In cats, hyperthyroidism is most commonly caused by autonomously functioning thyroid gland adenomas1–3 and is clinically and histologically similar to toxic nodular goiter in humans.1 In elderly humans, a mild to moderate lifetime deficiency of iodine increases the risk of thyroid gland autonomy and toxic nodular goiter.4,5 Hyperthyroidism in cats is hypothesized to be caused by the same mechanism as that proposed for the development of toxic nodular goiter in humans6 in that a mild to moderate iodine deficiency decreases production of T4 by the thyroid gland, which causes a reciprocal increase in the release of TSH through hypothalamic-pituitary gland feedback. This increase in the amount of circulating TSH increases the efficiency of iodine utilization by thyrocytes, allowing the serum T4 concentration to remain within reference limits. However, continuous hyperstimulation by circulating TSH leads to a chronically increased rate of metabolism in thyrocytes, which could subsequently cause genetic dysfunction, allowing the permanent activation of TSH-dependent stimulatory pathways in thyroid gland tissues. Such activating mutations of TSH receptors or associated G-proteins have been detected in cats7–9 and in humans with toxic nodular goiter.10,11 These mutations result in the formation of autonomous hyperplastic thyroid gland cells that produce T4 independent of TSH and the development of clinical hyperthyroidism.
In humans, high dietary intake of iodine is associated with reduced synthesis of thyroid gland hormones (the Wolff-Chaikoff effect). However, an escape mechanism typically allows the thyroid gland to adjust to chronically high exposure to iodine, and excessive iodine is not known to increase the risk of nodular goiter in humans.12 In contrast, rats with chronically high dietary intake of iodine have an increased risk of developing adenomatous thyroid gland changes, compared with rats fed lower amounts of iodine.13
In cats, it is therefore possible that a lifetime deficiency or excess of iodine could cause adenomatous changes in thyroid glands as cats mature. An association between the ingestion of canned food and the development of hyperthyroidism has been repeatedly detected in cats.14–18 Some canned cat foods contain excessive or deficient iodine concentrations, with a wide range of iodine concentrations reported.19,20 In adult humans, a dietary intake of 100 to 200 μg of iodine/d is recommended,5,12 and there is evidence of an increased risk of thyroid gland disease with long-term intake of iodine higher or less than this limited range.
Some dietary iodine supplementation programs in geographic regions of iodine deficiency have triggered hyperthyroidism in elderly humans with non-toxic nodular goiter.21,22 This form of iodine-induced hyperthyroidism is limited to those humans with preexisting, autonomously functioning nodules that are highly capable of converting iodine to thyroid gland hormones and that are not susceptible to the Wolff-Chaikoff effect. In addition, a study23 of Moroccan children before an iodine supplementation program was introduced and 1 year after the program was stopped revealed that hypothyroidism was more prevalent after than before the supplementation program. The increase in prevalence took place despite a similar degree of iodine deficiency prior to and after the program. This finding suggests that the thyroid gland is able to sustain thyroid gland hormone production better when the dietary intake of iodine is constantly suboptimal as opposed to when intake varies between suboptimal and optimal. Therefore, it is also possible that highly variable iodine intake could contribute to the development of hyperthyroidism in cats. Cats that consume various flavors of wet cat food reportedly have an increased risk of developing hyperthyroidism, compared with those that consume only 1 flavor.16
Direct interactions between changes in dietary intake of iodine and serum concentrations of thyroid gland hormones in cats have been reported.24–27 Two studies26,27 involving cats revealed decreases in serum free and total T4 concentrations after short-term (1- to 2-week) increases in iodine intake, and the decrease in serum T4 concentration was proportionate to the amount of iodine consumed. However, a third study24 revealed no significant difference in free T4 concentrations between cats fed a diet high in iodine versus those fed a diet low in iodine for a 3-month period.
Iodine is primarily removed from the body by renal excretion. In healthy cats, the UI concentration of a urine sample accurately reflects the amount of recent iodine intake, and the amount of fecal iodine excretion is constant in relation to body weight.19 Therefore, the UI concentration can be used to estimate dietary iodine intake in healthy cats.
In humans, CKD results in a marked decrease in renal clearance of iodide, with a concurrent increase in serum inorganic iodide concentration in humans with end-stage renal disease.28,29 Accumulation of iodide in the body happens despite a decrease in dietary intake (attributable to decreased appetite) and is persistent even after several weeks of restricting the amount of iodide consumed.30 Although it is uncommon for cats to be azotemic when hyperthyroidism is diagnosed, CKD is commonly diagnosed in hyperthyroid cats after treatment for hyperthyroidism.31–33 Hyperthyroidism hides the biochemical signs of renal disease by increasing the glomerular filtration rate and decreasing muscle mass, thus lowering the plasma creatinine concentration in affected cats.34 In most cats that develop CKD in the months after successful treatment for hyperthyroidism, it is likely that renal disease already existed when hyperthyroidism was diagnosed.
To the authors' knowledge, the relationship between UI concentration and dietary iodine intake has not been evaluated in hyperthyroid cats. Physiologic changes associated with hyperthyroidism, such as an increase in fecal volume, a decrease in gastrointestinal transit time, and renal disease, may alter the kinetics of iodide excretion by the body such that the UI concentration no longer accurately reflects the degree of iodine intake. In addition, hyperthyroidism causes polyphagia, which might increase the amount of iodine consumed.
The purpose of the study reported here was to compare UI concentrations in euthyroid and untreated hyperthyroid cats. Specifically, we sought to compare the variability of UI concentrations in cats prior to the diagnosis of hyperthyroidism with those of euthyroid cats > 8 years old, evaluate the effect of treatment on the UI concentration in hyperthyroid cats, determine the effect of renal disease on UI concentration, and compare UI concentrations in hyperthyroid cats at diagnosis with those of an age-restricted population of euthyroid cats. Another purpose was to estimate daily dietary iodine intake among those euthyroid cats that had a plasma creatinine concentration within the reference range.
Materials and Methods
Animals—The Royal Veterinary College feline research group examines, diagnoses, and treats cats at 2 nonreferral veterinary practices in London. At each visit, historical and clinical data are recorded, and blood and urine samples are collected for routine biochemical analysis and urinalysis. Data collected in these research clinics are recorded in a central database that was used to identify suitable cats for this retrospective study. Included diseased cats had one or more of the following conditions: CKD, hypertension, or hyperthyroidism. Healthy cats > 8 years old that had been evaluated for routine (every 6 to 12 months) health screening were also included. All cats were evaluated between September 1994 and May 2007. The ethics and welfare committee of the Royal Veterinary College approved the diagnostic protocol. Owner consent was obtained prior to sample collection.
Sample collection—Jugular venous blood samples were collected into heparinized tubes. Plasma was separated from the blood cells within 6 hours after collection, for measurement of plasma concentrations of various analytes (total protein, albumin, creatinine, urea nitrogen, bilirubin, cholesterol, potassium, chloride, sodium, calcium, phosphorus, and tT4) and plasma activities of alanine transferase and alkaline phosphatase.
Urine was collected by cystocentesis, and a full urinalysis (USG as measured by refractometry, semiquantative biochemical urinalysis, and microscopic sediment examination) was performed on the date of collection. Residual urine was centrifuged at 1,000 X g for 10 minutes and the supernatant stored at −80°C until analyzed for UI concentration. The UI analyses were performed in batches between June 2005 and August 2007.
Identification and evaluation of euthyroid cats—Healthy euthyroid cats were identified on the basis of clinical history and results of physical examination and routine laboratory analyses (plasma biochemical analysis, PCV determination, and urinalysis). None of the cats in this category was receiving medication other than antiparasitic drugs. To be classified as euthyroid, cats were required to have values for plasma and free T4 concentrations that were within reference ranges (19 to 55 nmol/L and 19 to 40 pmol/L, respectively) when measured; however, additional criteria were applied that varied according to the plasma creatinine concentration. For the purpose of the study, nonazotemic cats (plasma creatinine concentration ≤ 177 μmol/L) were required to have a plasma tT4 concentration < 40 nmol/L, and azotemic cats (plasma creatinine concentration > 177 μmol/L) were required to have a plasma tT4 concentration < 30 nmol/L. Different inclusion criteria for azotemic and nonazotemic cats were used because plasma tT4 concentration decreases with some nonthyroidal illnesses, including CKD.35,36 Cats with CKD that had tT4 measurements in the range of 30 to 55 nmol/L were excluded because azotemic cats with a tT4 value in this range may have hyperthyroidism.37
Identification and evaluation of hyperthyroid cats—Hyperthyroidism was diagnosed in cats on the basis of a plasma tT4 concentration that exceeded the upper reference limit (reference range, 19 to 55 nmol/L) or results of a T3 suppression test that exceeded the reference limit (tT4 > 25 nmol/L after administration of one 25-μg T3 tablet 3 times/d for 7 doses).38 Total T4 analyses for most cats, including all hyperthyroid cats and cats with CKD, were processed at 1 commercial laboratory; however, a subset of tT4 analyses in euthyroid cats was performed at the Royal Veterinary College laboratory by use of the same methods.
Cats identified with hyperthyroidism were treated with carbimazolea (5 mg, q 8 h, for 10 to 21 days), usually with a reduction to twice-daily dosing thereafter, depending on the clinical response. After initiation of treatment, cats were reexamined every 2 to 3 weeks until their condition was considered stable and every 6 to 8 weeks afterward. Some cats subsequently underwent unilateral or bilateral thyroidectomy on the basis of their age, health status, and tolerance of the medication and owner preference. Only successfully treated cats (ie, those with a plasma tT4 concentration < 40 nmol/L) were included in the study. For cats that did not tolerate orally administered medication, the treatment start date was considered to be the date of thyroidectomy. Renal function was assessed by measurement of plasma creatinine concentration and USG prior to and after successful treatment of hyperthyroidism.
Identification of cats with CKD—Chronic kidney disease was diagnosed in cats on the basis of clinical signs (eg, polydipsia, polyuria, or vomiting), results of renal palpation (ie, identification of subjectively small or misshapen kidneys), results of full urinalysis and bacteriologic culture of urine (when necessary), plasma electrolyte concentrations, plasma creatinine concentration, plasma urea nitrogen concentration, and PCV. Renal function was assessed in accordance with the staging system39 of the IRIS. Cats were classified as having normal renal function when the plasma creatinine concentration was < 140 μmol/L. Cats with a plasma creatinine concentration ≥ 140 μmol/L were separated into 2 groups: plasma creatinine concentration from 140 to 177 μmol/L (IRIS category 2a) and plasma creatinine concentration > 177 μmol/L (IRIS category ≥ 2b). For hyperthyroid cats, the IRIS category was based on the mean plasma creatinine concentration after treatment for hyperthyroidism (within 4 months of treatment initiation). For euthyroid cats, the plasma creatinine concentration on the date of urinalysis was used. Cats were considered to have CKD when the plasma creatinine concentration exceeded the upper reference limit (ie, > 177 μmol/L) and the USG was < 1.035. With the exception of hyperthyroid cats after treatment, cats were excluded from the study when the plasma creatinine concentration was > 250 μmol/L at the time of assessment and urinalysis; hence, only cats with a mild degree of CKD were included in the study. Cats with CKD were also excluded when they had received a prescription low-protein renal diet that may have influenced the UI concentration.
Measurement of UI concentration—A modified Sandell-Koltoff method40 was used to measure UI concentration. Before processing, urine samples were routinely diluted with distilled water (50:50); however, many concentrated samples (ie, USG > 1.050) required greater dilution than this, with the maximum dilution required being 1:19. For the assay, 200 μL of diluted urine was incubated with 0.8 mL of ammonium persulphate at 100°C for 1 hour to remove substances that might interfere with the assay results. Aliquots of the resulting digest (50 μL) were transferred to a 96-well plate and incubated for 30 minutes with 150 μL of arsenious acid and 25 μL of ceric ammonium sulfate. The resulting color change in the sample was determined spectrophotometrically by measuring the change in absorbance at 405 nm.
The assay was validated by measurement of intraassay and interassay coefficients of variation (n = 10 urine samples/test), determination of the limit of detection, analytic recovery of potassium iodate from feline urine to which it had been added, and dilutional parallelism of urine samples with a high measured iodide concentration. The limit of detection of the assay was calculated as mean + 1.96 SD of repeated measurements of a blank (distilled water) sample. The UI concentration was normalized to urine concentration,41,b as defined by a USG of 1.010, with a correction factor of (1 − USG) × 100, to yield an index of UI concentration. In a subset of cats, urine creatinine concentrations were available, and the correlation between UI normalized to urine creatinine concentration and UI normalized to solute concentration (USG) was evaluated.
VARIABILITY OF UI CONCENTRATION WITH TIME
Urinary iodide concentrations were determined in 5 urine samples collected at various time points from euthyroid cats > 8 years of age and from hyperthyroid cats prior to diagnosis and treatment. Urine samples were collected for a minimum period of 12 months (maximum, 6 years). These data were also used in power calculations to determine the number of samples required to detect a difference between hyperthyroid and euthyroid groups in subsequent analyses.
EFFECT OF TREATMENT ON UI CONCENTRATION IN HYPERTHYROID CATS
When samples were available, UI concentration was measured in hyperthyroid cats at the time of, or prior to, diagnosis and at 2 to 6 weeks and 3 to 6 months after treatment for hyperthyroidism began. To assess the effect of treatment type on UI concentration, a subgroup of 28 cats that underwent thyroidectomy was identified (9 were treated by surgery alone, 11 were successfully treated with carbimazole prior to surgery, and 8 were unsuccessfully treated with carbimazole prior to surgery). In that subgroup, UI concentration was measured prior to diagnosis and 2 to 8 weeks after surgery.
COMPARISON OF UI CONCENTRATIONS IN HYPERTHYROID AND EUTHYROID CATS
Urinary iodide concentration was measured in untreated hyperthyroid cats and was compared with UI concentration in euthyroid cats. Potentially confounding factors such as age, body weight, plasma tT4 concentration, reported appetite, and whether food was withheld from the cat prior to urine collection were also recorded. Cats were considered to have had food withheld prior to sample collection when the owner reported that the cat had not been fed for at least 12 hours. Cats for which with-holding of food had not been recorded or cats that were reported to have eaten in the 12 hours before sample collection were classified as not having had food withheld. Appetite was categorized as good or increased or as normal or decreased on the basis of clinical record entries made on the date of cystocentesis. Renal function was assessed as previously described, and euthyroid and hyperthyroid cats were classified in accordance with the IRIS system.
Estimation of iodine intake—Iodine intake was only estimated for nonazotemic euthyroid cats. It was estimated from the UI concentration with the assumptions that a typical cat produces 15 mL of urine/kg/d, with a USG of 1.05042 and a fixed rate of fecal iodine excretion of 14 mg/kg/d.19
Statistical analysis—Statistical analysis was performed by use of a standard statistical package.c Data were assessed for normality of distribution graphically and with the Kolmogorov-Smirnov test. Age and body weight were compared between euthyroid and hyperthyroid groups by means of an independent samples t test. Breed and sex were compared with a χ2 test with continuity correction. Data for UI concentration had a right-skewed distribution; therefore, values were logarithmically transformed (logUI). The correlation between the logarithm of UI values normalized to a USG of 1.010 and the logarithm of the ratio of UI concentration to urine creatinine concentration was examined by calculation of the Pearson correlation coefficient. The geometric mean (central 95% range) was used as a summary measure of the distribution of the data. The range containing the central 95% of the data was estimated as the inverse logarithm of (mean logUI ± 1.96 × SD logUI).
For the evaluation of variability in UI concentrations with time for the same cat, a measure of agreement was calculated by use of a separate 1-way ANOVA for data from hyperthyroid and euthyroid cats. In each ANOVA, the 5 logUI measurements from each cat represented a group. The SE of the measurement was multiplied by 2.83 to provide a measure of agreement that estimated the maximum likely difference between 2 UI concentration measurements for the same cat.43 The data were also examined graphically by means of a Bland-Altman approach44 in which the mean logUI for each cat was plotted against the associated SD for each cat.
Values of logUI for hyperthyroid cats before treatment were compared with those after treatment by use of a paired t test after confirmation that the differences were normally distributed. Results for that analysis are reported as mean difference (95% CI).
Values of logUI for hyperthyroid cats were compared with those of euthyroid cats by means of 2-way ANOVA, with IRIS category and hyperthyroid status as the 2 factors. The assumptions were verified via examination of the residuals. A value of P < 0.05 was considered significant for all analyses.
Results
Urinary iodide measurement—Interassay and intra-assay coefficients of variation were < 12% over the range of measured concentrations (70 to 591 μg/L), and the recovery of iodide from control samples of known iodide concentrations ranged from 100% to 113%. Mean ± SD apparent iodide concentration of a blank (distilled water) sample evaluated in triplicate was 6.9 ± 4.2 μg/L, with a calculated limit of detection of 15 μg/L. The assay had excellent dilutional parallelism (R2 = 0.999). There was a significant (P < 0.001) correlation between UI concentration normalized to USG and the ratio of UI concentration to urine creatinine concentration (n = 208 samples; R2 = 0.81).
Variability of UI concentration with time—The UI concentration was measured in samples from 18 cats > 8 years old (5 samples/cat; 3- to 12-month intervals between samples). In 10 hyperthyroid cats, UI concentration was measured before diagnosis and treatment of hyperthyroidism (prehyperthyroid cats). During the study period, 7 of these 10 prehyperthyroid cats were being treated for concurrent disease including hypertension (n = 4) or both CKD and hypertension (3). The 3 remaining prehyperthyroid cats were apparently healthy. The 8 euthyroid control cats had been evaluated as part of their regular monitoring for hypertension (2) and hypertension and CKD (3) or were apparently healthy (3).
Urinary iodide concentration in all cats was highly variable, with variability in UI concentration that exceeded 10-fold in some cats. The geometric mean UI concentration was 280 μg/L (central 95% range, 73 to 1,079 μg/L) for the hyperthyroid cats and 254 μg/L (69 to 929 μg/L) for the euthyroid cats. The Bland-Altman plot of mean versus SD logUI for each cat revealed a lack of a so-called funnel effect, indicating that the SD was independent of the magnitude of the measurement (Figure 1). There was no significant (P = 0.85) difference in mean SD of the logUI values between the 2 groups. There was little difference in the variability of UI concentrations in the prehyperthyroid (maximum likely difference in logUI, 0.886) and euthyroid groups (maximum likely difference in logUI, 0.908). Results of power calculations derived from these data suggested that a total sample size of 200 cats would yield, on the basis of a 2-tailed t test, a power of 0.8 to detect a difference in mean logUI between 2 groups (effect size, 0.4; A = 0.05).
Effect of treatment on UI concentration in hyperthyroid cats—Eighty hyperthyroid cats were identified that had been successfully treated with carbimazole, thyroidectomy, or both, and for which there were 2 stored urine samples (1 obtained prior to treatment and 1 obtained 2 to 6 weeks after treatment). The UI concentration was significantly (P < 0.001) higher 2 to 6 weeks after treatment, compared with the value before treatment (mean [95% CI] difference, 109 μg/L [56 to 162 μg/L]), and this difference was apparent in cats with and without evidence of CKD (Figure 2). Urinary iodide concentration in cats with well-controlled hyperthyroidism appeared to return to the pretreatment concentrations by 3 to 6 months after treatment. There was no significant difference (P = 0.46; n = 44) between UI concentrations measured before and 3 to 6 months after treatment (mean [95% CI] difference, 37 μg/L [–59 to 133 μg/L]).
To assess whether the increase in UI concentration after treatment was confined to cats treated with carbimazole, the effect of surgery on UI concentration was examined. Urinary iodide concentration was measured for each cat before treatment and 2 to 8 weeks after surgery (median, 28 days; range, 11 to 52 days; n = 28). The UI concentration increased significantly (P < 0.001) after surgery, compared with the value before treatment (mean difference, 233 μg/L [147 to 319 μg/L]). This change was evident in cats treated with surgery alone (P = 0.02; 9) and even in cats in which hyperthyroidism had been controlled with carbimazole prior to surgery (P = 0.04; 11; Figure 3).
Comparison of UI concentrations in hyperthyroid and euthyroid cats—Urinary iodide concentration was measured in 118 euthyroid cats and in 88 hyperthyroid cats before treatment for hyperthyroidism (Figure 4). Hyperthyroid cats had lower body weights (P < 0.001), were older (P = 0.01), and were less likely to be purebred animals (P = 0.001) than the euthyroid cats (Table 1).
Characteristics of euthyroid cats and hyperthyroid cats before treatment for hyperthyroidism,* stratified according to whether cats were azotemic (plasma creatinine concentrations, > 177 μmol/L) or nonazotemic (plasma creatinine concentration, b 177 μmol/L).
Characteristic | Euthyroid | Hyperthyroid | ||||||
---|---|---|---|---|---|---|---|---|
Nonazotemic | Azotemic | Nonazotemic | Azotemic | |||||
No. of cats | Value | No. of cats | Value | No. of cats | Value | No. of cats | Value | |
Age (y) | 76 | 12.7 ± 2.6 | 39 | 14.2 ± 3.9 | 53 | 14.1 ± 2.4 | 33 | 14.3 ± 2.6 |
Body weight (kg) | 77 | 4.6 ± 1.1 | 41 | 3.7 ± 1.0 | 55 | 3.3 ± 1.0 | 33 | 3.3 ± 0.5 |
Purebred breed | 77 | 16 (21) | 41 | 9 (22) | 55 | 1 (2) | 33 | 3 (9) |
Female sex | 76 | 46 (61) | 41 | 26 (63) | 55 | 30 (55) | 33 | 18 (55) |
Creatinine concentration (μmol/L) | 77 | 131 (118–148) | 41 | 209 (197–227) | 55 | 138 (110–155) | 33 | 218 (186–240) |
UI concentration (μg/L) | 77 | 350 (239–516) | 41 | 249 (118–397) | 55 | 253 (147–383) | 33 | 200 (135–336) |
Values for age and body weight are reported as mean ± SD. Values for breed and sex are reported as number (percentage). Values for analytes are reported as median (interquartile range).
Renal function (presence or absence of azotemia) in hyperthyroid cats was classified according to mean plasma creatinine concentration after treatment.
Hyperthyroid cats were older (P = 0.01), weighed less (P < 0.001), and were less likely to be purebred animals (P = 0.001) than the euthyroid cats.
The 2-way ANOVA in which the mean logUI of these 2 groups of cats was compared revealed a significant (P < 0.001) effect of renal disease and hyperthyroid status on logUI (Table 2). These results were the same irrespective of whether IRIS category in the hyperthyroid cats was estimated by use of plasma creatinine concentrations measured before or after treatment. A polynomial contrast indicated that the linear trend for logUI to decrease with IRIS category was significant (P < 0.001). There was no significant interaction between IRIS category and hyperthyroid status.
Results of a 2-way ANOVA to assess the difference in UI concentration (μg/L) between euthyroid cats and untreated hyperthyroid cats, taking into account the effect of renal function.
Variable | No. of cats | UI concentration | |
---|---|---|---|
Geometric mean | Central 95% range | ||
Thyroid gland function | |||
E | 118 | 304* | 82–1,132 |
H | 88 | 222 | 56–879 |
Renal function | |||
Normal | 82 (52 E; 30 H) | 315†,‡ | 97–1,020 |
IRIS category 2a | 50 (25 E; 25 H) | 290 | 83–1,014 |
IRIS category ≥ 2b | 74 (41 E; 33 H) | 209 | 46–959 |
Values are significantly different (P = 0.002) between groups.
Values are significantly different (P = 0.001) among groups.
The linear trend for logUI to decrease with increasing IRIS severity category was significant (P < 0.001).
E = Euthyroid. H = Hyperthyroid. Normal = Normal renal function (plasma creatinine concentration < 140 μmol/L). IRIS category 2a = Plasma creatinine concentration from 140 to 177 μmol/L. IRIS category ≥ 2b = Plasma creatinine concentration > 177 μmol/L.
Withholding food from cats for at least 12 hours before collection of urine (vs not withholding food) had no significant effect on logUI in the apparently healthy euthyroid cats (P = 0.15; n = 76) or in both groups of cats considered together (P = 0.89; 205). In the hyperthyroid group, there was no significant (P = 1.0) difference in mean logUI in cats reported to have a good to increased appetite (n = 43), compared with mean logUI in cats reported to have a reduced to normal appetite (45).
Estimated iodine intake in euthyroid cats > 8 years of age—The median (range) estimated iodine intake was 41 μg/kg of body weight/d (22 to 116 μg/kg of body weight/d) for the apparently healthy euthyroid cats. Therefore, none of the cats had an estimated iodine intake lower than the recommended minimum of 20 μg/kg of body weight/d.19
Discussion
In the study reported here, UI concentration was measured in client-owned hyperthyroid cats before and after treatment for hyperthyroidism. Results indicated that UI concentration was lower in untreated hyperthyroid cats, compared with UI concentration in euthyroid cats > 8 years of age, even when the effects of renal disease were taken into account.
The UI concentration of single urine samples was used in this study rather than 24-hour urine excretion profiles, and an indirect measure of solute concentration (USG) of the urine was used to normalize acquired UI concentration values with respect to urine concentration. This approach was validated by comparing the UI concentration normalized to USG with the UI concentration normalized to urine creatinine concentration; values for the 2 approaches were well correlated.
Overall, there was large within-cat variability with respect to UI concentrations measured at various times. This variability in UI concentration has also been detected in humans, in which it is suggested that at least 100 samples be collected from each group of humans used to compare UI concentrations in single urine samples.45 In our study, a power calculation based on the within-cat variability data also suggested an optimal group size was 100 cats to detect a difference in logUI between 2 groups.
Several potential confounding variables were difficult to assess and make adjustments for in the present study. It was initially hypothesized that the effect of polyphagia in hyperthyroid cats would increase the UI concentration, and therefore UI concentration was measured 2 to 6 weeks and 3 to 6 months after successful treatment in subsets of cats. Results indicated that UI concentration after treatment significantly increased (2 to 6 weeks after treatment) or was the same (3 to 6 months after treatment) as the pretreatment UI concentration. In addition, we did not detect any difference between cats reported to have a good or increased appetite and cats reported to be anorexic or have a normal appetite, although it should be considered that such measurements are highly subjective. Therefore, there was no evidence of a significant confounding effect of polyphagia on the results. Even if a confounding effect had existed, it would have tended to reduce the detected differences between the hyperthyroid and euthyroid groups.
Another finding of the present study was that UI concentration was significantly lower in cats with CKD than in cats without evidence of renal dysfunction. This finding is in agreement with findings of studies28–30 in humans that indicate that decreases in UI concentration and renal clearance of iodide are associated with a concurrent increase in plasma or serum iodide concentration. It would have been interesting to determine whether the decreased UI concentration in cats with mild CKD in the present study was associated with concurrent increases in plasma iodide concentration. Although high plasma iodide concentrations can inhibit production of thyroid gland hormones in the short term, this inhibition is eventually overcome with a concurrent increase in plasma TSH concentration in most clinically normal humans.46 However, in humans with preexisting thyroid gland disease and with certain other diseases such as renal disease, the Wolff-Chaikoff escape mechanism may fail, and the excess plasma iodide concentration may lead to hypothyroidism.30 Indeed, hypothyroidism is reportedly more common in humans with end-stage renal disease than in the general population.47 Renal disease in cats is associated with suppression of the serum tT4 concentrations35,36; however, this effect of renal disease is not believed to cause clinical hypothyroidism.
Iodide is excreted through the kidneys primarily via glomerular filtration,29 with partial tubular reabsorption,48 and therefore excretion of iodide is dependent on both the GFR and tubular function. In the present study, we estimated the GFR and renal function by measurement of plasma creatinine concentrations because direct measurement of the GFR was not possible. Results indicated that the UI concentration was lower in hyperthyroid cats than in euthyroid cats, even when decreased renal function was accounted for in the analysis. However, renal function of cats with untreated hyperthyroidism is difficult to measure accurately because of changes in the GFR and tubular function that develop secondary to the hyperthyroid state.34 Values for plasma creatinine or urea nitrogen concentration, or USG obtained before treatment for hyperthyroidism are not useful for predicting renal function after treatment.49 The increase in GFR associated with hyperthyroidism in cats34 might be expected to transiently increase UI concentration. The effects of hyperthyroidism on the tubular resorption of iodide are unknown; however, once a steady state is reached, the plasma UI concentration should once again reflect the amount of iodine ingested.
Assuming that pretreatment plasma creatinine concentrations were not representative of the true renal status of cats, we attempted to define the hyperthyroid cats as having renal disease or not on the basis of their post-treatment plasma creatinine concentration. Because the plasma creatinine concentration in hyperthyroid cats is almost invariably higher after rather than before treatment for hyperthyroidism, the approach used in the present study was likely to overestimate the effect of renal disease on UI concentration. This possible overestimation of the effect of renal function results from assigning hyperthyroid cats to a higher IRIS category for the ANOVA analysis because of the higher posttreatment plasma creatinine concentration. Indeed, as might be expected, the use of pretreatment plasma creatinine concentration in the analysis did not change the results. Our attempt to account for the effects of renal disease on UI concentration in the hyperthyroid cats prior to treatment may have been inaccurate. However, the effects of altered renal function on UI concentration would be expected to be only transient because, once the plasma iodide concentration stabilized, the amount of iodide filtered in the glomerulus would change proportionately, and a new steady state would be achieved in which iodine intake again matches iodide excretion. Thus, the only valid explanations for changes in iodine excretion (for an animal in a steady state) are changes in iodine intake or changes in the amount of iodine or iodide that is lost through routes other than the kidneys.
The proportion of ingested iodine excreted in feces may increase in hyperthyroid cats, compared with the proportion excreted in urine. A study50 of intestinal absorption of radiothyroxine and radiotriiodothyronine in clinically normal cats revealed that when thyroid gland hormones are secreted into the bowel through the bile and duodenal wall, they are poorly absorbed, leading to a net excretion of iodine in the feces. Most of this fecal iodine is organic iodine in T4, with approximately 10% excreted as free iodide. In the hypermetabolic state associated with hyperthyroidism, it is possible that the proportion of iodine excreted in the feces may increase, resulting in a decrease in the proportion of iodide excreted in the urine. A study51 of iodide, T4, and T3 metabolism revealed that fecal excretion of T4 was higher in hyperthyroid cats, compared with fecal excretion in euthyroid cats. It is therefore possible that the increase in UI concentration after treatment for hyperthyroidism and the difference between untreated hyperthyroid cats and euthyroid cats was at least in part a reflection of a change in the proportions of iodine or iodide excreted in feces versus urine.
In the present study, the marked but transient increase in UI concentration in the hyperthyroid cats in the weeks after treatment suggested that untreated hyperthyroid cats have a significant pool of stored iodine in the body. Treatment of cats with hyperthyroidism appeared to cause a significant increase in the loss of iodine from the body via the urine, presumably because of the deiodination of excess thyroid gland hormones. This increase in UI concentration was also detected in cats treated with surgery alone and is therefore not solely related to the release of iodine stored within the thyroid gland itself. Cats have a poor ability to deiodinate the inactive thyroid gland hormone rT3 and recycle their thyroid gland hormones. In cats, type I iodothyronine deiodinase is unable to use unconjugated rT3 as a substrate52,53 but can deiodinate sulphated rT3.53 This decreased ability to deiodinate inactive thyroid gland hormones in cats versus humans is reflected in a greater serum rT3/T4 fraction in euthyroid cats.53 Given that cats have a much higher metabolic clearance rate of tT4 than humans,51,54 it is possible that in addition to increased plasma tT4 and T3 concentrations, hyperthyroid cats will have a large circulating pool of inactive thyroid gland hormones. This pool may contribute to increased iodide excretion after treatment for hyperthyroidism.
If the UI concentrations in hyperthyroid cats in the present study were indeed reflective of iodine intake during the development of hyperthyroidism, then reduced iodine intake may be a risk factor for the development of the disease. In humans, results of large population-based studies4,55 suggest that even small decreases from optimal iodine intake, as assessed by UI concentration, can increase the risk of development of toxic nodular goiter, a disease with many similarities to hyperthyroidism in cats. Results of the present study indicate that hyperthyroid cats had a lower UI concentration at the time of diagnosis than did euthyroid cats. However, the influences on iodine metabolism in cats are complex; a simple cause-and-effect relationship of iodine intake on the development of hyperthyroidism has not been established. Establishment of such an effect would require prospective dietary trials; however, the duration and expense of these would be a considerable obstacle to study completion. Alternatively, a study of the relationship between dietary intake of iodine and plasma TSH concentration might shed light on this relationship. However, the only commercially available TSH assay validated for use in cats37 is a canine TSH assay, and this assay may not be sufficiently sensitive for this purpose.
ABBREVIATIONS
CI | Confidence interval |
CKD | Chronic kidney disease |
GFR | Glomerular filtration rate |
IRIS | International Renal Interest Society |
logUI | Logarithmic transformation of urinary iodide concentration |
rT3 | Reverse triiodothyronine |
T3 | Triiodothyronine |
T4 | Thyroxine |
TSH | Thyroid-stimulating hormone |
tT4 | Total thyroxine |
UI | Urinary iodide |
USG | Urine specific gravity |
Neomercazole, Amdipharm, Essex, England.
Syme HM, Elliott J. Comparison of urinary albumin excretion normalized by creatinine concentration or urine specific gravity (abstr). J Vet Intern Med 2005;19:466.
SPSS for Windows, version 15.0, SPSS Inc, Chicago, Ill.
References
- 1.↑
Gerber H, Peter HJ, Ferguson DC, et al. Etiopathology of feline toxic nodular goitre. Vet Clin North Am Small Anim Pract 1994;24:541–565.
- 2.↑
Hoenig M, Goldschmidt MH, Ferguson DC, et al. Toxic nodular goitre in the cat. J Small Anim Pract 1982;23:1–12.
- 3.
Peter HJ, Gerber H, Studer H, et al. Autonomy of growth and of iodine metabolism in hyperthyroid feline goiters transplanted onto nude mice. J Clin Invest 1987;80:491–498.
- 4.
Laurberg P, Pedersen KM, Vestergaard H, et al. High incidence of multinodular toxic goiter in the elderly population in a low iodine intake area vs high incidence of Graves disease in the young in a high iodine intake area—comparative surveys of thyrotoxicosis epidemiology in East Jutland Denmark and Iceland. J Intern Med 1991;229:415–420.
- 5.
Laurberg P, Jorgensen T, Perrild H, et al. The Danish investigation on iodine intake and thyroid disease, DanThyr: status and perspectives. Eur J Endocrinol 2006;155:219–228.
- 6.↑
Court MH, Freeman LM. Identification and concentration of soy isoflavones in commercial cat foods. Am J Vet Res 2002;63:181–185.
- 7.
Watson SG, Radford AD, Kipar A, et al. Somatic mutations of the thyroid-stimulating hormone receptor gene in feline hyperthyroidism: parallels with human hyperthyroidism. J Endocrinol 2005;186:523–537.
- 8.
Hammer KB, Holt DE, Ward CR. Altered expression of G proteins in thyroid gland adenomas obtained from hyperthyroid cats. Am J Vet Res 2000;61:874–879.
- 9.
Ward CR, Achenbach SE, Peterson ME, et al. Expression of inhibitory G proteins in adenomatous thyroid glands obtained from hyperthyroid cats. Am J Vet Res 2005;66:1478–1482.
- 10.
Tonacchera M, Pinchera A. Thyrotropin receptor polymorphisms and thyroid diseases. J Clin Endocrinol Metab 2000;85:2637–2639.
- 11.
Tonacchera M, Agretti P, Chiovato L, et al. Activating thyrotropin receptor mutations are present in nonadenomatous hyperfunctioning nodules of toxic or autonomous multinodular goiter. J Clin Endocrinol Metab 2000;85:2270–2274.
- 13.↑
Boltze C, Brabant G, Dralle H, et al. Radiation-induced thyroid carcinogenesis as a function of time and dietary iodine supply: an in vivo model of tumorigenesis in the rat. Endocrinology 2002;143:2584–2592.
- 14.
Martin KM, Rossing MA, Ryland LM, et al. Evaluation of dietary and environmental risk factors for hyperthyroidism in cats. J Am Vet Med Assoc 2000;217:853–856.
- 15.
Kass PH, Peterson ME, Levy J, et al. Evaluation of environmental, nutritional, and host factors in cats with hyperthyroidism. J Vet Intern Med 1999;13:323–329.
- 16.↑
Olczak J, Jones BR, Pfeiffer DU, et al. Multivariate analysis of risk factors for feline hyperthyroidism in New Zealand. N Z Vet J 2005;53:53–58.
- 17.
Scarlett JM, Moise NS, Rayl J. Feline hyperthyroidism—a descriptive and case-control study. Prev Vet Med 1988;6:295–309.
- 18.
Edinboro CH, Scott-Moncrieff JC, Janovitz E, et al. Epidemiologic study of relationships between consumption of commercial canned food and risk of hyperthyroidism in cats. J Am Vet Med Assoc 2004;224:879–886.
- 19.↑
Ranz D, Tetrick M, Opitz B, et al. Estimation of iodine status in cats. J Nutr 2002;132:1751S–1753S.
- 20.
Johnson LA, Ford HC, Tarttelin MF, et al. Iodine content of commercially-prepared cat foods. N Z Vet J 1992;40:18–20.
- 21.
Bourdoux PP, Ermans AM, Mukalay wa Mukalay A, et al. Iodine-induced thyrotoxicosis in Kivu, Zaire. Lancet 1996;347:552–553.
- 22.
Stanbury JB, Ermans AE, Bourdoux P, et al. Iodine-induced hyperthyroidism: occurrence and epidemiology. Thyroid 1998;8:83–100.
- 23.↑
Zimmermann MB, Wegmuller R, Zeder C, et al. Rapid relapse of thyroid dysfunction and goiter in school-age children after discontinuation of salt iodization. Am J Clin Nutr 2004;79:642–645.
- 24.↑
Kyle AH, Tarttelin MF, Cooke RR, et al. Serum free thyroxine levels in cats maintained on diets relatively high or low in iodine. N Z Vet J 1994;42:101–103.
- 25.
Tarttelin MF, Ford HC. Dietary iodine level and thyroid function in the cat. J Nutr 1994;124:2577S–2578S.
- 26.
Tarttelin MF, Johnson LA, Cooke RR, et al. Serum free thyroxine levels respond inversely to changes in levels of dietary iodine in the domestic cat. N Z Vet J 1992;40:66–68.
- 27.
Ranz DK, Kraft W, Rambeck WA. Influence of increasing nutritional iodine intake on thyroid hormones T3, fT3, T4 and fT4 in cats. Tierarztl Prax Ausg K Klientiere Heimtiere 2003;31:238–243.
- 28.
Koutras DA, Rigopoul GA, Marketos SG, et al. Iodine metabolism in chronic renal-insufficiency. Nephron 1972;9:55–62.
- 29.↑
Beckers C, van Ypersele de Strihou C, Coche E, et al. Iodine metabolism in severe renal insufficiency. J Clin Endocrinol Metab 1969;29:93–96.
- 30.↑
Sato K, Okamura K, Yoshinari M, et al. Reversible primary hypothyroidism and elevated serum iodine level in patients with renal dysfunction. Acta Endocrinol (Copenh) 1992;126:253–259.
- 31.
Adams WH, Daniel GB, Legendre AM, et al. Changes in renal function in cats following treatment of hyperthyroidism using I-131. Vet Radiol Ultrasound 1997;38:231–238.
- 32.
Graves TK, Olivier NB, Nachreiner RF, et al. Changes in renal function associated with treatment of hyperthyroidism in cats. Am J Vet Res 1994;55:1745–1749.
- 33.
Becker TJ, Graves TK, Kruger JM, et al. Effects of methimazole on renal function in cats with hyperthyroidism. J Am Anim Hosp Assoc 2000;36:215–223.
- 34.↑
Adams WH, Daniel GB, Legendre AM. Investigation of the effects of hyperthyroidism on renal function in the cat. Can J Vet Res 1997;61:53–56.
- 35.
Peterson ME, Gamble DA. Effect of nonthyroidal illness on serum thyroxine concentrations in cats—494 cases (1988). J Am Vet Med Assoc 1990;197:1203–1208.
- 36.
Mooney CT, Little CJ, Macrae AW. Effect of illness not associated with the thyroid gland on serum total and free thyroxine concentrations in cats. J Am Vet Med Assoc 1996;208:2004–2008.
- 37.↑
Wakeling J, Moore K, Elliott J, et al. Diagnosis of hyperthyroidism in cats with mild chronic kidney disease. J Small Anim Pract 2008;49:287–294.
- 38.↑
Peterson ME, Graves TK, Gamble DA. Triiodothyronine (T3) suppression test—an aid in the diagnosis of mild hyperthyroidism in cats. J Vet Intern Med 1990;4:233–238.
- 39.↑
Elliott J. Staging chronic kidney disease. In: Elliott J, Grauer FG, eds. BSAVA manual of canine and feline nephrology and urology. 2nd ed. Kundli, India: Replika Press Pvt Ltd, 2007;159–166.
- 40.↑
Gnat D, Dunn AD, Chaker S. Fast colorimetric method for measuring urinary iodine. Clin Chem 2003;49:186–188.
- 41.↑
Caple IW, Doake PA, Ellis PG. Assessment of the calcium and phosphorus nutrition in horses by analysis of urine. Aust Vet J 1982;58:125–131.
- 42.↑
Hawthorne AJ, Markwell PJ. Dietary sodium promotes increased water intake and urine volume in cats. J Nutr 2004;134:2128S–2129S.
- 43.↑
Petrie A, Watson P. Additional topics. In: Statistics for veterinary and animal science. Oxford, England: Blackwell Science Ltd, 1999;168–180.
- 44.↑
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–310.
- 45.↑
Andersen S, Karmisholt J, Pedersen KM, et al. Reliability of studies of iodine intake and recommendations for number of samples in groups and in individuals. Br J Nutr 2008;99:813–818.
- 46.↑
Namba H, Yamashita S, Kimura H, et al. Evidence of thyroid volume increase in normal subjects receiving excess iodide. J Clin Endocrinol Metab 1993;76:605–608.
- 47.↑
Kaptein EM, Quionverde H, Chooljian CJ, et al. The thyroid in end-stage renal disease. Medicine 1988;67:187–197.
- 48.↑
Walser M, Rahill WJ. Renal tubular reabsorption of iodide as compared with chloride. J Clin Invest 1965;44:1371–1381.
- 49.↑
Syme HM. Cardiovascular and renal manifestations of hyperthyroidism. Vet Clin North Am Small Anim Pract 2007;37:723–743.
- 50.↑
Hays MT, Hsu LC, Kohatsu S. Transport of the thyroid-hormones across the feline gut wall. Thyroid 1992;2:45–56.
- 51.↑
Hays MT, Broome MR, Turrel JM. A multicompartmental model for iodide, thyroxine, and triiodothyronine metabolism in normal and spontaneously hyperthyroid cats. Endocrinology 1988;122:2444–2461.
- 52.
Foster DJ, Thoday KL, Beckett GJ. Thyroid hormone deiodination in the domestic cat. J Mol Endocrinol 2000;24:119–126.
- 53.↑
Kuiper GGJM, Wassen F, Klootwijk W, et al. Molecular basis for the substrate selectivity of cat type I iodothyronine deiodinase. Endocrinology 2003;144:5411–5421.
- 54.
Broome MR, Hays MT, Turrel JM. Peripheral metabolism of thyroid hormones and iodide in healthy and hyperthyroid cats. Am J Vet Res 1987;48:1286–1289.
- 55.
Laurberg P, Pedersen IB, Knudsen N, et al. Environmental iodine intake affects the type of nonmalignant thyroid disease. Thyroid 2001;11:457–469.