the neuronal terminals of the hippocampus and amygdala by producing pyridoxal phosphate and affecting glutamatergic, gamma-aminobutyric acidergic (GABAergic), and glycinergic synapses (13).
Glutamic acid decarboxylase (GAD) acts as a major inhibitory neurotransmitter in the synthesis of gamma-aminobutyric acid (GABA) (14). A study by Ganesh R. and Janakiraman L. on 38 children with febrile convulsion and 38 children as a control group, aged between 3 months and 5 years, indicated that a serum zinc deficiency was significantly more prevalent in their case group than in the control group (15). Another study has reported that there is a correlation between disruption in Zn2+ homeostasis and fever seizure (16).
In studies by Papierkowski A., Mollah M.A., and Gündüz Z. et al., the mean serum zinc level in the febrile convulsion group was significantly lower than in the control group, which indicates the role of zinc in febrile seizure. Comparing the groups in terms of age and gender, no significant difference was found, similar to our study (17-19). Abdel Hameed Z.A. et al. (20), in a study on 100 infants in Egypt, observed that temperature had no significant difference between the case and control groups. But Berg A.T. (21), Ahmed B.W. (22), and our study showed the importance of temperature in febrile seizure. The geographic area can be the cause of this difference. Duangpetsang J. in a study from 2014 to 2017 reported that a high fever with electrolyte disturbance hyponatremia has an important role in FS (23). Sharifi R. et al., in a study in 2007-2014, showed the importance of family history in febrile seizure (24), which is similar to our results.

Conclusion
The findings of this study show that zinc deficiency is significantly associated with the occurrence of febrile seizures. Zinc supplementation in children can therefore be helpful for the prevention and treatment of FS.

Conflict of interest
The authors declare no conflicts of interest.