Iodine-131 and Thyroid Function: Ostroumova et al. Respond

Sun’s comments about the relationship between iodine-131 (131I), hypothyroidism, and simple diffuse goiter suggest a misunderstanding of our study findings. We reported a significantly higher—rather than lower—radiation-associated risk of hypothyroidism among study participants without goiter than in the participants with goiter (Ostroumova et al. 2013). Specifically, the excess odds ratio (EOR) per Gray of 131I thyroid dose was 0.50 [95% confidence interval (CI): 0.24, 0.90] in participants without goiter and 0.04 (95% CI: –0.09, 0.32) in those with goiter. We also reported a lack of significant variation of EOR per Gray for hypothyroidism by levels of urinary iodine (p = 0.23), although in the discussion we noted that iodine concentration in spot urine samples, unlike presence of diffuse goiter, reflects current levels of iodine intake and is subject to high within-individual variability. 
 
The territories of Belarus were known to be iodine deficient before the Chernobyl accident; in the Soviet Union there was a system of iodine prophylaxis that was discontinued by the mid-1980s (Kholodova and Fedorova 1992). In 1995–1998, five of the six Belarus regions were classified as having moderate iodine deficiency, whereas the Gomel region, most heavily contaminated with 131I, was classified as having mild iodine deficiency partly due to some iodine supplementation in this area after the Chernobyl accident (Arinchin et al. 2000). High prevalence of diffuse goiter detected by ultrasound in children and adolescents in the relatively uncontaminated Brest region (27.8%) and low prevalence in the heavily contaminated Gomel region (5.6%) (Arinchin et al. 2000) support the idea that these differences are attributed to different intake of dietary iodine and not to 131I exposure. Moreover, there is little evidence of a dose–response association between thyroid exposure and simple diffuse goiter in other radiation-exposed cohorts (Ron and Brenner 2010). 
 
As we described in the “Materials and Methods” of our article (Ostroumova et al. 2013), availability of individual direct measurements of thyroid radioactivity served as a key criterion for inclusion into the study. All study participants had direct measurements of thyroid radioactivity performed within 2 months after the accident. In the methods for dosimetry, we cited the article by Drozdovitch et al. (2013), in which dose reconstruction methods were described in detail. We also noted that intake of 131I on average accounted for about 95% of the estimated thyroid dose, whereas the contribution of other short-lived radioiodines, external exposures, and internal exposure from cesium-137 and cesium-134 was minor (Bouville et al. 2007). 
 
We appreciate Sun’s interest in our study and hope our response is useful.


Iodine-131 and Thyroid Function
http: //dx.doi.org/10.1289//dx.doi.org/10. /ehp.1307737 Ostroumova et al. (2013 reported an associa tion between iodine131 ( 131 I) dose and hypothyroidism in the Belarusian cohort, a cohort of individuals exposed to 131 I from fall out of the Chernobyl accident when they were ≤ 18 years of age. Ostroumova et al. also exam ined other thyroid outcomes: hyper thyroidism, auto immune thyroiditis, serum concentrations of thyroidstimulating hormone, and auto antibodies to thyro peroxidase.
It may not be appropriate to include participants with other thyroid outcomes in the analysis because those thyroid out comes could be indirectly associated with exposure. Chernobyl is in an iodinedeficient area (Ishigaki et al. 2001), and the preva lence of goiters among children ≤ 18 years of age has been reported at > 15% in this area (Hatch et al. 2011). Is high prevalence of goiters in the area caused by normal iodine deficiency or by the 131 I? If the goiters were caused by 131 I, the relation ship between the 131 I and hypo thyroidism is still unclear, even though Ostroumova et al. (2013) stratified the data according to the presence of goiters. Hypothyroidism can also cause goiters (Wilkins et al. 1954); thus, goiter is just a serious hypothyroidism. That could be the explanation for the higher excess odds ratio in the group with goiter compared with the group without goiter shown in Table 3 of Ostroumova et al. (2013). It would have been better for Ostroumova et al. to perform a stratified analysis on the relationship between 131 I and hypothyroidism based on the normal iodine level of the individual rather than the presence of goiter. Ostroumova et al. (2013) also claimed that the thyroid radioactivity of individuals from the Belarus cohort was based on a pre vious study (Stezhko et al. 2004). However, Stezhko et al. (2004) did not provide the details of the individual radio active iodine measurement. Were the original radio active iodine measure ments generated from a for mula or modeled based on food intake or soil contamination, or was the 131 I exposure level actually measured for each individual? The answer to this question is necessary because the two methods have different credibility. In addition, the exposure described by Stezhko et al. (2004) included 131 I as well as other radio active isotopes of iodine, not 131 I alone. I would like to know whether Ostroumova et al. (2013) separated 131 I from other radio active iodine isotopes. Cesium137 should also be considered as a potential confounder in the relationship between 131 I and hypothyroidism.
The author declares no actual or potential competing financial interests. Sun's comments about the relationship between iodine131 ( 131 I), hypo thyroidism, and simple diffuse goiter suggest a mis understanding of our study findings. We reported a significantly higher-rather than lower-radiationassociated risk of hypo thyroidism among study partici pants without goiter than in the participants with goiter (Ostroumova et al. 2013). Specifically, the excess odds ratio (EOR) per Gray of 131 I thy roid dose was 0.50 [95% confidence inter val (CI): 0.24, 0.90] in participants without goiter and 0.04 (95% CI: -0.09, 0.32) in those with goiter. We also reported a lack of significant variation of EOR per Gray for hypothyroidism by levels of urinary iodine (p = 0.23), although in the discussion we noted that iodine concentration in spot urine samples, unlike presence of diffuse goiter, reflects current levels of iodine intake and is subject to high withinindividual variability. The territories of Belarus were known to be iodine deficient before the Chernobyl accident; in the Soviet Union there was a system of iodine prophylaxis that was discontinued by the mid1980s (Kholodova and Fedorova 1992). In 1995In -1998, five of the six Belarus regions were classified as having moderate iodine defi ciency, whereas the Gomel region, most heav ily contaminated with 131 I, was classified as having mild iodine deficiency partly due to some iodine supplementation in this area after the Chernobyl accident (Arinchin et al. 2000). High prevalence of diffuse goiter detected by ultra sound in children and adolescents in the relatively uncontaminated Brest region (27.8%) and low prevalence in the heavily contami nated Gomel region (5.6%) (Arinchin et al. 2000) support the idea that these differences are attributed to different intake of dietary iodine and not to 131 I exposure. Moreover, there is little evidence of a dose-response association between thyroid exposure and simple diffuse goiter in other radiationexposed cohorts (Ron and Brenner 2010).
As we described in the "Materials and Methods" of our article (Ostroumova et al. 2013), availability of individual direct measure ments of thyroid radio activity served as a key criterion for inclusion into the study. All study participants had direct measure ments of thyroid radio activity performed within 2 months after the accident. In the methods for dosimetry, we cited the article by Drozdovitch et al. (2013), in which dose reconstruction methods were described in detail. We also noted that intake of 131 I on average accounted for about 95% of the estimated thyroid dose, whereas the contribution of other shortlived radio iodines, external exposures, and internal exposure from cesium137 and cesium134 was minor (Bouville et al. 2007).
We appreciate Sun's interest in our study and hope our response is useful.