Cases Referred from the Turkish National Screening Program: Frequency of Congenital Hypothyroidism and Etiological Distribution

Objective: The aim of this study was to evaluate cases referred from the congenital hypothyroidism (CH) newborn screening program. Methods: Infants referred to Pediatric Endocrinology between 30.09.2015 - 01.04.2018 because of suspected CH identified by National Neonatal Screening Program were prospectively evaluated. Results: Of the 109 newborns referred to our clinic, 60 (55%) were diagnosed with elevated neonatal thyroid stimulating hormone (TSH). The diagnosis of elevated neonatal TSH was made in 52 (47.7%) and eight (7.3%) infants at initial evaluation and after follow up, respectively of all referrals with 86.7% (52/60) diagnosed at initial visit. The median first and second heel prick times were 1.8 (0-7) and 8.72 (4-30) days. The median age at starting treatment of the infants diagnosed as a result of initial evaluation was 22.13 (7-53) days. Clinical findings associated with CH were present in 19 (36%) of patients. Etiology in patients diagnosed with elevated neonatal TSH on admission was: agenesis in one (2.08%); ectopia in one (2.08%); hypoplasia in 14 (29.16%); normal gland in situ 16 (33.3%); and hyperplasia in 16 (33.3%). The median time to normalization of TSH and free thyroxine concentrations after treatment initiation was 11.02 (4-30) and 9.03 (3-30) days, respectively. Conclusion: The rate of diagnosis in the first month was found to be 87%. The etiological incidence of both dysgenesis and dyshormonogenesis was equal at 33.3%. The majority of cases with normal thyroid gland will be diagnosed with transient hypothyroidism but some of them may be diagnosed with thyroid dyshormonogenesis so the rate of dyshormonogenesis will increase later after final diagnosis.


Introduction
Congenital hypothyroidism (CH) is the most commonly encountered endocrinological disorder of childhood worldwide. Its incidence was reported as approximately 1/4000 in the 1970s when the screening program was first used. In the last 20 years, an increase in the prevalence of CH has been reported worldwide, possibly as a result of a gradual reduction in referral screening cut-off values (1).
In a study conducted in Turkey in 2003, the incidence of permanent CH was reported as 1/2512 (2). In an evaluation of data between the years 2008 and 2010, the incidence of CH was reported as 1/888 in 2008, 1/592 in 2009, and 1/650 in 2010 (3).
The absence of disease-specific clinical manifestation in CH at birth and the fact that complications can be prevented with early initiation of treatment within the first few weeks after delivery, plus the fact that treatment is effective and inexpensive, led to the establishment of national screening programs. A CH screening program was first implemented in 1974 in Quebec, Canada, and in Pittsburgh and Pennsylvania in the US (4,5). In Turkey, although local screening was conducted before, nationwide CH screening in conjunction with phenylketonuria screening was introduced for the first time on 25 December 2006 in Turkey (6,7).
In Turkey, a thyroid stimulating hormone (TSH) bloodspot threshold value of 20 mIU/L was used in the first years of screening and subsequently this value was decreased to 15 mIU/L in January 2009, upon detection of cases of missed diagnosis. However, in 2016, the screening cutoff TSH threshold value was again increased to 20 mIU/L (3,8). According to the guidelines of the Turkish CH national screening program, an infant is considered to have passed the screening test when the blood spot TSH concentration is <5.5 mIU/L. Values between 5.5 mU/L and 20 mIU/L, are reported for a second evaluation. All cases with a capillary TSH concentration above 5.5 mIU/L in repeat blood samples or above 20 mIU/L in the first sample are referred to the appropriate center for evaluation by venous thyroid function testing, including thyroxine (T4) and TSH.
In this present study, we aimed to investigate the rate of diagnosis and the time of diagnosis in cases with elevated TSH referred from the screening program; to evaluate the clinical and laboratory findings in these cases; and also to determine the etiological distribution among these patients.

Materials and Methods
All infants referred to the Outpatient Clinic of Dr. Sami Ulus Obstetrics and Gynecology, Child Health and Disease Department of Ankara University, between the dates of September 30 th 2015 and April 1 st , 2018, because of suspected CH following newborn screening were prospectively evaluated. The study was approved by the Local Ethics Committee (no: 45/2015). Prior to inclusion in the study, informed consent was obtained from all parents.
In our study, serum T 4 and TSH levels were measured on the day of referral. The diagnosis and treatment plan was based on the Lawson Wilkins Pediatric Endocrine Society Guideliness for Congenital Hypothyroidism (9). Date of birth, date of admission, postnatal age, gender, gestational age in weeks, type of delivery and birth weight were recorded in all cases included in the study. Number, dates and results of heel prick tests were recorded. All cases were investigated for family and maternal history of thyroid disease, including a family history of CH. Consanguinity between parents, presence of iodine exposure in the mother and the baby and the iodine content of the salt used during pregnancy were recorded.
Body weight, height and head circumference were measured on admission. All patients were examined for clinical signs and symptoms associated with CH which included: lethargy, inactivity, hypotonia, difficulty in feeding, excessive sleepiness, constipation, abdominal distention, umbilical hernia, prolonged jaundice, galactorrhea, weak cry, hypothermia, cutis marmoratus, nasal congestion and dry and coarse skin. Physical examination was made for both additional congenital abnormalities and goiter. Serum thyroglobulin, TSH receptor binding antibody (TRB-Ab) and spot urinary iodine levels were measured in patients diagnosed with CH. The localization, volume and parenchymal echogenicity of the thyroid gland were evaluated by thyroid ultrasonography (USG). Thyroid gland localization and activity were evaluated by thyroid scintigraphy.
Serum TSH, free T 4 (fT 4 ), and free tri-iodothyronine (fT 3 ) concentrations were measured by chemiluminometric method using an Advia Centaur XP (Siemens Healthcare limited-Oakville) device. Thyroglobulin concentration was measured by immunoassay method on the Siemens Immulite 2000 analyzer device with the immulite 2000-thyroglobulin kit (Siemens Healthineers-United States). TRB-Ab level was measured on the Berthold 1B2111 device (Berthold, USA) by radioimmunassay with a Beckmann Coulter RRA Anti-R TSH kit (Beckman Coulter Company-Czech Republic). Spot urinary iodine level was measured on Agilent 7500 ICP-MS device (Agilent Technologies, USA) by using ICP-MS analysis technique.
Thyroid USG was performed by a radiologist using a 7.5-MHz linear probe with a Toshiba Aplio 500 US device (Toshiba Medical Systems Co. Ltd, Otawara, Japan). Thyroid volume was calculated for each lobe using the following formula: (D1xD2xD3/1000) x 0.523 where D1 is the longest longitudinal, D2 is the anteroposterior and D3 is the largest transverse diameters, in cm, for each lobe and the total volume was calculated as the sum of two lobes in mL. Thyroid volumes for the neonatal period with a value of less than 0.64 mL (10 th percentile) in patients were considered to constitute hypoplasia, those with a volume greater than the 95 th percentile value of 1.15 mL were considered to constitute hyperplasia and total volumes between 0.64 mL and 1.15 mL were considered normal (10). Thyroid scintigraphy was performed in the Nuclear Medicine Department of our hospital using General Electric Millenium device (General Electric; Elgems, Tirat Carmel, Israel) and gamma camera with technetium 99 pertechnetate.

Statistical Analysis
In the presentation of descriptive statistics; the data obtained by measurement were expressed as mean±standard deviation and (minimum-maximum) and categorical data as number (percentage). Cross-table analyzes, Pearson and Fisher's Exact chi-square tests were used to compare the qualitative characteristics of the groups. Compatibility to normal distribution of the numerical measurements in the groups was examined by independent groups for those with normal distribution in numerical measurements and with Mann-Whitney U test for those without normal distribution. IBM Statistical Package for the Social Science (SPSS; IBM Inc., Chicago, IL., USA) version 22 was used for all statistical analyzes. A p value of <0.05 was considered statistically significant.

Results
Of the 109 newborns referred from the neonatal screening program a total of 60 (55%) infants were diagnosed; 52 (47.7%) at initial evaluation and eight (7.3%) at follow up. Twenty nine (48.3%) of patients with elevated neonatal TSH were female and female to male ratio was 1:1.07. Gender, gestational age, term/preterm rate, type of delivery and birth weight were similar in patients with elevated neonatal TSH compared to babies referred but who were not diagnosed with elevated neonatal TSH (non-CH group) ( Table 1). Parental consanguinity rate was significantly higher (p=0.004) at 31.7% (19/52) in patients with elevated neonatal TSH compared to 8.16% (4/49) in the non-elevated neonatal TSH group. The family history of thyroid disease, family history of CH, the rate of iodized salt consumption and iodine exposure were similar in patients with elevated neonatal TSH compared to the non-elevated neonatal TSH group (

Discussion
In our study, only 60 (55%) of 109 cases referred from the national screening program were diagnosed with elevated neonatal TSH. In a study conducted in our country, it was reported that 114 (44.5%) of 256 cases referred from national screening program were diagnosed with CH (11). In our study, female to male ratio was calculated as 1/1.07 in patients with elevated neonatal TSH. While previous studies have shown that female preponderance in female to male ratio in patients with CH was 1.8/1, 1.4/1 (12), it has been pointed out that the dominance of gender in recent years has shifted to male direction as 1:1.14, 1/1.16 (8,13). Dysgenesis is more common in female sex, whereas dyshormonogenesis is much lower in girls (13). In our study, low rate of cases diagnosed with dysgenesis and milder cases due to lowering TSH screening cut-offs may be the cause of variability in gender distribution.
On admission to hospital, the mean age of the cases was found 27.47±14.2 (7-70) days in our study and 24.54±13.46 (4-168) days in the study of Kor and Kor (8). The mean age at diagnosis was 30.2±24.8 days in all CH cases; 22.13±10.35 in patients diagnosed as a result of initial evaluation and 82.62±28.53 days in patients diagnosed at follow up in our study. Kor and Kor (8) reported in their study that 223 (96%) of all 233 patients were diagnosed as a result of initial evaluation and the mean age of diagnosis was 19.87±7.63 (4-51); 10 (4%) of patients were diagnosed at follow up and the mean age of diagnosis was 43.71±14.02 (29-65) days. The mean age of diagnosis was found as 19.7±8.30 (5-60) days in the study of Peltek Kendirci et al (11), 38.1±58 (4-342) days in the study of Kuşdal et al (14) and 23±14 days in the study of Simsek et al (15). Early diagnosis is the most important aim of the screening program. In Turkey, in a study comparing the pre-screening and post-screening diagnosis time, prior to the introduction of screening diagnosis occurred at a mean of 292 days whereas it decreased to 35.2 days after screening was introduced (16). Bongers-Schokking et al (17) reported that there was no neurodevelopmental difference in the patients with CH whose treatment was initiated with appropriate dose in the first 13 days compared to a healthy group. In our country, despite the screening program having been implemented for 12 years, diagnosis is still occuring later than in other national programmes, although there is marked variability from center to center. This is still a major improvement over the diagnosis times reported prior to the introduction of screening in Turkey. However, there is the delay between second heel prick test and first clinic visit is nearly 19 days. Further analysis of the causes of delay between referral and initial diagnostics visits would enable some remedial organisation to be put in place.  (20,21), there has been an increase in the rate of dyshormonegenesis in recent studies in accord with our findings (19,22). An Italian national study conducted by Olivieri et al (12) reported the rate of dysgenesis in CH cases was 82% between years 1987 and 1998 although it decreased significantly compared to the previous period with a rate of 58% between years 1999 and 2006 due to significant chnages in the screening protocols used in Italy and, the authors speculate, the increased survival of many more premature infants. In the second period of the study, it was stated that the rate of detection of normal and hyperplastic thyroid gland increased compared to the first period, which again may be due to the changes in the Italian screning system and the greater proportion of unwell premature infants. In addition, it was noted that the rate of dyshormonogenesis was significantly higher in patients who had parental consanguinity (23). The consanguinity rate in our study was higher in patients with elevated neonatal TSH but there was no difference in consanguinity rate according to the etiology of elevated neonatal TSH In the group with dysgenesis, while consanguinity rate was 41.2%, it was 50% in the non-dysgenesis group (25% of whom had normal thyroid gland and 25% with hyperplasia). However, we think that the exact effect of consangunity in the etiological distribution of elevated neonatal TSH may be resolved by the evaluation of patients with permanent CH at follow up.

Study Limitations
Thyroglobulin and urine iodine levels were not evaluated at diagnosis in all patients with elevated neonatal TSH. Imaging was not performed in all patients with elevated neonatal TSH.

Conclusion
The rate of elevated neonatal TSH in the first month was 87%. The mean time of initiation of treatment was 22 (7-53) days. Dysgenesis rate was 33.3% and dyshormonogenesis rate was the same at 33.3%. The majority of cases with normal thyroid gland will be diagnosed with transient hypothyroidism but some of them may be diagnosed with thyroid dyshormonogenesis.