The role of season and climate in the incidence of congenital hypothyroidism in Kerman province, Southeastern Iran

Introduction

Congenital hypothyroidism (CH) is one of the most important preventable causes of mental retardation in the world and can only be prevented in case of prompt diagnosis and early treatment and on average occurs one in 3500–4500 live births globally [1], [2]. According to the studies conducted in Tehran, Iran which has a semiarid climate (based on an extended-De Martonne classification) [3], this disorder appears one in 400–900 live births, which is much higher than the world average [4]. CH is classified into transient and permanent types. The transient type improves automatically; while in the permanent type, the person has to use medication until the end of his life [5].

Some studies suggest that CH is associated with individual, genetic, and environmental factors, such as race, ethnicity, gender, birth weight, mother’s gestational age, parental consanguinity, parental education, birth season, and drug usage during pregnancy [6], [7], [8], [9], [10], [11]. In addition, immunological factors as well as iatrogenic factors are known to be the causes of CH [12]. However, there is disagreement about the impact of seasonal variation in the incidence of CH [13], [14], [15], [16].

In previous studies in Iran, Aminzadeh et al. investigated the association between seasonal changes in temperature and the prevalence of congenital hypothyroidism (CH) in Southwest Iran, and they reported that the prevalence of CH has a significant but negative correlation with temperature [17]. Hashmipour et al. in Isfahan showed that the highest incidence of CH was in summer and the lowest in winter [18]. Valizadeh et al. in Zanjan province, screened 18,000 neonates and reported that the highest hypothyroidism recall rates occurred in winter, they also showed no significant relationship between gender and incidence rate [19]. Rezaeian et al. in Hamadan province found that the season of birth significantly affected CH incidence so that the highest incidence of CH was in winter [10].

In the study conducted by Ordookhani et al. in Tehran province, a high incidence of CH was reported in winter that was 2.9 times greater than the average of the three other seasons [16]. In another study in East Azerbaijan province, the highest incidence of CH was in autumn and winter [20]. Nouri Shadkam et al. in Yazd province reported that the highest incidence of CH was in summer and spring [21]. In Markazi province, the highest incidence of CH was in autumn and the lowest incidence was spring [22].

In other parts of the world, Miyai et al. reported a higher incidence of CH in late autumn (October to December) in Osaka, Japan [15]. Nakamizo et al. also reported higher incidence of CH in winter (December to February) and lower incidence of CH in spring (April and May) and autumn (September and October) [23]. In the British Midland, higher incidences of CH were reported in fall between October and December [13]. Virtanen et al. found geographical and seasonal variations in the incidence of CH in Finland [24].

However, a number of previous studies have not seen a seasonal pattern for CH incidence; Rosenthal et al. observed no seasonal pattern in the incidence of CH in the North West of England, in Asian families compared with non-Asians [25]. No evidence of seasonal variations was found during the CH screening program in Saudi Arabia [26] either. A study in Italy did not show an association between birth date or season on CH incidence or TSH release [27]. Kaiserman et al. in Israel conducted a 10-year temporal analysis of primary CH, the average monthly incidence showed a small peak in August; however, monthly incidences of CH showed no significant periodicity [14].

The main purpose of this study was to investigate the probable role of the season and climatic factors in the incidence of CH in different regions of Kerman Province, Iran, which have different climates. The results of this study can be used in understanding the etiology and optimizing the allocation of resources in national congenital hypothyroidism screening programs.

Materials and methods

In this study, the data collected in the congenital hypothyroidism screening program from all hospitals and labor facilities as well as health centers since the beginning of the CH screening program (April 2005 to March 2011) from all regions under the cover of Kerman and Jiroft Universities of Medical Science was used. This data is checked and validated routinely. All children born in the province should pass this test and the results are sent to the Deputy of Health, CH Disease Registry of the Province.

Climatic factors, including temperature, humidity, and rainfall, were inquired from the meteorological office of Kerman Province for all cities of Kerman Province, including Kerman, Jiroft, Bam, Sirjan, Zarand, Baft, Shahre Babak, Ravar, Rabor, Orzooieh, Koohbanan, and Bardsir.

In the CH screening program, TSH is used as the initial test. Three to five days after birth, the blood sample taken from the heel of newborns is evaluated and infants with TSH values above 5 mIU/L are referred to confirm the diagnosis by venous TSH and T4 tests and physical examination. Infants who have TSH>10 mIU/L and T4<6.5 μg/dL in the second stage (between days 7–28) are considered hypothyroid and were treated with Levothyroxine. Serum TSH and T4 are measured by Kavoshyar Co. kits [28].

This data had been recorded in the entire province and was obtained from the Health Deputy at Kerman and Jiroft Universities of Medical Sciences. The authors of this article were not involved in the diagnosis of CH, and the diagnosis was done by the Kerman province’s health authorities and according to a standard national protocol.

A limitation of this study was its time frame. Although the incidence of CH was high in our study and the number of patients was high too, for more accurate results and better statistical power, longer data collection (at least 10–15 years) is suggested for future studies. Another limitation of this study was that we were not able to study the other risk factors of CH.

This project was approved by the Standing Committee on Ethics in Research of Kerman University of Medical Sciences (KMU), Kerman, Iran. Ethics Code: IR.KMU.REC.1395.338.

Results

From the beginning of April 2005 until the end of March 2011, a number of 288,437 infants were born in Kerman province, of whom 147,669 (51%) were male and 140,767 (49%) were female. Among them, 773 (0.26%) infants were diagnosed with CH in the 12 cities of Kerman province, in which 439 patients (57%) were male and 334 patients (43%) were female. The average incidence of CH in the province during this time was 2.68 per 1000 births. This number includes both transient and permanent disorders.

The highest incidence was in Rabor (9.72 per 1000 births) and the lowest incidence was in Bam (1.28 per 1000 births). Stratifying based on gender, the highest incidence in males was in Rabor (10.75 per 1000 births) and the lowest incidence in males was in Bam (1.18 per 1000 births); also the highest incidence of CH among females was in Ravar (8.79 per 1000 births) and the lowest incidence in females was in Koohbanan (0.87 per 1000 births). Table 1 shows that within both genders and total cases, there is a significant difference among CH incidence in various cities of Kerman province Iran (p<0.05).

Also in Kerman province, most males (13,472) and females (12,539) were born in January (winter) and most patients were born in October (81 cases, autumn). According to our data, the highest incidence of disease in both genders was in October and the lowest was in June (summer) (see Table 2).

In addition, Table 2 shows that there is a significant difference among the monthly distribution of CH incidence in males and in total (p<0.05), but these differences are not significant in females (p>0.05).

The CH incidence in males and females was 2.97 and 3.07 per 1000 live births, respectively. Although the occurrence of CH in females was more than males, the χ²-test showed that there was no significant relationship between gender and CH incidence (p=0.614).

The seasonal pattern of CH incidence was statistically significant in males and in total cases (p<0.05) and showed a higher incidence in autumn (September to November), but a significant difference was not seen in females (see Table 3 and Figure 1).

Figure 1:

Distribution of male, female and total infants with congenital hypothyroidism in four seasons.

The highest incidence of disease in both genders was in autumn and the lowest incidence was in winter for females and in summer (June to August) for males (Table 3).

The Pearson’s correlation between CH incidence and climatic factors in different cities of Kerman province and in different genders are presented in Table 4. Table 4 shows that there was a significant, but weak correlation between the incidence of CH in females born in Sirjan City, and humidity (ρ=0.277, p=0.02), males born in Orzooiyeh city, and humidity (ρ=0. 239, p=0. 04) and males born in Bardsir, and humidity (ρ=−0.246, p=0.03).

In addition, a significant correlation was seen between CH in males born in Shahrbabak city, and rainfall (ρ=0.367, p=0.002), and between the males born in Zarand city, and temperature (ρ=−0.274, p=0.02).

In order to study the univariate and the multivariate relationship (relation between CH incidence with each climatic factor while controlled for the other two climate factors (e.g. relation with temperature, controlled for humidity, and rainfall), we used negative binomial regression analysis. The results are presented in Table 5. Crude IRR and Adjusted IRR are the incidence rate ratio in univariate and multivariate, respectively.

The cross-correlation analysis between CH and climate variables was also done for different time lags up to 12 months, but the maximum correlation was 0.3 and none was significant.

Discussion

Congenital hypothyroidism (CH) is considered as one of the most common and preventable causes of mental retardation in the world and it can be prevented only by early and timely diagnosis and treatment [2]. In our study, 288,437 infants born in 12 cities in Kerman province of Iran were screened. The results showed a high incidence of CH in Kerman province that was equal to 2.68 per 1000 live births, whereas the incidence worldwide is 1 per 3000 to 1 per 4500 live births [29].

There are a few published studies investigating the reason for the high rates of CH in Iran. Findings from Isfahan strongly suggest the role of genetic factors as dominant etiologic factors in CH. These researchers suggest the possible involvement of other unknown genes in the pathogenesis of CH, which may follow non-Mendelian patterns of inheritance [30].

Reseachers have shown a high prevalence of parental consanguinity among cases of CH in Tehran [31], and Shadegan [32], and a higher prevalence of familial marriage, among parents of CH patients, in Isfahan [18]. Other reseachers from Iran studied the prevalence of thyroid abnormalities among first degree relatives of children with CH by ultrasonography and found that the frequency of thyroid gland developmental abnormalities was higher among parents and siblings of CH patients than the normal population [33].

Studies from Iran have also revealed factors such as twin births, birth weight, gestational age [34], birth season, jaundice at birth, age at pregnancy, maternal anemia and goiter, delivery type, father’s education, and smoking status, to have an effect on the incidence rate of CH [34], [35]. In addition, the interaction of environmental factors and other factors in some regions may play a role in the different rates of CH. Some researchers think that geographical differences and climate may play an important role in the occurrence of CH and recommended more studies to test this hypothesis [35]. Congenital hypothyroidism has different etiologies, and more studies are needed to find the etiologic factors of this disease. Congenital hypothyroidism in Iran may also be affected by environmental factors such as urbanization, as researchers have reported higher rates in urban than rural children [32].

The impact of environmental factors such as climatic conditions and seasonal changes in the incidence of CH is still unclear, and more studies are needed. So far, few studies in the world have investigated the effect of environmental and atmospheric factors on CH incidence. The present study is the first to evaluate the effects of several climatic factors such as temperature, humidity, and rainfall on the incidence of CH. We did not find any study in the literature that evaluated the effect of all climatic factors on CH incidence. However, a limited number of studies have been done on the effect of seasonal changes on the incidence of CH and have provided different results [14], [17], [18], [23], [27], [36].

Our result showed that there is a significant difference among CH incidence in various cities of Kerman province in Iran, especially Rabor and Ravar cities have the highest incidence of CH. May be unknown environmental factors that were not considered in this study lead to the increasing incidence of CH in both cities.

Our study showed a significant seasonal and a monthly difference for CH incidence in males and not females. Studies have reported increased risk of CH in girls in comparison to boys. Moreover, studies have reported considerably higher rates of both athyreosis and ectopia in girls in contrast to boys. It seems like girls are more susceptible to developing CH than boys [10]. A study from Hamadan, Iran, showed that the interaction of gender (female) and birth season (summer) increased the likelihood of CH significantly [10]. A study from Japan also found sex-specific seasonal patterns for CH incidence. They reported that from January to December, males had one, but females had two peaks [37].

There is a possibility of different risk factors for CH in different genders, but we did not find any clue in the literature.

Our result indicated that there was a significant but weak negative correlation between the incidence of CH and temperature in males born in Zarand city. A similar finding was reported from Ahwaz, Iran, in which the prevalence of CH had a significant negative correlation with temperature [1].

According to previous studies from different provinces in Iran, including Tehran [16], Hamadan [10], Zanjan [19], and Mashhad [38], CH happened significantly more in winter. However, other studies reported that the highest incidence of CH in summer and autumn seasons in Isfahan [18], in autumn and winter in East Azerbaijan [20], in summer and spring in Yazd province [21], and autumn in Markazi province [22].

Studies from other parts of the world have reported a higher incidence of CH in winter and lower incidence in spring and autumn in Japan [23], and a higher incidence of CH in fall between October and December in British Midland [13].

A few studies have reported significant seasonal differences in the incidence of CH and the reasons for these seasonal changes are not clear. Some have proposed the higher incidence of the disease in the autumn is related to viral infections [36]. However, the increased incidence rate of CH is probably related to many factors and identifying and relating these factors to seasonal changes requires more caution and research. Discrepancy may be due to differences in climate, living conditions, or various levels of iodine in different geographical areas.

Transient CH may be due to factors such as iodine deficiency or excess, maternal thyroid blocking hormone receptor antibodies, maternal use of antithyroid drugs, gene mutations, prematurity, and factors that affect the pituitary including drugs, prematurity, and maternal untreated hyperthyroidism [28], [39]. Iodine deficiency was considered the most important cause of CH in Iran before iodine supplementation, but this is not a major problem now and since 1997 Iran is considered as an iodine replete area [39].

We did not find any information about the incidence of agenesis and ectopia in different regions of Iran, although this would have been important in determining the aetiology of the variations of CH associated with climate.

We are not aware of any effect of seasonal variation on iodine resources or iodine nutrition in pregnant women in Kerman and we did not find anything in the literature either. An article from Tehran measured seasonal variations in urinary iodine concentration (UIC) among pregnant women and concluded that median UIC did not show seasonal fluctuations during pregnancy. They also mentioned that the Iranian national salt iodization program may be preventing seasonal fluctuations of UIC but still does not maintain median UIC within adequate and recommended ranges throughout pregnancy and extra iodine supplementation during pregnancy is recommended [40]. However, a study from Britain based on food consumption records has suggested that people’s iodine intake is less in summer and more in winter and thinks that these seasonal variations in intake were mainly due to seasonal differences in the iodine content of cow milk, which is related to farming practices [41].

In this study, the hypothesis of the effect of seasonal variations in the incidence of CH was investigated, although incidence of CH in various months and seasons was different, its relation to climatic factors was not clear. Based on the pattern of CH in Kerman province, it seems that we should look for factors that are not dependent on climatic factors, including socioeconomic status, parental inbreeding, genetic factors, and nutritional deficiencies of women during pregnancy. However, since there are controversies regarding the role of seasonal variations in the incidence of congenital hypothyroidism and some studies have reported a significant difference in the seasonal incidence of CH, further studies in different geographic regions, longer periods, and larger sample sizes are required. Presenting a more accurate statistics of the incidence of neonatal hypothyroidism and related factors requires implementing accurate CH screening programs and repeating the analysis with a larger population size as well as further studies that will present a more accurate picture of the risk factors of this disease among infants.

In general, given the high prevalence of congenital hypothyroidism in Iran and Kerman province in particular, and the role of this disease in developing mental retardation, neurological disorders, and physical illnesses, further research about CH to determine the main cause of disease and its risk factors are a priority in Kerman province and southeastern Iran.

Conclusions

It seems like the incidence of CH may be related to the month and season in Kerman province but has no clear relation to climatic factors. Probably, other variables which were not the subject of this study have a stronger impact on the incidence of this disease. In order to understand the etiology of congenital hypothyroidism in Kerman and the country more research is required.