In this study, the incidence of CH in Fars province from 2013 till 2016 was 1:313.66. In Iran, the incidence rate was 1:914 live births in Tehran [10], 1:357 in Isfahan[11], 1:1608 in Yazd [17] 1:1250 in Hamadan[19] 1:446 in Ahvaz [18] 1:453 in Mazandaran[16], and 1:307 in Markazi [24]. Our result is in line with Isfahan and Markazi provinces. The prevalence of CH for the rest of the world is 1:2118 in Alabama (a USA state) [25], 1:1684 in Oregon (a USA state) [26], 1:469 in Turkey [27], and 1:834 in Saudi Arabia [28]. The obtained results in Iran are far higher than the incidence rates in other parts of the world, which can be due to the gene pool, ethnicity, autoimmune factors, parental consanguinity, and iodine deficiency.[14] Due to cultural differences, parental consanguinity is more accepted in Iran. In addition, several studies showed that dyshormonogenesis, which is mainly caused by autosomal recessive inheritance, has higher prevalence rate amongst the CH patients in Iran.[11, 18, 29] Furthermore, pregnant women in Iran have moderate iodine deficiency.[30]
A 2007 study on CH in Fars province revealed an incidence rate of 1:1465.[29] Aside from the higher incidence rate in our study, we found a statistically significant increase in the prevalence of CH, even within the period, from 2013 to 2016. Shaghaghian et al. argue that the increase in the rate of CH prevalence in Fars can be attributed to better coordination between responsible organizations for the screening program, increased access to screening centers, increased doctor to patient ratio in the province, better training programs for pregnant women and the implementation of screening regulations for hospitalized neonates.[23]
Permanent vs. transient CH rates differ widely in the world and even within Iran. In Yazd, 45.5% of the confirmed CH cases were permanent [17], in Hamadan 64% [19], in Ahvaz 46% [18], in Isfahan 59.8% [11] and in Markazi 51.9% [24]; while the rest of CH cases were transient. The last study in Fars province (2007) reported 53.6% permanent CH cases.[29] Worldwide, 66.44% of CH cases in the state of Alabama (USA) [25], 70.6% in the state of Oregon (USA)[26], 75% in the state of Michigan (USA)[22], 62% in France [31], and finally, 46.6% of CH cases in China were permanent; while the rest were transient [32]. In our study, 66.04% had confirmed permanent CH, while only 33.96% had transient CH. Our results are more in line with studies from Hamadan and Isfahan in Iran, and Alabama (USA) and France. The difference in permanent CH rate worldwide can be attributed to iodine status and most importantly, differences in CH screening methods. For example, states in the USA use two round primary T4-reflex TSH method instead of the primary TSH method, which is used in Iran. Differences between provinces inside Iran can be mainly attributed to iodine levels. Despite successful universal salt iodization program since the 1990s and improvement in Iranian iodine status, there still remains a moderate iodine deficiency amongst the Iranian pregnant women.[30] Transient CH rate in Fars province has also decreased from 46.4% in 2007 [29] to 33.96% in our study, which can be explained by the improvement in urine iodine status in the Fars population.[33]
Effects of parental consanguinity and familial history of thyroid disorders on transient and permanent CH is controversial. While studies by Rabbiosi et al., Dorreh et al., and Saba et al., found first-degree familial thyroid disorders more prevalent in permanent CH patients [24, 34–35], Zhou et al., found them to be more prevalent in transient CH.[32] In our study, we did not find any statistical differences in the prevalence of familial thyroid disorders between the two groups, although it should be noted that both were higher than the general population. Familial thyroid disorders can account for transient CH based on maternal autoimmune diseases and trans-placental passage of autoimmune antibodies, while goiter and nodular disorders in familial history can account for permanent CH.[34] In our study, similar to Saba et al, parental consanguinity was more frequent in permanent CH[35], while in studies by Dorreh et al ., and Razavi et al., there were no statistical difference between the two groups.[19, 24] We also found caesarian section delivery to be more prevalent amongst permanent CH. The same result was obtained by Dorreh et al., in Markazi province.[24] However, Rabbiosi et al., found no significant difference with respect to the method of delivery between the two groups.[34]
In our study, the trend in reduction of TSH concentration in the span of ten outpatient visits was not statistically different between the transient vs. permanent CH; nevertheless, we found TSH serum concentration level to be generally higher throughout all visits in permanent CH cases. A closer look at TSH serum concentration level of the ten outpatient visits (Table 2) revealed that standard deviation for mean TSH level was much higher in permanent CH cases. In fact, 54.9% of permanent CH patients had experienced at least one uncontrolled TSH level (TSH ≥ 5) at some point during the three years of follow up vs. the 36% in transient CH cases. Similar results were obtained by Razavi et al., in Hamadan where TSH serum concentration level during the three years of follow up were significantly higher in the permanent CH subjects. (6.10 ± 6.18 vs. 2.92 ± 3.50 mIU/L in permanent and transient CH respectively, P < 0.001).[19]
Association between TSH serum concentration level at screening tests and the venous sampling with transient vs. permanent CH were inconsistent. Many studies found TSH serum concentration level at screening and the first venous sampling to be significantly higher in permanent CH [17, 19, 35, 24, 32]; while only a few studies found no significant difference between the two groups.[34, 36] We found TSH serum concentration level > 43.35 mIU/L in the venous sampling (initial TSH) with the sensitivity of 31.66% and specificity of 90.32% to be the best cutoff point among the three TSH measurements in this study. Zdraveska et al., found an initial TSH level of < 30.5 mIU/L to be the predictor for transient CH with sensitivity of 92% and specificity of 75.6%.[37]
Several cutoffs have been proposed for better predicting permanent or transient CH, using levothyroxine dose at the third year including cutoffs of > 2.86 µg/kg by Park ES et al., (Sensitivity: 88.9%, Specificity: 71%) [38] and > 3.96 µg/kg by Itonaga et al., (Specificity: 100%) [39] for permanent CH and cutoffs of < 1.3 µg/kg by Higuchi et al., (Sensitivity: 80%, Specificity: 84%) [36] and < 2.76 µg/kg by Park IS et al., (Sensitivity: 87.3%, Specificity: 67.6%) [40] for transient CH. In our study, a cutoff point of > 2.25 µg/kg with a sensitivity of 76.11% and specificity of 58.52% at the third year of follow up was the predictor for permanent CH.
The main limitation of this study is that despite recalls, we did not have access to follow-up data for more than 200 CH cases. As the screening program will improve overtime, further study on patient follow up with more thorough data is warranted.