Iodine excretion, variations

Geographical differences in iodine content in drinking water also exist in other countries. For instance, in Finland the iodine content in tap water from 21 cities was found to vary from 0.3 to 9.1 J,g/l (Hasanen, 1970). Variations from about 2pg/l in the south to about 12 o,g/l in the north (Felgentraeger, 1984) and from 0.7 to 5.5 xg/l (Dahl et at, 2003a) have been found in Germany and Norway respectively. Iodine intake, expressed as iodine excretion, in different geographical areas with significant variations in iodine content in tap water can be seen in Table 35.1. 

Figure 35.1 The figure shows the seasonal variations in iodine excretion in Danes with a milk intake of one or more glasses per day, n = 993 (unpublished results from the Danish Investigation of Iodine Intake and Thyroid Diseases).

The iodine content in milk shows seasonal variations, with the highest content during winter and the lowest content during summer (Hetherton Smyth, 1993 Rasmussen et al., 2002 Dahl et al., 2003c). This is reflected in a seasonal variation in iodine excretion, emphasizing the importance of milk as a source of iodine in some countries (Figure 35.1). 

Variation in urinary iodine excretion is important because it influences the study of iodine nutrition. Urinary iodine excretion exhibits large variations. The components of variation include preanalytical, analytical and biological variations. Biological variation consists of between- and within-individual variations, and can be broken down into chronobiological variation, i.e., diurnal and seasonal variations, and variations related to dilution, dietary peculiarities and supplement use. This is important for the evaluation and planning of studies of iodine nutrition, and it can be calculated that 500 urine samples depict population iodine nutrition level within 5%, while 125 urine samples are required for a value of 10%. Estimating 24h urinary iodine excretion lowers variation, and consequently the number of urine samples needed is reduced by around 20%. Similarly, it can be calculated that less than 10—15 urine samples from an individual may be misleading. 

Iodine excretion varies markedly within individuals and within groups. The variation in biological measures consists of analytical and biological variations (Harris eta/., 1970 Fraser and Harris, 1989 Andersen et a/, 2003). Biological variation is divided into two components variation within an individual and variation between individuals. The former is characterized... 

This chapter aims to provide a description of these variations in urinary iodine excretion, some components and determinants of this variation, and the importance of variation for the interpretation of measurements of iodine excretion used to describe iodine nutrition in groups and in individuals. 

What lessons can be learned from variation in iodine excretion ... 

Table 44.1 shows the analytical goals for urinary iodine excretion, and includes analytical goals for thyroid function tests for comparison, calculated from variation in a study using a routine laboratory setting (Andersen et ai, 2001). The analytical variation is less important for urinary iodine, because the biological variation is high in this set-up. 

Urinary iodine excretion displays very wide variations compared with most other biological analytes. 

Figure 44.4 gives an impression of the two components of biological variation in urinary iodine excretion within-individual variation (vertical) and between-individual variation (horizontal). 

In addition to large variations, both between individuals and within each individual, the individual variations differed widely between individuals with individual CVs, ranging from 20% to 70% (Andersen et ai, 2001). These were similar to the findings of a short-term study of 10 men and women with a higher mean iodine excretion (139 p.g/24h), where the CV%s varied between 14% and 78% (Rasmussen et al., 1999). 

Table 44.2 Variation in urinary iodine concentrations and estimated 24 h urinary iodine excretion in individuai sampies and in average annuai vaiues over 1 year...

This is a rather simple model to describe the components of variation that can be obtained from most statistical programmes using analysis of variance (ANOVA) facilities, and the interpretation here is that the biological variation in urinary iodine excretion is around 2.5 times larger than the variation between individuals. 

The iodine content of different foods varies widely, and variations in food choice are important for variations in iodine excretion. Thus, a meal of cod may supply 400 times more iodine than a meal of meat from a terrestrial animal (Andersen et al, 2002a), and iodine excretion may vary by a factor of 4 within the same society, depending on the composition of meals (Andersen et al., 2005). Variable bioavailability may add to the variation in iodine excretion... 

Variation in Urinary Iodine Excretion and Studies of Iodine Nutrition... 

Populations are evaluated in the study of iodine nutrition. Population iodine excretion is described by two variables mean and variation. When describing iodine excretion in populations, a common approach is to compare mean or median values. However, this value may be identical, even though the populations differ markedly in distribution (Andersen et al, 2004), and thus differ despite similar mean values. Variance is thus another parameter to consider in the description of iodine excretion, and comparison of variances may differ from comparison of mean values. 

Variation in population minary iodine excretion is often described as the number of individuals with minary iodine excretion below a certain value. Such a number is, however, a parameter in the description of the variance characteristic to the population, and not a fraction of the population with low urinary iodine, as interpreted in quite a number of studies. 

Variable fluid intake causes variable volumes of urine. Around 90% of dietary iodine is excreted in the urine, and variable urine volumes cause variable dilution of the iodine excreted in urine, and thus in the concentration of iodine in urine. If the variation in dilution is corrected, the variation in urinary iodine content is lowered. This can be achieved by correcting for an analyte that is excreted in parallel with urine volume. 

In the study of variation in urinary iodine excretion, an estimated 24 h urinary iodine excretion was calculated from age-, gender- and ethnic-specific creatinine excretions (Kampmann et al., 1974 Kesteloot and Joossens, 1996). The reduction in variation in iodine excretion is seen in Figure 44.5. 

The variation in estimated 24 h urinary iodine excretion is clearly lower than the variation in iodine concentration in spot urine samples. This improves the accuracy of calculations based on variation in iodine excretion (Andersen et al., 2007a). 

Figure 44.5 Variation in urinary iodine excretion expressed as crude urinary iodine content ( rg/l) and as 24 h urinary iodine excretion estimated from creatinine excretion in an age- and gender-matched group ( rg/24h).

The variation in urinary iodine excretion affects the reliability of estimates of population iodine nutrition. Low urinary iodine is seen in iodine-replete individuals due to random variation (Andersen et al., 2001). However, a high number of samples increases the reliability of the estimates of iodine excretion in a population, but what is the reh-ability of a study including a certain number of spot urine samples from a population ... 

Table 44.4 Number of spot urine samples needed to be 95% oonfident of being within a speoified range for crude urinary iodine con-oentration and for estimated 24 hour urinary iodine exoretion oaloulated from the variation in iodine excretion among heaithy men undertaking daily lives (Andersen et al., 2007b)...

In the smdy of variation in more than one variable, variations can be correlated. Such associations may be analyzed (Feldt-Rasmussen et al, 1989). The association between iodine excretion and serum TSH was studied in 15 healthy men in an area with mild-to-moderate iodine deficiency (Andersen et al., 2001). The association differed between individuals, and when individuals were grouped according to urinary iodine excretion levels, a negative correlation was found only in individuals with an iodine excretion below 50iig/24h (Figure 44.6). 

Variations in urinary iodine portray variations in iodine intake. However, huge variations in sample iodine content do not reflect similar variations in the iodine nutrition in that population. This is important for risk estimation in the evaluation of population iodine intake, and has important implications for studies of iodine nutrition. Furthermore, the reliability of studies of iodine nutrition and the number of urine samples needed to assess iodine excretion level can be calculated. 

Adjusting urinary iodine excretion for age-, gender-and ethnic-specific creatinine excretions markedly reduces variation in urinary iodine. 

Figure 119.3 Geographical variations in urinary iodine excretion in Denmark before iodine fortification of salt. Median urinary iodine excretion among inhabitants of various Danish cities before iodine fortification of salt, and the estimated number of people living in areas with different levels of urinary iodine excretion. Values were compiled from different studies of urinary iodine excretion, or estimated from measurements of groundwater iodine content. Geographical variation in iodine intake in Denmark is mostly determined by differences in groundwater iodine content. Pedersen etal., (1999) Rasmussen etal., (2000).

---