Ultrasonography is a useful diagnostic tool for examination of the thyroid gland in human and veterinary medicine. The superficial location of the thyroid gland allows ultrasonographic examination with high-frequency transducers. Such transducers provide high-resolution images that can be used to evaluate size, shape, echogenicity, and homogeneity of the thyroid gland.1 In humans, ultrasonographic evaluation of the thyroid gland is used to complement thyroid function testing during the initial evaluation for suspected thyroid gland disease.2–4 Ultrasonography is used in humans to assess thyroid gland size and echogenicity in patients with suspected autoimmune thyroiditis,4–6 evaluate palpable thyroid nodules and cystic lesions,1 identify nonpalpable thyroid tumors,7,8 guide core-needle biopsy procedures and fine-needle aspiration of thyroid nodules,9 manage maternal hyperthyroidism during pregnancy,10 and identify thyroid gland dysgenesis in children with primary congenital hypothyroidism.11
In dogs, ultrasonography is used to evaluate the origin, location, and margination of suspected thyroid tumors and to guide fine-needle aspiration and core-needle biopsy procedures for cytologic and histologic analysis of thyroid gland masses.12 Recently, ultrasonography has been used to assess thyroid glands in Golden Retrievers suspected to have hypothyroidism.13,a In dogs with hypothyroidism, thyroid gland size and volume were decreased and changes in echogenicity of the thyroid gland were identified, compared with findings in healthy dogs and euthyroid dogs with nonthyroidal illness. In that study,13 only Golden Retrievers were evaluated to negate breed variability in thyroid gland size as a potential confounding factor in the analyses of data from healthy, hypothyroid, and euthyroid dogs with nonthyroidal illness.
Studies in humans have identified positive correlations between thyroid gland size or volume and body weight,14–17 lean body mass,15 and age.14 To our knowledge, similar information has not been reported in dogs; however, differences in ultrasonographic characteristics (most notably thyroid gland size and volume) among breeds of dogs may exist. The objectives of this study were to determine ultrasonographic characteristics of the thyroid gland in healthy small-, medium-, and large-breed dogs and evaluate the relationships of thyroid gland size and volume with body weight and BSA. The hypotheses of this study were that there are differences in thyroid gland size and volume among breeds of dogs and that these differences are associated with the size of the dog. For purposes of this study, Miniature and Toy Poodles were included as small-breed dogs, Beagles were included as medium-breed dogs, and Akitas and Golden Retrievers were included as large-breed dogs. Recently reported ultrasonographic characteristics of the thyroid gland in healthy Golden Retrievers13,a were included in the study to allow comparison of data from 2 large breeds (Golden Retrievers and Akitas).
Materials and Methods
Dogs—Seventy-two adult healthy dogs were enrolled in the study from November 1999 to November 2002. Breeds included Golden Retrievers (n = 36), Akitas (12), Beagles (12), and Miniature (6) and Toy (6) Poodles. The Miniature and Toy Poodles were combined into 1 group. With the exception of 5 colony-raised Beagles, all dogs were owned by students, staff members, or clients of the Veterinary Medical Teaching Hospital. Informed consent was obtained from all dog owners. Mean age of the Akitas was 6.0 years (range, 1.3 to 13.5 years); there were 9 sexually intact or spayed females and 3 sexually intact or castrated males. Mean weight of the Akitas was 37.1 kg (range, 28.8 to 44.0 kg), and mean BSA was 1.12 m2 (range, 0.95 to 1.26 m2). Mean age of the Golden Retrievers was 4.0 years (range, 1.0 to 9.3 years); there were 16 sexually intact or spayed females and 20 sexually intact or castrated males. Mean weight of the Golden Retrievers was 31.3 kg (range, 22.4 to 46.2 kg), and mean BSA was 1.00 m2 (range, 0.79 to 1.30 m2). Mean age of the Beagles was 4.7 years (range, 2.5 to 12.0 years); there were 5 sexually intact or spayed females and 7 sexually intact or castrated males. Mean weight of the Beagles was 12.6 kg (range, 8.2 to 16.4 kg), and mean BSA was 0.54 m2 (range, 0.41 to 0.65 m2). Mean age of the Miniature and Toy Poodles was 6.3 years (range, 1.0 to 11.2 years); there were 9 sexually intact or spayed females and 3 sexually intact or castrated males. Mean weight of the Miniature and Toy Poodles was 4.3 kg (range, 2.5 to 6.4 kg), and mean BSA was 0.26 m2 (range, 0.19 to 0.35 m2). Body condition score was subjectively assessed for all dogs by use of a scale of 1 to 9 (score of 5 representing ideal condition). Mean (range) BCS was 5.0 (5 to 5), 5.5 (4 to 7), 5.6 (5 to 8), and 5.1 (5 to 6) for the Akitas, Golden Retrievers, Beagles, and Miniature and Toy Poodles, respectively. Weight and BSA were significantly (P < 0.01) different among all 4 breeds (Akitas, Golden Retrievers, Beagles, and Poodles). There were no significant differences in age, sex distribution, and BCS among breeds.
Assessment of health status in each dog was based on history; physical examination findings; and results of CBC, serum biochemical analyses, and tests of thyroid gland function (serum T4, fT4, and endogenous cTSH concentrations and presence of serum TGAA).
Ultrasonographic examination of the thyroid gland—Each dog was positioned in dorsal recumbency for ultrasonographic examination of the thyroid gland; the examination was performed without sedation or shaving of hair by use of a 10- to 5-MHz multifrequency linear-array transducer that was connected to a commercially available ultrasound machine.b Alcohol and acoustic gel served as the coupling agents. Examinations were performed by 1 of 2 board-certified radiologists (REP or VFS). The LL and RL of the thyroid gland were identified via a ventral approach; the larynx (cranial), trachea (medial), and common carotid artery (lateral) were used as landmarks. Longitudinal and transverse images of each thyroid lobe were obtained. Longitudinal images were produced by maximizing thyroid lobe length. Transverse images were obtained by rotating the transducer 90° at the widest portion of each thyroid lobe. Longitudinal and transverse images were recorded on radiographic film.
The size, shape, echogenicity, and homogeneity of each thyroid lobe were evaluated. On the basis of ultrasonographic appearance, the shape of each thyroid lobe was subjectively described as fusiform or elliptical (longitudinal images) and triangular or round to oval (transverse images). The echogenicity of each thyroid lobe was subjectively assessed as hypoechoic, isoechoic, or hyperechoic, compared with the surrounding musculature. The thyroid parenchyma was subjectively classified as homogeneous or heterogeneous on the basis of the presence or absence of a uniform distribution of echoes. Focal lesions within the thyroid parenchyma also were described. To determine thyroid lobe size, calipers were positioned at the capsular interface between the thyroid parenchyma and surrounding musculature and the distance in millimeters was recorded. On longitudinal images, the length of each thyroid lobe was defined as its maximal longitudinal axis and the height was defined as the maximal dimension perpendicular to the maximal longitudinal axis. On transverse images, the height of each thyroid lobe was defined as its maximal diameter along the axis of the ultrasound beam and the width was defined as the maximal dimension perpendicular to the height. All thyroid lobe measurements were performed on 3 separately obtained images (in each imaging plane), and mean values of the 3 measurements of thyroid lobe length, width, and height were used for determination of thyroid lobe size and volume. The volume of each thyroid lobe was estimated by use of the equation for an ellipsoid: π/6 (length × height × width).18–21 Thyroid lobe height was available from both longitudinal and transverse images. The height from transverse images was used to estimate thyroid lobe volume. The total thyroid gland volume was calculated by the summation of volumes of the left and right thyroid lobes.
Analytical methods—Food was withheld from each dog 12 hours or longer before collection of blood for a CBC; serum biochemical analyses; measurement of serum T4, fT4, and cTSH concentrations; and detection of serum TGAA. Serum was obtained and stored at -20°C until T4, fT4, and cTSH concentrations and the presence of TGAA could be determined. Clinicopathologic values in the serum biochemical profile were determined by use of a chemistry analyzer.c Serum T4 concentration was determined via an enzyme immunoassayd that had been validated previously for use with canine serum.22 Serum fT4 concentration was determined via a modified equilibrium dialysis techniquee that had been validated previously for use with canine serum.23 Serum cTSH concentration was determined via an immuno-radiometric assayf that had been validated previously for use with canine serum.24,25 Presence of serum TGAA was assessed via an indirect ELISAg that had been validated previously for use with canine serum.26 Complete blood counts were obtained by use of an automated cell counter.h
Analysis of data—Statistical analyses were performed by use of statistical software.i Analyses of mean thyroid lobe length, width, height, and volume; thyroid gland volume; and body weight, BSA, BCS, and age of the Akitas, Golden Retrievers, Beagles, and Miniature and Toy Poodles were performed by use of a 1-way ANOVA. If significant differences were identified, pairwise comparisons were performed by use of Tukey multiple comparison adjustments. Differences in the variables between Miniature and Toy Poodles were assessed by use of a t test. Wilcoxon signed rank tests were used for comparisons of size and volume of the left and right thyroid lobes in the same dog. Differences in total thyroid gland volume depending on sex and reproductive (neuter) status were evaluated with Mann-Whitney U and Kruskal-Wallis tests (on the basis of median values). Differences between categorical data were assessed by use of χ2 tests of homogeneity. Correlations were determined by use of Pearson correlations. Simple and multiple linear regression analyses were performed to predict thyroid lobe size and thyroid gland volume from body weight, BSA, and age. A value of P < 0.05 was considered significant. All data are presented as mean ± SD unless otherwise indicated.
Results
On the basis of history and results of physical examination, clinicopathologic analyses, and tests of thyroid gland function, all dogs were considered healthy. None of the dogs had a history of illness for at least 6 months prior to enrollment in the study, and the only treatments being administered at the time of the study were heartworm and flea preventives. Abnormalities identified on physical examination included mild obesity (4 dogs), mild systolic cardiac murmur (3), nuclear sclerosis (2), mild conjunctivitis (1), and bilateral entropion (1). Results of a CBC, serum biochemical analyses, and tests of thyroid gland function were within reference ranges in all Akitas, Beagles, and Miniature and Toy Poodles. In all Golden Retrievers, serum T4 and cTSH concentrations were within reference ranges (1.0 to 3.6 μg/dL and 0 to 0.6 ng/mL, respectively), and results for serum TGAA were negative. Serum fT4 concentration was within reference range (0.8 to 3.5 ng/dL) in 32 Golden Retrievers and was 0.7 ng/dL in 4 dogs (Table 1). None of these 4 Golden Retrievers had any historical or physical examination findings, clinical signs, or clinicopathologic abnormalities consistent with hypothyroidism or developed such signs or physical examination findings during the follow-up periods of 3 months (n = 1), 1 year (2), and 5 years (1). Both thyroid lobes appeared fusiform or elliptical in the longitudinal imaging plane and triangular or round to oval in the transverse imaging plane in all breeds (Figure 1). A fusiform shape of thyroid lobes was most common in the longitudinal imaging plane and was identified in 12 of 24 (50%) thyroid lobes in Akitas, 52 of 72 (72%) thyroid lobes in Golden Retrievers, 21 of 24 (88%) thyroid lobes in Beagles, and 18 of 24 (75%) thyroid lobes in Miniature and Toy Poodles. The shape of the thyroid lobe was more variable in the transverse imaging plane; a round to oval shape was identified in 17 of 24 (71%) thyroid lobes in Akitas, 14 of 24 (58%) thyroid lobes in Beagles, and 19 of 24 (79%) thyroid lobes in Miniature and Toy Poodles. A triangular shape was most common in Golden Retrievers (46/72 [64%] thyroid lobes). The frequency of the triangular shape of the thyroid lobe was significantly (P < 0.001) greater in Golden Retrievers, compared with each of the other breeds. In most dogs, the shape of the left, compared with the right, thyroid lobe was similar in the longitudinal and transverse imaging planes (8 and 7/12 Akitas, 30 and 24/36 Golden Retrievers, 9 and 6/12 Beagles, and 8 and 7/12 Miniature and Toy Poodles, respectively). The capsule of each thyroid lobe appeared smooth in all dogs except 3 Golden Retrievers, in which 1 or both lobes had an irregular surface; in those 3 Golden Retrievers, results of all tests of thyroid gland function were within reference ranges.
Typical longitudinal (A, C, and E) and transverse (B, D, and F) ultrasonographic images of thyroid lobes in healthy small-, medium-, and large-breed dogs. Representative images were obtained during evaluation of a Golden Retriever (large-breed dog; A and B), Beagle (medium-breed dog; C and D), andToy Poodle (small-breed dog; E and F). Margins of the thyroid lobes are marked by arrows or cursors (or both). Mean length, width, and height of the thyroid lobes were 24, 8, and 5 mm in Golden Retrievers; 20, 5, and 4 mm in Beagles; and 17, 3, and 3 mm in Miniature andToy Poodles, respectively. CR = Cranial. CA = Carotid artery. TR = Trachea.
Citation: American Journal of Veterinary Research 67, 1; 10.2460/ajvr.67.1.70
Thyroid function test results in healthy Akitas, Golden Retrievers, Beagles, and Miniature and Toy Poodles.
| Variable | Akita (n=12) | Golden Retriever* (36) | Beagle (12) | Toy and Miniature Poodles (12) | Reference range |
|---|---|---|---|---|---|
| Serum T4 (μg/dL) | 2.4 ± 0.9 | 2.0 ± 0.5 | 1.9 ± 0.6 | 1.9 ± 0.5 | 1.0–3.6 |
| Serum fT4 (ng/dL) | 1.1 ± 0.3 | 1.1 ± 0.4 | 1.3 ± 0.5 | 1.3 ± 0.5 | 0.8–3.5 |
| Serum cTSH (ng/mL) | 0.1 ± 0.1 | 0.1 ± 0.1 | 0.2 ± 0.1 | 0.2 ± 0.1 | 0–0.6 |
| TGAA positive sera | None | None | None | None | NA |
Echogenicity of the thyroid lobe was variable in all breeds. Echogenicity was hyperechoic, isoechoic, and hypoechoic, compared with the surrounding musculature in 18 (75%), 6 (25%), and 0 (0%) of 24 thyroid lobes in Akitas; in 47 (65%), 21 (29%), and 4 (6%) of 72 thyroid lobes in Golden Retrievers; in 13 (54%), 6 (25%), and 5 (21%) of 24 thyroid lobes in Beagles; and in 13 (54%), 10 (42%), and 1 (4%) of 24 thyroid lobes in Miniature and Toy Poodles, respectively. There was an increased frequency of hypoechoic thyroid lobes in Beagles, compared with all other breeds, but this difference was not significant (P = 0.07). Echogenicity of the right versus echogenicity of the left thyroid lobe was similar in most dogs (10/12 Akitas, 35/36 Golden Retrievers, 9/12 Beagles, and 8/12 Miniature and Toy Poodles). The echogenic pattern within the thyroid lobe was homogeneous in most dogs (16/24 [67%] thyroid lobes in Akitas, 62/72 [86%] thyroid lobes in Golden Retrievers, 18/24 [75%] thyroid lobes in Beagles, and 24/24 [100%] thyroid lobes in Miniature and Toy Poodles). In the remaining dogs, thyroid lobes had a heterogeneous distribution of echoes caused by hyperechoic foci (5 Golden Retrievers, 3 Beagles, and 1 Akita), hypoechoic foci (1 Akita), or a mottled appearance of both thyroid lobes (2 Akitas).
Within each breed, the mean lengths, widths, heights, and volumes of the left and right thyroid lobes were similar with the following exceptions: height of the LL versus height of the RL in Golden Retrievers and height and width of the LL versus height and width of the RL in Akitas (P < 0.05). Mean heights of thyroid lobes on longitudinal versus transverse images were similar in each breed except the RL height in Golden Retrievers, which was greater (P < 0.05) in transverse images.
In Miniature and Toy Poodles, mean lengths, widths, heights, and volumes of thyroid lobes were similar except the LL width, which was greater (P < 0.05) in Miniature Poodles (3.4 ± 0.3 mm) than in Toy Poodles (2.8 ± 0.5 mm). Mean length, width, and height of thyroid lobes were greater in Akitas and Golden Retrievers, compared with findings in Beagles and Miniature and Toy Poodles, and greater in Beagles, compared with findings in Miniature and Toy Poodles (Table 2). Mean total thyroid gland volume was similar in Akitas and Golden Retrievers; these values were greater than mean total thyroid gland volume in Beagles and Miniature and Toy Poodles. By use of the 95% confidence interval of the distribution of all total thyroid gland volumes, reference ranges for total thyroid gland volume were 63 to 1,912 mm3, 315 to 1,580 mm3, 128 to 713 mm3, and 73 to 244 mm3 in Akitas, Golden Retrievers, Beagles, and Miniature and Toy Poodles, respectively. There was overlap in total thyroid gland volumes between individual small-breed and medium-breed dogs and between medium-breed and large-breed dogs, but not between small-breed and large-breed dogs. Mean total thyroid gland volume was similar for male versus female and for sexually intact versus neutered Golden Retrievers and dogs of all breeds together.
Mean ± SD (range) size and volume of the thyroid lobes determined ultrasonographically in healthy Akitas, Golden Retrievers, Beagles, and Miniature and Toy Poodles.
| Variable | Akita (n=12) | Golden Retriever(36)* | Beagle (12) | Poodle (12) |
|---|---|---|---|---|
| Longitudinal image | ||||
| LL length (mm) | 24.2 ± 3.7†(3,4) | 24.8 ± 2.9†(3,4) | 20.0 ± 3.3†(1,2)‡(4) | 16.8 ± 1.7†(1,2)‡(3) |
| (18.0–31.6) | (18.9–32.4) | (11.5–22.7) | (13.2–19.1) | |
| RL length (mm) | 24.5 ± 3.4†(3,4) | 24.4 ± 2.6†(3,4) | 20.0 ± 2.9†(1,2)‡(4) | 16.8 ± 1.7†(1,2)‡(3) |
| (19.3–30.1) | (18.3–33.1) | (15.5–25.6) | (14.1–19.6) | |
| LL height (mm) | 5.0 ± 0.8†(3,4)‡(2) | 4.3 ± 0.6†(4)‡(1) | 4.1 ± 0.7†(1,4) | 3.2 ± 0.6†(1,2,3) |
| (3.5–6.5) | (3.2–5.8) | (2.9–4.8) | (2.0–4.1) | |
| RL height (mm) | 6.3 ± 1.5†(2,3,4) | 4.9 ± 1.0†(1,4) | 4.2 ± 0.9†(1)‡(4) | 3.0 ± 0.5†(1,2)‡(3) |
| (4.3–9.6) | (3.0–7.6) | (2.6–5.6) | (2.3–3.9) | |
| Transverse image | ||||
| LL height (mm) | 5.3 ± 1.5†(2,3,4) | 4.3 ± 0.7†(1,4) | 3.9 ± 0.7†(1)‡(4) | 2.9 ± 0.8†(1,2)‡(3) |
| (3.7–8.1) | (3.3–6.4) | (2.7–5.3) | (2.0–4.7) | |
| RL height (mm) | 6.4 ± 1.7†(3,4)‡(2) | 5.2 ± 1.3†(4)‡(1) | 4.1 ± 1.0†(1) | 3.0 ± 0.4†(1,2) |
| (3.5–9.1) | (3.1–8.5) | (2.6–5.9) | (2.0–3.5) | |
| LL width (mm) | 7.3 ± 1.7 †(3,4) | 7.9 ± 1.5†(3,4) | 4.9 ± 1.1†(1,2)‡(4) | 3.1 ± 0.5†(1,2)‡(3) |
| (5.1–9.6) | (4.0–11.0) | (3.4–6.3) | (2.3–3.9) | |
| RL width (mm) | 6.0 ± 1.6 †(4)‡(2) | 7.6 ± 1.9 †(3,4)‡(1) | 4.9 ± 1.1†(2)‡(4) | 3.0 ± 0.5 †(1,2)‡(3) |
| (4.1–9.4) | (2.8–11.8) | (3.5–6.2) | (2.2–3.9) | |
| Left lobe volume (mm3) | 486.0 ± 203.8†(3,4) | 446.9 ± 151.1†(3,4) | 199.5 ± 54.1†(1,2) | 78.0 ± 21.2†(1,2) |
| (218.2–905.2) | (178.2–805.4) | (71.9–281.7) | (46.2–126.3) | |
| RL volume (mm3) | 501.5 ± 246.4†(3,4) | 500.7 ± 189.9†(3,4) | 220.7 ± 97.5†(1,2) | 80.6 ± 23.2†(1,2) |
| (217.4–1,146.9) | (132.3–1,061.3) | (92.3–414.2) | (31.8–102.5) | |
| Total thyroid gland | 987.4 ± 420.1†(3,4) | 947.6 ± 311.7†(3,4) | 420.2 ± 133.0†(1,2) | 158.5 ± 38.8†(1,2) |
| volume (mm3) | (535.5–1935.7) | (424.6–1743.0) | (164.2–618.6) | (78.0–228.9) |
In superscript parentheses, 1 = Akitas, 2 = Golden Retrievers, 3 = Beagles, and 4 = Miniature and Toy Poodles.
Data reported previously.13,a,j
Significant (P < 0.01) difference between this value and values for groups identified in superscript parentheses.
Significant (P<0.05) difference between this value and values for groups identified in superscript parentheses.
The relationships of thyroid lobe size and total thyroid gland volume each with body weight and BSA in all breeds combined were examined. Left lobe and RL length (r = 0.73), width (LL, r = 0.76; RL, r = 0.62), and total thyroid gland volume (r = 0.73) correlated with body weight (P < 0.001; Figure 2). Thyroid lobe length (LL, r = 0.74; RL, r = 0.75), width (LL, r = 0.78; RL, r = 0.65), and total thyroid gland volume (r = 0.74) also correlated with BSA (P < 0.001; Figure 2). The height of both thyroid lobes in both imaging planes correlated with body weight and BSA (P < 0.001; r ranged from 0.51 to 0.57). There were weak but significant negative correlations between age of the dogs and thyroid lobe length (RL, r = -0.31; P < 0.01), width (LL, r = -0.37; RL, r = -0.29; P < 0.05), and total thyroid gland volume (r = -0.30; P < 0.01). Total thyroid gland volume did not correlate with BCS (r = 0.05; P = 0.690).
Relationship between total thyroid gland volume and body weight (r = 0.73; A) and between total thyroid gland volume and BSA (r= 0.74; B) in 12 healthy Akitas (closed circles), 36 Golden Retrievers (open circles), 12 Beagles (triangles), and 12 Miniature and Toy Poodles (squares).
Citation: American Journal of Veterinary Research 67, 1; 10.2460/ajvr.67.1.70
On the basis of significant relationships identified via simple linear regression, we examined whether thyroid lobe size or thyroid gland volume could be predicted with multiple linear regression models including body weight, BSA, and age. Multiple linear regression models revealed that age plus body weight and age plus BSA were significant (P < 0.001) determinants of RL length (R2 = 0.59 and 0.60, respectively), width (R2 = 0.43 and 0.46, respectively), and LL width (R2 = 0.66 and 0.67, respectively). Weight and BSA, but not age, were significant (P < 0.001) determinants of LL length (R2 = 0.55 and 0.56, respectively) and LL and RL height (R2 ranging from 0.26 to 0.39). In multiple linear regression models, significant (P < 0.001) relationships were also identified among thyroid gland volume, age, and BSA (total thyroid gland volume [mm3] = 118.6 - 31.8 × age [years] + 936.2 × BSA [m2]; R2 = 0.59) and among thyroid gland volume, age, and weight (total thyroid gland volume [mm3] = 311.8 - 33.3 × age [years] + 23.7 × body weight [kg]; R2 = 0.58). However, plotting total thyroid gland volumes predicted from the model with age and body weight versus total thyroid gland volumes calculated from thyroid lobe measurements revealed that this model does not provide a good fit (Figure 3).
Relationship between predicted total thyroid gland volume (based on age and body weight) and calculated total thyroid gland volume (based on thyroid lobe measurements) in 12 Akitas, 36 Golden Retrievers, 12 Beagles, and 12 Miniature and Toy Poodles. The predicted total thyroid gland volumes were derived by use of the following equation: total thyroid gland volume (mm3) = 311.8 - 33.3 × age + 23.7 × body weight, where age is in years and weight is in kilograms. For this equation, R2 = 0.58 (P < 0.001); ideally, all datum points should be located close to the line of identity. Age and body weight alone do not fully predict the calculated total thyroid gland volume, and there may be other factors determining thyroid gland volume in healthy dogs.
Citation: American Journal of Veterinary Research 67, 1; 10.2460/ajvr.67.1.70
Discussion
In most dogs in each breed, the thyroid capsule had a smooth surface; the thyroid parenchyma had a homogeneous echogenic pattern and was usually hyperechoic or isoechoic, compared with the surrounding musculature, and the shapes and echogenicities of the left and right thyroid lobes were similar. Although most dogs had fusiform thyroid lobes on longitudinal images, thyroid lobe shapes were quite variable. This was especially true for shapes identified on transverse images, where the frequency of a triangular shape of the thyroid lobe was greater in Golden Retrievers than in the other breeds. The clinical relevance of this finding is not known. In 1 study13 of Golden Retrievers with hypothyroidism, the frequency of a round to oval shape of the thyroid lobe was greater than that of a triangular shape of the thyroid lobe in transverse imaging planes. Unfortunately, the common finding of a round to oval-shaped thyroid lobe in transverse imaging planes in the breeds evaluated in the present study, including Golden Retrievers, suggests that the shape of the thyroid lobe on transverse images is not useful for differentiating between euthyroid and hypothyroid dogs.
Thyroid lobe echogenicity was variable within all breeds. Thyroid hypoechogenicity was an uncommon finding except in Beagles (5/24 [21%] thyroid lobes in Beagles vs 0/24 [0%] thyroid lobes in Akitas, 4/72 [6%] thyroid lobes in Golden Retrievers, and 1/24 [4%] thyroid lobes in Miniature and Toy Poodles). Hypoechoic thyroid gland parenchyma is a common ultrasonographic finding in humans with autoimmune thyroiditis27,28 and Golden Retrievers with hypothyroidism.13 It is not known whether hypoechoic thyroid lobes are normal for Beagles or represent subclinical lymphocytic thyroiditis in some of the Beagles evaluated in our study. Unlike the other breeds in the present study, 5 of 12 Beagles were colony raised. In 1 research colony, lymphocytic thyroiditis was identified via histologic examination of thyroid tissue samples from 29 of 276 (11%) Beagles that did not show clinical signs of hypothyroidism.29 None of the Beagles in our study had clinical or laboratory evidence of hypothyroidism, and serum TGAA test results were negative in all dogs. Because histologic examination of thyroid tissue was not performed, subclinical lymphocytic thyroiditis remains a potential explanation for the trend toward an increased frequency of hypoechoic thyroid lobes detected ultrasonographically in the Beagles of our study, compared with the frequency of hypoechoic thyroid lobes in the other breeds.
Although a few differences were identified, the size and volume of the left and right thyroid lobes were essentially similar within each breed of dog. An ultrasonographic study21 in newborn infants also revealed that the left and right thyroid lobes were of similar size and volume and identified a strong positive correlation between LL and RL volumes. However, the volume of 1 lobe could not be used to reliably predict that of the other lobe, and measurement of all 3 dimensions of each thyroid lobe and calculation of thyroid lobe and total thyroid gland volumes were recommended to determine normalcy of the thyroid gland.
Results of our study indicated marked variations of thyroid gland size and volume in dogs within breeds and overlap of size and volume ranges among breeds. Comparisons of mean values for thyroid gland size and volume among breeds revealed significant differences; thyroid lobe volume and total thyroid gland volume were similar in the large breeds (Akitas and Golden Retrievers), and these values were greater in the large-breed dogs, compared with values in the medium-breed (Beagles) and small-breed (Miniature and Toy Poodles) dogs. Many measurements of thyroid lobe size revealed similar differences among breeds; in addition, thyroid lobe size in medium-breed dogs was significantly greater than that in small-breed dogs. When data from all dogs were analyzed together, there were significant and strong positive correlations of thyroid gland size and volume with body weight and BSA. These findings suggest that differences in thyroid gland size and volume exist among breeds of dogs and that these differences may be related to size of the dog rather than its breed. Positive correlations between thyroid gland volume and body weight14,15,30,31 and between thyroid gland volume and BSA15,16,31 have been identified in humans. Interestingly, there was no significant difference in total thyroid gland volume between female and male dogs in our study nor between sexes in same-age groups of children,19,31 but thyroid gland volume is larger in adult men than in adult women as a consequence of differences in body weight and lean body mass between sexes.14,15,30
Multiple linear regression models that included total thyroid gland volume as the response variable and age and body weight or BSA as predictors identified age, weight, and BSA as significant determinants of total thyroid gland volume in healthy Akitas, Golden Retrievers, Beagles, and Miniature and Toy Poodles. On the basis of coefficients of determination (R2) from these models, weight plus age and BSA plus age accounted for 58% and 59% of the variation in total thyroid gland volume, respectively. Similar coefficients of determination were obtained when weight, BSA, and age were used for predictions of thyroid lobe length and LL width, but predictions of RL width and thyroid lobe height were less adequate. Unfortunately, plotting the predicted total thyroid gland volume derived from the regression model, including age and body weight versus the calculated total thyroid gland volume derived from the ultrasonographic measurements, revealed that this regression model did not provide a good fit, particularly at volumes > 1,000 mm3. Possible explanations include the small number of breeds (and hence only 4 groups of body weights) evaluated in the present study and the likelihood that other factors also determine total thyroid gland volume in dogs.
Because the calculation of total thyroid gland volume includes all 3 dimensions of both thyroid lobes, this variable appears most comprehensive for assessing normalcy of thyroid gland dimensions. In 1 study,13 decreases in thyroid lobe volume and total thyroid gland volume were the most consistent ultrasonographic findings supportive of hypothyroidism in Golden Retrievers, followed by a decrease in thyroid lobe size. Results of the present study suggest that body size of dogs should be considered when assessing thyroid gland size and volume in dogs with suspected hypothyroidism. A weak but significant negative correlation between total thyroid gland volume and age was also identified, suggesting that age of the dog may also have to be considered when assessing thyroid gland size and volume. Body weight and BSA had equally strong relationships with total thyroid gland volume in the dogs of the present study, but most dogs had approximately the same BCS (ie, 4 to 6/9). In obese dogs, BSA may be a more reliable variable than weight for estimating normal thyroid gland volume.
The present study has provided normative data for ultrasonographic assessment of the thyroid gland in healthy small-breed (Miniature and Toy Poodles), medium-breed (Beagles), and large-breed (Golden Retrievers and Akitas) dogs. The significant increase in thyroid gland size and volume of small through large breeds and the strong positive correlation of thyroid gland size and volume with body weight and BSA of all dogs in our study suggest that differences in thyroid gland size and volume associated with the size of the dog should be taken into account in diagnostic imaging of canine thyroid glands. The present study also identified wide variability in shape, echogenicity, size, and volume of the thyroid gland within breeds of healthy dogs and suggests that subtle changes in these variables may not be useful in distinguishing between the euthyroid and hypothyroid states. Further studies are needed to assess the extent of ultrasonographic changes of the thyroid gland in different breeds of dogs with hypothyroidism and better define the value of ultrasonography in identifying hypothyroidism in dogs.
| BSA | Body surface area |
| T4 | Thyroxine |
| FT4 | Free thyroxine |
| cTSH | Canine thyroid-stimulating hormone |
| TGAA | Thyroglobulin autoantibody |
| BCS | Body condition score |
| LL | Left lobe |
| RL | Right lobe |
Brömel C, Nelson RW, Samii V, et al. Ultrasonographic evaluation of the thyroid gland in Golden Retrievers (abstr). J Vet Intern Med 2001;15:297.
ATL HDI 3000, ATL, Bothell, Wash.
Hitachi 717 chemistry analyzer, Roche Diagnostics, Indianapolis, Ind.
Hitachi 912 chemistry analyzer, Vet-T4 assay, Roche Diagnostics, Indianapolis, Ind.
Free T4 kit, Nichols Institute Diagnostics, San Juan Capistrano, Calif.
Coat-A-Count Canine TSH IRMA, Diagnostic Products Corp, Los Angeles, Calif.
Canine thyroglobulin autoantibody enzyme immunoassay kit, Oxford Biomedical Research, Oxford, Mich.
Baker System 9000, ABX Diagnostics, Irvine, Calif (until July 2001); Advia 120 Hematology System, Bayer Diagnostics, Tarrytown, NY (after July 2001).
BMDP Statistical Software Inc, Los Angeles, Calif (1990); S-Plus, Insightful Corp, Seattle, Wash (2003); StatXact, Cytel Statistical Software, Cambridge, Mass (2004).
Brömel C, Nelson RW, Pollard RE, et al. Ultrasonographic evaluation of the thyroid gland in canine breeds predisposed for hypothyroidism (abstr). J Vet Intern Med 2002;16:632.
References
- 1↑
Barraclough BMBarraclough BH. Ultrasound of the thyroid and parathyroid glands. World J Surg 2000; 24: 158–165.
- 2↑
Reschke KKlose SKopf D, et al.Role of ultrasound in the diagnosis of thyroid autonomy. Exp Clin Endocrinol Diabetes 1998; 106: S42–S44.
- 3↑
Premawardhana LDParkes ABAmmari F, et al.Postpartum thyroiditis and long-term thyroid status: prognostic influence of thyroid peroxidase antibodies and ultrasound echogenicity. J Clin Endocrinol Metab 2000; 85: 71–75.
- 5
Yeh H-CFutterweit WGilbert P. Micronodulation: ultrasonographic sign of Hashimoto thyroiditis. J Ultrasound Med 1996; 15: 813–819.
- 6
Takamatsu JYoshida SYokozawa T, et al.Correlation of antithyroglobulin and antithyroid-peroxidase antibody profiles with clinical and ultrasound characteristics of chronic thyroiditis. Thyroid 1998; 8: 1101–1106.
- 7
Karwowski JKJeffrey RBMcDougall IR, et al.Intraoperative ultrasonography improves identification of recurrent thyroid cancer. Surgery 2002; 132: 924–928.
- 8
Papini EGuglielmi RBianchini A, et al.Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab 2002; 87: 1941–1946.
- 9↑
Harvey JNParker DDe P, et al.Sonographically guided core biopsy in the assessment of thyroid nodules. J Clin Ultrasound 2005; 33: 57–62.
- 10↑
Cohen OPinhas-Hamiel OSivan E, et al.Serial in utero ultrasonographic measurements of the fetal thyroid: a new complementary tool in the management of maternal hyperthyroidism in pregnancy. Prenat Diagn 2003; 23: 740–742.
- 11↑
Kreisner ECamargo-Neto EMaia CR, et al.Accuracy of ultrasonography to establish the diagnosis and aetiology of permanent primary congenital hypothyroidism. Clin Endocrinol 2003; 59: 361–365.
- 12↑
Wisner ERNyland TG. Ultrasonography of the thyroid and parathyroid glands. Vet Clin North Am Small Anim Pract 1998; 28: 973–991.
- 13↑
Brömel CPollard REKass PH, et al.Ultrasonographic evaluation of the thyroid gland in healthy, hypothyroid, and euthyroid Golden Retrievers with nonthyroidal illness. J Vet Intern Med 2005; 19: 499–506.
- 14↑
Hegedüs LPerrild HPoulsen LR, et al.The determination of thyroid volume by ultrasound and its relationship to body weight, age, and sex in normal subjects. J Clin Endocrinol Metab 1983; 56: 260–263.
- 15↑
Wesche MFWiersinga WMSmits NJ. Lean body mass as a determinant of thyroid size. Clin Endocrinol 1998; 48: 701–706.
- 16
Gomez JMMaravall FJGomez N, et al.Determinants of thyroid volume as measured by ultrasonography in healthy adults randomly selected. Clin Endocrinol 2000; 53: 629–634.
- 17
Sari RBalci MKAltunbas H, et al.The effect of body weight and weight loss on thyroid volume and function in obese women. Clin Endocrinol 2003; 59: 258–262.
- 18
Chanoine JPToppet VLagasse R, et al.Determination of thyroid volume by ultrasound from the neonatal period to late adolescence. Eur J Pediatr 1991; 150: 395–399.
- 19
Vitti PMartino EAghini-Lombardi F, et al.Thyroid volume measurement by ultrasound in children as a tool for the assessment of mild iodine deficiency. J Clin Endocrinol Metab 1994; 79: 600–603.
- 20
Wisner ERThéon APNyland TG, et al.Ultrasonographic examination of the thyroid gland of hyperthyroid cats: comparison to 99mTcO4– scintigraphy. Vet Radiol Ultrasound 1994; 35: 53–58.
- 21↑
Perry RJHollman ASWood AM, et al.Ultrasound of the thyroid gland in the newborn: normative data. Arch Dis Child Fetal Neonatal Ed 2002; 87: F209–F211.
- 22↑
Horney BSMacKenzie ALBurton SA, et al.Evaluation of an automated, homogeneous enzyme immunoassay for serum thyroxine measurement in dog and cat serum. Vet Clin Pathol 1999; 28: 20–28.
- 23↑
Schachter SNelson RWScott-Moncrieff C, et al.Comparison of serum-free thyroxine concentrations determined by standard equilibrium dialysis, modified equilibrium dialysis, and 5 radioimmunoassays in dogs. J Vet Intern Med 2004; 18: 259–264.
- 24
Peterson MEMelian CNichols 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: 1396–1402.
- 25
Williams DAScott-Moncrieff JCBruner J, et al.Validation of an immunoassay for canine thyroid-stimulating hormone and changes in serum concentration following induction of hypothyroidism in dogs. J Am Vet Med Assoc 1996; 209: 1730–1732.
- 26↑
Scott-Moncrieff JCAzcona-Olivera JGlickman NW, et al.Evaluation of antithyroglobulin antibodies after routine vaccination in pet and research dogs. J Am Vet Med Assoc 2002; 221: 515–521.
- 27
Mazziotti GSorvillo FIorio S, et al.Grey-scale analysis allows a quantitative evaluation of thyroid echogenicity in the patients with Hashimoto's thyroiditis. Clin Endocrinol (Oxf) 2003; 59: 223–229.
- 28
Schiemann UAvenhaus WKonturek JW, et al.Relationship of clinical features and laboratory parameters to thyroid echogenicity measured by standardized grey scale ultrasonography in patients with Hashimoto's thyroiditis. Med Sci Monit 2003; 9: MT13–MT17.
- 29↑
Benjamin SAStephens LCHamilton BF, et al.Associations between lymphocytic thyroiditis, hypothyroidism, and thyroid neoplasia in Beagles. Vet Pathol 1996; 33: 486–494.
- 30
Berghout AWiersinga WMSmits NJ, et al.Determinants of thyroid volume as measured by ultrasonography in healthy adults in a non-iodine deficient area. Clin Endocrinol 1987; 26: 273–280.
- 31
Semiz SSenol UBircan O, et al.Correlation between age, body size and thyroid volume in an endemic area. J Endocrinol Invest 2001; 24: 559–563.
