Seizure activity is one of the most common neurologic

Transcription

1 J Vet Intern Med 2006;20: Euthyroid Sick Syndrome in Dogs with Idiopathic Epilepsy before Treatment with Anticonvulsant Drugs Thilo von Klopmann, Irene Christine Boettcher, Annett Rotermund, Karl Rohn, and Andrea Tipold Euthyroid sick syndrome is a common finding in dogs and is attributable to nonthyroidal illness or treatment with any of a variety of drugs such as phenobarbital. In dogs with epilepsy, treatment with anticonvulsant drugs can lead to subnormal plasma thyroid hormone concentrations despite normal thyroid function. One-hundred thirteen dogs with seizure activity were retrospectively evaluated to determine the influence of idiopathic epilepsy (IE) on thyroid hormone concentrations. Blood samples were taken from 60 dogs with IE before initiation of anticonvulsant therapy. Control groups consisted of 34 dogs with IE and receiving anticonvulsants and 19 dogs with secondary epilepsy. Thyroid concentrations consistent with euthyroid sick syndrome were diagnosed in 38% of dogs with untreated IE without clinical signs of hypothyroidism or concomitant diseases. There was a significant correlation (r , P 5.01) between seizure frequency and plasma thyroid hormone concentrations: the longer the interval between 2 seizure events, the higher the serum total thyroxine concentration. There was no correlation between the degree of alteration of thyroid hormone concentrations and the time span between the most recent seizure event and blood collection, the type of the most recent seizure event, the duration of the complete seizure history, or the predominant seizure type. These results suggest that IE can be a reason for euthyroid sick syndrome in dogs. The effect of phenobarbital on plasma thyroid hormone concentrations must be investigated in future studies, as it might be less pronounced than expected. Key words: Nonthyroidal illness; Phenobarbital; Seizure frequency; Thyroid hormones. Seizure activity is one of the most common neurologic disorders in dogs and cats. 1 Up to 10% of all cases of neurologic dysfunction are examined because of seizure activity. 2 The prevalence of idiopathic epilepsy (IE) amounts to 2% of the total canine population, with higher incidences in particular breeds. 3 5 Formerly, IE was estimated to be responsible for seizures in up to 80% of the cases. 6 In more recent investigations, IE was diagnosed in only about 50% of the dogs with seizure activity because advanced imaging techniques allowed the detection of other causes. 2 The other 50% were suffering from secondary or reactive epilepsy. 7 Whereas an underlying cause for seizure activity in IE cannot be found, there are structural lesions in the central nervous system or metabolic changes associated with the other forms. 7 There is considerable evidence for the development of euthyroid sick syndrome attributable to nonthyroidal illness or as a result of concomitant medication (eg, anticonvulsant therapy). 8,9 In euthyroid sick syndrome, plasma thyroid hormone concentrations are subnormal despite normal thyroid function. Hypothyroidism can be a cause of seizure activity in dogs. 5,9 Because primary hypothyroidism is a common endocrine disorder in From the Departments of Small Animal Medicine and Surgery (von Klopmann, Boettcher, Tipold), Endocrinology (Rotermund), and Biometry, Epidemiology and Information Processing (Rohn), University of Veterinary Medicine, Hannover, Germany. Preliminary results were presented as a poster presentation at the 17 th Annual ECVN Meeting, Glasgow, Great Britain, 2004 (Bayer award for best poster presentation). Abstract published in J Vet Intern Med 2005;19:296. Reprint requests: Thilo von Klopmann, Department of Small Animal Medicine and Surgery, University of Veterinary Medicine, Bischofsholer Damm 15, Hannover, Germany; thilo.von.klopmann@tiho-hannover.de. Submitted April 29, 2005; Revised July 25, October 11, 2005; Accepted December 8, Copyright E 2006 by the American College of Veterinary Internal Medicine /06/ /$3.00/0 dogs, 10 measurement of thyroid hormone concentrations is an important tool in the diagnosis of epilepsy in dogs. Interpretation of laboratory results can be difficult because several factors influence thyroid hormone concentrations. 8 Numerous studies have investigated and described the effect of anticonvulsant drugs and other substances frequently used in veterinary practice on plasma thyroid hormone concentrations. 8,11 13 For example, phenobarbital treatment decreases total thyroxine (TT 4 ) concentration and can increase canine thyroidstimulating hormone (ctsh) concentration. 8,12 14 In 13% of dogs with seizure activity, plasma TT 4 concentration was lower within the 24 hours preceding blood collection than was TT 4 concentration measured at a second time point not associated with seizure activity. 12 Therefore, it is postulated that epilepsy per se is not responsible for lowered serum TT 4 concentration in dogs, albeit seizure activity might have an influence. 12 To our knowledge, no other published data indicate that IE itself has an influence on thyroid hormone concentrations. The aim of the study reported here was to retrospectively evaluate thyroid hormone concentrations in 113 epileptic dogs to identify an influence of IE itself on thyroid hormone concentrations before initiation of any anticonvulsant treatment. Material and Methods Inclusion Criteria Between July 2003 and March 2005, 320 dogs with seizure activity were presented to the Department of Small Animal Medicine and Surgery, University of Veterinary Medicine Hannover. Dogs of various breeds and ages, either sex, and with various seizure types were available for this study. Dogs had to meet the following inclusion criteria: measurement of plasma thyroid hormone concentrations (TT 4, ctsh); sufficient information concerning seizure history and pretreatment (anticonvulsant medication before blood sampling for thyroid hormone measurements); and sufficient diagnostic testing for a diagnosis of idiopathic or secondary epilepsy. Of the 320 dogs with seizure activity, 113 met these criteria and were included in the study.

2 Euthyroid Sick Syndrome in Dogs 517 Diagnostic Evaluation In all dogs, diagnostic testing consisted of physical and neurologic examinations, CBC, and biochemical analysis including plasma TT 4 and ctsh concentrations. All blood samples were taken before anesthesia. Advanced imaging techniques were performed in 76 dogs: computed tomography (CT) a in 6 cases and magnetic resonance imaging (MRI) b in 70 cases. Cerebrospinal fluid (CSF) analysis was performed in 75 of the 76 dogs. The TT 4 and ctsh plasma concentrations were measured using a commercial chemiluminescence assay. c In 27 dogs, a thyrotropin-releasing hormone stimulation test (TRH stimulation test) was performed in addition to baseline TT 4 and ctsh concentration measurements. In the group of untreated dogs, this test was performed before initiation of any medication or anesthesia. The protocol for the TRH stimulation test included measurement of plasma TT 4 and ctsh concentrations before (baseline) and after intravenous administration of protirelin d at a dosage of 10 mg/kg of body weight. Blood sampling occurred at 20 minutes (ctsh) and 2 and 3 hours (TT 4 ) after administration of protirelin. 15 Plasma cholesterol concentration was available for 93 dogs. Groups According to Diagnosis The 113 dogs of this retrospective study were assigned to 5 groups: IE was diagnosed in dogs in which clinical and neurologic examination, blood and CSF analytic, and CT or MRI findings were within normal limits. Dogs were assigned to group A if there was no prior anticonvulsant treatment before blood sampling or to group C if such medication had been given. Dogs for which CSF analysis and advanced imaging techniques could not be performed were also included in the IE group if clinical and neurologic examination and blood work results were normal. These dogs were defined as presumptive IE and were assigned to group B if untreated or to group D if treated previously. Dogs with an underlying disease responsible for seizure activity were categorized as secondary epileptic, regardless of whether seizures were caused by structural brain lesions or metabolic disorders (group E). Seizure History Evaluation Seizure history was reviewed concerning the type and duration of the most recent seizure event and the time span between the most recent seizure event and blood collection for thyroid hormone concentration measurements in dogs suffering from IE without pretreatment (groups A and B). In addition, the time span between the 1st seizure and blood collection, the type of seizure most frequently occurring in this time span, and the frequency with which the seizures occurred were evaluated in dogs of groups A and B. Seizures were classified as focal, complex-focal, generalized, cluster, or status epilepticus. 7,16 The occurrence of more than one seizure event within 24 hours was classified as a cluster. Status epilepticus was defined as the occurrence of several subsequent seizures without complete regain of consciousness. 7 Groups According to Thyroid Hormone Concentrations and Time of Blood Collection According to thyroid hormone plasma concentration measurement, the dogs of these 5 groups were classified as follows: euthyroid sick, hypothyroid, or normal. The diagnosis of euthyroid sick status was based on the presence of the following criteria: TT 4 concentration below the reference interval, ctsh concentration within the reference interval, lack of clinical signs of hypothyroidism such as alopecia, skin lesions, or lethargy, and plasma cholesterol concentration within the reference interval. Furthermore, dogs of groups A and B were subsequently allocated to 2 subgroups according to the time point of blood collection: sampling within a 24-hour period following the most recent seizure event (A,24,B,24 ), or after a longer period (A.24,B.24 ). Reference Intervals Reference intervals for thyroid hormone concentration measurements were obtained from a previous study. 15 Normal concentration of TT 4 ranged from 1.4 to 4.5 mg/dl; TT 4 # 1.3 mg/dl was considered decreased. The upper limit for normal ctsh concentration was 0.45 ng/ml. 15 Concerning the TRH stimulation test, adequate stimulation was considered as a significant increase in ctsh concentration and a TT 4 increase of at least 0.5 mg/dl above baseline values. 9 The reference interval for cholesterol was mg/dl. Statistical Analysis Statistical evaluation was carried out using the SAS statistical software. e Goodness of fit for normal distribution of thyroid hormone parameters was assessed by visual assessment of normal probability plots and the Kolmogorov-Smirnov test. Because the data, which are from a ratio scale, were neither normally nor lognormally distributed, nonparametric methods were used (ie, tests would be performed without making any strict distributional assumptions). For the statistical models, the Kruskal-Wallis test and the Wilcoxon two-sample test were used. Analyses of associations were performed by calculating Spearman correlation coefficients. Significance was defined as P,.05. Results Patient Characteristics Within the observation period, 113 dogs presented to the Department of Small Animal Medicine and Surgery with seizure activity met the inclusion criteria for this study. Idiopathic epilepsy was found in 94 dogs (83%; groups A D), and secondary epilepsy was found in 19 animals (group E; 17%). Because of our inclusion criteria concerning measurement of plasma thyroid concentrations, these percentages do not reflect the whole population suffering from seizure activity. Of the 94 dogs with IE, 42 (37%) belonged to group A (IE), 18 (16%) to group B (presumptive IE), 24 (21%) to group C (IE with pretreatment), and 10 (9%) to group D (presumptive IE with pretreatment). Group E included 4 dogs with brain tumors, 4 with encephalitides, 2 with brain malformations, and 9 with metabolic diseases. Three of the 9 dogs with metabolic diseases suffered from hypothyroidism. Eighty-nine (79%) of all 113 animals were purebred dogs of 41 different breeds, predominantly Golden Retriever (n 5 6; 5%), German Shepherd Dog (n 5 6; 5%), Labrador Retriever (n 5 6; 5%), Beagle (n 5 5; 4%), Small Munsterlander (n 5 5; 4%), Schnauzer (n 5 5; 4%), Boxer (n 5 4; 4%), Cocker Spaniel (n 5 4; 4%), Husky (n 5 4; 4%), Border Collie (n 5 3; 3%), Dachshund (n 5 3; 3%), Irish Setter (n 5 3; 3%), and Jack Russell Terrier (n 5 3; 3%). The other 24 of 113 patients (21%) were dogs of mixed breeding. The study population consisted of 55 (49%) reproductively intact male, 19 (17%) neutered male, 26 (23%) reproductively intact female, and 13 (11%) spayed female dogs. The age range of all dogs was years (median, 4.0 years) at the time of blood collection. The median age was lowest in group-a (2.9 years) and

3 518 von Klopmann et al Table 1. Type and duration of the most recent seizure event prior to blood collection. All Groups Group A/B Group C/D Group E Seizure Type n n 5 60 n 5 34 n 5 19 Focal 2 (1.8%) 0 (0%) 0 (0%) 2 (11%) Complex-focal 6 (5.3%) 4 (7%) 1 (3%) 1 (5%) Generalized 59 (52.2%) 37 (62%) 18 (53%) 4 (21%) Cluster 44 (38.9%) 18 (30%) 15 (44%) 11 (58%) Status epilepticus 2 (1.8%) 1 (1%) 0 (0%) 1 (5%) Seizure duration (median, min) n 5 95 n 5 53 n 5 28 n 5 14 n, No. of dogs; group A/B, idiopathic epilepsy (IE) without anticonvulsant treatment; group C/D, IE with anticonvulsant treatment; group E, secondary epilepsy. significantly higher (P 5.003) in group-e (7.6 years) dogs. Seizure Characteristics Most of the study dogs manifested generalized seizures (93%) in the form of a single generalized seizure, cluster, or status epilepticus (Table 1). Duration of the most recent seizure was assessed in 95 dogs as well as possible by the owners (Table 1). The seizure durations in dogs of the pretreated groups were shorter than those in dogs of the untreated groups, but differences were without statistical significance (P 5.35; Table 1). Additionally, in 106 dogs, the time span was known between the last seizure event and blood collection (Table 2). Plasma Cholesterol Concentration The median plasma cholesterol concentration was 235 mg/dl (range, mg/dl, n 5 93). Concentration was lowest in dogs of group A/B (n 5 55). The cholesterol concentration ranged from 193 to 457 mg/dl in group-e dogs. Values were above the reference interval in 1 dog of group A (491 mg/dl) and in 2 dogs of group E (416, 427 mg/dl). The thyroid hormone concentrations (TT 4 of 2.9 mg/dl, ctsh of 0.03 ng/ml) were normal for the dog of group A which did not manifest clinical signs of hypothyroidism. One dog of group E had a cholesterol concentration of 427 mg/dl, a low TT 4 (0.27 mg/dl) concentration, and a very high ctsh concentration (9.7 ng/ml), and was diagnosed as hypothyroid. In the other dog of group E, an Arnold Chiari-like malformation was diagnosed and thyroid hormone concentrations were within the reference interval (TT 4 of 2.8 mg/dl, ctsh of 0.12 ng/ml). Plasma TT 4 and TSH concentrations and TRH Stimulation Test Results Plasma TT 4 concentrations in the dogs of the various groups are illustrated in Figure 1. Statistically significant differences (P 5.78) were not detected in TT 4 concentration between the groups A.24, B.24 and A,24,B,24. The ctsh concentration in dogs of group A/B and group A.24,B.24 ranged from 0.03 to 0.45 ng/ ml, and from 0.03 to 0.29 mg/dl in dogs of group A,24, B,24. Plasma TT 4 concentration in the medically treated, IE-affected dogs (group C/D) was significantly lower than that in dogs of group A/B (P,.001; Fig 1). The ctsh concentration in dogs of group C/D ranged from 0.03 to 0.32 ng/ml. Dogs with secondary epilepsy (group E) had significantly lower TT 4 concentration than that in dogs of group A/B (P 5.003; Fig 1). In dogs of this group, ctsh concentration ranged from 0.03 to 9.7 ng/ml. Three dogs of this group were diagnosed as being hypothyroid. The ctsh concentration did not differ significantly among the various groups (P..05). Dogs of groups B and D, without evaluation of advanced imaging techniques, did not have significant differences in TT 4 (P ) and ctsh (P 5.45) concentrations, compared with those that underwent CT or MRI procedures (groups A and C). In 27 cases, the Table 2. Interval between the most recent seizure event and blood sampling. All Groups Group A/B Group C/D Group E All Groups. 24h Group A.24, B.24 All Groups, 24h Group A,24, B,24 Interval n n 5 58 n 5 29 n 5 19 n 5 71 n 5 44 n 5 35 n 5 16 Range 0 600d 0 150d 0 21d 0 600d d d 0 24h 0 24h Median 4.0d 4.5d 5.0d 1d 9.0d 9.0d 6.0h 12.0h n, No. of dogs; d, days; h, hours; group A/B, idiopathic epilepsy (IE) without anticonvulsant treatment; group C/D, IE with anticonvulsant treatment; group E, secondary epilepsy; all groups. 24 h, all dogs with no seizure event within the last 24 hours preceding blood collection; group A.24,B.24, untreated IE dogs with no seizure event within the last 24 hours preceding blood collection; All groups, 24 h, all dogs with seizure event within the last 24 hours preceding blood collection; Group A,24,B,24, untreated IE dogs with seizure event within the last 24 hours preceding blood collection.

4 Euthyroid Sick Syndrome in Dogs 519 Fig 1. Plasma thyroxine (T 4 ) concentrations in the various groups of dogs. Group A/B, idiopathic epilepsy (IE) without anticonvulsant treatment; group A.24,B.24, untreated dogs with IE and no seizure event within the past 24 hours preceding blood collection; group A,24,B,24, untreated dogs with IE and seizure event within the past 24 hours preceding blood collection; group C/ D, IE with anticonvulsant treatment; group E, secondary epilepsy; ESS, euthyroid sick syndrome (box plots with gray filling); Black bars, median plasma T 4 concentrations. presumptive diagnosis of euthyroid sick syndrome or normal thyroid status was confirmed by results of the TRH stimulation test. Baseline values indicated a median TT 4 concentration of 1.0 mg/dl and a median ctsh concentration of 0.08 ng/ml. After administration of protirelin, an appropriate enhancement of ctsh and TT 4 concentrations was measured. The median ctsh concentration increased to 0.46 ng/ml, the median TT4 concentration at 2 and 3 hours after administration of TRH increased to 1.8 and 1.7 mg/dl, respectively. Results of thyroid hormone concentration measurements were consistent with euthyroid sick syndrome in 50.4% of the study population (Table 3). In these dogs, cholesterol concentration ranged from 126 to 362 mg/ dl. Hypothyroidism was diagnosed in 2.7% of all dogs, all of which were from group E. In the dogs with IE and without pretreatment (group A/B), thyroid hormone concentrations consistent with euthyroid sick syndrome were detected in 38% of the cases (Table 3). In these dogs, the median TT 4 concentration was 1.1 mg/dl (Fig 1), and the median ctsh concentration was 0.07 ng/ml. In 13 of these dogs, a TRH stimulation test was performed to exclude hypothyroidism (Fig 2). After administration of protirelin, an appropriate increase of ctsh and TT 4 concentrations was obvious in all 13 dogs (Fig 2). The percentages of findings consistent with euthyroid sick status were higher in dogs diagnosed with IE and pretreatment (group C/D) and in dogs with secondary epilepsy (group E; Table 3). Seventeen (74%) of the 23 dogs with presumptive euthyroid sick syndrome in group A/B had no seizure activity within the last 24 hours preceding blood collection (Group A.24,B.24 ), whereas only 6 (26%) dogs had seizures during the last 24 hours (Group A,24, B,24 ). In the group of untreated dogs with IE (group A/B), there was only a very weak correlation without significance between TT 4 and ctsh concentrations and the time from the most recent seizure event to blood collection, the duration of this seizure event, the time from the onset of the 1st seizure to blood collection, and the predominant seizure type. There was no association in dogs of group A/B between TT 4 concentration and the type of the most recent seizure event before blood collection (P 5.24). However, the seizure frequency was negatively correlated to the TT 4 concentration (r 5.363, P 5.01) in this group (Fig 3). Plasma TT 4 concentration was lower in dogs with a short seizure-free interval than in those with seizures occurring at greater intervals. Discussion Results of this study indicate that thyroid values consistent with euthyroid sick syndrome are seen in dogs with IE but without anticonvulsant therapy and other concomitant diseases. Changes in thyroid hormone concentrations in epileptic dogs before initiation of anticonvulsant treatment were investigated in a previous study. 12 Plasma T 4 concentration was lower in dogs with seizure activity during the last 24 hours before blood collection for thyroid hormone concentration measurement. 12 However, there was no significant alteration in thyroid hormone concentrations in dogs in which the last seizure occurred more than 24 hours before blood collection. 12 It is postulated that epilepsy per se is not responsible for lowered serum TT 4 concentration in Table 3. Classification of dogs by results of combined thyroid hormone concentration measurement. All Groups Group A/B Group C/D Group E n n 5 60 n 5 34 n 5 19 Euthyroid sick 57 (50.4%) 23 (38%) 23 (68%) 11 (58%) (TT 4 : # 0.3 mg/dl, ctsh:, 0.45 ng/ml) A(n518), B(n55) C(n518), D(n55) Hypothyroid 3 (2.7%) 0 (0%) 0 (0%) 3 (16%) (TT 4 :, 1.1 mg/dl, ctsh: ng/ml) Normal 53 (46.9%) 37 (62%) 11 (32%) 5 (26%) (TT 4 : mg/dl, ctsh:, 0.45 ng/ml) n, No. of dogs; group A/B, idiopathic epilepsy (IE) without anticonvulsant treatment; group C/D, IE with anticonvulsant treatment; group E, secondary epilepsy; TT 4, total thyroxine; ctsh, canine thyroid-stimulating hormone.

5 520 von Klopmann et al Fig 2. Results of thyrotropin-releasing hormone (TRH) stimulation testing in 13 dogs with suspicious euthyroid sick syndrome of group A/B (IE without anticonvulsant treatment). (A) Plasma total T 4 (TT 4 ) concentrations: baseline, baseline TT 4 concentrations before protirelin administration; 2 h, TT 4 concentrations 2 hours after stimulation with protirelin; 3 h, TT 4 concentrations at 3 hours after stimulation with protirelin; black bars, median plasma T 4 concentrations. (B) Concentrations of canine TSH (ctsh): baseline, baseline ctsh concentrations before protirelin administration; ctsh concentrations at 20 minutes after stimulation with protirelin; black bars, median ctsh concentrations. dogs. 12 In contrast to that study, we found that thyroid findings consistent with or suspicious for euthyroid sick syndrome are present in dogs with untreated IE. Findings consistent with euthyroid sick status also were found in dogs that had the most recent seizure event more than 24 hours before blood collection, indicating that a short interval between the most recent seizure event and time of blood collection for thyroid hormone concentration measurement has no major influence on thyroid hormone concentrations. This statement is Fig 3. Correlation between seizure-free interval and plasma T 4 (TT 4 ) concentration in group A/B. Abscissa: seizure-free interval between 2 seizure events in days as a measure of seizure frequency. Ordinate: plasma T 4 concentration (TT 4 ; mg/dl), total thyroxine concentration in micrograms per deciliter. Spearman correlation coefficient indicates a significant correlation between TT 4 concentration and the seizure-free interval (r 5.363, P 5.01). supported by the statistical analytic results of this study, which indicated no significant difference in the TT 4 concentration dependent on the time span (more or less than 24 hours) between the most recent seizure event and blood sampling. In dogs with no seizure activity during the last 24 hours, ctsh concentration was slightly higher, indicating a certain degree of feedback from low TT 4 concentration to increased release of ctsh by the pituitary gland. This effect was not seen in dogs in which the most recent seizure event occurred within the last 24 hours preceding blood collection. According to the study reported here, it could be postulated that seizure activity attributable to IE may be sufficient to induce euthyroid sick syndrome. The significant negative correlation between seizure frequency and the alteration of TT 4 concentration in dogs of group A/B supports this statement; the longer the interval between 2 seizure events (seizure-free interval), the higher the TT4 concentrations, indicating that subsequent and frequently occurring seizure events are able to induce euthyroid sick syndrome. The ctsh concentration was less affected and had no significant changes or correlations throughout the entire study population, except in dogs with hypothyroidism. However, follow-up examination of ctsh concentration was not performed to investigate whether serial ctsh concentration trends in the same animal might indicate development of thyroid disease, because several studies indicated that ctsh concentration might be influenced by anticonvulsants. 8,11 With the exception of seizure frequency, there was no evidence of an influence on thyroid hormone concentrations by any other seizure parameter such as the type and the duration of the last seizure event, duration of the entire seizure history, or the predominant seizure type. Therefore, it could be postulated that recovery from euthyroid sick status attributable to IE is more closely

6 Euthyroid Sick Syndrome in Dogs 521 related to the time span without seizures than to the severity of seizure activity or all the other aforementioned parameters. Considering that cholesterol values are increased in up to 75% of dogs with hypothyroidism, we used this variable additionally to distinguish between hypothyroidism and euthyroid sick syndrome, which was only suspected when all parameters clinical examination, plasma cholesterol, TT 4, and ctsh concentrations, and results of TRH stimulation test in some of the dogs were normal. Subclinical or developing hypothyroidism could not completely be excluded in these animals of this study. Euthyroid sick syndrome attributable to nonthyroidal illness or concomitant medication for several diseases is well known in humans and dogs. 9,17 Drug-induced mechanisms responsible for alterations of thyroid hormone concentrations may include increased metabolism by the liver because of liver enzyme induction (eg, phenobarbital) and displacement of thyroid hormones from their protein binding sites (eg, acetylsalicylic acid). 8,11 Furthermore, suppressed ctsh secretion attributable to glucocorticoid administration and altered sensitivity of the thyroid gland to ctsh result in lowered thyroid hormone concentrations in humans. 8 Various studies have indicated similar changes in thyroid hormone values in dogs, although results are controversial. There was no significant effect of phenobarbital on thyroid hormone concentrations during a 3-week treatment period, although others observed significantly lower TT 4 and significantly higher ctsh concentrations within this period. 13,14 Effects on the metabolism of thyroxine secondary to phenobarbital in other species also have been documented Nonthyroidal illnesses leading to altered thyroid hormone concentrations include cardiac failure, kidney and liver insufficiency, immune-mediated disorders, sepsis, hyperadrenocorticism, hypoadrenocorticism, and diabetes mellitus. 9,17 The severity of illness is directly correlated to the change in thyroid hormone concentrations. 9 This phenomenon might also be associated with IE, since seizure frequency as an indicator for the severity of illness seems to be the most important factor in the induction of presumptive euthyroid sick syndrome in dogs with IE. Several methods are available for the measurement of thyroid hormone concentrations. However, numerous studies have indicated that measurement of serum TT 4 concentration alone has high sensitivity, specificity, and accuracy, and was determined in a variety of euthyroid sick conditions as well as in confirmed hypothyroid and euthyroid patients. 21,22 Evaluation of ctsh concentration as a single diagnostic tool for canine hypothyroidism has sensitivity of 82% and specificity of 100%. 21 However, in some cases of otherwise confirmed hypothyroidism, ctsh concentration was within the reference interval. 21,22 In the study reported here, we measured plasma TT 4 and ctsh concentrations in combination and were able to distinguish between euthyroid sick, hypothyroid, and clinically normal dogs. Our data were further supported by the fact that there were no clinical signs of hypothyroidism such as alopecia, skin lesions, or lethargy, and that cholesterol values were within normal limits. In addition, in 27 study dogs, results of the TRH stimulation test including TT 4 as well as ctsh stimulation patterns confirmed the diagnosis made by the combined measurement of TT 4 and ctsh. In these cases, the TRH stimulation test seemed to have no crucial advantages over the described testing procedures. 23 Additional variables such as free T 4, or plasma triiodothyronine concentration, or results of thyroglobulin autoantibody testing were not available for the present retrospective study to further support our data. According to our study, adequate diagnostic testing of epileptic dogs is necessary to distinguish between euthyroid sick status and hypothyroidism. Combined measurement of total T 4 and canine TSH concentrations is a timesaving and reliable method and should be performed before anesthesia and anticonvulsant treatment, regardless of the time elapsed since the most recent seizure event. To exclude subclinical or developing hypothyroidism, additional testing procedures such as the TRH stimulation test, measurement of FT 4 and plasma triiodothyronine concentrations, and thyroglobulin autoantibody testing should be considered in future studies. Because, in a substantial subset of patients, IE alone is associated with euthyroid sick syndrome, a combined effect of anticonvulsants and IE on thyroid hormone concentrations must be considered. Therefore, in dogs with seizures, future prospective studies comprising sequential data must be conducted to differentiate between the effects on thyroid hormone concentrations purely induced by IE and those in response to anticonvulsant drugs such as phenobarbital, as the latter may be less pronounced than expected. Footnotes a Somatom spiral AT, Siemens, Erlangen, Germany b Magnetom impact plus 1.0 Tesla, Siemens, Erlangen, Germany c DPC Biermann, Los Angeles, CA d TRH Ferring, Ferring GmbH, Kiel, Germany e SAS, version 9.1, SAS Institute, Cary, NC References 1. Thomas WB. Seizures and narcolepsy. In: Dewey CW, ed. A Practical Guide to Canine and Feline Neurology. Ames, IA: Iowa State University Press; 2003: Jaggy A, Bernardini M. Idiopathic epilepsy in 125 dogs: A long-term study. Clinical and electroencephalographic findings. J Small Anim Pract 1998;39(1): Berendt M, Gredal H, Pederson LG, et al. A cross-sectional study of epilepsy in Danish Labrador Retrievers: Prevalence and selected risk factors. J Vet Intern Med 2002;16: Cunningham JG, Farnbach GC. Inheritance and idiopathic canine epilepsy. J Am Anim Hosp Assoc 1988;24: Gandini G, Jaggy A, Kathmann I. Großhirn. In: Jaggy A, ed. Atlas und Lehrbuch der Kleintierneurologie. Hannover, Germany: Schlütersche Verlagsgesellschaft 2005: Schwartz-Porsche D. Seizures. In: Braund KG, ed. Clinical Syndromes in Veterinary Neurology, 2 nd ed. St. Louis, MO: Mosby Yearbook, 1994:

7 522 von Klopmann et al 7. Podell M. Seizure classification in dogs from a nonreferralbased population. J Am Vet Med Assoc 1995;206: Daminet S, Ferguson DC. Influence of drugs on thyroid function in dogs. J Vet Intern Med 2003;17: Feldman EC, Nelson RW. Hypothyroidism. In: Feldman EC, Nelson RW, eds. Canine and Feline Endocrinology and Reproduction, 3 rd ed. Philadelphia, PA: WB Saunders; 2004: Kooistra HS, Diaz-Espineira M, Mol JA, et al. Secretion pattern of thyroid-stimulating hormone in dogs during euthyroidism and hypothyroidism. Domest Anim Endocrinol 2000;18(1): Daminet S, Croubels S, Duchateau L, et al. Influence of acetylsalicylic acid and ketoprofen on canine thyroid function tests. Vet J 2003;166: Gaskill CL, Burton SA, Gelens HCJ, et al. Effects of phenobarbital treatment on serum thyroxine and thyroid-stimulating hormone concentrations in epileptic dogs. J Am Vet Med Assoc 1999;215(4): Gaskill CL, Burton SA, Gelens HCJ. Changes in serum thyroxine and thyroid-stimulating hormone concentrations in epileptic dogs receiving phenobarbital for one year. J Vet Pharmacol Ther 2000;23: Daminet S, Paradis M, Refsal KR, Price C. Short term influence of prednisone and phenobarbital on thyroid function in euthyroid dogs. Can Vet J 1999;40: Hoppen HO, Lohmann P, Schlote S, et al. Die Messung von caninem TSH zur Diagnostik der Hypothyreose des Hundes. Der praktische Tierarzt 1997;78: Berendt M, Gram L. Epilepsy and seizure classification in 63 dogs: A reappraisal of veterinary epilepsy terminology. J Vet Intern Med 1999;13: Scott-Moncrieff JCR, Guptill-Yoran L. Hypothyroidism. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine, 6 th ed. St. Louis, MO: WB Saunders 2005: Yuksel A, Kartal A, Ceani A, Yalcin E. Serum thyroid and pituitary response to thyrotropin-releasing hormone in epileptic children receiving anti-epileptic medication. Acta Paediatr Jpn 1993;35: Hood A, Klaassen CD. Differential effects of microsomal enzyme inducers on in vitro thyroxine (T(4)) and triiodothyronine (T(3)) glucuronidation. Toxicol Sci 2000;55: Deda G, Akinci A, Tezic T, Karagol U. Effects of anticonvulsant drugs on thyroid hormones in epileptic children. Turk J Pediatr 1992;34: Dixon RM, Graham PA, Mooney CT. Serum thyrotropin concentrations: A new diagnostic test for canine hypothyroidism. Vet Rec June 1996;15: Peterson ME, Melian C, Nichols R. Measurement of serum total thyroxine, triiodothyronine, free thyroxine, and thyrotropin concentrations for diagnosis of hypothyroidism in dogs. J Am Vet Med Assoc 1997;211: Scott-Moncrief JCR, Nelson RW. Change in serum thyroidstimulating hormone concentration in response to administration of thyroid-releasing hormone to healthy dogs, hypothyroid dogs, and euthyroid dogs with concurrent disease. J Am Vet Med Assoc 1998;213: