Abstract
Objectives
To investigate the prevalence of congenital hypothyroidism (CH) among neonates in East Azerbaijan, Iran.
Methods
A cross-sectional study conducted from 2009 to 2010. Most heel blood samples were taken between three and seven days of birth. Thyroid stimulating hormone (TSH) was tested using the enzyme-linked immunosorbent assay method. Neonates with abnormal screening results (TSH>20 mIU/L, T4<6.5 mg/dL) were re-examined. Data were analysed using SPSS software.
Results
A total of 62,459 infants were screened. Of 1562 re-called subjects (re-call rate 2.5%), 94 cases with CH were detected, yielding a CH prevalence of 1:666 (female:male ratio 1:1.4).
Conclusion
The prevalence of CH in East Azerbaijan province is higher in comparison with worldwide levels. These results emphasize the importance of the neonatal screening programme.
INTRODUCTION
Congenital hypothyroidism (CH) is one of the most common causes of preventable mental retardation, but can be prevented by early diagnosis and treatment. 1 The first screening programmes were performed in North America in 1972. 2 Today, most industrialized and some developing countries conduct routine newborn screening. Programmes in developed countries record prevalences between 1:3000 and 1:4000. 3 Although newborn screening for CH has been a routine practice in developed countries for more than 40 years, developing countries have addressed this concern only in the past decade. 4 In Iran, screening programmes for CH were first carried out by Azizi et al. 5 in Tehran (Iran's capital) in 1987. The first screening for CH in East Azerbaijan was performed in 2005. The aim of this study was to investigate the prevalence of CH among neonates screened in East Azerbaijan, Iran in 2009–2010.
METHODS
The study area of East Azerbaijan province is one of the largest health regions in the northwest of Iran, with a population of about 3.6 million. The number of annual live births is about 63,000.
We conducted a cross-sectional study on 62,459 neonates in East Azerbaijan province, from 21 March 2009 to 20 March 2010. All neonates were screened according to the CH screening protocol. In accordance with the design of the screening programme, samples were collected in 38 centres in the 19 cities of the province. Heel prick blood samples were taken by trained nurses, mostly within 3–7 days of birth, and transferred to 903 Whatman filter papers. They were dried and immediately transferred to the screening laboratory of the province by express mail service. Thyroid stimulating hormone (TSH) was tested using the enzyme-linked immunosorbent assay (ELISA) method.
In accordance with the screening programme protocol, neonates were re-called based on the level of the first TSH measurements. Among neonates aged 3–7 days, TSH≥5 mIU/L; and for infants aged greater than eight days, TSH≥4 mIU/L was considered abnormal. For such cases additional confirmation tests with T4, TSH and T3 resin uptake (T3RU) were performed. In addition, if the TSH level was ≥20 mIU/L, treatment was initiated and the thyroid function laboratory studies were repeated. If the results of the second set of tests were within normal limits, the neonate was considered as a case of transient TSH elevation, and treatment was halted.
All re-called neonates were examined clinically by a paediatric endocrinologist. From the secondary measurements (between 7 and 28 days of birth), neonates were considered as hypothyroid if T4 was <6.5 μg/dL and TSH was >10 mIU/L. 6 In both premature and full-term neonates whose T4 measurements were low according to their weight, 7 complementary tests including T3RU and free T4 index were performed and treatment was started if the results were abnormal. In a few neonates whose TSH levels were slightly more than normal upper limits (5<TSH<10), a thyrotropin-releasing hormone test was also performed. 8
Neonates with confirmed hypothyroidism received levothyroxine (10–15 μg/kg/day). Variables such as gender, weight, maternal age, birth weight, birth term (full term or premature, and if premature, gestational age at birth), day of sampling, maternal history of thyroid disorder, time of beginning of treatment and parents’ consanguinity were recorded.
TSH was measured by ELISA using Iran Padtan Elm Co kits. The sensitivity of TSH test was 0.5 μIU/mL. The accuracy of TSH test was assessed using both intra- and interassay. Intra-assay coefficients of variation at TSH concentrations of 17.7, 78.4 and 162.4 mIU/L were 8.6%, 9.8% and 9.1%, respectively. The interassay coefficients of variation for different methods at TSH concentrations of 15, 26.5 and 54.9 mIU/L were 11.3%, 9%, and 8.8%, respectively.
STATISTICAL ANALYSIS
Data were analysed using SPSS 11.5 software. Frequency, mean and standard deviation for demographic data and TSH levels in neonates were estimated. Qualitative variables were compared using the Chi-square test and quantitative/qualitative variables were compared by t-test. All P values were two-sided with P < 0.05 being considered significant.
RESULTS
Of 62,459 neonates screened, 32,479 (52%) were male and 29,982 (48%) were female with a male-to-female ratio of 1.08; 92.8% of all neonates in the population were screened. In 76% of neonates, samples were taken at 3–5 days of birth, and in 89% the result of the test was determined by 13 days of life.
Distribution of serum TSH levels among screened neonates (n = 62,459)
TSH, thyroid stimulating hormone
Among neonates with CH, 55 (58. 5%) were male and 39 (41. 5%) were female (male:female ratio 1.4:1). The mean birthweight of CH neonates was 3300 ± 651 g (range: 1350–5200 g) and the mean maternal age was 26.5 ± 6.5 years (range 15–42 years). Of the CH cases, 85 (90.4%) were full term and nine (9.6%) were premature. There was a statistically significant difference between maternal age and hypothyroidism (P = 0.03). There was no statistical correlation between birthweight (P = 0.2) or gestational age (P = 0.3) and hypothyroidism. The prevalence of CH among male and female neonates was 1.7/1000 (1:588) and 1.3/1000 (1:769), respectively, but the difference was not statistically significant (P = 0.38).
In three CH neonates (3.2%), the mother also had a history of thyroid disease. There was no parental consanguinity among 62 (66%) neonates. In 22 (23.4%) cases, parents had first-cousin consanguinity and in 10 (10.6%) there was second-cousin consanguinity between parents. There was no statistical correlation between hypothyroidism and parental consanguinity (P = 0.41).
Frequency of congenital hypothyroidism by season
DISCUSSION
The prevalence of CH among neonates in East Azerbaijan has not previously been determined, but this study shows that the estimated prevalence is one in 666 live births. This is approximately 4.5–6-fold that was reported from the developed countries. 9 CH prevalence varies throughout the world, but the worldwide prevalence is one in 3000–4000 live births. 3 It has been reported as 1:2736 in Turkey, 10 1:781 in Pakistan 11 and 1:10,000 in African Americans. 12 In Iran, the reported prevalence of CH among different populations has varied. Our results are in line with those of previous studies (l in 914, l in 1465 and 1 in 370 live births in Tehran, Fars and Isfahan province, respectively 13–15 ). The studies have reported a high CH prevalence in Iran.
Varying prevalences of CH reported from different parts of the world may be due to several factors. Iodine deficiency is a known risk factor for CH, 16–18 but this problem has been eradicated in Iran. 19 In East Azerbaijan province, studies show a significant decrease in the number of school age children with goiter (caused by iodine deficiency) and have reported that urinary iodine excretion levels are within the normal range defined by the World Health Organization. 19,20 A recent study showed that in an iodine-sufficient area, median urine and milk iodine concentrations in neonates and mothers were within the acceptable range. 19,21 Frequent application of iodinated antiseptics, used during vaginal and caesarean delivery in Iran, 22 is one of the probable causes of transient neonatal hypothyroidism, especially in premature neonates. 23,24 However, currently there is no active surveillance or policy for avoidance of iodinated antiseptics in hospitals in Iran.
Some studies suggest that CH prevalence in premature compared with full-term neonates is two-fold or greater. 25 In our study, 4.37% of all neonates and nine of the 94 CH cases were premature, showing that 0.33% of premature neonates had hypothyroidism. The observed difference may be due to our small sample size.
An important finding in this study is that only maternal age was found to be related to hypothyroidism. In 14 of the 94 CH neonates maternal age was more than 35 years. In some studies, CH incidence in neonates born to older mothers has been higher than in those with younger mothers. 26 We found no significant association between CH and the other variables such as sex, weight, gestational maturity of the neonates or parental consanguinity. This is in line with the results obtained in one other study, which also showed no significant relationship between parental consanguinity and CH, 27 but many studies in Iran have shown that parental consanguinity in cases of neonatal CH was 1.5 times that among neonates without CH. 28–30
The reported CH re-call rate also varies between different populations (varying from 0.16% to 3.3% 31–33 ). Our re-call rate of 2.5% is in the higher ranges. The difference may be due to the sampling method, or a different way of performing laboratory tests, and may also reflect the levels of iodine deficiency in different regions. 28,32
Recent studies suggest that nearly all screening programmes report a female predominance, approaching 2:1 female to male ratio. 26 In our study, this ratio was 1:1.41, with male preponderance. The female:male ratio varies in different studies (e.g. 6:1 34 in Estonia and 3:1 35 in Saudi Arabia). The difference may be due to the high prevalence of consanguineous marriages in our region, or more probably to our small sample size. To our knowledge, there have been no systematic long-term follow-up studies of the sex ratio among newborns with transient versus permanent hypothyroidism in Iran. In three of the CH neonates in our study (3.2%) the mother also had a history of thyroid disease, supporting the importance of genetic factors as a possible cause of congenital hypothyroidism. 32
Seasonal variation in CH incidence has been described in some studies. 36,37 In our study the highest prevalence (2.3 in 1000 live births) was found in autumn, and the lowest prevalence in summer. The seasonal variation suggests that environmental factors (e.g. viral infections) may be important in the development of CH; however, the seasonal variation in CH incidence in our study cannot be generalized. Further multicentre studies of longer duration and larger sample sizes will be required to determine the correlation between CH and seasonal variation in Iran.
CONCLUSION
This study reveals that the prevalence of CH in East Azerbaijan province is higher than the comparable worldwide levels, and emphasizes the need for constancy and re-inforcement of the neonatal screening programme. Comprehensive and complementary studies for recognizing related risk factors should be a priority for health system research in this province.
Footnotes
ACKNOWLEDGEMENTS
The study was funded by the Deputy of Research at the Tabriz University of Medical Sciences.