Effect of Anticonvulsant Dosages of Potassium Bromide on Thyroid Function and Morphology in Dogs
A placebo-controlled experiment was performed to evaluate the effect of potassium bromide on the canine thyroid gland. Basal total thyroxine, free thyroxine, and basal thyrotropin serum concentrations were evaluated over a 6-month period in potassium bromide-treated and control dogs. A thyrotropin-releasing hormone stimulation test was also performed in all dogs at the beginning and conclusion of the study. Thyroid histopathology was compared between treated and control dogs at the end of the study. No difference was detected in any parameter between the two groups at the end of the study. A decline in thyroid hormone concentrations over the course of the study did occur in both groups of dogs. Potassium bromide does not appear to have a significant effect on canine thyroid function or morphology.
Introduction
A number of nonthyroidal factors may influence the diagnosis of hypothyroidism in dogs, including age, breed, concurrent disease states, and administration of certain drugs.1–4 Glucocorticoids, sulfonamides, and phenobarbital are among some of the drugs known to affect the hypothalamic-pituitary-thyroid axis (HPTA).5–10 Experimental studies in rats suggest another commonly used anticonvulsant, potassium bromide (KBr), also affects thyroid function.1112 Treatment with bromide salts resulted in decreased serum thyroxine (T4) concentrations and parenchymatous goitre in rats.11–17 Concurrent elevations in thyroid-stimulating hormone (TSH) were not consistently found in these studies.
A recent prospective case series investigated the effect of phenobarbital alone, KBr alone, and phenobarbital and KBr together on thyroid function in the dog.18 This study failed to identify a significant effect of KBr on serum thyroid hormone concentrations. However, thyroid function was not evaluated prior to initiation of anticonvulsant therapy. Changes in serum thyroid hormone concentrations in response to KBr administration may have gone undetected without knowledge of hormonal status before treatment. In addition, it is possible that the effects of KBr on thyroid function were masked by concurrent administration of phenobarbital in dogs treated with dual therapy. To date, no prospective, placebo-controlled studies have investigated the effect of KBr on canine thyroid function, nor have they evaluated thyroid histopathological changes in response to KBr administration.
The purpose of this experiment was to test the hypothesis that administration of KBr in normal dogs significantly alters the function and histology of the thyroid gland.
Materials and Methods
Dogs
Ten adult (approximately 1 to 2 years of age), sexually intact, laboratory hound dogs (four male, six female) were housed and maintained in a US Drug Administration (USDA)-approved laboratory animal facility. All dogs received a standard laboratory ration.a Water was provided ad libitum. A physical examination, complete hematological and serum biochemical analyses, urine specific gravity (USG), serum basal total T4 (TT4)b and TSHc concentrations, serum free T4 (fT4) concentration,d anticanine thyroglobulin antibody (TgA) optical density (OD),e and a thyrotropin-releasing hormone (TRH) stimulation test were evaluated in each dog at baseline. The TRH stimulation test protocol was performed as previously described,19 and percentage change in TSH concentration was calculated using the following equation:
\(\frac{(postTRH\ TSH\ {-}\ preTRH\ TSH)\ {\times}\ 100}{preTRH\ TSH}\)Total T4 and TSH concentrations were measured using chemiluminescent enzyme immunoassays validated for use in the dog.19 Free T4 concentrations were measured by equilibrium dialysis validated for use in dogs.20
Experimental Protocol
Dogs were assigned such that baseline TT4, TSH, and sex were comparable between control and experimental groups. Five dogs (three female, two male) assigned to the experimental group were dosed orally with KBr dissolved in distilled water to a 200 mg/mL solution. Treated dogs received a loading dose of KBrf (100 mg/kg body weight, per os [PO], q 12 hours) for 2 days. This was followed by a maintenance dose of KBr (30 mg/kg body weight, PO, q 24 hours) in the evening for 180 days. The five control dogs (three female, two male) received an equivalent loading dose volume of distilled water placebo, PO, q 12 hours for 2 days followed by an equivalent maintenance dose volume, PO, q 24 hours in the evening for 180 days. The KBr dose was adjusted on day 120, with the aim to achieve serum bromide concentrations within the target range of 250 to 300 mg/dL by the end of the study. Dogs with serum bromide concentrations less than the low end of the target range on day 120 received a KBr dose increased by 5 mg/kg body weight for the remainder of the study. Potassium bromide or distilled water was administered each evening in a meatball of white bread and canine maintenance ration.g
Physical examinations were performed daily for the first week, every 2 weeks for the first 28 days, and then once every 30 days until completion of the study. All dogs were observed daily, and body weights were measured at baseline and on day 177.
Blood samples were collected in the morning from all dogs on days 3, 30, 120, and 177 for serum biochemical analysis and basal TT4 and TSH concentrations. A fT4 by equilibrium dialysis concentration and TRH stimulation test were repeated on day 177. Blood samples for serum bromide measurement were collected in the morning from KBr-treated dogs on days 3, 30, 120, and 177, and from control dogs on day 177. Serum bromide concentration was measured using a gold colorimetric assayh validated for use in dogs.21 All blood samples were processed on the day of collection. Serum was removed from each blood tube within 45 to 60 minutes. Serum samples were divided into four to five aliquots and stored in plastic tubes. These tubes were stored in a −20°C freezer within 90 to 120 minutes of collection and remained frozen until sample analysis was performed.22 Assays performed at Purdue University were completed within 48 to 72 hours of freezing. Serum samples for fT4 by equilibrium dialysis were shipped frozen on ice via overnight mail. Samples for serum bromide measurement were shipped via regular mail. All analyses requiring shipment to outside laboratoriesij were batched at each collection interval and mailed within 3 to 5 days of sampling.
Thyroidectomy and Tissue Processing
Unilateral thyroidectomy was performed in all dogs on day 182. Dogs were premedicated with butorphanol tartrate (0.3 mg/kg body weight, intramuscularly [IM]) and acepromazine maleate (0.03 mg/kg body weight, IM). Anesthesia was induced with intravenous (IV) sodium pentothal and maintained with isoflurane through a semiclosed rebreathing system. All dogs were positioned in dorsal recumbency, and the right lobe of the thyroid gland was approached through a midline skin incision. Unilateral thyroidectomy was performed using an extracapsular technique.2324 Thyroid tissue wet-weights were recorded in grams immediately after removal and expressed as g/kg body weight.
All thyroids were processed for light microscopy. Tissue was fixed in 10% neutral-buffered formalin. Formalin-fixed tissues were paraffin-embedded, and 5- to 6-μm sections were prepared.
Hematoxylin and eosin-stained sections were evaluated by a boarded pathologist (DeNicola) masked to group assignments. Each slide was assessed for evidence of thyroid activation. Thyroid activation refers to the following criteria that occur in response to hormone stimulation: enhanced vascularization, increased microfollicular development (MFD), decreased intrafollicular colloid staining (IFC), increased number of columnar follicular epithelial cells, and increased follicular epithelium mitotic figures. Increased MFD and decreased IFC were scored for each slide on a scale of 1 to 4 [Table 1]. Vascularity was scored by averaging the number of intermediate-sized blood vessels counted with the 10× objective in 10 fields. To evaluate the prevalence of mitotic figures in thyroid sections, calculation of a mitotic index (number of mitotic figures per follicle/total number of epithelial cells per follicle) was attempted. The predominance of columnar follicular epithelial cells was subjectively assessed.
Statistical Methods
Statistical comparisons were performed using a statistical software program.k The independent and joint influences of group (experimental versus control) and time on serum hormone concentration or body weight were evaluated by repeated measures analysis of variance (ANOVA).25 Thyroid wet-weights and thyroid histopathology scores were analyzed by the Wilcoxon’s signed rank sum test.26 Significance was defined as P<0.05 for all statistical analyses.
Results
Physical Examinations
Neither clinical signs of hypothyroidism nor evidence of bromism were identified in any of the dogs. There was a significant (P<0.0001) gain in body weight over time, but the weight change did not differ between experimental and control groups (P=0.46). However, the interaction for group X time was significant (P<0.01), reflected by more rapid weight gain of dogs in the experimental group. The difference in body weights between dogs in the experimental and control groups at day 177 was not significant (P=0.15) [Table 2].
Hematology, Serum Biochemical Analysis, and USG
Hematological and serum biochemical analyses and USG were within reference ranges in each dog at initiation of the study. Abnormalities identified on serum biochemical analyses of KBr-treated dogs were consistent with those expected due to administration of KBr (falsely elevated chloride and low anion gap).27 All other results were within reference ranges.
Anticanine Thyroglobulin Antibody
The median TgA OD for all dogs was 0.037 (range, 0.009 to 0.153). Eight dogs were negative for TgA, having ODs less than twice the OD of the negative control (negative control, 0.062 OD).28 Two dogs (one each from the treatment and control groups) were weakly positive for TgA [Table 2].
Serum Bromide Concentration
From day 30 to completion of the study, all serum bromide concentrations in KBr-treated dogs were within or exceeded the therapeutic range (88 to 300 mg/dL) recommended for epileptic dogs treated with KBr monotherapy.29 Three dogs exceeded the target serum bromide concentration (250 to 300 mg/dL) on day 120. One of these three dogs was within the target range, and one exceeded it on day 177. The third dog had a serum bromide concentration within the therapeutic range but below the target range. At the end of the study, four of five dogs were within the therapeutic range, one of which was within the target range of 250 to 300 mg/dL. The remaining dog had a serum bromide concentration of 313 mg/dL [Figure 1]. On day 177, serum bromide concentrations in three of five control dogs measured 0 mg/dL. Two control dogs had serum bromide concentrations of 8 and 9 mg/dL, respectively; however, on repeat analysis, bromide concentrations were 0 mg/dL in both dogs.
Thyroid Hormone Analysis
There was no significant difference between experimental and control groups for serum basal TT4, basal TSH, fT4, or results of TRH stimulation tests [Tables 2, 3; Figures 2, 3]. There were significant decreases in TT4 (P<0.0001) and fT4 (P=0.0002) over time, but these changes did not differ between the experimental and control groups [Figures 2, 3]. At baseline, one control dog had a basal TT4 concentration less than the established reference range (1.3 to 4.0 μg/dL) and a TSH concentration in the reference range (0 to 0.65 ng/mL). The postTRH TT4 concentration in this dog was consistent with euthyroidism (1.7 μg/dL). On day 177, five dogs (three KBr-treated dogs and two control dogs) had basal TT4 concentrations below the reference range. The fT4 of one KBr-treated dog was below the reference range (9 to 40 pmol/L) on day 177. Thyroid-stimulating hormone concentrations in all dogs were within the reference range (0 to 0.65 ng/mL). Four of the five KBr-treated dogs had postTRH TT4 concentrations >1.5 μg/dL. One KBr-treated dog had a postTRH TT4 concentration of 1.3 μg/dL. On day 0, percent TSH concentration change following TRH stimulation was <100% in all dogs. All KBr-treated dogs and four of five control dogs had a percent TSH concentration change following TRH stimulation >100% on day 177.
Thyroid Wet-Weight
No significant difference in thyroid wet-weight (P>0.05) was found between experimental and control dogs [Table 4]. Thyroid wet-weights (mean±standard deviation [SD]) of the experimental and control groups were 0.037±0.017 and 0.045±0.015 thyroid g/kg body weight, respectively.
Thyroid Histopathology
In the scored categories of MFD and IFC [Table 4], no significant difference was found between the KBr-treated dogs and control dogs (P>0.05). However, most dogs (9/10 for MFD and 8/10 for IFC) scored outside of normal for these categories. The degree of vascularity identified in thyroid sections was considered appropriate and was not different between the treatment and control groups. Calculation of a mitotic index for thyroid sections was not possible because of the difficulty of differentiating apoptotic cells from mitotic cells. In general, mitotic figures appeared primarily in areas of microfollicular development; however, they were rarely seen in these regions (0 to 1/40× objective field). With respect to the prevalence of columnar epithelium in thyroid sections, no morphological differences were identified when microfollicular structures were compared with normal-appearing follicles. Tall cuboidal to low columnar epithelial cells prevailed in microfollicular structures compared to the primarily low cuboidal epithelial cells of normal-appearing follicles. Neither normal follicular nor microfollicular regions of thyroid sections were different between the treatment and control groups. No significant inflammatory infiltrates consistent with thyroiditis were found in any of the thyroid sections.
Discussion
The ability of the thyroid to concentrate iodide and subsequently incorporate this halide into thyroglobulin is essential to thyroid hormone synthesis. Experimental studies performed in rats and in vitro using sheep thyroid slices demonstrated that the thyroid was able to concentrate other ions of the group VII elements (i.e., fluorine, chlorine, bromine, iodine, and astatine).30–33 In addition to the relative lack of anion specificity of the iodide uptake mechanism, halogen ions were also found to act as competitive inhibitors of iodide accumulation and cause release of accumulated iodide.3132 Based on these findings, it was hypothesized that bromide treatment might decrease thyronine synthesis. Experimental studies performed in rats confirmed that administration of bromide salts resulted in significantly decreased serum T4 and triiodothyronine (T3) concentrations with concurrent goitrous alterations of the thyroid gland.11121415 Further experiments with bromide-treated rats were performed to clarify the mechanisms by which bromide caused these abnormalities. Van Leeuwen, et al., investigated thyroid peroxidase activity in thyroid homogenates from sodium bromide-treated rats.16 Their findings suggested that in addition to inhibiting iodide uptake, bromide strongly inhibited the oxidation of iodide to iodine by hydrogen peroxide and, to a lesser degree, inhibited iodinated tyrosine residue coupling to thyronine.
The recommended therapeutic range for serum bromide concentrations in epileptic dogs treated with KBr monotherapy has been reported as 88 to 300 mg/dL. A target serum bromide concentration of 250 to 300 mg/dL was chosen for this study to ensure maximal effect on thyroid function. In addition, most epileptic dogs treated with KBr monotherapy require bromide concentrations in the upper therapeutic range for seizure control.29 Although not all KBr-treated dogs reached this target range, 90% of dogs reached or exceeded the therapeutic range recommended for KBr monotherapy from day 30 to completion of the study. No dogs that exceeded recommended serum bromide concentrations exhibited signs of bromide toxicity. Low serum bromide concentrations identified in two control dogs were considered to be spurious, as on repeat analysis these concentrations were found to be zero. In addition, both lipemia and hemolysis can cause false-positive results in serum bromide concentration measurement by gold colorimetric assay.
Serum basal hormone concentrations (TT4, TSH, and fT4) did not differ between the two groups at any time point in the study. This is in agreement with previously published work that also failed to find an effect of bromide administration on canine thyroid function.18 Bromide is a relatively weak competitive inhibitor of iodide uptake by the thyroid gland, and compared to iodide, it is concentrated by the thyroid to a much smaller degree.30–32 Although bromide administration in rats caused depression of thyroid hormone production, this was not observed in dogs. Reasons for this may include species differences in iodide uptake, thyroglobulin organification, tyrosine residue coupling, or thyroid hormone metabolism. The duration of exposure to bromide may have been inadequate to result in derangement of canine thyroid function or histology. In addition, most studies in rats utilized higher doses of bromide than were used in this study. Small sample size could have contributed to the absence of an identifiable bromide effect on the canine thyroid. However, the power of the study using five dogs per group was such that there was a 90% chance of detecting a 1.0 μg/dL difference in TT4, 0.13 ng/mL difference in TSH, and an 8 pmol/L difference in fT4 between the groups.
In this study, both control and experimental groups developed a statistically significant decrease in serum TT4 and fT4 concentrations over time. Five dogs had TT4 concentrations below the reference range, and one dog also had a depressed fT4 concentration at the end of the study. No clinical signs of hypothyroidism were identified in either group of dogs, apart from weight gain. In addition, the basal TSH concentrations and results of TRH stimulation tests at the end of the study did not support overt hypothyroidism in any of the dogs. A percentage change in post-TRH stimulation TSH of >100% was considered supportive of euthyroidism.19 One KBr-treated dog did have a decreased postTRH TT4 concentration; however, this dog had a 238% increase in TSH concentration in response to TRH stimulation. Another dog had a percent change in TSH concentration following TRH stimulation <100% on day 177. However, TT4 and fT4 were within reference ranges, consistent with euthyroidism. The TRH stimulation test is less reliable than the TSH stimulation test in the dog, because some euthyroid dogs fail to respond to TRH as illustrated in all dogs on day 0. Although none of the dogs exhibited a TSH response to TRH stimulation, all dogs were considered euthyroid based on the absence of clinical signs, normal fT4 and TSH concentrations, and appropriate TT4 response to TRH stimulation. This was in contrast to all dogs on day 177 in which TRH stimulation resulted in percent-TSH concentration changes ranging from 88% to 991%. This difference in TSH response to TRH stimulation was likely due to lower basal thyroid hormone concentrations at the end of the study. Based on previous work evaluating TSH response to TRH stimulation in healthy dogs, hypothyroid dogs, and euthyroid dogs with nonthyroidal illness, a percentage-change in postTRH stimulation TSH concentration of >100% was considered supportive of euthyroidism.19 Although TSH stimulation is a superior test of thyroid function and considered the gold standard for the diagnosis of hypothyroidism, this test was not performed due to the lack of availability of bovine TSH.
The cause of the decline in thyroid hormone concentrations over time in all dogs in this study is unknown. Because a similar decline in hormone values occurred in both TT4 concentrations performed at Purdue University and in fT4 by dialysis concentrations performed at Michigan State University, interassay variability was an unlikely cause for this finding. Although thyroid hormone concentrations have been shown to decrease with age, this occurs over a period of years as opposed to 6 months.1 All dogs had been housed in the same facility for 6 months prior to initiation of the study. During this time, all dogs were utilized in a technician course where basic venipuncture, catheterization, and restraint techniques were performed. These procedures were unlikely to have contributed to decreased basal thyroid hormone concentrations. The weight gain observed in all dogs of this study was likely the result of added food intake (meatball dosing) to their usual laboratory ration over a 6-month period.
Criteria chosen for evaluation of thyroid histopathology were based on those identified in bromide-treated rats.1112141517 Most (9/10) dogs in this study showed some degree of thyroid activation (increased MFD and decreased IFC), although these changes were patchy in their distribution. In all thyroid sections reviewed, vascularity and prevalence of columnar follicular epithelium were normal and appropriately distributed. Due to the difficulty in differentiating cells undergoing mitosis versus pyknosis, a mitotic index could not be reliably calculated. However, the overall numbers of mitotic/pyknotic cells were small and considered within normal limits for either process. Since thyroid biopsies were only collected at the end of the study and there were no differences identified between the groups, it is unlikely that bromide treatment was the cause of the thyroid activation identified.
Thyroid activation is usually caused by increased serum TSH concentrations in response to decreased serum thyronine.3435 Despite changes consistent with mild thyroid activation in some dogs of this study, an increase in basal serum TSH concentration was not identified. This discrepancy may simply reflect the heterogeneous nature of thyroid histology. Measurement of iodide to bromide ratios in the extracted canine thyroid glands may have better allowed the authors to identify reasons for differences between the rat and dog studies.36
Anticanine thyroglobulin antibodies were measured in an attempt to exclude from the study dogs with underlying thyroiditis. Two dogs (one each from the KBr-treated and control groups) had ODs that were positive, being slightly greater than 200% of the negative control. The significance of these results was questionable since anticanine thyroglobulin antibody ODs from dogs with lymphocytic thyroiditis are usually much greater (400% to 500%) than those of the negative control.28 For this reason, these two dogs were not excluded from the study. Thyroid histopathology at the end of the study did not detect evidence of thyroiditis in any dog.3437
Conclusion
In this study, KBr administration for 6 months to young, healthy adult dogs did not have a significant effect on the function or morphology of the canine thyroid gland compared to the control group. Although there was no difference in thyroid function between the groups at any time, the fact that both control dogs and KBr-treated dogs exhibited a significant decline in TT4 and fT4 over time complicated the results of this study. It was not possible to determine the cause of the decreased thyroid hormone concentrations over the course of the study; however, because the decrease occurred in both the control group and the treatment group, the change did not appear to be related to bromide administration. Further studies are necessary to confirm that the results observed in this study are repeatable in dogs with stable thyroid function over time.
Purina 5006; Ralston Purina Company, St. Louis, MO
Immulite Total T4; Diagnostic Products Corp., Los Angeles, CA
Immulite Canine TSH; Diagnostic Products Corp., Los Angeles, CA
Free T4 by equilibrium dialysis; Nichols Institute Diagnostics, San Juan Capistrano, CA
Canine thyroglobulin autoantibody immunoassay kit; Oxford Biomedical Research, Inc., Oxford, MI
Malinckrodt Analytical Reagent; Malinckrodt Baker Inc., Paris, KY
Canine canned Science Diet maintenance; Hill’s Pet Nutrition, Inc., Topeka, KS
Gold-chloride colorimetric assay; Catachem, Inc., Bridgeport, CT
Free T4 by equilibrium dialysis; Animal Health Diagnostic Laborarory, Endocrine Diagnostic Section, Michigan State University, East Lansing, MI
Serum bromide; Veterinary Diagnostics Laboratory, University of Ohio, Columbus, OH
PROC Univariate normal, PROC MIXED, PROC N PAR1WAY; SAS Institute, Cary, NC
Acknowledgments
The authors thank Dr. Andrew Nuijens, Carrie Kniola, and Lisa Holeman for their technical assistance. Dr. Julia Hawthorne is also thanked for her guidance in thyroidectomy procedures.
). Upper and lower dotted lines indicate upper and lower limits of the anticonvulsant therapeutic range for KBr monotherapy. The box outlined by the upper and middle dotted lines indicates the target range for serum bromide concentrations.
Citation: Journal of the American Animal Hospital Association 39, 2; 10.5326/0390193
Citation: Journal of the American Animal Hospital Association 39, 2; 10.5326/0390193
Citation: Journal of the American Animal Hospital Association 39, 2; 10.5326/0390193
Scatterplot of serum bromide concentrations in each potassium bromide (KBr)-treated dog(). Upper and lower dotted lines indicate upper and lower limits of the anticonvulsant therapeutic range for KBr monotherapy. The box outlined by the upper and middle dotted lines indicates the target range for serum bromide concentrations.
Serum basal total thyroxine (TT4) concentrations in KBr-treated (- -•- -) and control (—•—) dogs. Values are graphed as mean±standard deviation. Dotted lines indicate upper and lower limits of the reference range.
Serum free thyroxine (fT4) concentrations in KBr-treated and control dogs. Values are graphed as mean±standard deviation. Dotted lines indicate upper and lower limits of the reference range.
